The International Congress and Exhibition on Noise Control Engineering

The 2001 International Congress and Exhibition on Noise Control Engineering The Hague, The Netherlands, 2001 August 27-30 Abstracts Edited by Rinus Boone Technical Program Chairman Permission hereby granted for any person to reproduce a fractional part of any abstract herein, provided that the permission is obtained from its author(s) and credit is given to the author(s) and to this abstract book. An author or his or her research sponsor may reproduce his or her abstract in full, crediting this abstract book. This permission is not assignable. ISSN: 0105-175x ISBN: 90-806554-1-4 Organisation Organising Committee: Tjeert ten Wolde – General Chairman Rinus Boone – Technical Program Chairman Rob Hoffman – Exhibition manager Erik Druyvesteyn – Secretary Jan Granneman – Treasurer Hans Verspoor – Sponsoring Paul de Vos – Publicity Rob Witte – Social Program and Facilities Congress Secretariat: Mrs. Gerda Boone, Voorweg 105-A, 2715 NG Zoetermeer, The Netherlands e-mail: gerda.boone@wxs.nl The International Advisory Committee Marion Burgess, Australia Judith Lang, Austria André Cops, Belgium Samir N.Y. Gerges, Brazil Gilles Daigle, Canada Jing Tian, China Josef Novák, Czech Republic Jørgen Kragh, Denmark Hannu Nykänen, Finland Jean Tourret, France Joachim Scheuren, Germany Konstantin Vogiatzis, Greece Fülöp Augusztinovicz, Hungary N. Tandon, India The Technical Program Committee Chairman Rinus Boone, University of Technology Delft Co-chairman Tjeert ten Wolde, TNO-TPD Gijs-Jan van Blokland, M+P Noise & Vibration Consultants Gerard Bekebrede, Directorate General of Civil Aviation, RLD Giovanni Brambilla, 17th ICA, Rome Internoise 2001 Abstracts 3 Alessandro Cocchi, Italy Masaru Koyasu, Japan Adriaan Houtsma, The Netherlands Keith Ballagh, New Zealand Truls Gjestland, Norway Zbigniew Engel, Poland José Palma, Portugal Yuri Bobrovnitskii, Russia Salvador Santiago, Spain Ulf Sandberg, Sweden Beat Hohmann, Switzerland Zerhan Karabiber, Turkey Geoff Leventhal, United Kingdom Paul Schomer, U.S.A. Harry Brouwer, National Aerospace Laboratory NLR Niek Doelman, TNO-TPD Steve Elliott, Institute of Sound and Vibration Research Eddy Gerretsen, TNO-TPD and Eindhoven University of Technology Henk Miedema, TNO Prevention and Health Paul Sas, K.U. Leuven Paul de Vos, AEA Technology Rail BV Rob Witte, dgmr Consulting Engineers Main Sponsors Ministry of Housing, Spatial Planning and the Environment (The Hague) TNO TPD (Delft) Ministry of Transport and Public Works and Water Management (The Hague) ONRI, the Association of Dutch Consulting Engineers (The Hague) NS Railinfrabeheer (NS Rail infrastructure Management, Utrecht) The European Commission (Brussels) DCMR Environmental Protection Agency (The Hague) Port of Rotterdam Delft University of Technology, Lab. of Acoustical Imaging and Sound Control University of Technology Twente (Enschede) IGB Leerdam (Leerdam) Peutz & Associates consultants (Zoetermeer) AEA Technology Rail b.v. (Utrecht) dgmr consultants (The Hague) 4 Internoise 2001 Abstracts Table of contents Guidelines Tuesday August 28 Keynote Address by Henk Miedema Session A1-1: Costs of environmental noise control Session A1-2: Cost-benefits of noise control Session A1-3: Road traffic noise Session A1-4: Measurement of road traffic noise Session A2-1: Active noise control in ducts Session A2-2: Active noise control in enclosures Session A2-3: Active control of structural vibrations Session A3-1: Annoyance 1 Session A3-2: Annoyance 2 Session A3-3: Combined effects of noise on humans Session A4-1: Human effects of vibration Session A4-2: Noise and recreation Session A4-3: Rating Prominent Discrete Tones Session A5-1: Outdoor Machinery noise Session A5-2: Noise policy worldwide Session A5-3: Vibro-acoustic sources Session A6-1: Application of new measurement technologies to building acoustics Session A6-2: Fan noise Session A6-3: Vehicle interior noise Session A7-1: Effectiveness of aircraft noise control Session A7-2: Energy flow methods Session A7-3: Sound power and intensity measurements Session A8-1: Outdoor noise prediction 1 Session A8-2: Outdoor noise prediction 2 Session A9-1: Low frequency noise control Session A9-2: Mufflers, silencers and pipes Session A9-3: Aero-acoustic sources 7 9 9 14 17 21 23 29 32 36 40 45 48 57 60 61 64 69 73 76 79 86 92 96 98 102 109 111 118 124 124 129 131 134 138 142 5 Wednesday August 29 Plenary Lecture by Jiri Tichy Session B1-1: Tyre/road noise modeling Session B1-2: Low noise tyre design Session B1-3: Low noise road design and application Session B2-1: Active control of structural vibrations (continued) Session B2-2: Active control of sound radiation and transmission Session B2-3: Active control in aircraft engines Internoise 2001 Abstracts Session B2-4: Active noise control at the ear Session B3-1: Quiet façade and insulation Session B3-2: Effects of noise on sleep and stress and cardiovascular risks Session B3-3: Effects of noise on children Session B4-1: Noise and neighbours Session B4-2: Sound quality Session B4-3: Soundscapes Session B5-1: Railway noise Session B5-2: Beamforming arrays Session B6-1: Resilient elements and materials Session B6-2: Shipboard noise control Session B6-3: Building acoustics Session B7-1: Aircraft noise sources Session B7-2: Detection and identification of noise sources Session B7-3: Aircraft noise modeling and monitoring Session B8-1: Noise barrier attenuation Session B8-2: Military weapon noise mitigation Session B8-3: Numerical modeling 146 147 151 156 159 161 167 171 178 183 187 191 199 202 204 211 218 221 224 225 229 231 233 238 240 244 247 251 254 261 268 271 277 283 Thursday August 30 Plenary Lecture by Ulf Sandberg Session C1-1: Measurement of tyre/road noise Session C1-2: Absorbing materials and surfaces Session C2-1: Active noise control at the ear (continued) Session C2-2: Practical aspects of active control Session C2-3: Occupational noise Session C3-1: Hearing loss by leisure activities Session C3-2: Effects of low frequency noise Session C4-1: Industrial noise Session C4-2: Impulsive noise: effects, prediction and control Session C5-1: Noise mapping Session C6-1: Structure borne sound Session C7-1: Acoustical diagnosis methods Session C7-2: Instrumentation and measurement techniques Session C8-1: Nearfield Acoustic Holography Authors index 6 Internoise 2001 Abstracts Guidelines In this abstractbook all abstracts have been arranged according to the sessions of the congress. Each abstract starts with a heading showing the session code, a starting time, the paper number with the type of paper and the I-INCE Classification. The session code starts with A, B or C to indicate the day: Tuesday, Wednesday or Thursday, respectively. Next follows the room number of the session, followed by a serial number given to the session. Oral presentations in the parallel sessions take one timeslot of 20 minutes (including discussions and time for room change), except for the tutorial papers. For the latter category two timeslots have been designated. The poster presentations need some further explanation. The poster presentations have also been assigned to the sessions and in this book their abstracts are placed at the end of each session. If there is more than one poster within a session one or two timeslots have been designated for a short oral introduction of the posters by the authors. In that case you find the starting time after the session code in the same way as for the oral presentations. Notice that there are several introductions during the same time slot. The session chairmanship will take care of that. There are also sessions with only one poster presentation. In that case no time slot has been allocated for an introduction of the poster by the author. The poster abstracts have a second heading showing their posterboard number, the day of display (Tuesday, Wednesday or Thursday) and a time period during which the author is expected to be present for explanations at his or her posterboard. All posters are on display for one full congress day, on Tuesday and Wednesday between 9:40 and 18:40 h and on Thursday between 9:40 and 16:00 h. The time schedule is such that the author explanations at the posterboard location are always after the corresponding introductory in the oral session. Consequently, many Posters are on display one day after their oral session. Internoise 2001 Abstracts 7 8 Internoise 2001 Abstracts Tuesday August 28 Keynote Address 8:20 – 9:20 h Henk M.E. Miedema TNO-PG, P.O. Box 2215, 2301 CE Leiden NOISE & HEALTH: HOW DOES NOISE AFFECT US? Paper number 740 I-INCE Class 60 Many people are annoyed by environmental or neighbourhood noise. They report that noise interferes with communication, distracts from concentrated activities, or disturbs the sleep. These are clear but broad indications that noise causes adverse effects. Knowledge of the underlying, more elementary effects induced by noise helps us in getting a better understanding of the nature of the reported effects. Moreover, more elementary insights can bring effects to our attention that people are not aware of, it may give information on the existence of vulnerable sub-populations, and it may provide information on the role of non-acoustical factors. These points are discussed after an overview of research findings concerning the working mechanisms of noise. The influence of noise on the following four important processes are discussed: auditory perception, attention, activity of the hypothalamus-pituitary-adrenal (HPA) axis, and sleep. The weight of evidence is that the influence of environmental noise on these processes, affects our daily functioning in a potentially reversible way. However, because environmental and neighbourhood noise exposures usually persist over time, there is little benefit from the reversibility of the effects and they actually often occur permanently during (a part of) life. Moreover, sustained primary reversible effects may cause severe irreversible effects (reduced cognitive ability, cardiovascular diseases). Session A1-1: Costs of environmental noise control Chaired by Soren Rasmussen and Tom Worsley Session A1-1, 9:40 h Paper number 314 (Invited, Oral) Jakob Oertli, Frank Elbers and Paul van der Stap I-INCE Class 67 Swiss Federal Railways, Rail Environmental Center, Parkterrasse 14, CH-3000 Bern 65, Switzerland THE STAIRRS PROJECT: A COST-BENEFIT ANALYSIS OF DIFFERENT MEASURES TO REDUCE RAILWAY NOISE ON A EUROPEAN SCALE Several countries have enacted severe noise control standards and European noise legislation is being considered as well. Detailed studies in Switzerland, the Netherlands as well as a UIC sponsored study on two freight lines have demonstrated very high costs for noise control, however - as shown by results to be presented in the paper - considerable savings can be achieved by choosing those noise control strategies with the best cost-benefit ratio. In an effort to determine optimal strategies on a European as well as on national levels, the E.U. 5th Framework Programme is co-financing the STAIRRS project (Strategies and Tools to Assess and Implement noise Reducing measures for Railway Systems). Work Package 1 is designed to develop the software necessary to undertake large scale cost-benefit analyses. The acoustically relevant geographic, traffic and track data have been Internoise 2001 Abstracts 9 collected for 11'000 km of line length in seven European countries. Noise calculations are undertaken with the Eurano 2001 software programme. Standard cost-benefit analysis methodologies were adapted to fit the requirements of noise control projects. A specifically developed extrapolation methodology enables studies on any geographic area of interest, be it individual countries, the EU or Europe as a whole. In parallel an optimisation algorithm allows determining optimal strategies for a specific line under given constraints. The STAIRRS project coordinator is the European Rail Research Institute, the work package leader are the Swiss Federal Railways with the participation of AEAT Technology (NL), German Railways, French Railways, PSI -A Kalivoda Consult (A),the Swiss Federal Institute of Technology and the Free University of Brussels. Session A1-1, 10:00 h Alfred Rust AVL List GmbH, Graz, Austria Paper number 390 (Invited, Oral) I-INCE Class 67 ENVIRONMENTAL NOISE CONTROL AT SOURCE: PASSENGER CARS AND TRUCKS Legislation and market demands drive the development of vehicles and engines including also the environmental aspects by legislative regularizations of exhaust emissions, noise, safety, material recycling etc. Much effort is applied to fulfil the current and the even more stringent future legislative noise limits. The final cost related to noise control arises indirectly from increased development effort and directly from additional hardware cost. The paper gives a survey on the current measures and future technologies of vehicle noise control. The technical approaches for the noise control of passenger cars and trucks are described and the effectiveness and cost of such noise control measures are taken into consideration. Session A1-1, 10:20 h Paper number 456 (Invited, Oral) Lars Ellebjerg Larsen and Hans Bendtsen I-INCE Class 67 Danish Transport Research Institute, Knuth-Winterfeldts Allé, Bygning 116 Vest, DK-2800 Lyngby, Denmark COSTS AND PERCEIVED NOISE REDUCTION OF POROUS ASPHALT PAVEMENTS In 1999 ÿster S¯gade in Copenhagen was rebuilt and paved with test sections of two layer porous asphalt and a reference section of dense asphalt concrete. The project includes an extensive measuring program on noise, road surface, and traffic safety. There has also been done a before-after questionnaire survey on noise annoyance among those living along the road. When the pavements were new the noise reduction on the test sections was 5-6 dB compared with the reference section, and a further 2,5 dB compared with the old pavement. The questionnaire survey reveals reductions in noise annoyance along the reference section as well as the test sections, with the greatest reduction along the test sections. The costs connected with porous asphalt pavements are calculated and compared with ordinary pavements. Finally the effects on noise levels and annoyance are compared with noise abatement through noise barriers and insulation of the facade using special windows. Two layer porous pavements are more expensive than ordinary pavements but cheaper than noise barriers and sound insulation. 10 Internoise 2001 Abstracts Session A1-1, 10:40 h Frank Elbers Paper number 579 (Invited, Oral) I-INCE Class 67 dBvision, Vondellaan 6, NL- 3521 GD Utrecht, The Netherlands COST-BENEFIT CALCULATION TO PREVENT MORE THAN 65 DB(A) FOR RAILWAY NOISE IN THE NETHERLANDS Noise legislation for railways is enacted since 1987 in The Netherlands. This legislation prevents more than 57 dB(A) for new situations, an increase when reconstructions are planned and a limited growth. To reduce annoyance for high noise levels a large reduction program is part of this legislation. This program prevents 65 dB(A) levels near houses. To prevent these values three kind of measures are taken into account: noise barriers, window insulation and reduction of noise from trains over steel bridges. This article presents the results of a cost-benefit analysis for the reduction by barriers for four different scenarios: 1. no source reduction, 2. replacement of wooden sleepers by concrete sleepers, 3. scenario 2 with additional rail dampers, 4. scenario 2 with a 7 dB(A) reduction of all Dutch vehicles with cast iron thread brake. Also the cost-benefit results are show for measures on steel bridges. The Dutch network has over 1.000 steel railway bridges. Results are shown to sort these out with a costbenefit focus on a top 25 list. Per bridge one or two kinds of measures are taken into account. Session A1-1, 11:00 h Margit Bonacker Paper number 673 (Invited, Oral) I-INCE Class 67 konsalt GmbH., Luruper Chaussee 125, D-22761 Hamburg, Germany EVALUATING A NOISE BARRIER ALONG A MOTORWAY WITH RESPECT TO ITS COSTS AND EFFECTIVENESS In the city of Magdeburg, a noise barrier will be constructed along a highway in a residential area which is highly affected by street noise. To analyse the sociopsychological effects of the noise barrier, a field study (panel) in three waves is conducted: one wave took place before the construction of the noise barrier, the second wave will take place one month after the construction, and another wave is planned for twelve months later. In the study, personal interviews with the same sample of residents are used to examine the degree of individual noise annoyance before and after the noise has been diminished. To control the actual noise load, acoustical data are related to the sociopsychological data. Also, a control group is included in the investigation, who will not profit by the noise barrier. Additionally, a number of respondents keep precise records of their activities during the day in order to see how much they are affected by the road noise, e.g. avoid certain activities or close windows, turn up the radio because of the noise. The results of the study are expected to indicate how the effects of a change in noise exposure relate to the actual reduction of noise load and to relate the costs and benefits of noise control. Session A1-1, 11:20 h Paper number 158 (Invited, Oral) Jochen Richard and Christian Popp COSTS RELATED IMPLEMENTATION OF NOISE ABATEMENT MEASURES I-INCE Class 67 PLANUNGSBÜRO RICHTER-RICHARD, Südstrasse 52, D-52064 Aachen, Germany The upcoming publication of the EU guidelines on noise abatement planning means that this environment-oriented set of planning instruments will be given a much higher status in most member states than heretofore. In addition to technical questions, questions will also be asked regarding the Internoise 2001 Abstracts 11 costs not only of the setting up of noise abatement planning but also of their implementation in particular. The cost of the implementation of the action plan on noise abatement planning must be seen in conjunction with the measures to be taken and these latter must be seen in the context of municipal planning policies. Additional costs need not necessarily arise in every case. As a rule of thumb it can be said: the stronger municipal development is geared towards sustained development (e.g. framework of an LA-21-process), the more economical the implementation of the noise abatement planning will be. Noise abatement can also bring in money. Lower rents are paid at the side of loud roads, there is a reduced level of maintenance, potential renewal zones are formed which have to be restored with the aid of state finances. In such areas the property prices necessarily drop after the rent levels have dropped. Measures for the abatement of noise can reverse this trend - so noise abatement not only costs money, but also, with noise abatement, money can be earned. Noise abatement measures and the costs associated with them will always be a part of a co-ordinated plan for sustainable urban development. Session A1-1, 11:40 h Henrik Pettersson Paper number 15 (Contributed, Oral) I-INCE Class 67 Metso Paper, Inc., Air Systems, Pansiontie 56, FIN-20240 Turku, Finland EXTERNAL NOISE CONTROL IN THE PULP AND PAPER INDUSTRY - EFFICIENCY AND ECONOMY Industrial noise sources can roughly be divided into internal and external noise sources. The major external noise sources of a paper mill are process and machine room ventilation, vacuum pumps and steam exhausts. External sound attenuation efficiency and costs are discussed in this paper. The noise control targets from authorities, customers and acoustics consultants will dictate the scope and costs of sound attenuation. Considerable cost reductions can be achieved if the noise control design starts at the early stage of a project. Additionally, depending on the noise level requirements, absorptive silencers, resonators or enclosures are used. Fans and vacuum pumps create broad band noise. When the requirements are strict, a combination of absorptive and reactive silencers must be used to cover the whole frequency range. Active Noise Control is yet another method of reducing noise, especially at the lower frequencies, where it is difficult and expensive to attenuate noise with passive means. In Active Noise Control, original noise and anti-noise cancel each other and the net result is greatly diminished sound. Active noise control systems are also introduced. Sensors and actuators are using ElectroMechanical Film (EMFi) technology. EMFi is thin and flexible plastic film having a constant charge. The film converts the electrical energy to vibration and sound. Further development for reducing sound attenuation costs includes deeper study of noise phenomena in duct systems and new ventilation systems with integrated sound attenuation properties. Session A1-1, 12:00 h Paper number 233 (Contributed, Oral) A.I. Koffeman and A.J. Kerkers Lichtveld Buis & Partners BV, PObox 156, 3500 AD Utrecht, The Netherlands I-INCE Class 67 COST OPTIMAL REDUCTION OF NOISE IN LARGE INDUSTRIAL AREAS - A METHOD TO SELECT MEASURES It is always a big issue in noise reduction programs which noise levels in the neighbourhood can be achieved economically. Especially near large industrial areas it is seldom possible to comply with the 12 Internoise 2001 Abstracts preferred values at reasonable cost. This means that an acceptable noise level must be found and an optimization must take place to get the best ‘value for money’. A method, based on the decrease of the number of dwellings with a noise level above the preferred value was developed. The method leads to an order of individual noise reduction measures and uses the cost to reduce the noise level for one dwelling with 1 dB as a decision factor. The strength of the method is, that measures at different and far apart sources can be compared. It is also a big help for the decision makers in industry and at the authorities to be aware of the cost of reducing the noise level at a single dwelling with 1 dB. The method was applied successfully at several large industrial areas in the Netherlands. With a few alterations it is also applicable to design the optimal distribution of noise sources for new developments. Session A1-1, 13:20 h Paper number 616 (Contributed, Oral) Félippe Van Eekhout and Ferry Koopmans I-INCE Class 67 Adviesbureau Peutz & Associés B.V., PO box 66, NL-6585 ZH MOOK, The Netherlands THE PARADOX OF A COMPLEX VIBRATION-ISOLATED SYSTEM AND COST EFFICIENCY The compression of the molding process in a metal foundry usually involves vibration related problems in the civil area. Currently there are two common compression methods: hydraulic and impact compression. The former method is related with a highly expensive, conventional machine, which causes comparatively lower vibration levels. The latter is related with a much cheaper machine, which however causes high vibration levels, but on the other hand, has the capacity to produce more complex, smoother and more precisely products. The choice to be made is to choose between high and low costs, high and lower quality of the products, and between high and low vibration levels in the civil area. The key question is if a good quality product is possible at low costs and with low vibration levels. Therefore it is necessary to investigate the possibilities to reduce the vibration levels using a vibration isolated system, starting from a impact compression method. By means of measurements on similar machines, transfer measurements on the soil and finite element method (FEM) calculations, a complex isolated system can be designed and realized. In practice is shown that a complex isolated system can be very (cost) effective. POSTER Session A1-1 Paper number 42 (Contributed, Poster) Posterboard AP-1 Tuesday, Author present 14:00 – 15:00 h Ulrich Rückert and E.h. Gerhard Heimerl I-INCE Class 67 University of Stuttgart, Institute of Railway and Transportation Planning and Engineering, Pfaffenwaldring 7, 70569 Stuttgart, Germany ECONOMIC EVALUATION OF THE LOW-NOISE-TECHNOLOGY ON THE GERMAN RAILWAYNETWORK Noise emissions are often a serious problem for the railways which can reduce their acceptance among their neighbours decisively. Of particular interest is the reduction of night-time noise emissions which are predominantly caused by goods trains. A study has been carried out to evaluate if vehicle- and track-based measures (the Low-Noise-Technology (LNT) has proved to reduce emission levels by 20 Internoise 2001 Abstracts 13 dB(A)) can play a major role in future noise-abatement policies for new and upgraded lines instead of building high barriers or walls next to the track. Two lines with together 190 km track have been acoustically calculated with both methods. In the case of the LNT the reception points with exceeded reception levels reduce by 50 to 75 % compared with conventional measures, although barriers of up to 5,6 m height have been imposed. The investments and costs of both measures have been calculated and extrapolated to the concerning 3700 km part of the German railway network. The LNT can provide better noise protection and save infrastructure investments of about 700 million Euro and annual costs of about 70 million Euro. On the other side costs for the LNT freight cars have to be drawn into account which lead to a deficit of between 63 million Euro and 93 million Euro per year but significant improvements and cost degression can be achieved in the future. If you mention the net-wide noise-reducing effects of low noise vehicles, the way of reducing the emissions of freight cars is highly effective. Session A1-2: Cost-benefits of noise control Chaired by Jacques Lambert and Matti Vainio Session A1-2, 13:40 h Matti Vainio Paper number 545 (Invited, Oral) I-INCE Class 67 European Commission, Environment Directorate-General, Avenue de Beaulieu 9 , B-1160 Bruxelles, Belgium COMPARISON OF HEDONIC PRICE AND CONTINGENT VALUATION METHODS IN URBAN TRAFFIC NOISE CONTEXT In empirical economic literature, hedonic price method (HPM) has been used extensively to give a willingness-to-pay estimate (WTP) for noise reduction while contingent valuation method (CVM) has been used only recently. Unfortunately it has not usually been possible to compare the results. In this paper, the results of two studies will be reported: one applying HPM and the other CVM using the same population. The question is, are these different methods giving WTP estimates of the same order of magnitude? The studies were carried out in an urban environment (Helsinki, Finland) where a cross section sample of 1500 transactions of owner occupied apartments was used for the HPM study. As all apartment characteristics and prices were from one real estate agent, the quality of the data were very high. A sample of 700 these apartments was selected for the CVM study (mail survey). With a response rate of 60% the CVM results can be compared with those from HPM. In the CVM survey some light is also given to the issue what traffic externality people are most bothered with (noise, dust, visual intrusion etc.). Session A1-2, 14:00 h Paper number 68 (Contributed, Oral) Hand Nijland and Jan Jabben RIVM, P.O.B. 1, 3720 BA Bilthoven, The Netherlands I-INCE Class 67 EFFECTS, COSTS AND BENEFITS OF NOISE MEASURES IN THE NETHERLANDS This paper presents the results of a recent study on the effects, costs and benefits on a national scale of the implementation of a number of measures aimed at reducing noise emissions by trains and road traffic. In the Netherlands, it is the first time a cost/benefit analysis of a variety of noise abatement 14 Internoise 2001 Abstracts measures has been carried out using different methodologies. The results of the study have been used in the process of national policy making. A GIS system was used to calculate effects for 2030. Costs were derived from recent studies carried out by a number of acoustical consultancy firms, and benefits estimated by calculating the value of the effect (i.e. people not being annoyed by noise any more after implementation of the measures). Since the value attached to nature and recreational areas becoming less noisy was not calculated, the benefits (as presented here) represent an underestimation of the overall benefits. Two different methodologies, ‘hedonic pricing’, employing observed prices of dwellings and ‘contingent valuation’, employing the willingness to pay for the improvement of living conditions were used. Finally an analysis of the sensitivity of the results to underlying assumptions has been carried out. The main conclusions follow: - implementation of source measures will make a considerable contribution to noise abatement in the Netherlands; - the measures will be the most costeffective if implemented at all levels (national, regional and local); - the societal costs will be surpassed by the societal benefits; - use of different methodologies in estimating societal benefits has made no substantial difference in the outcome. Session A1-2, 14:20 h Paper number 223 (Contributed, Oral) Sylvie Guerrand and David De Almeida Direction de la Recherche et de la Technologie SNCF, France I-INCE Class 67 AN OPTIMISATION PROCESS FOR IMPROVING COST BENEFIT ANALYSIS SOFTWARE TOOLS The search for a trade-off between environmental noise legislation and the development of railway transport has led to the setting up of decision tools dedicated to the analysis of the benefits and costs induced by solutions for the reduction of noise of railway lines. So far these tools are able to study the impact of "scenarios" built with a set of solutions, for one line. Within the European project STAIRRS, the improvement of such a tool called Eurano is planned to provide a software able to work at a European scale as well as on one specific line scale. One of these improvements consists in building an optimisation process which can give automatically for one or several lines the best combination of solutions under several constraints. For instance the optimisation can answer a question of the type "what is the best combination of solutions to protect a maximum number of people at the lowest cost". To built this optimisation process, the data (track, traffic, population density, geography etc.) must be well structured. An appropriate algorithm for the scale of the problem must be chosen to achieve relevant calculation times. The data structure, the selected algorithm and first tests carried out with a representative set of data will be presented in the paper. Session A1-2, 14:40 h Paper number 33 (Contributed, Oral) Thomas Baumgartner, Thomas Mettier and Sabina Pfister UNS, ETH Center HAD, CH-8092 Zurich, Switzerland I-INCE Class 68 NOISE IN THE LCA OF FREIGHT TRANSPORT. A METHODOLOGICAL TRANSFER FROM ROAD TO RAIL AND RESULTS FROM THE ASSESSMENT OF TRANSPORTS Life Cycle Assessments (LCA) of products and services so far mostly ignore noise impacts even though noise is recognised as a significant environmental and health problem. Contradictions between the time and space indifferent perspective of LCA and the importance of concrete noise emission and reception situations for the severity of human health impact explain largely this situation. However, failure to account for noise impacts is a serious shortcoming in the comparative LCA of transport Internoise 2001 Abstracts 15 services. Mueller-Wenk (1999) has suggested a procedure to integrate noise emissions of freight transports on road into the Eco-Indicator'99, a damage-oriented LCA method, and calculated weighting factors for the Swiss situation. We have adapted this procedure for the four steps of fate, exposure, effect and damage analysis assess noise from freight transports on Swiss rail. We arrived at weighting factors of between 0.10 DALY (average short distance freight train over 1000 km, daytime) and .46 Daly (average long-distance freight train over 1000 km, night-time) where DALY represents "disability adjusted life years lost". Restating the weighting factors of Mueller-Wenk based on our method results in about 2.5E-4 DALY for 1000 km with truck or pick-up. Including noise in Eco-Indicator ‘99 increases the number of EI’99 points for each ton-kilometre (tkm) by around 30% for both road and rail. The relative advantage of freight transport by rail increases slightly when noise is included as an impact category. The noise balances of train and road do not differ much but react very much to variations in the load factors. This result might be biased against rail transport as the data reflects the situation in the early nineties and rail might have improved noise abatement to some extent since that time. Session A1-2, 15:00 h Paper number 221 (Contributed, Oral) I-INCE Class 68 Aude Lenders, Nancy Da Silva, Thomas Baumgartner and W. Hecq Centre for Economic and Social Studies on the Environment, ULB, Avenue Jeanne 44, CP 124, B-1050 Brussels, Belgium COST-BENEFIT ANALYSIS OF RAIL-NOISE MITIGATION PROGRAMMES AT EUROPEAN LEVEL: METHODOLOGICAL INNOVATIONS FROM EURANO TO STAIRRS EURANO (UIC 1999) evaluated the efficiency of noise reduction measures for two European freight corridors: (i) A maximum benefit is achieved with an investment of about Euro 60 000/km/year. (ii) Solutions with rolling stock improvement are optimal. The results were based on a benefit defined as the reduction in the number of persons above 60 dB(A). The costs were based on unit costs of measures, capitalised over 40 years. The STAIRRS project (2000-2002), a follow-up of EURANO, includes a cost-benefit analysis based on about 10 000 km of modelled rail lines in seven countries. Benefits are defined in terms of dB(A) experienced by all persons living in the rail corridors modelled. They are to be weighted by the number of persons in each year benefiting from a reduction as different measures provide different time profiles of noise reductions. We assume a preference of people for an early reduction in noise and therefore will experiment with a time weighting factor. The problem of different lifetimes of measures making up a noise mitigation programme will be approached in two ways: (i) Costs will be discounted over the lifetime of each measure in an investment programme and benefits will be weighted in function of the average lifetime of a programme; (ii) Costs will be modelled for each programme over a 50-year period, but assuming an indefinite assurance of noise reduction. Using different discounting factors for both approaches should clarify the uncertainty implied by these two approaches. 16 Internoise 2001 Abstracts Session A1-3: Road traffic noise Chaired by Jan Jabben and Judith Lang Session A1-3, 15:20 h Paper number 563 (Contributed, Oral) J. Jabben, C.J.M. Potma and W.J.R. Swart I-INCE Class 13 National Institute of Public Health and the Environment RIVM, Antonie van Leeuwenhoeklaan 9, P.O. Box 13720 BA BILTHOVEN, The Netherlands CONTINUOUS MONITORING OF NOISE EMISSION FROM ROADWAYS The advisability and practicability of monitoring transport noise by means of continuous measurements on sites nearby the sources was investigated. A permanent measurement location was set up at a site located along the A2 roadway (major northsouth road connection in the Netherlands) at the elevation of Breukelen (nearby Utrecht). The study aims at monitoring the future development of noise emission from roadways. The measured data was combined with traffic-volume and -velocity data to differentiate between the various vehicle categories and to determine noise emission levels independent of traffic volume. The paper will present results showing that continuous monitoring of noise levels on sites along the roadway will offer possibilities for keeping emission data up to date, checking model predictions and evaluating the effect of policies on noise emission issuing from roadways. Also options for monitoring noise from other sources will be explored in a general context. Session A1-3, 15:40 h Paper number 753 (Contributed, Oral) Francesco Borchi and Monica Carfagni I-INCE Class 13 Dipartimento di Meccanica e Tecnologie Industriali, Università di Firenze, Via di Santa Marta, 3, 50139 Firenze, Italy A HYBRID NUMERICAL-EXPERIMENTAL METHOD FOR MODELING SOUND EMISSIONS OF A MOVING VEHICLE There are many approaches for modeling a traffic source made of several vehicles, for instance it can be modeled by a linear source. The problem arises when you have to model a single moving vehicle characterized by an engine speed dependent sound power spectrum and a time varying position. This work lead to a method to deal with the described problem. The method consists of changing the continuos problem into a discrete one. Many point sources were displaced, one for each position assumed by the vehicle in different instants. For each instant, one source only was switched on and its sound power level was evaluated using both sound pressure measurements and the receiver/source transfer function. As a test site for the methodology we used the Mugello Gran Prix Italian Circuit. The case study was a Ferrari F1 which raced along the circuit in a trial performance. For each lap a time history of sound pressure spectrums was recorded, using several receivers near the circuit. The sound pressure spectrum was used to calibrate the sound power spectrum of the point source representing the model at each instant. At the end of the procedure we obtained the sound power spectrum of each point source, and we could simulate instant by instant the environmental noise caused by the F1 race car. Moreover the commercial software, tuned using the results of our measurements, was used to predict the noise in some different environmental situations. The described Internoise 2001 Abstracts 17 approach proved to be an efficient way for simulating the noise reduction solutions for Gran Prix circuit. Session A1-3, 16:00 h Paper number 12 (Contributed, Oral) Sayed A. Ali and A. Tamura I-INCE Class 52 Department of Architecture, Faculty of Engineering, Yokohama National University, 79-5 tokiwadai, Hodogaya-Ku, Yokohama city, 240-8501 Japan DIAGNOSTIC STUDY FOR TRAFFIC NOISE PROBLEMS IN GREATER CAIRO, EGYPT Noise pollution has become a common concern to all members of the society. The noise problem is complex and is related to the increase in industrialization, transportation and the larger number of vehicles and motor vehicles that are poured and added into the already overcrowded streets. Measurements taken in Greater Cairo city, Egypt for traffic noise levels indicates that noise levels in the city are higher than those set by international agencies to protect public health and welfare in residential areas. So that makes Greater Cairo city in front of Capitals from aspect of noise level and continuous periods. In our paper we have presented: (1) Factors that lead to traffic noise problems as the following: significant increase of volume of the city, populations, and numbers of cars, using horn, high speeds, bad behavior of drivers, sounds emit from engine and vehicle parts, bad paving of road, using bad tires, sound from road-tire interaction, bad planning of roads, no special lanes for service, bad relationship between pedestrians and traffic, busy platform, vibrations of bridges, and un-using abatement. (2) Study countermeasures taken, these contain the efforts of Egyptian Ministry of Environment, City Halls, Traffic Administration of the city, Recommendations of Symposium of Noise & Traffic problems in Greater Cairo 1988, and Recommendations of Report of Egyptian Academy of Scientific Research and Technology, and Egyptian National Research Center 1994. This to evaluate this situation, and put planning for future to make Greater Cairo quiet city. (3) Conclusions and remarks will be discussed. Session A1-3, 16:20 h Sigmund Olafsen Paper number 2 (Contributed, Oral) I-INCE Class 72 Hjellnes COWI AS, P.O.Box 91 Manglerud, N-0612 OSLO, Norway ACCURACY OF PRACTICAL ROAD TRAFFIC NOISE MEASUREMENTS There is a great need for simple and fast methods for the measurement of road traffic noise. Several factors affect the road traffic noise levels. In a given point, the level varies with the amount of traffic, the driving patterns of the vehicles and the weather conditions. A simple measurement method was presented at Transport Noise and Vibration’98 in Tallinn (Immission measurement of road traffic noise in a practical approach). Data of noise emission for individual vehicles can be found in an article accepted for publication in Intíl Journal of Vehicle Design (Strategies to reduce road traffic noise by making each vehicle quieter). This new paper will give updated information about the basis for the earlier method. Two approaches have been pursued in order to improve the knowledge of the accuracy of practical road traffic noise measurements: 1) Three series of one hour measurements of noise from the vehicle stream; 2) Several new series of measurements of emitted noise from each vehicle in a real traffic situation. The results clearly show that the number of vehicles during a measurement is not critical. Variations in driving pattern and in weather conditions are much more important. The paper will give guidelines as to expected accuracy of road traffic noise measurements as a function of 18 Internoise 2001 Abstracts measurement time, distance from the road and number of passing vehicles during the measurement session. Session A1-3, 16:40 h Paper number 317 (Contributed, Oral) Roland Sottek, Winfried Krebber and Klaus Genuit HEAD acoustics GmbH, Ebertstr. 30 a, 52134 Herzogenrath, Germany SIMULATION OF VEHICLE EXTERIOR NOISE I-INCE Class 76 Within the framework of the European project SVEN (Sound Quality of Vehicle Exterior Noise) tools and methods will be developed to improve the situation of traffic noise in cities and suburbs. In order to achieve this aim, the effects of vehicle noise on human hearing have to be explored. Suitable measurements and analyses are used for the objective description of vehicle exterior noises and traffic noises correlating with the subjective judgement and the physiological effects. This paper introduces a hybrid model for the calculation of binaural vehicle exterior noises. The model uses recordings of the main excitation signals (exhaust system, intake, engine, tires) and binaural filters describing the transfer paths from the sources to an artificial head standing at a distance of several meters outside the vehicle. The calculation method allows the simulation of any driving situation. A simulated pass-by situation is used as an example to show the effects of the partial sound events on the entire vehicle exterior noise. Session A1-3, 17:00 h Paper number 662 (Contributed, Oral) Hans-Holger Bartel, Frank Heidebrunn and Christian Popp Federal Environmental Agency, Berlin (Germany) I-INCE Class 76 ROAD TRAFFIC NOISE IMPACT MODEL FOR THE FEDERAL REPUBLIC OF GERMANY In Germany the Federal Government has to report one year after its first convene the state and the development of harmful environmental effects (such as air-pollution and noise). This report has to be given according to Article 61 of the Federal Immission Control Act. Concerning noise the report has to distinguish road traffic noise, railway noise, aircraft noise and noise from industrial and commercial plants. Basic figures for this report are the numbers of exposed citizens due to the noise sources and differentiated into noise level classes as well as into sizes of the towns and cities based on a calculation model for the whole German territory. A new calculation model was created to satisfy the changed requirements in consequence of Germany´s reunification. In addition this model is able to look back until the year 1980 and it can make predictions based on a lot of different input parameters. Furthermore the model offers possibilities to adjust calculation methods (e.g. suggested by EU-WG 3), assessment periods (e.g. as suggested by EU-WG 1) and changes in the emission data. The model also enables its user to calculate and compare different scenarios depending on political decisions and technical developments. The mentioned characteristics of the model needed to develop a computer program with a high input on the statistical level and relatively complicated calculation algorithms as well as interface-systems to connect emission and propagation modules. For the time being the model works for road traffic noise. The part for railway noise is in preparation. The presentation will put emphasis on the road traffic model and the herewith produced results. Internoise 2001 Abstracts 19 POSTERS Session A1-3, 17:20 h Paper number 164 (Contributed, Poster) Posterboard BP-1 Wednesday , Author present 9:40 – 10:40 h Joël Lelong Inrets - Laboratoire Transport-Environnement. F - 69675 Bron Cedex, France PASSENGER CARS. POWER UNIT AND TYRE-ROAD NOISE, DRIVING BEHAVIOUR: WHAT ARE THE STAKES? I-INCE Class 13 The noise emission values for single vehicles defined in the French "Guide du Bruit des Transports Terrestres" result from measurements and data analyses lead between 1973 and 1977. To update these values (which has been maintained in prediction software used today), passby noise of passenger cars have been measured on a test track under various running conditions (acceleration /deceleration, constant speed). After a brief description of the experimental disposition, this paper presents the results obtained with various-motorized vehicles (gas, diesel, hybrid and electric). At low speed (< 50 kmph), it appears that the influence of the motorization on emitted noise can reach 7 dB. The rate of acceleration - directly related to the driving behaviour -is another important parameter greatly affecting the emitted noise. At high speed, for the road surface of which the test track is fitted, the characteristics of the tyres are not so significant: their influence doesn't exceed 3 dB. Session A1-3, 17:20 h Paper number 348 (Contributed, Poster) Posterboard BP-2 Wednesday, Author present 9:40 – 10:40 h Pedro E. Solana, Miguel A. Picard and Juan V. Arizo I-INCE Class 52 Depart. de Física Aplicada, High Technical School of Industrial Engineering. Polytechnic University of Valencia, Camino de Vera, 14. D5-B. 46022. Valencia. SPAIN NOISE DISTURBANCE IN THE URBAN SITES OF HIGH DENSITY The usual way to evaluate the grade of nuisance by the noise level in external population areas it is the elaboration of surveys and the measurement of the noise levels. In this work, we have settled down a relationship between the equivalent level and four grades of bother. On the other hand, the measurement of the existent noise, when this oscillates along the day, we require measurement during most of day. Small time intervals, in the order of ten minutes, carried out in every hour, could be misleading. They are graphs enclosed to show the daily equivalent level between 8-20 hours in function of the level observed in a particular hour of the day. In order to find relationships between the daily equivalent level and the equivalent measured level in periods of less time is being investigated. By effecting short measurements in a certain schedule sector of the day the equivalent daily level and the grade of perceived bother could be induced. In the following stage, by the use of models the noise level could be evaluated by means of the traffic and establish its acoustic incidence. This study is important for the elaboration or revision of acoustic maps, decreasing the costs. Implemented the model in a computer, the acoustic map and the annoyance, we could set down almost in real starting time from the traffic measurements of and the measured noise by placing sound record receivers. 20 Internoise 2001 Abstracts Session A1-4: Measurement of road traffic noise Chaired by Dick Schumacher Session A1-4, 17:40 h Paper number 676 (Invited, Oral) R. Le Salver, U. Tornar and E. Vivier PSA PEUGEOT - CITROËN, F92250 LA GARENNE COLOMBES, France ACEA PROPOSAL FOR A NEW METHOD TO MEASURE THE NOISE EMITTED BY A CAR IN URBAN DRIVING CONDITIONS I-INCE Class 72 Based on statistics coming from measurements of the driving conditions for 39 different cars, in 8 different cities of Europe and Japan, the ACEA proposal for a new method to measure the noise emitted by a vehicle, in urban driving conditions is based on the simulation on the ISO test track of: - vehicle speed 50km/h in front of the microphones; - acceleration of the vehicle: 90th percentile of the maximum acceleration during the acceleration phases in urban driving conditions; - partial load factor of the engine: ratio between the 90th percentile of the used acceleration and the full throttle acceleration of the vehicle at the engine speed used during the maximum acceleration in urban driving conditions. The noise measured in these conditions is higher than the 95th percentile of the noise emitted during the driving time. The explanation of the proposal method and its technical background are the content of our proposed presentation. Session A1-4, 18:00 h Heinz Steven Paper number 565 (Invited, Oral) I-INCE Class 81 TÜV Automotive, Herzogenrath (Germany) IMPROVEMENT OF THE TYPE APPROVAL NOISE LIMITATION FOR MOTOR VEHICLES The noise emission from road traffic causes one of the most important ecological damage in the European Union. Since 1972 the limitation of the noise emission of motor vehicles is included in the type approval legislation in the EU. Although the noise limits have been reduced significantly several times till today, the noise impact of the population has been increased. This is - besides the enormous increase of the traffic load in the streets – due to the fact, that the effectiveness of the type approval test method with respect to the noise emissions in real life traffic is rather poor. About two years ago a work item on improving the test method was started within ISO. This item is also discussed in the ECE working party on noise (GRB) and a working group on "Road Traffic Noise" of the EUcommission. Several proposals from EU member states as well as from the vehicle manufacturers have been worked out. The present status of the discussion and the effectiveness of a further reduction of noise limits for the noise emission reduction in real traffic situations will be reported. Internoise 2001 Abstracts 21 Session A1-4, 18:20 h D.F. De Graaff Paper number 590 (Invited, Oral) I-INCE Class 81 M+P Consulting Engineers, The Netherlands A SPEED AND ACCELERATION LIMIT IN THE NOISE TYPE APPROVAL OF VEHICLES WILL ENABLE SILENT CARS TO REVEAL THEIR SILENCE In order to lower traffic noise emission more adequate, a more representative type approval measurement method and a set of more effective limit values are needed. For this purpose many researchers have investigated the driving behaviour (vehicle speed, acceleration etceteras) of vehicles in urban traffic. In the underlying study not only the driving behaviour of 15 different vehicles (ranging from 0.6 to 18 tons), but also the noise emission, and even the resulting noise immission on the adjacent dwellings is determined. Furthermore the noise emission is broken down into its three main noise sources: engine, exhaust and tyre/road. The resulting d-Base opens a wealth of additional knowledge and insights. Amongst others: 1. The difference in noise emission between the vehicles is very much speed dependent. It ranges from 30 dB(A) at 10 km/h to 6dB(A) at 50 km/h; 2. In 50 km/h steady flowing traffic, tyre/road noise is the dominant noise source for almost all vehicles; 3. At 30 km/h medium acceleration (1 m/s2 typical), the dominant noise source various from tyre/road for some passenger cars, to exhaust for a sports cars and engine for the delivery vans and trucks; 4. The relevance of these two driving circumstances for the noise immission various per vehicle, but is typically in the order of 2/3 to 1/3; 5. Proposals for a new type approval measurement method should reckon with this speed dependent difference. Silent vehicles will only reveal their silence at low speed; 6. This study ones again shows the trinity of traffic noise abatement: vehicles, tyres and road-surfaces. POSTER Session A1-4 Paper number 274 (Contributed, Poster) I-INCE Class 56 Posterboard BP-3 Wednesday, Author present 9:40 – 10:40 h Juan M. Barrigón, Valentín Gómez, Juan A. Méndez and Rosendo Vílchez Universidad de Extremadura / Escuela Politécnica / Departamento de Física, Avda. De la Universidad s/n Cáceres 10071 (SPAIN) COMPARATIVE STUDY OF THE CITY OF CACERES (SPAIN) IN NORMAL AND TRAFFIC REDUCTION CONDITIONS Ciceres, a city in the middle west part of Spain (declared Patrimony of the Humanity by the UNESCO) has a irregular urbanistic distribution with a high proportion of narrow streets and avenues. This implies high levels of noise due to the traffic. By a categorisation procedure of the streets, noise levels (Leq, Lmax, Lmin, L10, L50 and L90) were measured in normal traffic conditions, at different hours in working days. With the celebration of the day without car, all the above mentioned levels were again measured in the coincidental points with the normal conditions. Frequency distributions for Leq, Lmax and Lmin in 1/3 octave bands were obtained in the day without car. In both situations traffic flows were counted. A decrease of about 50% of vehicles running in the day without car was detected, and as a average from the whole set of measurements, 3 dBA less for the energy-averaged sound level. Some comments are added relatives to the result of the day without car. 22 Internoise 2001 Abstracts Session A2-1: Active noise control in ducts Chaired by Gérard Mangiante and J. Scheuren Session A2-1, 9:40 h Gérard Mangiante Paper number 13 (Tutorial, Oral) I-INCE Class 37 Laboratoire de Mécanique et d'Acoustique, Centre National de la Recherche Scientifique, 31 Chemin Joseph-Aiguier, 13402 Marseille Cedex 20 (France) THE MECHANISM OF ACTIVE SOUND ABSORPTION IN A DUCT: EQUIVALENT CIRCUIT ANALYSIS Active noise control in a duct can be obtained according to two concepts: (i) reflecting the primary noise upstream in the duct by the use of a single monopole secondary source acting as a barrier, (ii) absorbing the noise by the use of a tripole secondary source working as a "sucker". With this latter method the recurrent question is: where does the power go? In this paper, the analogies between plane sound waves and electromagnetic waves along a lossless bifilar line are used to analyze the mechanism of sound absorption by a secondary source. The paper is outlined as follows: Kirchhoff's network model of an active noise control system in a duct with reflections is introduced in Sec. I. With this model the system is represented by an equivalent circuit including both generators corresponding to primary sources, and controlled generators, corresponding to secondary sources. The energy balance equations for the main types of secondary sources are given and the cases where the secondary source absorbs acoustic power are investigated. Section II considers this power absorption for secondary sources obtained with loudspeakers. A gyrator network is used to represent the coupling equations and an equivalent electromechanical circuit of an active system is given. From this circuit, the equations permitting the analysis of the power radiation or absorption are established and discussed. The last section describes our conclusions. Session A2-1, 10:20 h Paper number 22 (Invited, Oral) Mingsian R. Bai, Yuanpei Lin and Jienwen Lai National Chiao-Tung University, Taiwan I-INCE Class 37 MULTIRATE SIGNAL PROCESSING APPROACH FOR REDUCTION OF DELAY IN SPATIALLY FEEDFORWARD ANC SYSTEMS Electronic delay has been a critical problem in active noise control (ANC) systems. In particular, excessive delays would cause severe causality problems in a spatially feedforward duct ANC system with critical space limitation. This paper suggests a multirate approach for minimizing the electronic delay in the control loop. In this approach, analog filters widely used in conventional digital controllers are waived by direct decimation and interpolation of discrete-time signals. The computation efficiency is further enhanced by a polyphase method, where the phases of lowpass finite impulse response (FIR) filters must be carefully designed to avoid unnecessary delays. Frequency domain optimization procedures based on norms are utilized in the FIR filter design. The proposed method was implemented by using a floating-point digital signal processor (DSP). Experimental results showed that the multirate approach is effective for suppressing a broadband random noise in a spatially feedforward duct ANC system, with length of only 30 cm. Internoise 2001 Abstracts 23 Session A2-1, 10:40 h Paper number 188 (Invited, Oral) René Boonen and Paul Sas KU Leuven, B-3001 HEVERLEE, Belgium I-INCE Class 37 MODIFIED SMITH COMPENSATION FOR FEEDBACK ACTIVE NOISE CONTROL IN DUCTS In active noise control, the performance of the popular adaptive feedforward controllers suffer from their low convergence speed, during alteration of the operational conditions of the acoustic source. Feedback controllers can react almost immediately on alteration of the acoustic regime. However, the high number of acoustic resonances and, more important, the acoustic time delay restricts the control bandwidth considerably. The design of feedback controllers is in most cases very critical. In this paper, control bandwidth is improved by applying a time delay compensator. The proposed compensator is based on the Smith-compensator. Normally, the Smith-compensator requires a model of the acoustical plant. The plant model is reduced to the actuator dynamics, which is easily identified and remains almost constant, even in different acoustical circumstances. The improvement of the control bandwidth can be approximately one decade in frequency, resulting in a considerable higher noise reduction within the control bandwidth. Session A2-1, 11:00 h Seppo I. Uosukainen Paper number 334 (Invited, Oral) I-INCE Class 37 VTT Building and Transport, Air Handling Technology and Acoustics, P.O. Box 1804, FIN – 02044 VTT, Finland UNIDIRECTIONAL JMC ACTUATORS IN THE ACTIVE ATTENUATION OF NOISE IN DUCTS WITH UNIFORM AXIAL FLOW The JMC actuators need a monopole and a dipole radiator. The planar monopole and dipole source types and their source strengths are defined in the case of uniform flow. Further, the source strengths needed when the JMC method is applied with the uniform flow present are discovered. Both the definitions of the source strengths and their actual values needed in the JMC method depend on the Mach number of the flow. In feedforward active noise control systems in ducts, a natural solution to avoid the acoustic feedback from the cancelling actuator to the reference detector is to use a unidirectional actuator. Ideal unidirectional three- and two-element source configurations, based on the JMC method, are introduced in the no-flow case and in the case of uniform axial flow. In the twoelement actuator, the elements act as a dipole with the monopole effect included. The well-known unidirectional Swinbanks' solution always needs an inter-channel delay corresponding to the acoustic propagation time between the elements of the actuator. With the axial flow present, the delay further depends on the Mach number of the flow. The JMC based solutions can be realized with no interchannel delays, even in the case of flow, which is advantageous in the practical implementation of digital control systems. 24 Internoise 2001 Abstracts Session A2-1, 11:20 h Paper number 386 (Invited, Oral) Vesa Välimäki, Jari Kataja, Marko Antila I-INCE Class 37 Helsinki University of Technology, Lab. of Acoustics and Audio Signal Processing, P. O. Box 3000, FIN–02015 HUT, Espoo, Finland UNIDIRECTIONAL SOLUTION FOR ACTIVE NOISE CONTROL IN DUCTS This paper reviews our recent work on active noise control (ANC) in an air-conditioning ducts using unidirectional detectors and actuators. The system is a feedforward FIR controller where we have investigated various signal processing methods and configurations using a pair of reference microphones and secondary sources. The paper first discusses a microphone pair as a unidirectional reference detector. Its main goal is to reduce the feedback from the secondary source to the reference signal, which is one of the major problems in feedforward ANC systems. Each of the two microphone signals is filtered using one of many appropriate techniques before the two signals are combined. The idea is to amplify the desired reference signal, which is the downstream going wave in the duct, while at the same time to cancel the anti-noise signal, which propagates upstream from the secondary source. There are various ways to accomplish this. The simplest configurations employ short delay lines and inversion, while the more sophisticated ones use optimized FIR or IIR filters for each signal branch. The overall delay caused by filtering must be kept at minimum not to risk the causality of the ANC system. Unidirectional secondary sources are also examined. In addition to the reciprocal detector structures, adaptive unidirectional source configurations are presented, which can automatically learn the properties of the loudspeakers and the acoustic field in the duct. Finally, experimental results obtained with a real-time demonstration ANC system are presented, which show the advantages of the use of unidirectional detectors and actuators. Session A2-1, 11:40 h Paper number 460 (Invited, Oral) Marko Antila, Jari Kataja and Veli-Jukka Ollikainen VTT Automation, Acoustics Engineering, Tampere, Finland RING SENSORS FOR PROPAGATING WAVE DETECTION IN DUCTS I-INCE Class 37 Acoustic sensors are an important part in an active noise control system in ducts. Flush mounted or wire suspended pressure microphones are commonly used as sensors to both detect the primary sound field and to measure the remaining error signal. Such sensors detect the sound pressure at a single point. Large area sensors or sensor arrays are beneficial for detecting the propagating plane wave. A way to construct large area sensors is to use pressure sensitive Electromechanical Film (EMFi). These sensors can be attached as rings to the inner wall of a duct. With this arrangement, a sensor integrates the sound field at a short section of the duct and can be made less sensitive to turbulent airflow disturbance. To gain further insight to the operation of the sensors, we modelled and simulated the propagation of sound in a semi-infinite duct and the influence on the sensors. If the diameter of the duct is small enough compared to the length of the duct, the sound can be assumed to propagate as plane wave. This assumption was studied with the simulations. For experiments, we constructed circular sensors using Electromechanical Film (EMFi) and attached them to the inner wall of a ventilation duct. The dimensions of the duct were 6 metres in length and 49 centimetres in diameter. A loudspeaker was installed at one end of the duct and the other end was covered with absorptive material to make the duct acoustically semi-infinite. The operation of the sensors was verified by the measurements. Internoise 2001 Abstracts 25 Session A2-1, 12:00 h Shiro Ise Paper number 684 (Invited, Oral) I-INCE Class 37 Faculty of Engineering, University of Kyoto, Yoshida-honmachi, Sakyo-ku, Kyoto 606-8501, JAPAN ACTIVE NOISE CONTROL FOR HIGH ORDER MODES IN A DUCT BASED ON THE BOUNDARY SURFACE CONTROL PRINCIPLE Based on the boundary surface control (BSC) principle, by designing a inverse system which control the pressure and the particle velocity on a entire surface enclosing a volume, we can perfectly control the sound field within the volume. In the case of applying this principle to the active noise control, we can perfectly control within the volume by controlling the entire surface so that the pressure and the particle velocity are zero. Because it is practically impossible to control the entire surface, the surface should be divided into small elements. That means if we locate error sensor on each element and design the system so that the output signals of the error sensors are zero, we can control the volume enclosed by the surface where the error sensors are located. In the case of the ANC system in a duct, we do not need to use so much error sensors but we locate error sensors at the each element in a cross section of the duct. In this paper, in order to realize the ANC system in a duct system based on the BSC principle and control noise at high order mode, we study the condition of the error sensors numerically and experimentally. Session A2-1, 13:20 h Paper number 428 (Contributed, Oral) Ondrej Jiricek and Marek Brothanek CTU-FEE, Dept. of Physics, 166 27Prague 6, Czech Republic I-INCE Class 37 APPLICATION OF MULTI-CHANNEL FEED-FORWARD ACTIVE CONTROL IN A DUCT The paper deals with active attenuation of fan noise in a duct. If the cross-section of the duct is large enough, higher modes have to be assumed in the frequency range of ANC, Multi-channel system has to be assumed. Algorithms and arrangements were first simulated by means of a numerical model and then tested on an experimental duct. Experiments performed done on a duct of rectangular crosssection of dimensions similar to actual ventilating or air-conditioning ducts. As the cut-off frequency was about 500 Hz, the first cross-mode was assumed and a two-channel ANC system was used. An axial flow fan with five blades and maximum volumetric flow rate up to 1.5 cubic meters per second was used. Flow speed was regulated by regulation of fan revolutions. The digital control unit was based on Texas Instrument floating point DSP. Multi-channel feed-forward algorithm with passive feedback compensation was implemented. Satisfactory attenuation in frequency band from 20 Hz to 700 Hz was achieved. Session A2-1, 13:40 h Paper number 184 (Contributed, Oral) I-INCE Class 38 Seiji Watanabe, Kensaku Yanagimoto, Takahiro Ito, Shuichi Sakamoto, Ryohichi Ichimiya and Katsuya Tanifuji Tsuruoka National College of Technology, Sawada104 Inooka Tsuruoka-City, YAMAGATA, Japan ACTIVE NOISE CONTROL USING SLIT RESONATORS IN CYLINDRICAL DUCT EDGE Purpose of this study is clarifying the sound insulation characteristics of the circular slit resonator with additional active noise sources. The hybrid system which consists of the circular slit resonator with loudspeakers and the active noise control technique(ANC) is expected to improve the sound 26 Internoise 2001 Abstracts attenuation performance of resonator. This paper describes sound insulation characteristics of circular slit resonator that combined ANC system for silencing leakage noise at ventilation aperture of resonator. The circular slit resonator composed of thin air layer that is radial expansion with very short axial length was set up machinery wall. The circular thin air layer is driven by four loudspeakers that are controlled using Filtered X-LMS algorithm to minimize sound pressure at outside error microphone. The primary noise source is supplied in the sine wave, and the frequency of the sine signal is changed from 50[Hz] to 5[kHz] every 50[Hz]. The experimental results show that this hybrid system has two noise absorption regions. There are 30 to 45[dB] noise reductions in the frequency range 150[Hz] to 900[Hz]. By combining ANC with the circular slit resonator, attenuation performance at resonant frequency in the high range 900[Hz] to 5[kHz] are increased greatly. Session A2-1, 14:00 h Paper number 641 (Contributed, Oral) P.R. Fraanje, M. Verhaegen and N.J. Doelman I-INCE Class 76 University of Twente, Faculty of Applied Physics, Systems and Control group, P.O. Box 217, 7500 AE ENSCHEDE, The Netherlands SUBSPACE IDENTIFICATION FOR ACTIVE NOISE CONTROL IN A DUCT In Active Noise Control (ANC) problems the filtered-X LMS algorithm and its variants are widely used. In these algorithms the (feed forward) controller has a Finite Impulse Response (FIR) structure. However, in broadband disturbance cancellation problems FIR controllers will be of very large order. For broadband disturbance cancellation the filtered-U LMS algorithm was proposed, in which the controller has an Infinite Impulse Response (IIR) structure. These sample-by-sample update methods as well as their block update variants have some important drawbacks, like slow convergence, step size choice, choice of optimal orders, and IIR controllers might become unstable. Recently, we proposed the Blockwise Subspace Identification for ANC (SIANC) method [Fraanje,2000] to overcome these problems. In the method, a state-space controller is estimated and updated using a block of N samples, based on the method of subspace model identification. The method provides an estimate of the order of the controller and an accurate estimate of the controller which minimizes the residual sound (though, it does not explicitly minimize the variance of the residual signal). Simulation experiments with this method on a realistic duct problem showed considerable increase of the convergence speed compared with the block filtered-U LMS algorithm. The paper presents the comparison of the SIANC method and the block filtered-U LMS algorithm applied on a real-life acoustical duct. Internoise 2001 Abstracts 27 POSTERS Session A2-1, 14:20 h Paper number 259 (Contributed, Poster) Posterboard AP-2 Tuesday, Author present 15:00 – 16:00 h J. Hernandez, L. Tardy, E. Gonneau and J.P. Guilhot LAMI-Universite Paul Sabatier, 31062 Toulouse Cedex, France MODAL ACTIVE NOISE CONTROL IN DUCTS: IMPLEMENTATION OF DIFFERENT ALGORITHMS I-INCE Class 37 Paul Lueg invented the Active Noise Control in 30's. Unwanted acoustic waves produced by a primary source are cancelled by anti-phase waves given by a secondary source: this is the superposition principle. Since 80's, active methods have developed with the Digital Signal Processors which allow real time treatment. We used feedforward control strategy in rectangular and cylindrical duct cases. In the rectangular case, we controlled the plane mode by using a single input - single output system and by implementing the FxLMS algorithm. We proposed a version of this algorithm with a variable step size because of the duct closing conditions. The reductions were improved of 10 dB. We also realized a feedback compensation by recursive control. In the cylindrical case, we have been interested in the reduction of the noise produced by an air fan for particular frequencies to realize a narrowband modal control. We decided to control the fundamental and the first harmonic for the two principally present modes: (0,0) plane mode, (1,0) mode which has been decomposed in two orthogonal (1,0)+ and (1,0)- components. The Fxlms algorithm has been implemented in its classical and notch versions by using Independent Modal Control and Fully Coupled Control strategies. We could only realize IMC control in real time because of computational time. We obtained reduction of 14 dB ((0,0) mode), reductions of 17 dB and 10 dB ((1,0) mode for the fundamental and the first harmonic).We decided to implement fast versions and other algorithms as RLS to improve our results. Session A2-1, 14:20 h Paper number 470 (Invited, Poster) Posterboard AP-3 Tuesday, Author present 15:00 – 16:00 h E. Gonneau LAMI Universite Paul Sabatier 118 route de Narbonne 31062 Toulouse France I-INCE Class 37 AN APPROACH TO OPTIMIZE AND IMPROVE ACTIVE NOISE CONTROL IN DUCTS Due to recent developments in adaptive signal processing and digital signal processing hardware, active noise control (ANC) is nowadays a growing concern for many researchers. The paper presented here discusses with single and multiple channel ANC, both for controlling the noise field in smalldimensions ducts. First, in the case of single channel ANC, i.e. for non-modal control, we develop a theoretical and numerical method to implement a variable step size filtered x-LMS on a fixed point DSP. The step size is controlled by the variance of the filtered reference signal. The purpose is to improve convergence of the conventional filtered x-LMS which does not always converge in an acceptable manner, particularly when the magnitude of the filtered reference signal is small. Effectiveness of the method developed here is demonstrated by numerical simulations for a simplified ANC system and by real time experiments made on the duct, and specially when closing conditions are different. Then, we consider ANC of sound radiated by axial fans in ducts. The modal controller and its optimization are presented. We are interested with the comparison of independent modal 28 Internoise 2001 Abstracts control (IMC) and coupled modal control (CMC) efficiency, both simulated from transfer functions and reference signal, measured on the real system. The (0,0) mode and the (1,0) mode have been controlled at fundamental frequency and its first harmonic. Because of its computational cost, only coupled modal control has been implemented in real time with classical FXLMS algorithm and an adaptive notch filter. Results are also presented with the use of a variable stepsize algorithm. Session A2-2: Active noise control in enclosures Chaired by Leonardo Lecce Session A2-2, 14:40 h Paper number 619 (Invited, Oral) Leonardo Lecce, Massimo Viscard and Francesco Di Maso I-INCE Class 38 Dipartimento Progettazione Aeronautica, Università di Napoli “Federico II”, Via Claudio, 21 80125 Napoli – Italy ACTIVE NOISE CONTROL WITHIN THE GENERATING/PUMPING GROUPS OF THE HYDROELECTRIC PLANT OF PRESENZANO (ITALY) This work is related to the implementation of an active noise control system to reduce the hydraulic turbomachine's noise within the four generating/pumping groups (250 MW each) of the Presenzano power plant. During a preliminary stage of the activity it were in fact measured a very high overall noise levels especially at the turbine "zero" floor level. It was also verified that noise spectra presented a typical profile with a set of high frequencies components and moreover two very shaped low frequencies components (pure tones at 150 Hz and 200 Hz that represent the third and fourth harmonics of the turbine’s BPF). At these last frequencies, passive noise control systems are poor effective in front of relevant costs and an active approach was so decided to be tested and taking into account the noise genesis and peculiarities a pure acoustic active control architecture, microphones/loudspeakers based, was chosen to be assessed. A set of actions were so planned in the aim to assess such a system, involving theoretical, numerical and experimental aspects. With special reference to this last aspect a dedicated, integrated system, including acoustic sensors with relative conditioning system, low frequencies designed loudspeakers with relative power amplifiers and a12in/12outdigital controller(DSP based) implementing a FIR based XLMS control algorithm, was built up and tested. In accordance with a "classical" control aimed procedure, a first noise field characterisation was performed by the use of a 32,microphones grid, all around the zero floor level. After that, 20 loudspeakers were positioned all around the target area and relative FRF (microphone's related) were measured and successively used in an optimisation process (genetic algorithm's based) aimed at sensors/actuators configuration assessment. A set of configuration were so determined and expected attenuation calculated and compared to be successively experimentally tested. Two main results appeared from the experimental activities: by one side, the very good agreement between the numerical expected data and the experimental ones that represented an encouraging confirmation of the correct procedure and algorithms implementation , on the other side the opportunity effort the problem with a sub-set "silence area" system approach more than with a global one. A mean reduction of 15 dB at 150 Hz and 7.5 dB at 200 Hz was in fact measured at the error sensors but, due to environmental acoustic condition, it was not enough spatially distributed as operative conditions Internoise 2001 Abstracts 29 should require. A more in depth investigation was so decided to be performed and is actually on course whose results seems to be very encouraging in the aforementioned direction. Session A2-2, 15:00 h Paper number 450 (Invited, Oral) I-INCE Class 38 Hannu Nykänen, Marko Antila, Jukka Lekkala, Roberto Aimasso, Paolo Delzanno and Görgün Necati VTT Automation, PO Box 1307, FIN-33101 Tampere, Finland ACTIVE NOISE CONTROL IN CARS BASED ON UTILISING ELECTROMECHANICAL FILM (EMFI) BASED ACTUATORS In the European Union funded FACTS project the development of Active Noise Control (ANC) systems for vehicles utilising wide surface area acoustic actuators based on ElectroMechanical Film (EMFi) technogoly was started. In the paper will be described the experiences of the project concerning the cars. Two car demonstrations were built, one for the "traditional" active engine order noise control and one for the attenuation of broad band noise radiation of the dash panel, caused by the vibration of the dash panel. The results show that the noise reduction using membrane mode EMFi actuators is fully effective at higher engine orders (the noise peaks are reduced to background noise level), when the frequencies are above 200 Hz and the noise levels are reasonable (below 80 dB). Also the broad band noise radiation caused by the dash panel vibrations could be attenuated some 5 dB in the frequency range 150 - 300 Hz, but some amplification was present at higher frequencies. Session A2-2, 15:20 h Paper number 131 (Invited, Oral) I-INCE Class 38 Sven Johansson, Sven Nordebo, Per Sjösten, Mathias Winberg and Ingvar Claesson Department of Telecommunications and Signal Processing, Blekinge Institute of Technology S-372 25 Ronneby, Sweden ACTIVE CONTROL OF SOUND USING THE ACTUATOR-INDIVIDUAL NORMALIZED FILTERED-X LMS ALGORITHM The control algorithm is a central part of an active noise control system. The computational power of the DSP hardware is limited, and therefore algorithms with high computational complexity allow fewer control sources and sensors to be used, often with reduced noise attenuation as a result. In applications where noise within large volumes is to be reduced, such as in aircraft cabins, a large multi-channel system is needed to control the relatively complex interior sound field. Here it is of great importance to keep down the computational complexity of the control algorithm so that a large number of control sources and sensors can be used, and also a large number of frequency components can be controlled. This paper presents results from practical experiments using an ANC system based on the actuator-individual normalized Filtered-X LMS algorithm. The algorithm combines low computational complexity with high performance. For that reason the algorithm is suitable for use in large multi-channel systems in order to reduce the computational power required by the control system. 30 Internoise 2001 Abstracts Session A2-2, 15:40 h Paper number 426 (Contributed, Oral) Min-Chun Pan,Yung-Chang Lin and Ming-Chang Jeng DESIGN AND FABRICATION OF ACTIVE ROCKET-FIRING NOISE REDUCER I-INCE Class 38 Department of Mechanical Engineering, National Central University, Chung-Li, Taiwan 320, ROC In the study, the approach of active noise control (ANC) is used to minimize the noise up to 140 dB in the control room, which results from rocket firing. This problem dealing with the overwhelming noise can be categorized as being three dimensional in space, transient, non-stationary and relatively broadband in noise characteristics. A noise reducer has been designed and fabricated to actively reduce the transient and non-stationary noise during continual rocket firing. To tackle this problem, firstly, two sets of recorded rocket-firing noise have been analysed to investigate their characteristics. The sound pressure level of firing noise lasts from 140 dB high to 110 dB (or 120 dBA to 90 dBA) in 0.9 sec duration. Most of the noise energy stays in-between 50 Hz and 350 Hz after A-weighting. Secondly, two anti-noise control algorithms were considered, i.e., adaptive least-mean-square method (adaptive LMS) and LMS methods, and their effectiveness was compared. The transfer function of the ANC system has been well identified using both time-domain and frequency-domain methods. All experiments were conducted in a full-scale model of the control room, which was built to imitate the sound field. In the current phase, a single-input-single-output active noise reducer was designed and fabricated. To assess the validation of this noise reducer, varying pure-tone and varying multiple-tone were tested during the implementation stage. 20-dB noise reduction can be achieved. Those noises can be cancelled effectively near the working personnel’s ears in the control room. As to more realistic assessment, 12-dB reduction can be yielded for real continual rocket-firing noise. Session A2-2, 16:00 h Paper number 640 (Contributed, Oral) G. Nijsse, M. Verhaegen and N.J. Doelman I-INCE Class 76 University of Twente, Faculty of Applied Physics, Systems and Control Group, P.O. Box 217, 7500 AE ENSCHEDE, The Netherlands STATE SPACE MODELING OF ACOUSTICS FOR ACTIVE CONTROL In Active acoustical control (AAC) applications, like active noise control (ANC) and sound reproduction systems, mathematical models of the plant to be controlled are of fundamental importance for the performance of the controller. Commonly in AAC, finite impulse response models (FIRMs) are adopted i.e. both the plants (e.g. in ANC, the secondary path and the intrinsic feedback path), and the controller are represented by FIRMs. However, a problem with FIRMs is that they can require a large number of coefficients, increasing the computational complexity of AAC applications. This is especially pronounced in poorly damped plants which require many coefficients in the FIRM for a proper representation of the plant. In the current paper, we adopt a hybrid modeling approach. We propose to model the plants in AAC by state space models (SSMs), and to represent the controller by a FIRM. By using this hybrid approach we are able to reduce the computational burden significantly. Another benefit of the use of SMMs is that they pave the way to a broad spectrum of tools such as the use of modern control theory for designing reliable (feedback) controllers. One of these tools is the SSM inner/outer factorization, which we adopt in the current paper. We demonstrate by simulations on a realistic acoustical channel, that by using a SSM inner/outer factorization of the secondary path, we are able firstly to reduce the number of computations compared to FIRM Internoise 2001 Abstracts 31 inner/outer factorization and secondly we are able to speed up the convergence. Also, we generalize the results from [Elliott, 2000] to systems with an arbitrary number of inputs and outputs. POSTER Session A2-2 Paper number 768 (Contributed, Poster) Posterboard AP-4 Tuesday, Author present 16:40 – 17:40 h J. Romeu, S. Jiménez and R. Capdevila LEAM, Politechnical University of Catalonia, Colom 11, Terrassa 08222 Spain FORCED ACOUSTIC UNLOADING STRATEGY IN ACTIVE NOISE CONTROL I-INCE Class 38 To achieve global attenuation in enclosed sound fields using active noise control, especially in diffuse sound field conditions, the secondary source must be placed very close to the primary source. Otherwise, reductions in the error microphone can produce an increase in the space average pressure in the room. To avoid this undesirable effect, a strategy to cause "forced" acoustic unloading of the primary source is presented. This consists in placing the error microphone near the primary source, so that the acoustic pressure on its surface is reduced and its power output diminishes, providing a global attenuation of the enclosed sound field. This strategy was applied on electric motor, and better results than those with the error microphone far away from the sources were obtained, although results have been sensitive to the relative positions of the microphone and secondary source. Session A2-3: Active control of structural vibrations Chaired by Osman Tokhi and Jing Tian Session A2-3, 16:20 h Paper number 399 (Invited, Oral) T. Meurers and S.M. Veres School of Eng. Sciences, University of Southampton, SO17 1BJ, UK FSF BASED FEEDBACK CONTROL WITH ONLINE SECONDARY PATH ESTIMATION I-INCE Class 40 Active noise control is a fast growing area of interest. However, little research effort is aimed at producing efficient controller tuning algorithms. The filtered X-LMS algorithm is commonly used, which has two significant disadvantages; a reference signal and a model for the secondary path are required. The model is normally estimated off-line in advance and remains fixed for the control period. This could lead to instability in the case of changing dynamics. Moreover, the number of taps in the filter realising the controller is usually high, and low correlation between the reference and the disturbance results in poor performance. In this paper a new frequency selective filter (FSF) based controller is presented with online secondary plant estimation. Due to the constant update of the plant model, slow changes in the dynamics can be tracked and instability avoided. The controller operates in feedback and therefore no reference signal is needed. A bank of FSFs is selected so that they cover the spectrum of the disturbance frequencies. Parallel real-time computation for a large number of FSFs is 32 Internoise 2001 Abstracts feasible, which can increase efficiency. The cost of computation is low or is on par with filtered XLMS. The method is illustrated within simulations and in a laboratory duct system. The disturbance spectrum measured at the error microphone consisted of three major disturbance frequencies and additional white noise. The attenuation achieved was about 37dB, which clearly shows the usefulness of this feedback control approach. Session A2-3, 16:40 h Paper number 187 (Invited, Oral) L. Pettersson, L. Hakansson, I. Claesson and S. Olsson I-INCE Class 46 Department of Telecommunications and Signal Processing, Blekinge Institute of Technology, 372 25 Ronneby, Sweden ADAPTIVE CONTROL OF MACHINE-TOOL VIBRATION BASED ON AN ACTIVE TOOL HOLDER SHANK WITH AN EMBEDDED PIEZO CERAMIC ACTUATOR In the turning operation chatter or vibration is a common problem affecting the result of the machining, and, in particular, the surface finish. Tool life is also influenced by vibration. Severe acoustic noise in the working environment frequently occurs as a result of dynamic motion between the cutting tool and the workpiece. These problems can be reduced by active control of machine-tool vibration. However, machine-tool vibration control systems are usually not applicable to a general lathe and turning operation. The physical features and properties of the mechanical constructions or solutions involved regarding the introduction of secondary vibration usually limit their applicability. An adaptive active control solution for a general lathe application has been developed. It is based on a standard industry tool holder shank with an embedded piezo ceramic actuator and an adaptive feedback controller. The adaptive controller is based on the well known filtered-x LMS-algorithm. It enables substantial reduction of the vibration level by up to 40 dB at 3.4 kHz. Session A2-3, 17:00 h Paper number 398 (Invited, Oral) G.S. Agleitti, R.S. Langley, E. Rogers and S.B. Gabriel The University of Southampton, Southampton, SO17 1BJ, UK I-INCE Class 46 PROBABILISTIC ASSESSMENT OF ROBUST CONTROLLER PERFORMANCE FOR THE SUPPRESSION OF MICROVIBRATIONS Microvibrations are defined as vibrations in the frequency range of up to 1 kHz and their suppression is a critical task in many areas. The most important of this is in spacecraft where they arise due to the functioning of equipment necessary for other tasks, e.g. the motion of reaction wheels, gyroscopes etc. This can lead to such vibrations propagating through the spacecraft structure and therefore impacting on the performance of highly sensitive payloads, such as optical mirrors whose useful operation requires highly accurate suppression of these effects. The nature of microvibrations is such that only active control techniques can be effective. This has led to an increasing volume of work on the application of modern robust control theory to this area. The use of such techniques requires the development of an appropriate mathematical model. The authors have previously developed an efficient Lagrange-Rayleigh-Ritz (LRR) modelling technique. The control theory used to design controllers based on the models produced by this (and other) procedures represent unmodelled/uncertain effects as either structured or unstructured uncertainties and the controllers are intended to maintain stability and robustness under such effects (or perturbations). It can often be difficult, however, to confirm that the actual control system does, in fact, perform well under realistic Internoise 2001 Abstracts 33 perturbations due, in the main, to the lack of a suitable setting in which to study this problem. In this paper, results on the use of Monte-Carlo simulation based studies, to estimate the probability of an unstable plant, are reported. Session A2-3, 17:20 h George Juraj Stein Paper number 400 (Invited, Oral) I-INCE Class 46 Institute of Materials and Machine Mechanics, SLOVAK ACADEMY OF SCIENCES, Racianska 75, SK-831 02 Bratislava 3, SLOVAK REPUBLIC POWER FLOW IN A SIMPLE LINEAR ACTIVE VIBRATION CONTROL SYSTEM AND SYSTEM VIBRATION CONTROL EFFICIENCY The proposed contribution will deal with power flow in simple linear active vibration control systems. In contrary to the time domain approach of Mahajan and Redfield (Journal of Vibration and Acoustics, Vol. 120 (1998), pp. 571-578) the problem will be treated in the complex plane, in similar way as in electricity. The power flow in a single degree of freedom oscillatory system combined with an ideal "sky-hook" type active actuator will be analysed, using the mechanical impedance approach, as described by Hsuhen, W,-J. in Journal of Sound and Vibration Vol. 216(3) 1998, pp. 399-412. Some analytical and simulation results will be presented for this type of active vibration control system, indicating the energetic cost of using active system actuator. Further some ways of describing the frequency dependent efficiency of the vibration control system will be introduced and analysed in respect to variation of system parameters. This approach may improve the insight into the costs and benefits of the active vibration control system in energetic terms especially concerning the actuator power demand. There are still some discrepancies between theoretical actuator power demand, as derived e.g. by Ballo (Vehicle Systems Dynamics, Vol. 24, 1995, pp. 681-691) and the real requirements measured on an operating dummy system by Stein (Proc. of the Instn. of Mech. Engrs. Vol. 209, Part D, 1995, pp. 227-234 and Vol. 214, Part D, 2000, pp. 533-544), hence this problem deserves some attention. Session A2-3, 17:40 h Paper number 402 (Invited, Oral) M.N.H. Siddique and M.O. Tokhi I-INCE Class 46 Department of Automatic Control and Systems Engineering, University of Sheffield, UK VIBRATION SUPPRESSION OF FLEXIBLE-LINK MANIPULATORS USING PD-LIKE FUZZY CONTROL Conventional model-based control has been applied successfully to many industrial processes of wellknown plant dynamics. The performance of such control mainly depends on the parameters during operation. If there are uncertainties in the system model, as in the case of a flexible-link manipulator, the performance of the controller may become unsatisfactory. A fuzzy logic controller (FLC) can easily be designed to deal with such an unknown dynamic system. The conventional approach to fuzzy logic controller design is to generate a fuzzy rule set based upon the system states of error, and change of error, thus producing a two-input single-output control rule base. This provides a response with fast rise time and minimal peak overshoot, but with possible oscillatory behaviour around the set point. Much of the research on PD-like FLC has considered the error and change of error as inputs. In this paper a PD-like FLC is developed for a flexible-link manipulator to suppress the system vibration. Two different types of PD-like FLCs are developed based on considering the error and change of error 34 Internoise 2001 Abstracts and the error and velocity. A comparative assessment of the performance of each type of controller with different parameters and rule-bases is carried out. The end point acceleration of the flexible link manipulator is measured and used for the selection of the inputs of the FLC. Experimental results demonstrating the performance of the FLC strategies in vibration suppression of the system in comparison to a bang-bang input are presented and discussed. Session A2-3, 18:00 h Paper number 176 (Contributed, Oral) C. Roger Halkyard and Brian R. Mace I-INCE Class 42 Department of Mechanical Engineering, University of Auckland, Private Bag, Auckland, New Zealand REAL-TIME ESTIMATION OF FLEXURAL WAVE AMPLITUDE IN A BEAM IN THE PRESENCE OF A NEARFIELD Many potential applications for active vibration control involve structures that are relatively lightly damped and have dimensions that are relatively small in terms of wavelengths. Furthermore, the dynamic characteristics of these structures are often not known accurately, and vary from structure to structure or under different operating conditions. In these situations it is necessary that the control system can adapt or tune itself, and is relatively compact and insensitive to reflections within the structure. An adaptive, wave-based control system that can operate effectively within the presence of nearfields can fulfil these requirements. This paper describes the experimental implementation of such a system, in which adaptive feedforward control is applied to flexural vibration in a beam. A conventional filtered-X LMS algorithm is adopted, with an estimate of propagating wave amplitude used as the cost function. This estimate is obtained by digitally filtering and combining the outputs of an array of three sensors situated within the nearfield of the control actuator. The performance of the wave-based system is compared with that of a conventional adaptive control system based on a single point measurement. It is shown that the wave-based approach gives better broad-band attenuation than the conventional approach, as predicted in simulations. Session A2-3, 18:20 h Paper number 507 (Contributed, Oral) Shi-Hwan Oh and Youngjin Park NOVIC, Center for Noise and Vibration Control, Taejon 305-701, Korea I-INCE Class 46 ACTIVE VIBRATION CONTROL OF MULTI-POINT MOUNTING SYSTEMS WITH FLEXIBLE STRUCTURES Driving of the engine makes unbalance forces which induces vibration to the engine mount system. Active vibration control must be performed to reduce the vibration and the propagation of structureborn sound. In this study, the engine system is modeled as 3-dim. vibration system including flexible structures and an effective active noise control method is proposed. Also, appropriate actuator and sensor locations and types are selected. The miniature of the engine vibration system with multi-input multi-output is built and an active vibration control with multiple filtered-X LMS algorithm is applied to it. The applied control method was effective to reduce the transmitted vibration power through the rubber mount It showed the feasibility of the control of the engine vibration systems with flexible structures. Internoise 2001 Abstracts 35 Session A3-1: Annoyance 1 Chaired by Soames Job and Aad Houtsma Session A3-1, 9:40 h Paper number 16 (Invited, Oral) David Southgate, Nick Fisher and Donna Perera I-INCE Class 52 Australian Commonwealth Department of Transport and Regional Services, GPO Box 594, Canberra, 2601, Australia THE PERSON-EVENTS INDEX - A SIMPLE BUT INVALUABLE AIRCRAFT NOISE ASSESSMENT TOOL In the mid 1990s there was a public outcry following the opening of a new runway at Sydney Airport. There were strident community claims that the Environmental Impact Statement (EIS) for the project had been misleading because it 'hid' the fact that the purported noise benefits of the project were only achieved through concentrating the noise on a small number of suburbs. The noise assessment in the EIS had used conventional thinking - a reduction in the number of persons within an airport’s noise contours is a positive outcome. However, minimisation of the number of persons within the contours can be achieved in several ways. 'Real' noise reductions can be made, for example, by re-routing flights over non-residential areas. On the other hand this outcome can also be achieved simply by focussing the noise on a smaller number of people - the same macro noise exposure result but a very different outcome for the community. The Person-Events Index (PEI) is a simple tool, based on the N70, which, amongst other things, allows the decision maker and the community to very clearly see whether claimed environmental benefits of a project will simply be achieved by concentrating, rather than by reducing, aircraft noise. The PEI is not designed to determine whether Option A is better than Option B. It is a tool to build public confidence in decision making processes through improving the transparency of conventional assessment approaches. Session A3-1, 10:00 h Paper number 56 (Invited, Oral) M.M. Haines, S.L. Brentnall, S.A. Stansfeld Department of Psychiatry, Queen Mary, University of London, UK I-INCE Class 63 A QUALITATIVE STUDY OF AIRCRAFT NOISE ANNOYANCE IN CHILDREN: THEORETICAL IMPLICATIONS Results from recent quantitative research consistently demonstrate that children are a high risk group vulnerable to the adverse effects of noise exposure especially effects on: cognitive performance, motivation and annoyance. The results from these quantitative studies can not provide complete answers as to how children perceive noise exposure, the meaning of noise exposure and the annoyance reaction. The aims of this paper are to discuss the theoretical implications of the results from two qualitative studies. The aim of these studies were to explore children's a) perception of noise exposure; b) perceived risk of noise pollution; and c) the annoyance response. The first study conducted interviews in a focus group format in a sample unselected by noise exposure. The second study, individual interviews were conducted with a purposively selected sample exposed to a specific noise source namely aircraft noise. The results from both studies indicated that noise exposure affected children's quality of life whilst they were participating in a range of activities from school work, homework and playing. In both the studies children report being most affected by neighbours' noise, 36 Internoise 2001 Abstracts road traffic noise and aircraft noise. Children exposed to high levels of aircraft reported more disturbance in everyday life. The emotional response of children to noise was consistent with adult reactions and it would seem that child noise annoyance is the same construct. These results will be discussed in terms of their theoretical implications for annoyance research. Session A3-1, 10:20 h Gooitske Marsman Paper number 323 (Invited, Oral) I-INCE Class 63 RIGO Research en Advies BV, De Ruyterkade 139,1011 AC Amsterdam, The Netherlands DOES THE PRESENCE OF AN AIRPORT INFLUENCE THE WAY PEOPLE EXPERIENCE THEIR LIVING AND URBAN ENVIRONMENT? In the next years the international airport Schiphol will grow. A new runway means more flights, traffic and activity in the surroundings of the airport. The reverse of this growth is the pressure that will come on the liveability experienced by the people living around Schiphol. To be able to influence the liveability in the region you need knowledge about the aspects which determine the way how people experience their living and urban environment. Especially you should have information about the role aircrafts annoyance plays in this. Therefore a project is started in which the indicators of liveability in the region of Schiphol are examined. Liveability is defined as the appreciation of the urban environment by the inhabitants. In this project we assume that the judgement of people about the liveability is the result of the judgements they make about the separate aspects of the environment. These are called indicators. The indicators vary from physical aspects of the dwelling to the annoyance of aircraft. In this paper the influence of the indicators, which are connected to the presence of the airport, will be discussed. In other words, which role plays aircraft annoyance (e.g. noise and smell) in the way people appreciate the urban environment. Session A3-1, 10:40 h George Dodd Paper number 113 (Contributed, Oral) I-INCE Class 61 University of Auckland, New Zealand NOISE, NOISE SENSITIVITY AND ATTENTION A rigorous definition for NOISE has been proposed which would permit logical distinctions between the different categories of noise severity identified in New Zealand legislation. In support of this proposal experiments have been carried out on human subjects to see if it is possible to identify easilymeasurable objective indicators of when a person is attending to sound and when they are not. This has also involved work on ways of measuring the degree of noise sensitivity of individuals. A revision of the Weinstein noise sensitivity questionnaire is proposed. The results from the attention experiment suggest it could be possible to distinguish the category of noise sensitivity of a person by objective measurement of their physiological responses. Internoise 2001 Abstracts 37 Session A3-1, 11:00 h Paper number 80 (Contributed, Oral) I-INCE Class 63 Ken Hume, Daniela Terranova, Callum Thomas and Martin Gregg Departments of Biological Sciences and Environmental & Geographical Studies, Manchester Metropolitan University CAN COMPLAINTS ABOUT AIRCRAFT NOISE BE USED AS AN INDEX OF ANNOYANCE? Objectives: The impact of aircraft movements on the local community surrounding an international airport was assessed for annoyance due to noise level and the time-of-day of the event. Method: Information from 2,667 complaints was collected from Manchester Airport, over the period July to December in 1998 and 1999 along with associated information on noise monitoring and aircraft movements. Results: The vast majority of complaints concerned aircraft noise. The complaints were categorized and ranked into (a) specific aircraft-noise-events, 41% (b) general non-specific concern about the level of noise nuisance, 31% and (c) aircraft considered to be 'off-track', 29%. 575 noise complaints were matched with the monitoring data for these specific aircraft movements. The outdoor threshold for a noise complaint was 75 PNdB, while the average noise level that caused a complaint was 97 PNdB. As predicted, the louder the noise the more complaints were generated. There was a marked hourly variation over the 24 hours in both flight frequency (movements per hour) and complaints. However, the hourly patterns in flight frequency and complaints were clearly distinct. Calculations of the complaints per aircraft movement for each hour of the day showed a striking 24h pattern with night flights (23.00-06.00) causing on average nearly 5 times (highly significantly) more complaints than the rest of the day (06.00-23.00). Conclusion: The louder the noise, above 75 PNdB, the more complaints were generated. Night-time, particularly during the hours of sleep, was the time of greatest sensitivity to aircraft noise (the highest value of 33 complaints per 1000 flights, was between 01.00-02.00 and the lowest, with 1 complaint per 1000 flights, at 08.00-09.00). These results could be used to help plan airport operations/flight movements to minimize community disturbance. Session A3-1, 11:20 h Paper number 100 (Contributed, Oral) Kenji Furihata and Takesaburo Yanagisawa I-INCE Class 63 Shinshu University, Faculty of Engineering, 4-17-1 Wakasato, Nagano-shi, 380-8553 Japan A PATH MODEL APPLIED TO FIELD HEARING DATA ON ROAD TRAFFIC NOISE ANNOYANCE CONSTRUCTED WITH THE INDUCTION METHOD It is well known that psychoacoustic annoyance depends on the loudness, the tone color, and the temporal structure of sounds. And then, it is found that fear and noise sensitivity have a large impact on annoyance. However, words representing the degree of annoyance vary with region and other factors. In many papers published up to date, the factors above were not all considered during selection of the rating words. In this paper, path analysis is applied to noise annoyance constructed with the induction method of reasoning which discovered a general psychological scale from the rating words representing psychological effects for road traffic noise by residents in Nagano City. The model is based on the postulate that an individual’s annoyance data obtained through hearing test in the field at vehicle noise sources is a function of 19 antecedent factors. Of these, 14 (traffic volume, heavy vehicles, crossing, traffic lane, kinds of tires, distance, hearing places, garden wall, gender, age, kinds of subjects, length of residence, ownership of car and license) are exogenous, that is variables within the model are not dependent on any other factors, and the remaining five LAeq(5min), standard deviation, repeated average time of peak levels, average duration time and noise sensitivity) are 38 Internoise 2001 Abstracts endogenous. The results show that the strongest direct effects are LAeq(5min), noise sensitivity, hearing places, distance and kinds of tires. The strongest indirect effects are distance and hearing places. It can be said that the path model can suppose the effects of noise and non-noise factors. Session A3-1, 11:40 h Paper number 344 (Contributed, Oral) Dirk Schreckenberg, Rudolf Schuemer and Ulrich Moehler ZEUS GmbH, Universitaetsstr. 142, D-44799 Bochum, Germany I-INCE Class 66 RAILWAY-NOISE ANNOYANCE AND 'MISFEASANCE' UNDER CONDITIONS OF CHANGE Several studies under steady state conditions showed relationships between residents' noise annoyance and their attitudes towards the authorities, especially their belief that the noise could be reduced, but that those in authority do nothing about it ('misfeasance'). It was assumed that these relationships are even stronger under conditions of expected changes in exposure. Residents at an already existing railway line, which was planned to be extended and broadened, were interviewed before and after the extension. In the before condition, those residents, who mistrust the intentions of the planners, have more negative expectations with regard to their future annoyance. Furthermore the greater the mistrust, the more the expected future annoyance exceeds the actual annoyance. In addition, the mistrust in the before condition also correlates with the later actual annoyance after the extension. Even stronger correlations between mistrust and expected future noise annoyance were found in a survey in which residents at two planned new railway lines were interviewed before the construction of the new lines had begun. These results confirm the importance of effective strategies to inform the residents fully and honestly in an early stage of the planning process, in order to minimise the mistrust of those affected. POSTERS Session A3-1, 12:00 h Paper number 46 (Contributed, Poster) Posterboard AP-5 Tuesday, Author present 13:20 – 14:20 h Ayse Erdem Aknesil I-INCE Class 52 Yildiz Technical University, Architectural Faculty, Building Physics Department, Istanbul, Turkey A PREDICTION FOR POSSIBLE INCREASE IN NOISE POLLUTION FOR THE ISTANBUL'S NEW AIRPORT, AS THE CITY EXPANDS TOWARDS IT The aim of this work is to make comparison between the Atatürk Airport in European part of Istanbul which has caused so many noise problems to its neighbourhood and Kurtküy Airport in Asian part of Istanbul whose construction started recently. This new airport can create similar noise pollution problems in the future. Years ago Atatürk Airport was far from the residential and industrial areas but year by year it found itself in the centre of the city because of unplanned city development. So the effects of aircraft noise on individuals who live at their dwellings and the people work in the industrial areas and offices increased. Yesilküy Airport exceeded its capacity relative to early days and got larger with the addition of new buildings. Furthermore, a new airport Asian side also became a necessity. The solution is to construct another airport for Istanbul in Kurtküy region. The area which the new airport is being built is very similar to the Yesilküy area years ago by its empty environment. In this Internoise 2001 Abstracts 39 paper, the evaluation of Yesilküy airport from an out-of city low traffic into its current situation is studied, and the possibility of similar development for the Kurtküy Airport is discussed. Session A3-1, 12:00 h Paper number 101 (Contributed, Poster) Posterboard AP-6 Tuesday, Author present 13:20 – 14:20 h Kenji Furihata and Takesaburo Yanagisawa I-INCE Class 63 Shinshu University, Faculty of Engineering, 4-17-1 Wakasato, Nagano-shi, 380-8553 Japan A PATH MODEL APPLIED TO SOCIAL SURVEY DATA ON ROAD TRAFFIC NOISE ANNOYANCE CONSTRUCTED WITH THE INDUCTION METHOD Words representing the degree of annoyance vary with region and other factors. In many papers published up to date, however, the factors above were not all considered during selection of the rating words. In this paper, path analysis is applied to social survey data collected with the induction method of reasoning which discovered a general annoyance scale from the rating words representing psychological effects for vehicle noise sources by residents in Nagano City. The model is based on the postulate that an individual’s annoyance obtained the whole judgment through social survey at road traffic noise is a function of 16 antecedent factors. Of these, five (distance, gender, age, length of residence, and license) are exogenous, that is variables within the model are not dependent on any other factors, and the remaining 11 (windows closed, sensitivity to noise, LAeq, adaptability, driving manners, relaxation disturbance, read disturbance, listening interference to the radio or TV, speech interference, work efficiency, sleep disturbance) are endogenous and therefore hypothesized to be partially determined by specified exogenous variables. The results show that the strongest direct effects are relaxation disturbance, listening interference to the radio or TV, sleep disturbance, sensitivity to noise, speech interference. The strongest indirect effects are distance, sensitivity to noise, LAeq, driving manners, adaptability and windows closed. Overall the model explains R2=0.60 of the variation in the annoyance reported by the 523 survey respondents. It can be said that this path model is able to explain the definition of annoyance collected with the induction method. Session A3-2: Annoyance 2 Chaired by Pieter Stallen and J.M. Palma-Oliveira Session A3-2, 13:20 h Rainer Guski Paper number 305 (Invited, Oral) I-INCE Class 63 Fac. Psychology, Ruhr-Uni Bochum, D-44780 Bochum, Germany ABOUT THE VALIDITY OF GLOBAL ANNOYANCE JUDGMENTS Global annoyance judgments are made during systematic surveys of noise-exposed people, and the data is mainly used for descriptions of dose-response relationships, as well as for planning purposes. Although the reliability of global annoyance judgments is sufficiently high, we know rather little about the validity of the annoyance data. In the psychological and social science, the term validity refers to the meaning of data, e.g., how close is the global annoyance judgment related to the annoyance feelings of the individual respondent, how close is the global annoyance judgment related to other noise effects, how close is it related to actual long-term acoustic data, how close to future long-term acoustic data? The question of validity is by no means of purely academic interest - it is essential for 40 Internoise 2001 Abstracts planning processes, noise abatement programs, and the determination of limits for tolerable noise exposure. For instance, it is questionable whether dose-response curves from quasi-stationary noise situations are suited for predicting dose-response data in changing noise situations. Furthermore, it is questionable whether technical noise abatement programs are effective in reducing noise annoyance responses, if they ignore social and psychological factors. The paper gives an overview of several concepts of validity (e.g., internal / external / construct / predictive / ecological validity), discusses the use of empirical data for assessing validity, points to open questions, and proposes lines of thought in order increase some validity aspects of global annoyance judgments. Session A3-2, 13:40 h Paper number 760 (Invited, Oral) I-INCE Class 63 Irene Van Kamp, Hanneke Kruize, A. Ellis M. Franssen and Ronald G. De Jong RIVM, P.O. Box 1, 3720 BA, Bilthoven, The Netherlands INDICATORS OF ANNOYANCE; A PSYCHOMETRIC APPROACH In the past 30 years a variety of concepts and measures are introduced in the noise field such as annoyance, noise sensitivity, interference with daily activities, attitudes toward the noise producer, stress due to noise, perceived controland coping with noise. Little agreement has been achieved on the theoretical meaning of these concepts and the way they should be measured. Especially the issue of construct validity needs further attention. Part of the problem is due to the fact that research in this field has been mostly a-theoretical aimed at dose-response relations. Some uniformity has been reached in the way non-specific annoyance is measured. In this paper it will however be argued on the basis of extensive psychometric analyses on annoyance related measures that this only covers a small part of the total reaction to noise. The interrelation’s between measures of perceived loudness, interference, non-specific annoyance and stress due to noise show a pattern that supports the notion that they are really indicators of one underlying construct. Session A3-2, 14:00 h Paper number 463 (Invited, Oral) S. Hallmann, R. Guski and R. Schuemer Ruhr-University Bochum, Department of Psychology, 44780 Bochum, Germany COGNITIVE PROCESSES IN GLOBAL NOISE ANNOYANCE JUDGEMENTS I-INCE Class 66 Which aspects lead a noise-exposed person to a global noise annoyance judgement during an interview? At present this question can hardly be answered. As a first step we took the empirical approach and asked respondents about their cognitions. The study focused on the following questions: Which cognitive processes take place when affected individuals make a summarized judgement about the degree of their noise annoyance? How influential are certain single factors? The effect of cognitive processes on the global noise annoyance judgement was analyzed by the method of half-structured personal interviews. In three residential areas, each containing a dominating source of noise, 60 subjects were interviewed. Subsequent to the interviews acoustic measurings were carried out in these areas. First the subjects were asked to state how much they felt annoyed by road traffic noise at all, without restricting their judgement to a certain period of time. Except for the time specification, the global noise annoyance judgement was ascertained in accordance with the agreements of the ICBEN (International Commission on the Biological Effects of Noise) Team No. 6. Afterwards the respondents were asked to spontaneously name those aspects leading to their particular judgement. Furthermore they had to assess certain aspects with respect to the importance for the annoyance Internoise 2001 Abstracts 41 judgement and finally rate the five most important aspects according to their influence on the judgement. The collected verbal data were transcribed, evaluated by using Qualitative Content Analysis, and finally combined with the numeric ratings and acoustic data. First results are presented. Session A3-2, 14:20 h Julie Hatfield Paper number 754 (Invited, Oral) I-INCE Class 66 The Department of Psychology, The University of Sydney, 2006, Australia WHAT IS REACTION TO NOISE AND HOW SHOULD IT BE MEASURED? Research and regulations often focus on the relationship between reaction to noise and noise exposure. However, our understanding of this relationship may be muddied by conceptual and methodological problems related to reaction. For example, do we have a clear conception of exactly what reaction is, and how best to measure it? "Reaction" is commonly understood to mean psycho-emotional response, but in the noise literature it has become virtually synonymous with "annoyance". "Annoyance" is too limited a concept of psycho-emotional response for theoretical and practical purposes. General measures of "dissatisfaction with" and "affectedness by" the noise are more valid and reliable than measures of annoyance. However, these measures may capture effects of noise such as activity disturbance and health problems, or variables typically thought to modify noise reaction (such as attitude and sensitivity). For theoretical purposes (e.g. understanding relationships between psychoemotional response and activity disturbance) we might not wish to regard such variables as part of reaction (although they may be consequences of it).However, for regulatory purposes, consideration of more general reaction (including activity disturbance etc.) seems appropriate. Such conceptual issues are seldom addressed. Various methodological issues are also problematic. Potential confounding factors (e.g. problems in matching areas, other sources of pollution, noise sensitivity) will be considered. For example, matching residents of high and low noise areas in terms of socioeconomic status is an insufficient control for potential confounders. For example, areas with low socioeconomic status and low noise are likely to have other problems such as air pollution which might complicate noise reaction, and individuals with high noise sensitivity might self-select into low noise areas reducing the apparent relationship between noise exposure and reaction. Session A3-2, 14:40 h John S. Bradley Paper number 626 (Invited, Oral) I-INCE Class 61 Institute for Research in Construction, National Research Council, Montreal Rd., Ottawa, Canada, K1A 0R6 DERIVING ACCEPTABLE VALUES FOR PARTY WALLS SOUND INSULATION FROM SURVEY RESULTS Acceptable values of party wall sound insulation were derived from the grouped subjective ratings obtained from a field survey of 600 subjects and the airborne sound insulation of their 300 party walls. The subjects lived in both row housing and multi-floor apartment buildings in three different Canadian cities. The survey results clearly demonstrate that residents are influenced by the amount of sound insulation between them and their neighbour and that this is largely independent the type of housing or whether they are owners or renters. Spontaneous responses show increasing negative ratings of their buildings with decreasing sound insulation. Residents with poor sound insulation are more likely to want to move, be awoken by noises and to have trouble falling asleep due to noises as well as to think 42 Internoise 2001 Abstracts that their neighbours are less considerate. Directly elicited response are used to determine STC 55 as a realistic goal for better sound insulation and STC 60 as a more ideal goal that would practically negative effects of neighbour's noises. Session A3-2, 15:00 h Paper number 377 (Contributed, Oral) K. Kanda , T. Yano and T. Tobata I-INCE Class 66 Kumamoto National College of Technology, Dept of Electronic Control, Sue 2659-2, Nishigoshi, Kikuchi, Kumamoto 861-1101, JAPAN ANALYZING THE MODIFIERS FOR THE STANDARD NOISE ANNOYANCE SCALE BY FUZZY RELATION Standard noise annoyance scales in various languages are necessary to precisely compare the data obtained from social surveys conducted throughout the world. An experiment was carried out in Japan to construct a standard noise annoyance scale by using the standard questionnaire and the procedure recommended by the ICBEN team #6. A standard 5-point verbal scale was proposed, based on data set from 1102 subjects of various generations in four areas of Japan. In order to understand the vagueness in the concept "noise annoyance" and in its modifiers well, fuzzy relation was applied to the data treated as a crisp set. In this paper, we showed that a fuzzy approach analyzed the degree of noise annoyance and had some important advantages than ordinary statistical methods. The fuzzy matrix was composed by the result of two evaluations, preference and intensity. The matrix has a similarity relation and a fuzzy ordering relation if the condition details are satisfied. Using the matrix of modifiers' values, cluster analysis was conducted and distances between modifiers were derived. As a result, considering uncertainty and vagueness makes it easier to analyze the noise annoyance modifiers. Session A3-2, 15:20 h Paper number 607 (Contributed, Oral) I-INCE Class 66 Toru Tabata, Takashi Yano, Kazunobu Kanda, Kei Eguchi and Fumio Ueno Kumamoto National College of Technology, 2659-2, Suya, Nishigoshi, Kikuchi, Kumamoto 861-1102 (Japan) FORMULATION OF FUZZY MEMBERSHIP FUNCTION FOR ANNOYANCE LEVEL MODIFIERS In this paper, a social survey-based proposition model on much realistic and accurate fuzzy set in natural language is introduced. Community responses to noise annoyances are usually measured with questionnaires by using categorical verbal labels. In order to analyze characteristics of comparable description words for noise annoyance scales, a large experiment were performed by Yano et al. in 1999-2000.In the experiment with 1103 questionnaires, the intensities of the standard 21 annoyance modifiers were rated by making their position on continuous scale of 10 cm lines between the lowest (not at all) and highest (extremely) annoyance level. The indications on the continuous scale for these 21 modifiers are translated to possibility distributions. The much smoothed fuzzy set curves can be obtained by utilizing the processing of movement average method. Comparing the relation between the noise annoyance scales and experimental results of intensity in the annoyance response , it became clear that the membership function of fuzzy set for all modifiers can be represented as the numerical formula, the combination of a(x-b) to the power c. This is derived by the representation of indistinct meanings of adjective phrases of the fuzzy set theory. These coefficients of the formula are almost fitting to the meanings used in natural language. Our proposed model is much useful for further work on obtaining dose-response relationship. Internoise 2001 Abstracts 43 POSTERS Session A3-2, 15:40 h Paper number 474 (Contributed, Poster) Posterboard AP-7 Tuesday, Author present 16:20 – 17:20 h S. Hallmann, R. Schuemer and R. Guski Ruhr-University Bochum, Department of Psychology, 44780 Bochum, Germany A SYSTEMATIC OVERVIEW OF QUESTIONNAIRES USED IN NOISE EFFECT STUDIES I-INCE Class 66 One of the long-term aims of the ICBEN (International Commission on the Biological Effects of Noise) Team No. 6 is the development of a proposal of questionnaires to be applied in the field of noise effects research. The ultimate aim is to elaborate a proposition on how global and specific noise effects should be ascertained in social surveys. To support this intention the work-group "Ecological Noise Research" has started to work out a systematic overview of existing questionnaires on an international level. This overview intents to provide researchers with the possibility to compare different noise-effect and moderating variables as well as response formats and the structure of many different questionnaires. After inquiring systematically names and addresses of appropriate noise researchers, we asked them to provide us with questionnaires they had used in their own field studies of noise effects. These questionnaires are evaluated using the method of Qualitative Content Analysis. Concurrently, an Access database has been developed, in which the results of the analyses are presented and administered. The database will be placed in the Internet and will be made accessible for every research team. Although the project is not completed yet, the structure and operating of the database can already be demonstrated. Session A3-2, 15:40 h Paper number 510 (Contributed, Poster) I-INCE Class 76 Posterboard AP-8 Tuesday, Author present 16:20 – 17:20 h Masato Akagi, Mizue Kakehi, Masataka Kawaguchi, Masaharu Nishinuma and Akihiro Ishigami School of Information Science, Japan Advanced Institute of Science and Technology, 1-1 Asahidai, Tatsunokuchi, Ishikawa 923-1292, Japan NOISINESS ESTIMATION OF MACHINE WORKING NOISE USING HUMAN AUDITORY MODEL The aim of this paper is to construct a model to estimate noisiness of unstable machine noises. Since noises of working forklifts, one of machine noises, contain moving, lifting and idling sounds and are unstable, it is difficult to estimate noisiness of the noises using loudness as the same as the case of stable noises. This paper proposes a new noisiness estimation model based on a model of human auditory mechanisms, which can deal with not only stable but also unstable noises. The proposed model is constructed using a gammatone filterbank and Medis's hair-cell model. The gammatone filterbank models a frequency analysis mechanism of the basilar membrane and the hair-cell model mimics a neural firing mechanism of inner hair-cells such as large response for onset and adaptation for constant input. Then, noisiness is estimated using the mean and time derivative of firing-rates calculated from outputs of the model. To evaluate the model, noisiness estimation experiments were carried out. The noisiness values used for the experiments were noisiness values of idling, moving and 44 Internoise 2001 Abstracts lifting sounds of forklifts scored by psychoacoustic subjective evaluation. The evaluation results show that all correlation coefficients between the noisiness values from psychoacoustic experiments and their estimation by the model exceeded 0.94. The correlation coefficient using loudness was about 0.7. These results suggest that the proposed model can estimate noisiness of unstable machine noises. In the presentation, we will deal with other machine noises such as air conditioner as well as forklifts. Session A3-3: Combined effects of noise on humans Chaired by Paul Schomer and Joos Vos Session A3-3, 16:00 h Michel Maurin Paper number 197 (Contributed, Oral) I-INCE Class 63 INRETS-LTE, case 24, 69 675 Bron cedex, France THE MEASUREMENT OF COMBINED ANNOYANCES IN MULTI-NOISE EXPOSURE This note deals with noise annoyance measurement. When annoyance is recorded on an ordered category scale, there is a measurement technique (Falmagne, Roberts) providing an annoyance quantification on an interval scale; the Adams and Messick method (Suppes Zinnes). All of these yields a psychophysical response for annoyance versus noise level (Maurin, Maurin and all 1988). When there are many sources and a multi-exposure situation, one may combine Adams and Messick and conjoint measurement technics in order to get an additive conjoint measurement for overall annoyance (Maurin Vernet). But when additive conjoint measurement conditions are not fulfilled (and then no conjoint additive measurement), it is possible to continue with specific probabilistic properties of conditional expectations (Feller), because in the Adams and Messick method the final numerical measurement is an expectation. The note shows the way to adapt the general conditional properties to the measured annoyance µ = E(Y) knowing two different noise exposures X1 and X2 (and possibly more) ; the measured annoyance is given by µ12(X1,X2) = µ1(X1) + µ2(X2) + c12(X1,X2) with a specific term c12 for interaction of two exposures. Then the additive conjoint measurement is a special case (c12 = 0) of this approach, no longer with the specific drastic conjoint measurement conditions, but possible subtleties of interactions. Session A3-3, 16:20 h Paul D. Schomer Paper number 698 (Invited, Oral) I-INCE Class 66 US Army CERL, Champaign, Illinois, USA USE OF THE NEW ISO 226 EQUAL LOUDNESS CONTOURS AS A FILTER TO ASSESS NOISE ANNOYANCE Assessment of the annoyance of combined noise environments has been the subject of much research and debate. There is a need to assess the urban/suburban noise environment using a single, allencompassing method. Currently, most countries use some form of the A-weighted equivalent level (ALEQ) to assess the annoyance of most noises. Many adjustments have been suggested to provide a method whereby all noise can be assessed using ALEQ. But A-weighting only operates in the frequency domain. It provides a constant filter that is independent of sound level. Schomer (2000) Internoise 2001 Abstracts 45 suggested the use of the equal loudness-level contours (ISO 226, 1987) as a dynamic filter that changes with both sound level and frequency. He showed that loudness-level-weighted sound exposure level (LLSEL) and loudness-level-weighted equivalent level (LL-LEQ) can be used to assess the annoyance of environmental noise. Compared with A-weighting, loudness level weighting better orders and assesses transportation noise sources, and it better assesses sounds with strong lowfrequency content. Also, with the addition of a 12-dB adjustment, loudness-level weighting better orders and assesses highly impulsive sounds vis-à-vis transportation sounds. Now there is a proposed modification to ISO 226. This modification is to be circulated as a Draft International Standard by summer 2001. This paper compares calculations of loudness-level-weighted sound exposures for various environmental sounds using the old and the new ISO 226 curves. Session A3-3, 16:40 h Joos Vos Paper number 601 (Invited, Oral) I-INCE Class 61 TNO Human Factors, P.O. Box 23, 3769 ZG Soesterberg, The Netherlands CRITERIA FOR THE AUDIBILITY OF SHOOTING SOUNDS Objective: At some firing ranges, many millions of shots are produced per year. As a result, the rating sound level in the surrounding residential areas might be affected by inaudible or hardly audible shooting sounds produced at distant sites. To discriminate between the relevant and the irrelevant sounds, we developed a measure for the audibility of the shooting sounds. This research was financed by the Dutch Ministry of Defence. Methods: In the laboratory, we determined detectability thresholds for nine sound types produced by various firearms ranging in caliber from 7.62 to 155 mm. Detectability was investigated for two outdoor-to-indoor sound level reduction types (windows open or closed) and for three operating conditions in which the listeners had to perform their tasks (listening to the bangs only or, as a secondary task, reading short stories or watching sound movies). Results: In all conditions we could very well predict the detectability thresholds from the A-weighted and Cweighted sound exposure levels of the bangs. For outdoor levels, the thresholds in the conditions with closed windows were at least 5-10 dB higher than those in the conditions with the windows open. In both facade conditions, the thresholds determined while reading were at least 5 dB, and the thresholds determined while watching video were at least 15 dB higher than the thresholds in the condition in which the subjects had no secondary task. Conclusions: For various listening and operating conditions, we could adequately predict the detectability from just two acoustic measures of the shooting sounds. Session A3-3, 17:00 h Paper number 620 (Invited, Oral) R.F. Soames Job and Julie Hatfield The Department of Psychology, The University of Sydney, Australia I-INCE Class 66 RESPONSES TO NOISE FROM COMBINED SOURCES AND REGULATION AGAINST BACKGROUND NOISE LEVELS Our understanding of the effects of noise from combined sources on reaction, and other potential consequences of noise exposure (e.g. sleep disturbance, cardiovascular disease), is inadequate, despite an array of theories and data pertaining to this issue. Existing data are ambiguous, at least in part because measurement techniques are unclear. Nonetheless, understanding the interactive effects of noise from combined sources is critical to effective regulation. For example, background noise level is often considered when assessing whether a new noise source is "acceptable", on the assumption that 46 Internoise 2001 Abstracts the noise will have a more profound impact when background levels are low. However, there may be complex interactions between the various noise sources comprising the background noise and the new noise source. Given the potential problems associated with regulation against background noise (e.g. "noise creep"), it is important that background noise be employed sensibly, which requires a better understanding of the interactive effects of noise from combined sources. Session A3-3, 17:20 h Paper number 190 (Contributed, Oral) A.G.M. Dassen, J. Jabben and J.H.J. Dolmans I-INCE Class 66 National Institute of Public Health and the Environment RIVM, P.O. Box 1, NL-3720 BA, Bilthoven, The Netherlands DEVELOPMENT AND USE OF EMPARA: A MODEL FOR ANALYSING THE EXTENT AND EFFECTS OF LOCAL ENVIRONMENTAL PROBLEMS IN THE NETHERLANDS The EMPARA model (Environmental Model for Population Annoyance and Risk Analysis) will be presented. EMPARA was developed in order to calculate and analyse the extent of annoyance, due to local environmental problems in the Netherlands. National inquiries have shown that these local environmental problems (noise, local air pollution, odour and external risks) cause annoyance and concern to more than 50% of the Dutch population. EMPARA is used to support government policy makers by giving them insight into the extent and effects of these environmental problems, as well as the possibility to evaluate measures. The presentation focuses on the method of calculating local problems like noise on a national scale. First, the set up of the EMPARA system is explained on the basis of the conceptual design scheme, showing how driving forces are translated into a spatial distribution of sources and emissions, and how land-use and demographic developments are incorporated to get the loads and exposures. In order to maintain a flexible and easy to handle system, it is important to balance the level of detail over all parts of the system. Therefore simplifications and assumptions have to be made that can affect the system’s accuracy. The presentation is concluded with a discussion of the sensitivity analyses, experimental validations and model comparisons carried out to understand and quantify the error margins of the EMPARA results. Session A3-3, 17:40 h Paper number 362 (Contributed, Oral) I-INCE Class 66 Julie Hatfield, R.F. Soames Job, Normal L. Carter, Peter Peploe, Richard Taylor and Stephen Morrell Department of Psychology, University of Sydney, Australia THE VALUE OF THE IMPORTANCE OF ACTIVITIES WHEN PREDICTING REACTION FROM ACTIVITY DISTURBANCE A few studies have identified that noise-induced activity disturbances contribute to negative reactions to noise. The present community survey around Sydney Airport (N=1,015) examined whether the prediction of reaction by activity disturbance is improved when the importance of the activity is considered. A structured interview assessed noise-induced disturbance of several activities (e.g. conversation, relaxing, entertaining) and reaction to noise (annoyance and general reactiondissatisfaction, affectedness). Subjects either rated the importance of each activity directly, or indicated how much time they spend on the activity. Multiplicative indices were computed to weight each activity disturbance by its importance (measured in both ways). Weighted and unweighted activity disturbance were entered stepwise in regression equations on reaction. Results are discussed in Internoise 2001 Abstracts 47 terms of the value of considering the importance of activities when predicting reaction from activity disturbance. Session A4-1: Human effects of vibration Chaired by Setsuo Maeda and Neil J. Mansfield Session A4-1, 9:40 h Yoshiharu Yonekawa Paper number 247 (Tutorial, Oral) I-INCE Class 49 National Institute of Industrial Health, 6-21-1, Nagao, Tama-ku, Kawasaki, 214-8585, Japan ESTABLISHMENT OF FUNDAMENTAL EVALUATION METHODS FOR HUMAN VIBRATION This paper presents series of psychological experiments in order to establish fundamental evaluation methods for human vibration in the 1960's and an early 1970's. Human vibration means whole-body vibration and hand-transmitted vibration in which include x,y,z directions. First, perceptional thresholds of vibration were observed to clarify the lowest level of human sensation of vibration. Equal sensation contours of vibration were observed to determine vibration greatness level(VGL) like loudness level in acoustic field. It was also observed a relationship between VGL and vibration greatness(VG) to sum VGL values of frequency components of the vibration concerned like the relationship between loudness level and sone scale. Finally, emotional response (unpleasant or intolerable ) were observed against vibration greatness level(VGL). Then, it is possible to obtain a relationship between vibration greatness level(VGL) and unpleasant level or intolerable level. Additionally, VGL of pulsed vibration was observed and a time constant of human sensation of vibration was estimated. In the practical use of these evaluation methods, VAL of the vibration concerned is converted into VGL, then VG is calculated to sum values of VGL. Finally, unpleasant level or intolerable level of the vibration was determined using conversion table from VGL into unpleasant or intolerable level. Session A4-1, 10:20 h Paper number 69 (Invited, Oral) I-INCE Class 49 Setsuo Maeda, Yoshiharu Yonekawa, Nobuhisa Okamoto, Yoshihiro Hirao and Akinori Yokota Department of Human Engineering, National Institute of Industrial Health, 21-1, Nagao, 6-Chome, Tama-Ku, Kawasaki 214-8585, Japan WHOLE-BODY VIBRATION PERCEPTION THRESHOLDS WITH DIFFERENT POSTURES This paper presents the results of laboratory experiment concerned with perception thresholds for whole-body vibration on standing, sitting on the floor, sitting on the seat and recumbent subjects. Thresholds of subjects exposed to x-, y- and z-axis sinusoidal vibration from 1 to 80 Hz were determined for four different postures. It was clear that the threshold curves, as a function of frequency, had different characteristics for vertical vibrations than for horizontal vibration in the different postures. Significant differences were found between perception thresholds for horizontal vibrations on the different postures. 48 Internoise 2001 Abstracts Session A4-1, 10:40 h Neil J. Mansfield Paper number 134 (Invited, Oral) I-INCE Class 49 Department of Human Sciences, Loughborough University, Loughborough, Leicestershire, LE11 3TU, United Kingdom LOCALIZED VIBRATION AT THE AUTOMOTIVE SEAT-PERSON INTERFACE The vibration characteristics of the occupied car seat are usually measured using a semi-rigid pad containing accelerometers placed between the seat and the driver. This enables the investigator to measure the dynamic response of the seat and to assess the vibration at one discrete point on the seat surface. This paper reports a study where the results obtained using such a measurement device were compared with localised vibration measurements made at 5 positions along the thigh; between the ischial tuberosity and the front of the seat. Measurements of vibration were made on a car seat with three occupants on two roads and with three repeats for each occupant-road-transducer combination. Results show that vibration at the front of the seat was greater than that measured between the ischial tuberosities using standard techniques. If it is assumed that vibration perception thresholds are constant along the rear of the thigh, then vibration in car seats would be perceived more easily than indicated by measures in the centre of the seat. This might explain some discrepancies between objective and subjective assessments of car seat dynamic comfort. Session A4-1, 11:00 h Kjell Spång Paper number 149 (Invited, Oral) I-INCE Class 49 Ingemansson Technology AB, Box 276, SE-401 24 Goteborg, Sweden QUALITY OF METHODS FOR ASSESSMENT OF ADVERSE EFFECTS ON HUMANS OF VIBRATION INCLUDING SINGLE EVENT SHOCKS AND REPETITIVE SHOCKS Operators of off-road machinery are subjected to vibrations which may involve a potentially adverse health effect. The paper includes a presentation of results of a systematic study of the ability of various methods for assessment of effects on humans of vibration to take into account single event shocks and repetitive shocks. It also includes a comparison of results of assessment of health risks by application of the various methods to a broad range of vibration records. The methods studied are the methods existing in the present version of ISO 2631-1 (1997): r.m.s. (the basic method), running r.m.s., and fourth power vibration dose value. In addition to the existing methods in ISO 2631-1, ahuman response method which is proposed by ISO TC108 SC4 WG10 for assessment of repetitive shocks has been studied. The methods studied have been applied to systematically varied idealised time histories and records from measurement of seat vibration in off-road machinery at travelling over more or less rough surfaces. Session A4-1, 11:20 h Paper number 250 (Invited, Oral) Tsunetaka Matoba and Tatsuya Ishitake I-INCE Class 49 Department of Environmental Medicine, Kurume University School of Medicine, Kurume 830-0011, Japan WHICH IS MORE EFFECTIVE IN HUMAN BODY, VIBRATION, NOISE OR BOTH? Many workers are daily exposed to vibration and noise, which affect their health. It causes using vibratory tools such as chain saws and riveters, and vibratory machines such as bulldozers and Internoise 2001 Abstracts 49 helicopters. The research concerning the health disorders has been performed according to epidemiology, measurement and evaluation of vibration exposure, ergonomics, and medical examination. In the researches of the pathophysiology of health disorders, there are two streams in the world: One is by Portugal research group, and the other is by western countries' group. For the health disorders, Portugal research group names it as "vibroacoustic disease", and other countries do as "vibration syndrome or disease". It is a reasonable name of "vibroacoustic disease" because the generation of vibration from the generator is actually associated with noise. According to these reasons, we should consider the health disorders of workers due to using vibratory machines, resulting from noise as well as vibration. In the paper, we will discuss the pathophysiology and clinical features of health disorders due to working with vibratory tools or machines. Session A4-1, 11:40 h Paper number 260 (Invited, Oral) T. Ishitake and T. Matoba I-INCE Class 49 Department of Environmental Medicine, Kurume University School of Medicine, 67, Asahi-machi, Kurume, 830-0011 JAPAN HOW DOES WHOLE-BODY VIBRATION AFFECT GASTRIC MOTILITY IN THE HUMANS? To clarify the suppressive mechanism and the responses to vibration frequency on gastric motility by Whole-body vibration exposure (WBV), electrogastrography (EGG) was measured in 9 healthy male volunteers under different frequency of vibration and pretreatment of autonomic nerve blocking agents. Sinusoidal vertical vibration at each of 7 different frequencies (4 Hz, 6.3 Hz, 8 Hz, 12 Hz, 16 Hz, 31.5 Hz and 63 Hz) were randomly given to the subject seated on a platform of vibrator for 10 min. The magnitude of exposure at 4Hz was 1.0 ms-2 (rms). The magnitudes of the other frequencies gave the same frequency-weighted acceleration according to ISO 2631 (1997). The amplitude of EGG wave and the power spectrum corresponding to the slow wave component was decreased by vibration exposure of 4, 6.3, 12 and 16 Hz. There was no significant changes in the amplitude of relative power of slow wave component at 8 and 63 Hz of vibration exposure. Under the influence of anticholinergic and an alpha-adrenergic blocking agents, the power of slow wave component were decreased. A further decrease was observed during vibration exposure. A beta-adrenergic blocking agent led to a marked increase in the amplitude of EGG and its power spectrum. With pretreatment by a betaadrenergic blocking agent, however, vibration exposure reduced both of them. WBV may produce the suppression of gastric motility, depending on the vibration frequency. The responses may be mediated by neurohumoral effects as well as the mechanical effects. Session A4-1, 12:00 h Paper number 373 (Invited, Oral) Gülin Birlik and Önder Sezgin SPEED REDUCERS ON THE ROADS I-INCE Class 49 Middle East Technical University, Department of Engineering Sciences, 06531 Ankara, TURKEY In order to reduce the speed of cars and thus to minimize the in-city accidents, speed reducers, i.e. artificial humps (or bumpers) are placed on the roads. Road bumps on the urban routes are well known for many years. Nowadays, however, they are placed, more or less, on each road in the city. Transient or rather impulsive vibrations, therefore, became concern of every driver. Expressing it in other words whole body vibration is not only a problem of off-road vehicles, but also of the private cars, buses and mini-buses. In this study five types of speed reducers, i.e. type A (one row of thin tiny metal boxes and 50 Internoise 2001 Abstracts rubber coated narrow (half) metal cylindrical shell), type B (wide asphalt hump), type C (two rows of thin tiny metal boxes, type D (narrow rubber coated (half) metal cylindrical shell) and type E (43 rows of narrow very thin strips or bands) are assessed based on a vehicle, i.e. passenger car, speed (V = 20, 30 and 40 km/hr) and back (rather low back) cushions. Response variables are chosen to be az (vertical) and ax (fore and aft) accelerations at the (seat)/(beneath the ischial tuberosities of the driver) (i.e. azs or axs) (scat back)/(waist or back) of the driver (i.e., azw (or azb)) interfaces. In the measurements rectangular time weighting function and linear averaging mode were chosen. Frequency resolution was 0.25 Hz. The comparison of floor az versus frequency curves showed that (A, D), (C, E) and (B) types have their peak az (floor) values in (3 Hz < f < 10 Hz), (f ≥ 15 Hz) and (f < 5 Hz) frequency ranges. Since chest, abdominal pains and muscle contractions are affected by the vibrations in the 4-9 Hz frequencies, B type is analyzed in detail. A 50% increase in the speed of the car while crossing over the bumper caused a shift of 1.5 Hz in the peak of transmissibility, Hp (based on power spectral density function method) (for V = 30 km/hr, peak Hp = 2.58 at f = 5.25 Hz). The presence of back support (either a 100% polyurethane foam or air filled cushion) happened to be of no considerable help in reducing axh accelerations. Besides axw accelerations, (without cushion peak axw = 0.26 m/s2 at f = 3 Hz, V = 30 km/hr) on the bumper were increased when a back cushion (with cushion, peak axw = 0.34 m/s2 at f = 3 Hz, V = 30 km/hr) was present. Preventive measures for the reduction of the speed of cars seem to be hazardous to men health and isolation efficiency of the seats seems to be not adequate, on the bumpers, whatever the speed of the car is. Session A4-1, 13:20 h Paper number 622 (Invited, Oral) Riei Ishida and Setsuo Maeda I-INCE Class 49 Department of Urban Environment Systems, Faculty of Engineering, Chiba University, 1-33, YayoiCho, Inage-Ku, Chiba 263-8522, Japan EFFECTS OF VIBRATION SOURCE CHARACTERISTICS ON WHOLE-BODY VIBRATION PERCEPTION THRESHOLDS On the evaluation of environmental vibration in building, the consideration of the flow of evaluation such as 'vibration source - propagation - building - human body' is very important. But, the series of flow has never been concretely proposed before. As for traffic-induced vibration, it has been already pointed out by the author that the spectrum of the building vibration is affecting by the characteristics of vibration sources such as road classification, pavement roughness and traffic conditions (type of vehicle, loading condition and vehicle velocity) and the characteristics of soil propagation and buildings. But, it has not become clear whether the differences of the spectrum of vibration sources are affecting on the whole-body vibration perception thresholds. Therefore, in this paper, in order to clarify the effects of vibration sources on whole-body vibration perception, the frequency-weighted vibration acceleration values with Wd, Wk (ISO 2631-1), Wb(BS 6841), Wg(JIS C 1510) and Wcomb(ISO 2631-2) frequency weightings were calculated with the vibration signals on soil or buildings obtained by parametric field experiments. And the frequency-weighted vibration acceleration values were compared with a scale of vibration perception according to the BS 6841 and ISO 2631-1 standards. The effects of the spectrum of the building vibration depending on the characteristics of vibration sources such as road classification, pavement roughness and traffic conditions (type of vehicle, loading condition and vehicle velocity) and the characteristics of soil propagation and buildings on the whole-body vibration perception thresholds are discussed. Internoise 2001 Abstracts 51 Session A4-1, 13:40 h Paper number 645 (Invited, Oral) I-INCE Class 49 Roberto Torres, Guillermo Tirado, Amaury Roman, Ricardo Ramirez, Halima Colon, Amilcar Araujo, Fernando Pais, Wochiech Marciniak, Joao Nóbrega, Armando Bordalo e Sá, Joao M.C. Lopo Tuna, M. Salomé, N.A.A. Castelo Branco and Mariana Alves-Pereira Virtual Research Group, San Juan, Puerto Rico VIBROACOUSTIC DISEASE INDUCED BY LONG-TERM EXPOSURE TO SONIC BOOMS Background. Vibroacoustic disease (VAD) is a whole-body pathology induced by long-term exposure to low frequency noise (LFN) (< 500 Hz, including infrasound), characterized by proliferation of extra-cellular matrices, and reflected by thickening of cardiovascular structures. VAD has been identified in many LFN-exposed workers. The unexpected pathology observed among the island habitants of Vieques, Puerto Rico was in stark contrast to that obtained in San Juan. Since the 1970's, part of Vieques has been permanently used as a military testing-site. These tests involve sonic booms that result in acoustic environments with a predominance of LFN, reaching 120 dB around 30 Hz. Goal. To determine whether or not the habitants of Vieques are suffering from VAD. Material and Methods. Volunteer study population: N=50, permanent residents of Vieques since birth. Control population: 50 age-matched and healthy permanent residents of Ponce, since birth. The HP Sonnos 1800 was used for echocardiography, and results were analyzed by two independent teams. RESULTS. All Vieques habitants had thickened pericardia. All subjects had some additional sort of abnormality: thickening of the aortic valve (27 cases), mitral valve (15); Regurgitation of the mitral valve (24), of the aortic valve (19). 28 individuals, one of whom was 11 years old, had 3 or more abnormal parameters. This is in contrast with the control population. Discussion. The general population of Vieques is suffering from VAD due LFN exposure caused by military testing. Once again it is proven that acoustic phenomena affects the whole-body, and VAD is, in fact, an environmental disease. Session A4-1, 14:00 h Marek A. Ksiazek Paper number 126 (Contributed, Oral) I-INCE Class 49 Institute of Mechanics and Machine Design, Cracow University of Technology, Al. Jana Paw•a II, 37 , 31-868 Kraków, Poland VIBRATION ISOLATION OF A SITTING HUMAN BODY CONSIDERED AS AN ACTIVE BIOMECHANICAL SYSTEM In this paper a vibration isolation of sitting human body represented by an active biomechanical system is considered. The paper is composed of the following parts: - description of the principal features of the active biomechanical model of the sitting human body; - analysis of the dynamical system composed of the existing passive vibration isolation systems (VIS) and the active biomechanical model, procedure of synthesis of optimal vibration isolation system (VIS) for the considered active model. In the first part the structures and the parameters of the active 2 DOF biomechanical model of the sitting human body are presented. In the second part the dynamical properties of the system: passive seat - active model subjected to random excitation were analysed. In the third part the procedure of calculation of the optimum vibration isolation system for the active biomechanical model is proposed. The results are presented in analytical and graphical forms. 52 Internoise 2001 Abstracts Session A4-1, 14:20 h Paper number 243 (Contributed, Oral) I-INCE Class 49 Wan-Sup Cheung, Byung-Jae Choi, Hyu-Sang Kwon and Hee-Joon Eun Acoustics and Vibration Group, KRISS, Korea HEAD VIBRATION MEASUREMENT DEVICES AND THEIR UNCERTAINTY CHARACTERISTICS This paper addresses issues encountered in measuring the general 6-degree-of-freedom motion of human head. This measurement requires the usage of multiple translational accelerometers, such as the well-known ‘bite-bar’, which consists of three or four measurement blocks connected by the two or three connecting rods to position each block in a fixed distance. Each measurement block for the bitebar has a single linear accelerometer, dual or three ones of which measurement axis or axes are oriented for the indirect measurement of rotational acceleration. There are many possible ways of measuring the three orthogonal angular accelerations using multiple linear accelerometers installed in the three or four measurement blocks. But, a standardized way of measuring the angular acceleration components using the multiple linear accelerometers has been not yet established. After reviewing previous work related to the topics, we have found that detailed theoretical and experimental researches on the bite-bar are not enough to understand rigorously what expected performance can be obtained and what measurement errors and limits are involved. It has been apparent that those questions are obviously answered by carrying out the angular acceleration calibration of the bite-bar and that calibration results enable us to see its real performance and related measurement uncertainties. This observation has motivated the reappraisal of better configuration ways of the measurement blocks of the bite-bar and their theoretical formulation for analysis of angular acceleration measurement errors. At this moment, experimental attempts have been already made for three different configurations of the bite-bar, and their error analysis and performance comparison work is in progress. Of course, this paper will introduce a systematic calibration procedure and its results will be demonstrated in the presentation of this paper. Session A4-1, 14:40 h Paper number 472 (Contributed, Oral) Gen Tamaoki and Takuya Yoshimura I-INCE Class 49 Department of Mechanical Engineering, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji-shi, Tokyo, 192-0397, JAPAN DEPENDENCY OF DYNAMIC CHARACTERISTICS OF HUMAN BODY ON INPUT VIBRATION AMPLITUDE UNDER EXPOSITION TO MULTIDIRECTIONAL EXCITATION The purpose of this study is to understand dynamic characteristics of sitting human body which is exposed to multidirectional base excitation. By performing a vibrational test, transfer functions from the seat to the mouth and to the forehead are measured with respect to three translational directions. In this paper, the dependency of the transmissibility of transfer functions on the input vibration amplitude is examined experimentally. As a result, some non-linearity of the human body is revealed; with the increase of input vibration amplitude, the transmissibilities near resonances become small and the resonant frequencies decrease. These characteristics are clearly observed in rotational motions. But some characteristics are observed independent of the input vibration amplitude. That is, there are a few resonant peaks in each translational direction and the coupling between the fore-and-aft motion and the vertical motion. It is experimentally confirmed this dynamic coupling is caused by the pitch motion of the head. In constant input vibration amplitude, dynamic characteristics can be considered Internoise 2001 Abstracts 53 as linear because measured transfer functions can be modelled by superposition of small number of modal parameters. POSTERS Session A4-1, 15:00 h Paper number 25 (Invited, Poster) I-INCE Class 49 Posterboard AP-9 Tuesday, Author present 15:40 – 16:40 h Nuno A.A. Castelo Branco, Jose I.G. Fragata, Artur Aguas, Antonio de Sousa Pereira, Emanuel Monteiro and Nuno R. Grande Center for Human Performance, Alverca, Portugal THE PERICARDIUM IN NOISE-EXPOSED INDIVIDUALS BACKGROUND. Exposure to low frequency noise (LFN) (<500 Hz, including infrasound) can induce the development of vibroacoustic disease (VAD). One of the main features of VAD is the proliferation of the extra-cellular matrix which induces thickening of cardiovascular structures (particularly the pericardium), and that can be visualized through echocardiography. Despite the increased thickening of the pericardia, no diastolic dysfunction has ever been identified in VAD patients. Studies relating gross anatomy to echocardiography findings are very few. GOAL. To verify whether there is a corresponding anatomical thickening of the pericardia, as viewed though echoimaging, and, if so, investigate what elements of this organ are involved. In this report we perform histological and ultrastructural studies of the pericardia of 9 individuals, occupationally exposed to LFN for over 15 years, and who were submitted to cardiac surgery for other reasons. METHODS. Pericardial fragments were prepared for histological and ultrastructural studies by standard methods. RESULTS. The most important findings deal with the real thickening of the pericardium, the dynamic arrangements of mesothelial cells of the serosa leaflet, the plasticity of the cells found among the multifascicular waveform collagen fibers, and multiple images featuring processes of cell aging and death. DISCUSSION. Echo-imaging of pericardial thickening has a real, anatomical correspondence. The features identified in this study partially explain why no important diastolic changes are observed in spite of the pericardial thickening. Session A4-1, 15:00 h Paper number 26 (Invited, Poster) I-INCE Class 49 Posterboard AP-10 Tuesday, Author present 15:40 – 16:40 h Mariana Alves-Pereira, M. Salome Castelo Branco, Jerzy Motylewski, Ana Pedrosa and Nuno A.A. Castelo Branco Dept. Environmental Sciences & Engineering, Nova University of Lisbon, Portugal AIRFLOW-INDUCED INFRASOUND IN COMMERCIAL AIRCRAFT INTRODUCTION. An in-depth study of noise was performed on onboard several commonly used Portuguese airline aircraft (B737-2, B737-3, A319, A320, A310, A340, EBR145, F100), where low frequency noise (LFN) (<500 Hz), including infrasound (<20 Hz) was assessed. These noise assessments detected higher levels of infrasound in the cockpits than in the cabins. GOAL. to mathematically treat the values of sound pressure levels to demonstrate that a) Cockpits have statistically significant higher levels of infrasound than the cabins and b) Aerodynamic flow causes the 54 Internoise 2001 Abstracts increased levels of infrasound in the cockpit. METHODS. Sound pressure levels were measured and frequency distribution analyses were obtained from 1.6-20 Hz. Data were also obtained at different flight altitudes and speeds. RESULTS. There was a statistically significant difference (s.s.d.) (F = 41.9346, p<0.001) between the cockpits and cabins, for the levels of infrasound. Different speeds induced a s.s.d. (F=8.4181, p<0.001) in the levels of infrasound, as did aircraft type, i.e. its aerodynamic properties (F=113.5263, p<0.001). Speed and aircraft type were also found to s.s. influence the levels of infrasound (F=73.600, p<0.001). A s.s. correlation was found between the levels of infrasound and altitude (p<0.001). CONCLUSIONS. Infrasound is an acoustic component inherent to the cockpit environment, but not to the cabin. The levels of infrasound in the cockpit vary with speed and altitude, and can be explained by the impact of airflow on the nose of the aircraft. This is conditioned by aircraft design. Session A4-1, 15:00 h Paper number 27 (Invited, Poster) I-INCE Class 49 Posterboard AP-11 Tuesday, Author present 15:40 – 16:40 h Amilcar Araujo, Fernando Pais, Joao M.C. Lopo Tuna, Mariana Alves-Pereira and Nuno A.A. Castelo Branco Santa Maria University Hospital, University of Lisbon, Medical School, Portugal ECHOCARDIOGRAPHY IN NOISE-EXPOSED FLIGHT CREW Background. Echocardiography is the method of choice to diagnose and monitor patients with vibroacoustic disease (VAD), which is a whole-body pathology is caused by long-term exposure (years) to low frequency noise (LFN) (less than 500 Hz, including infrasound). To evaluate the noiseinduced thickening of cardiovascular structures, echos were administered to volunteer members of Portuguese airline companies. Methods. Study population: selected 30 male pilots, ave. age 48 (range 30-61), and 30 female flight attendants, ave. age 43 (range 27-57). Total exposure time of all individuals was over 8 yr. Echo-Doppler studies were performed and recorded on coded videotapes. Three morphological parameters were evaluated: thickening of the pericardium and mitral & aortic valves. Results. There is no statistically significant (s.s.) difference (p = 0.63) regarding thickening of cardiac structures when thickening scores (TS) of pilots and flight attendants are compared. When broken up into 2 age groups (less than 45 yr and over 45 yr),there is an s.s. difference (p = 0.000) between the TS of pilots and flight attendants of the less than 45 yr age-group. There is no evidence of an s.s. difference (p = 0.26) between the over 45 yr age groups. The mean thickening for the less than 45 yr group is larger in the pilot group than in the flight attendant group (6.56 + 1.67 vs. 4.00 + 1.25). Conclusion. The pericardium and mitral and aortic valves are thickened in these professionals, indicating that a risk for the development of VAD is non-trivial. Internoise 2001 Abstracts 55 Session A4-1, 15:00 h Paper number 229 (Contributed, Poster) Posterboard AP-12 Tuesday, Author present 15:40 – 16:40 h Tommaso Meloni Swiss Agency for the Environment, Forests and Landscape 3003 Berne, Switzerland PROTECTION OF MAN AGAINST VIBRATION: CRITERION - APPLICATION I-INCE Class 49 Most installations, machines or vehicles in operation produce vibrations. In respect of the air transmission (sound) the effects on persons are quite well studied. But in respect of the other way of transmission via solid material the effects on persons can't be neglected. Humans feel vibration in physical contact with buildings and hear sound born from the vibrating structures. Among the range of these influences the dominant effects during daytime are identified as subjective effects by name disturbance and annoyance. During night-time vibration and structure born noise affect on sleep. Wake-up reactions are the dominant effects under all possible specific effects on sleep. These reactions are widely used as criterion to apply the protection of man against vibration and structure born noise. For the daytime evaluation specific properties of vibration and structure born noise and their combined effects has to be considered. An ordinance dealing with protection for man against vibration and structure born noise is now in preparation in Switzerland. The primary intention is to protect humans in rooms sensitive to noise and vibration. These rooms are in apartments or in industrial buildings that are regularly occupied by persons for longer periods. The relation between the effects on man including protection criterions and the provided legal procedure will be discussed in the paper. Session A4-1, 15:00 h Paper number 366 (Contributed, Poster) I-INCE Class 49 Posterboard AP-13 Tuesday, Author present 15:40 – 16:40 h Seong Keol Kim, Seth W. White, Anil K. Bajaj, Patricia Davies, Kihong Park and Joon Hyun Kim Graduate School of Automotive Engineering, Kookmin University, 861-1, Chongnung-dong, Songbukgu, Seoul 136-702, Korea MODELING THE VIBRATION RESPONSE OF MANNEQUINS IN CAR SEATS ABOUT A STATIC EQUILIBRIUM POINT A simplified modeling approach to predict the vibration response, about an operating point, of mannequin occupied car seats is demonstrated to be feasible. A two-dimensional model, consisting of rigid bodies interconnected by pin joints with torsional dampers, springs and dampers has been developed. The springs and dampers are linear and the nonlinearity in the model is due to geometric effects, although at vibration levels experienced under normal driving conditions a linearized version of the model predicts responses. To produce good initial conditions for the model parameters (stiffness and damping coefficients)a mannequin-seat system was initially broken down into subsystems, and approximate values for the model parameters were determined from experiments on these subsystems. The effects of changing model parameters on the natural frequencies, the mode shapes and resonance locations in frequency response functions are shown. Reasonable qualitative as well as good quantitative agreement between experimental and simulation frequency response estimates is obtained. 56 Internoise 2001 Abstracts The application of these simplified models is in car seat design, where they would facilitate evaluation of the effects of seat design changes prior to prototyping. Session A4-2: Noise and recreation Chaired by Constantinos Vogiatzis and Marion Burgess Session A4-2, 15:20 h Paper number 663 (Invited, Oral) Miguel Arana and Amando Garcia Physics Department, Public University of Pamplona, 31006 – Pamplona (Spain) EVALUATION OF THE NOISE PRODUCED BY LEISURE ACTIVITIES (BARS, PUBS AND DISCOTHEQUES) IN DIFFERENT URBAN ZONES OF SPAIN I-INCE Class 52 The sound sources related with many leisure activities (open air concerts, fun fairs, pyrotechnic fires, street celebrations, etc.) have a very acute social importance in many urban areas. Within this group of sound sources, it should be specially mentioned the bars, pubs, discotheques and similar sites, that can produce a quite important direct and indirect noise impact on the nearby residents. This is a very important problem in many cities of Spain, touristic or not, where these activities are the target of frequent complaints, specially during weekends, when they remain open most of the night and, in warm weather, with many people staying in middle of streets, chating and drinking. In the present paper the main general characteristics of this problem have been analized. In particular, the results of two different surveys carried out in Valencia (east of Spain) and Pamplona (north of Spain) are presented and discussed. A special attention has been devoted to evaluate the negative effects of these leisure activities on the urban residents of these zones, whose sleep and night rest can be seriously affected. The need for a most severe noise control of such annoying activities is also discussed. Session A4-2, 15:40 h Paper number 670 (Invited, Oral) Constantinos Vogiatzis SSE Ltd, 48 Ventouri Str, 15561, Holargos – Athens Greece I-INCE Class 52 NOISE IN THE TOURIST CITY OF RHODES - 10 YEARS OF EFFORTS IN ACOUSTIC ENVIRONMENT REHABILITATION City of Rhodes (tourist resort in Southern Greece) has initiated for the 1st time in 10 years ago -a large programme aiming at the rehabilitation of the urban acoustic environment. Environmental noise in tourist cities was not covered by relevant laws in that period, taking into account also the difficulty of establishing laws for a parameter of life quality attached to the city life rythme properly i.e. urban structure, economic model, cultural & meteorological elements, inhabitants & tourist's perception of the noise climate). The last 10 years the city has implemented a number of global programs and the use of anti-noise measures of different types of application regarding also differences on the human reactions to noise of both inhabitants and tourists mainly from Northern European countries. According to the relevant noise map of 1999 compared to the noise map of 1990 a considerable diminution was observed, which for the index Leq day was established from 74,1 dB(A), to only 69,4 dB(A). The city of Rhodes action plans for rehabilitation of the urban acoustic environment consists mainly at the proper valorisation spaces sound quality, referred as "space spirit" which is of main importance for touristic zones take into account the maintain of an upper grade of acoustic quality. Internoise 2001 Abstracts 57 Session A4-2, 16:00 h Paper number 671 (Invited, Oral) Fotini Honianaki and Constantinos Vogiatzis 30, 28th October str, 1554, Athens, Greece I-INCE Class 52 ACOUSTIC CLIMATE IN TOURIST TRADITIONAL SETTLEMENTS IN GREEK ISLAND - THE CASE OF MYKONOS Tourism consists the most important productive activity on the island of Myconos at the Egean sea (Greece) which is a popular tourist destination with international magnitude. During the summer period, the islands’ population is estimated to exceed itself by twice as much. A great number of leisure and restauration centers is available while the circulation of vehicles and especially motorcycles has risen in significant ways. A great density of relating fenomenos appears in the internal area of the most important traditional settlement of the island which is Chora of Myconos. A vaste Measurement's program have already executed in the touristic period during specified innerperiods of the day and in locations where land use and the already mentioned activities differentiate completely. The evaluation of the results regarding the relating research has proved the connection between the parameters of the acoustic environment and the land use in the immediate and the surrounding area as well as the period of the day in addition with the magnitude characterising the relating activities. The formation of the acoustic environment depends on the structure of traditional urban tissue. Session A4-2, 16:20 h Paper number 710 (Invited, Oral) N.H. Krog and B. Engdahl I-INCE Class 52 National Institute of Public Health, Department of Environmental Medicine, Postbox 4404 Nydalen, N-0403 Oslo, Norway DOSE-RESPONSE RELATIONS FROM TWO OUTDOOR RECREATIONAL AREAS IN NORWAY The special peace and quiet of nature is shown to be an important component of what people seek in outdoor recreation (Vaagbö, 1993). In recent years there has been growing concern about the effects of noise on outdoor recreationists (Kariel, 1990; Anderson et al., 1993; McDonald et al., 1995; Fidell et al., 1996; International Recreational Noise Symposium, 1998; FICAN Symposium on the Preservation of Natural Quiet, 1999; Krog et al., 2000). In 1998 the former main airport of Norway, Fornebu, was closed down, and another airport, Gardermoen, was extended to become the new main airport. The relocation of the airport offered the opportunity to study visitors’ reactions to different levels of aircraft noise in the same outdoor recreational areas. Two recreational areas were studied, one close to the old airport (Bygd¯y), and one close to the new airport (Romeriks Âsen). Both areas are important local recreational areas. Field studies with simultaneous interviewing and sound recordings were conducted both before and after the change. The main purpose was to establish dose - response relations. To obtain dose - response relations, contingent on the distinguishing features of outdoor recreation, annoyance is controlled for expectations to the recreational experience. Two models of exposure are tested and discussed: one containing parameters that require an observer to distinguish the aircraft noise from the overall noise exposure, and one that does not. 58 Internoise 2001 Abstracts Session A4-2, 16:40 h A.E. Watson Paper number 538 (Invited, Oral) I-INCE Class 52 Acoustic Consultancy Services, 2 Belhaven Terrace Lane, Glasgow, Scotland COMPARISON OF METHODS FOR NOISE MANAGEMENT AND CONTROL AT UK MOTOR SPORT VENUES Over the past few years, different methods have been introduced in the UK to monitor and control environmental noise impact from motorsport venues. This followed the initiative taken by the Association of Motor Racing Circuit Owners and The Royal Automobile Club Motor Sports Association in producing Guidelines for Noise Control at venues. This paper compares the effectiveness of some of the systems in use at large multi purpose venues. Session A4-2, 17:00 h Marion Burgess Paper number 354 (Invited, Oral) I-INCE Class 52 Acoustics & Vibration Unit, School of Aerospace & Mechanical Engineering, University College, UNSW, Australian Defence Force Academy, Canberra, ACT 2600 AUSTRALIA COST EFFECTIVE MANAGEMENT OF NOISE FROM AMATEUR MOTOR SPORTS The aim of a noise management plan is to control the noise such that the surrounding residential areas are not subject to excessive noise impact. This paper will describe the approach to noise management that has been developed which meets the requirements for environmental noise criteria but with minimum cost to the amateur motor sports organisation. The complainants are located at some distance from the facility and subject to many noises other than from the motor sports, hence automatic monitoring at this location is not appropriate. The first stage of the program is for simultaneous attended monitoring in the residential area plus automatic logging near the track using a short time period, 5 sec. From the noise levels in the residential area the number of event credits for each type of motor sports are established by the government agency. For subsequent events it is only necessary for the automatic logging to be undertaken near to the track to establish that the noise levels during the event. This approach to noise monitoring requires no skills for the event organisers and is low cost as the independent agency only programs the logger and downloads the data. The Government agency has a record of the noise during the event should there be a complaint received. The effects of improved mufflers or changes to the operation of the event can also be monitored in an efficient manner. Session A4-2, 17:20 h Christoph Lechner Paper number 30 (Contributed, Oral) I-INCE Class 38 Dr.-Ambros-Giner-Weg16, A-6065 Thaur, Austria CONTROL OF MUSIC NOISE WITH SOUND LEVEL LIMITERS The presented report was drawn up by a study group under the supervision of the Federal Environment Agency, Austria. The problem being discussed is the selection and adjustment of sound level limiters. Restaurants and bars are often situated in buildings with flats inside. To avoid disturbing people living in these flats sometimes sound level limiters are being used in music systems. To ensure that music is not being played too loud in restaurants and bars where normally only background music is allowed the authorities generally require limiters. In practise not all of them work all right. Sometimes the Internoise 2001 Abstracts 59 wrong type of limiters are used, sometimes the equipment is not adjusted the right way. The disturbance is normally caused by low frequencies of the music sound. To find a suitable signal for the calibration of limiters the music signals of various CDs were analysed. The musical variety reached from country music to hard rock. By simulating the function of sound level limiters pink noise was found as the best suitable signal for adjusting. It also was found out that in general only one type of limiters is possible. Moreover a simple and quick to use guidance for determining the admissible music sound level in restaurants was developed that bases on the Austrian standard. To meet these requirements sometimes different signals are necessary. All these signals are on a CD attached to the report. Session A4-2, 17:40 h Nicholas P. Miller Paper number 338 (Invited, Oral) I-INCE Class 66 Harris Miller Miller & Hanson Inc., 15 New England Executive Park, Burlington, MA 01803 USA EFFECTS OF MILITARY AIRCRAFT OVERFLIGHTS ON RECREATIONAL USERS OF NATIONAL PARKS This study was initiated as part of the cooperative US Air Force / National Park Service efforts to understand and effectively manage the potential adverse effects military air crew training can have on the National Parks. Through simultaneous sound data acquisition and Park user interviews, data were collected that provided a basis for determining how military jet overflights can affect visitor experience at a site in White Sands National Monument, New Mexico. Several useful findings resulted from the analysis. First, visitors can distinguish between the concepts of ‘annoyance’ and ‘interference’ produced by aircraft sound. Annoyance is an emotional reaction, while interference is more of an objective judgement. Visitors can find that the sound of aircraft interferes with the natural soundscape, but are not necessarily annoyed. Visitors believe annoyance results if the interference is often or severe enough. Second, visitors tend to be less annoyed by aircraft noise if they remember learning that they could hear or see aircraft while in the Park. This finding shows the importance of informing visitors about possible aircraft overflights - i.e., managing visitor expectations. Finally, aircraft noise is likely to produce less annoyance if aircraft fly over in close succession, rather than widely spaced, one at a time. Session A4-3: Rating Prominent Discrete Tones Chaired by Bob Hellweg Session A4-3, 18:00 h Paper number 624 (Invited, Oral) Dirk Sagemuehl and Lothar Schmidt Bayer AG, Dept. ZT ñ TE PAT SST, Building H1, 51368 Leverkusen, Germany REVISION OF E DIN 45681 (TONALITY) I-INCE Class 81 The draft of the German standard DIN 45681 "Bestimmung der Tonhaltigkeit von Geräuschen und Ermittlung des Tonzuschlages für die Beurteilung von Geräuschimmissionen" (Tonality estimation of sounds and calculation of the penalty for the rating of noise immission) was published in Jan. 1992. Since then there have been multiple applications. Reports show comparisons between penalties calculated in accordance to the DIN standard and those defined by experienced people. The 60 Internoise 2001 Abstracts amendments of the DIN standard are based on these comparisons. The basic concept of the DIN standard and the major amendments will be presented. Session A4-3, 18:20 h Paper number 649 (Invited, Oral) Robert D. Hellweg Jr., Anne C. Balant and Matthew A. Nobile I-INCE Class 81 Robert D. Hellweg, Jr., Compaq Computer Corp. 200 Forest Street Marlboro, MA 01752 USA A COMPARISON OF TWO METHODS FOR THE EVALUATION OF PROMINENT DISCRETE TONES - RECOMMENDATIONS This paper presents results of the study of prominent discrete tones by the Inter Committee Working Group (ICWG) from Information Technology and Telecommunications Equipment(ITTE). There are two objective methods that evaluate the potential annoyance of prominent discrete tones in the noise emissions of products: (1) the Tone-to-Noise Ratio (TNR) method (ISO 7779-1999, ECMA-74-1999, ANSI S1.13-1995) and (2) the Prominence Ratio (PR) method (ANSI S1.13). Progress and earlier results were reported at INTER-NOISE 99, the 139th Meeting of the Acoustical Society of America, INTER-NOISE 2000 and NOISE-CON 2000. This paper presents results of a round robin involving 40 signals (some from ITTE products and some artificial sounds) which were recorded and distributed to participants as .wav files. Each of several ITTE company laboratories provided subjective ratings and a sub-set of the labs determined the objective Tone-to-Noise Ratios and Prominence Ratios. This paper presents comparisons of the objective and subjective ratings from the different laboratories. The relative success of the two objective methods in predicting the subjective ratings is discussed, as are the underlying issues and recommendations of the task group, which include an increase in the threshold of prominence and a further increase at lower frequencies. Session A5-1: Outdoor Machinery noise Chaired by Giorgio Billi and Brian Ross Session A5-1, 9:40 h Carol Drutowski Paper number 54 (Invited, Oral) I-INCE Class 11 The Toro Company, 8111 Lyndale Avenue South, Bloomington MN, 55420, USA LAWN MOWER NOISE EMISSION Variety of lawn mower types: Rotary, cylinder, hover, and flail are the most common classification of lawn mowers. Uses in Europe: Single spindle rotary lawn mowers are used mainly for consumer applications such as lawns and gardens of limited dimensions (700 to 1500 m2). Multi-spindle rotary lawn mowers are used mainly for commercial applications on expanses of grass with dimensions greater than 1500 m2, such as lawns and gardens, road sides, sports fields, cemeteries, campuses, and golf courses. Noise Sources: Noise emission of a lawn mower is related to the driving source such as an IC engine and the cutting tool such as the rotating blade. A small percentage of the noise emission comes from other sources. Noise and Physical Law: Rotary lawn mower noise depends on the rotation speed of the blade which results in the cutting of the grass stem and its consequent transportation for collection. The minimum rotation speed of the blade tip in order to achieve this with an acceptable quality of cut is between 273 and 345 Km/hrs. Reducing Lawn Mower Noise: Single blade lawn mowers can be reduced by using an electric motor only. This, however, greatly reduces their Internoise 2001 Abstracts 61 geographical reach. Multi-spindle rotary mowers cannot be reduced by using an electric motor because they are used mainly in wide area applications. Large reductions in the noise emitted from larger area mowers are unfeasible and will destroy their utility. Session A5-1, 10:00 h Cosette Dussaugey Paper number 396 (Invited, Oral) I-INCE Class 82 Mtps/FEM, 39-41 Rue Louis Blanc, 92400 Courbevoie, France PRESENTATION OF THE GUIDELINES ON DIRECTIVE 2000/14/EC NOISE EMISSION IN THE ENVIRONMENT BY EQUIPMENT FOR USE OUTDOORS" The Fifth Environmental Action Programme identifies noise as one of the most pressing environmental problems in urban areas and the need to take action with regard to various noise sources and in its Green Paper "Future Noise Policy". The European Commission addressed noise in the environment as one of the main local environmental problems in Europe, The European Parliament and the Council of the European Union have worked with the Commission, to create and adopt a new directive that the Member States shall implement. That directive has been published in July 2000 for application 18 month later. This new directive supersedes 9 existing directives but it introduces new system of certification of the machines and various other aspects. The commission invited the member states and the industry to prepare guidelines to explain it and to ensure that it will be applied in a common way all over European Union. The guideline is addressed to the member states who have to implement the directive; to the public who need to be informed; to the industry who has to apply the requirements; to the notified bodies . The structure and main points of the guideline could be developed during this presentation. Session A5-1, 10:20 h Volker K.P. Irmer Paper number 421 (Invited, Oral) I-INCE Class 83 Umweltbundesamt, 14193 Berlin, Bismarckplatz 1, Germany THE GERMAN BLUE ANGEL AWARD AND THE USE OF LOW NOISE PRODUCTS The Directive relating to the noise emission in the environment by equipment for use outdoors states that Member States shall not prohibit, restrict or impede the placing on the market or putting into service in their territory of equipment which complies with the provisions of the Directive. If an equipment meets the noise limits and is properly marked with the sound power level guaranteed by the manufacturer this equipment may be sold in whole Europe. On the other hand Article 17 of the Directive states that the provisions of the Directive shall not prevent Member States to lay down measures to regulate the use of equipment in areas which they consider sensitive, including the possibility of restricting the working hours of the equipment. If, however, work must be continued, a solution may be that in these areas special low noise equipment may be used. The German Blue Angel Award may be an indication which products meet the requirements of ‘low noise’ equipment. 62 Internoise 2001 Abstracts Session A5-1, 10:40 h Paper number 627 (Contributed, Oral) John P.J. Oostdijk, Hans Kuypers and Jan H. Granneman Adviesbureau Peutz, Postbus 696, 2700 AR Zoetermeer I-INCE Class 11 ACOUSTIC ASPECTS OF THE DEVELOPMENT OF SILENT LOADING AND UNLOADING DEVICES Increasing traffic jam problems during the day in combination with the change of opening times of shops has enlarged the demand of supplying shopping centres during the evening and night. Since loading and unloading activities near adjacent residences can be very disturbing to inhabitants rather rigid peak noise limits are applied in The Netherlands. A research program ‘Peak’ has been developed by the governmental agency Novem to reduce the sound due to (un)loading trucks near shops. This program aims at measures at the sources, operational measures and measures at the location of (un)loading. Several feasibility studies regarding these aspects have been executed. In this paper the measures that have been examined at the source will be presented for a number of devices that are used during loading and unloading of trucks. The devices are the trailer, the cooling unit, the hydraulic loading flap and the containers. The measures that are taken to reduce the sound in different pilot plans will be described. The results of measurements before and after measures were taken will be presented. From the results of the measurements can be concluded that in many cases sufficient reduction can be reached by taking the appropriate measures. An exception has to be made for the manoeuvring of trucks. The economical aspects of the measures will be discussed. It is expected that on relatively short notice the measures will seriously be considered by companies who want to (un)load trucks during the evening and night. Session A5-1, 11:00 h Paper number 582 (Contributed, Oral) Jukka Tanttari and Ari Karjalainen VTT Automation, P.O. Box 1307, FIN-33101 Tampere FINLAND EXPERIMENTS ON A SCREW COMPRESSOR SOURCE PROPERTIES I-INCE Class 12 To understand fluid machines as noise sources, mathematical models of the source are needed. Usually it is not possible to derive these models purely theoretically, so experiments based on "black-box" models offer an alternative. In order to characterize the gas discharge of a twin-screw compressor as an acoustic source, the discharge was considered as a linear time-invariant one-port source. The experiments needed to the source description (the source volume velocity and -impedance) were based on the so-called two-load method, in which the source is assumed to be "deterministic". In this paper, the authors present their experiences and draw conclusions about the usability of the approach from point of view noise control engineering. Session A5-1, 11:20 h Paper number 767 (Contributed, Oral) Jean R. Jacques and Jean-Paul Lacore I-INCE Class 82 Institut National de Recherche et de Sécurité (INRS), Mission Normalisation, 30 rue Olivier Noyer, 75680 Paris Cédex 14, France INDICATIVE VALUES OF NOISE EMISSION IN SAFETY STANDARDS - A PILOT ACTION European directives 98/37/EC on machinery safety (first published in 1989) and 2000/14/EC on noise from equipment used outdoors (published in July 2000) require manufacturers to provide quantitative Internoise 2001 Abstracts 63 data on the noise emission of the machines they put on the European market. Such data are potentially extremely valuable because, for machines of a given family, they describe the state of the art regarding the noise emission of that particular family. A key issue is how these data could or should be used, in particular in safety standards. To investigate this, a pilot action was conducted on pneumatic screwdrivers available on the European market. It included collecting information about the market, selecting the machines to be tested, choosing industrial partners, test methods and machine parameters against which noise emission would be plotted, investigating various possibilities to process and condense the so-obtained clouds of noise emission data. The paper describes the different steps and difficulties met at each of them. It presents the pros and the cons of different possible ways to derive indicative values of noise emission and highlights precautions to take in determining such values for other families of machines. Session A5-2: Noise policy worldwide Chaired by Martin Van Den Berg and William Lang Session A5-2, 11:40 h M. Van den Berg Paper number 154 (Tutorial, Oral) I-INCE Class 82 Ministry of Environment, The Hague, The Netherlands THE IMPLEMENTATION OF THE EU-NOISE POLICY: PROBLEMS AND SOLUTIONS The consequences of the EU-Directive on noise policy for EU-Member States become more clear now that the draft Directive nears completion. The impact will be substantially different between Member States already having a noise policy and those without regulation. In both cases Member States will encounter similar problems, and so it may be helpful to exchange information on solutions. Examples of common problems of the "haves" are translation of limit values in the new indicators and adaptation of calculation methods, while for the "havenots" the collection of information and appointment of competent authorities may be a difficult first step. An overview of known transposition problems will be presented, followed by suggestions of generalized approaches which may reach from more technical to institutional. Session A5-2, 13:20 h William W. Lang Paper number 95 (Invited, Oral) I-INCE Class 82 29 Hornbeck Ridge, Poughkeepsie, NY 12603-4205, U.S.A. CURRENT NOISE POLICIES OF THE UNITED STATES OF AMERICA American noise policies represent a continuation of those that were adopted several decades ago, but the intention and will to implement these policies have changed. The Federal regulatory process involves the approval of a law by the U.S. Congress, followed by rule-making and enforcement by an agency of the Federal government. Noise control regulations may also be issued by State and local governments. The Aircraft Noise Control Act of 1968 directed the Federal Aviation Administration to establish standards and regulations that have resulted in significant reductions in the noise of individual aircraft and in the vicinity of airports. The Noise Control Act of 1972 assigned to the Environmental Protection Agency (EPA) the primary role in controlling environmental noise. This role was effectively terminated in 1982 when funding by the Congress to enforce the noise standards 64 Internoise 2001 Abstracts and regulations of EPA was withdrawn. The Quiet Communities Act of 1978 authorized EPA to provide financial assistance to State and local governments to initiate or expand their own noise control programs. Towards the end of the decade of the 1980s, all Federal funding was withdrawn and State and local programs were cut back severely. The Federal Highway Administration provides noise standards to guide State highway departments in planning and designing new highways. The Occupational Safety and Health Administration sets permissible noise exposures at the workplace, but enforcement today is lax. Except for aircraft, current U.S. noise policies have been largely ineffective in controlling the sources of environmental and occupational noise. Session A5-2, 13:40 h Lawrence S. Finegold Paper number 759 (Invited, Oral) I-INCE Class 82 Finegold & So, Environmental Noise Consultants, 1167 Bournemouth Court, Centerville, Ohio 45459, USA A NON-EUROPEAN VIEW OF RECENT NOISE POLICY ACTIVITIES IN EUROPE The European Union (EU) has embarked on a quite ambitious and sorely needed program to develop a series of Directives to manage environmental noise exposure. Two of the most important of the new European Commission (EC) regulations are the Outdoor Machinery Directive (2000/14/EC) and the new Draft Directive on Assessment and Management of Environmental Noise (COM (2000)468). These legislative documents, along with other existing EC Noise Directives, will certainly have a major positive impact on the noise exposure of the populations of European nations. These collective documents also provide the conceptual model needed to effectively manage environmental noise exposure throughout the world. However, in the short run at least, not all of the impacts of these highly commendable EC efforts will be positive for non-European countries. From the global noise management perspective, a dialog needs to be established between the EC and other organizations throughout the world to determine how to best achieve a balanced approach to noise control that takes into account the varying interests of all nations. In addition, socio-cultural differences between countries in various areas of the world will require different concepts for noise management. For example, in many countries, the "polluter pays" principle is not yet accepted. This paper discusses both the positives and negatives of the evolving set of EC Noise Directives from the broader international perspective. The overall view supports maximizing the effectiveness of what has been achieved in Europe, while taking into account the financial interests and socio-cultural differences existing in non-European countries. Session A5-2, 14:00 h Hideki Tachibana Paper number 143 (Invited, Oral) I-INCE Class 83 Institute of Industrial Science, University of Tokyo, Komaba 4-6-1, Meguro-ku, Tokyo 153-0041, Japan CURRENT NOISE POLICIES IN JAPAN In Japan, the national government enacts and enforces noise policies (laws, regulations and standards). They are within the jurisdiction of the Environmental Agency. The local governments legislate ordinances for the enforcement of the national regulations/standards and they can enact their own ordinances, i.e., noise regulations/standards to satisfy local requirements. In this presentation, the formulation of the national noise policies is introduced. The fundamental law, "Basic Environment Internoise 2001 Abstracts 65 Law" (1968, revised in 1993) defines basic principles and directions for environmental policies, and directs the government to establish Environmental Quality Standards (EQS) as administrative targets. According to this law, three EQSs have been established up to this time; "EQS for Noise" (1971, revised in 1998), "EQS for Aircraft Noise" (1973) and "EQS for Shinkansen Super-express Railway Noise" (1975). In the new version of the first EQS, which is mainly dealing with road traffic noise, the desirable noise level is specified for general living areas and roadside areas, and for two time categories of daytime and night-time, by the equivalent continuous noise level. The general noise regulation, "Noise Regulation Law" (1986, amended 1999) specifies the maximum permissible noise levels for factory noise, construction noise and motor vehicle noise. Besides these noise regulations and standards, an "Environmental Impact Assessment Law" was enacted in 1997 and enforced in 1998. This law sets forth procedures and other provisions to ensure that proper consideration shall be given to environmental protection issues relating to such large-scale development projects as roads, railways, airports, power plants. Session A5-2, 14:20 h Paper number 355 (Invited, Oral) Marion Burgess and Warren Renew I-INCE Class 83 Acoustics & Vibration Unit, School of Aerospace & Mechanical Engineering, University College, UNSW, Australian Defence Force Academy, Canberra, ACT 2600 AUSTRALIA APPROACHES TO ENVIRONMENTAL NOISE POLICY IN AUSTRALIA Australia is a federation of six states and two territories. Legislation for environmental noise is the responsibility of each of the Australian states and territories. The Federal government has the responsibility for national issues such as aircraft noise and also to encourage harmonisation of the legislation and regulations among the states and territories. For some decades there has been a document on environmental noise produced by Standards Australia but it is up to each state or territory to call up part or all of this Standard. For general environmental noise some states use comparison with background as the criteria while others define the criteria levels based on land use zones. Both approaches have their advantages and drawbacks. This paper will compare and contrast the different legislation and regulations and discuss the issue of 'cross border' disputes. Session A5-2, 14:40 h Selma Kurra Paper number 739 (Contributed, Oral) I-INCE Class 82 Division of Physical Environment, Faculty of Architecture, Istanbul Technical University, Turkey KEY PROBLEMS IN IMPLEMENTATION OF TURKISH NOISE CONTROL REGULATIONS AND ADAPTABILITY TO EU DIRECTIVES Noise control issues exist in Turkish legislation since 1926, however the articles concerning the different aspects of the problem are scattered widely in various regulations. The Noise Control By-law which came into force in 1986, comprises the detailed actions and incentive measures to be taken against environmental noise. After more than 10 years of experience, a recent survey conducted amongst all of the interested parts, has revealed the inefficient use of the By-law, urging a revision study also by providing conformity with the Directives and Standards of European Commission. In this paper, evaluation of the survey results is presented as key problems in implementation of the noise control legislation in Turkey, together with a summary of the report prepared as an action plan within national environmental strategies against pollution. Comparisons between the existing noise limits and 66 Internoise 2001 Abstracts those proposed in the documents, namely; Green Paper and the new document for the European Parliament related to environmental noise, are made and an approach including the basic principles for adaptability and reconciliation with regards to contents and terms is explained. POSTERS Session A5-2, 15:00 h Paper number 94 (Contributed, Poster) I-INCE Class 82 Posterboard AP-14 Tuesday, Author present 15:40 – 16:40 h Miguel Arana, Maria Luisa San Martin, Antonio Vela and Amando Garcia Acoustics Laboratory. Physics Department. Public University of Navarra. Pamplona, Spain A COMPARISON OF THE NOISE REGULATIONS PREVAILING IN SEVERAL CITIES AND AUTONOMOUS COMMUNITIES OF SPAIN In the last twenty years, many local governments and Autonomous Communities of Spain have developed regulations to control the environmental noise within the limits of their responsibility. At the present time, there does not exist any specific legislation on this subject for the whole country. All the prevailing regulations contain all the basic aspects of urban environmental noise, such as definition of the evaluation indices, equipment and measurement conditions, maximum limits permitted in different environments, and administrative or economic penalties. We have carried out a detailed comparative study of ten of these regulations, including all the largest cities of Spain. The main objective of the present study has been to harmonize the most positive aspects of the different regulations in order to accomplish an integrative proposal for a General Noise Law in this country, now in preparation. We have also carefully considered the effects of the future noise European regulation on the present contents of all these spanish regulations. The most important conclusion of our comparative study is that, even though the maximum permitted limits are very similar in all the considered regulations, the used noise indices and the measurement conditions show important differences. It has been also observed that most of the spanish noise regulations do not devote enough attention to the important aspect of acoustic isolation of buildings, as one of the key tools to control the environmental urban noise. Session A5-2, 15:00 h Paper number 99 (Contributed, Poster) Posterboard AP-15 Tuesday, Author present 15:40 – 16:40 h Y. Yasuda, K. Furihata and T. Yanagisawa I-INCE Class 82 Department of Electrical and Electronic Engineering, Faculty of Engineering, SHINSHU UNIVERSITY, 4-17-1 Wakasato, Nagano 380-8553, Japan VERIFICATION OF SAFETY REGULATIONS TO FURTHER REDUCE THE NOISE GENERATED BY INDIVIDUAL MOTOR VEHICLES In Japan, many locations (areas facing roads) do not attain environmental quality standard values (70dB or less for daytime, and 65dB or less for night-time), mainly places located in major urban regions and along main roads. For this reason, it is concerned with ways to further reduce the noise generated by individual motor vehicles. From this aspect, first, noise rating scales for the amount of noise from individual vehicles, which can be evaluated not only physically (LAE and LAmax [dB]) Internoise 2001 Abstracts 67 but also psychologically (annoyance), were composed experimentally with the road vehicle noise simulator proposed in the previous paper (inter-noise 99). Secondly, to attain the above standard values, safety regulation values were verified in Nagano City. From the noise rating scales obtained, it can be found as follows: (a) In the case of steady running noise, ‘little annoying’ corresponds to LAE=76dB (LAmax=74dB) and ‘annoying’ to 82dB (80dB). (b) In the case of accelerated running noise, ‘little annoying’ corresponds to 69dB (65dB) and ‘annoying’ to 78dB (74dB). Next, from the results (each LAeq (10min) is from 71.0dB to 74.4dB at a straight road, and 68.3dB to 71.7dB at a crossroad) measured at 7.5m from a microphone to a traffic line for 24 hours on weekdays, it can be predicted that, to attain the above standard values for all times and anywhere, it is necessary to regulate all of individual motor vehicles under LAmax=74dB. Consequently, it can be said that all safety standards should be implemented 74dB or less. Session A5-2, 15:00 h Paper number 242 (Contributed, Poster) Posterboard AP-16 Tuesday, Author present 15:40 – 16:40 h Y. Mishina, T. Ogura, A. Hayashi, M. Omiya and K. Kuno Daido Institute of Technology, Nagoya 457-8530 Japan I-INCE Class 83 THE NEW AND THE OLD STANDARDS IN THE ENVIRONMENTAL NOISE MEASUREMENT IN JAPAN The standards for the environmental noise in Japan had been measured and estimated based on the median values of sound pressure levels, L50, for many years. However, they were revised in September 1998 and enforced in April 1999. Since then, the measurement and the estimation based on the equivalent sound pressure level, LAeq, have been used. In this report, the status quo about the achievement of standards for environmental noise has been analysed by the new and the old estimations, namely L50 and LAeq. At the same time, the standards for areas adjacent to the roads (the space adjoining the main traffic road), which have been newly defined this time, are also analysed using the existing research data of over 2000 samples which have been collected in Nagoya City. The population of Nagoya City is about 2 millions. As the result, we have come to the conclusion that the achievement ratios of the standards for environmental noise based on LAeq is about 53%. We have also found out that about 24% of residents is disturbed by the noise caused by the traffic in case of areas adjacent to the roads. Namely, the traffic noise level is over the standards in these areas. On the other hand, the problems concerning the measurement based on LAeq are also presented to a certain level in this study. Session A5-2, 15:00 h Paper number 384 (Invited, Poster) Posterboard AP-17 Tuesday, Author present 15:40 – 16:40 h D.-J. Kang and Y.-H. Kim NOVIC, Dept. of M.E., TAEJON 373-1, Korea NOISE POLICIES IN REPUBLIC OF KOREA I-INCE Class 83 The Korean constitutional law provides that all the citizens have the right to enjoy a healthy and pleasant environment and the government and the citizens have to try their best to preserve it. According to the constitutional law the Environmental Pollution Prevention Act was first established in 1963 and it was revised as the Environmental Conservation Act in 1977. However this act was ineffective in controlling the various aspects noise and vibration problems since it regulates all 68 Internoise 2001 Abstracts different types of environmental issues such pollution of air and water resources and conservation of natural resources. Hence it was subdivided in to detailed categories in 1990 including the Organic Environmental Policy Act, the Air Pollution Control Act, the Noise and Vibration Regulation Act, the Aircraft Noise Control Act, the Industrial Safety and Health Act, the Housing Construction Expediting Law. According to the Organic Environmental Policy Act the Minister of Environment can launch the long-term environmental conservation with the approval of the Cabinet council and the noise and vibration subcommittee is organized to provide technical advice. Majority of noise-related issues are covered by the Noise and Vibration Regulation Act that extends to workplace noise and vibration, traffic noise, automobile noise, environmental noise, etc. The occupational noise, building noise and vibration, and aircraft noise are regulated according to the Industrial Safety and Health Act, the Housing Construction Expediting Law, and the Aircraft Noise Control Act, respectively. Session A5-3: Vibro-acoustic sources Chaired by Yuri I. Bobrovnitskii and Mogens Ohlrich Session A5-3, 15:20 h Mogens Ohlrich Paper number 608 (Tutorial, Oral) I-INCE Class 41 Ørsted-DTU, Acoustic Technology, Technical University of Denmark, Building 352, DK-2800 Lyngby, Denmark STRUCTURE-BORNE SOUND SOURCES AND THEIR POWER TRANSFER A review is given of the current development stage of methods for characterising structure-borne sound sources. The paper focuses on the characterisation of vibrating machinery sources and on estimating the vibrational power they inject into foundations and receiving structures. Because of the high complexity of most machinery structures and their inherent excitation mechanisms the vibration source strength generally has to be determined from measurement. Practical receiving structures are also very complex and their input properties often have to be measured if reliable estimates of the power transmission are to be made. However, it has been found that a course prediction of the injected power can be made by using simple, approximate expressions for the foundation mobilities. Session A5-3, 16:00 h Yuri I. Bobrovnitskii Paper number 324 (Invited, Oral) I-INCE Class 41 Mechanical Engineering Research Institute, M.Kharitonievsky Str., 4, Moscow 101990, Russia MODELS OF ACOUSTIC SOURCES: A SURVEY Models of acoustic sources used in the literature for noise prediction are surveyed. Most attention is paid to mid frequency range models based on representation as a superposition of elementary sources. These are: (1) models of the Neumann and Dirichlet types; (2) modelling by the secondary sources (the Huygence principle), and (3) modelling by equivalent sources. Advantages and restrictions of the models are discussed. A model that is the most adequate for treating the noise prediction problem at middle frequencies is described. Computer simulation results are presented. Internoise 2001 Abstracts 69 Session A5-3, 16:20 h Paper number 226 (Invited, Oral) Pieter Van Beek, Martijn Termeer and Christ De Jong TNO TPD, P.O. Box 155, 2600 AD, Delft, The Netherlands I-INCE Class 41 AN EXPERIMENTAL AND NUMERICAL STUDY OF THE APPLICABILITY OF STRUCTURE-BORNE SOUND SOURCE DESCRIPTORS Structure-borne sound power transmission from a vibrating machine to its foundation structure is a complex process that involves multiple contact points and multiple force components and is influenced by the structural mobilities of both machine and foundation. This explains why a general and practicable characterisation of structure-borne sound sources is still lacking. Following Mondot and Petersson, Moorhouse has recently suggested some sound power based source descriptors. This paper describes a combined numerical and experimental study of the applicability of these descriptors. Two cases have been studied: simplified models of a diesel engine and an electric motor. The power transmission and source descriptors are written in terms of free velocities (internal vibration mechanism) and mobilities (structural properties). It is demonstrated that multi-point and multicomponent effects are important for a correct characterisation of both sources. Measurements of the free velocities of the source at the interface positions are necessary for a correct description of the internal vibration mechanisms. On the other hand, it is difficult and cumbersome to measure large quantities of mobilities (especially for rotational degrees of freedom). Therefore, it is suggested that an approach of using measurements in combination with numerical (FEM) model calculations is promising, although a good match between experimental data and the numerical model is necessary. Session A5-3, 16:40 h Vladimir I. Popkov Paper number 331 (Invited, Oral) I-INCE Class 41 Krylov Shipbuilding Research Institute, Moskovskoe shosse, 44,196158, St. Petersburg, Russia CHARACTERISTIC OF MECHANISMS AS THE SOURCES OF VIBRATION The characteristics of mechanisms as the sources of vibration are necessary for objective monitoring and comparing between each other the mechanisms with different weights and principles of operation. These characteristics are also necessary for decreasing the vibration of mechanisms with attached structures and vibroisolators. In this report the possible ways of mechanisms characteristic as the sources of vibration are considered: – by the complex forces or velocities of vibration at the region of working units and also by transfer impedances and mobilities of mechanism structures; – by the complex forces or velocities of vibration at the region of fastening the mechanism to supporting and service structures and also by input, transfer impedances and mobilities of mechanism structures; – by complex oscillatory powers and also generalized effective forces, root mean- square vibration and effective mechanical impedances of mechanism structures. The fields for the optimum using the different ways of mechanisms characteristic as the sources of oscillations when determining their vibration jointly with different foundations and vibroisolators attached to mechanisms. The ways for monitoring vibration of mechanisms and efficiency of means for vibroisolating and decreasing vibration of foundations are presented. 70 Internoise 2001 Abstracts Session A5-3, 17:00 h Paper number 532 (Invited, Oral) I-INCE Class 41 Andy Moorhouse, Charles Pezerat, Goran Pavic and Thomas Allevard Acoustics Research Unit, University of Liverpool, Liverpool L69, 3BX, UK INDEPENDENT CHARACTERISATION OF A STRUCTURE-BORNE SOUND SOURCE USING RIFF TECHNIQUES RIFF techniques have previously been used to measure the forces at the interface of a vibrating machine mounted on a steel plate. In this paper a variation of the method is used to measure independent properties of a source. The independence is achieved by carrying out two measurements, the first with the machine in normal operation, and the second with it switched off whilst still mounted on the plate. In this way, the effect of the plate on the source could be removed mathematically. The independent source properties are the free velocity, locked force and characteristic power (the product of free velocity and blocked force). Results of some numerical simulations are presented, followed by actual measurement results. Session A5-3, 17:20 h T.M. Tomilina Paper number 339 (Invited, Oral) I-INCE Class 41 MODELING FORCED VIBRATING STRUCTURES BY EQUIVALENT ELEMENTARY SOURCES Mechanical Engineering Research Institute, M.Kharitonievsky Str., 4, Moscow 101990, Russia An efficient algorithm for analysis of the sound field radiated from elastic forced vibrating structure of complicated geometric shape is presented. It is based on modelling the structure by a set of elementary (point, ring, or line) sources located inside (or outside) the structure. A procedure of optimization of the algorithm parameters is given. Computer simulation examples are presented on predicting the interior noise of an elastic forced vibrating shell as well as the exterior noise of composite structures. The results demonstrate high efficiency of the algorithm. Session A5-3, 17:40 h Paper number 170 (Contributed, Oral) Anand N. Thite and David J. Thompson I-INCE Class 43 Institute of Sound and Vibration Research, University of Southampton, Highfield, Southampton SO17 1BJ, UK USE OF REGULARIZATION TECHNIQUES TO IDENTIFY STRUCTURE-BORNE NOISE SOURCES The transfer path analysis technique has wide applications in the automotive industry as well as other sectors. It relies on the identification of structure-borne and air-borne noise sources, which is often an inverse problem. The strong modal behaviour of structures, combined with errors in the measurement of operational responses and frequency response functions, can result in large errors during the matrix inversion. These errors can be minimised by employing regularization techniques while inverting the frequency response function matrices. Regularization techniques have been employed for a long time in the field of digital image processing and more recently in relation to nearfield acoustic holography to improve source identification. The structure-borne noise source identification differs from these applications due to the modal behaviour of the structures. In this study singular value rejection, Tikhonov regularization and iterative inversion techniques have been used in identifying coherent forces on a flat rectangular plate. The different threshold criteria for singular value rejection are compared. The ordinary cross validation technique is used to choose the regularization parameter in Internoise 2001 Abstracts 71 Tikhonov regularization. In the iterative solution the expressions for bias and random errors are used to arrive at a optimal iteration number which compromises these two errors. All these techniques are compared and their relative advantages are evaluated. Session A5-3, 18:00 h Paper number 430 (Contributed, Oral) John D. Fieldhouse and Carl Beveridge The University of Huddersfield, Queensgate Huddersfield HD1 3DH England I-INCE Class 76 NOISE REDUCTION OF A TWIN CALLIPER DISC BRAKE BY ANGULAR POSITIONING OF THE CALLIPERS A noisy disc brake vibrates in a wavelike manner with a diametral mode of vibration. The mode shape exhibits antinodes with angular spacing of a specific size related to the noise frequency and mode order. The sizes of the antinodes vary slightly with brake pressure which causes a slight change in noise frequency. If the angular spacing of the callipers is arranged such that an integer of antinodes between the callipers is not possible then the disc is inhibited from vibrating in a given mode and noise at that frequency is prevented. The paper demonstrates that at particular angular settings of the callipers certain frequencies exist yet as the angular settings are changed the disc mode order and related frequency changes. It is shown that the frequencies can be related to the angular spacing between the callipers and the free mode angular spacing of the disc waveform antinodes. An analysis of the geometry allowed the brake noise frequency to be predicted for a certain angular setting of the callipers and specific angular settings to be predicted when the brake would be quiet. It is shown that this analysis was successfully applied in practice on the test rig and on the problematic vehicle. It is shown that the theory may be extended to single calliper brakes and adjustment of the pad/disc interface geometry. Session A5-3, 18:20 h Paper number 560 (Contributed, Oral) Andreas Meier and Manfred Schmidt Müller-BBM GmbH, Munich, Germany PREDICTION OF NOISE IN ROOMS CAUSED BY DRILLING IN BUILDINGS I-INCE Class 76 Noise by drilling in buildings often leads to annoyance of the inhabitants. It is known, that the attenuation of this noise is low, when it is transmitted over several floors from the point of drilling. In buildings with high acoustical requirements, e.g. in hospitals or large office buildings, it is of interest to use drilling techniques with low noise emission. During a consulting the noise emission and propagation of drilling machines was examined. In particular the purpose of the client was to get information about airborne sound levels in rooms of buildings when different drilling techniques are applied. This contribution focuses on a comparison of different prediction methods for the transmission of drilling noise in buildings. It takes into account methods according to EN 12354 and SEA methods and considers the limits of their application. Measurements carried out will be compared to results calculated by different prediction methods. 72 Internoise 2001 Abstracts POSTER Session A5-3 Paper number 650 (Contributed, Poster) I-INCE Class 41 Posterboard BP-4 Wednesday, Author present 9:40 – 10:40 h Wim Hendricx, Herman Van der Auweraer, Alberto Pezzutto and Francesca Garesci LMS International NV, Interleuvenlaan 68, B-3001 Leuven (Belgium) A STRUCTURAL DYNAMICS STUDY OF BRAKE SQUEAL Friction-induced vibration may cause excessive, undesired noise. Brake noise, in particular brake squeal, is a typical example. This problem is investigated in the EUREKA project Brake Noise. Operational measurements have shown that brake squeal goes along with stationary deformation shapes of the rotating disc. This means that, while rotating, the nodal lines of the disc deformation are standing still with respect to an immobile observer. However, referring to the rotating disc, these nodal lines are then moving along the disc. A possible explanation is that during rotation and braking, different modes are switching and exchanging energy. The energy input to sustain the vibration is provided by the friction force. This phenomenon shows some similarities with aerodynamic flutter where an instability of an airplane wing can be caused by the continuous switching between a bending and a torsional mode of the wing. A simplified test stand (rotating disc, 1 brake pad) was developed to get physical insight into the fundamental dynamics of the friction contact and of the resulting behaviour of the brake components and their interaction during braking. Experimental modal analysis as well as numerical (FE) models of the different parts of the set-up were developed and related to deformation data measured during squeal. The importance of the dynamics behaviour of the surrounding assembly components was assessed and found to be significant. A multibody simulation model including the models for the flexible parts was developed and allowed to predict the generation of squeal vibrations. Session A6-1: Application of new measurement technologies to building acoustics Chaired by Hideki Tachibana Session A6-1, 9:40 h Michael Vorländer Paper number 609 (Tutorial, Oral) I-INCE Class 72 Institute of Technical Acoustics, Technical University Aachen, D-52056 Aachen, Germany CATEGORIZATION OF MODERN MEASUREMENT TECHNIQUES IN BUILDING ACOUSTICS In the past decades several new approaches for measurement of reverberation time and sound levels in buildings have been developed. Partly these techniques were known from measurements in auditoria or in electroacoustic applications, for instance. However, the practical circumstances in measurements of sound insulation in the laboratory and in the field require detailed investigation of the applicability of the techniques. At first, not much information was available concerning time variance effects like wind or temperature drift while measuring sound insulation of facades. On the other hand, MLS and Internoise 2001 Abstracts 73 sweep techniques were applied in noisy environments with great success due to their rejection of uncorrelated noise. In the meantime several authors have reported theoretical and experimental results of the benefits and the limitations of modern digital techniques. This contribution is an approach for a classification of modern techniques (MLS, FFT, Sweep, Time-stretched pulse ) regarding efficiency (excitation length, processing time), signal to noise ratio, immunity against stationary/impulsive background noise, non-linearities and time variances. Session A6-1, 10:20 h Werner Scholl Paper number 690 (Invited, Oral) I-INCE Class 73 Fraunhofer Institut fuer Bauphysik, Nobelstrasse 12, D-70569 Stuttgart, Germany A MODIFIED TAPPING MACHINE TO REPRESENT A WALKING PERSON In structure borne sound a strong interaction of sound sources and excited structures may take place. That means that the difference of input forces of two different sources (e.g. a walking person and a standard tapping machine) acting on the same structure may not only depend on the sources but also on that structure. It also means that the rank order of acoustical performance (e.g. impact sound insulation) of structures may depend on the type of sound source. Regarding the impact sound pressure level of a person walking on a floor, one approach to get realistic measurement results with respect to the walking noise would be to use an artificial impact source which the same dynamic characteristics as the walker. Measurements of the source impedance of walking people showed, that it can be simulated by a mass-spring-mass system which can be realised by modifying the existing ISO 140 standard tapping machine. Comparisons of the impact sound pressure level of a walking person, the standard tapping machine and the modified machine on a concrete and a wooden joist floor showed that the modified tapping machine fits the walker's results much better than the unmodified tapping machine. The modification will become an informative part of the ISO 140-11 (laboratory measurement of impact sound improvement on lightweight floors) to make it possible to collect more experience. Session A6-1, 10:40 h Paper number 556 (Invited, Oral) Hideki Tachibana, Hiroo Yano and Fumiaki Sato Institute of Industrial Science, University of Tokyo Komaba 4-6-1, Meguro-ku, Tokyo 153-0041, Japan I-INCE Class 51 SOUND INSULATION MEASUREMENT BY VARIOUS KINDS OF DIGITAL SIGNAL PROCESSING TECHNIQUES In the measurements of sound insulation properties of building walls/facades using an artificial sound source, the measurement accuracy is often deteriorated by the influence of background noise in the field. To reduce this problem, various kinds of digital signal processing techniques are contrived and applied to improve the signal/noise ratio. In this paper, the applications of the direct cross-correlation method, M-sequence modulation cross-correlation method, synchronized integration method, MLS method and Stretched Pulse Method are introduced and their relative merits are discussed. 74 Internoise 2001 Abstracts Session A6-1, 11:00 h Takayuki Hidaka Paper number 630 (Invited, Oral) I-INCE Class 51 Takenaka R&D Institute, 1-5-1, Otsuka, Inzai, Chiba, 270-1395 Japan CURRENT MEASURING METHODS FOR OBJECTIVE PARAMETERS IN ROOM ACOUSTICS This paper is a review of recent progress made in the field of room acoustical measurement. A large number of researches and experiences have led to identification of series of significant objective parameters confirmed to be contributing to acoustical quality evaluation to some extent. Furthermore, remarkable advancement of digital devices now allows high-speed and high-accuracy execution of measurement techniques in every genre of acoustics. Accordingly, the conventional room acoustical measurement has been replaced by the measurement of exact impulse response in a room, and various objective parameters are numerically calculated on the relevant software. The discussion in this paper is thus presented based on this scheme. To begin with, the survey results of recent studies are provided. Then, based on the author's own experiences in this field is presented a report on measurement technologies to realize efficient or labor-saving implementation of acoustic measurement, i.e., measuring equipment and systems, practical analysis procedures, digital signal processing methods, and statistically efficient number of source locations and receiver positions. In addition, several essential issues that arise from the very application of new technologies are discussed. Lastly, the paper outlines application of the similar room acoustical measurement technique to 1/10 scale model experiments. Session A6-1, 11:20 h Paper number 637 (Invited, Oral) K. Inoue, M. Yasuoka and H. Tachibana I-INCE Class 51 College of Science & Technology, Nihon University, Surugadai 1-8-14, Kanda, Chiyoda-ku, Tokyo, Japan REDUCTION OF FLOOR IMPACT SOUND ON FLOOR FINISHING FOR WOODEN FLOOR STRUCTURE We are studying the measuring methods of the reduction of floor impact sound on floor finishing materials (structure) for the standard Wooden floor structure in ISO TC/43,SC-2,WG-22. This is the experimental measuring method using reverberation rooms. In ISO-WG, three types standard wooden floor structures is offered for measurement. In this paper we studied on the reduction of floor impact sound by floor vibration characteristics using tapping-machine and heavy and soft impact source. Session A6-1, 11:40 h Hans G. Jonasson Paper number 408 (Invited, Oral) I-INCE Class 51 SP Swedish National Testing and Research Institute, SE-50115 Borås, Sweden A SIMPLIFIED METHOD TO DETERMINE IMPACT SOUND IMPROVEMENT ON LIGHTWEIGHT FLOORS A new simple method to determine the impact sound improvement of floor coverings on lightweight floors has been tested with promising results. Instead of testing the covering on a complete floor it is tested only on a superstructure mounted directly on the traditional concrete slab used for tests according to ISO 140-8. The idea is to simulate the upper part of the floor only as the most important factor reducing the improvement relative the one obtained on a concrete slab is the resiliency of the Internoise 2001 Abstracts 75 upper panel of the floor construction. Some tests are reported and good agreement is obtained compared to simple theory modelling the floor covering as a spring and the resilient panel as a point impedance. There is a possibility that this simple method could be an alternative to the more complicated method proposed by ISO TC 43/SC 2 WG 22. In this proposal the requirement is to build a complete light-weight floor for the test. There is a risk that the high cost will prevent people from using such a method. Session A6-1, 12:00 h Paper number 438 (Contributed, Oral) Noboru Nakasako and Shin-ichi Nakagawa I-INCE Class 74 School of Biology-Oriented Sc. & Tech., Kinki Univ., Nishi-Mitani 930, Uchita-cho, Naga-gun, Wakayama 649-6493 Japan IDENTIFICATION AND PROBABILISTIC PREDICTION METHODS FOR ACTUAL SOUND INSULATION SYSTEMS USING INDEPENDENT COMPONENT ANALYSIS ON THE BASIS OF NON-STATIONARITY This paper describes a trial on the identification and probabilistic prediction for sound insulation systems contaminated by a background noise with unknown statistics, especially from a practical viewpoint. Since the sound source of background noise is quite different from that of objective sound, Independent Component Analysis (ICA) can be introduced to evaluate the effects of the objective noise and background noise separately and to identify the sound insulation systems. Usually, there is a time delay in an actual sound field at each observation point and are many reflections from the floor, walls, etc. It was pointed out that the performance of separation becomes worse when the observation signals are composed of source signals with time delays or with the effect of the room (i.e., convolutive mixture). By paying our attention to the fact that a sound fluctuation in the field of noise evaluation is measured through a sound level meter with time averaging operation and the environmental noise very often fluctuates non-stationarity, ICA based on the non-stationarity can be adopted to estimate the sound insulation systems. That is, by considering the sound insulation wall as a linear system on an intensity scale, the system parameters are estimated on the basis of statistical independency and non-stationarity. Then, by using the estimated parameters of the sound insulation system, the output probability can be easily predicted when the other input noise passes through the same insulation system. Finally, the validity and effectiveness of the proposed method are experimentally confirmed, by applying it to the actually observed data. Session A6-2: Fan noise Chaired by Jean Tourret Session A6-2, 13:20 h Göran Kårfalk Paper number 557 (Invited, Oral) I-INCE Class 14 ABB Ventilation Products AB, S-351 87 Växjö, Sweden LOW NOISE DESIGN OF AXIAL FANS A test rig for testing possible measures to design an axial fan with low noise was set up. All measurements were made in 1/24-octave bands to be able to follow all resonances. The performance in the form of flow and fan pressure was measured and the comparisons of sound are made as Specific 76 Internoise 2001 Abstracts Sound Power Level. In the project the importance of resonances in the axial direction in a duct of limited length is demonstrated. This is important also for testing according to standards. The question whether the fan is a velocity or a pressure source is raised. Further the modes in the section are important and the fan design, blade number, diameter and speed, should be made considering this. The incoming turbulence will create noise when hitting the fan wheel. The noise production depends on the blade shapes and especially the tip clearance is important at high frequencies. When comparing Sound Power Levels it is important to have a good performance and well designed guide vanes will improve this. Session A6-2, 13:40 h Paper number 409 (Contributed, Oral) Jeonghan Lee, Jaekwon Ahn and Soogab Lee Seoul National University, SEOUL 151-742, Korea I-INCE Class 11 ANALYSIS AND OPTIMIZATION OF AERODYNAMIC NOISE IN AUTOMOTIVE ENGINE-COOLING FANS Response surface method is employed in optimizing the acoustic performance of automotive enginecooling axial fans. The effects of modifications in blade geometry on noise reduction are investigated. Taking the far-field noise level as the objective, a quadratic response surface is constructed utilizing D-Optimality condition as the candidate-points selection criteria. It is shown that the quadratic model exhibits a reasonable fitting capability resulting in the blade design with low far-field noise level. Session A6-2, 14:00 h Paper number 462 (Contributed, Oral) A. Karjalainen and J. Hyrynen I-INCE Class 14 Tampere University of Technology, Dept. of Mathematics, PO Box 692, FIN 33101, Tampere, Finland EXPERIMENTS ON DESIGN PARAMETERS AFFECTING COOLING FANS IN AN INDUCTION MACHINE. PART 1: SPECIFIC SOUND POWER LEVEL An experimental investigation was carried out in order to find out a practical minimum of noise emission for an induction machine cooling fan construction. The main aim was to detect the most influencing design parameters on the noise emission and also the aerodynamic performance of the experimental set up. The set up consisted of an induction machine mock up including an axial fan and it's cover section. The mock up was attached to a pressure plenum, which was placed in a semianechoic chamber. In order to acquire pressure rise, volumetric flow and sound power for a sufficient amount of operating points on the fan curve, the test chamber was provided with an auxiliary fan that was situated outside the semi-anechoic chamber. Based on literature surveys, manufacturing limitations and standards the following parameters chosen for the study were blade wheels, tipclearance, impeller axial positioning inside inlet cone, shape and length of inlet cone and the distance between the impeller and motor. Due to a large number of parameters and individual tests a method known as DOE (Design of experiments) was applied to reduce the number of measurements. Results from this approach are reported in a separate article. The acoustic and aerodynamic results are presented in the form of specific sound power levels as a function of flow. The results indicate that the most import parameter affecting specific sound power level is tip clearance. Internoise 2001 Abstracts 77 Session A6-2, 14:20 h Paper number 526 (Contributed, Oral) I-INCE Class 14 Kyungseok Cho, Soogab Lee, Changho Park and Taeyoung Park Center for Environmental Noise & Vibration Research (CENVR) School of Mechanical and Aerospace Engineering, Seoul National University, Seoul, Korea CORRELATION OF NOISE AND BLADE SURFACE PRESSURE IN AUTOMOTIVE ENGINECOOLING FAN Automotive engine cooling fan is one of the major sources of automobile noise. The most annoying features can be characterized by the discrete frequency noise at the blade passing frequency. These tonal noises are generated by blade interaction with inflow and physical configuration such as shroud. Measurements of the blade surface pressure on the rotating blade are conducted and correlated with the far field sound pressure. The results show good coherence between surface and sound pressure. Power spectral density of the surface pressure reveals that there exists some broad hump, which associated with broadband noise. Session A6-2, 14:40 h Paper number 115 (Contributed, Oral) K. Yanagimoto, S. Watanabe, T. Ito and R. Ichimiya I-INCE Class 34 Tsuruoka National College of Technology, 104 Sawada Inooka, Tsuruoka Yamagata, 997-8511, Japan PASSIVE CONTROL OF TONAL FAN NOISE BY USING CIRCULAR SLIT RESONATOR WITH ADDITIONAL NOISE SOURCE This paper describes a passive noise control technique to suppress the blade tones of axial flow fans by using circular slit resonator. The circular slit resonator with ventilation aperture composed of thin air layer that is radial expansion with very short axial length was set up machinery wall. The circular thin air layer acts as an acoustic resistance, inertance and capacitance. The noise was reduced greatly at a natural frequency of the circular thin air layer that shows symmetrical circular mode of standing wave. The insertion loss, IL, of the circular slit resonator indicates narrow peak at resonant frequency that is not harmonic and it strongly depends on the slit depth and slit width. However, it is expected that additional secondary noise sources, which oscillate the circular thin air layer, will absorb the harmonic fan noise. The circular slit resonator that is designed to reduce a fundamental blade rotational noise f1[Hz] was set on the inflow side of the axial flow fan casing. Secondary sound sources (four loudspeakers) are mounted into the circular membrane of the slit resonator. These sources are driven with electrical signals that synchronized with the rotation of the fan blade, and their amplitudes and phase are adjusted to give maximum reduction for the blade tone levels. The experimental results show that the target harmonic frequencies could well be attenuated by using the circular slit resonator with additional secondary noise sources. 78 Internoise 2001 Abstracts POSTER Session A6-2 Paper number 464 (Contributed, Poster) Posterboard AP-18 Tuesday, Author present 15:20 – 16:20 h J. Hyrynen and A. Karjalainen VTT Automation, PO Box 1307, FIN 33101, Tampere, Finland EXPERIMENTS ON DESIGN PARAMETERS AFFECTING COOLING FANS IN AN INDUCTION MACHINE. PART 2: RESULTS FROM DESIGN OF EXPERIMENT I-INCE Class 14 An experimental investigation has been carried out in order to find out a practical optimum for an induction machine cooling fan construction. The main task was to detect the most influential design parameters on the noise characteristics and also the aerodynamic performance of the experimental set up. Results from plain specific sound power level approach are presented in an other paper. Due to a large amount of parameters and individual tests, the introduction of a method known as DOE (Design of Experiments) was carried out. Four structural factors were chosen to vary in the experiments. These factors were tip-clearance, impeller axial positioning inside inlet cone, shape and length of inlet cone and the distance between the impeller and motor. The specific sound power level of the fan set up was regarded the most interesting response in the DOE-method, but also the effects on both the flow coefficient and pressure coefficient were investigated. The DOE method has a benefit of being able to create a prediction model of the effects of the chosen factors on responses. The effect of one of these combinations was verified by doing an additional experiment at this particular location in the experimental space. Another benefit of using DOE is to evaluate the importance of the factors, in some cases it is even possible to eliminate some. This is very helpful when designing further experiments. The results showed that varying the values of factors had a great influence on the responses, especially the specific sound power level. The most influential factor seemed to be the tip clearance. Session A6-3: Vehicle interior noise Chaired by Herman Vanderauweraer and Samir N.Y. Gerges Session A6-3, 15:00 h Paper number 569 (Tutorial, Oral) P.J.G. Van der Linden, K. Wyckaert and H. Van der Auweraer LMS International, Interleuvenlaan 68, 3001 Leuven, Belgium MODULAR VEHICLE NOISE AND VIBRATION DEVELOPMENT I-INCE Class 76 Fundamentally, vehicles are assemblies of subsystems or modules; engine, tyres, dashboard, steering system, body structure, wheel suspensions, etc. These modules are developed in parallel at multiple locations and often for more than one vehicle. On top of this, the integration of modules in a complete vehicle is also becoming a parallel rather than sequential development activity. This has a strong impact on vehicle NVH development. The noise and vibration (NVH) quality of a vehicle will become apparent not too long before its planned production start, on the first fully trimmed and equipped prototype vehicles. Under normal costs and product release pressure, this does not leave much scope for further NVH development. The obtainable NVH quality therefore relies strongly on the specification and assessment process of the subsystems as well as on the early part of the integration Internoise 2001 Abstracts 79 process. Recent improvements of substructuring and transfer path analysis techniques can contribute significantly to the reliability and practicality of modular NVH assessment and development. This does not only cover the individual subsystem design, but also the assessment of proposed packaging and subsystem interfacing solutions. The accuracy of modular NVH assessment is improved by these techniques in two ways: 1. Top-down; from vehicle behaviour to required subsystem behaviour and parameters. The transfer path analysis techniques for airborne and structure borne effects can be used in many situations to derive adequate subsystem NVH requirements from complete vehicle NVH requirements; 2. Bottom-up; estimating the vehicle NVH response from subsystem tests and numerical models. Modal based synthesis and frequency response function based synthesis can be used to build assemblies of subsystems. In situations where the subsystem’s dynamics and/or acoustics are strongly coupled, individual subsystem requirements (top-down) will not necessarily ensure a good assembled system behaviour. Substructuring can produce estimations of the NVH properties by combining data from existing and newly proposed subsystem solutions. The paper will give an overview of the methods and some test-based case examples are used to clarify the procedure. Session A6-3, 15:40 h Paper number 723 (Invited, Oral) I-INCE Class 33 Otto Von Estorff, Martin Wandel, Rüdiger Rezek and Stefan Schirmer TU Hamburg-Harburg, Eissendorfer Str. 42, D-21073 Hamburg, Germany MODELING OF MULTI-LAYERED PORO-ELASTIC PANELS FOR VEHICLE INTERIOR ACOUSTICS The paper concentrates on the modeling of multi-layered poro-elastic panels which are widely used for noise control and sound insulation purposes. The panels consist of a combination of porous and elastic/visco-elastic layers. Each layer is discretized by special finite elements. These rely on particular choices of basic variables (displacement, stress and pressure) and have specific features leading to variable modeling options. After a brief summary of the calculation process, a multi-layered panel consisting of a heavy visco-elastic layer, a porous layer, and an elastic steel plate is investigated in detail. Particularly, combining the structural finite elements with a boundary element model which represents the surrounding air, the sound radiation of the panel is considered. Emphasis is placed on the comparison of the calculated results with measurements. It can be observed that the agreement in a wide frequency range is very good. Some minor differences due to a phase shift occur. It is clearly shown that a proper modeling of the porous material is essential. Session A6-3, 16:00 h Paper number 543 (Invited, Oral) Virginie Linet, Arnaud Lepage and François Van Herpe I-INCE Class 76 PSA Peugeot Citroën - Direction Recherche et Innovation Automobile, Route de Gisy, F- 78 943 Vélizy Villacoublay, France IMPROVEMENT OF BODY PANEL FE MODELS THROUGH VIBRATION AND 3D-SHAPE IMAGE MEASURING METHOD The noise reduction is a major concern of car development. The use of predictive calculation methods during the development process is now a common practice, allowing theoretically to save time and costs and in the same time to enhance the vehicle acoustical quality. In the low frequency range (0200Hz), the noise is especially transmitted to passenger ears through the body panels vibrations (firewall, floor, roof, doors, windscreen, and ..) that can be coupled to the acoustical cavity giving rise to booming noise. This coupling effect in now well understood and calculation methods based on 80 Internoise 2001 Abstracts boundary finite elements exist already and works well assuming that the structural models are reliable. This is indeed a critical point. Very often, the predictive calculation of frequencies and mode shapes shows great differences with the actual dynamic behaviour of the tested structure, even for body-inwhite panels. That means that the mathematical simplified model is far from the physical parameters of the structure. The structural modelling of these flexible panels is made at the earliest stage of development from the CAD data from established and improved modelling rules. Among the numerous errors factors that arise together, the calculations are very sensitive to the real physical parameters of the piece under study (mass, thickness, geometrical shape, elastic characteristics of the material) and are also influenced, of course, by boundary conditions and assembly process. In this paper, we propose a strategy for validation and improvement of body panel FE models using vibration and 3D-shape image measuring methods. The main advantage of these methods is to provide a whole field information allowing in the same time to enhance the quality of the validation test data and to speed-up the measurements. The first optical method proposed is based on pulsed laser ESPI (Electronic Speckle Pattern Interferometry) and allows to record vibration patterns with a density of up to one million of simultaneous measured points. The high sensitivity and spatial density are a real benefit to validate and improve the large body panels models. However, ESPI records are temporally limited. The laser shots have to be triggered relatively to an external signal coming either from an accelerometer or a force gauge. We therefore developed a hybrid methodology combining the use of accelerometers and laser ESPI that we will explain and demonstrate through application examples. Session A6-3, 16:20 h Paper number 617 (Invited, Oral) Charles Q. Zhang, Marie-Odile Schott and Bruno Mahieux RENAULT SA, 1 av de Golf 78288 Guyancourt France I-INCE Class 76 NUMERICAL PROCESS FOR REDUCING THE LOW FREQUENCY INTERIOR NOISE OF A PASSENGER CAR This paper presents a numerical - experimental approach for assistance of the structure design to achieve the acoustical comfort target of a passenger car. This approach can be split into three steps in a vehicle development project. In the early design stage, subsystem numerical models are applied in order to select architectural concept and to determine the most important designs parameters. These models are often combined with experimental data from the actual production cars or from competitor cars to complete lacking information. During the detail drawing stage, both body in white and trimmed body finite element models are exploited for optimizing the structure design to achieve the specified technical requirements. The trim part models are built up based on the design data combined with a data base which contains updated parameters of actual production cars. During the physical prototype validation stage, the numerical models are first correlated to experimental data and then they are used to find technical solutions for the remained acoustical problems. In the end of this step, we have an updated numerical model which is considered as a reference model for the next vehicle project. The main advantage of this process is that it makes numerical analysis more reliable and more reactive with respect to design evolution. Moreover, know-how is enriched from one project to another through a database. Internoise 2001 Abstracts 81 Session A6-3, 16:40 h Paper number 623 (Invited, Oral) L. Cremers, M. Tournour and C.F. McCulloch LMS International, Interleuvenlaan 68, B-3001 Leuven, Belgium I-INCE Class 76 PANEL ACOUSTIC CONTRIBUTION ANALYSIS BASED ON ACOUSTIC TRANSFER VECTORS In order to refine the interior vibro-acoustic performance of vehicles, especially in the lower frequency range, it is of interest to make analyses to show the relative significance of different parts of the vehicle structure as they contribute to the interior sound - panel acoustic contribution analysis. The interior field of the vehicle can be modelled with boundary elements, and multi-domain BE methods enable different properties to be applied in some regions, such as the absorbent behaviour of the seats. In order to solve the BE model of the vehicle interior, a new methodology of Acoustic Transfer Vectors (ATVs) has been implemented. This gives much more rapid computation of panel contributions, especially when re-computation is required due to changes in vibration boundary conditions (for different loadcases on the structure) or due to changes in the selected panels (for example, to assess the relative importance of different regions of a structural panel, such as the car roof). The paper explains the principles of the ATV methodology. Details of the implementation of ATVs for use in panel acoustic contribution analysis are then given, and illustrated with some practical examples. Session A6-3, 17:00 h Paper number 652 (Invited, Oral) Ricardo Pinto, Patricia Strasser and Claudio Bertolini RIETER Automotive, Winterthur, Switzerland I-INCE Class 76 DEVELOPMENT AND VALIDATION OF A SEA MODEL FOR THE SIMULATION OF AIRBORNE NOISE FROM THE ENGINE COMPARTMENT The efficiency of vehicle’s sound packages is evaluated by some OEM’s through acoustical transfer functions, which give a measure of the fraction of the acoustic energy that, radiated by different sources (engine, muffler and tail tube) reaches the passenger’s ears. There are several ways of defining these acoustical transfer functions. Some OEM’s define them in terms of ratio between the SPL nearby an artificial sound source and the SPL inside the passenger compartment (Ps/Pin). Sound package suppliers are in this case directly responsible for the achieving of the targets set for Ps/Pin transfer functions at the beginning of a new car project. It is clear then that, both for the OEM’s and for its suppliers, it would be very useful to be able to analytically estimate, already in this very first phases of the project, the Ps/Pin transfer functions. SEA is a natural candidate to carry out this analytical estimation. Several Ps/Pin transfer functions are of interest when one considers all the airborne sources which are present in a vehicle (powertrain, mufflers, exhaust tail tube, tyres, etc.). For each one of these transfer functions, it is not unreasonable to assume that, at least in the low-mid frequency range, sound transmission into the passenger compartment takes place mainly through the part of the vehicle that is closer to the source. For this reason, the SEA model of the ‘corresponding’ vehicle part for each source can be kept separate. In this way a more efficient modeling of the whole vehicle can be achieved. This work concerns the modelling of the airborne noise path from the power train to the interior and therefore it refers mainly to the dash area. When a first model is ready, a validation procedure must be considered. Such a validation procedure is usually carried out using ideal acoustical load cases. The typical ‘ideal’ acoustic load case is a diffuse sound field. This validation allows an iterative refinement of the model, through which it can be assured that not only the model is 82 Internoise 2001 Abstracts reasonably accurate in ‘overall’ terms, but also that the contributions of the different transmission paths included in the model are in agreement with the experimental data. Subsequently, this validated model may be used to simulate the relative Ps/Pin transfer functions. To do this, some further steps are necessary. The emission and the receiving acoustic environments must be modeled and characterized in terms of absorption. The acoustic loads (i.e. the SPL that excite the different passenger compartment’s boundaries when Ps/Pin transfer functions are measured) must be measured and applied as excitations to the developed model. The ‘secondary components’ contribution (doors, floor etc.) must be included into the model. Session A6-3, 17:20 h Paper number 486 (Contributed, Oral) Hélène Giardi and Xavier Bohineust I-INCE Class 21 PSA Peugeot Citroën - Division of Research and Automotive Innovation, Route de Gisy 78943 VélizyVillacoublay Cedex (France) DIAGNOSTIC METHOD APPLIED TO AERO-ACOUSTIC VEHICLE ANALYSIS Aeroacoustic vehicle is still a matter of understanding and fully a numerical approach can not be used in the design vehicle cycle. Like other fields of NVH vehicle design and development of innovative products, it is required to combine physical understanding and hybrid tools to evaluate the acoustical contribution of each component. This paper presents an experimental approach that allows to analyse the aeroacoustical transfer paths and which will be able in the future to integrate acoustical sources models or structural acoustics of fully trimmed vehicle models. Furthermore, this approach is compatible with binaural acoustical simulation of modification previously established by PSA in the context of a European Project (ISMA 94, ISMA 96, FISITA 97). The diagnostic method is based on an airborne transfer path analysis method in reverberant chamber and aeroacoustic wind tunnel. A classical masking method is used with new dedicated light structure and removable insulated complex panels that allow a high insulation index level. The fully trimmed vehicle passenger compartment is divided in around 15 parts that can be subdivided in panels and seals contribution. In a first step, it is demonstrated that the individual transfer path of each panel can be used to calculate the global binaural contribution at the front and rear head location in the vehicle using reverberant room measurement. Then the same demonstration is made in hemi-anechoic wind tunnel at several speeds and several yaw angles. Thus, for each innovative design modification, one could establish the ranking of each contribution allowing a transmission loss modification and an aerodynamic analysis of airflow modifications. It is illustrated that numerical or experimental data basis can be used to evaluate design modification of innovative products in term of target level or binaural acoustic quality index. Detailed analysis of external acoustic field and internal acoustic transfers and responses are used to established an equivalent loading panel acoustic pressure which can be used at design step on new innovative concept. Internoise 2001 Abstracts 83 Session A6-3, 17:40 h Paper number 83 (Contributed, Oral) I-INCE Class 45 Takayuki Koizumi, Nobutaka Tsujiuchi , Koji Yamazaki, Tomoyuki Iwagase and Takaaki Takenaka Doshisha University, Tomoyuki Iwagase, Toyo Tire & Rubber Co. Ltd. OPTIMIZATION PROCEDURE OF RUBBER MOUNTS FOR INTERIOR NOISE REDUCTION OF PASSENGER CAR Road noise, a factor of interior noise, generated by tire-suspension-body vibration, can be reduced by stiffness changes of rubber mounts installed in the suspension attachment points. However, generally, its stiffness properties are barely changed because of difficulty of compatibleness with riding comfort. Therefore, in this study, we propose a mount stiffness optimization procedure that enable to not only reduce road noise but enhance riding comfort. For road noise reduction, it is important to grasp which component vibration on the system contributes mainly to the interior noise. So, the relationship between road noise and vibration of the suspension system is investigated by road noise contribution analysis based on the multiple inputs/single output (MISO) data analysis procedure. The inputs of the MISO system are the accelerations on various locations of the suspension system, and the output is the sound pressure at the driver’s ear location. These data required for the analysis are measured by the inoperation test on an actual rough road. By this analysis, we specified the contribution ranking of various components to the interior noise, and the best location and direction to evaluate the dynamic characteristics of the suspension system. Furthermore, the properties of rubber mounts which are highly sensitive to road noise and riding comfort respectively are specified by sensitivity analysis using frequency response functions. By this analytical method, it can be evaluated that the effects of the rubber mount’s stiffness modification, and also can be considered the frequency dependence of its stiffness with relatively easy and accurate way. Session A6-3, 18:00 h Paper number 635 (Contributed, Oral) A. R. Mohanty and B. Suresh Kumar PREDICTION OF TRANSMISSION LOSS OF LAYERED BARRIER MATERIALS I-INCE Class 35 Department of Mechanical Engineering, Indian Institute of Technology, Kharagpur 721 302, INDIA In this paper the transmission loss of various layered barrier materials are predicted using available mathematical formulations. The prediction results are compared with those measured experimentally. A description of the developed test rig for the measurement of the TL is presented. The A sound intensity probe is used to measure the transmission loss of the barrier materials. The effects of metal stiffeners on the TL of the layered barrier materials are also studied experimentally. There is excellent agreement between the predicted TL and the measured TL of the layered barrier materials. Such layered barrier materials have wide application in automobiles for NVH reduction. 84 Internoise 2001 Abstracts POSTERS Session A6-3, 18:20 h Paper number 48 (Contributed, Poster) Posterboard BP-5 Wednesday, Author present 9:40 – 10:40 h Woo Seok Hwang Taegu University, Kyungbuk 712-714, Korea ANALYSIS OF BOOMING NOISE USING RIGID BODY INFORMATION OF COMPONENTS I-INCE Class 76 Booming noise generated by the structural vibration is one of the typical NVH problems in a car. Its mechanism for generation and transfer is so complex that various methods have been suggested so far. Through several basic tests, we found that the powertrain and subframes, which are the main sources and paths of the booming noise, keep the rigid body motions over the RPM range of the low frequency booming. This paper presents a technique to predict the low frequency booming noise in a car using the rigid body information of the related sub-components including the engine and subframes. The rigid body information comes from the CAD data, from which we can predict the constant FRF’s for each subsystem. For the trimmed body including the cabin cavity, we measure the FRF’s using the experimental technique such as impact hammer test. The FRF-based substructuring method combines those FRF’s and the mounting stiffness at each connecting point to predict the noise level in the cabin. The noise level predicted by this method shows good agreement with those measured during the runup tests of a real car. This method enables us to suggest a design modification for the noise reduction and predict the effect of design changes. Since the mechanism of this technique is very similar to the finite element formulation, we can apply it to the complex system with ease. Session A6-3, 18:20 h Paper number 118 (Contributed, Poster) Posterboard BP-6 Wednesday, Author present 9:40 – 10:40 h Sang-Kwon Lee I-INCE Class 76 Acoustics and Dynamic Laboratory Department of Mechanical Engineering, Inha University, Inchon, Korea REDUCTION OF INTERIOR NOISE USING NOISE PATH ANALYSIS AND RUNNING MODAL ANALYSIS This paper presents a practical method for reduction of interior noise and improvement of sound quality in compartment of passenger car. The tested vehicle has a booming noise problem at rear passenger seats. In order to identify the transfer path of interior noise, the running modal analysis, the vibro-acoustic frequency transfer response and the noise path analysis are systematically employed. Using these various methods, it has been founded that the rear part of the roof of the test car was a noise source for the booming noise. Through the modification of the roof, the booming noise has been reduced to 5 dBA and sound quality inside car also has been improved. Internoise 2001 Abstracts 85 Session A6-3, 18:20 h Paper number 263 (Contributed, Poster) Posterboard BP-7 Wednesday, Author present 9:40 – 10:40 h Shinichi Ohno and Toru Yamazaki I-INCE Class 56 Kanagawa Institute of Technology, Department of Mechanical Engineering, 1030, Shimoogino, Atsugi-shi, Kanagawa-ken, 243-0292, Japan A SURVEY OF THE LEVEL OF INTERIOR NOISE OF PASSENGER CARS A data-base about the level of interior noise of passenger cars was made using data taken in Japan for nearly 400 passenger cars in the period over 30 years since 1960’s. In this paper, based on the database, decrease in the level of interior noise during this period, relation between interior noise level and vehicle velocity, relation between interior noise level and engine displacement, etc. are shown. In addition the effect of the regulation on automotive exterior noise on interior noise level is shown. Session A7-1: Effectiveness of aircraft noise control Chaired by Ichiro Yamada and Kare Liasjo Session A7-1, 9:40 h Michel Vallet AEDIFICE Lyon, France AIRCRAFT NOISE REGULATIONS FOR LONG TERM LAND PLANNING AND SHORT TERM ACTION Paper number 328 (Tutorial, Oral) I-INCE Class 52 Aircraft Noise around airports is a major problem, which limits the extension of the airport and of the traffic. The cost-effectiveness and cost-benefit of noise control and regulations for airports noise have to be assessed. Clearly, people living around airports, prefer a limit in term of number of noise events, and don't understand the acoustic indexes. This view doesn't consider the continuous global decrease in the maximum level in modern aircrafts. This paper deals with the necessary distinction between the long term regulations mainly for land planning purpose and global noise regulations and control, and the short term actions, which need complementary acoustical restrictions. A recent psycho-social survey carried out in France confirms the importance of the peak of annoyance in the evening and at night. A review of the regulations adopted in Europe, United States and Japan shall be performed to assess the cost benefit of the complementary noise regulation adopted (SEL, Lmax and quota counts at night, and not only time of day penalty) to limit the extra annoyance due to the continuous aircraft traffic increase. Session A7-1, 10:20 h Paper number 104 (Invited, Oral) I-INCE Class 13 Dion F. Oude Lansink, Foort De Roo, Marinus M. Boone and Stephan P. Galis TNO TPD (Institute of Applied Physics), P.O. Box 155, 2600 AD Delft, The Netherlands MEASUREMENT OF AIRCRAFT NOISE EMISSION UNDER OPERATIONAL CONDITIONS Prediction of noise impact from air traffic requires input data concerning the noise emission of aircraft as a function of the modes of operation. Up to now these data are only acquired by the manufacturer 86 Internoise 2001 Abstracts during the certification tests of aeroplanes. The certification tests are carried out during controlled flyovers at measuring positions directly under the flight path. The atmospheric conditions during the tests and the ground impedance of the measurement terrain will in most cases be as close as possible to standardised conditions. The sound levels measured under the flight path are assumed to be equivalent to levels at more lateral positions with the same source-receiver distance. This methodology does not guarantee that the actual emission values in normal practice will be similar to the values derived from the certification measurements. Therefore a need for verification of the input data has arisen. On request of the Dutch civil aviation authority a methodology is being developed to measure the noise emission of aircraft during normal operations around airports. A measurement method for emission values should be closely related to the simulation method for the noise impact prediction. The measurement method under development will be related to the Swiss method of modelling and measurement, but will be based on segmentation rather than on full simulation of the flight path in the noise impact predictions. This approach should be sufficiently easy to handle in the everyday practice of noise impact calculations. For the short term the measurement method will be used in a system of regular verification of the noise emission input data, based on LA,max values. On the longer run also a LA.x (=SEL) based data set may be developed and verified on a regular basis. Session A7-1, 10:40 h Paper number 359 (Invited, Oral) Naoaki Shinohara, Saburo Ogata and Ichiro Yamada I-INCE Class 24 New Tokyo International Airport Promotion Foundation, 118 Nakanodai, Higashi-Sanrizuka, Narita, Chiba 286-0112, Japan LONG-TERM LEVEL FLUCTUATION OF AIRPORT NOISE DUE TO METEOROLOGICAL CONDITIONS Long-term level fluctuation of airport noise was investigated, in relation to meteorological conditions, using observation data of unattended noise monitoring system of airport noise including fly-over noise, engine run-up noise and other airport ground noise due to APU operation, towing and so on. This paper not only describes the procedure that classifies the observed noise events being based on information about sound arrival direction and airport operation, but also shows results of level fluctuation of noise events due to individual sound sources respectively in relation to meteorological conditions. Session A7-1, 11:00 h Paper number 174 (Invited, Oral) John S. Bradley and J.A. Birta I-INCE Class 31 Institute for Research in Construction, National Research Council, Montreal Rd., Ottawa, Canada, K1A 0R6 THE ATTENUATION OF AIRCRAFT NOISE BY WOOD STUD WALLS In North America, both single-unit and multiple housing is frequently built using wood stud construction. The overall performance of many common wood stud constructions is limited by excess low frequency sound transmission and with these constructions indoor aircraft noise levels are dominated by low frequency sound. A study to develop improved methods for determining the sound insulation of buildings against aircraft noise has included laboratory sound insulation measurements of 41 wood stud wall constructions. The results demonstrate the relative importance of the principal parameters: mass of surface layers, details of the wood stud system, structural breaks, and cavity Internoise 2001 Abstracts 87 insulation. Optimum designs are shown to out-perform a brick wall, which is often cited as an ideal goal for exterior wall sound insulation. In some cases less material is shown to produce better sound insulation and hence a more cost effective solution. Session A7-1, 11:20 h Paper number 98 (Invited, Oral) Maurice Yeung, K.S. Chan, C.L. Wong and K.Y. Fung Environmental Protection Department, Hong Kong SAR Government A REVIEW OF AIRCRAFT NOISE PROBLEM IN HONG KONG I-INCE Class 52 The new Chek Lap Kok Airport located at the semi man-made island at the southwest corner of Hong Kong commenced operation in 1998 as a successor to the old Kai Tak Airport located at the heart of the city. The relocation of airport has brought immediate relief to some 380,000 people who once had to endure very high aircraft noise levels. Tremendous efforts have been spent on planning of the new airport including careful site selection, impact assessment and a more stringent planning criterion of NEF25 to avoid replica of the severe noise problem adhered to the old airport. As a result, the NEF 25 contour of the new airport covers largely sea areas. Only a small number of acoustically insulated village houses close to the airport marginally lie within the NEF 25 contour. Despite all these planning efforts, noise from aircraft flyby still triggered considerable member of public complaints when it started operation. Special flight procedures and arrangements have been adopted to further minimize noise impact from the new airport operation and a state-of-art aircraft noise and flight track monitoring system has been installed to keep track of the noise picture as well as to validate the noise assessment conducted before. This paper will give a retrospective view of aircraft noise problem in Hong Kong. Other aspects like planning considerations for the new airport from site selection, adoption of more stringent noise criterion, noise assessment process and the experience gained after its operation including complaint handling, noise monitoring and validation of assessment would also be covered. Session A7-1, 11:40 h Paper number 364 (Invited, Oral) Michihiro Sugai and Katsumi Igarashi I-INCE Class 52 New Tokyo International Airport Authority, NAA-Bldg., Narita Airport, Narita, Chiba 282-8601 JAPAN EFFECTIVENESS OF NOISE COUNTERMEASURES AT NARITA AIRPORT Aircraft noise is one of the most serious problems that affect communities surrounding airports. Strategy for noise mitigation is, therefore a matter of great concern as a mainstay of environmental policies at Narita Airport. This paper briefly describes noise measures which the Narita Airport Authority (NAA) has carried out and then examines their effectiveness for noise-effect mitigation. The measures, which are focused on noise reduction at sound sources(restrictions of noisy aircraft and aircraft movements, and improvement of flight procedures),structural improvement to the airport (wooded buffer zones, noise mitigation embankments and noise suppressor facilities) and countermeasures in the surrounding communities(compensation to soundproofing and removal of houses, land use control and other local treatments), are based on a Law on the Prevention of Troubles due to Aircraft Noise around Public Airdrome, as well as a Law for Special Measures to Control Aircraft Noise Around Specified Airports. The paper refers to the circumstances in legislation of these laws and their effectiveness for control of airport noise problems. 88 Internoise 2001 Abstracts Session A7-1, 12:00 h Gerard Bekebrede Paper number 763 (Invited, Oral) I-INCE Class 52 Ministry of Transport, Public Works and Water Management, Directorate General of Civil Aviation, P.O. Box 90771, 2509 LT The Hague, The Netherlands THE POLICY POTENTIAL OF NEW TECHNOLOGY TO IMPROVE THE ENVIRONMENTAL CHARACTERISTICS OF AVIATION This paper will discuss the issue that the potentials of the technology can not always fully be used due to market/economic reasons and that policy measures (such as more stringent noise standards and rules for phase out of the noisiest aircraft) can help to enhance the introduction of new technology. With respect to policy measures international organizations are active, such as the environmental committee of the European Civil Aviation Conference (ECAC) and ICAO’s Committee on Environmental Protection CAEP. CAEP’s task is to give recommendation to ICAO’s Council. In January 2001, CAEP held its 5th meeting. The meeting was attended by the delegations of 18 UN countries and observer organizations. The meeting which was attended by about 200 people made several recommendations including: - a new noise standard which is 10 decibels lower, on a cumulative basis, than the current Chapter 3 standards in ICAO Annex 16, Volume I; - procedures for recertification of existing aircraft meeting the new standard; - more stringent noise standards for helicopters; - further development of elements necessary for an open emission trading system for international aviation. This paper will discuss the discrepancy between technological potential, market potential and policy potential of new technology and will describe the process in the international organizations. Session A7-1, 13:20 h Paper number 653 (Contributed, Oral) I-INCE Class 13 Federico Miyara, Susana Cabanellas, Patricia Mosconi, Vivian Pasch, Marta Yanitelli, Juan C. Rall and JorgeVazquez Fac. Cs. Ex. Ing. y Agrim., Univ. Nac. Rosario, Riobamba 245 bis, 2000 Rosario, Argentina THE ACOUSTICAL GEOMETRY OF AIRCRAFT OVERFLIGHTS Noise perceived in the surroundings of an airport is an important cause of annoyance because aircraft are very high sound power noise sources and, as a result of the lack of obstacles, propagation conditions are fairly good. Several publications, such as ICAO's Annex 16 on aircraft noise as well as Circular No 205of the same entity, suggest methods to predict noise levels in the surroundings of an airport, including how to compute several descriptors which have proved useful in airport environmental noise assessment. However, they require intensive use of data regarding noise emission from different aircrafts, which in practice may be difficult or expensive to obtain. In this paper we show how these data may be easily derived from field measurements performed during aircraft overflights. From an aircraft overflight noise signal, acquired with a digital audio tape (DAT) recorder, the audio signal is computer-processed to plot its spectrogram. This is used to estimate the aircraft's speed and to derive the aircraft's altitude and elevation angle. Once the geometry of the overflight has been secured, the last step involves matching the propagation-corrected measured noise at several instants with the corresponding aircraft position. This allows to draft the polar pattern of the aircraft viewed as a noise source. Internoise 2001 Abstracts 89 Session A7-1, 13:40 h Paper number 224 (Contributed, Oral) H.B.G. Ten Haven and S. Jonkhart COCKPIT, NOISE-LOAD GUIDANCE AROUND SCHIPHOL AIRPORT I-INCE Class 52 National Aerospace Laboratory NLR, PO Box 90502, 1006 BM Amsterdam, The Netherlands Together with Amsterdam Airport Schiphol, NLR is developing a short term noise-load analysis and steering system for Schiphol Airport: the Cockpit. Cockpit combines the outcome of two data flows to predict the expected noise-load over a time frame (the enforcement year of Schiphol, starting the 1st of November ending 31st of October). These data flows are: a prediction of the expected traffic for the coming part of the year, and the actual noise-load and traffic data from the FANOMOS monitoring system. The traffic predictions are driven by the slot distribution. On the basis of the distribution the system can verify expected noise-loads for a coming period. Slot distributions are processed using average weather conditions including expected deviations, incorporating amongst others crossing SID avoidance, preferential runway use, and typical ATC decisions to combine runways under specific circumstances. By law: the maximum noise-loads in a number of so called zone-points are prescribed and the Airport has to declare the expected noise-load in these points for the coming enforcement year. Both values are used to evaluate the expected noise-load during the year to identify so-called critical (zone)-points. In these points the system can analyze contributions in various ways. The results of these analyses are used to define measures. The measures can be included in the traffic generation/prediction to evaluate their efficiencies to decide what measure should be taken to support the Airport to keep within its boundaries. Session A7-1, 14:00 h Paper number 612 (Contributed, Oral) P. Bouter and F.D. Van der Ploeg I-INCE Class 83 Adviesbureau Peutz & Associés B.V., P.O. Box 66, NL-6585 ZH MOOK, The Netherlands AIRCRAFT NOISE MANAGEMENT IN THE NETHERLANDS VS. LIMITED PREDICTION CAPABILITIES OF CAUSAL AIRCRAFT NOISE CALCULATION MODELS. Causal aircraft noise calculation models (such as FAA/INM) are widely used as an analysis tool to assist in assessing the impact of aircraft noise in the vicinity of airports. Typical applications are land use planning (by means of a contour-set) and assessing noise control measures such as noise abatement procedures, alternative runway lay-out and/or tracks (by means of the total area within a certain contour, or the number of residents within the contour). In the Netherlands, this concept of aircraft noise calculation models is extended to the set-up of limit-values along the contour-boundary, including enforcement and sanction procedures. The main argument is that only in separate points calculations can be supported by measurements. These regulations, down to the level of single points, to be predicted at least 10 years ahead, make high demands on the calculation procedure and set-up. A sensitivity analysis of causal aircraft noise calculation models was executed, covering aspects such as: - interpretation/implementation of officially prescribed (Dutch) calculation method; - horizontal and vertical dispersion of tracks and flight-profiles; - noise abatement measures; - wind, capacity, destination and noise dependent runway usage; - growth of traffic volume vs. modern quieter aircraft; - and many more. The results show that accurate predictions (e.g. within 1 dB(A)) are hardly possible, whereas even 1 dB(A) inaccuracy corresponds to 20-25% uncontrolled traffic volume. Possible solutions are empirical or hybrid noise calculation methods, which are currently under investigation. 90 Internoise 2001 Abstracts These methods may achieve higher accuracy, at the cost of loss of causality; these methods rely on time-independence of the empirically determined effects. POSTERS Session A7-1, 14:20 h Paper number 121 (Contributed, Poster) Posterboard AP-19 Tuesday, Author present 15:00 – 16:00 h W. Dobrzynski and U. Stöcker I-INCE Class 52 Deutsches Zentrum für Luft- und Raumfahrt (DLR), Institut für Entwurfsaerodynamik, Lilienthalplatz 7, 38108 Braunschweig, Germany REGULATIONS FOR NOISE ABATEMENT AROUND SMALL AIRFIELDS IN GERMANY - A CHALLENGE FOR THE GENERAL AVIATION AIRCRAFT INDUSTRY? Increasing noise complaints are recorded from residential areas surrounding small airfields in Germany. As a consequence the German Ministry of Transport, Building and Housing set in force noise related traffic regulations for General Aviation (GA) aircraft, which were meant as an impetus for the aircraft operators to realize available noise reduction technologies. These regulations (German "Landeplatz-Lärmschutzverordnung") impose operational restrictions - at leisure times in particular for aircraft without a noise certificate and for aircraft with noise levels which do not fall 4 to 7 dB below their relevant (Chapters 6 or 10) noise certification level. Exceptions are granted for crosscountry flights. The philosophy of this regulation is to reduce noise originating from the most annoying circuit flights of training aircraft but not hamper GA-aircraft business operations. Seeking for a non restricted operation, therefore each individual aircraft type needs a noise certificate to establish its compliance with the required noise limits. This turns out as a hardship for older aircraft, which only comply with former - less stringent - noise limits, or which even do not have a noise certificate at all (if they got their airworthiness certificate before 1971). Such aircraft need an adequate retrofitting with low-noise propellers or engine exhaust mufflers. The respective technology has been developed in recent years, coming up with multiblade propellers, geared engines and innovative muffler designs. Based on the knowledge of respective noise reduction potentials the adequate technical solution (and the ensuing cost/benefit relation) has to be determined for each one aircraft type individually. Session A7-1, 14:20 h Paper number 417 (Contributed, Poster) Posterboard AP-20 Tuesday, Author present 15:00 – 16:00 h Gianni Cesini, Antonio Iannotti, Eugenio Mattei I-INCE Class 76 Dipartimento di Energetica - University of Ancona, Via Brecce Bianche, 60100 Ancona - Italy EVALUATION OF THE AIRCRAFT NOISE AROUND A REGIONAL AIRPORT BY COMPUTATIONAL METHODS AND EXPERIMENTAL VALIDATION The study of the acoustic impact at the airport of Ancona-Falconara, a regional airport situated in central Italy near the Adriatic Sea, has been carried out in order to evaluate the acoustic influence of the airport activities in the surrounding areas. Through the use of a predictive model, several conditions on either air traffic or flights’ different distribution among different directions of landing Internoise 2001 Abstracts 91 and take-off have been supposed. The evaluation of the noise produced by the airport infrastructure has been carried out using the evaluation level of the airport noise Lva, as it is defined in the Italian regulations. Moreover, the LAmax has been analysed for each type of aircraft both during take-off and landing in order to evaluate the maximum noise due to the traffic of every aircraft. Although this situation is very limited in time, it can create some problems to the people who live in the surrounding areas of the airport. Afterwards, phonometric recording has been carried out both with movable and fixed measurement stations in different points around the airport; such measurements have both had the purpose of checking the results given by the predictive model and the task of analysing the acoustic influence of every single air movement in order to define possible noise abating procedures afterwards. Session A7-1, 14:20 h Paper number 461 (Contributed, Poster) Posterboard AP-21 Tuesday, Author present 15:00 – 16:00 h J.J. Busink I-INCE Class 76 National Aerospace Laboratory (NLR), P.O. Box 90502, 1006 BM, Amsterdam, The Netherlands EFFECTS OF INCREASED NOISE STRINGENCIES ON FLEET COMPOSITION AND NOISE EXPOSURE AT SCHIPHOL AIRPORT In order to limit the effects caused by air traffic in terms of community noise, the International Civil Aviation Organisation (ICAO) is currently studying the possibility of increased noise stringencies. Recent studies focus on the phase-out of aircraft that do not meet the Chapter 3 standards by a margin of 8dB, 11dB or 14dB. This margin is the total difference of the 3 noise certification points with the Chapter 3 certification limits. On behalf of the Dutch CAA, an investigation into the effects of these increased stringencies on the fleet composition and noise exposure at Amsterdam Schiphol airport is carried out, using a number of possible phase-out scenarios. The time frame for these scenarios is 1999-2020, with the increased stringencies taking effect in 2002/2003. Firstly, for each of the years 1999, 2002, 2010 and 2020, the fleet exposure level is calculated, based on fleet projections by the airlines and airports. In a second stage, the proposed phase-out scenarios are implemented and fleet exposures are determined for 2010 and 2020 showing the effect of the increased stringencies. The results so far show that increased stringencies of 8 dB and 11 dB will have a limited effect on the noise exposure and fleet composition at Amsterdam Schiphol Airport. Only the 14 dB stringency will have a more significant effect at Amsterdam Schiphol Airport. Session A7-2: Energy flow methods Chaired by Jean Louis Guyader Session A7-2, 14:40 h K.S. Sum and J. Pan Paper number 769 (Invited, Oral) I-INCE Class 75 Department of Mechanical and Materials Engineering, The University of Western Australia, Nedlands, Western Australia 6907 ESTIMATION OF REVERBERATION TIME USING BAND-LIMITED POWER FLOW METHOD The discrepancy between reverberation times measured by steady state method and by transient decay method is well known, So far no clear explanation has been obtained. In this paper, the steady state 92 Internoise 2001 Abstracts band-limited energy in an enclosure and band-limited power flow into modally reactive boundaries are derived to describe the energy balance relationship and thus the reverberation time in a frequency band. This reverberation time is then compared to that obtained from the transient decay of the sound field based on the modal analysis. The comparison sheds some light towards the nature of the reverberation curves and the understanding of the discrepancy mentioned above. Session A7-2, 15:00 h Paper number 235 (Invited, Oral) Rajendra Singh and Seungbo Kim The Ohio State University Columbus, OH 43210-1107, USA I-INCE Class 76 VIBRATION POWER FLOW CONSIDERATIONS ARISING FROM MULTI-DIMENSIONAL ISOLATORS Much of the vibration isolation research has focused on uni-directional behavior of the system. For many real-life problems, the role of rotational and shear stiffness components must be understood and a multi-dimensional formulation needs to be developed, especially at higher frequencies. In this study, the effects of multi-dimensional isolator on vibration power transmitted to the receiver structure are analytically investigated using the mobility synthesis method. Rigid body and Timoshenko beam models are employed to describe source and path respectively. Infinite and finite beam or plate structures are used to represent a compliant foundation. Also, linear, time-invariant system assumption is made. Several measures of vibration isolation performance are studied using power flow consideration, and sample parametric design studies are presented. Sound power and pressure are also calculated in order to examine their relations to vibration power transmitted to the receiver. Session A7-2, 15:20 h Paper number 490 (Invited, Oral) I-INCE Class 76 Jeffrey S. Vipperman, Deyu Li, Ilya V. Avdeev and Steven A. Lane University of Pittsburgh, Department of Mechanical Engineering, 648 Benedum Hall, Pittsburgh, PA526USA CHARACTERIZATION AND CONTROL OF SOUND RADIATION IN A COMPLEX FAIRING STRUCTURE A complex mock-scale composite fairing structure will be modeled and experimentally characterized for radiation control purposes. The tapered cylindrical structure is formed from multiple triangular shaped ribbed panels that are custom-fabricated from graphite-epoxy prepreg. The ribbed stiffners tend break up the structure into regions of localized structural response in addition to the expected global behavior of the tapered structure. A finite element analysis (FEA) model of the structure and interior acoustic cavity are formed and coupled using a modal-interaction approach. Experimental characterization is then performed by conducting structural measurements with a laser vibrometer and acoustic measurements using an internal microphone array. The experimental data are used to corroborate the FEA models. For both the analytical and experimental data, the structural response will be decomposed into radiation modes. A comparison between the modeled and measured responses will be made with specific attention to the accuracy of the modal-interaction approach and it's ability to accurately suggest placement of transducers and predict control performance. Internoise 2001 Abstracts 93 Session A7-2, 15:40 h Paper number 636 (Invited, Oral) L. Maxit and J.L. Guyader STATISTICAL POWER FLOW ANALYSIS BETWEEN COUPLED SUBSYSTEMS I-INCE Class 76 Laboratoire Vibrations Acoustique - INSA de Lyon 69621 VILLEURBANNE Cedex - France Statistical Energy Analysis (SEA) is based on assumptions which can be unrealistic in some situations. It is known that one of them occurs when subsystems are highly damped. Indeed, in this case the vibratory wave field is not diffuse and the exponential decay of the direct field predominates. The goal of this paper is to propose an alternative approach to SEA for such situations. Contrary to SEA which starts from the basic power flow relation between two oscillators, the present approach is based on an approximate power flow relation between two coupled multi-modal systems excited by a random point force of white noise spectrum. The advantage of this relation is that it permits to take into account correlations between modes within each subsystem. Using the results of the dual modal formulation and extending the formalism to several coupled subsystems, one obtains a system of equations linking the power flows through the different junctions. The results yielded by the present approach for different damping values are compared to classical SEA results in the case of coupled beams. For systems composed of three or more subsystems and having a modal overlap factor greater than one, the prediction is significantly improved compared to SEA. Another example of coupled plates shows that the method is well suitable for point excitations. In conclusion, the present approach is an interesting possibility for extending SEA to systems highly damped or excited by a localised forces. Session A7-2, 16:00 h Paper number 385 (Contributed, Oral) Matthieu Noël, Frank Simon and Jean-Claude Pascal I-INCE Class 72 Laboratoire d’Acoustique de l’Université du Maine, (UMR CNRS 6613), Le Mans, France ONE DIMENSIONAL THEORETICAL AND EXPERIMENTAL WAVE NUMBER EXTRACTION FOR STRUCTURAL INTENSITY Wave number is strongly related to energy propagation speed in structures. Thus, most of experimental structural intensity formulations need prior knowledge of wave number. In the case of structure with simple boundary conditions, a theoretical expression of the wave number exists, which is not the case for complex boundary conditions. Different methods exist to extract the experimental wave number (two or three accelerometers, Spatial Fourier Transform identification by Wigner distribution, and wave extraction). The authors validate their approach by comparison of the extracted wave number with theoretical expression. Respective advantages and drawbacks are sometimes analysed. The purpose of this paper is to compare simulated and measured structural intensity obtained with three methods of wave number extraction). In a first stage, in- and out-of-plane motion of a beam are simulated with wave approach, for a highly damped cantilever beam (TEFLON 570x70x10mm3) with viscous dampers at the free end, excited by in or out-of-plane force. Once the wave number computed according to the three methods, structural intensity is simulated using wave-decomposition. Wave number and associated structural intensity are then measured by the same three methods. These theoretical and experimental results are also compared to the exact analytical expression of structural intensity. The sensitivity of wave number and structural intensity to different parameters (phase mismatch between accelerometers, spatial sampling frequency) is briefly tackled. 94 Internoise 2001 Abstracts Session A7-2, 16:20 h Paper number 178 (Contributed, Oral) E.C.N. Wester and B.R. Mace Industrial Research Limited, PO Box 2225, Auckland, New Zealand I-INCE Class 75 STATISTICAL MEASURES OF ENERGY FLOW IN COMPLEX SYSTEMS USING A WAVE-BASED APPROACH Although SEA is a powerful analysis tool, traditional assumptions of high reverberance, high modal overlap and weak coupling significantly limit the range of systems to which it can be applied with confidence. This paper describes a wave-based approach which addresses many of these limitations. The approach uses junction and subsystem scattering matrices to describe wave propagation in the system, and the eigenvalues and eigenvectors of a certain function of these matrices to describe the energy flow between subsystems. A notional ensemble of systems is generated by assuming that the phases of the eigenvalues are random, although not independent, variables, uniformly distributed in the interval [-pi,pi]. If the system is not heavily damped, the statistical distributions of eigenvalues and eigenvectors are independent, and the effects of correlations between eigenvalues are not important in determining energy flow. These simplifications are valid for a wide range of systems in practice, and lead to a computationally inexpensive method for estimating the average and variance of energy flow. The validity of the method is demonstrated for a number of irregular, multi-plate systems by comparing predictions with the results of Monte Carlo simulation, and with the average flow predicted by traditional SEA. Session A7-2, 16:40 h Denis Blanchet Paper number 502 (Contributed, Oral) I-INCE Class 76 Vibro-Acoustic Sciences, 200 E. Big Beaver Road, Troy, MI, 48083, USA A NEW AUTOMATED SEA TOOL FOR THE ANALYSIS OF INTERIOR NOISE OF AUTOMOBILES A key goal in the development of SEA has been to reduce the number of idealizations and expert judgements required by the analyst, thereby making accurate results less dependent upon the expertise of individual users. This paper propose to develop and demonstrate prototype technology for automated interior noise analysis of automobiles. The objectives of this work are to efficiently evaluate interior trim components - for sound quality, weight reduction and cost savings; to automate the creation of SEA models for one-third-octave band interior sound pressure levels due to road noise and engine noise; to speed up the noise and vibration engineering cycle to help reduce vehicle development time and improve quality. This is done through the development of model template and modeling wizard. Internoise 2001 Abstracts 95 POSTER Session A7-2 Paper number 146 (Contributed, Poster) Posterboard AP-22 Tuesday, Author present 17:40 – 18:40 h o-Won Lee, Suk-Yoon Hong and Young-Ho Park I-INCE Class 76 Department of Naval Architecture and Ocean Engineering, Seoul National University RADIATION NOISE ANALYSIS USING THE RESULTS OF POWER FLOW FINITE ELEMENT METHOD In this paper the problem of the noise radiation induced by the structural vibration of the submarine is considered. The power flow finite element method which is used for the prediction of the vibration energy density and intensity of a complex structure in mid to high frequency band is here used as a method concerning the vibration analysis of the submarine structure. The sound field around the submarine structure is studied by solving the surface pressure about the structure by a boundary element method. At this time the vibration energy density calculated by power flow finite element method is used as the boundary conditions of BEM. The characteristics of the vibration and radiation of the submarine structure are predicted and the results are compared with the results of the SYSNOISE which is widely used in the radiation problems. Session A7-3: Sound power and intensity measurements Chaired by George Maling Session A7-3, 17:00 h Gustav A. Sehrndt Paper number 232 (Contributed, Oral) I-INCE Class 72 Beratung zur Lärmminderung, Niesertstr.42, DE 48145 Münster, Germany SOME REMARKS ON MICROPHONE POSITIONING FOR SOUND POWER LEVEL DETERMINATION Positioning of microphones according standard ISO 3744 should be chosen carefully. Directivity and tonal character of the source should be considered. There are several possibilities to distribute measuring points on hemispherical measuring surfaces. For parallelepiped surfaces machine related investigations seem to be necessary. Some experience determining the environmental correction by use of a reference sound source are reported, especially trials with a generalized spiral path according standard ISO 6926. Possible implications on applying the standard ISO 3744 are discussed. 96 Internoise 2001 Abstracts Session A7-3, 17:20 h Paper number 420 (Contributed, Oral) Rodrigo P.B. Costa-Félix and Igor A. Aquino Lab. Ensaios Acust.- DIAVI-INMETRO, CEP 25250-020 RIO DE JANEIRO, Brazil I-INCE Class 72 REPEATABILITY OF SOUND POWER MEASUREMENTS OF BLENDERS IN REVERBERATION ROOM A set of sound power measurements was performed in repeatability conditions by the comparison method in a reverberation room with a fixed position array of six microphones. The noise sources were six mixers and a complementar reference sound source. A total of five or six sound power measurements were performed on each noise source, and the standard deviation of repeatability (sr) was computed for each octave band from 125 Hz to 8 kHz, and in dB(A).With the results obtained, values of sr for mixers were compared with results for other noise sources found in the literature. The results in each octave band and in dB(A) will be used in a uncertainty expression derived for sound power measurement of sources under a labelling program. Session A7-3, 17:40 h Paper number 632 (Contributed, Oral) M. Ohya, T. Naito, K. Iwahashi and T. Koshiroi Rion Co., Ltd, 3-20-41, Higashi-motomachi, Kokubunji, Tokyo 185-8533 JAPAN I-INCE Class 72 A METHOD OF MEASURING THE PHASE DIFFERENCE BETWEEN PROBE CHANNELS FOR A PP PROBE In this paper, a method of measuring the phase difference of a p-p probe is reported. The p-p probe is a sensor for measuring a sound intensity and is consisted of a pair microphone closely set by face to face. In order to measure a sound power level by applying the method of ISO 9614-2, the sound intensity instrument which is stated IEC 61043 shall be used. Furthermore IEC 61043 requires to evaluate the phase difference of the instrument before starting the measurement of a sound power level by using it. Hence it is important to correctly evaluate the phase difference of the p-p probe using for the sound intensity instrument. A residual intensity testing device is generally used for a purpose of the measurement of the phase difference of the p-p probe. The residual intensity testing device normally has a little bit phase difference of itself. Therefore some bias error is caused by the phase difference of the testing device when the evaluation of a p-p probe is executed by using it. This report shows an improved method of the check of a p-p probe phase difference by using the residual intensity testing device. The method is able to avoid the unknown phase error of the residual intensity testing device. Finally, a sound intensity meter SI-50 which newly developed by Rion is introduced. The new instrument applies the proposed method of measuring this phase difference and adopts IEC 61043 Class1 . Internoise 2001 Abstracts 97 Session A8-1: Outdoor noise prediction 1 Chaired by Jørgen Kragh and Tony Embleton Session A8-1, 9:40 h Jørgen Kragh Paper number 672 (Tutorial, Oral) I-INCE Class 76 DELTA, Building 356, Akademivej, DK-2800 Kgs. Lyngby, Denmark NEWS AND NEEDS IN OUTDOOR NOISE PREDICTION The intention is to present what is needed to be able in the future to predict outdoor noise for various purposes, illustrated by examples of potentials and limitations of existing tools. Various needs call for a variety of methods: country-wide (or region-wide or EU-wide) noise mapping at one extreme requires other tools than assessing the need for compensation or mitigation at one individual dwelling. The LDEN averaged over a reference year will be used in Europe as a metric for environmental noise impact on people. We therefore have to deal with the combination of varying source noise emission and sound path attenuation. The variation can be caused by daily or seasonal variation in source operation (e.g. traffic intensity) and weather conditions. Old "empirical" models are being replaced by new "theoretical" models with wider fields of application. This trend has been advanced by increasing computer speed and availability. Attempts are being made to separate source and propagation in order to allow independent up-dating of source and propagation models. The availability of source data will be commented upon. Session A8-1, 10:20 h Hans G. Jonasson Paper number 404 (Invited, Oral) I-INCE Class 72 SP Swedish National Testing and Research Institute, Box 857, SE-50115 Borås, Sweden SOURCE MODULES IN THE NEW NORDIC PREDICTION METHODS FOR ROAD AND RAIL TRAFFIC NOISE In the new Nordic prediction methods for road and rail traffic noise the source emission is completely separated from the sound propagation. The sound propagation is dealt with in exactly the same way for both rail and road traffic and industrial noise. Each type of source has a source module which together with the point source propagation module constitute the complete prediction method. This paper will describe background and details of these source modules including source modelling and how the sound power level is determined from measurements of the sound exposure level during passby. The sound power level is distributed among a number of sub-sources and it is used, if asked for, to calculate maximum sound pressure levels during pass-by. In order to get accurate input data in third octave bands new emission measurements have been carried out on road vehicles. In the data base information on road surface, temperature, driving conditions, vehicle type, etc is given. The categorization is at present detailed and it might be simplified later on. Great efforts have been made to separate emission and propagation completely from each other. The source models are used both to determine the sound power level and to calculate propagation effects in more complicated cases. 98 Internoise 2001 Abstracts Session A8-1, 10:40 h Kurt Heutschi Paper number 49 (Invited, Oral) I-INCE Class 76 Swiss Federal Laboratories for Materials Testing and Research, EMPA, CH-8600 Dübendorf, Switzerland NEW SWISS SOURCE MODEL FOR ROAD TRAFFIC NOISE The road traffic noise model that is used in Switzerland today was published in 1986. The source is described as the Leq at a reference distance with input parameters speed, number of vehicles, the portion of trucks and the grade of the road. This paper presents a new source model that is based on the maximum pass by level given separately for the engine and tyre noise of passenger cars and trucks. The emission values were measured at freely flowing traffic in a distance of 7.5 m. For the sake of a well defined ground reflection, situations were selected with an essentially hard ground. The emission values were split into the two components engine and tyre noise using measurements of controlled pass-by events without engine noise and taking into account a priori knowledge about tyre noise. This separation leads to an elegant consideration of the influences of pavement and grade of the road. The source model is currently completed by investigating the representative source height. The height of traffic noise is of greatest importance in calculating the shielding effect of noise barriers. By measuring this shielding effect and comparing it with calculations the parameter source height is determined by best fit. For suitable geometries the sensitivity of the shielding effect lies in the order of 0.3 dB per 10 cm change in source height. One problem that has to be dealt with is the presence of the ground reflection which smears the results to some extent. Session A8-1, 11:00 h Foort De Roo Paper number 103 (Invited, Oral) I-INCE Class 76 TNO TPD, P.O. Box 155, 2600 AD Delft, The Netherlands Trends in sound propagation modelling Most of the current official noise impact prediction models are based on empirical modelling techniques. This means that physical propagation phenomena are described by regression relationships and other approximative formulae, based on empirical data. Because of shortcomings of this type of modelling a more theoretical approach has been followed by many researchers over the past 10 years. This has resulted in a series of models describing the sound propagation phenomena by means of mathematical expressions giving the theoretical formulations of the physical mechanisms involved. Such models have the advantage that the underlying theory generally covers a very wide range of applications and conditions. Their range of application is therefore only limited by simplifying assumptions in the development of the model and not by the availability of empirical data. The drawback may be that not every propagation effect can be modelled by each model. This implies that every modelling principle has its own strengths and weaknesses. This paper will give an overview of the pros and cons of the recent developments in modelling and will discuss the prospects for the development of a harmonised European reference propagation model. Internoise 2001 Abstracts 99 Session A8-1, 11:20 h Truls Berge Paper number 130 (Invited, Oral) I-INCE Class 76 SINTEF Telecom and Informatics, N-7465 Trondheim, Norway ROAD VEHICLE DRIVING PATTERNS AND NOISE EMISSION In developing the new Nordic prediction model for environmental noise, Nord2000, one of the tasks has been to establish a more accurate source model for road vehicles. As part of this work, the individual noise emission levels of several thousands of vehicles in the Nordic countries have been measured, according to a new NORDTEST method. This method has been developed to measure imission relevant noise levels of road vehicles. The general layout of the method will be described, together with the results from measurements performed at different locations in Norway. A special developed measuring unit has been used, which makes it possible to relate the noise levels to the driving behaviour and to the general acoustical condition of the vehicle. The results will be used to verify the proposed source model in Nord2000, as well as the relationship between driving pattern and noise emission levels. The measurements performed in Norway are compared with the results from the other Nordic countries. All the Nordic data are planned to be part of a noise emission database for road vehicles. Such a database can be an important tool to study the effect of noise reducing measures such as new tyre/noise regulations, improved type approval test methods and more stringent noise emission limits for road vehicles. Session A8-1, 11:40 h Paper number 150 (Invited, Oral) Jean-François Hamet, Vincent Steimer INRETS-LTE, 25 avenue François Mitterrand, F69675 Bron Cedex, France I-INCE Class 76 MODELLING PASS-BY NOISE OF HEAVY TRUCKS BY POWER UNIT NOISE AND ROLLING NOISE Controlled passby noise levels of different heavy trucks have been measured on a test track under various load configurations: tractor alone, tractor + empty trailer, tractor + loaden trailer. Three gear ratios conditions were used: the gear ratio adapted to each speed and the ratios next above and next below. It appears that each vehicle can reasonably be modelled by two sources which correspond to the power and the rolling (not only the tire) units. Noise level discrepancies are however observed on some vehicles and are presently being studied. New measurement campaigns are planned to get further information on this matter. Session A8-1, 12:00 h Paper number 371 (Invited, Oral) Shinichi Sakamoto and Hideki Tachibana I-INCE Class 76 Institute of Industrial Science, University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo, 153-8505 Japan NUMERICAL STUDY ON SOUND PROPAGATION FROM DEPRESSED/SEMI-UNDERGROUND ROADS Noise propagation from depressed or semi-underground roads are much complicated because of multiple reflection inside the constructions. In the prediction model of road traffic noise, - Model-1998 - proposed by the Acoustical Society of Japan, the effect of such multiple reflection is calculated by summation of the contributions from a series of image sources mirrored by the sidewalls. In case 100 Internoise 2001 Abstracts where a semi-underground road with considerably large overhangs, however, the method cannot be applied because of the limitation of applicability of geometrical acoustics. In such a case, other approaches based on wave theory are necessary. In this study, sound propagation from various types of depressed or semi-underground roads is investigated using 2- dimensional Finite Difference Method. Differences of sound propagation due to a variety of the shapes in cross section, effects of absorption treatment of walls, ceilings and drainage asphalt pavement on the road surfaces are studied by numerical analysis and visualization technique of transient sound propagation. In addition, aiming to make a simplified noise prediction method, modeling of hypothetical sound sources to simulate noise radiation from opening of semi-underground road is studied. Session A8-1, 13:20 h Paper number 301 (Contributed, Oral) J. Picaut and M. Bérengier I-INCE Class 23 Laboratoire Central des Ponts et Chaussées, Section Acoustique Routière et Urbaine, Route de Bouaye, BP 4129, 44341 Bouguenais Cedex, France QUALITATIVE EVALUATIONOF SOUND REFLECTION BY A BUILDING FACADE Because of the economical cost and the environmental impact of noise in cities, a lot of researches have been done in sound propagation modelling in urban areas. Most of the analytical or numerical models need to simulate the sound reflection by a building façade. The specular reflection apart, only some statistical reflection laws (like the Lambert's law) are used in practice, although their existence have not been proved for real façades. Moreover, some recent results also suggest there are considerable differences between sound fields in a street resulting from different kinds of boundaries. At this time, we think that it is necessary to develop new reflection laws which take into account a real architectural façades description. As a preliminary step of this further work, we present in this paper a qualitative approach in order to study the sound reflection by a building façade. First, using the wellknown Rayleigh criterion to determine what values of wavelength, surface roughness and angle of incidence does specular reflection change into diffuse scattering, one can determine the architectural elements which participate to the specular reflection or the diffusion of sound. Moreover, using the concept of "the Fresnel zones", we also show that is possible to find the location and the size of the active scattering region of the building façade. With the help of this qualitative approach, it will be possible to develop new reflection laws, by considering only the architectural elements which take part in the diffusion of sound. Session A8-1, 13:40 h Paper number 360 (Contributed, Oral) Teruo Iwase, Tetsuya Sakuma and Koich Yoshihisa I-INCE Class 24 Faculty of engineering, Niigata University, 8050 Igarasi 2nocho, Niigata City, 950-2181 Japan MEASUREMENTS AND SIMULATIONS ON SOUND PROPAGATION ABOVE SNOW FIELD To know characteristics of sound propagation above snowfield in winter season of northern countries, field measurements were performed on some kinds of snow depths over ground. Snow depth conditions were as about 50cm in 1998, about 100cm in 1999 and deeper almost reached to 2m in ten years ago for filed measurements. From these measurements, it was found that frequency characteristics of sound propagation had specific wide and deep dips that mean sound attenuation for each snow depth condition. For example case that both heights of sound source and receiver were about one-half meter, sound propagated with strong sound attenuation in frequency range lower than Internoise 2001 Abstracts 101 250 or 500Hz including main spectra of human voice. By frequency characteristics as comb-tooth type filter, sound easily propagated in frequency range such as 1kHz or 2kHz including spectra of noisy over tone of instruments music. It was also found that deep dip appears at lower frequency as increasing depth of snow. To establish the prediction method of sound propagation above snowfield, authors tried numerical calculation. This was based on really measured acoustic surface impedance of snowfield. Authors also tried scale model experiments by using fibrous material to simulate snow layer. As the simulated results, both prediction methods made agreement with field measurements. Session A8-1, 14:00 h Jan Hofmann Paper number 482 (Contributed, Oral) I-INCE Class 24 Swiss Federal Laboratories for Materials Testing and Research, EMPA, CH-8600 Dübendorf, Switzerland OUTDOOR SOUND PROPAGATION USING COMBINED RAY TRACING A ray tracing model has been developed to calculate weapon noise. As weapons can be heard up to several kilometres and propagation takes place close to the ground, a detailed consideration of terrain and meteorology is important. The calculation model combines two ray tracing techniques. The basic technique is energetic ray tracing in three dimensions, using the full information of topography and obstacles. Two dimensional ray tracing that maintains the proper phase relationship is performed along the most significant propagation paths to improve the results of the energetic ray tracing. To handle diffuse reflections, solid rock structures and forest borders are divided into discrete areas to collect all incoming sound rays. Secondary sources are placed at each of these areas, emitting the diffuse portion of the collected rays. To model the influence of atmospheric turbulence, jitter is assigned to each point in space. The minimal and maximal size of turbulence can be chosen. This technique guarantees that rays with nearly identical paths will have similar local conditions during the propagation. This computer model might be of general use for the calculation of outdoor sound propagation. Session A8-2: Outdoor noise prediction 2 Chaired by Gilles Daigle and Tony Embleton Session A8-2, 14:20 h Paper number 407 (Invited, Oral) I-INCE Class 76 Jérôme Defrance, Michel Bérengier and Jean-François Rondeau CSTB, 24, rue Joseph Fourier, F-38400 St-Martin-d’Hères VALIDATION AND EVOLUTION OF THE ROAD TRAFFIC NOISE PREDICTION MODEL NMPB-96. PART 2: IMPROVEMENTS BASED ON THEORETICAL METHODS An important experimental campaign has been carried out in France on several typical road sites in order to validate the NMPB-96 model (see part 1: Comparison between calculation and measurement results). The experimentation performed under various meteorological conditions which have been precisely measured, has shown a good agreement between prediction and measurements in most cases. However, an improvement of the method on some particular points is still in progress. In this second part, different problems have thus been studied such as the height of the equivalent road source, especially in the case of a viaduct site, the reflection of sound on the sloped ground in presence of a depressed road, the validity of the model for a small height barrier in addition to an embankment. For 102 Internoise 2001 Abstracts each situation, reference results are obtained using ‘exact’ ray and numerical calculation software: MICADO (BEM in homogeneous medium), PARABOLE (GFPE in inhomogeneous medium) and PROPATE (ray theory in inhomogeneous turbulent medium). Impedance discontinuity effects including or not meteorological conditions (positive vertical sound speed gradient) have also been investigated. Session A8-2, 14:40 h Paper number 447 (Invited, Oral) F. Besnard, H. Lefèvre and V. Zouboff I-INCE Class 76 Service d’Etudes Techniques des Routes et Autoroutes 46 av. A. Briand, BP 100, 92225 BAGNEUX Cedex, France VALIDATION AND EVOLUTION OF THE ROAD TRAFFIC NOISE PREDICTION MODEL NMPB-96 PART 1/ COMPARISON BETWEEN CALCULATION AND MEASUREMENT RESULTS The French road traffic noise prediction model NMPB-96 takes into account the influence of the real weather conditions of the site on sound propagation. In order to complement the model validation and to prepare its final version, a measurement campaign took place on six sites bordering motorways. On each site, sound levels at short and large distance (up to 400 m from the road), weather parameters and traffic were measured on 10 minute periods during at least two weeks so as to observe a large diversity of weather conditions. These 10 minute samples have been classified and averaged according to the weather conditions. The sound attenuation due to propagation has been compared with the predictions given by NMPB-96 for conditions favourable to propagation (positive sound speed gradient) in all directions and for homogeneous atmosphere conditions (roughly no sound speed gradient). Measurement results for whole daytime and night-time periods have also been compared to predictions based on the relevant distribution of weather conditions. The agreement between calculation and measurement results is satisfactory, even at large distances, for microphones with a direct view of the road, and also when the road is screened off by a vertical obstacle. For microphones located below the road level (road on embankment or viaduct), the model overestimates the measurements by 3 to 5 dB(A). A few other specific situations have also been considered to need enhancements in the model. The work carried out to upgrade NMPB is described in part 2: Improvements based on theoretical models. Session A8-2, 15:00 h Paper number 493 (Invited, Oral) I-INCE Class 76 Christopher W. Menge, Grant S. Anderson and Douglas E. Barrett Harris Miller Miller & Hanson Inc., 15 New England Executive Park, Burlington, MA USA 01803 EXPERIENCES WITH USDOT'S TRAFFIC NOISE MODEL (TNM) This paper begins with a brief overview of the source and transmission models used in TNM. The paper discusses critical elements of the model’s implementation necessary for accurate noise prediction from ground-based line sources. Comparisons with measurements showing the accuracy of the model will be presented. Since the TNM has been approved for use in the US since March 1998, recent experiences with the model on several projects have yielded insights into many aspects of PCbased noise prediction from highways as well as other sources. These include data input efficiency, computation speed, operating systems, and the flexibility afforded by Computer-Aided Design (CAD) and Geographical Information Systems (GIS) tools. The paper concludes with a discussion of the cost Internoise 2001 Abstracts 103 and benefit implications of more-precise modeling, such as TNM's, compared to older models that are less precise and have some built-in conservatism to compensate. Session A8-2, 15:20 h Paul H. de Vos Paper number 519 (Invited, Oral) I-INCE Class 76 AEA Technology Rail bv, Concordiastraat 67, P.O. Box 8125, NL 3503 RC Utrecht, The Netherlands OPPORTUNITIES FOR IMPROVED HARMONISED NOISE PREDICTION METHODS IN EUROPE The draft Directive on Environmental Noise, issued by the European Commission, prescribes the use of LDEN as the harmonised indicator for noise from roads, railways, industrial activities and aircraft. The Directive intends to prescribe harmonised methods for the assessment of environmental noise from these sources. The Directive concludes, that none of the prediction methods presently available in the Member States of the EU is adequate to be used as the preferred harmonised method. Therefore optimised harmonised methods shall be developed for future use. This paper presents the outlines of such harmonised methods as proposed in the 5th Framework project HARMONOISE. By separating the noise generation model from the noise propagation model, the latter is made suitable to be used for any noise source. This is the first step of harmonisation. For this principle to be applied consequently, the source should be described in terms of physical parameters such as sound power level and 3D directivity. For railway noise special attention should be paid to the characteristic maintenance aspects. The roughness levels of wheels and rails are predominant parameters of the rolling noise generation, which have not been taken into account in most available models. By describing the source in terms of sound power level a better link with legal noise emission limits, type approval limits for new rolling stock and relevant measurement methods is achieved. Also the relation with scientific tools for noise reduction design can be significantly improved. Thus a comprehensive system of quantities and standards is set up. Session A8-2, 15:40 h Heinz Steven Paper number 542 (Invited, Oral) I-INCE Class 76 TÜV Automotive, Herzogenrath (Germany) CATEGORISATION OF ROAD TRAFFIC SITUATIONS AND CORRESPONDING NOISE EMISSION DATA Within the framework of a research project of the German environmental agency the exhaust emission model "MOBILEV" has been developed. This model combines exhaust emission factors for vehicle classes (vehicle category, vehicle size and emission standard classes) with traffic and fleet composition data to calculate the emissions of road traffic in dependence of road category, traffic situation and traffic load. The reference year ranges from 1980 to 2020. That means, that the model also contains estimates for future emission standards (EURO III, EURO IV, EURO V). The emission factors are based on typical driving pattern for different road categories, traffic situations and road gradients. The same driving pattern were used to calculate noise emission factors for different vehicle classes. To be able to differentiate between tyre/road noise and propulsion noise, driving pattern data were used that contain vehicle speed data as well as engine speed and engine load data. The influence of type approval noise limit reductions on the noise emission for different traffic situations is modelled based on measurement results of a series of research projects. Tyre/road noise and propulsion noise are separately calculated. The tyre/road noise is calculated on the basis of an average tyre in dependence 104 Internoise 2001 Abstracts of the road surface, ranging from porous asphalt to transversally grooved cement concrete. The propulsion noise calculation takes into account the increase of noise with engine speed and engine load. As results of the calculation LD, LE and LDEN are provided, it is also possible to get the contributions of each vehicle class to the overall Leq on an hourly basis. Session A8-2, 16:00 h Birger Plovsing Paper number 669 (Invited, Oral) I-INCE Class 76 DELTA, Building 356, Akademivej, DK-2800 Kgs. Lyngby, Denmark NORD2000. COMPREHENSIVE MODEL FOR PREDICTING OUTDOOR SOUND PROPAGATION. EXPERIENCE FROM VALIDATING THE MODEL. A Nordic project is in progress to elaborate prediction methods for various types of environmental noise sources. The prediction methods are based on a common comprehensive sound propagation model which provisionally has been completed at the end of year 2000. The propagation model which is based on geometrical ray theory and theory of diffraction has been elaborated for calculating onethird octave band attenuation in a homogeneous atmosphere. The model has subsequently been extended to include refraction by geometrical modifications of rays. The comprehensive model is applicable for any terrain profile assuming that the terrain is approximated by a number of straight segments. Each segment is characterised by its surface impedance and roughness. The model combines solutions for base models (flat terrain, one or two screens with a flat surface before, after or between the screens) by means of Fresnel-zone interpolation. The theoretical models have been extended to include effects of partial coherence between rays and averaging effects to obtain better agreement with outdoor sound propagation measurement results. The method has been validated by comparison with measurements and with predictions by accurate prediction methods. Experience from the validation has been or will be used to improve the model. The paper will describe the experience gained so far by validating the model. Session A8-2, 16:20 h Paper number 675 (Invited, Oral) J.D. Van der Toorn, T.C. Van den Dool and W.J.A. Van Vliet TNO TPD, Stieltjesweg 1; Postbox 155, NL-2600 AD Delft, The Netherlands I-INCE Class 76 SOUND EMISSION BY MOTOR VEHICLES ON MOTORWAYS IN THE NETHERLANDS: 1974-2000 In 1974, 1983 and 1996 noise emission of motor vehicles in freely flowing traffic on motor-ways have been measured for a road surface of dense hot rolled asphalt. The classical, single micro-phone, passby method has been applied to make the three series of experiments directly comparable. In 1996 a synthetic acoustic antenna has been applied in parallel, in order to get a vertical source strength distribution - as an input for LAeq prediction methods - and to effectively separate source strength and transmission effects, that bother the interpretation of classical measurement results. The classical measurement results in dB(A) show that the sound emission of the different vehicle types did not change significantly in the last two decades, although type test limits for sound emission have been lowered up to 10 dB(A) during that period. The classical data for 1996 have been complemented with results of measurements that M+P performed in 1999 for lower driving speeds. From the complete set the actual acoustical source strengths of vehicles have been derived as a function of driving speed and frequency, for the benefit of the update of existing prediction schemes. The acoustical antenna has been applied in 1/1 octave bands. One or two partial sound source strengths have been detected, Internoise 2001 Abstracts 105 depending on octave band and on vehicle type. The refined source description enables to improve the accuracy of predictions of LAeq and of sound barrier effects, for example, with more modern prediction schemes. Session A8-2, 16:40 h Paper number 523 (Contributed, Oral) K. Yoshihisa, K. Tatsuda and Y. Okada I-INCE Class 24 Faculty of Science and Technology, Meijo University, Shiogamaguti 1-501, Tempaku-ku, 468-8502, Nagoya, Japan TEMPORAL FLUCTUATION OF ATMOSPHERIC ABSORPTION AT HIGH ALTITUDES DURING A YEAR In order to predict outdoor sound propagation more accurately, it is necessary to examine the fluctuation of the attenuation coefficients for atmospheric absorption. There are large variations with time in actual meteorological conditions on which the attenuation coefficients depend strongly. These natural variabilities are often neglected in environmental impact assessments. The Attenuation coefficients were calculated from meteorological data observed at Tateno Observatory in Japan by using the calculation method described in ISO 9613-1. The meteorological data were observed at 9:00 and 21:00 in the year of 1991 by using a radiosonde. The calculated results show that the attenuation coefficients for 4kHz at the height of 31m above the ground are higher in winter than those in summer, although the attenuation coefficients at the height of 3000 m become high in summer, and that the ranges of variation of attenuation coefficients at a height of 3000m tends to be higher than those at a height of 31m. Next, the effects of atmospheric absorption on aircraft noise propagation were examined using the calculated attenuation coefficients at high altitudes up to 5000 m. The sound pressure levels of noises at the ground generated from an aircraft that flies overhead at various heights from 200m to 5000m were calculated, assuming that the atmosphere is modeled by a stack of horizontal layers. As a result of this study, it has been found that the attenuation coefficients for atmospheric absorption vary strongly due to the changes of actual meteorological condition during a year in Japan. Session A8-2, 17:00 h Paper number 112 (Contributed, Oral) Cornelius (Neil) Huybregts and Stephen Samuels Marshall Day Acoustics, Collingwood, Victoria, Australia I-INCE Class 76 EXPERIENCES WITH IMPLEMENTING THE AMERICAN FHWA TRAFFIC NOISE MODEL (TNM) IN AUSTRALIA The American FHWA Traffic Noise Model (TNM) has been evaluated in Australia by a number of organisations over the last couple of years. These evaluations have included two major investigations into the accuracy of the TNM calculation method and trial practical implementations on noise barrier design projects. One of the major investigations, conducted in Victoria, indicated that the TNM method may be more accurate than the UK Calculation of Road Traffic Noise (CRTN) method, and calibration factors were estimated, using a risk analysis approach, for the use of TNM in Victoria. However, the results of more recent investigations indicate that the TNM method may not necessarily be more accurate, but is at least of comparable accuracy to that of the CRTN method. Implementation of the TNM model has not gone smoothly in Australia. There appear to be significant problems with 106 Internoise 2001 Abstracts the TNM software package, notably long calculation run-times and difficulties with importing terrain data. Session A8-2, 17:20 h Paper number 452 (Contributed, Oral) I-INCE Class 76 G. Taraldsen, M. Bjørhus, V. Henriksen, U.R. Kristiansen, S. A. Storeheier SINTEF Telecom and Informatics - Acoustics, 7465 Trondheim, Norway A COMPUTATIONAL MODEL FOR OUTDOOR SOUND PROPAGATION We present a linear mathematical model for outdoor sound propagation based on a two- domain airground formulation. In both subdomains we use the same type of system of partial differential equations, modeling wave propagation in a porous medium with rigid frame. In the air domain, however, the choice of parameters ensures that the system reduces to the usual linearized equations for sound propagation in air. We propose boundary condition to be used on the artifical boundaries that are needed in a corresponding numerical model, and construct interface conditions designed to give a stable and physically correct coupling between the air and ground subdomains. The numerical method for the solution is based directly on the two-domain air-ground formulation of the model, but also a domain decomposition of the two physical subdomains into a larger number of subdomains. We use the Chebyshev spectral collocation method for spatial discretization, and a second order explicit Runge-Kutta method for the time integration. A simple correctional technique is used for imposing the boundary conditions and interface conditions in each time step. The simulation results are compared with measurement results from sound propagation outdoors. Session A8-2, 17:40 h Lowi Sturrus Paper number 646 (Contributed, Oral) I-INCE Class 52 DGMR Consulting engineers, P.O. Box 82223, 2508 EE The Hague THE ATTENUATION OF SOUND IN RESIDENTIAL AREAS To gather more knowledge about the reducing effect of residential areas on the propagation of sound, dgmr consulting engineers is investigating the acoustical situation for a number of locations. The purpose of the measurements is twofold: - We want to get insight in the value of the effect and the way different aspects (reflections and screenings) contribute to the total effect; - Is it possible to establish the reducing effect with good accuracy by making use of a calculation method. For each location the reducing effect is determined on the basis of the measurements. These results are compared with calculated values based on several calculation methods like the ISO-standard ISO-9613, the German guideline VDI-2714, the Dutch I.C.G.-publications GF-HR-01-03 and GF-HR-01-05 and/or a method developed by dgmr. From this several conclusions can be drawn. The results and the conclusions of the investigation will be presented in the paper. Internoise 2001 Abstracts 107 POSTERS Session A8-2, 18:00 h Paper number 139 (Contributed, Poster) Posterboard BP-8 Wednesday, Author present 9:40 – 10:40 h Nese Yugruk Akdag Yildiz Technical University, Architectural Faculty, 80750, Besiktas, Istanbul, Turkey EFFECTS OF REFLECTIONS FROM SURFACES ON ENVIRONMENTAL NOISE I-INCE Class 23 The subject of noise control effects the main planning principles both macro (urban planning scale) and micro (building planning scale) scales. In order to realise the comfort conditions in terms of outdoor sound, noise caused by traffic and industry should by controlled by land use and zone planning. In terms of urban acoustics, one of the factors effects environmental noise is the sound reflection from the surfaces round the noise source. The aim of this paper is to determine that how the reflections from surfaces increase noise in the different conditions. Especially, taking into consideration road traffics, the effects of reflections from facades were investigated by using a software. It has been determined that there are some relations between noise increasing, size of reflecting surface and the receiver distance from the noise source. Findings will be illustrated with tables and graphics. Session A8-2, 18:00 h Paper number 270 (Contributed, Poster) I-INCE Class 23 Posterboard BP-9 Wednesday, Author present 9:40 – 10:40 h François-Xavier Bécot, Pontus J. Thorsson and Wolfgang Kropp INRETS, Transport and Environment Laboratory, F-69675 Bron, France NOISE PROPAGATION OVER INHOMOGENEOUS GROUND USING AN EXTENDE RAYLEIGH INTEGRAL METHOD Sound radiation from a source placed over finite impedance ground is of major interest in several areas of acoustics, such as outdoor sound propagation or sound radiation from tyres. The general case of a radiating point source over a surface with inhomogeneous impedance and arbitrary shape is usually handled by integral equation methods. For the two-dimensional case, an analytical far field approximation based on plane wave expansion has been proposed in literature for homogeneous ground impedance and flat terrain. In this paper, the problem of a coherent line source radiating above a ground surface of arbitrary impedance and profile is solved by using an alternative method to the boundary integral equation. Distributed sources are placed at the ground surface, and the resulting sound field is expressed through the Rayleigh integral. The source strengths are adjusted to fulfil the prescribed locally reacting ground impedance. The obvious singularity problem is solved analytically by deforming the integration path of the Rayleigh integral. The calculation time for this method is drastically reduced for flat ground compared to standard boundary element methods. One advantage of the present method is the possibility to use arbitrary order of the primary source. Moreover, this solution is valid for the near field and for the far field. Satisfying accuracy in the results can be achieved by an adapted discretization of the ground surface, to the detriment however, of the calculation speed. The validity of the Rayleigh integral method has been confirmed through comparisons with other models and with measurements. 108 Internoise 2001 Abstracts Session A8-2, 18:00 h Paper number 454 (Contributed, Poster) I-INCE Class 24 Posterboard BP-10 Wednesday, Author present 9:40 – 10:40 h Jérôme Defrance, Jean-François Rondeau, Jean-Marc Abramowitch and Nicolas Barriere CSTB, 24, rue Joseph Fourier Centre de Villaroche, F-38400 St-Martin-d’Hères, France THEORETICAL INVESTIGATION ON THE DISPERSION OF ROAD TRAFFIC NOISE MEASUREMENT RESULTS When traffic noise measurement results are shown to the residents, they sometimes invoke very particular meteorological conditions during experiments in order to cancel their validity. This refusal may be justified when measurements have been performed under meteorological situations not representative of the site. Residents and measurers thus need to know for a given typical site (embankment, cutting, with a barrier added) the dispersion of the measured LAeq at a particular receiver due to the changing of meteorological conditions along the year. To meet this request, numerical calculations have been carried out using a fast resolution of the Parabolic Equation (GFPE) with a set of 45 sound speed profiles representative of a large range of meteorological conditions, night and day. The results presented as the statistical probability of respecting a LAeq value are obtained (for given receiver position, orientation of the road, type of site and climatic zone) by crossing, for each sound speed profile, the calculation with its probability of occurrence in one full year. Specific indices as L50 and (L25-L75) can then be obtained. Session A9-1: Low frequency noise control Chaired by M. Koyasu Session A9-1, 9:40 h Paper number 747 (Contributed, Oral) Sarah B. Knight, Jack B. Evans and Chad N. Himmel I-INCE Class 12 Engineered Vibration Acoustic & Noise Solutions, 5806 Mesa Drive, Suite #380, Austin, Texas; 78756; USA CASE STUDY: AIR COOLED CHILLERS WITH ROTARY HELICAL SCREW COMPRESSORS TONAL NOISE MITIGATION AT HOSPITAL NEAR PATIENT ROOMS, RESIDENTIAL NEIGHBORHOODS AND OPEN PARK Air cooled with rotary helical (screw) compressors were producing disturbing tonal noise in patient rooms, an adjacent open park, as well as nearby neighborhoods. A low frequency ‘buzz’ or ‘hum’ produced a distinct tone measured at 117 Hz. The primary noise source was compressor discharge and casing radiation. Analysis was undertaken and recommendations developed to decrease the tonality of the chiggers by 5 to 8 dB in the125 Hz third octave band within the surrounding neighborhoods, to reduce perceived annoyance according to Composite Noise Rating (CNR) analysis acceptability levels. Mitigation recommendations included: (a) attenuate radiated compressor noise by wrapping the compressors with an insulation and limp mass noise barrier jacket, (b) attenuate compressor discharge noise by inserting pulse diffusers in the compressor discharges, (c) control the intake noise by providing acoustic louvers, (d) control compressor noise escaping through the discharge by providing discharge attenuators, and (e) reduce reflected sound from building surfaces. Sound level measurements were made after the implementation of the first two recommendations. JEA found that Internoise 2001 Abstracts 109 ~ a 5 dB tonal reduction had been attained by the implementation of the two recommendations. Rather than the mass loaded jacket that had been recommended, a sheet metal enclosure was constructed. The sheet metal faces of the enclosure re-radiated compressor noise into the environment. ‘Before’ and ‘after’ measurements, the attenuation design, and effectiveness of the mitigation efforts will be discussed. Several different types of mitigation between the source and receivers were taken into consideration, which makes this case study particularly interesting. Session A9-1, 10:00 h W. Soede Paper number 351 (Contributed, Oral) I-INCE Class 43 ARDEA Acoustics and Consult, Harmen Doumastraat 24, NL-2321 JL LEIDEN, The Netherlands RESEARCH FOR A GUIDELINE FOR LOW-FREQUENCY NOISE INDUCED BY VIBRATIONS OF UNDERGROUND RAILWAYS Nowadays, underground construction becomes more and more important to enable new infrastructure. New tunnel-bore techniques enable the construction of underground roads and railways near or below houses and offices in densely packed city areas. Due to the short distance between the foundation of the buildings and the new tunnel it is often important to apply measures against vibrations and low frequency noise. Towards vibrations, design-targets can be based on special standards like the Dutch SBR-guidelines and the German DIN4150. However, no adequate guidelines can be found for low frequency noise. Participants within the Dutch Center for Underground Construction (COB) have initiated a project to create a guideline to handle low frequency noise induced by vibrations of underground railways. This was felt necessary to facilitate new projects in the Netherlands like the subway in Amsterdam (North/Southline), the high-speed train Paris-Amsterdam (HSL) and the cargoline to Germany (Betuweroute). The project should answer questions for a preferred maximum noise level on a train passage and an upper boundary. This upper boundary is felt necessary for situations where the preferred level can not be reached with available techniques or appear to be too costly. Those levels should be based on knowledge on annoyance by human beings due to this special type of low frequency noise. The project is divided in an inventory of available knowledge around the world and background research to raise insight on the annoyance in relation to other types of noise. The results of a literature search by the participants in the project and an inventory of opinions of several Dutch and German experts in the field will be summarized and discussed. Session A9-1, 10:20 h Hiroaki Ochiai Paper number 381 (Contributed, Oral) I-INCE Class 61 Kobayasi Institute of Physical Research, 3-20-41 Higashi-motomachi Kokubunji Tokyo 185-0022 Japan THE STATE OF THE ART OF THE INFRA AND LOW FREQUENCY NOISEPROBLEM IN JAPAN There has been an increase in the number of the complaints against infra and low frequency noise in Japan. However, there are still no regulations for the management of this noise. A techinical committee of low frequency noise proposed a method for the measurement and evaluation of infra and low frequency noise in 1998. Based upon this proposal, Environmental Agency of Japan issued a manual for the measurement of low frequency noise in October in 2000, they also had an intention to gather field data in all over the country. In this report, we describe the state of the art of the infra and 110 Internoise 2001 Abstracts low frequency noise problems in Japan. We also describe the content of the manual and the future issues on this subject. Session A9-1, 10:40 h Rob Hallows Paper number 114 (Contributed, Oral) I-INCE Class 69 Winstone Wallboards NZ Ltd, PO Box 12 256 Penrose, Auckland, New Zealand LOW FREQUENCY FOOTFALL SOUNDS: AUDIBILITY AND PERFORMANCE OF LIGHT TIMBER FRAMED FLOORING SYSTEMS An audio recording of a bag-drop simulating footfall sound was recorded on a light timber framed/plasterboard intertenancy floor-ceiling system. Subjects listening to the recording via headphones performed three tasks: - Judgement of just noticeable differences for high and low levels of impact sound; - Judgement of threshold of detectability of impact sound in the presence of typical background noise; - Judgement of doubling impact loudness. Additionally, impact sound transmission of a wide variety of floor-ceiling systems were measured in controlled conditions in a reverberation suite. Low frequency transmission of the different systems were compared to illustrate the range of impact levels for different types of systems (eg higher mass; floating floors; floor coverings). The significance of the impact performances of the different floor-ceiling systems was assessed using the results from the subjective sensitivity experiments. The assessment was used to suggest categories of performance quality for light timber framed floors at low frequencies. Session A9-2: Mufflers, silencers and pipes Chaired by Eddy Gerretsen and Alan H. Marsh Session A9-2, 11:00 h Paper number 696 (Contributed, Oral) J.L. Horner, X. Huang, J.J. McGuirk, G.J. Page and Y. Zheng I-INCE Class 26 Dept. of Aero. And Auto. Engineering, Loughborough University, Loughborough LE11 3TU, U.K. THE PERFORMANCE OF A PRESSURE-BASED CFD ALGORITHM IN SOME BENCHMARK COMPUTATIONAL AEROACOUSTIC TEST PROBLEMS There has been recently significant growth in interest in simulating the acoustic properties of aeroengine intake duct systems. The background application is, for example, acoustic fatigue assessment, where the acoustic excitation arises from rotating shock systems attached to transonic fan blades. Much work has also been underway to develop Computational Fluid Dynamic (CFD) methods, which are adequate for shock capture and aerodynamic simulation, to also be in a position of accurate resolution of acoustic phenomena. To this end, workshops on Computational Aeroacoustics (CAA) have been organised, particularly by NASA, and have lead to a series of CAA benchmark problems. To date most CFD methods applied to these problems have been derived using explicit density-based algorithms and have implied that high-order accuracy (at least fourth order) is essential. The proposed paper will describe some recent work applying the alternative pressure-based CFD approach to some of the NASA CAA benchmark problems. This is not a fourth-order accurate method, but it will be demonstrated that adequate results can still be obtained, particularly for strongly non-linear problems such as will be important close to rotating shock systems. Results for a series of 1D test problems relevant to duct acoustics will be presented; these will cover linear and non-linear plane wave Internoise 2001 Abstracts 111 propagation, including the well-known Morfey-Fisher test case of saw-tooth or N-wave formation. Predictions for these non-linear problems are, in terms of amplitude and phase errors, as good as or better than previous results presented in the NASA workshop, implying that further study of the pressure-based approach for CAA problems is worthwhile. Session A9-2, 11:20 h Paper number 75 (Contributed, Oral) Wan-Ho Jeon and Joonkeun Lee Digital Appliance Research Lab., LG Electronis, 327-23, Kumchon-gu, Seoul, Korea I-INCE Class 34 A STUDY ON THE NOISE REDUCTION METHOD OF A CENTRIFUGAL COMPRESSOR The turbo chiller uses centrifugal compressor, which operates at about 14500 rpm. Due to the high rotating velocity of the impeller, the noise of chiller makes one of the serious problems. The possibility of the sound reduction by using absorbing material and silencer is studied in this paper. The generated sound propagates through the duct and then transmitted and radiates to the outer field. Higher frequencies noise is transmitted through the elbow duct, but lower frequencies noise are transmitted through the condenser wall. Measurement shows that the elbow duct is the main radiator of noise. So, the use of silencer inside the duct is one of the effective methods. The use of sound absorbing material on the elbow duct shows 4dB noise reduction in the high frequency. The use of sound absorbing material on the condenser wall shows 4dB noise reduction in the mid frequency range. Session A9-2, 11:40 h Paper number 116 (Contributed, Oral) Yadong Lu, Xiaodong Li, Jing Tian and Wen Wei I-INCE Class 34 Institute of Acoustics, Chinese Academy of Sciences, 17 Zhongguancun St., P. O. Box 2712, Beijing 100080, P. R. China THE PERFORATED PANEL RESONATOR WITH FLEXIBLE TUBE BUNDLE AND ITS ACOUSTICALMEASUREMENTS A new sound absorptive structure, i.e. the perforated panel resonator with tube bundle, which has been granted the Chinese patent, is presented in this paper. The sound absorbing structure consists of the perforated plate and back sheet and side plates which form an enclosed resonant cavity as well as a flexible tube bundle. Each flexible small tube connects with the hole of the perforated facing plate. The length of the tube can be much greater than the cavity depth, and be made different from one another to absorb sound at a different frequency. The incorporation of the tube bundle into the acoustical perforated panel is expected to widen the absorptive bandwidth of frequency and increase sound absorption. In fact, through the acoustic measurements in the impedance tube, the following conclusions can be derived. Not only the absorption coefficient of the innovative acoustical absorbing structure is much bigger than that of the common perforated panel absorber, but also its frequency bandwidth of absorption is broaden considerably comparing with that of traditional perforated panel absorber. Except for these, there are two resonant peaks in the frequency spectrum of sound absorption for the perforated panel resonator with tube-bundle. Its entire absorptive properties have come up to and even exceed that of the two-layered microperforated panel absorber. It can be widely used in the case of strictly space limitation and expecting to increase the sound absorption at lower frequencies and further widen the frequency range of absorption. 112 Internoise 2001 Abstracts Session A9-2, 12:00 h Paper number 202 (Contributed, Oral) J. Xia and A. Cummings Department of Engineering, University of Hull, Hull HU6 7RX, UK THE ACOUSTIC IMPEDANCE OF LEAKS IN EXHAUST SILENCER BAFFLES I-INCE Class 34 Baffle leakage in exhaust silencers is caused by spot-welding of the baffles to the chamber walls, which leaves an irregular and intermittent gap around the edge. These leakage paths can affect the behaviour of resonators that are partly bounded by such baffles and need to be taken into account in predictive design methods for silencers. The acoustic characteristics of baffle leaks do not appear to have been previously investigated. In this paper, both steady flow tests and acoustic tests for the characterisation of baffle leakage are reported. Steady flow tests yield data on both the flow area and the zero-frequency acoustic resistance. These data may be used in prediction methods based on existing models for the acoustic impedance over the full frequency range. Comparison between measured and predicted impedance data shows satisfactory agreement for engineering purposes. The leakage flow area inferred from the steady flow tests compares well to the directly measured flow area. It is concluded that steady flow tests can yield useful data for acoustic predictions, from very simple test methods. Non-linear and through-flow effects are also discussed here. Session A9-2, 13:20 h Paper number 300 (Contributed, Oral) Jing Tian and Ziaodong Li NOISE CONTROL FOR AN AIR-CONDITIONER I-INCE Class 34 Institute of Acoustics, Chinese Academy of Sciences, P.O. Box 2712, Beijing 100080, P. R. China The specification of noise level for household appliances becomes more and more important in the competition of the mainland market of China. One reason is that other specifications for appliances come to almost the same level for different manufacturers. The other is that the consumers in this market are much more sensitive to noise because their apartments are generally smaller than in western countries. The refrigerator, for example, might have to be placed in the bedroom. It is presented that there is a gap between the national standards on noise levels for household appliance products and for the environments in living rooms. Manufacturers are striving to fit the lower level of living environments. That needs a thorough understanding of the noise from a given appliance. Noise control for air-conditioners is a representative problem in household appliances. The main noise sources include a centrifugal blower, an axial fan, an electrical motor and a compressor. Noise comes from both the direct radiation in air and the structural vibration. In this paper, noise from a windowfixed air-conditioner is measured and discussed. Noise sources are separated and characterized with their coherent and non-coherent power spectrum analyses. Micro-perforated panels, flow-steady mess, structural stiffness and other methods are applied to attenuate different types of noise accordingly. Internoise 2001 Abstracts 113 Session A9-2, 13:40 h Paper number 508 (Contributed, Oral) I-INCE Class 34 Jeong-Guon Ih, Chul-Min Park, Yoshio Nakayama, Senji Kitahara and Hideo Takao Korea Advanced Institute of Science and Technology, Science Town, Taejon, Korea (305-701) ACOUSTICAL EFFECT OF THE NONLINEAR PROPERTY OF POROUS WOVEN HOSES IN ENGINE INTAKE SYSTEMS The porous tube of coated woven fabric is recently used as an effective component of intake systems of many internal combustion engines to reduce the intake noise. The acoustic characteristics of porous woven hoses have been specified by the so-called 'porous frequency', which is experimentally determined from the ambient pressure, the internal volume of the hose, and the ratio between the specific volume flow rate through the hose wall and the effective pressure drop at this specific volume flow rate. The latter ratio is determined by the wall property according to the weaving condition and the coating condition of porous hose. In addition, it is noted that the physical meaning of the companyspecified ‘porous frequency’ is very similar to the flow resistance through a porous material in general. Because of the aforementioned nonlinear property of the porous woven hose, the porous frequencies measured at different radius and/or different length are different even if the hoses possess the same wall property. In this paper, the nonlinear factor of the porous woven hose is investigated experimentally. It is shown that the porous frequency of the porous hose having other radius or length is predicted, based on this nonlinear factor. Developed method can be used to specify the acoustic characteristics of porous woven hoses more precisely than before and thus the transmission or insertion loss of the hose can be predicted with precision. Session A9-2, 14:00 h Paper number 555 (Contributed, Oral) A.D. Sahasrabudhe, Nimesh Bhandari, Ashish Shrivastava AN EXPERT SYSTEM FOR THE DESIGN OF SILENCERS I-INCE Class 34 Department of Mechanical Engineering, Indian Institute of Technology, Guwahati, INDIA The problem of noise from engines and design of silencer elements is known for quite some time. Design of silencers is being carried out through the extensive use of computers for quite some time. However, there has been a long felt need for developing an expert system for the design of silencer. The paper deals with the development of an Expert system to design silencers based on the standards desired by the user. The basis for the design is the engine noise characteristics data which is either available in the form of Noise Level (dB) vs. frequency (Hz) graph or is collected online while the engine is running and fed into the software after FFT analysis. In the previous case, the data is retrieved and stored in the tabular form by scanning the graph using image processing techniques. This is then used to determine the global as well as local maxima corresponding to the maximum level of noise generation. An appropriate silencer element is then designed based on maximum noise level encountered from the engine and the desired level of noise based on noise standard or regulation. The constraints on space and geometry are also taken into account. An algorithm for the design of silencers comprising simple expansion chamber elements and extended tube elements has been developed and implemented. In this algorithm, design is based on hierarchy of complexity of elements based on Munjal et. al. Accordingly, an attempt is first made to design a silencer with the simplest type of element. If this does not succeed, algorithm attempts to design next higher element. The development of algorithm for multiple expansion chambers and Helmholtz resonators is being attempted. 114 Internoise 2001 Abstracts Implementation of the algorithm avoids the conventional iterative procedure for design of silencers and reduces the overall design cycle time. The software thus developed has been tested for the case of noise levels of the order of 100 dB and up to a frequency limit of 4000 Hz. The results have been satisfactory and encouraging. Session A9-2, 14:20 h Paper number 578 (Contributed, Oral) Chang-Myung Lee and Jong-Hoon Sim Dept. of Automotive Eng. University of Ulsan, Mugue-dong, Nam-ku, Ulsan, Korea I-INCE Class 34 DEVELOPMENT OF TRANSMISSION LOSS PREDICTION SOFTWARE OF A MUFFLER ON WEB BASE To predict transmission loss of muffler, several methods have been introduced such as numerical methods called FEM or BEM, and a 4-pole parameter method using analytical approach. Both methods, however, have limitation to predict transmission loss of muffler since acoustic circuit of main muffler is so complicated. Although 4-pole parameter method can be a powerful tool to predict transmission loss of a muffler, analytical approach only can not provide enough acoustic circuit of muffler. Therefore, if the numerical method is coupled into analytical approach in 4 pole parameter method, it will be a very powerful technique called a hybrid method. To adopt numerical method on 4pole parameter method, it can be utilised only with numerical data base. Thus, we have to run a commercial FEM or BEM software to get numerical data base for each case of acoustic element. Which means we need at least 100 or 200 data base for one acoustic element of muffler. To prepare a powerful package with the hybrid method for transmission loss prediction of muffler, we need more data base. Whenever we update data base of acoustic element of muffler, we do not need to prepare update version for each users if we utilize web network. If the package is developed on web base, it can be used by any users of world simultaneously. If we update the software once with a server computer, any users in world can use the software with new version. Specially, if we have to update a commercial software frequently or we need to distribute a software quickly with low price, software on web base will be another way in acoustic area also. Session A9-2, 14:40 h Paper number 448 (Contributed, Oral) Michihito Terao, Hidehisa Sekine and Shinpei Ohkawa I-INCE Class 37 Kanagawa university; Rokkaku-bashi 3-27-1, Kanagawa-ku, Yokohama, 221-8686, Japan EFFECT OF ACOUSTIC LINERS ON SOUND TRANSMISSION OF WALL APERTURES FOR VENTILATION A wall aperture for ventilation often becomes a fatal week point in the sound insulation performance of a room of a dwelling house. Sound efficiently goes through the aperture especially when the frequency coincides with one of the resonant frequencies of the axial modes. G. P. Wilson and W. W. Soroka gave an approximate solution for the transmission loss of a circular aperture in a plane rigid wall of a finite thickness when a plane sound wave incidents normally on the aperture and when the perimeter wall of the aperture is rigid and loss free. In this study, the sound transmission characteristics of fundamental wall apertures of ventilation use were investigated by a numerical approach for not only normal incidence but also oblique incidence, and for not only rigid perimeter but also acoustically lined perimeter of the aperture. The numerical approach employed was a boundary element method whose effectiveness was confirmed by experiments carried out in an anechoic room. Internoise 2001 Abstracts 115 Consequently, it is found that the insertion loss of the wall with ventilation aperture is independent of the direction of the incident wave. Introduction of acoustic lining on the perimeter wall of the aperture make the axial resonant frequency lower, and dissipates significantly the sound of frequency higher than the first axial resonance, but give only a little attenuation of the sound of the first axial resonance. Session A9-2, 15:00 h Paper number 53 (Contributed, Oral) Ferrari Giancarlo and Onorati Angelo OPTIMIZATION OF I.C. ENGINES INTAKE AND EXHAUST SYSTEMS I-INCE Class 76 Department of Energetics - Politecnico di Milano, Piazza L. da Vinci 32, 20133 Milano - Italy In the field of reciprocating internal combustion engines, the calculation of unsteady wave motion in the exhaust duct system allows to predict the effect of muffler geometry on radiated noise and engine performance. Geometrical parameters of i.c. engine intake and exhaust systems are generally optimised both from the acoustic and engine performance point of view. At the moment onedimensional fluid dynamic codes let to calculate wave actions through these complex systems, modeling the effects of its typical components. So they are current and useful tools for design and development of new engines, because they allow to choice a few optimised configurations, which can be finally adjusted by experimental tests. In particular, during last decade an extensive research work has been carried out by the authors in this field, to develop an advanced fluid dynamic model for the simulation of one-dimensional unsteady flows in engine ducts. It can solve the fundamental conservation equations both by classical and innovative shock-capturing numerical methods and allows to follow the propagation of typical exhaust chemical species along the exhaust and intake ducts. The complex geometry of typical silencers, including abrupt area changes, axial side-branches, flow reversals, perforated pipes and sound absorbing material, can be successfully represented via a one-dimensional non-linear approach. In order to highlight the features of this model, the paper reports some typical results of a comprehensive optimisation work done on the intake and exhaust systems of different types of engines. Comparisons between predicted and measured data will show the code accuracy in predicting pressure pulses just before and after the main components, and their influence on global engine performance. Session A9-2, 15:20 h Paper number 78 (Contributed, Oral) I-INCE Class 76 Ronaldo Fernandes Nunes, Denis de Lima, Celso Figueiredo Nogueira and Marcos Antonio Argentino Humaitá IT Services Latin America, Al. Campinas, 1070 – Jd. Paulista – 01404002 – São Paulo / SP AIR INTAKE SYSTEM MODELING FOR VIBRO-ACOUSTIC ASSESSMENT This paper describes the use of Vibrato-Acoustics numerical modelling for the prediction of an Air Intake System noise level for a commercial vehicle. The aim of this work it is to assess the VibratoAcoustics behaviour of the Air Intake System and the influence of it in the sound pressure level of the vehicle. For the commercial vehicle with 5 cylinders engine, a sweeping from 0 to 350 Hz was done, aiming on verifying the influence of the generated sound pressure level with respect to the engine rotating frequency, i.e., 2nd 1/2 order at 175 Hz and 4200 rpm engine peak rotation, and 5th order at 350 Hz. The results have shown that there is a big potential to influence the radiated sound pressure level (at high frequencies) for the areas represented by the intake duct, medium duct and air filter. Overall, the acoustic analyses have shown that an increase in stiffness on this zones can reduce the 116 Internoise 2001 Abstracts acoustic energy emitted to the exterior. As a conclusion, the use of the Helmholtz Resonators close to the turbine region have been suggested to reduce the sound pressure level in the air intake system when sized for specific frequencies, as for example in the range between 10 to 40 Hz, which was verified in the experimental measurements. POSTER Session A9-2 Paper number 18 (Contributed, Poster) Posterboard AP-23 Tuesday, Author present 16:00 – 17:00 h Ming Une Jen and Ming-Hung Lu I-INCE Class 34 Industrial Technology Research Institute, Bldg. 11, 195-3 Chung Hsing Road, Section 4, Chutung, Hsinchu, Taiwan 310, R.O.C. REFINEMENTS OF A PRODUCTION MUFFLER TO MEET THE DEFINED EXHAUST NOISE TARGET For the development of exhaust system, using the available production muffler is a preferred means to reduce the manufacturing cost. However, in order to meet the specific designed NVH target, the muffler modification and refinement process is unavoidable. This paper presents a case study of the exhaust noise development, and the specific emphasis is placed on the trouble-shooting process, muffler design modification and experimental NVH verifications. The first modification of production muffler was the proportional size reduction due to the package concern. Out of different size reductions, the one with relative lowest exhaust noise was chosen, but it can not meet the target below 2500 rpm. Further order analysis shows that the 6th and 8th orders are the dominant components below 3000 rpm and they both correspond to the single noise frequency of 256 Hz. To solve the problem, an individual Helmholtz resonator was designed and verified at first. Then the original builtin resonator inside the muffler was tuned to the optimum dimensions to account for the Helmholtz effect. It was verified that 5 to 10 dB(A) improvement of the 6th and 8th order exhaust noise at major concerned speeds for the final muffler version. It is shown that the final muffler version can fully satisfy the target of exhaust noise as well as the targets of pass-by and stationary noise, and raises no engineering concern. The trouble-shooting methodology and muffler design rules employed in this case study can serve as a useful reference for the related field. Internoise 2001 Abstracts 117 Session A9-3: Aero-acoustic sources Chaired by Alex de Bruijn and Marli Vogels Session A9-3, 15:40 h Hajime Fujita Paper number 182 (Contributed, Oral) I-INCE Class 11 College of Science and Technology, Nihon University, 1-8 Kanda Surugadai, Chiyoda-ku, Tokyo 1018308, Japan THE CONTROL OF AEOLIAN TONE GENERATED FROM TWO-DIMENSIONAL CYLINDERS WITH CROSS SECTION MANIPULATION Control of the aerodynamic noise generated from two-dimensional circular and square cylinders are studied experimentally in a low noise wind tunnel. The characteristics of the peak level and the Strouhal number of the Aeolian tone for square cylinders of 2x2 cm cross section with 0, 5 to 25% rounded corners, a half square with semi-circular cross section cylinder (U shape cross section cylinder) and circular cylinders of diameter of 0.8 and 1 cm wound with thin spiral wires are studied in the velocity range between 10 and 30 m/s. The effects of the corner rounding or the spiral wire have been known only qualitatively so far. In this study, these effects are presented quantitatively with the consideration of the flow characteristics, particularly the separation of the flow and the formation and distraction of the Karman vortices. From the series of experiments, it is concluded that the peak level is lower when the separation point of the flow is firmly fixed at a sharp edge while the possible movement of the separation point over a round surface results in higher Aeolian tone generation. The effect of the spiral wire of diameter, d, wound on a circular cylinder of diameter, D, depends on the diameter ratio d/D and the spiral pitch, p. The formation of the Karman vortices can be distracted and the peak level is considerably decreased when d/D is more than 0.2 and p/D is more than 3. Session A9-3, 16:00 h Matjaz Prek Paper number 189 (Contributed, Oral) I-INCE Class 11 University of Ljubljana, Faculty of Mechanical Engineering, Askerceva 6, SI – 1000 Ljubljana, Slovenia THE IMPACT OF GEOMETRICAL PARAMETERS ON HYDRODYNAMIC NOISE GENERATION Noise-source mechanisms for simple fluid valve model were observed and compared with predictions. Hydroacoustic scaling laws for the noise sources were derived by dimensional analysis of governing equations. Comparison of the predictions with experimental data indicated that the scaling laws could be further improved with the re-examination of the influence of impact parameters. This was made by extended analyze of the influence of geometrical parameters on acoustical characteristics. Additional parameters were determined statistically with non-linear multiple regression analysis of the experimental data. For the applied geometrical and hydrodynamic parameters the estimated correlation between observed and predicted values are good. 118 Internoise 2001 Abstracts Session A9-3, 16:20 h Paper number 444 (Contributed, Oral) Yannick Knapp, Eric Bertrand and Antoine Buffel du Vaure Université d'Avignon IUT-GCE, Agroparc BP1207, 84911 AVIGNON, FRANCE I-INCE Class 11 EXPERIMENTAL DETERMINATION OF WAVE EQUATION SOURCE TERMS FOR AEROACOUSTIC THEORIES The localisation and quantification of aero-acoustic noise sources is a difficult task because of the complex generation, propagation and interaction mechanisms of such sources. Most aero-acoustic theories are based on the development and coupling of equations of both fluid mechanics and acoustics (if ever there was a separation). The principal disagreement of such theories lies in the problem that the source terms appearing in the resulting wave equation are difficult to measure because they involve double spatial differentiation of velocity components (Lighthill or PowellHowe). Particle Image Velocimetry is a technique used in fluid mechanics for the experimental determination of two-dimensional velocity fields. This technique is based on the determination of local spatial correlations of pairs of tomographic recordings. This recordings, consisting in images of a seeded flow illuminated by a pulsed laser light sheet, are performed with a digital camera allowing the use of FFT algorithms. The knowledge of the local displacement of the seeding particles and the time between the light pulses provides the determination of local velocity components with high accuracy. The determination of most of the source terms by a proper post-processing of these measurements is here made easy by the use of this technique. The work presented here shows the use of this technique in several experimental configurations ,the results obtained for the tests performed, and the limitations and expectations of this technique. Session A9-3, 16:40 h Paper number 685 (Contributed, Oral) G. Hübner and V. Wittstock ITSM, University of Stuttgart, Pfaffenwaldring 6, D-70550 Stuttgart, Germany I-INCE Class 12 INVESTIGATIONS OF THE SOUND POWER OF AERODYNAMIC SOURCES AS A FUNCTION OF STATIC PRESSURE Usually the description of aerodynamically generated sound is based on monopole, dipole or quadrupole source elements. For such ideal elementary sound sources the sound power P depends on the gas density rho and the sound velocity c according: P proportional rho times c power n with n=-1, 3, -5 depending on the source characteristic. But for real aerodynamic sources theoretical derivations assume additional rho-effects such as caused by the Reynolds number influence and the correlation length of the flow structure. Whilst the issued experimental investigations of the sound power generation by aerodynamic sound sources are carried out under more or less undefined acoustical conditions the paper presents results of true sound power determinations using the reverberant field method. The special reverberant room allows measurements in a range of static gas pressures between 1 and 100 kPa. The measurements are carried out for a whirler configuration varying the number, the diameter and the length of the rotating cylinders and for a system of radially assembled cylinders such as usual for a radial fan. Internoise 2001 Abstracts 119 Session A9-3, 17:00 h Paper number 383 (Contributed, Oral) Yongbo Zhu, Ku Liu, Mingkun Chen and Jing Tian I-INCE Class 21 Institute of Acoustics, Chinese Academy of Sciences, 17 Zhongguancun St. P. O. Box 2712, Beijing 100080, P. R. China EXPERIMENTAL INVESTIGATIONS AND THEORETICAL STUDIES ON RIJKE TUBE To investigate the principle of thermoacoustic interaction, a series of experiments on a heat duct (Rijke tube) are done. The influence of heat source location and temperature on the sound pressure and frequency in Rijke tube is provided. Besides the linear characters, the nonlinear phenomena in Rijke tube are presented, such as instantaneous peculiarity, heat source temperature saturation, and how the variation of inlet velocity, outlet acoustical condition, and outlet temperature affects the acoustic field in tube. The results show that the acoustic change in Rijke tube can be highly nonlinear. Finally, a mathematical model for unsteady heat-acoustic interaction in Rijke tube is developed, and a solution is then given by modal analysis. Then the method is used to explain the behaviors of Rijke tube. But up to now, to predict the acoustic field in Rijke tube quantitatively is quite difficult. Session A9-3, 17:20 h Paper number 3 (Contributed, Oral) Andrei Abramian and Sergei Vakulenko I-INCE Class 42 Institute for Problems in Mechanical Engineering, Bolshoy pr. V.O. 61, 199178 St. Petersburg, Russia CAUSE OF UNKNOWN DISCRETE SPECTRUM IN GENERAL SPECTRUM OF STRUCTURES, INTERACTING WITH FLUID Some reasons, which cause formation of undetermined discrete spectrum in the general spectrum of a structure interacting with a liquid, have been studied. They are the following: 1) Localized modes existing in the construction; 2) Inadequate model of the liquid; 3) Non-linear effects. 1) Localized modes causing extra radiation occur because of trapped modes effect. The trapped modes effect was found and studied as a phenomenon of infinite-length structures. The same effect has been found in finite-length construction. Influence of boundary conditions and number of elastic-mass inclusions on the trapped modes has been investigated. The conditions under which the trapped modes correspond to structure multiple natural frequencies of the second order have been found. Author has sequentially solved the problems of the evaluation of the spectrums for the following hinged structures: - a finite length beam on the elastic foundation with two similar elastic supports; - a circular cylindrical shell supported by two circular frames. To solve these problems, the solutions for finite length structures were expressed in terms of Green function for the similar infinite length structures. 2) The second reason has been studied by the example of frequencies of a structure composed of coaxial cylindrical shells submerged into a liquid. The influence of different physical liquid models on the discrete spectrum has been studied. 3) From the nonlinear dynamic point of view, discrete spectrums can be explained by the existence of periodical, subharmonic and quasiperiodic regimes of the structure. The solutions of the problems of nonlinear beam-liquid interaction and the liquid-beam on nonlinear supports interaction have been given. 120 Internoise 2001 Abstracts Session A9-3, 17:40 h Paper number 175 (Contributed, Oral) A.C. Fagerlund, M.P. Norton, D.G. Karczub and F.W. Catron Fisher Control International, Marshalltown, Iowa, USA I-INCE Class 53 PIPING SYSTEM COMPONENT NOISE MODELING FOR IN-PLANT NOISE PREDICTION Tools for predicting industrial noise levels are essential for adequate and economic design of plants to meet occupational and boundary noise limits. An essential element of noise prediction in refineries, power stations and offshore platforms is the sound power due to control valves and piping. The estimation of control valve noise is well established, but an important consideration often overlooked is the noise due to piping and piping components. The principles of noise generation and propagation are the same as for a control valve, but the modeling of noise generated by piping components such as bends and equal tees requires different techniques to those used for control valves. The applicability of non-dimensional scaling of internal wall pressure spectra1 to fill this gap is demonstrated here for bends and equal tees in high-speed gas flow. This technique utilizes experimental data from a selected range of pipe sizes, pressures and flow rates to provide predictions for any combination of pipe size, pressure and flow rate of practical interest. The components tested produced localized external radiated noise levels well above downstream noise levels. The ultimate goal of this work is to assist the plant designer/owner in minimizing plant capital costs by avoiding the unnecessary specification of low-noise valves where they are not required. Application of piping system noise predictions can also avoid noise limits being exceeded unexpectedly due to piping component noise sources such as bends and equal tees, which results in rectification costs and delays. POSTERS Session A9-3, 18:00 h Paper number 39 (Contributed, Poster) Posterboard BP-11 Wednesday, Author present 9:40 – 10:40 h Ruggero Bartolini and Giacomo Bisio Via all'Opera Pia 15 A - 16145 Genoa - ITALY THERMOACOUSTICS IN PIN-ARRAY STACK STRUCTURES AND OTHER STRUCTURES THERMODYNAMIC REMARKS I-INCE Class 21 In the thermoacoustic analysis, it has been shown that productive and dissipative processes occur mostly at different distances from the surface of the thermoacoustic solid element. Viscous dissipation occurs adjacent to the solid surface, while thermoacoustic occur mostly at about a thermal penetration depth away from the surface. Thus, it is reasonable to expect that changing the curvature of the surface can change the ratio of productive to dissipative effects. Consequently, the pin-array stack structure presents a lower ratio of dissipative zone to thermoacoustic zone, followed by the plate structure and then by the pore structure. The pin-array stack has, possibly, the ideal geometry for a thermoacoustic engine. However, it is currently the most difficult stack to fabricate. In this paper a Table with similarities and differences among several systems, including pin-array stacks, RVC stacks and pore stacks, is reported. The following distinguishing criteria have been assumed: (i) the presence or absence of constructive irreversibilities, (ii) the presence or absence of a valuable mass transfer and then of a valuable power, (iii) the range of the employed frequencies. The systems characterized by Internoise 2001 Abstracts 121 irreversibilities having a "constructive" role present low exergy efficiency, and for these the reference to reversibility is devoid of any meaning. In these cases, indeed, useful effect tends to zero as an infinitesimal of higher order than entropy production rate. On the contrary, when irreversibilities have not a "constructive" role, even if they are always present, the efficiency values can be higher, but only with an expedient choice of the various parameters. This Table seems suitable to remark basic differences among devices that are usually considered similar. Session A9-3, 18:00 h Paper number 60 (Contributed, Poster) Posterboard BP-12 Wednesday, Author present 9:40 – 10:40 h Gaku Minorikawa, Shoji Suzuki and Motohiro Moriya Dept. of Mech. Eng. Hosei University, Koganei-shi, Tokyo, Japan I-INCE Class 11 STUDY ON THE REDUCTION OF KARMAN VORTEX INDUCED SOUND BY CROSSING TWO CYLINDERS A cylindrical object such as heat exchanging pipe, sensor, stay and so on is located in a uniform flow, the discrete sound called the Karman vortex induced sound will be generated. Since the sound is caused by the periodic vortex shedding from the cylinder, to suppress the vortex shedding is important for the reduction of the sound. In this paper, the method of crossing two cylinders was carried out, especially on condition that the attached cylinder is much shorter than the first cylinder. It was also assumed that the geometrical arrangement of the second cylinder affects the amount of the sound reduction. In this paper, the tests were carried out in which the length of the second cylinder and crossing angle between two cylinders were varied. Also the effect of the sound reduction in case the second cylinder was attached upstream to the first one was compared to the downstream positioning. Further, in order to examine the effect of the sound reduction, the coherence function between the pressure fluctuations on the first cylinder surface and the sound was measured. As a result, it was cleared that the upstream positioning broke the vortex shedding and lead the better effect of the sound reduction. Session A9-3, 18:00 h Paper number 253 (Contributed, Poster) Posterboard BP-13 Wednesday, Author present 9:40 – 10:40 h Wan-Jo Jeon I-INCE Class 76 Digital Appliance Research Lab., LG Electronics, 327-23, Kasan-dong, Seoul, Korea A NUMERICAL STUDY ON THE ACOUSTIC CHARACTERISTICS OF CENTRIFUGAL IMPELLER WITH SPLITTER Centrifugal fans are widely used and the noise generated by these machines causes one of the most serious problems. In general, the centrifugal fan noise is often dominated by tones at BPF (blade passage frequency) and its higher harmonics. This is a consequence of the strong interaction between the flow discharged from the impeller and the cutoff in the casing. However, only a few researches have been carried out on predicting the noise because of the difficulty in obtaining detailed information about the flow field and casing effects on noise radiation. The objective of this study is to develop a prediction method for the unsteady flow field and the acoustic pressure field of a centrifugal fan, and to calculate the effects of small vanes that are attached in original impeller - Splitter impeller. We assume that the impeller rotates with a constant angular velocity and the flow field around the impeller is incompressible and inviscid. So, a discrete vortex method (DVM) is used to model the 122 Internoise 2001 Abstracts centrifugal fan and to calculate the flow field. The force of each element on the blade is calculated by the unsteady Bernoulli equation. Lowson's method is used to predict the acoustic source. The splitter impeller changes the acoustic characteristics as well as performance. Two-splitter type impeller and splitter impeller which splitter locates in jet region are good for acoustic characteristics. Internoise 2001 Abstracts 123 Wednesday August 29 Plenary Lecture 8:20 – 9:20 h Jiri Tichy Paper number 357 I-INCE Class 38 The Pennsylvania State University, P.O. Box 30, State College, PA 16801, USA APPLICATIONS OF ACTIVE NOISE CONTROL Although the principles and potential for active control of sound have been explored for more than 50 years, the early stages of active control technology development can be linked to the progress in signal processing and the availability of control electronics. Today, the control electronics needed for active control implementations is readily available. Further progress of active control technique depends primarily on sufficient insight into the properties of specific sound fields and elastic waves on structures to determine the feasibility and practical schemes to develop active control systems. I essence, active control is based on destructive sound interference which is induced by secondary sources. The selection of their location is crucial to achieve a sufficient degree of attenuation. In addition, the secondary sources have to be controlled by strategically located sensors to secure optimal performance. A successful development of active control systems requires optimization of all the mentioned components. After a brief review of the general principles, the paper will concentrate on the main areas of existing applications such as active control in ducts, enclosures, barriers, radiation into free space, sound transmission, use of smart materials etc. A brief analysis of underlying principles will precede the presentation of installation examples and achievements in various application areas. Better understanding of the feasibility and further potential of application development as well as encouragement to apply this technology will be emphasized. Session B1-1: Tyre/road noise modeling Chaired by Thomas Beckenbauer and Jean-François Hamet Session B1-1, 9:40 h Paper number 706 (Tutorial, Oral) Ard Kuijpers and Gijsjan Van Blokland M+P noise and vibration consultants, the Netherlands TYRE/ROAD NOISE MODELS IN THE LAST TWO DECADES: A CRITICAL EVALUATION I-INCE Class 76 During the last two decades various efforts have been made to model, with a certain degree of physical abstraction, the noise generated by the interaction between a rolling tyre and the road surface. Without the intention of being complete, we have tried to situate a selection of the most generally known models in the research field of tyre/road noise by making a critical evaluation. First an overview of various tyre/road noise mechanisms is given. Then, an assessment of the tyre/road noise models is made in terms of modelled noise generation mechanisms, frequency range, speed exponents, experimental validation etc. With this paper and presentation we try to facilitate the discussion about what issues are considered to the most important for future tyre/road noise models. 124 Internoise 2001 Abstracts Session B1-1, 10:20 h Paper number 654 (Invited, Oral) Jean-François Hamet, Philippe Klein INRETS-LTE, 25 avenue François Mitterrand, F69675 Bron Cedex, France I-INCE Class 76 USE OF A ROLLING MODEL FOR THE STUDY OF THE CORRELATION BETWEEN ROAD TEXTURE AND TIRE NOISE The influence of road texture profile on tyre noise is the object of experimental studies since the last twenty years. The recent development of rolling tyre models offers possibilities of getting some further insight in this matter. The road profile texture being measured in 2D (along a line), the tyre road interaction is also generally handled in 2D in the rolling models. The question which raises is how much this affects the resulting correlations between the road texture and the tyre noise evaluated by the model. It is indeed acknowledged that a tyre rolling on a transversely textured road makes a noise quite different than when rolling on a randomly textured road having the same longitudinal profile. The question will be addressed in this presentation. A way of considering a 3D tyre road interaction from 2D road texture profiles will be proposed and the resulting correlations between texture and noise will be compared. The model used is the analytic model initially proposed by Kropp where the tyre is described by an orthotropic plate. Session B1-1, 10:40 h Andrej Pietrzyk Paper number 655 (Invited, Oral) I-INCE Class 76 Goodyear Technical Center, 7750 Colmar-Berg, Luxembourg PREDICTION OF THE DYNAMIC REPONSE OF A TIRE The dynamic properties of tyres in extended frequency range are of general interest, both for the interior and exterior noise. At lower acoustic frequencies (below 500 Hz) the transmission of force from the tyre/road contact zone to the body of the vehicle is important for the interior noise problems. At higher frequencies, the structure borne noise originating from the vibration of tyres makes substantial contribution to the tyre noise perceived in the environment, which becomes most intense in the frequency range around 1000 Hz. Within the current research project RATIN, sponsored by the European Commission, Goodyear Technical Center, Luxembourg, participates in the studies of tyre vibration. The predictions are based on the simulations employing the Finite Element Method. In the proposed paper, the results of the simulations will be shown in relation to the measurement results. The important contributions to noise transmission to the car body will be identified. The feasibility of predictions at higher frequencies will be assessed. The FE results will be related to the predictions of existing semi-analytical tyre noise/vibration models. Session B1-1, 11:00 h Paper number 656 (Invited, Oral) I-INCE Class 76 Regina A.G. Graf ,Chih Y. Kuo, Ann P. Dowling and Will R. Graha Cambridge University Engineering Department, Trumpington Street, Cambridge CB2 1PZ, UK TYRE/ROAD INTERACTION NOISE Tyre noise is generated from vibrations caused by the impact and release of tread blocks entering and leaving the contact patch. These acoustics sources are then differentially amplified by the tyre/road geometry, resulting in far-field noise. The amplification is strongest in the ‘horn’ between the tyre belt and road surface, so that contributions from local vibrations in this region dominate the far field. This Internoise 2001 Abstracts 125 makes the problem intractable to fully numerical (FEM/BEM) approaches. Instead, we adopt a simplified strategy, whereby the tread vibrations are estimated from a flat plate model. The amplification of the sound they produce is then determined from an acoustic BEM computation. In this paper, we present the vibration model and initial far-field sound predictions; the horn effect calculation is described in Graf et al. The tyre structure around the contact patch is represented by an infinite, flat elastic plate, to which the tread blocks are attached. The forces and displacements at the block faces can be found by expressing the displacements in terms of the system impulse response and the block forces, and applying known values of force/displacement as boundary conditions. This approach has been implemented for strictly normal forces and displacements; however, its extension to the general case is straightforward. The vibration model is validated by comparison with experimental transfer functions between block pairs and with measured tread block accelerations on a rolling road. Far-field noise predictions and data are then used to assess the extent of the normal displacement contribution. Lastly, future developments of the model are discussed. Session B1-1, 11:20 h Paper number 657 (Invited, Oral) I-INCE Class 76 Wolfgang Kropp, Krister Larsson, Frédéric Wullens, Patrik Andersson, François-Xavier Bécot and Thomas Beckenbauer Department of Applied Acoustics, Chalmers University of Technology, S-41296, Göteborg, Sweden THE MODELLING OF TYRE /ROAD NOISE - A QUASI THREE-DIMENSIONAL MODEL The increased focus on road traffic noise has intensified the research activities concerning the investigation of tyre/road noise generation as one of the major noise sources. While during the seventies and eighties research effort considered mainly the understanding of phenomena behind the generation processes, today, prediction methods are the main subject of ongoing activities. These prediction methods are based on different philosophy and approach. They cover the wide range from FE approaches, modelling details as exact as possible, to statistical approaches quantifying the influence of main parameters influencing tyre/road noise. The paper focuses on a quasi-three dimensional model for the prediction of tyre/road noise. The model is based on the description of the main physical properties of tyres, the contact between tyre and road and the radiation from tyres. It includes the properties of the road surface. By means of this model possibilities and limitations of tyre/road noise modelling are discussed with respect to the required quality of the input data. The results from the model are compared with coast-by measurement for tyres on surfaces with different properties. Session B1-1, 11:40 h Paper number 714 (Invited, Oral) Udo Nackenhorst and Otto Von Estorff I-INCE Class 76 Institut für Baumechanik und Numerische Mechanik, Universität Hannover, Appelstraße 9A, D-30167 Hannover, Germany NUMERICAL ANALYSIS OF TIRE ROLLING NOISE RADIATION - A STATE OF THE ART REVIEW The sound radiation of rolling tires is a major source of traffic noise. By means of numerical simulations, it will be possible to get a deeper understanding of the physics of rolling noise phenomena and thereafter, for the optimization of the interacting counterparts, namely the tires and the road. The current presentation starts with a short state of the art review of numerical methods for the computation of the noise radiation of rolling tires. In addition, future needs of development are pointed 126 Internoise 2001 Abstracts out. In detail, a subsequent three step strategy will be presented and discussed: 1. The highly nonlinear behavior of stationary rolling wheels is computed by using the finite element method; 2. Based on the results of step one, the vibrations of the rolling wheel excited by the tread impact and the characteristics of the roads surface are calculated, again by means of the finite element method; 3. The sound radiating from the rolling wheel is investigated by either finite elements or boundary elements. A foundation of a numerically efficient treatment of this problem is a relative kinematics description of the rolling wheel, namely an Arbitrary Lagrangian Eulerian (ALE) description of rolling bodies, which allows a time independent description of steady state rolling in a fully nonlinear frame of reference. Furthermore, special emphasis has to be laid into the interpretation of the vibrations of rotating structures. The excitation mechanisms will be classified and solution methods will be suggested for the tread impact and the roads surface roughness. The air surrounding the wheel is modeled either by finite elements or by boundary elements. In the first case, special infinite elements are needed to take into account the wave radiation to infinity. In the case of the boundary element method, on the other hand, an infinite acoustical medium is already included in the formulation. In both approaches the vibrating surface of the tire, calculated with the mechanical model, is taken as a specific boundary condition for the acoustical model. In addition, the impedance of the road surface as well as other boundary conditions, such as reflecting parts of the car body, can be defined. The discussion of the state of the art and future work will be completed by representative examples. Session B1-1, 12:00 h Paper number 716 (Invited, Oral) Yong-Joe Kim and Stuart Bolton I-INCE Class 76 1077 Ray W. Herrick Laboratories, School of Mechanical Engineering, Purdue University, West Lafayette, IN 47901-1077, U.S.A. MODELING TIRE TREADBAND VIBRATION A static tyre was driven radially at a point on its treadband and measurements of the resulting radial treadband vibration were made around the treadband circumference by using a laser Doppler velocimeter. By performing a circumferential wave number decomposition of the measured spacefrequency data, the wave propagation characteristics were visualized. In an attempt to understand these experimental results in details, the tyre treadband was modelled as a ring-like, circular cylindrical shell with air pressure acting on its interior surface. The model makes allowance for general boundary conditions at the lateral edges of the ring, and the sidewall of the tyre is modeled as a distribution of springs and dampers. Both analytical and finite element methods were applied to obtain the vibration response. These solutions were compared with both the experimental result and the response of the infinite length plate driven by point excitation. This shell model was found to explain the propagation characteristics of the waveguide modes that are visible in the experimental results. In particular, the low frequency, flexural wave characteristics associated with finite curvature and which result in efficient radiation of sound were well reproduced. Fast in-plane waves that are potentially significant sound radiators at high frequencies were also well reproduced by this model; the excitation of these modes results from wave coupling within the curved shell. Internoise 2001 Abstracts 127 Session B1-1, 13:20 h Paper number 717 (Invited, Oral) Thomas Beckenbauer and Ard Kuijpers Müller-BBM GmbH, Germany I-INCE Class 76 PREDICITION OF PASS-BY LEVELS DEPENDING ON ROAD SURFACE PARAMETERS BY MEANS OF A HYBRID MODEL Based on research work for the German Highway Research Institute (the "Sperenberg project", carried out for the BASt) the combination of a statistical and a physical model (hybrid model) for the prediction of pass-by levels depending on road surface parameters will be presented. A physical model is used to separate physical input parameters such as the roughness of the road surface or the radial deformation of the tyre. In a second step physical descriptions are used to calculate a set of new input parameters like the contact pressure distribution within the tyre/road contact patch. The new parameters are appropriate to serve as independent variables for the statistical model. Based on carefully recorded and documented measuring values for the input parameters and the pass-by levels of 840 different tyre/road surface combinations the statistical model is used to quantify the relation between the input parameters and the pass-by level. The physical part of such a model needs to represent the main properties of the noise generation mechanisms in a qualitative way. The statistical part takes care for the quantification and reduces thus the efforts that are required for a pure physical model. Actual limitations and further developments of the model are discussed. Session B1-1, 13:40 h Paper number 61 (Contributed, Oral) I-INCE Class 76 Takeshi Toi, Naoki Ishii, Nobuyuki Okubo, Zempachiro Adachi and Toshiyuki Ikeda Chuo University, Dept. of Precision Mechanics, 1-13-27 Kasuga Bunkyo-ku, Tokyo 112-8551, Japan RADIATION NOISE FOR ROTATING TIRE UNDER CONSTANT SPEED USING•MODAL ANALYSIS AND BOUNDARY ELEMENT ANALYSIS Noise from rotating tire is studied by not only running on a road or drum test but also numerical simulation, however it is not clear to find the cause of noise generating mechanism and take prompt measures. In this research, a vibration deformation is predicted by sound radiation from a rotating tire under constant speed and the noise reduction is carried out by structural modification. First, the contribution of each vibration mode shape is estimated by using modal analysis and boundary element analysis and verified by numerical simulation. Next, the proposed method is applied to a rotating tire to predict the vibration deformation at some frequencies of the noise problem. The advantage of this method is as non-contact measurement and to obtain the contribution of each vibration mode shape. Finally, the highest contribution of the vibration mode shape is focused for structural modification to reduce the noise of the rotating tire. Session B1-1, 14:00 h Paper number 102 (Contributed, Oral) Foort De Roo, Eddy Gerretsen and Era H. Mulder TNO TPD, P.O.Box 155, 2600 AD Delft, The Netherlands Predictive performance of the tyre-road noise model TRIAS I-INCE Class 76 Recently the development of the TRIAS model (Tyre Road Interaction Acoustic Simulation) has been finalised and its predictive accuracy has been tested in a validation study. The model encompasses 128 Internoise 2001 Abstracts simulation modules for all major aspects of tyre-road noise generation, namely: force excitation by pavement surface texture, tyre vibration response, influence of pavement parameters, air-pumping volume flow generation, tread pattern resonances, sound radiation, horn effect and propagation. The paper will summarise the various modelling techniques that were used and will discuss the prediction accuracy achieved to-date. The stronger and weaker points of the methodology chosen and the possibilities for further improvements will be reviewed. Session B1-1, 14:20 h Paper number 459 (Contributed, Oral) I-INCE Class 76 Sandra Buss, Reinhard Weber, Werner Liederer and Volkert Mellert University of Oldenburg, Physics/ Acoustics, D-26111 Oldenburg, Germany SUBJECTIVE AND OBJECTIVE CHARACTERIZATION OF TYRE NOISE Subjective evaluations of tyre noise play an important role during improvement and development of noise reduce tyres. A combination of categorical judgements and the characterization of specific noise characteristics is applied for tyre assessments at various velocity ranges. More than 800 subjective tyre set assessments over a three years period undergo various analyses. Statistical test procedures are applied to reveal the internal structure of the assessments. In order to find an objective description of noise phenomena noise recordings are psychoacoustically preprocessed. Similar to the MP3 procedure irrelevant information is removed in a first step. Auditory important contours are processed. An attempt is made to relate noise phenomena to special characteristics in the obtained prominent contours. Session B1-2: Low noise tyre design Chaired by Foort de Roo and Federico Mancosu Session B1-2, 14:40 h Paper number 423 (Invited, Oral) Reiner Stenschke and Peter Vietzke Umweltbundesamt, Postfach 330022, 14191 Berlin, Germany TYRE/ROAD NOISE EMISSIONS, ROLLING RESISTANCE AND WET BRAKINGBEHAVIOUR OF MODERN TYRES FOR HEAVY-DUTY VEHICLES (STATE OF THE ART) I-INCE Class 13 Following the publication in 2000 of the latest results with regard to the noise and use characteristics of modern car tyres, corresponding data are now available for tyres for heavy duty vehicles (HDVs). 32 tyre types currently available on the market, with sizes commonly used for lorries and busses, produced by different manufacturers and including both steered-wheel tyres and drive-wheel tyres, were investigated to determine tyre/road noise emissions, rolling resistance, and wet braking behaviour. In addition to presenting the measuring results as such, the contribution also discusses the relation between tyre noise and wet braking behaviour and between rolling resistance and wet braking behaviour. This sheds light on the question of whether or not HDV tyres with a low rolling resistance and/or low noise emissions have poorer wet braking characteristics. The measuring results obtained form an important basis for the adaptation of EU tyre-noise control legislation to technical progress as well as for discussions about the establishment of stricter requirements for, e.g., award of the German "Blue Angel" ecolabel. Internoise 2001 Abstracts 129 Session B1-2, 15:00 h Paper number 167 (Invited, Oral) A. Fadavi, D. Duhamel and H.P. Yin I-INCE Class 76 LAMI Unit Mixte ENPC-LCPC, 6 et 8 avenue Blaise pascal, cité Descartes, Champs sur Marne, 77455 Marne la vallee, France TYRE/ROAD NOISE: FINITE ELEMENT MODELLING OF TYRE VIBRATIONS Within the problem of tire road noise this work uses a new finite element method to study the vibrations of the tire which is the most important noise source. Tire characteristics have a significant effect on tire road noise and can be included more easily in a FEM model. This paper presents a new detailed three dimensional finite element analysis model which takes into account the contact of the tire with the road and the tire characteristics such as the internal pressure and tire geometrical parameters. The tire and the wheel are modeled by the 6-node linear triangular and the 8-node linear brick finite elements. The tire is a viscoelastic rubber containing rebbars in its treads and its belts. The wheel is modeled by an annular steel plate of the same amount of mass. Natural frequencies and mode shapes of this tire-wheel unit rolling on the ground are calculated. For this purpose, first the large strain theory is used to obtain the steady state equilibrium. The linear vibrations of the tire are then calculated around this equilibrium state. Modal behavior can be analysed as a function of different parameters such as internal pressure, rotation velocity, weight of the vehicle, properties of the materials, distribution of rebbars and the type of the contact with the ground. The contact of the tire with a smooth ground could be extended to a rough contact. In addition, the frequency response functions of this tire are predicted, using their predicted natural frequencies and mode shapes. Session B1-2, 15:20 h Paper number 440 (Invited, Oral) Eric D. Schoen and Eddy Gerretsen TNO TPD, PO Box 155, NL-2600 AD Delft, The Netherlands I-INCE Class 76 SYSTEMATIC STUDY OF TYRE PARAMETERS TO REDUCE THE VIBRATION NOISE DUE TO TYRE-ROAD INTERACTIONS Over recent years, TNO TPD has developed a comprehensive, physical model for Tyre-Road Interaction Acoustics Simulation (TRIAS). In this paper we report on the results of a study, using TRIAS, to systematically reduce the noise production of tyres, resulting from vibrations induced in the tyre. The parameters varied in the study are: the circumferential and lateral tyre dimension, the width of the contact area, the bending stiffnesses (Bx, By, S) of the tyre carcass, the contact stiffness of the rubber, the mass of the tyre and the percentage groove area in the tread pattern (averaged). The mathematical procedure followed consists of three stages: an initial mapping of noise production within the 9D parameter space, a fixed-gradient optimisation and a final set of runs to find the optimum around the best result of the fixed-gradient phase. The initial mapping indicates that for minimisation of the noise produced due to tyre vibrations, the tyre should be decreased in diameter, in width, in circumferential bending stiffness and in percentage groove area, while it should be increased in mass, lateral dimension and lateral bending stiffness. In our initial parameter choices we have also indicated ranges within which the parameters can be varied. With the use of the optimal gradient for 9parameter unconstrained optimisation, the groove percentage would drop below its lower limit. Therefore we have used the optimal gradient for the remaining eight parameters to converge to the optimum. 130 Internoise 2001 Abstracts Session B1-2, 15:40 h G.J. Kim Paper number 594 (Invited, Oral) I-INCE Class 76 KUMHO TIRE R&D Center, 506-040Kwangju, Korea FLEXURAL WAVE PROPAGATION AND SOUND RADIATION OF THE TYRE SHELL Tyre noise is generated by several mechanisms. With a modern tyre, wall vibration, air pumping and air resonant radiation are all considered to be important. But tyre noise generating mechanisms are still not clear due to the complication of tyre vibration behavior. Vibrations of the tyre shell are the combination of several different wave types which appear at different frequencies. In a low frequency range, where the tyre behaves like a elastically supported beam, the circular ring model is used to analyze the dispersion relations. Above 300Hz, which is the transition point from one-dimensional to two dimensional waveguide properties of the passenger car tyre, a cylindrical shell model is used to analyze flexural waves propagation. Two important features on the wave propagation, wave-guide behavior and the curvature effect of the tyre wall are analyzed. In consideration of noise radiation from tyre waves, most of the tyre waves observed in this study are inefficient sound radiators since their wavenumbers are larger than the acoustic wave number. As a result, It is observed that one of the most important features in sound radiation of a tyre shell is acoustically excited wave motion of the tyre wall. Session B1-3: Low noise road design and application Chaired by Michel Berengier and Gijs-Jan van Blokland Session B1-3, 16:00 h Yves Meunier Paper number 302 (Invited, Oral) I-INCE Class 35 USIRF/ROUTIÈRE MORIN : Continental Square, 3 Place de Londres - BP 10764 - 95727 ROISSY CHARLES DE GAULLE CEDEX FRANCE LOW-NOISE PAVEMENTS: THE SOLUTIONS OFFERED BY THE FRENCH ROADBUILDING INDUSTRY In France, tyre-road noise measurements carried out on new or maintenance road construction sites constitutes a national database allowing a pavement raking in three main families (low noise pavements, intermediary and noisy pavements). From this classification, we conclude that two main factors have an important effect on a low noise pavement formulation. These are: the granularity and the void content which is connected to the acoustic absorption. For about twenty years, drainage asphalts have been the first low noise pavements developed in France. More recently, we took a new orientation and we began to study twin-lay porous asphalt concretes. In addition, since five years, French road companies have developed a new concept of low noise very and ultra thin layer bituminous concretes. These new pavement structures have interesting characteristics which fulfil the previous two requirements (granularity and the void content) together with satisfactory mechanical and skid resistance properties. Concerning their acoustical properties, we observe an average improvement around 3 to 4 dB(A) in terms of LAeq (6.00:22.00) with respect to the French reference 0/10 bituminous concrete pavement. Nevertheless, considering their present knowledge of the Internoise 2001 Abstracts 131 phenomena involved and of the measuring methods they practice, French companies wish that the acoustic performances level of those products remains a qualitative objective and do not become contractual in deals context. Session B1-3, 16:20 h Paper number 337 (Invited, Oral) S.M. Phillips, S. Kollamthodi and P.A. Morgan CLASSIFICATION OF LOW NOISE ROAD SURFACINGS I-INCE Class 35 Transport Research Laboratory, Crowthorne, Berkshire, RG45 6AU. United Kingdom An increasingly important aspect of the development of low noise road surfaces is the definition and certification of its acoustic properties. In the UK, a number of complementary methods are being developed for such assessments. One of the most important of these is the noise test attached to the HAPAS (Highway Authorities Products Approval Scheme) type approval system. This test method is based upon the ISO standard Statistical Pass-by (SPB) method and results in a classification rating related to a standard surfacing. To supplement this test and to allow the acoustic properties of road surfaces to be determined at any location a system based upon the Close-proximity (CPX) measurement procedure. This is carried out using the TRITON system developed by TRL. This system is used for measuring the rolling noise of specified test tyres at suitable speeds. Combining the results of several tyres allows a representative noise classification measure to be determined. In addition surrogates for direct noise measurements are under development with high-speed laser-based texture measurement systems. These systems are being used to routinely survey the high-speed road network and could underpin future noise maps. This paper describes the development of these tests and the relationships between the results obtained with each. Session B1-3, 16:40 h Paper number 457 (Invited, Oral) Hans Bendtsen, Lars Ellebjerg Larsen and Jørgen Kragh I-INCE Class 35 Danish Transport Research Institute, Knuth-Winterfeldts Allé, Bygning 116 vest, DK-2800 Lyngby, Denmark TWO-LAYER DRAINAGE ASPHALT - NOISE REDUCTION AND CLOGGING An urban test road with 4 pavements was constructed in Copenhagen in 1999. One section with dense asphalt concrete as a reference. Three sections with two-layer drainage asphalt. Maximum 5 or 8 mm chippings in the top layer. The total layer thickness was 55 mm, 70 mm and 90 mm. The aim is to study the performance of the pavements over their lifetime. To avoid clogging the pavements are cleaned twice a year. A comprehensive measurement program has been designed. Yearly measurements of noise, pavement characteristics, permeability and traffic safety are carried out. The results from the noise measurements on individual vehicles in ordinary traffic are presented. When new, the noise reduction at the thickest two-layer drainage asphalt, compared to the reference section, was 6 dB. At the two other test sections the corresponding noise reduction was 5 dB. Results nine months later indicate some changes. The reduction at the thickest two-layer is reduced from 6 to 4 dB. The reduction of the thinnest fine graded pavement is reduced by 1 dB. The two-layer with 8 mm chippings still has the same reduction of 5 dB. The measurements of permeability show some tendencies of clogging. The pavements will be intensively cleaned in December 2000. Results of noise measurements before and after this cleaning will be presented and correlated with measurements of permeability. 132 Internoise 2001 Abstracts Session B1-3, 17:00 h Paper number 695 (Invited, Oral) Gijsjan Van Blokland and Ard Kuijpers M+P noise and vibration consultants, the Netherlands TYPE APPROVAL AND COP TESTS FOR LOW NOISE SURFACES I-INCE Class 35 The developments of low noise surfaces is going rapidly and several new types are being produced nowadays. Application of these surfaces in every day road building practice requires the availability of generally accepted test methods to establish the acoustic properties and the be able to control the performance after being produced and during its life time. In this paper an assessment method for the acoustic quality of road surfaces, which is closely connected to the noise immission calculation scheme, will be presented. We will discuss the statistical and methodological aspects and we will address the issue of conformity of production testing. In this presentation data from several very low noise surfaces will be used as illustration of the method. Session B1-3, 17:20 h Paper number 707 (Invited, Oral) I-INCE Class 35 Massimo Losa, Gaetano Licitra, Michel Berengier and Mauro Cerchiai University of Pisa - Civil Engineering Department, Via Diotisalvi, 2 – 56126 Pisa – ITALY PHYSICAL CHARACTERISTICS OF ROAD PAVEMENTS AND NOISE EMISSION In the past decade a great effort has been made in order to reduce the tyre-road noise by working on surface characteristics of pavements. In many cases, this goal is conflicting with other pavement requirements regarding road safety, like skid resistance, and durability. In order to evaluate reliably the influence of road pavement characteristics (texture, porosity, acoustic absorption, flow resistance for water and air) on their functional performances (noise emissions, skid resistance, drainage, durability) experimental controls have been performed on road sections of porous asphalt pavements. In this paper the analysis of the results of both laboratory tests, performed on specimens taken by pavements, and in situ tests, performed on the same road sections, is presented. Primarily, tests have been carried out to control that pavements are safe and stable; then, tests have been performed in order to measure surface characteristics and noise emissions. Relationships have been found between: - flow resistance for water, for air and pavement drainage; - acoustic absorption coefficient in the impedance tube and pavement surfacing absorption factor carried out according to ISO 13472-1; - pavement surfacing absorption factor and sound equivalent level spatial variability; - pavement texture profile, for pavements characterised by the same acoustic absorption coefficient, and noise level measurements: pass by method (ISO/CD 11819-1), on board method with self powered vehicle (ISO/CD 11819-2), inside various vehicles at different speeds. These relationships allow defining guidelines useful in order to design silent pavements that are economic, safe and stable. Session B1-3, 17:40 h Paper number 120 (Contributed, Oral) G.R. Watts, R. Stait, N.S. Godfrey and R.E. Layfield OPTIMISATION OF TRAFFIC CALMING SURFACES I-INCE Class 52 Transport Research Laboratory, Crowthorne, Berkshire, RG45 6AU. United Kingdom Early, mainly empirical studies, on Traffic Calming Surfaces (TCSs) at TRL led to the development of rumble areas and strips that were subsequently employed by local authorities to alert drivers to hazards ahead. The main application foreseen at this early stage was on the high-speed approaches to road Internoise 2001 Abstracts 133 hazards in mainly rural areas. For this reason resident’s reactions were not given high priority in these early trials although it became clear that there was considerable anecdotal evidence that noise disturbance occurred at some sites and rumble devices were not generally recommended for roads in urban areas because of this issue. There is a need therefore to optimise TCSs so that the driver is alerted without consequential disturbance to residents. This paper reports on an ongoing study where a more fundamental approach was taken to designing appropriate TCSs. This involved a consideration of vehicle resonances and the means of exciting these with a suitable road surface profile without the generation of significant external noise. It was found that for some surfaces significant horizontal vibrations were generated in vehicle suspensions which were readily transmitted into the driver’s cab. The paper describes the results of test track measurements of noise and vibration that has led to the identification of potentially suitable designs. Session B2-1: Active control of structural vibrations (continued) Chaired by Osman Tokhi and Jing Tian Session B2-1, 9:40 h Paper number 44 (Contributed, Oral) C.S. Chou and W.D. Wu I-INCE Class 47 National Taiwan University, Institute of Applied Mechanics, Taipei 106, Republic of China VIBRATION CONTROL OF A BEAM BY CO-LOCATED AND NON-COLOCATED PIEZOELECTRIC SENSOR AND ACTUATOR Direct feedback control and non-collocated control of a vibrating beam is studied in this paper. Based on the idea of inverting and amplifying the motion of a beam at a point to generate a control motion to annihilation the vibration of the beam. A control system composed two similar electromechanical transducers, one as the sensor to detect the vibration at a point and the other as the actuator to apply the control force (motion). When the sensor and the actuator of the control system are placed collocated onto the structure, they are connected to an inverting amplifier to form a closed-loop direct feedback system. For the case of the sensor and the actuator are placed non-collocated, an electronic device developed from the frequency response function of the beam is used to process the sensed signal and generate the control signal to the actuator. Experimental results showed that the amplitude of vibration of the beam under control was reduced significantly. 134 Internoise 2001 Abstracts POSTERS Session B2-1, 10:00 h Paper number 62 (Contributed, Poster) Posterboard BP-14 Wednesday, Author present 13:20 – 14:20 h Jian-Da Wu, Fu-cheng Su and Hai-Ping Lin I-INCE Class 46 Department of Vehicle Engineering & Technology, Da-Yeh University, 112 Shan-Jeau Road, DaTsuen, Chang-Hwa, Taiwan A STUDY OF FEEDFORWARD AND FEEDBACK ACTIVE VIBRATION CONTROL SYSTEM This paper describes the single degree active vibration control systems. Two active techniques are compared in terms of control structures and control algorithms. The adaptive structure using filtered-x least-mean-square algorithm (FXLMS) with synthetic reference and the linear quadratic gaussian (LQG) control are employed for controller synthesis. The control algorithms are implemented by using a digital signal processor (DSP). Experiments are carried out to evaluate the proposed system for attenuation of the vibration from shaker. The experimental results indicate that both algorithms achieved the vibration control in the test model. The comparison of both structures is also described in this paper. Session B2-1, 10:00 h Paper number 177 (Contributed, Poster) Posterboard BP-15 Wednesday, Author present 13:20 – 14:20 h C. Roger Halkyard and Brian R. Mace I-INCE Class 42 Department of Mechanical Engineering, University of Auckland, Private Bag, Auckland, New Zealand ADAPTIVE CONTROL OF FLEXURAL BEAM VIBRATION IN THE PRESENCE OF A NEARFIELD The use of wave-based quantities in active vibration control has several potential advantages over conventional approaches based on isolated measurements. These include increased flexibility in the choice of control strategy and reduced sensitivity to unmodelled reflections. However, such approaches require real-time estimates of the relevant wave amplitudes. This paper describes a method by which the amplitude of a propagating wave may be estimated from the outputs of three error sensors, applicable to feedforward adaptive control of flexural vibration in beams. It is assumed that the control system must be compact, and thus that the error sensors lie within the nearfield generated by the control force. This is in contrast to most investigations, in which far field conditions are assumed, leading to control systems that are too large for most practical applications. The wave amplitude is estimated by digitally filtering and combining the outputs of the error sensors. Design of the necessary filters is discussed, along with issues of causality and required filter length. The resulting wave amplitude estimate is used as the cost function in simulations of active vibration control, using a filtered-X LMS algorithm. System parameters are varied, and their effect on the overall control achieved is discussed. The performance of the system is compared with a conventional control system, and it is shown that the wave-based approach can offer significantly better broad-band attenuation. Internoise 2001 Abstracts 135 Session B2-1, 10:00 h Paper number 206 (Contributed, Poster) I-INCE Class 46 Posterboard BP-16 Wednesday, Author present 13:20 – 14:20 h Ruud Vermeulen, René Lemmen, Arthur Berkhoff and Jan Verheij TNO TPD, P.O. Box 155, 2600 AD Delft, The Netherlands ACTIVE CANCELLATION OF UNWANTED EXCITATION WHEN MEASURING DYNAMIC STIFFNESS OF RESILIENT ELEMENTS Laboratory measurement of the vibro-acoustic transfer properties of resilient elements will be dealt with in a series of five International Standards (ISO-10846-1 through 10486-5, partly still forthcoming). The measured quantity that is standardised, is dynamic stiffness. Because of the multidirectional nature of the vibration transmission, the standards cover separate measurements of stiffness for three perpendicular excitation directions. For each of the three orthogonal translational excitations, the input for the unwanted directions have to be at least 15 dB lower. To meet this requirement over a large frequency band may be quite difficult, especially in case of 'transverse' excitation. Then special measures have to be taken to suppress rotation inputs. In this paper active vibration control on the input side of a test element is shown to be useful to suppress unwanted input rotations. The applicable frequency range depends on the controller available. The results presented are measured with a rather outdated controller, but show the feasibility of the method and good suppression result up to 170 Hz. The driving signal was a swept sine from 20-500 Hz in 60 seconds. New tests with a state of the art controller have to be performed to cover a wider frequency range. Session B2-1, 10:20 h Paper number 368 (Contributed, Poster) Posterboard BP-17 Wednesday, Author present 13:20 – 14:20 h J.W. Heo, J. Chung, J.M. Park, Y.W. Lee and Y.S. Seo I-INCE Class 46 Department of Mechanical Engineering, Hanyang University, 1271 Sa-1-dong, Ansan, Kyunggi-do 425-791, Korea VIBRATION AND NOISE REDUCTION OF AN OPTICAL DISK DRIVE BY USING A DYNAMIC VIBRAION ABSORBER Vibration and noise reduction of an optical disk drive by using a dynamic vibration absorber is presented. In optical disk drives such as CD-ROM or DVD drives, vibration becomes one of serious problems to be resolved as the rotating speed and the storage capacity increase. In fact, it is practically impossible to overcome the off-track and off-focusing errors by the servo control without reducing vibration. Especially, when the optical disk drive rotates a wavy disk, the disk drive suffers from severe vibration. For vibration reduction, a dynamic vibration absorber is attached to the feeding deck of the disk derive. Before design of the vibration absorber, the frequency response for the disk vibration is investigated by a laser vibrometer, so that it is found that the resonance occurs at 58 Hz. A similar result is obtained when the vibration of the feeding deck is measured by using an accelerometer. To reduce vibration around the resonance frequency, a dynamic vibration absorber is designed in this study. The vibration absorber consists of a steel ring and a rubber bobbin, which play roles of inertia mass and spring, respectively. By using a FEM code of ANSYS, the required fundamental natural frequency of the vibration absorber is determined. After a manufactured absorber is attached to the feeding deck, vibration and noise test are performed. The test result shows that considerable vibration and noise are reduced. 136 Internoise 2001 Abstracts Session B2-1, 10:20 h Paper number 576 (Contributed, Poster) Posterboard BP-18 Wednesday, Author present 13:20 – 14:20 h Lucyna Leniowska and Ryszard Leniowski FEEDBACK ACTIVE CONTROL OF THE CIRCULAR PLATE VIBRATION I-INCE Class 42 Institute of Technology, Pedagogical University of Rzeszów, Rejtana 16, 35-350 Rzeszów, Poland The paper presents a simulation of active vibration control for the circular plate clamped at the edge as well as its experimental implementation. The axially - symmetric vibrations of the plate are measured by the application of several pairs of strain sensors located along the plate radius. The standard linear system theory is applied to the state equation of considered system and an optimal reduction of plate vibrations are obtained for centrally located point control force using LQR controller with position and vibration velocity errors feedback signals. With the use of the formulated algorithm, implemented on a multi-input board, the ‘secondary’ control force was determined. The paper presents the simulation results as well as results obtained for ‘close-loop’ system on an experimental plant. Session B2-1, 10:20 h Paper number 577 (Contributed, Poster) Posterboard BP-19 Wednesday, Author present 13:20 – 14:20 h Ryszard Leniowski and Lucyna Leniowska I-INCE Class 42 Department of Automatic Control and Computer Science, Technical University of Rzeszów, W.Pola 2, 35-902 Rzeszów, Poland DISCRETE-TIME MODEL IDENTIFICATION FOR ACTIVE VIBRATION CONTROL Flexible systems are typically represented by non-linear differential equations. These equations are high order involving, numerous unknown or poorly known parameters. For these reasons, the dynamic models must be identified. There are many techniques used in the identification process. Paper presents the methods of discrete-time model identification for purpose of real-time active vibration control of planar circular plate. It contains parameter estimation of an Output Error Model for SIMO system and introduces a neural net structure, which appears to be suitable for non-linear identification and direct control applications. In particular, this paper presents detailed information about identification experiment, selection of model structure, estimation and model validation for the clamped circular plate. Obtained numerical values and structure of dynamical models have been used to develop the Model Reference Adaptive Controller for vibration cancellation. Internoise 2001 Abstracts 137 Session B2-2: Active control of sound radiation and transmission Chaired by Jiri Tichy and J. Scheuren Session B2-2, 10:40 h Paper number 127 (Tutorial, Oral) S.J. Elliot and P. Gardonio I-INCE Class 38 Institute of Sound and Vibration Research, University of Southampton, Highfield, Southampton, SO17 1BJ, UK ACTIVE CONTROL OF SOUND TRANSMISSION In a number of applications, particularly in the aerospace field, it is desirable to have panels which are light and strong, but which also have a high sound transmission loss over a broad range of frequencies. For a given weight and strength the transmission loss of a conventional panel is subject to fundamental physical limits and active solutions have been sought. A long-term goal in active control has been to integrate sensors, actuators and controllers into the structure of the panel to improve its transmission loss and create a "smart panel". An increasing understanding of the physical mechanisms of active control with realistic actuators and sensors, and the control strategies which are most appropriate, together with the falling cost of signal processing devices and MEMS transducers suggests that such a goal is coming closer. This paper will review the actuator, sensor and control approaches which are currently being investigated in this application. Session B2-2, 11:20 h Paper number 237 (Invited, Oral) Sabih I. Hayek and Jungyun Won ACTIVE CONTROL OF RADIATED ACOUSTIC POWER FROM ELASTIC PLATES I-INCE Class 38 Department of Engineering Science and Mechanics, University Park, PA 16802, U.S.A. An infinite line force excites an infinite Timoshenko-Mindlin plate in contact with an infinite acoustic medium. Part of the input mechanical power that is injected into the plate contributes to the structural intensity in the plate and the remainder is injected as acoustic intensity in the acoustic medium. In this paper, the radiated power from the structure into the acoustic medium as well as the total output structural intensity in the plate is to be controlled by secondary line force(s) on the plate. The object function to be minimized is either the total radiated acoustic power, or a weighted sum of that and the total structural intensity output in the plate. These object functions can be minimized by varying the position of the controller(s). These positions were judiciously picked to be in increments of one-half or full fluid-loaded structural wavelengths. Results were obtained for frequencies below and above the coincidence frequency. 138 Internoise 2001 Abstracts Session B2-2, 11:40 h Paper number 325 (Invited, Oral) Kris Henrioulle and Paul Sas I-INCE Class 38 Katholieke Universiteit Leuven, Department of Mechanical Engineering - PMA, Celestijnenlaan 300B, B-3001 Leuven, Belgium A PVDF SENSOR/ACTUATOR PAIR FOR ACTIVE CONTROL OF SOUND TRANSMISSION In this work, a simple and effective control system is developed to reduce the sound transmission through a single plate. The control system is characterised by three elements. First, the control system aims at reducing the sound power output at low frequencies. This is done by measuring and exciting the first radiation mode of the plate. Second, a PVDF polymer sensor and actuator is used. The shape of the electrode is such that only the first radiation mode is sensed and excited. Third, a collocated sensor actuator configuration is chosen because it allows the implementation of direct velocity feedback. This is a simple and effective control algorithm that increases the damping of the plate without spill-over to higher frequencies. Previous studies have analysed the behaviour of this configuration and predicted the performance. These simulations have shown some problems related to the coupling between sensor and actuator foil. This paper deals with the experimental validation of the simulations and presents the control results. The experiments show an increase in the sound transmission loss through the plate in the low frequency region below 300 Hz. Session B2-2, 12:00 h Paper number 449 (Invited, Oral) Ricardo A. Burdisso and Leonardo R. Molisani I-INCE Class 38 Vibration and Acoustics Labs., 153 Durham Hall, Mechanical Engineering Dept., Virginia Tech, Blacksburg, Virginia 24061-0238, U.S.A. CONTROL OF STRUCTURALLY RADIATED NOISE USING PASSIVE/ACTIVE WEAK RADIATING CELLS A new approach has recently been developed to attenuate low frequency structurally radiated noise. The treatment consists of incorporating into the surface of the structure an array of weak radiating cells (WRC), which converts the structural response to that of an equivalent array of dipole sources. The cells are mechanical devices designed to effectively create a dipole acoustic source over a relatively wide frequency band. The concept was validated experimentally on a plate driven with white noise over the 0-1600 Hz frequency range. The results showed a reduction in sound power levels of 10.3 dB over the frequency range of 100-1600Hz. Furthermore, attenuations of up to 25 dB were obtained at the frequency where the cells behave as a perfect dipole. Here, a model of a simply supported plate including an array of cells is developed to gain insight into the performance of the sound attenuation mechanism of the system. The model is also used for design optimization of the weak radiating cells of the plate experiments that predicts 15 dB sound power attenuation in the 01600 Hz band. To improve the performance at low frequency, the model was also extended by incorporating independent SISO feedforward control systems to each cell. The error signal to be minimized by the control system is the volume velocity of the cell monitored using two accelerometers while the control input is a force-pair acting between the plate and the cell. The results show that the hybrid cells further improve the performance of the system by expanding the frequency range where a dipole source is formed by the cells. Internoise 2001 Abstracts 139 Session B2-2, 13:20 h Paper number 480 (Invited, Oral) M.R.F. Kidner and M.E. Johnson Virginia Polytechnic Institute and State University, Blacksburg VA 24060 USA 2D SHAPED SENSORS FOR FEEDBACK CONTROL I-INCE Class 38 The performance of feedback control systems is inherently limited by instability. Careful design of the feedback loop can place unstable regions far outside the control bandwidth and increase the gain margin. This is often achieved by using temporal pass band filters that greatly reduce the amplitude outside the control bandwidth. However phase roll off within the pass band of a temporal filter is unavoidable. Spatial filters have been shown to have no phase roll off within their pass band and hence can be used to enhance the control bandwidth of feedback systems. As an extension to work done on shaped sensors as spatial filters in one dimension, spatial filtering in two dimensions is considered. It is shown that by modifying the desired spatial sensitivity of the sensor by that of a uniformly sensitive disc any desired window sensitivity can be presented to flexural waves from all directions on a two dimensional structure. A force actuator is developed which has low pass filtering properties by means of applying the force to the structure via 'footprint', which can also take the form of any desired spatial window. The combination of these two modified transducers allows the creation of sensor actuator pairs that are collocated and have natural low pass properties. It is also shown that band pass sensors can be created. This considerably enhances the stability of the feedback system in which they are implemented. Theoretical and experimental results are presented as well as design criteria. Improvements in closed loop stability are also demonstrated. Session B2-2, 13:40 h Paper number 606 (Invited, Oral) I-INCE Class 38 Jacqueline M. Munn, Colin D. Kestell, Ben S. Cazzolato and Colin H. Hansen Department of Mechanical Engineering University of Adelaide SA 5005, Australia REAL TIME FEEDFORWARD ACTIVE NOISE CONTROL USING VIRTUAL SENSORS Active noise control (ANC) is a potential solution for low frequency tonal noise in vehicular cabin interiors. While ANC can significantly reduce low frequency noise at a microphone error sensor, the zone of attenuation may be so small that a nearby observer might not experience any noise improvement at all. Using an energy density error sensor can increase the size of the attenuation zone around the sensor, so that the observer becomes partly enveloped in the region of reduced noise. However, it is also possible to reduce the noise specifically at the observer location, by using either a forward prediction “virtual microphone” or “virtual energy density sensor”. Pressure or energy density can be estimated (and hence minimised) at the observer location by extrapolating the signal from a number of remote microphones. Publications by the authors to date have presented the results from either control simulation models or ‘off-line’ control examples where the potential for control is evaluated mathematically using experimental data. This paper presents the results of real time control using an adaptive controller and various ‘virtual’ error sensors in both a free field and a long narrow duct. 140 Internoise 2001 Abstracts Session B2-2, 14:00 h Paper number 700 (Invited, Oral) Tian Ran Lin and Jie Pan I-INCE Class 38 Department of Mechanical and Materials Engineering, The University of Western Australia, Nedlands, WA 6907 Australia SOUND RADIATION FROM A BOX-TYPE STRUCTURE The finite element and boundary element methods have been used to analyze the sound radiation properties of a box-type structure. The symmetrical properties of the mode shapes of the box-type structure directly give information about which group of the mode shapes are effective sound radiators. This paper presents results of typical radiation patterns of 6 groups of modes. The radiation efficiency of such structure will also be discussed in the paper. Session B2-2, 14:20 h Paper number 109 (Contributed, Oral) S.E. Wright and H. Atmoko School of Engineering University of Huddersfield HDI 3DH UK PERFORMANCE OF ELECTRONICALLY CONTROLLED ACOUSTIC SHADOWS I-INCE Class 38 The properties of a basic free-field cancelling system have been considered in Inter-Noise 2000. To produce sound attenuation over a wide angle from large complex sources, multi-units of these basic systems (multi-channels) are required. The proposed paper considers the control conditions necessary for these multi-channel systems to operate efficiently. The adaptive performance and the resulting acoustic shadow characteristics from these systems are then considered. To operate free-field multichannel systems efficiently, two conditions need to be met. Condition I is that for a multi-channel freefield system to converge effectively, it is necessary to operate the system in the center of it's stability regions and align all the stability regions of the system. Methods of automatically determining, centering and aligning these stability bands are considered. Condition II is that for large channel number system, the control robustness has to be optimised based primarily on the system geometry. The paper considers how to determine the multi-channel system robustness and how to avoid instability peaks for the system to remain stable. The adaptive modal convergence characteristics and the overall adaptive speed of these multi-channel systems are then considered. These are established in terms of frequency, propagation distance, size and complexity of the primary source, the geometry of the system and the number of channels. Finally, the acoustic shadow performance in terms of cancellation depth, shadow angle, shadow profile and number of channels is considered. The shadows are then compared with theoretical predictions. Session B2-2, 14:40 h Arthur P. Berkhoff Paper number 425 (Contributed, Oral) I-INCE Class 38 TNO TPD, P.O. Box 155, 2600 AD Delft, The Netherlands ACTIVE STRUCTURAL ACOUSTIC CONTROL USING BROADBAND RADIATION MODES Weighting schemes for actuator arrays and sensor arrays are presented that can be used for the active reduction of broadband sound radiated from plates. These weighting schemes are obtained by maximizing the coupling of the arrays to the acoustic field in a certain frequency band. Unlike previous radiation mode weighting schemes for sensors, which lead to the optimum weighting coefficients at a single frequency, the present formulation optimizes the coupling in a broadband Internoise 2001 Abstracts 141 sense. The weighting scheme for the actuator array is obtained by maximizing the coupling in some frequency range, while having a minimum coupling in another frequency range. The technique can be used to obtain good coupling to the in-band modes with a minimum coupling to the out-of-band modes, resulting in a minimum amount of spillover. It is shown that the technique can be used with good results in feedforward and feedback systems for the active reduction of broadband sound transmitted through plates using piezoelectric patch actuators and structural sensors. Session B2-2, 15:00 h Paper number 661 (Contributed, Oral) Sadao Akishita, Yasunori Inoue and Yuuichiroh Mitani Ritsumeikan Univ., 525-8577Kusatsu, Japan I-INCE Class 38 SOUND TRANSMISSION CONTROL WITH INTELLIGENT COMPOUND PLATE BY USING ROBUST FEEDBACK CONTROL This paper describes active feed-back control systems constructed with compound metal plate containing piezoelectric material, PZT, which functions as sensors and actuators and is called ‘Intelligent Compound Plate’. The bi-morph type PZT plate is sandwiched by two sheets of stainless steel plate for reinforcing the strength of the structure and the passive insulation of sound. The dimension of the plate is rather small, 40mm width, 150mm length and 1.2mm thickness, for the plate is supposed as a proto-type model of the future active sound insulation structure. Active feed-back control system is introduced for supplementing the sound insulation effect, which is deficient in the passive structure over low frequency range. Double robust feed-back control is applied for the system, where the inner model control (IMC) system is constructed in the basis of the system, and also the Hinfinity feed-back control algorithm is constructed on the identified control system involving IMC. The double robust control system certifies not only the stability of the system which tends to be deficit in the case of closely positioning of sensors and actuators, but also the high sound insulation performance. The goal of the control is to decrease the transmitted sound pressure level by more than 10dB within the frequency range lower than 1000Hz. The control experiment proves the effectiveness of the control systems. Session B2-3: Active control in aircraft engines Chaired by Rudolf Maier Session B2-3, 15:20 h Paper number 308 (Invited, Oral) I-INCE Class 38 J. Zillmann, R. Maier, W. Just, I. Borchers, E. Bouty, H. Antoine, L. Enghard and U. Tapken EADS, Corporate Research Center Germany, 81663 Munich, Germany ACTIVE CONTROL OF FAN NOISE BY ACTIVE STATORS This paper is devoted to the active cancellation of tonal fan noise by means of novel actuators. The actuators are flush mounted on the stator vanes to ensure, that the secondary source is nearest to the primary, rotor-stator interaction noise source. The fan rig consists of a ducted rotor and stator with 48 rotor blades and 10 stator vanes. Each vane was equipped with four piezoceramic actuators developed by Dornier resulting in 40 actuators. 32 induct microphones were flush mounted to the duct wall and serves as error microphones. A narrowband filtered-x-LMS algorithm has been used to minimize the 142 Internoise 2001 Abstracts sum of squared sound pressure levels measured by the error sensors. Three cluster configurations have been applied to reduce the control complexity. ANC was successfully applied to the 1st blade passage frequency of the fan. Additional 64 microphones have been used by DLR to calculate a modal decomposition of the induct sound field. The complexity of the control process is analysed by means of an eigenvector decomposition of a transfer function between the actuators and error sensors. The work described in this paper has been carried out in the framework of the BRITE-EURAM project "Reduction of Engine Source Noise through Understanding and Novel Design" (RESOUND). It addresses the challenge of reducing noise at source, in particular turbomachinery noise, through engine component aeroacoustic design and through devices that can be integrated within the engine structure. Session B2-3, 15:40 h Paper number 320 (Invited, Oral) R. Maier and J. Zillmann EADS Corporate Research Center Germany, D-81663 Munich/Germany I-INCE Class 38 ACTIVE CONTROL OF NOISE FROM TURBOMACHINES - CONTROL ALGORITHMS AND EXPERIMENTAL RESULTS The noise radiated by stationary gas turbines or aeroengines is usually reduced by acoustic absorbers, whose efficiency is limited due to the confined length of the inlet and outlet duct sections. In particular dominant tonal components containing a large amount of the generated sound power are not attenuated sufficiently. Higher sound level reductions can be obtained by additional active control of the tones. In the framework of AG Turbo Turbotech II - supported by the German Ministry for Research and Technology (BMBF) - DLR, EADS and MTU Aero Engines studied the noise reduction potential of different active noise control (ANC) configurations and control algorithms for the application to turbomachines at a fan model of 1 m diameter. Theoretical investigations with regard to the realisation and design of the ANC test configuration were performed by MTU (see paper by Traub et al.). DLR provided the test stand and executed the radial mode analysis of the sound field (see paper by Tapken et al.). This paper deals mainly with the control algorithms used for the active noise control experiments. Two types of control algorithms have been used. A narrowband filtered-x-LMS algorithm has been applied to minimize the sum of squared sound pressure levels measured by the error sensors. Furthermore a modal control technique has been tested. The modal controller minimizes the sum of squared modal amplitudes for certain circumferential modes. ANC was successfully applied to the 1*BPF- and the 2*BPF-tone (BPF = Blade Passage Frequency) of the fan. Session B2-3, 16:00 h Paper number 424 (Invited, Oral) I-INCE Class 38 Tatsuya Ishii, Kenichiro Nagai, Hideshi Oinoma and Katsumi Takeda National Aerospace Laboratory, 7-44-1 Jindaiji-Higashi-Machi, Chofu-Shi 182-8522 Tokyo Japan EXPERIMENTAL STUDY ON NOISE CANCELLATION OF TURBOMACHINES This paper introduces recent research activities conducted in National Aerospace Laboratory in Japan. The contents are the noise cancellation of turbomachines, the noise reduction of supersonic jet plume, and so on. In the noise cancellation of turbomachines, fundamental experiments are carried out aiming at canceling spinning modes by imposing the anti-phased spinning modes. A ducted prop fan with a diameter of 450mm is employed as a primary noise source. An array of 16 loudspeakers or of 16 piezo-electric sound sources is located on the duct surface upstream of the fan blades and releases Internoise 2001 Abstracts 143 secondary modes anti-phased to the primary ones. Error sensors composed of microphones or PVDF are flushed with the duct surface. As a result of experiments, it has been found that the secondary mode reduces the primary mode as much as 13dB with little spillover of the secondary sound. The piezo-electric devices provided a promising potential for secondary sources and sensors of ANC. The noise reduction against supersonic jet screech tones is investigated with regard to devices such as a tab and the ejector. The tab, the much smaller obstacle compared to the nozzle diameter, contributes to breaking the feed-back structure of the screech tone as well as enhancing the mixing of jet plume with surrounding air. In addition to the acoustic measurement and thrust measurement of the 25mm-nozzle with tabs and ejectors, we tried the visualization of the shock cell structure developed at the downstream of the nozzle. Session B2-3, 16:20 h Ricardo A. Burdisso Paper number 451 (Invited, Oral) I-INCE Class 38 Vibration and Acoustic Laboratories, Department of Mechanical Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0238, USA REDUCTION OF INLET NOISE FROM TURBOFAN ENGINES USING HERSHEL-QUINCKE WAVEGUIDES An innovative implementation of the Herschel-Quincke (HQ) waveguide concept for the reduction of tonal and broadband noise from a turbofan engine is experimentally and analytically investigated. The concept consists of mounting circumferential arrays of HQ-waveguides on the engine inlet. Previous work on the HQ concept has been restricted to the control of simple one-dimensional plane waves in ducts1-5 and, more recently, to the attenuation of higher-order modes in two-dimensional ducts6. In the present study, the potential of HQ waveguide for reducing noise in cylindrical ducts in the presence of higher-order modes is investigated by applying the system to turbofan engines. The experimental work is carried out on a running Pratt and Whitney JT15D turbofan jet engine, where circumferential arrays of tubes are installed around the inlet of the engine. Effects of single and multiple arrays of HQ tubes are investigated. Measurements show high reductions in the blade passage frequency tone (more than 8 dB sound power level) and in the broadband noise (up to 3 dB in the 03200 Hz frequency range). In order to predict the noise reduction and have insight into the noise control mechanisms involved in the HQ system, a modeling technique was also developed. The approach involves modeling the tubes-inlet interfaces as finite piston sources that couple the acoustic field inside the inlet with the acoustic field within the HQ waveguides. The effects of the configuration and number of the HQ waveguides on the performance of the system will be discussed. Session B2-3, 16:40 h Paper number 478 (Invited, Oral) I-INCE Class 38 Ulf Tapken, Lars Enghardt, Wolfgang Neise and Peter Schimming DLR, Institute of Propulsion Technology, Department of Turbulence Research, Müller-Breslau-Str. 8, D-10623 Berlin, Germany ACTIVE CONTROL OF NOISE FROM TURBOMACHINES - RESULT OF RADIAL MODE ANALYSIS The noise radiated by stationary gas turbines or aeroengines is usually reduced by acoustic absorbers, whose efficiency is limited due to the confined length of the inlet and outlet duct sections. In particular dominant tonal components containing a large amount of the generated sound power are not attenuated sufficiently. Higher sound level reductions can be obtained by additional active control of the tones. In 144 Internoise 2001 Abstracts the framework of AG Turbo Turbotech II - supported by the German Ministry for Research and Technology (BMBF) - DLR, EADS and MTU Aero Engines studied the noise reduction potential of different active noise control (ANC) configurations and control algorithms for the application to turbomachines at a fan model of 1 m diameter. Theoretical investigations with regard to the realisation and design of the ANC test configuration were performed by MTU (see paper by Traub et al.). EADS provided and operated a 32-channel controller using filtered-x-LMS-algorithms and a modal control concept alternatively (see paper by Maier et al.). DLR provided the test stand and executed the radial mode analysis of the sound field resulting in the inlet with and without the application of ANC. For this purpose 96 precision microphones were placed wall flush mounted in a special duct section. The results of the radial mode decomposition are presented, which provided detailed information on the effectiveness of the ANC-techniques and allowed the calculation of the achieved sound power reductions. Session B2-3, 17:00 h Paper number 531 (Invited, Oral) Paul Traub, Fritz Kennepohl and Klaus Heinig MTU Aero Engines GmbH, Section Acoustics, Munich, Germany I-INCE Class 38 ACTIVE CONTROL OF NOISE FROM TURBOMACHINES - A NUMERICAL SIMULATION The noise radiated by stationary gas turbines or aeroengines is usually reduced by acoustic absorbers, whose efficiency is limited due to the confined length of the inlet and outlet duct sections. In particular dominant tonal components containing a large amount of the generated sound power are not attenuated sufficiently. Higher sound level reductions can be obtained by additional active control of the tones. In the framework of AG Turbo Turbotech II - supported by the German Ministry for Research and Technology (BMBF) - DLR, EADS and MTU Aero Engines studied the noise reduction potential of different active noise control (ANC) configurations and control algorithms for the application to turbomachines at a fan model of 1 m diameter. DLR provided the test stand and executed the radial mode analysis of the sound field (see paper by Tapken et al.). EADS provided and operated a 32channel controller using filtered-x-LMS-algorithms and a modal control concept alternatively (see paper by Zillmann et al.). MTU's task in this project was to develop a computer code to simulate the sound field in the intake of the fan-rig by the use of active control. The primary objective of the numerical study was to specify the numbers and locations of secondary sources (loudspeakers) and error sensors (microphones) for achieving good active control performance. The findings of the theoretical study have been verified by the ANC tests on the fan model. Internoise 2001 Abstracts 145 Session B2-4: Active noise control at the ear Chaired by Boaz Rafaely Session B2-4, 17:20 h Boaz Rafaely Paper number 611 (Tutorial, Oral) I-INCE Class 36 Institute of Sound and Vibration Research, University of Southampton, Southampton, SO17 1BJ, England ACTIVE NOISE REDUCING HEADSET - AN OVERVIEW Analogue active noise-reducing headset has been one of the most successful applications of active control of sound, with recent digital noise reducing headset promising even further improvement in performance and flexibility. This paper presents an overview of the active noise-reducing headset, from passive attenuation, through analogue control to digital control. Potential performance with analogue controllers and potential benefits and limitations of digital controllers will be discussed. Session B2-4, 18:00 h Paper number 133 (Invited, Oral) M. Winberg, S. Johansson and I. Claesson I-INCE Class 36 Department of Telecommunications and Signal Processing, Blekinge Institute of Technology, Sweden AN ACTIVE HEADSET BASED ON HYBRID ACTIVE NOISE CONTROL TECHNOLOGY In helicopters, the low frequency noise generated by the rotors and engines often masks and jeopardizes safe communication. In addition, pilots are likely to suffer from damage to their hearing due to the high sound levels in the headset produced to overcome the noise caused by increased speaker levels. A feasible approach is to reduce the low frequency noise by using active techniques combined with a method for reducing the noise in the intercom microphone signal, with lower speaker levels as a result. Helicopter noise consists of tonal components embedded in broadband noise. In order to achieve an efficient attenuation of the primary noise inside the headset, a combination of a digital feedforward controller and an analog feedback controller is employed. Spectral Subtraction is used to suppress the background noise in speech signals. This paper evaluates a combination of the two techniques and their application to real data. Session B2-4, 18:20 h Paper number 262 (Invited, Oral) Nils Westerlund, Mattias Dahl and Ingvar Claesson I-INCE Class 36 Blekinge Institute of Technology, Department of Telecommunications and Signal Processing, Soft Center, S-372 25 RONNEBY, Sweden IN-EAR MICROPHONE EQUALIZATION EXPLOITING AN ACTIVE NOISE CONTROL An active headset that employs active noise control (ANC) for noise reduction substantially reduces the influence of the low frequencies inside the cap. This implies an indirect high-pass filtering of the sound inside the auditory canal. This paper shows that the above mentioned high-pass filtering property is convenient when combining an ANC headset with an in-ear microphone (ear-mic) for communication purposes since the mouth-to-ear channel represents a fairly simple low-pass filtering. Hence, the mouth-to-ear channel is to some extent equalized by the ANC high-pass filtering, the quality of the speech signal in the auditory canal is improved and the speech intelligibility is increased. 146 Internoise 2001 Abstracts The equalization will also decrease the demand of dynamic range and resolution of electronics used. By that, combining an active headset with an ear-mic serves two purposes: Protecting the user from harmful noise and enables the user to communicate over some channel using the speech signal in the auditory canal. Session B3-1: Quiet façade and insulation Chaired by Tor Kihlman and A. Harold Marshall Session B3-1, 9:40 h Tor Kihlman Paper number 159 (Tutorial, Oral) I-INCE Class 52 Department of Applied Acoustics, Chalmers University of Technology, 412 96 Gothenburg, Sweden QUIET SIDE AND HIGH FACADE INSULATION - MEANS TO SOLVE THE CITY NOISE PROBLEM The common long-term goal for road traffic noise, Lden<55 dB cannot be reached for all dwellings neither in sprawled nor in compact cities during a foreseeable future. Strategies to improve the situation then include the use of highly insulating facades and the exploitation of the spatial noise level variations. Even though many dwellings are exposed to Lden=65-70 dB, outdoor levels in the range 40-50 Lden are not rare in the cities: courtyards in European cities being the typical case. The EU draft directive on ambient noise includes the use of facade insulation and the protection of quietness. This session has the aim to present some of the scientific foundation for these measures and discuss their potentials. Questions to answer are: What are the effects on the annoyance and sleep disturbance when dwellings have good facade insulation and/or one noisy and one quiet side? To what extent do existing, noise exposed dwellings also have a quiet side? What are the potentials to increase - also at low cost - the access to quietness in the urban renewal process? Some results from studies by the author and others will be presented. Session B3-1, 10:20 h Paper number 34 (Invited, Oral) Truls Gjestland and Bård Støfringsdal SINTEF Telecom and Informatics, N-7465 Trondheim, Norway THE INFLUENCE OF A QUIET FACADE ON ROAD TRAFFIC ANNOYANCE I-INCE Class 52 An experimental procedure has been developed to assess how high noise levels on one side of a house can be compensated by having a correspondingly lower noise level on another side. Experiments of this nature may help us understand the importance of having at least one quiet facade of a dwelling. The procedure will be demonstrated and some preliminary experimental results will be presented. Session B3-1, 10:40 h Paper number 157 (Invited, Oral) Mikael Ögren and J. Forssén Dept. of Applied Acoustics, Chalmers, SE-412 96 Göteborg, Sweden PREDICTION OF NOISE LEVELS IN SHIELDED URBAN AREAS I-INCE Class 52 In shielded urban areas the transfer paths for the traffic noise can be due to a variety of effects: multiple reflections, diffraction, refraction, and turbulence scattering. A typical situation with shielding is the quiet side of a tall building. For nearby sources some effects might be dominating, and Internoise 2001 Abstracts 147 other effects for distant sources. Predictions of noise levels are made using a ray-based model for a variety of shielded situations. The predictions are based both on sources close to the receiver and more distant ones. The results are compared with measurements, and the importance of the different transfer paths and sources are discussed. Session B3-1, 11:00 h Paper number 195 (Invited, Oral) Evy Öhrström and Annbritt Skånberg DOES ACCESS TO QUIET AREAS REDUCE ADVERSE HEALTH EFFECTS? I-INCE Class 52 Department of Environmental Medicine, Göteborg University, Box 414, SE 405 30 Göteborg, Sweden The aim of the study is to obtain knowledge of how various adverse health effects, behaviours and self-estimated noise sensitivity are related to individual noise exposures including access to a quiet side of the dwelling. Method: Adverse health effects are evaluated by questionnaires in cross-sectional studies in the field. Five pairs of study sites are selected. Each pair consists of one residential area with buildings exposed to high level traffic noise on the street side, the opposite side being quiet, and one residential area with buildings exposed to a more moderate traffic noise on both sides of the dwelling. To clarify the effect-exposure relationship for various adverse health effects and the importance of having access to a quiet side of the residence, a much more detailed assessment of the noise immission in comparison with previously performed studies is made. Assessments of the individual noise exposure of day, evening and night are made for the indoor and outdoors situation considering the location of bedroom, living room, balcony and areas for outdoors recreation. The noise exposure is assessed for LAeq, LAmax and number of noise events, L90 and L10. Results: Preliminary results will be reported from one of the five study sites. Annoyance, interference with various activities (e.g. concentration, communication and recreation) and perceived sleep quality and wellbeing will be analysed in relation to individual noise exposure indoors and outdoors. The importance of noise sensitivity, behaviours and other coping strategies will be taken into account. Session B3-1, 11:20 h Paper number 318 (Invited, Oral) Birgitta Berglund and Mats E. Nilsson I-INCE Class 52 Institute of Environmental Medicine, Karolinska Institutet and Department of Psychology, Stockholm University, SE-106 91 Stockholm, Sweden VARIATION IN PERCEIVED SOUNDSCAPE DUE TO SHIELDING BUILDING AND FACADE The perceived soundscape in residential areas varies in space and time. This variation is caused by many factors, especially the particular pattern of sound emissions and acoustic transformations due to shielding buildings and their insulation and facade reflections. Residents living in road-traffic noise exposed apartment buildings (60-65 dB LAeq,24h) participated in structured walks in their indoor and outdoor environments. During these walks, which included 6 listening stations, they repeatedly assessed their perceived soundscapes with regard to discernable sounds, perceived loudness and qualitative characterization. By master scaling perceived loudness of the total soundscapes and expressing these in pink noise equivalents (PNE), the following was found from the walks: (a) The effect of shielding buildings on the outdoor perceived soundscapes are from exposed to quiet side 15 dB PNE; (b) the effect of traffic-noise exposed insulating facades on the outdoor to indoor perceived soundscape is for the traffic noise exposed side 15 dB PNE, and for the quiet side 31 dB PNE; by opening the window 300 mm, this difference is reduced to 3 and 15 dB PNE, respectively. Thus, 148 Internoise 2001 Abstracts residents’ individually perceived soundscapes can be assessed under field situations and the resulting loudness measurements calibrated to a common scale. These perceptual measurements are helpful in evaluating the efficiency of shielding buildings, facade designs and insulation of walls. Session B3-1, 11:40 h Paper number 756 (Invited, Oral) R.F. Soames Job and Julie Hatfield Department of Psychology, University of Sydney, Sydney, 2006, Australia I-INCE Class 52 HOME INSULATION AS COUNTERMEASURE AGAINST THE EFFECTS OF INCREASED NOISE IN THE VICINITY OF SYDNEY AIRPORT: A SURVEY OF EFFECTIVENESS AND CONCERNS Home insulation is one of the main countermeasures against the effects of noise. However, its effectiveness is still in dispute. We report a small, in depth survey of residents whose homes were insulated at government expense, when noise increased in their area as a result of reconfiguration of Sydney Airport's runways. Participants were interviewed before and after home insulation was installed. Insulation generally reduced reaction to noise (annoyance, dissatisfaction and affectedness), noise-induced activity disturbance (including sleep disturbance), and subjective health effects of noise. However, reaction remained beyond generally accepted levels, with 40-50% of respondents "seriously affected" by noise after insulation. Eighty percent respondents were satisfied with the noise reduction in the bedroom, but only 40% were satisfied for living rooms. Further, some residents were concerned that sealing their homes negatively impacted their comfort or health, and found the air-conditioning systems noisy (although a good form of heating). There was also concern regarding the expense of repairs following insulation installation, and of running the air-conditioning. The acceptability and usefulness of insulation as a noise countermeasure may depend on the climate, and culture in relation to use of natural ventilation. POSTERS Session B3-1, 12:00 h Paper number 47 (Contributed, Poster) Posterboard BP-20 Wednesday, Author present 13:20 – 14:20 h Barbara Lebiedowska I-INCE Class 52 Technical University of Lodz, Department of Building Physics and Building Materials, 6, Politechniki Av., 90-538 Lodz, Poland PROTECTION FROM COMMUNICATION NOISE OF HOUSING INFRASTRUCTURE ON BASIS OF RONDO ONZ IN WARSAW Nowadays, urban acoustic climate is defined mainly by road and motor transport. Due to the absence of ring roads in major city agglomerations in Poland, through traffic is directed chiefly through housing and administrative estates. Consequently, permissible equivalent noise levels are significantly exceeded. The article will present suggestions for the protection of housing estates from excessive noise penetration through architectural and urban solutions, applicable in city districts with dense building developments. The solutions put forth are based on computer simulations using the NMPB counting algorithm and Mithra software. Internoise 2001 Abstracts 149 Session B3-1, 12:00 h Paper number 92 (Invited, Poster) Posterboard BP-21 Wednesday, Author present 13:20 – 14:20 h Tor Kihlman and Sajed Abukhader I-INCE Class 52 Department of Applied Acoustics, Chalmers University of Technology, 412 96 Gothenburg, Sweden LONG-TERM NOISE ABATEMENT PLANNING - CASE STUDIES IN GOTEBORG, SWEDEN The main goal of the project is to develop a long-term strategic noise abatement plan for Göteborg’s urban area as a case study. The noise exposure situation in this city has been evaluated. An extensive database of map layers (GIS system) and data of theoretically calculated sound levels in front of 1700 facades have been used. The process is in a micro level, i.e. involves counting the population within neighbourhoods also considering the possible access to quiet sides/yards in houses. Almost half the population of the metropolitan area of Göteborg was found to have an acceptable acoustic environment, i.e. LAeq, 24hr< 55 dB. Additionally, approximately one third of those who have LAeq, 24hr> 55 dB may have access to a quiet side/yard which may even have a good acoustic environment, i.e. LAeq, 24hr < 40 - 45 dB. A detailed analysis of selected areas has followed through collecting data from several sources: site visits, archive of the Municipality, architectural engineers, etc. Data included: age of houses, renovation type and year, window type, ownership, location and top-view shape of houses, floor/dwelling plans, population no., facade sound levels, asphalt type, traffic flow and type, etc. On this basis, the long-term plan is being worked out. Session B3-1, 12:00 h Paper number 162 (Invited, Poster) I-INCE Class 52 Posterboard BP-22 Wednesday, Author present 13:20 – 14:20 h Tor Kihlman, Wolfgang Kropp, Evy Öhrstrom and Brigitta Berglund Department of Applied Acoustics, Chalmers University of Technology, 412 96 Gothenburg, Sweden SOUNDSCAPE SUPPORT TO HEALTH. A CROSS-DISCIPLINARY RESEARCH PROGRAMME The Swedish research programme Soundscape Support to Health aims at developing methods and models and providing tools for predicting and optimising acoustic soundscapes in connection with traffic and city planning, including the production of new dwellings and rebuilding of dwellings in noise pollution areas, with respect to effects on health. The programme started late in 1999 and runs over 4 years. It is a cross disciplinary programme between three departments in three universities. It consists of 8 projects including studies of annoyance, sleep disturbance, soundscape perception, sound propagation in cities and finally town planning with regard to noise. It is financed by the Foundation for Strategic Environmental Research in Sweden. Session B3-1, 12:00 h Paper number 679 (Contributed, Poster) Posterboard BP-23 Wednesday, Author present 13:20 – 14:20 h Tommaso Meloni Swiss Agency for the Environment, Forests and Landscape, 3003 Berne, Switzerland FACTORS MODERATING THE EFFECT OF NOISE BARRIERS I-INCE Class 52 More often than not, on noise research the dose-response relationship shows a modest explained variance. A field study was carried out to investigate the effects of 13 different noise barriers, and 500 people exposed to noise were surveyed using questionnaires. First it seems that the acoustic 150 Internoise 2001 Abstracts effectiveness of the noise barriers is responsible for the reduction in annoyance. However, a detailed analysis shows that further factors moderate the relation between the noise parameters and the reaction to noise. It seems that simply introducing a noise barrier can cause a reduction in annoyance to an extent equivalent to 5 dB. This confirms the fact that the outdoor noise levels and the acoustic design of the surroundings are among the principle parameters influencing annoyance. In addition, annoyance is determined by individual acceptance of noise barriers. The degree of acceptance splits the reaction curve with a separation around 4 units on the rating-scale of 11 units of annoyance. The design of the noise barrier partly affects this splitting. There is a higher degree of acceptance of naturally constructed and transparent noise barriers than for ones that are specially formed. Other determining factors for annoyance will be discussed. Session B3-2: Effects of noise on sleep and stress and cardiovascular risks Chaired by Willy Passchier-Vermeer and Wolfgang Babisch Session B3-2, 13:20 h Paper number 151 (Tutorial, Oral) Wolfgang Babisch and Hartmut Ising I-INCE Class 66 Federal Environmental Agency, Department of Environment and Health, P. O. Box 33 00 22, 14191 Berlin, Germany NOISE INDUCED STRESS IS A RISK FACTOR IN CARDIOVASCULAR DISEASE The risk factor concept of cardiovascular diseases (CVD) implies that numerous external and internal factors play a role in the development of CVD. Noise may be counted together with smoking, lack of physical activity etc. to the external risk factors while biochemical alterations such as increases of cholesterol or fibrinogen are internal risk factors. Chronic influences of external factors may over many years lead to increases of the mentioned internal risk factors and thus to artheriosclerosis and CVD - for example myocardial infarction. One of the conditions that long term noise exposure will enhance this pathomechanism is a noise induced chronic increase of stress hormones such as adrenaline, noradrenaline and/or cortisol. In several work noise studies it was shown that persons exposed for years to intensive noise (Leq > 95 dB(A) had chronic elevations of stress hormones and internal risk factors. A few studies revealed additionally an increased risk for hypertension, especially in workers with complex activities. The connection between environmental noise (indoor Leq < 40 dB(A)) and stress reactions during sleep is explained by recently discovered functions of the amygdala. This region of the brain stem plays an important role in the auditory warning system and is able to differentiate between neutral sounds and those implying danger i.e. an approaching lorry. In several studies it was demonstrated that traffic noise immissions into the bedroom with average maximal levels of 35 dB(A) or more have the potential of causing chronic increases of night-time cortisol. The described process of stress hormone increases and internal CVD risk factor development leads in the long term to an increased CVD risk. By assessment of different lifetime risks the traffic noise related CVD risk will be compared quantitatively to other environmental risks. Internoise 2001 Abstracts 151 Session B3-2, 14:00 h Paper number 225 (Invited, Oral) G. Bluhm, E. Nording and N. Berglind I-INCE Class 62 Department of Environmental Health, Stockholm County Council, Stockholm, Sweden INCREASED PREVALENCE OF HYPERTENSION IN A POPULATION EXPOSED TO ROAD TRAFFIC NOISE A relation between chronic noise exposure and hypertension has been reported in some epidemiological studies but the evidence is still conflicting. Our objective was to study the prevalence of hypertension related to road traffic noise in residential settings. A questionnaire regarding environmental related health effects was distributed to a random sample of 1000 individuals, 19-80 years old in an area with busy traffic outside Stockholm. After using defined exclusion criteria 568 persons were included in the study. Different prediction models for road traffic noise were used to classify the individuals into four exposure categories with Leq(24h); 40-45 dBA, 45-50 dBA, 50-55 dBA, and >55 dBA respectively. At noise levels >50 dBA (n=357) and <50 dBA (n=211) a medical diagnosis of hypertension was reported by 16% and 8% respectively. Adjusting for potential confounding factors resulted in a prevalence odds ratio (POR) for hypertension of 1.8 (95%CI 1.03.2). Corresponding POR for men and women were 1.4 (95% CI 0.6-3.1) and 2.5 (95% CI 1.0-6.2). The risk was somewhat higher in younger age groups (<50 years) with a POR of 2.7 (95% CI 0.9-8.7). With longer residential time (>10 years) POR was 2.8 (95% CI 1.2-6.4). Adjusted POR:s in the four exposure categories were 1.0, 1.5 (95% CI 0.6-4.4), 2.0 (95% CI 1.0-4.2), and 2.2 (95% CI 0.8-6.0) respectively. In conclusion a relation between road traffic noise exposure and hypertension was indicated especially among women and in younger age groups. A long residential time seemed to be of major importance. Session B3-2, 14:20 h Paper number 265 (Invited, Oral) C. Maschke, K. Hecht and H. Niemann Institute for Technical Acoustics,TU-Berlin, Einsteinufer 25, 12587 Berlin, Germany EFFECTS OF NOISE BEFORE, DURING AND AFTER PREGNANCY I-INCE Class 62 In the recent years the preventive-medical assessment of noise in Germany gives higher attention to risk-groups like children, elderly and sick persons. An additional risk-group that until today is hardly be considered, are pregnant women. Pregnancy, birth and postpartale life-phase represent high psychobiological demands which are related with multiple health risks. It has turned out that 10 - 20% of pregnant women suffer from pregnancy-depressions and over 60% from postpartum-depressions. In approximately 10% of the cases, the duration is more than one year. These depressions are frequently related with sleep disturbances, internal unrest, daily tiredness, and a loss of self-esteem. Since the sleep of pregnant women is frequently interrupted due to the pregnancy, the greater difficulty falling asleep under the exposure to noise represent an increased demand. After delivery, this situation remains, sometimes even more serious than during pregnancy. The poly-phasic sleep of the newborn leads in the mother, who shows a mono-phasic sleep, to sleep-interruptions, that already are a burden to the sleep of the young mother even without noise. 152 Internoise 2001 Abstracts Session B3-2, 14:40 h Paper number 345 (Invited, Oral) Andy Smith, Neil Rich, Sue Wilson and David Nutt Centre for Occupational and Health Psychology, Cardiff University, UK I-INCE Class 62 SUBJECTIVE AND OBJECTIVE ASSESSMENT OF THE EFFECTS OF NOISE, NOISE SENSITIVITY AND NOISE DISTRUBED SLEEP ON HEALTH The present research combines two different approaches to examine associations between community noise exposure, sleep disturbance, noise sensitivity and subjective reports of health. The first approach involves a longitudinal community survey which allows identification of cross-sectional and lagged associations as well as assessment of adaptation and response to change. Following this a cohort study has been conducted involving assessment of sleep by actimeters. Similarly, objective indicators of noise sensitivity have been examined and compared with subjective reports. The main objective has been to determine the profile of associations revealed using subjective and objective indicators and these data will be presented. Session B3-2, 15:00 h Paper number 392 (Invited, Oral) Gunn Marit Aasvang and Bo Engdahl I-INCE Class 62 National Institute of Public Health, Department of Environmental Medicine, Section for Air Pollution and Noise, Box 4404 Nydalen, N-0403 Oslo, Norway SLEEP DISTURBANCES DUE TO NIGHT-TIME NOISE FROM RAILWAY, AIRCRAFT AND ROAD TRAFFIC: PRESENTATION OF A LONGITUDINAL PROJECT AND AVAILABLE SURVEY DATA A longitudinal project on transportation noise and its effect on physiological and subjectively reported sleep will be presented. So far, a social survey is carried out to obtain data on subjective sleep quality and night time noise annoyance. Further studies will evaluate the effect of noise abatement measures on both subjective and physiological sleep. A questionnaire was mailed to a community sample (n=8000) in and around Oslo exposed to either prevailing aircraft, railway or road traffic noise. The questionnaire contained items concerning physical and mental health and subjective health complaints, in addition to questions about sleep quality, noise annoyance and noise sensitivity. Data on subjective sleep quality and night time noise annoyance are coupled with calculated noise data for each respondent, to obtain dose-response relationships. An extensive noise mapping by the transportation authorities will form the basis for a noise abatement program. Respondents from the survey, who will be affected by the noise abatement, will be invited to participate in a follow up study. This study will evaluate the effect of noise abatement measures on sleep from railway and road traffic noise. Both subjective and physiological (EEG) sleep variables will be obtained in the natural setting of the respondents own home, before and after noise reduction. The paper presents available data from the social survey and study design for the whole project. Internoise 2001 Abstracts 153 Session B3-2, 15:20 h Paper number 602 (Invited, Oral) M. Schümann and H. Neus I-INCE Class 62 Arbeitsgruppe Epidemiologie der BAGS und des IMDM/ Univ.Klinik Hamburg-Eppendorf, Winterhuder Weg 29, D22085 Hamburg / Germany ASSESSMENT OF NOISE RELATED CARDIOVASCULAR RISK: EVALUATION OF UNCERTAINTY IN POPULATION RISK ASSESSMENT Environmental health impact assessment with respect to noise might contribute to public health priority setting and environmental decision-making. Based upon recent reviews on epidemiological evidence noise induced cardiovascular risk will be evaluated, including sensitivity analyses with respect to selected model parameters. The aim of this study is to determine the influence of traffic noise levels on population cardiovascular risk (ICD9:410-414) and total mortality. For quantitative risk assessment several model parameter sets must be considered: distribution of the environmental noise levels, distribution of the population exposure prevalence, dose-effect relation (exposureresponse function), exposure-effect-latency, age- and sex-related specific mortality as well as total population mortality rates and age- and sex-structure of the target population. The absolute and relative impact of noise exposure at a level higher than 65 dB (A) on cardiovascular mortality (excess cases in the population and lifetime lost) is calculated. The model input parameters are chosen or estimated for the German population. The combination of these parameter distributions in a linear risk assessment model gives the basis for the calculation of the population attributable risk estimates. Since the scientific basis of a noise related health impact assessment is uncertain at least with respect to the parameters relative risk and latency, a sensitivity and uncertainty analysis will be presented in order to evaluate the degree of influence of the parameter sets in the model. Session B3-2, 15:40 h Paper number 227 (Invited, Oral) Stephen A. Stansfeld, Sarah L. Brentnall and Mary M. Haines I-INCE Class 66 Department of Psychiatry, Barts and The London, Queen Mary School of Medicine and Dentistry, Queen Mary, University of London, UK INVESTIGATING THE EFFECTS OF NOISE EXPOSURE ON STRESS HORMONE RESPONSES IN CHILDREN Aircraft noise exposure has been associated with impairments of cognitive performance and health and raised annoyance levels in children. It was anticipated that these effects would be accompanied by psychophysiological disturbance indicated by raised 'stress' hormone levels. Results from two studies of children, 9-11 years old, exposed to aircraft noise around London (Heathrow) Airport are presented: The Schools Environment and Health Study and the West London Schools Study. In the first study, salivary cortisol was measured prior to cognitive testing and immediately after testing in the morning in 114 pupils from 4 high aircraft noise schools (>66 dBA) and 124 pupils from 3 low noise schools (<57 dBA). In the second study, adrenaline, noradrenaline and cortisol were measured in overnight urine samples in 96 pupils from 8 high aircraft noise schools (>63 dBA) and 108 pupils in 8 low noise schools (<57 dBA). Although associations between catecholamine levels and self reported stress levels were found in the second study, there was no association between hormone levels and aircraft noise exposure. The methodological and theoretical implications of these results will be discussed in the context of other studies, which have found effects of aircraft noise on hormone levels. 154 Internoise 2001 Abstracts Session B3-2, 16:00 h Paper number 666 (Invited, Oral) Ric Van Poll, Masja Straetemans and Nancy A. Nicolson RIVM-LBM, PO Box 1, 3720 BA BILTHOVEN, The Netherlands AMBIENT NOISE IN DAILY LIFE: A PILOT STUDY I-INCE Class 66 Noise is one of the most important ambient stressors in urban residential environments because of the large number of people adversely affected. Although the acute health effects of ambient noise are not dramatic, long-term exposure to ambient noise may contribute to the development of chronic disease through stress pathways. In this study we examined the stress inducing potential of ambient (traffic) noise in daily life. Stress effects studied were the psychological (annoyance) and the physical (saliva cortisol) effects due to traffic noise exposure. Subjects (N=30) were residents from two areas differing in average traffic sound level pressure (65 versus 50 dB(A), respectively HE- and LE-group). The HEgroup lived near a major traffic highway. Subjects recorded daily noise experiences (‘moments’) by filling in a short questionnaire and collecting a saliva sample 10 times per day for four consecutive days. The questionnaire contained questions on cognition, mood, context (time, place, person), and noise (annoyance, intensity, source etc.). The HE-group reported higher levels of annoyance throughout the day than the LE-group (mean: 3.0 versus 2.5, range 1-7). Also the HE-group reported more noise moments to be annoying (score > 2) than the LE-group(57 versus 33%). Not surprisingly, road traffic was reported more frequently by the HE-group to be the most annoying noise source (30 versus 8%). Activities appeared to be more frequently disturbed by noise in the HE-group (21 versus 8%). Cortisol levels (diurnal rhythm) appeared not to be different between the two groups. The relationship between annoyance and cortisol levels will be explored in an additional analysis (MLA). Diary methods in conjunction with ambulatory physiological assessments seem a promising tool for in-depth studies of theperson-environment transaction. Psychological stress (annoyance) due to noise could be well assessed. The results of the physiological stress measure are still under scrutiny. Session B3-2, 16:20 h Paper number 192 (Contributed, Oral) I-INCE Class 62 M. Basner, H. Buess, N. Luks, H. Maass, L. Mawet, E.W. Mueller, U. Mueller, C. Piehler, G. Plath, E. Rey, A. Samel, M. Schulze, M. Vejvoda and J. Wenzel DLR, Institute of Aerospace Medicine, D-51170 Cologne, Germany THE INFLUENCE OF NOCTURNAL AIRCRAFT NOISE ON HUMANS - A NEW COMPREHENSIVE APPROACH The DLR Institute of Aerospace Medicine in Cologne investigates the influence of aircraft noise at night on humans. 128 healthy volunteers aged 18-65 are being examined in four studies during 13 consecutive nights (23:00-07:00) in our sleep laboratory. The results will be validated in 2 field studies investigating 64 volunteers, thus 2496 nights in total will be observed. Various physiological signals are sampled: polysomnographic measurements (EEG, EOG, EMG), the electrocardiogramme, finger pulse amplitude, respiratory movements and actigraphy. The stress parameters cortisol, epinephrine and norepinephrine are extracted from nocturnal urin samples. The AGARD-Stress battery is used to examine the possible interference of aircraft noise with mental performance. In order to measure state and stress reactions psychometric questionnaires are filled out twice a day. In 9 of the 13 nights of the laboratory studies aircraft noise with varying frequencies of occurrence (4 to 128 events) and noise levels ranging from 50 to 80 dB LASmax (LAeq 31.2-52.6 dB) is presented by loudspeakers. Correct playback is assured by remeasurements of terz levels in each bedroom. The noise has been recorded in Internoise 2001 Abstracts 155 bedrooms of residents living near the airport with half closed windows using a class 1 noise level meter. Since all parameters are stored simultaneously on hard disk, analysis and correlation of single noise events with physiological parameters are possible. The design of the laboratory studies will be presented in detail. As in December 2001 data collection of three of the four groups will be finished, some preliminary results may also be presented. POSTER Session B3-2 Paper number 279 (Invited, Poster) I-INCE Class 62 Posterboard CP-1 Thursday, Author present 9:40 – 10:40 h Toshihito Matsui, Takashi Uehara, Takashi Miyakita, Kozo Hiramatsu, Yasutaka Osada and Takeo Yamamoto Asahikawa Medical College, Asahikawa 078–8510, Japan ASSOCIATION BETWEEN BLOOD PRESSURE AND AIRCRAFT NOISE EXPOSURE AROUND KADENA AIRFIELD IN OKINAWA There are only few community studies that show the relationship between aircraft noise and cardiovascular functioning. We examined the effects of aircraft noise on blood pressure of residents living around military airports in Okinawa, Japan. The Japanese government suggests that citizens over 40 years old should receive health examination on the basis of Health and Medical Service Act for the Elderly. The data obtained by the health examination for the years of 1994 and 1995 (28,781 subjects) were analysed with respect to systolic blood pressure and diastolic blood pressure. Multiple logistic regressions were applied to the rates of subjects having higher blood pressure. Age, sex, BMI and their interaction terms were included in the logistic model as confounding factors. The results showed clear dose-response relationships between blood pressure and noise exposure level expressed by WECPNL. The odds ratio of the noise-exposed group with WECPNL over 85 was 1.3 referred to the control. The significant increase of odds ratio was also found in the noise exposed group with WECPNL from 75 to 80. Similar results were obtained on the rate of hypertension. Highly significant dose-response relationships were found, and the odds ratio of the highest noise-exposed group was 1.4 referred to the control. Session B3-3: Effects of noise on children Chaired by Mary Haines and Staffan Hygge Session B3-3, 16:40 h Paper number 55 (Invited, Oral) I-INCE Class 61 M.M. Haines, S.A. Stansfeld, S.L. Brentnall, J.A. Head, B. Berry, M. Jiggins and S. Hygge Department of Psychiatry, Queen Mary, University of London, UK WEST LONDON SCHOOLS STUDY AIRCRAFT NOISE EXPOSURE AT SCHOOL AND CHILDREN'S MENTAL HEALTH AND ANNOYANCE The objective of the West London Schools study was to test whether: a) cognitive impairments (in reading, memory and attention); and b) stress responses (noise annoyance and self reported stress) 156 Internoise 2001 Abstracts previously found in children are attributable to aircraft noise exposure, after adjustment for confounding factors both at the school and individual level, and to examine whether children exposed to high levels of social disadvantage are at greater risk of noise effects. The cognitive performance and health of 451 children aged 8-11 years attending ten schools in high aircraft noise areas (16-hr outdoor Leq >63 dBA) was compared with children attending ten matched control schools exposed to lower levels of aircraft noise (16-hr outdoor Leq < 57 dBA). Health and cognitive tests were group administered in the schools. Analyses used multi-level modelling to take account of school level factors. Noise exposure was associated with impaired reading on difficult items and raised annoyance after adjustment for age, main language spoken, household deprivation. There was no variation in the size of the noise effects in vulnerable sub-groups of children. High levels of noise were not associated with impairments in memory, attention, mean reading score nor self-reported stress. Aircraft noise was weakly associated with hyperactivity and psychological morbidity. The cognitive results indicate that chronic aircraft noise exposure does not always lead to generalised cognitive effects but, rather, more selective cognitive impairments on difficult cognitive tests in children. It can be concluded that chronic noise exposure is associated with raised noise annoyance in children. Session B3-3, 17:00 h Paper number 220 (Invited, Oral) Staffan Hygge, Gary W. Evans and Monika Bullinger University of Gävle, SE-801 76 Gävle, Sweden MEDIATORS OF NOISE EFFECTS IN CHILDREN'S COGNITION I-INCE Class 63 Despite several decades of careful, thoughtful thinking and empirical work on noise and cognitive performance, we still have very little in the way of a clear and detailed theoretical model of how noise affects performance. Possible mediating links between noise on the independent side and cognitive outcomes on the dependent side, need to be formulated and tested. These mediators include other cognitive and perceptual measures, as well as motivation and psychophysiological processes. The paper will formulate and discuss a few of these mediational hypotheses, expand on them and try to find out if at least some of them are testable with the three-wave children data from the new and old airports in Munich. Session B3-3, 17:20 h Paper number 412 (Invited, Oral) I-INCE Class 63 H.H. Emmen, B.A.M. Staatsen, J.B. Deijen, P.H. Fischer and I. Van Kamp TNO-Voeding Zeist, The Netherlands NEUROBEHAVIORAL MEASUREMENTS IN CHILDREN LIVING AROUND SCHIPHOL AIRPORT; FURHTER METHODOLOGICAL CONSIDERATIONS Unfortunately results of epidemiological studies conducted in children living close to airports suggest that effects of aircraft noise on cognitive performance may occur at high levels of exposure. The available data are insufficient to estimate the impact in the population around Schiphol Airport. Therefore a feasibility study was carried out in 1995. This study was aimed at testing the reliability and suitability of selected automated behavioral methods (Neurobehavioral Evaluation System, NES) that measure cognitive and psychomotor functioning of children in a school environment and questionnaires designed to measure psychological well-being, annoyance and sleep quality, in children. In addition the potential differences in psychological functions between groups of children exposed to different levels of aircraft noise were explored. The study was part of the Health Impact Internoise 2001 Abstracts 157 Assessment Schiphol Airport programme, that is being conducted in response to the planned expansion of the Airport. Attention, psychomotor performance, perceptual coding, learning and memory were repeatedly measured in two groups of 80 children in the age of 8-11 years at school, hereby contrasting a high and low (control) exposure (to aircraft noise) group. The children were tested twice during school hours in the period May-June 1995 with a 4-6 week interval between testing. The average noise level (Laeq, 24 hours) in the high exposed area was 59 dB(A) at the first session and 52 dB(A) at the second session. Precise information on the noise level in the control group is lacking. The test-retest repeatability and precision of most of the tests and rating scales was fairly good with a correlation coefficient above .70 good enough to detect differences in the selected neurobehavioral parameters. Small but statistical significant differences were found between the high and low exposed groups for psychomotor performance (Hand Eye Coordination) and attention (Switching Attention). In the high exposed areas significantly more children reported annoyance (76 vs 40%). The reduction in noise levels between the first and second session did not result in consistent neurobehavioral changes. The pilot study demonstrated the feasibility of measuring a wide range of neurobehavioral parameters using an automated test (NES) and questionnaires in the school environment. No firm conclusions regarding the relationship between aircraft noise and performance in children can be drawn; the relatively small number of children and the lack of precise exposure information being limiting factors. It is recommended that future research should test at least 500 children selected from locations with known exposure levels. The contrast in aircraft noise exposure should be maximised by a good selection of locations and estimates of (individual) exposure levels. In the design of a new study (RANCH) which started this year, the recommendations of the pilot study are accounted for. We will investigate 500 children, all living in the vicinity of Schiphol Airport. Both automated behavioral methods and paper and pen tests will be used. In addition noise contours will be used to assess the (individual) exposures. Session B3-3, 17:40 h Paper number 475 (Contributed, Oral) Kjell Holmberg and Pär Lundquist National Institute for Working Life, P.O. Box 7654. SE-907 13 Umeå, Sweden NOISE LEVELS AT SCHOOL I-INCE Class 63 At school the activity in the classroom is the dominating noise source during class. Still, many complaints are raised by teachers and students about the background noise and different acoustical aspects of the classroom. An example of the situation in Swedish schools is a survey by the Swedish National Board of Occupational Safety and Health pointing out that the second most common environmental problem at school is ‘noise, sound and acoustic problems’. The purpose of the present study was to evaluate noise levels during class. Level distribution in time was considered, as was reverberation time for the specific room. Students perceived mood was studied. Sound levels were recorded in 25 classrooms chosen at three different, still typical schools in the northern part of Sweden. The lessons under study were all about the same type of schoolwork carried out under similar circumstances. Background noise was recorded, as was the noise during classroom work. The recordings were analysed according to equivalent sound levels and temporal distribution of level. The class levels ranged from 43 to 68 dB(A). Reverberation time did not seem to affect class level and sensations. 158 Internoise 2001 Abstracts POSTER Session B3-3 Paper number 467 (Contributed, Poster) Posterboard CP-2 Thursday, Author present 9:40 – 10:40 h Pär Lundquist, Anders Kjellberg and Kjell Holmberg National Institute for Working Life, P.O. Box 7654. SE-907 13 Umeå, Sweden EVALUATION OF CLASSROOM NOISE: INSTRUMENT FOR SELFREPORTED MOOD I-INCE Class 66 School serves a variety of functions. The most obvious function is to educate. Many aspects of the school children's mood and attitudes such as their concentration, interest, attention, fatigue, and arousal probably affect their scholastic performance. These qualities may, in turn, be affected by e.g. the physical and psychosocial environment. The aim of this study was to develop a mood-rating instrument to identify effects of noise in the classroom environment, that are of relevance for the children's scholastic performance. It is desirable that the instrument is easy to administer and quickly completed. The target group is children in upper compulsory school in Sweden. An adjective checklist containing 45 mood-describing adjectives was constructed and answered by 280 students. Thirteen of the items had a non-response rate above ten percent and were excluded. The remaining 32 items were subjected to factor analyses. 443 students were used to cross-validate the obtained factor structure. The analyses showed that the adjective checklist reflected two slightly negatively correlated latent factors. The two factors looked like opposites on a bipolar dimension, but were only moderately correlated and were therefore kept apart. The reliability of the two resulting scales was high and the model explained a large part of the item variance. Session B4-1: Noise and neighbours Chaired by Jef van Dongen Session B4-1, 9:40 h Jef E.F. Van Dongen Paper number 19 (Invited, Oral) I-INCE Class 51 TNO Prevention and Health (TNO-PG), P.O. Box 2215, 2301 CE LEIDEN, The Netherlands NOISE ANNOYANCE FROM NEIGHBOURS In the Netherlands noise annoyance from neighbours is experienced in approximately 1/3 of all households (= 2.2 million). To support policy actions to improve the situation a field study was performed into the dose-effect relationships, i.e. the relationships between sound insulation and dwellings on the one hand and the ability to hear sounds from neighbours and the annoyance caused on the other. The study also aimed to ascertain the extent to which factors other than noise from neighbours alone such as social, environmental and individual-related factors influenced dose-effect relationships. Verbal questionnaires were taken in 660 of the most common types of dwellings in the Netherlands. Acoustic quality was established in 202 of these dwellings. The units Ilu, Ilu;k (both indexes of airborne sound), Ico, (impact sound), Ga;k sound insulation to the exterior) and the GWK (a calculated sound insulation category) were used as parameters for the dose (the sound insulation quality). It was found that the annoyance clearly diminishes in proportion to the quality of the sound insulation (if measured in Ilu or GWK). For example this is the case with the ‘general annoyance’ and Internoise 2001 Abstracts 159 with the annoyance caused by TV/radio/audio equipment being on loud, people walking heavily on floors and by showering and taking baths in neighbouring dwellings. Factors that affect the annoyance are the age of the respondent, individual sensitivity to noise, understanding with the neighbours, tolerance, degree of satisfaction with the dwelling in general and with its state of construction and the degree of satisfaction with home environment. Session B4-1, 10:00 h Eddy Gerretsen Paper number 603 (Invited, Oral) I-INCE Class 51 TNO TPD, P.O.Box 155, NL-2600 AD Delft, The Netherlands CLASSIFICATION OF ACOUSTIC QUALITY OF DWELLINGS - BACKGROUNDS OF A RENEWED DUTCH STANDARD The acoustic quality of dwellings relates to airborne and impact noise from neighbours, reduction by the facade of outdoor noise and noise from technical installations. In the past the requirements for these aspects were primarily based on the acoustic performance in existing situations which traditionally were considered as 'acceptable'. The aim of the renewal of the Dutch standard for acoustic requirements was to introduce several classes of acoustic quality, not just minimum requirements, and to specify for all mentioned aspects requirements, which were expressed in a comparable way and were balanced. To deduce these quality classes use was made of a global model for the relations between source, transmission and receiving sound levels as well as a global model for the perception of the noise. To that end it was also tried to combine the results of several social surveys on noise from neighbours, however different in goal and method these surveys were. These backgrounds to the new standard (NEN 1070) will be given as well as the necessary simplifications to be able to derive a practical standard for the building industry. Session B4-1, 10:20 h Paper number 621 (Invited, Oral) I-INCE Class 51 Julie Hatfield, R. F. Soames Job, Norma Carter, Peter Peploe, Richard Taylor and Stephen Morrell Department of Psychology, University of Sydney, Australia NOISE FROM NEIGHBOURS: PREVALENCE, IMPACT ON REACTION, AND DETERMINANTS Noise from neighbours may be a considerable noise nuisance. It may also influence residential satisfaction, which may in turn influence responses to other noises (e.g. aircraft noise). In a community survey, conducted as part of the Sydney Airport Health Study, 1015 respondents indicated whether they ever heard a range of noises from their neighbours (e.g. neighbours' television, barking dogs). We assessed whether respondents who heard noises from their neighbours (compared to those who did not) were less satisfied with their place of residence and more annoyed by aircraft and traffic noise. We also considered whether sociodemographic variables (e.g. age, gender, employment status, personality, noise exposure), influenced whether respondents reported hearing noise from their neighbours. In a laboratory study, 150 undergraduate psychology students identified three outside noises which they hear regularly where they live and rated their dissatisfaction with these noises. We assessed the importance of noise from neighbours to these students' noise reaction. Results are discussed in terms of degree of noise nuisance presented by noise from neighbours background noise be employed sensibly, which requires a better understanding of the interactive effects of noise from combined sources. 160 Internoise 2001 Abstracts Session B4-1, 10:40 h Rainer Thaden Paper number 643 (Invited, Oral) I-INCE Class 51 Institute of Technical Acoustics, RWTH Aachen, Templergraben 55, 52056 Aachen, Germany, INVESTIGATION ON THE RELATION BETWEEN SINGLE NUMBER RATINGS AND SPEECH INTELLIGIBILITY In practice building engineers deal with single number ratings of sound insulation quantities. The method of transforming the frequency-dependent sound insulation in single number ratings is criticised because the reference curve is not always appropriate for modern constructions. Particularly low frequencies are problematic. In R'w, for instance, frequencies below 100 Hz and above 3150 Hz are not taken into account. It is thus possible that very different sound insulation curves come up with same single number ratings. In listening tests regarding speech intelligibility, where a listener in the receiving room had to understand spoken words, it was found that room situations with same single number ratings had very different speech intelligibility. It was tried to find a rating scheme which takes the speech privacy into account, similar to the spectrum adaption terms C and Ctr. Rating schemes known in building acoustics and quantities from speech research will be considered. To create audio signals for these tests, an algorithm for auralisation of sound insulation was developed and used together with a sentence test. Session B4-2: Sound quality Chaired by Jean-Dominique Polack and Reinhard Weber Session B4-2, 11:00 h Paper number 758 (Invited, Oral) Takeo Hashimoto and Shigeko Hatano Seikei University, 3-3-1 Kichijoji Kitamachi, Musashino-shi, Tokyo 180-8633 Japan EFFECT OF SEAT VIBRATION AND VISAL SCENERY TO THE PERCEPTON OF SOUND QUALITY OF CAR INTERIOR NOISE I-INCE Class 61 The perception of sound quality of car interior noise is much influenced by the simultaneous exposure of seat vibration and visual scenery from the wind screen. Our results show that simple exposure of noise is strengthened by the simultaneous exposure of seat vibration and that is weakened by the simultaneous exposure of visual scenery. Session B4-2, 11:20 h Robert Barti Paper number 693 (Contributed, Oral) I-INCE Class 61 Director of acoustic department of La Salle of Ramon Llull, University Pg. Bonanova, 8. 08022 Barcelona, Spain ACOUSTIC QUALITY OF DOORSLAM Door slam identifies the quality of a car. Car's manufacturer wants to improve their design with high quality products. Actually people want better finished products. The sound emitted by a door slam is very important and let to the driver "to know" how strong and compact the door car is. Is just a subjective sensation, but consumers decides his buying with the sound of door slam. The doors of cars are very complex structure. The door is made in steel. Other parts are mounted in the door, the window, loudspeaker, electric motor, etc. The sound is good when it has short time duration, and high Internoise 2001 Abstracts 161 frequency components free. In this paper a study of subjective sensation with a jury is doing. Some commercial cars where analysed in order to know the correlation between the preferences of the jury test and some acoustic indicators. A new tool for automatic classification of door slam acoustic quality is showed. Session B4-2, 11:40 h Paper number 155 (Contributed, Oral) M. Shafiquzzaman Khan I-INCE Class 63 Division of Railway Technology, Department of Vehicle Engineering, Royal Institute of Technology, Teknikringen 8, 100 44 Stockholm, Sweden DETERMINATION OF SOUND QUALITY IN SWEDISH PASSENGER TRAINS Sound quality for the passenger trains is one of the main concerns for the railway companies. The comfort of the passengers related to the train noise is measured in dBA and such measurement often fails to describe annoying noises. The purpose of the study is to determine sound quality for the passenger trains. The study includes both subjective and objective studies. The subjective study utilizes a questionnaire survey in order to determine annoying noises to the passengers. The questionnaire was distributed randomly to 380 passengers in four types of trains. The results showed that passengers who sat near to the entrances were more annoyed than those who sat in the middle of the compartment. Non-identified sources of noises such as rattling, squeaking, rail-joints and beating are causing most annoyance to the passengers. Besides train noises, mobile telephone and children cry is also reported to cause annoyance to the passengers. The objective study comprised recording and reproduction of the non-identified sources of noises. The objective study was administered to see the masking effects on annoying noises. Thirty-nine subjects from various professions performed the listening tests. The listening tests comprised of three sessions, namely original and modified nonidentified sources of noise stimuli. The stimuli were modified by adding background levels of 62 dBA and 65 dBA to each of the stimuli. The results showed a reduction of annoyance for overall acoustical comport in passenger trains if the background level is raised to 65 dBA. Session B4-2, 12:00 h Takeshi Akita Paper number 431 (Contributed, Oral) I-INCE Class 63 Department of Architecture, Faculty of Engineering, Tokyo Denki University, 2-2 Kanda-Nishiki-cho Chiyoda-ku, Tokyo 101-8457, JAPAN INTERACTION BETWEEN HEARING AND VISION: EFFECTS OF ONE MODAL TASK ON THE OTHER MODAL REACTION In usual environment, a person recognizes his environment using several modality, especially hearing and vision. Some papers report that there exists cross modal effect and interaction between the sense of hearing and sight. In the previous research, no interaction between primal sensory processing of auditory and visual information in brain was reported. However, it is supposed that some interaction may exist in more advanced psychological stage. In the present research, one modal reaction when a subject pays attention to the other modal task is examined to reveal interaction between auditory and visual sense in cognition level. Two experiments are carried out for that purpose. In experiment 1, subjects evaluate impression of seven kinds of sound that are presented when they do visual task that requires them to write down many meaningless characters. In experiment 2, subjects react to visual targets that are presented on the display as soon as possible when they do auditory task that requires 162 Internoise 2001 Abstracts them to memorize the contents of stories. These results are compared with normal evaluation of the sound and reaction to the visual target. Results show that auditory impression is more detailed under normal condition than task condition. They also show that normal visual reaction is faster than the reaction under task condition. It is suggested that much attention to one modality causes reduction of attention to the other modality. Reduced attention to non-attended modality seems to lead to decrement of auditory and visual recognition. Session B4-2, 13:20 h Yunseon Ryu Paper number 191 (Contributed, Oral) I-INCE Class 72 Bruel & Kjaer Sound & Vibration Measurement, Skodsborgvej 307 Naerum 2850 Denmark DEVELOPMENT OF THE ACOUSTICALLY GOOD EXHAUST SYSTEM IN PASSANGER'S VEHICLE USING SOUND QUALITY TECHNIQUES The exhaust system could be a dominant acoustical source in the passenger's vehicle. It would be very important to obtain the acoustically good exhaust system, in order to control the cabin interior sound in automotive. In order to obtain the acoustically good exhaust system in automotive, many kinds of exhaust system should be measured, and simultaneously those results should be compared by the sound quality parameters. In this paper, in order to develop the methodology determining sound quality parameters, acoustic simulator is introduced, combining the time domain analysis and convolution analysis. As an example to verify the reliability of this method, several kinds of measurements are carried out, and the acoustically good exhaust system is selected, based on this proposed method. Session B4-2, 13:40 h Paper number 322 (Contributed, Oral) Marleen Adams, Patrick Van de Ponseele and Filip Deblauwe LMS International NV, Interleuvenlaan 68, 3001 Heverlee, Belgium I-INCE Class 79 SOUND QUALITY EQUIVALENT MODELING FOR VIRTUAL CAR SOUND SYNTHESIS The pressure on development cycles in the automotive industry forces the acoustical engineers to create awareness of sound quality in the early stages of development, perhaps even before a physical prototype is available. Currently, designers have few tools to help them listen to their "virtual" models. For the design of a synthesis platform of in-vehicle binaural sound, the sound should be modeled with almost identical sound quality perception. A concept is presented where the total sound of a vehicle is split in a number of components, each with its own sound characteristics. These characteristics are described in a signal model that allows the analysis of an existing sound into a limited number of signal components: orders- frequency spectra, time envelopes and time recordings. The introduction of this sound synthesis concept in a hybrid model consisting of networks of transfer paths, force inputs and noise sources will allow the evaluation of the impact of changes to structural components on the sound quality of in-vehicle sound. This publication will report on the experience gained and the progress made in these fields. Internoise 2001 Abstracts 163 Session B4-2, 14:00 h Lucille Lamotte Paper number 561 (Contributed, Oral) I-INCE Class 79 MicrodB, 7b, allée Claude Debussy, 69130 Ecully, France LOCALIZATION AND QUANTIFICATION OF SOUND QUALITY The acoustic quality is become a demand in many domains. To qualify it, it is useful to affect the good acoustic metrics to the right noise source. This paper develops a method that combines localization technique and metrics computation. The first step concerns the acquisition of signal. While an array or a simple sensor in far field records many sources, the human ear separately perceives sources. From this kind of measurement, it is difficult to get the signature of each source like human ear does. The solution is to affect a near field sensor to each source, this sensor should be a microphone or an accelerometer. Or a more expensive solution is an array of microphone adapted to the focusing technique. The second step consists in separating each sources. Phenomenon that are visible on focused signal are not detectable from raw signals on the array. Or an accelerometer can traduce audible shocks that are melted to other close sources in the acoustic signature (for example with piston slap). At least, the right metrics can be selected. In this step, different signal processing techniques like filtering, time/frequency, sonogram, are necessary to detect the phenomenon in time and frequency domain. Acoustic metrics like roughness, impulsiveness, kurtosis Ö that are computed on the right frequency band are more pertinent. This method has been applied for engine in stationary or run-up condition with good results. Resonance are affected to the right component, modulation aspects due to the belt are quantified in a right scale. POSTERS Session B4-2, 14:20 h Paper number 123 (Contributed, Poster) I-INCE Class 76 Posterboard BP-24 Wednesday, Author present 15:20 – 16:20 h Beatriz Sánchez, Juan Lladó, Javier Celma and Miguel A. Luzón Benedicto Mechanical Engineering Department, CPS; University of Zaragoza (Spain) NOISE PREDICTION MODEL OF URBAN SQUARES Sound quality of the cities concerns everybody. Noise level constantly increases due to the traffic that produces serious effects. However, in the city there are particular areas with a sound perceived as of positive quality. To evaluate these sound characteristics and to develop a predictive model of positive sound quality, several squares of the city of Zaragoza (Spain) with different environments have been acoustically identified by means of the equivalent sound level, Leq; and the frequency spectra. The mathematical model has been obtained from the statistical analysis of the measurements carried out at several points of the squares and it considers the influence of three sound sources, identified through the frequency spectra: traffic, pedestrians and fountains. This predictive model of positive sound quality will help to plan new open areas as well as to improve the existing ones. 164 Internoise 2001 Abstracts Session B4-2, 14:20 h Paper number 261 (Contributed, Poster) Posterboard BP-25 Wednesday, Author present 15:20 – 16:20 h Yasushi Takano, Toru Inaba and Susumu Sasaki I-INCE Class 63 Mechanical Engineering Research Laboratory, Hitachi, Ltd. 502 Kandatsu, Tsuchiura, Ibaraki, 3000013 Japan SOUND QUALITY EVALUATION METHOD OF AUTOMOBILE ALTERNATORS Sound quality of alternators used in automobiles is discussed. It is known that a rapid change in number of rotation of an alternator sometimes results an inadequate noise from cooling fans fixed to the rotors. This noise is referred to a word pronounced as "Myah". It is found that the primary reason of this "Myah" sound is caused by the change of the frequency band with maximum noise level. Sound quality can be improved by controlling the center frequency of this frequency band within certain range. Session B4-2, 14:20 h Paper number 283 (Contributed, Poster) Posterboard BP-26 Wednesday, Author present 15:20 – 16:20 h Pál Zoltán Bite and Fülöp Augusztinovicz I-INCE Class 61 Budapest University of Technology and Economics, Department of Telecommunications, H –1111, Budapest, Stocek u. 2, Hungary EVALUATION AND NOISE IMPACT OF CAR AUDIO SYSTEMS This paper presents the effects of, and the problems caused by, high-end audio systems used in passenger cars. Two major aspects of the problem were investigated: first, what is the quality of the sound reproduction that can be achieved in the car interior, and how, and to what extent, car audio systems can contribute to the noise pollution of residential areas. Correspondingly, both the interior and the surrounding sound field of the test vehicles was investigated for various operating conditions of the car (including switched off engines), with three different kind of music pieces. In order to assess the environmental impact, the overall insertion loss of the car was measured. Standard by-pass measurements were also performed to demonstrate the difference in sound levels between a standard car audio and a car equipped with a high-end audio system. The behavior of the car as an acoustic cavity was evaluated by means of the Boundary Element Method (BEM) and verified by measurements. For better illustration of the growing problem, we calculated noise maps for a peaceful holiday resort too. The results have shown that currently used high-end audio systems are not really appropriate for good quality sound reproduction, caused both by disadvantageous acoustic characteristics of the cavity and artificially modified transfer characteristics of the electroacoustic system. Significant exterior sound level differences were found between different audio systems, ranging up to 20 dB. Exterior noise predictions have shown that high-end systems can significantly deteriorate the noise environment, especially for quiet places in the night period. Internoise 2001 Abstracts 165 Session B4-2, 14:40 h Paper number 361 (Contributed, Poster) Posterboard BP-27 Wednesday, Author present 15:20 – 16:20 h Jaeheon Kim, Ku Tae Kang and Seung Dong Yeo R&D Division, Hyundai Motor Company, Korea I-INCE Class 11 A STUDY ON MECHANISM OF INTAKE RUMBLE NOISE AND OPTIMIZATION OF INTAKE MANIFOLD The rumble noise is necessarily improved in view of the sound quality and the passenger’s comfort. This article presents the mechanism of intake rumble noise and optimization of the shape of intake manifold. Simple analysis authenticates the initial hypothesis on the mechanism of intake rumble noise, which is also verified by numerical analysis and experiment. Compared to the basic manifold, the modified manifold mitigates half-order harmonics of intake noise, consequently improves the sound quality of interior noise. Session B4-2, 14:40 h Paper number 534 (Invited, Poster) I-INCE Class 63 Posterboard BP-28 Wednesday, Author present 15:20 – 16:20 h Myriam Mzali, Danièle Dubois, Jean-Dominique Polack, Fabien Letourneaux and Franck Poisson SNCF- Direction de la Recherche et de la Technologie, Unité Physique du Système Ferroviaire et Confort (SFC), 45, rue de Londres, 75379 PARIS cedex 08, France MENTAL REPRESENTATION OF AUDITORY COMFORT INSIDE TRAINS: METHODOLOGICAL AND THEORETICAL ISSUES Improve auditory comfort inside trains is an important task for SNCF. The only criteria that can be currently provided to train builders is a threshold level in dBA. This method seems not to be sufficient. Therefore, a study developing an evaluation method of the auditory comfort is being conducted. The approach chosen is based on passengers perception and their mental representation. This takes place in three steps: - First, information had been collected from client survey to highlight what is important in auditory comfort for passengers and define work orientations; - Second, simultaneous psychological and physical measurements have been performed leading to a correlation between this two aspects; Third, listening tests are performed to confirm previous results and to define new acoustic criteria. The work presented in this paper concerns the third part of the study and specifically the validation of hypothesis set out from client survey results. Sound class had been build up according to the cognitive process performed (for example, we have already shown that "announcements" are not perceived in the same way as "background noise"). Listening test of categorisation allows to compare class obtained in real conditions (different sensory stimulation are mixed) and in experimental ones (only auditory stimulation is present). Then, we will conclude about reliability of mental representation of sound sources through different listening conditions. 166 Internoise 2001 Abstracts Session B4-3: Soundscapes Chaired by Peter Lercher and Brigitte Schulte-Fortkamp Session B4-3, 15:00 h Peter Lercher Paper number 264 (Tutorial, Oral) I-INCE Class 66 Institute for Hygiene and Social Medicine, University of Innsbruck, A-6020, Sonnenburgstrasse 16, Innsbruck, Austria CONTEXTUAL AND NON-CONTEXTUAL PERSPECTIVES IN RESEARCH AND MANAGEMENT OF NOISE ANNOYANCE: AN ENVIRONMENTAL HEALTH PERSPECTIVE Several reviews (Job 1988, Fields 1993, Miedema & Vos 1999) have tried to quantify the importance of factors that are known to moderate/modify the reported annoyance response toward main noise sources. Undoubtedly, the use of such a checklist is helpful for the initial assessment of potential moderation. However, the simple transfer of average importance from aggregated data to the actual response of group in a probably different situation and context is not advisable. Additionally, the common dichotomization between acoustical and non-acoustical factors may further contribute to incomplete solutions in the management of noise problems: interactions between sound, person, situation, and context are neglected and this omission may result in wrong conclusions about the true relative importance of the "factor-mix". There is a strong need for re-verification of this aggregated information in the specific socio-cultural, economic, environmental, and situational context. This procedure requires a thorough embedding of theories (Cohen et al 1986, Fidell 1990, Stallen 1999) and methods appropriate for this ambitious task. The methods of psychoacoustics and acoustic ecology, the psychological construct of perceived control, the socio-ecological construct of "personenvironment fit", and the psychosomatic construct of "adaptational capacity" are possible helplines along the path to a better understanding of the structure and the processes that often lead to major deviations in the annoyance response. Examples will support the understanding of this approach. Session B4-3, 15:40 h Paper number 591 (Invited, Oral) Dick Botteldooren, Andy Verkeyn and Peter Lercher Ghent University (INTEC), St-Pietersnieuwstraat 41, B-9000 Gent (Belgium) I-INCE Class 66 HOW CAN CLASSICAL ANALYSIS AND FUZZY MODELING HELP US FIND OUT HOW ROAD NOISE MODIFIES PEOPLE'S REACTION TO RAILWAY NOISE? The reaction of people to railway noise may be influenced by the presence of road noise in various ways. In laboratory research the masking effect of one noise source by another is well studied and proven. Although the underlying concept (loudness psychophysics) is different from the one used in field studies (annoyance), masking may also be observed in these field studies under certain conditions. However, there is more. The sensitivity to noise of inhabitants of regions close to highways may be lower and additional situational and environmental variables can contribute to the difficulty of the problem. A large representative environmental noise data set (N = 2007) gathered in Austria was used as the bases for this research. Classical statistical and fuzzy techniques are applied in parallel to screen for the effects mentioned above. The classical analysis uses multivariate regression techniques to determine a stable set of variables to explain the inherent variance. The fuzzy analysis uses a rule-based model that is constructed to predict noise annoyance. Specific attention is paid to Internoise 2001 Abstracts 167 variables such as road noise level, distance to the railway track and the road, direction of bedroom and living room windows to the sources, and combined variables such as noise sensitivity and expectation. Session B4-3, 16:00 h Paper number 292 (Invited, Oral) Patrick Fourcade and Nicolas Tixier ACROE, INPG, 46 av. Félix Viallet, 38 031 Grenoble Cedex, France I-INCE Class 66 MODELING OF SOUND AMBIENCES: CONTRIBUTION OF THE PHYSICAL MODEL To consider the notion of architectural or urban ambiences, works of Cresson showed the need for an interdisciplinary and in situ approach. The study of sensitive phenomena, like the listening, the touching, the looking, etc or in general way multisensitive perceptions, implies taking into account their physical, social and aesthetic dimensions. This complexity results in seeking tools and concepts for the analysis as well as for the design which are transverse with the various disciplines and the various sensory methods. This approach is based on the three following concepts : 1. the ambiences as a whole of active relations in context ; 2. the ambiences as a dynamic construction; 3. the ambiences as a sensitive and exprimable experience. To model ambiences, means finding a numerical tool which is compatible with these elements of this theory of ambiences. The physical model created and developed by Acroe is a model which aims to the emergence of significant phenomena (sound, visual, tactile and multisensitive). It is known as physics because it uses the rules and the descriptors of physics. It is dynamic by nature. For the application to the sound ambiences, it is generating acousticospatio-temporal patterns, i.e. of patterns accessible to our senses. From a sound effect, "the metabole effect" [a perceptive effect describing the unstable relations between the elements composing a sound ensemble], one will study and one will listen to how this connection between sound ambience and a physical model is made possible. Session B4-3, 16:20 h R. Klaeboe Paper number 289 (Invited, Oral) I-INCE Class 66 Institute of Transport Economics. P.O.Box 6110, Etterstad, N-0602 Oslo, Norway THE POSSIBLE IMPACT OF THE NEIGHBOURHOOD SOUNDSCAPE ON EXPOSURE-EFFECT RELATIONSHIPS The construction of new tunnels and new road systems altering the traffic flows dramatically in existing town areas are always interesting when seen as natural experiments al-lowing the relationships between changes in traffic and noise exposure to be studied along with the associated changes in noise reactions. Often noise exposure is only assessed at people’s dwellings not taking the effect of the neighbourhood soundscape into account. The impact of the homogenity or heterogenity of the neighbourhood soundscape on noise annoyance at the dwelling is assessed on the basis of a beforestudy in the Norwegian town of Drammen. Such heterogenity has in previous cross-sectional studies in Oslo been shown to be of importance for noise annoyance at people’s dwellings. Overall traffic reductions after changes in the traffic system has previously been shown to also increase the homogenity of road traffic noise exposure in the neighbourhood. This hypothesis will be checked in after studies commencing in 2002. 168 Internoise 2001 Abstracts Session B4-3, 16:40 h James D. Foch Paper number 249 (Contributed, Oral) I-INCE Class 68 Foch Associates, 18 Windward Hill, Oakland, CA 94618, USA SOUNDSCAPE CHARACTERIZATION IN PARKS Recently the U.S. National Park Service has recognized natural sound and natural quiet as resources, entitled to the same protection as air quality, water quality, wildlife, archaeological artifacts, and other park resources. A key part of this new policy is characterizing the soundscape in a park. In most parks, the soundscape varies appreciably during each day, and from day to day. Part of this variation is secular, and part is random. By dividing the day into 48 half-hour periods, and pooling the results for each half-hour period for every day of the month, much of the randomness can be tamed. Such analysis has been applied to measurements of one second Leqs for 14 months at Devils Tower National Monument in Wyoming. The results show strong diurnal and monthly variations, but recurring patterns year to year. For July 1996 and July 1997, in particular, bootstrap calculations of confidence intervals have shown that most of the 1997 results (for L90, L50, and L10 at half hour intervals) fall within the 1996 confidence intervals. These calculations indicate that long term measurements, and suitable statistical analysis, can have enduring value. This approach should provide a reproducible, quantitative characterization of the soundscape in parks, equally useful for gauging natural soundscape degradation, natural soundscape restoration, and trial of new policies for managing the natural soundscape. Session B4-3, 17:00 h Kozo Hiramatsu Paper number 181 (Invited, Oral) I-INCE Class 66 Kozo, Mukogawa Women’s University, Nishinomiya 663-8558, Japan SOUNDSCAPEGRAPHY: THE NEED, METHOD AND UTILITY Need: Soundscapegraphy is description of soundscape. It is an alternative term for ethnography. Sonography is defined as the art of soundscape notation including spectrogram and the graphic level recorder, but beyond these it also attempts to register the geographic distribution of sound events. Soundscapegraphy is more than sonography in the sense that soundscape is sonic environment with emphasis on the way it is perceived and understood by the individual, or by a society. Method: Thus soundscapegraphy requires besides sonography the perception and understanding of sonic environment by the individual, or by a society. Sonography is a description of basically objective sonic environment. Soundscapegraphy is a description of objective and subjective sonic environment. Subjective sonic environment is described on the basis of the individual or community response to sounds heard in a certain area obtained by questionnaire survey, interview etc. Life histories of the individuals residing in the area and living the soundscape can be important information for soundscapegraphy. Documents related to the sonic environment will also be beneficially used for soundscapegraphy. Utility: Soundscapegraphy is the description of total sonic environment which environmental assessment presumably requires the present state of sonic environment before altered by development or whatever. Comparison of sonic environments can really be conducted by means of the comparison of total sonic environments. Soundscape museum presents the lived sonic environment through recorded sounds and some other information. Soundscapegraphy should be the one presented in the soundscape museum. Internoise 2001 Abstracts 169 Session B4-3, 17:20 h B. Schulte-Fortkamp Paper number 618 (Invited, Oral) I-INCE Class 66 Acoustics /Physics Oldenburg University D-26111 Oldenburg, Germany ACCEPTING THE SOUNDSCAPE - DOES IT INFLUENCE THE NOISE ANNOYANCE? When judging road-, rail-, and air traffic noise concerning annoyance it is well known that the judgment of one source as well as of combined sources are influenced by acoustical characteristics and subjective meaning. Recently research work by Fields, Guski, Lercher, Job, Schulte-Fortkamp have evaluated those effects convergentely. Since there are some new investigations regarding the relevance of soundscapes it becomes clear that the specific constellation of noise events may force a typical understanding of a defined environment. Maffiolo 98, Klaboe 2000, Lercher 99, Lorenz 99, Job 98, Winkler 98have shown the influence of landscapes, visual parameters, and attitudes regarding the environment. The meaning of soundscapes will be discussed concerning a definition of soundscapes which may help to understand the relevance regarding noise annoyance. POSTERS Session B4-3, 17:40 h Paper number 342 (Contributed, Poster) Posterboard CP-3 Thursday, Author present 9:40 – 10:40 h Nicholas P. Miller I-INCE Class 66 Harris Miller Miller & Hanson Inc., 15 New England Executive Park, Burlington, MA 01803 USA A PROPOSAL FOR ACOUSTIC DATA COLLECTION IN PARKS AND WILDERNESS AREAS The US National Park Service is currently developing a guidance document that describes procedures to document and preserve the existing acoustical environments (soundscapes) and also to manage noise intrusions in the US National Parks. The purpose of the guidance document is to provide technical guidance for the collection, analysis and assessment of data that can be used for preservation of National Park Soundscapes. The document defines terminology, describes measurement procedures, characterizes sources of sound, and recommends sound-level metrics and analysis methods. This paper presents some of the preliminary work on procedures for documenting soundscapes, including sound source characterization, acoustic data sampling strategies and several methodologies for acoustic data collection. Session B4-3, 17:40 h Paper number 524 (Contributed, Poster) Posterboard CP-4 Thursday, Author present 9:40 – 10:40 h Hideo Shibayama I-INCE Class 66 Shibaura Institute of Technology, 3-9-14 Shibaura Minato-ku Tokyo, 108-8548 JAPAN THE EVALUATION OF THE COMPLEXITY BY THE FRACTAL DIMENSION OF THE ENVIRONMENTAL NOISE After advanced economic growth passed, we paid attention to society of taking harmony development from not only the side of the industry and the economy but also the point of the quality of the environment. Exceeding the range of prevention of pollution, we understood importance for the 170 Internoise 2001 Abstracts preservation of the natural environment and the earth environment. Environmental impact assessment law was carried out in June 1998 in Japan. In this law, equivalent noise level was adopted in place of median of the noise level used so far as an amount of foundation evaluation. But, it doesn't have a clause toward the quality of the environment noise. We know that there is unpleasant sound even if the level of the sound is low. Auditory experiments for psychological effects were done from a viewpoint of the complexity of a sound waveform. Fractal dimension was chosen as a measure of the complexity. A difference in fractal dimension can be recognized as a difference in the sound quality. There are many complaints for the environment noise around the roadside in Tokyo which is the Japanese capital and fractal dimension of the sound waveform measured in that area is less than 1.4. Session B5-1: Railway noise Chaired by Michael Dittrich and David Thompson Session B5-1, 9:40 h Paper number 735 (Invited, Oral) Manfred T. Kalivoda, Monika Kudrna and Gerard Presle psiA-Consult ltd., Wiener Gasse 146/3, A-2380 Perchtoldsdorf, Austria RAIL TEST SITES PATERNION - ACOUSTIC AND RAIL VIBRATION MEASUREMENTS I-INCE Class 11 The Austrian Federal Railways have tested three different track systems. One system is the standard Austrian mono-bloc concrete sleeper track, the second system is the noise optimised French bi-bloc concrete sleeper track with Nabla fasteners and the third system consisting in three sub-systems was a newly developed double-H shaped sleeper with quasi-continuously supported rail. Aside the issues of safety, track stability and LCC differences in noise and vibration emission are of major interest. The noise and vibration study has been carried out using the best available measurement methodology provided by the 4th Framework Programme project MetaRail. The track sections were in very good condition, rail roughness was reasonably beneath the proposed ISO limit, noise propagation conditions were the same for all five test sites and a special train with disc and block braked coaches was used. Pass-by noise measurements showed an average 2 dB higher A-weighted emission level for the three double-H sub-systems compared with the Austrian and French track. Third octave analysis of the data gave a clear indication then which part of the track/wheel systems contributes how much to total level. Finally it could be shown that spatial decay alone is not able to describe the noise emission behaviour of a track. Session B5-1, 10:00 h Paper number 132 (Invited, Oral) David J. Thompson and Tianxing Wu I-INCE Class 13 Institute of Sound and Vibration Research, University of Southampton, Highfield, Southampton SO17 1BJ, United Kingdom THE EFFECTS OF NON-LINEARITIES AT THE WHEEL/RAIL INTERFACE ON THE GENERATION OF ROLLING NOISE A railway wheel and track may be considered as linear systems that are joined by a non-linear contact stiffness. The system can be excited either by a vertical relative displacement, such as roughness or discontinuities, or by instabilities in transverse creep forces which lead to squealing noise. Models for rolling noise usually consider a linearised contact spring, which allows the model to be expressed in Internoise 2001 Abstracts 171 the frequency domain. This paper considers the response of a wheel/track system in the time domain to excitation by roughness of different amplitudes. By comparing the response when a linear or a nonlinear contact spring is included, it is concluded that a linear model is a good approximation provided that the r.m.s. roughness amplitude is smaller than about 0.35 times the static deflection of the contact spring under the wheel load. For larger amplitudes of roughness loss of contact can occur which leads to a modified excitation force and consequently sound radiation. Session B5-1, 10:20 h Paper number 439 (Invited, Oral) M.G. Dittrich and M.H.A. Janssens TNO TPD, P.O. Box 155, 2600 AD Delft, The Netherlands I-INCE Class 13 MEASUREMENT PROCEDURES FOR DETERMINING RAILWAY NOISE EMISSION AS INPUT TO CALCULATION SCHEMES An overview of new procedures is presented for measuring railway noise emission data that can be used as input to statutory calculation schemes. Initially developed for the Dutch railway noise calculation scheme, these procedures are generalised so as to be broadly applicable. Two new procedures were developed to enable the assignment of new rolling stock into existing or new train categories: - a simplified method for assigning a rail vehicle to a category and for track access; comprehensive methods for characterisation of vehicle and track. The comprehensive methods take the various noise sources into account: traction noise, rolling noise and aerodynamic noise. Also the main influence factors on rolling noise, rail roughness and track response, and wheel roughness and vehicle response are characterised separately. This results in data with which contributions from the track and from the vehicle can be separated and allows for data exchange between different sites. The procedures are consistent with the prEN ISO 3095 draft standard for exterior noise type testing, but include additional steps to obtain input data for calculation schemes. Session B5-1, 10:40 h Paper number 498 (Invited, Oral) Ulf Orrenius and Roger Enblom I-INCE Class 13 Bombardier Transportation, Center of Competence for Noise and Vibration, SE-721 73 Västerås, Sweden ACOUSTIC OPTIMIZATION OF CAR-BODY STRUCTURES: WEIGHT EFFICIENT SETTING OF SUBSYSTEM REQUIREMENTS One key objective of a modern rail-way coach is to keep the noise generated by rail-wheel contact and other bogie sources out. Accordingly, the car-body design is partly governed by needs for high acoustical transmission loss, and a significant fraction of the car-body weight may be due to such needs. However, control of noise should be accomplished without adding more weight and costs than necessary. The design challenge is therefore to reach the required noise levels within given weight budget at a reasonable cost. In the car-body design process, requirements on interior noise levels are broken down to subsystem level. For each element, e.g. a door or window, transmission loss requirements are stipulated. To arrive at an efficient distribution of requirements in a systematical way, an optimization procedure has been developed. The procedure is based on a standard room acoustic model for airborne sound transmission and an experimentally determined model for the sound pressure distribution around the car-body. Assuming mass-law behavior for the subsystems in combination with a quality function given by the design, an optimal mass distribution is determined. 172 Internoise 2001 Abstracts As a constraint, a minimum sound pressure level difference between the bogie cavity and the coach interior is stipulated. The procedure uses a commercially available optimization tool and can be applied early in the design process to arrive at a cost and weight effective design while taking into account design limitations, such as a given door configuration. A calculation example is given to illustrate the potential of the optimization process. Session B5-1, 11:00 h James T. Nelson Paper number 746 (Invited, Oral) I-INCE Class 21 Wilson, Ihrig & Associates, Inc., 5776 Broadway, Oakland, CA 94618, USA WHEEL SQUEAL NOISE CONTROL WITH WHEEL AND RAIL VIBRATION ABSORBERS The noise reduction effectiveness of wheel and rail vibration absorbers were measured at the New Jersey Transit and Portland Tri-Met in the United States. The tests included treatment of tangent and curved track rail vibration absorbers and treatment of resilient Bochum wheels with wheel vibration absorbers. The tests indicated little reduction of wayside rolling noise exposure and maximum noise level, but singing rail was eliminated, leading to a qualitative improvement in wayside noise. Wheel squeal was not eliminated at curved track, but there was a reduction of the probability and duration of wheel squeal with the wheel vibration absorbers. The rail vibration absorbers significantly modified the rail vibration spectrum at both tangent and curved track. The pinned-pinned mode of vibration was eliminated with the rail vibration absorbers. Wayside one-third octave noise reductions and Fourier spectra are presented. The tests were conducted under Project C-3A of the Transit Cooperative Research Program, and administered by the Transportation Research Board. Session B5-1, 11:20 h M. Jaecker-Cueppers Paper number 411 (Invited, Oral) I-INCE Class 82 Umweltbundesamt, Postfach 330022, 14191 Berlin, Germany RAILWAY NOISE ABATEMENT IN THE EUROPEAN UNION - THE WORKING GROUP RAILWAY NOISE OF THE EUROPEAN COMMISSION With the Green Paper "Future Noise Policy" of November 1996 the European Commission began to step up its activities in the field of European noise abatement policy including the railway sector. The Commission has convened various working groups (WG) on noise reception and emissions to develop proposals for the new European noise policy. The WG Railway Noise started its work in December 1999. It has the following tasks: - In order to increase the competitiveness of the European railways the Commission has enacted and plans to enact Directives on the Interoperability of the European Railway systems (high speed systems: 1996, conventional systems: draft 1999). These Directives will be complemented by technical specifications, among others for the noise emissions of railbound vehicles. For this purpose noise limits and measurement procedures must be developed; - The WG Railway Noise shall evaluate the Action Plan of the international railway association UIC, the Community of European Railways CER and the international association of private car owners UIP to replace the cast-iron brake blocks of freight wagons by composite blocks (emission reduction of about 8 dB(A)); - Furthermore, the WG Railway Noise shall evaluate the principal options for an improved strategy to mitigate railway noise taking into account the economic aspects of noise reduction (noise limits for vehicles, track-related measures, economic instruments). A general overview on the activities of the WG Railway Noise will be presented (with a more detailed report on its work in other presentations in the Railway Noise Session of Internoise 2001). Internoise 2001 Abstracts 173 Session B5-1, 11:40 h Peter Hübner Paper number 665 (Invited, Oral) I-INCE Class 82 SBB BahnUmwelt-Center, Parkterrasse 14, CH 3000 Bern 65, Switzerland THE ACTION PROGRAMME OF UIC, UIP AND CER 'ABATEMENT OF RAILWAY NOISE EMISSIONS ON GOODS TRAINS' Railways are by far the most environmentally friendly means of transportation. Railway noise, especially that resulting from freight trains, poses the only relevant impact on the environment. To ensure the long-term support of rail transportation by decision makers it is necessary to remove this negative impact. The railways have therefore initiated an action programme in the UIC, CER and UIP organisations with the aim of reducing noise from freight traffic. Noise can be reduced by smooth rail surfaces which in turn are achieved either by disc brakes or – at a much lower expense – by replacing cast iron brake blocks with synthetic ones. The action programme is pursuing two major paths: Starting immediately, all new wagons are fitted with K-blocks; this possibility was accepted by the UIC in autumn of 2000. Those wagons already in service will be retrofitted with LL-blocks, which have the same braking characteristics as cast iron blocks, however still require some development. Studies comparing different noise reduction possibilities demonstrate that rolling stock improvement is much less expensive than infrastructure measures. Choosing an optimal mix of noise control measures results in significant financial savings. If the existing fleet is to be retrofitted quickly without reducing the competitiveness of the railways, the money saved by infrastructure should at least in part be used for rolling stock noise measures. Out of the STAIRRS project – which is discussed in detail in other papers at the conference – a tool will become available to determine the optimal mix of noise control measures. The aim of the railways is a significant reduction of noise within the next decade. Session B5-1, 12:00 h Paper number 737 (Invited, Oral) I-INCE Class 82 Ulrik Danneskiold-Samsøe, Uffe Degn, Christopher Maxon, Louise Rebien Villefrance, Friedrich Krüger, Manfred Kalivoda, Bernd Barsikow, Bela Buna, and Marco Masoero Ødegaard & Danneskiold-Samsøe A/S, Titangade 15, 2200 COPENHAGEN, Denmark A STUDY OF EUROPEAN PRIORITIES AND STRATEGIES FOR RAILWAY NOISE ABATEMENT The study consists of a review of current European legislation and documentation of cases, where technical measures against railway noise have been successfully applied. Secondly, the potential for further noise reduction will be studied and a classification of "good practise", "promising new technology", and "promising research results" will be applied. Thirdly, a strategy for future activities of the Commission concerning reduction of rail noise will be proposed. All types of rail vehicles from high-speed trains to trams are included in the study. Furthermore, noise reduction measures applied to the track, bridges and marshalling yards are included. The paper comprises the first public presentation of this very recent project. 174 Internoise 2001 Abstracts Session B5-1, 13:20 h Paper number 267 (Contributed, Oral) Toshiki Kitagawa, Yasuo Zenda, Yufuko Abe and Yukie Ogata I-INCE Class 13 Noise Analysis Group, Railway Technical Research Institute, 2-8-38, Hikari-cho, Kokubunji-shi, Tokyo, Japan SOUND RADIATED BY VIBRATION OF RAILWAY WHEELS Railway noise, such as aerodynamic noise, driving device noise, concrete support structure noise and wheel/rail noise are generated by various parts of railway cars, and noise measured at the track wayside consists of these noises. In order to reduce railway noise effectively, we have developed a method of analyzing railway noise and investigating the contributions of these noises. Using this method we investigated railway noise, and got the results which wheel noise were major noise source in case of cars with one type of wheels. Wheel noise is composed of the sum of two components: tread radiation (radial vibration) and web radiation (axial vibration), and, in previous papers, sound radiation of web vibration was considered to be higher than that of tread vibration. We investigated the character of the wheels by impulse response, and measured the noise around the bogies at running. As the results of tests, it was proved that wheels vibrated and radiated sound intensively at many natural frequencies of both web and tread vibrations. So, in our investigation sound radiation of both vibrations had same contribution of wheel noise. We will report the results of impulse response test and field test and numerical solutions of wheel vibration and sound by using FEM and BEM simulation method, and explain the mechanism of sound radiation by the vibration of wheels. Session B5-1, 13:40 h Paper number 241 (Contributed, Oral) I-INCE Class 31 Seigo Ogata, Kyoji Fujiwara, Toshiki Kitagawa and Kiyoshi Nagakura National Traffic Safety and Environment Laboratory (Ministry of Land, Infrastructure and Transport) 7-42-27, Jindaiji-higashi-machi, Chofu, Tokyo, Japan EXPERIMENTAL EXAMINATION WITH ELEVATED STRUCTURAL MODEL CONCERNING MULTIPLE REFLECTION OF NOISE BARRIER IN NOISE FORECAST OF THE ORDINARY RAILWAYS Noise barriers have been constructed to reduce the railway noise. However, as multiple reflections between noise barrier and car body occurs, we can't accurately estimate the effects of the multiple reflections reflection in predicting method of ordinary railway noise. In this paper we investigate the effects of multiple reflections using a technique of scale model experiment. The experimental results show that noise level varies due to multiple reflections in accordance with the difference of the distance between noise barrier and car body. Then we think out that the railway noise prediction method in due consideration of the difference of the multiple reflections. Session B5-1, 14:00 h Paper number 293 (Contributed, Oral) Andrew E.J. Hardy and Richard R.K. Jones AEA Technology Rail, PO Box 2, London Rd, Derby, DE24 8YB, UK MEASURING THE SOUND POWER PRODUCED BY A MOVING TRAIN I-INCE Class 72 Although assessing the sound power radiated from a source such as a moving train is often the underlying aim of many sound measurements it is usually the sound pressure level that is measured. However, the sound pressure at a given point depends on factors that are independent of the source. Internoise 2001 Abstracts 175 This is why standard conditions for measuring the sound pressure level have been developed. However, with complex noise sources such as moving trains, such standards can easily become so restrictive that it is very difficult to make reliable and consistent measurements. It is estimated that on the UK railway less than 10% of the track miles have levels of rail roughness low enough to allow testing to be carried out in compliance with many standards. To compound this difficulty, once the requirement for an open, level, site is also included it is found that in the UK there is one suitable site available, on average, every 125 miles of track. An investigation has therefore been undertaken to identify the optimum microphone position for measuring the sound pressure level from a passing train. Too close would mean the measurements were in the near field but moving further away would limit the number of possible sites. The investigation resulted in the measurement of the sound power from a moving train and demonstrated that the optimum positions were different from those commonly used in published standards. Session B5-1, 14:20 h Paper number 744 (Contributed, Oral) N.A. Nilsson and H. Malker Acoustic Control AB, Tumstocksvägen 1, 187 66 TÄBY, Sweden THE SADDLE-PROFILE-RAIL - A SQUEAL FREE RAIL DESIGN I-INCE Class 21 By studies of a non-linear lumped parameter model, it was revealed that the squeal vibration level in the wheel could be reduced by reducing the mechanical impedance of the rail at squeal frequencies. The impedance reduction of the rail should be of the extent that the rail impedance is less than or comparable to the wheel impedance at squeal frequencies. The reduction of impedance was achieved by a reduction of the apparent mass at squeal frequencies. In the lumped model the non-linear friction coefficient vs creep velocity is modelled by a non-linear friction element. Each mode of the wheel and the rail is modelled as a damped single degree of freedom system. The crabbing velocity of the bogie in a curve is modelled by a huge mass pulling the system at approximately constant velocity. A saddle like profile that is covering the railhead gives the desired lower apparent mass in the new rail design. The profile is also decoupled dynamically from the railhead by a thin rubber mat. A test section 20-m long in a curve of radius 50-m has been built. Narrow band FFT analysis of the recorded sound pressure time history reveal that the sound pressure level at the first and second squeal frequency is reduced by 30-37 dB compared to the squeal sound pressure levels from the adjacent track sections. This means that the Saddle Profile Rail is totally abolishing the wheel/rail curve squeal. The technique has been developed within the BRITE EURAM project SQUEAL (BRPR-CT97-0477)and is protected by two patent applications. 176 Internoise 2001 Abstracts POSTERS Session B5-1, 14:40 h Paper number 269 (Contributed, Poster) Posterboard BP-29 Wednesday, Author present 15:20 – 16:20 h Frank Rysgaard Mortensen, Per Finne and Lars Deigaard RAIL GRINDING AS A NOISE CONTROL MEASURE I-INCE Class 13 ScanRail Consult, Transport and Environment, Pilestraede 58, DK-1112 Copenhagen K, Denmark Usually, railway noise impact on the environment is reduced by means of either noise barriers or noise insulation of buildings. However, the obvious way of reducing the source emission using rail grinding is very often neglected. It is well-known, that the wheel-rail contact is bound to produce rolling noise as it is virtually impossible to make both parts perfectly smooth. The cost and effectiveness of different strategies of rail grinding is investigated and compared to other noise reduction measures relevant for railways. Session B5-1, 14:40 h Paper number 455 (Contributed, Poster) Posterboard BP-30 Wednesday, Author present 15:20 – 16:20 h Jan Lub and Jan Van den Brink AEA Technology Rail BV, P.O. Box 8125, 3503 RC Utrecht, The Netherlands NOISE CLASSIFICATION OF ROLLING STOCK AND TRACK I-INCE Class 82 Active noise policy is required to regulate on a reasonable time scale the process of reducing railway noise at the source. The reduction is necessary because of the conflicting policies of growing demand of transport on one hand and community noise control on the other. Cost-benefit studies of different measures show that reduction of rolling noise is preferable to measures like sound barriers. Policy focuses on the stimulation of measures on railway systems by using a mix of instruments. The lack of a European system of noise generation limits for rail vehicles makes a system of classification of vehicles as an instrument for legislation necessary, which is aimed at optimisation of the balance between traffic intensity, track capacity and noise production. In WP2 of the 5th Framework project STAIRRS a classification method will be one the main results. In March 2000 a provisional classification scheme was proposed based on the technical features of rolling stock responsible for noise differences. An intermediate classification proposal, for an important part based on information obtained from operators, supply industry and legislators to be collected in a workshop in March 2001, will be delivered in June 2001. The final classification proposal will be ready in 2002. In STAIRRS tools are being developed for a more sophisticated approach, bringing into account the noise contribution of wagon and track separately, closely connected with the trend of separating responsibilities between infrastructure owners and train operators. Data collected during the project will demonstrate the feasibility of the classification method. Internoise 2001 Abstracts 177 Session B5-1, 14:40 h Paper number 745 (Contributed, Poster) Posterboard BP-31 Wednesday, Author present 15:20 – 16:20 h N.A. Nilsson and H. Malker Acoustic Control AB, Tumstocksvägen 1, 187 66 TÄBY, Sweden I-INCE Class 21 STUDYING THE GENERATING MECHANISMS OF WHEEL/RAIL CURVE SQUEAL BY A NONLENEAR LUMPED PARAMETER MODEL With the aim to find ways to treat the rail in order to abolish wheel/rail curve squeal a non-linear lumped parameter model has been developed. In the lumped model the non-linear friction coefficient vs. creep velocity is modelled by a non-linear friction element. Each mode of the wheel and the rail is modelled as a single degree of freedom system. The crabbing velocity of the bogie in a curve is modelled by a huge mass pulling the system at approximately constant velocity. The model suggests that squeal is generated with a frequency of approximately the first wheel mode frequency. The model also suggests that rail lubrication and wheel damping will lower the squeal sound pressure level. Increasing the internal loss factor of a normal rail was found to have no effect on the squeal sound pressure level. This was due to the fact that the impedance of the rail is so much higher compared to the impedance of the wheel that the "wheel just slides around on top of a rail almost at rest". Since the rail is performing almost no movements then almost no energy is consumed by a rail damping treatment. If however the wheel/rail system is changed in a way that brings the rail impedance to be equal to or less compared to the wheel impedance, then squeal can be reduced. This more, with respect to rail treatment, favourable impedance relation could be achieved either by increasing the wheel apparent mass (which is may not the most practical solution of the squeal problem) or reducing the apparent rail mass at squeal frequencies. The method of reducing the apparent rail mass is described in a separate paper at this conference. The lumped parameter modelling technique has been developed within the BRITE EURAM project SQUEAL (BRPR-CT97-0477). Session B5-2: Beamforming arrays Chaired by Marie-Agnes Pallas and Jean-François Piet Session B5-2, 15:00 h Paper number 597 (Invited, Oral) Marie-Agnès Pallas and Joël Lelong INRETS/LTE, 25 avenue F. Mitterrand, case 24, 69675 Bron cedex, France INVESTIGATION OF THE ROLLING NOISE SOURCES ON THE TRAM OF NANTES WITH A MICROPHONE ARRAY I-INCE Class 72 As part of a research program on the acoustic emission of ground transportation, the tram of Nantes has been investigated. In relation with the development of a noise model of the tram, microphone array measurements have been achieved. Rolling noise being the main noise source, the investigations aimed at pointing out noise differences between driving bogies and bearing bogies, as well as the influence of the track construction. The measurement and processing involved an horizontal line array with focussing and dedopplerization. Acoustic cartographies for each pass-by of the tram in thirdoctave bands were drawn. Tram constant speeds spread from 10 to 50 km/h. Some accelerating runs were also measured. Three types of track construction were tested : classical track (concrete bed with absorbing pads between the rails and the sleepers), floating slab on a bridge, rails on ballast. 178 Internoise 2001 Abstracts Measurements of squeal noise in a curve were also achieved. The classical track appears to be the less noisy, with very few contribution of the rail except at higher speed. The measurements on the bridge show mainly the vibration of the whole structure, hiding even the pass-by of the wheels at lower frequencies. For the track on ballast, the contribution of the rail arises clearly and the rail may radiate even higher as the tram is leaving than between bogies. The contribution of each noise source can be estimated. Session B5-2, 15:20 h Paper number 432 (Invited, Oral) I-INCE Class 72 Anders Nordborg, Alexander Martens, Jörg Wedemann and Ludger Willenbrink Deutsche Bahn AG, Völckerstr. 5, DE-80939 München, Germany WHEEL/RAIL NOISE SEPARATION WITH MICROPHONE ARRAY MEASUREMENTS The spiral microphone array developed at Deutsche Bahn AG produces excellent resolution and S/Nratio over a broad frequency range. For the first time, it may be possible to directly separate rail and wheel noise from each other. Possibilities and limitations are discussed and illuminated with measurement results. Session B5-2, 15:40 h Paper number 553 (Invited, Oral) A. Mast, Marinus M. Boone and T.C. Van den Dool I-INCE Class 72 Laboratory of Acoustic Imaging and Sound Control, Department of Applied Physics, Delft University of Technology, The Netherlands AN EXPERIMENTAL MICROPHONE ANTENNA ARRAY FOR MEASURING AIRCRAFT NOISE In recent years, aircraft noise pollution has become a public issue. Governments try to set tight limits for this aircraft noise pollution. In accordance with the Dutch government, a new integrated measurement and calculation system for aircraft noise is being developed in the Netherlands. The immision levels of the aircrafts have to be measured at specific locations and these measurements form the basis for the noise maintenance. An existing NOise MOnitoring System (NOMOS) is based on single omnidirectional microphones. These measurements are strongly influenced by other noise sources like trains, cars, and reflections and by wind turbulence. Applying array technology in measuring aircraft noise improves the SNR significantly. Being able to decompose the noise for their angles of incidence gives the opportunity to measure the noise from specific angles, and therefore omitting unwanted noises from other directions and reducing windnoise. From an earlier investigation, a sparse line array was preferred, consisting of 15 microphones, measuring the octave bands of 125 Hz to 2000 Hz. This frequency range is sufficient to measure the A-weighted immision levels of representative aircrafts. Computer simulations showed good results and for verification this array was built and tested. Because of the size and robustness, very small microphones, also used in hearing aid applications, were used. Outdoor measurements, next to official NOMOS measurement units for comparison, were performed. Temporary results show a SNR improvement of 5 to 10 dB, due to minimization of the influence by unwanted noise sources. The SNR improvement is dependent on the angular spacing of the sources. Further research is performed on improving the angular resolution. Internoise 2001 Abstracts 179 Session B5-2, 16:00 h Paper number 380 (Contributed, Oral) G.L. Oh, C.L. Yap, S.K. Pang DSO National Laboratories, 20 Science Park Drive, Singapore 118230 SOUND LOCALISATION IN THE ATMOSPHERE USING A BIG SENSOR ARRAY I-INCE Class 72 Localisation of sound in the atmosphere using microphone array techniques have been widely addressed in the literatures. Using an array of microphones, one can apply adaptive beam-forming or super-resolution techniques (such as MUSIC) to determine the direction of arrival of multiple signals impinging on array. In this paper, we investigate the performance of sound localisation in the atmosphere using a big microphone array. Two array configurations are investigated. Array I is an under-sampled array. It is a box-shaped array with two end clusters and a kink along the array length. The length of the array is about 26 times of the wavelength of signal of interest spanning several hundred meters long. The array comprises 54 microphones. The horizontal distance between each pair of microphone is about two wavelengths except for the end clusters where the microphones are spaced at distance half wavelength apart. The end clusters of microphones are designed to remove ambiguities that will arise when array is under-sampled. The second array, Array II, is a smaller L-shaped array. Its total aperture length is about 9 times the wavelength of signal of interest. It comprises 26 microphones spaced at distance about half wavelength apart. For localising the multiple signals, we apply one form of the linearly constrained minimum variance adaptive beam-former, namely the constrained least mean-squares algorithm derived by FROST. The technique is applied to the real acoustic data collected using Array I & Array II respectively. In the analysis, we first address the signal coherence and ambiguities issues. Finally we compare the localisation results between Array I & Array II, and show that Array II is preferred for localisation. Session B5-2, 16:20 h Paper number 418 (Contributed, Oral) I-INCE Class 72 K.G. Degen, A. Nordborg, A. Martens, J. Wedemann, L. Willenbrink and M. Bianchi Deutsche Bahn AG, Völckerstr. 5, DE-80939 München, Germany SPIRAL ARRAY MEASUREMENTS OF HIGH-SPEED TRAIN NOISE A high performing microphone array enables localization and quantification of noise from high-speed trains. So far, most reported results suffer from bad resolution and a high degree of measurement noise due to disturbing side lobes. However, the new spiral array, developed at the Deutsche Bahn AG's Research Center, produces results with a good resolution and S/N ratio, demonstrated by high-speed train measurements, including the German ICE 1, ICE 2 and ICE-T trains as well as the Italian ETR 500 train, partially equipped with bogie shrouds and multi-layer damped wheel. The importance of the relevant sound sources are discussed. 180 Internoise 2001 Abstracts Session B5-2, 16:40 h Paper number 512 (Contributed, Oral) I-INCE Class 72 Motoatsu Miwa, Hui Wang, Hideo Shibayama and Toshiya Miyauchi Shibaura Institute of Technology, Tokyo, 108-8548, Japan ESTIMATION OF DIRECTION OF INFRASOUND ARRIVAL BY USING AN ARRAY MICROPHONE This paper describes the spatial smoothing method for estimating a Direction of arrival (DOA) of a sound source in existence of a reflection sound and simulation results. In existence of a reflection sound, it is impossible to estimate the DOA of a sound source and the reflection sound by using only the MUSIC method. However we can estimate the DOA of a sound source and a reflection sound by using both the spatial smoothing method and the MUSIC method. The simulation for estimating the DOA of the sound generated by the sound source and reflected sounds is carried out. The frequency of the sound source is 20Hz, and SN ratio is 10dB. The DOA of the sound source and the reflection sound is 10deg and 70deg respectively. A linear microphone array consists of four microphones and the each spacing is 8.5m. As mentioned above, we obtained the result that cannot estimate the DOA of the sound source and the reflection sound by using MUSIC method. It is possible accurately to estimate the DOA of sound source and reflection sound by the proposed method. Session B5-2, 17:00 h N. Roosnek Paper number 658 (Contributed, Oral) I-INCE Class 72 Roosnek Research & Development, Vlaskamp 82, 2892 AC The Hague, The Netherlands PASSIVE RANGING WITH FOUR MICROPHONES IN A SPATIAL CONFIGURATION For high precision aircraft noise measurements a system has been developed with highly accurate tracking capabilities, based on correlation and nonlinear least square optimization. History and the state equation has been taken into account resulting in an extended Kalman filter optimization procedure. The accuracy is so high that passive ranging is possible for moving and stationary objects, with an error depending on distance, velocity and environmental parameters. The sensitivity towards the environmental parameters will not be treated. Session B5-2, 17:20 h Paper number 254 (Contributed, Oral) Young-Chul Choi and Yang-Hann Kim NOVIC-KAIST, Taejon-shi, 305-701, Korea I-INCE Class 74 NOISE SOURCE IDENTIFICATION IN A REVERBERANT FIELD USING SPHERICAL BEAMFORMING Identification of noise sources, their locations and strengths, has been taken great attention among the noise control engineers because of its practical importance. Useful methods of these kinds can be classified in two kinds; One is what is based on beamforming method, another one is what is utilizing the technique of acoustic holography. The beamforming method has advantage to the other method. That is the method use less microphone. Therefore, it is practically feasible. The method normally assumes that noise sources are located at a free field. However, sound in a room and tunnel consists of that coming directly from the source plus sound reflected or scattered by the walls and by objects in the room. In contrast to exterior sound field, reflections are added to sound field; therefore, the sound strength predicted by the conventional method may give unacceptable errors. The main objective of this paper is to identify noise sources in the reverberant field using spherical beamforming and Internoise 2001 Abstracts 181 spherical MUSIC method. The validity of these methods to identify the location of noise sources has been tested by the experiments which were performed in an acryl duct. POSTERS Session B5-2, 17:40 h Paper number 341 (Contributed, Poster) Posterboard CP-5 Thursday, Author present 9:40 – 10:40 h L. Koop, K. Ehrenfried, A. Dillman and U. Michel TU-Berlin, HF1, Straße des 17. Juni 135, D-10623 Berlin, Germany IN-FLOW MEASUREMENTS WITH MICROPHONE ARRAYS I-INCE Class 72 During the last years, microphone arrays have become a significant measurement tool to locate sound sources. One major advantage of in-flow measurements consists in the absence of acoustic wave propagation through an unsteady, refracting and scattering shear layer. However, in-flow array sensors are subject to noise generated by the turbulent boundary layer on the fairing. Another problem consists in the fact that for downscaled wind tunnel models, higher signal frequencies have to be recorded in order to preserve Strouhal similarity. In this work, theoretical simulations have been performed to systematically investigate the effects of coherent and incoherent noise on array measurements. The signal-to-noise-ratio was calculated for different kinds of algorithms available from literature. In addition microphone positions were optimized in order to get maximum side lobe suppression for a large frequency range. The theoretical simulations show that for optimized microphone positions, robust-adaptive and dynamic-adaptive algorithms can significantly improve the performance of the array. Tests of the optimized algorithms in wind tunnel experiments prove that aerodynamic sound sources can be detected with high accuracy and reliability, although the turbulent boundary layer noise is dominant in the frequency spectrum. Session B5-2, 17:40 h Paper number 422 (Contributed, Poster) Posterboard CP-6 Thursday, Author present 9:40 – 10:40 h Y. Nii, N. Takahashi and H. Matsumiya I-INCE Class 72 National Institute of Advanced Industrial Science and Technology, Namiki 1-2-1, Tsukuba, Ibaraki 305 Japan BROADBAND NOISE SOURCE LOCALIZATION BY LINE MICROPHONE ARRAY POSITIONED NEAR A WIND TURBINE ROTOR BLADE Broadband noise is a dominant noise source of recently developed wind turbine generator systems, and a tool to evaluate the noise source and thus a measure to reduce the noise are strongly expected. A line microphone array was fabricated for the test to localize the noise of a 15 m diameter wind turbine rotor blade. Evaluation method of microphone data for source reconstruction was based primarily on the Acoustic Telescope while the acoustic dipole was assumed as a source model instead of the monopole following the dipole feature of the source. The array was positioned near the blade for measurements aiming to have accurate source distribution along the blade span direction. Prior to the fabrication of the array system and the tests, numerical simulations had been carried out to investigate feasibility of applying such array system to the noise source rotating with fast pass-by speed and to 182 Internoise 2001 Abstracts obtain necessary parameters to be used at the tests. The results of the tests had shown a performance as expected. Session B5-2, 17:40 h Paper number 489 (Contributed, Poster) Posterboard CP-7 Thursday, Author present 9:40 – 10:40 h Yan, Chye Hwang and Oh, Geok Lian DSO National Laboratories, 20 Science Park Drive, Singapore 118230 I-INCE Class 74 TACKLING THE COCKTAIL PARTY PHENOMENON USING A SMALL APERTURE LINEAR ARRAY The ability to focus on a conversation at a distant location in a crowded environment (cocktail party effect) using a small aperture microphone array is an essential step in many audio processing systems. In our setup, we assume that the environment is stationary; locations of noise sources remain fixed and reflection from walls and ceiling are negligible. Linearly constrained minimum variance filter (LCMV) and generalized sidelobe canceller (GSC) are two popular existing beam forming techniques that are suitable for the abovementioned problem. We have conducted two experiments in a busy and noisy cafeteria. In the first experiment, we attempt to focus on a controlled conversation at a location that is about 7.5m away from an array of 6 microphones with an aperture size of 20cm. For the second experiment, an additional identical microphone array is placed at a location that is 7.5m away from the conversation and 1m away from the first array. In this paper, we will first present a procedure that we have developed for determining the appropriate constrain matrix of the GSC. We will assess the effectiveness and computational requirement of GSC and LCMV using the results from the first experiment. Using the second experiment, we will outline a new methodology for exploiting the 2 array setups to obtain better beam focusing. Session B6-1: Resilient elements and materials Chaired by Rajendra Singh Session B6-1, 9:40 h Paper number 171 (Contributed, Oral) B.M. Efimtsov, L.A. Lazarev, S.N. Baranov I-INCE Class 23 Central Aerohydrodynamic Institute (TsAGI), 17 Radio Street, 107005, Moscow, Russia DECAY OF SOUND TRANSMISSION INTO A LAYERED SHELL BY RESONANT ELEMENTS An orthotropic cylindrical shell of finite length with butt-ended walls limiting the acoustic volume is considered. The internal shell surface is covered with layers of crumbly fibrous material and with an inertial panel with resonant elements. The butt-ended walls are assumed to be acoustically soft or acoustically rigid. The external shell surface is subjected to the effect of the sound pressure field, nonuniform in space. On the basis of analysis of the sound energy flow into the shell the effects of sound decay in it, due to resonant elements of different type, are studied. One of the physical phenomenon of sound transmission decay is realized with the use of these elements: phenomenon of soft reflection, phenomenon of rigid reflection, phenomenon of pressure compensation or of volumical velocity. Consideration of the shell with acoustically soft butt-ends permit revealing and explaining the principal regularities in the attenuation of sound transmitted, which are associated with the behavior of elastically acoustic system ‘shell - layered soundinsulating structure with resonant elements - acoustic volume’ and evaluating the relations between the parameters of these subsystems at which the Internoise 2001 Abstracts 183 significant sound transmission decay is revealed and degenerates. These regularities also manifest is themselves in the case of considering the shell with acoustically rigid butt-ends, only in a more veiled form. Session B6-1, 10:00 h Leif Kari Paper number 530 (Contributed, Oral) I-INCE Class 46 MWL/Department of Vehicle Engineering, Kungliga Tekniska Högskolan, 100 44 Stockholm, Sweden DYNAMIC STIFFNESS OF RUBBER ISOLATORS - RADIUS AND LENGTH DEPENDENCE Embarked upon a waveguide solution, the dynamic stiffness radius and length dependence for cylindrical vibration isolators is resolved. It constitutes a major practical application; in particular, predicting the audible dynamic stiffness alteration due to the geometric shift of a given isolator. The rubber is modeled as nearly incompressible, with the deviatoric viscoelasticity based on a fractional order derivative model. The main advantage of the viscoelastic model is the minimum parameter number required to successfully model the material properties over a broad structure borne sound frequency domain. In general, the results show a rapidly growing stiffness magnitude as the isolator shortens or widens, while being slightly disordered, displaying a possible magnitude or phase peak splitting/merging, occasionally with a pitchfork phase curve 'bifurcation'. The driving point 'bifurcation' however, disappears as the resonances and antiresonances are required to alternate. A simple scaling law fails to satisfactorily model the stiffness variation due to a single length or radius shift, while successfully predicting the results of an isolator shape invariant shift. The small deviations arise from a disregarded material property shift. Session B6-1, 10:20 h Paper number 552 (Contributed, Oral) I-INCE Class 47 Fuxing Yin, Susumu Takmori, Yoshiaki Ohsawa, Akira Sato and Kohji Kawahara Materials Engineering Laboratory, National Institute for Materials Science, Tsukuba 305-0047, Japan THE DAMPING BEHAVIOR AND PRACTICAL APPLICATIONS OF THE M2052 DAMPING ALLOY The M2052 damping alloy in the nominal composition of Mn-20Cu-5Ni-2Fe was developed with a superior damping capacity and strength balance. The damping behavior of the alloy under different influencing parameters, such as temperature, strain-amplitude and frequency, was characterized in different measuring methods. It was found that the damping capacity of the alloy reached a higher level only below a critical temperature and was highly strain-amplitude sensitive. The damping mechanism of the alloy was discussed according to microstructural characterization. Some examples for practical application were also described in this paper. 184 Internoise 2001 Abstracts Session B6-1, 10:40 h Paper number 476 (Contributed, Oral) G. Jansens, W. Lauriks, J.F. Allard and G. Vermeir I-INCE Class 72 Laboratorium voor Akoestiek en Thermische Fysica, Katholieke Universiteit Leuven, Celestijnenlaan 200 D, B-3001 Heverlee, Belgium MEASUREMENT OF THE DYNAMIC SHEAR MODULUS ON POROELASTIC FOAMS The prediction of the acoustical properties of multilayered systems, including poroelastic layers using the full Biot theory is limited by the absence of material data. One of the parameters that is difficult to measure is the dynamic rigidity of the porous frame. Current measurement methods are limited to the low audible frequency range (typically below 400 Hz) and require special shapes of the sample (cube, cylindrical rod or very thin samples). A new method for measuring the dynamic shear modulus of the frame of poroelastic foams in the medium and high audible frequency range (1 to 4 kHz) has been developed. This method is based on the measurement of the velocity of a Rayleigh-type surface wave, which is known to be related to the shear velocity. The latter is directly related to the shear modulus. Technically, the experiment is very simple: the surface wave on the air-air saturated porous layer interface is directly excited through mechanical excitation of the frame, and the time-of-flight of a sine burst is measured with an accelerometer. One of the advantages is that there is no requirement whatsoever about the shape of the sample under investigation. Furthermore this new technique has the advantage that it is possible to detect possible anisotropy in the material. Results of measurements and simulations using the full Biot model will be shown, as well as comparisons with other methods. POSTERS Session B6-1, 11:00 h Paper number 90 (Contributed, Poster) Posterboard BP-32 Wednesday, Author present 13:20 – 14:20 h Rongying Shen, Zhiyong Zhang and Qiang Wang I-INCE Class 37 National Laboratory of Vib., Shock and Noise, Institute of Marine Equipment Research, 10 Heng Shan Road, Shanghai, 200031, China EXPERIMENTAL INVESTIGATION OF THE VIBRATIONAL POWER TRANSMISSION IN FLUIDFILLED PIPING SYSTEMS INVOLVING FLEXIBLE HOSE Flexible rubber hoses are widely used in fluid-filled pipe system not only for the purpose of the misalignment accommodation, but also for the vibration and noise reduction. However, the mechanism and effect of the rubber hose on the vibrational power transmission through pipe system has not been studied adequately yet. In this paper the vibrational power transmission in fluid-filled pipe system involving a flexible rubber hose is studied in detail. The test arrangement consists of the components of the straight pipe and the hose. The experimental method of measuring the axial quasilongitudinal, bending vibrational power flow of the pipe structure is developed. The effect of the accelerometer input error on the decomposing result of wave amplitude is analyzed. It is found that for a certain pipe system vibrating in axial or bending direction and in the range of frequency there is a optimal measuring field arrangement which makes the effect of the input error be least. The crosssensitivity of the accelerometer is explored successfully and the experimental feedback method is suggested. The experiment result gives a guideline on the vibrational power transmission between the Internoise 2001 Abstracts 185 pipe and the hose and is useful for the designation and usage of the flexible rubber hose in actual piping systems. Session B6-1, 11:00 h Paper number 161 (Contributed, Poster) Posterboard BP-33 Wednesday, Author present 13:20 – 14:20 h L. Bujnoch, L. Lapcik Jr., P. Augustin and A. Pistek I-INCE Class 47 Dep. of Physics and Material Engineering, T. G. Masaryka 275, 762 72 Zlin, Czech Republic MEASUREMENT OF DYNAMIC STIFFNESS OF RECYCLED RUBBER BASED RAILWAY TRACK MATS ACCORDING TO DB-TL 918.71 Measurement of dynamic stiffness were performed on three samples of material (based on recycled rubber) used as a mats for railway track applications. Measurements were performed at three static preloads, 0.03, 0.06 and 0.1 MPa, in the frequency range 10-100 Hz. A steady increase in dynamics stiffness with increasing frequency was observed in the sample studies, ranging from 0.1 to 0.23 N.mm-3. At the same time, the measured dynamic stiffness shifted to higher absolute values with increasing static preload. Session B6-1, 11:00 h Paper number 736 (Contributed, Poster) Posterboard BP-34 Wednesday, Author present 13:20 – 14:20 h Petr Bris, Stanislav Vasut and Ludek Bujnoch I-INCE Class 47 The Tomas Bata University in Zlin, Faculty of Technology, Dep. of Physics and Material Engineering, Nam. T. G. Masaryka 275, 762 72 Zlin, Czech Republic DECREASE OF NOISE ORIGINATED FROM BALL IMPACT ON THE PLAYING FIELD SURFACE The contribution presents the influence of rubber composite as the playing field surface upon the decrease of noise caused by the ball impact. It is known that concrete and wooden playing field surfaces radiate at the ball impact the rather great noise which is more or less understood both by players and visitors as the natural "part" of the ball game. In case the grounds are located close to the house-building and above all close to schools or directly in their buildings, as a rule, that noise is often disturbing in these spaces. The work compares mutually the measuring results of noise resulted by ball impact on the concrete surface of the indoor stadium and on the surface formed by the rubber based composite and on the partly covered surface by the composite square of size 1x1 m, too. The experiment was executed for the basketball ball impact from the height of 2 m. For the measuring of sound pressure level, the noise analyzer B - K 2146 was used. The results showed the significant influence of rubber surface upon noise decrease in the indoor stadium, at the same time the lowest values of sound pressure were reached at the whole covering of the indoor stadium surface by the rubber composite. 186 Internoise 2001 Abstracts Session B6-2: Shipboard noise control Chaired by Christ de Jong Session B6-2, 11:20 h Paper number 347 (Invited, Oral) Raymond Fischer and Leo Boroditsky Noise Control Engineering, Inc., 799 Middlesex Tnpk., Billerica, MA 01821, USA CONTROL OF DIESEL INDUCED SHIPBOARD NOISE BY THE USE OF ISOLATION MOUNTS I-INCE Class 54 Vibration isolation mounting is critical to shipboard noise control, particularly on medium and high speed diesel drive vessels. This paper assesses methods used to design and select isolation mounts, to predict mount performance, and to ensure correct installation. Isolation mounting of propulsion diesels and gensets will be considered for both single and double stage mount systems. The importance of airborne flanking of the mount system itself and the importance of "secondary" (or airborne coupled) structureborne transmission paths will be explained. Predicted and measured mount transmission losses are examined along with the expected airborne and structureborne noise associated with isolation mounted diesel systems. Examples of correctly and incorrectly installed isolation mount systems are presented. Session B6-2, 11:40 h Yutaro Yokokura Paper number 193 (Invited, Oral) I-INCE Class 54 Industrial Machine & Plant Development Center, IHI Co., Ltd., 1, Shin-Nakahara-cho, Isogo-ku, Yokohama, 235-8501, Japan DEVELOPMENT OF SHIP'S CABIN NOISE PREDICTION METHOD APPLICABLE ON EARLY DESIGN STAGE For predicting ship's cabin noise, Statistical Energy Analysis has been applied as theoretical method. But SEA method cannot be calculated without detailed specification of structure, then it is not used effectively on initial design stage. Also it requires a great deal of work for the structure modelling. On the other hand, empirical or semi-empirical method based on measured data are effective on that stage. However, ship structure or specification has been changed with the times, it is necessary to modify and refine parameters in order to maintain and improve the accuracy. So, the data-bank of the noise and vibration data of ship's main engine, diesel generator and so on which are main sources of cabin noise was developed, which makes it easy to collect and adjust these data and possible to statistical analysis. And also, in order to find transmission characteristics of structure born noise of latest ships, we measured then on board. At the same time, we estimate then using SEA for the ships. Comparing these result, we defined TL parameters of structure born noise. And also we derived the insertion loss of interior wall, and the contribution of each wall, floor, ceiling from measured data and analysis which are necessary to find cabin noise. Using these result, we developed the practical prediction method applicable on initial design stage. Internoise 2001 Abstracts 187 Session B6-2, 12:00 h Paper number 208 (Invited, Oral) Ruud Vermeulen and Christ De Jong TNO TPD, P.O. Box 155, 2600 AD Delft, The Netherlands HYBRID MODELLING OF MACHINE FOUNDATIONS I-INCE Class 54 Machines are often resiliently mounted to prevent the transmission of vibrations to their foundation structure. To obtain a good vibration isolation it is important to have a sufficiently large mobility mismatch between the machine footing and the resilient mount and between the resilient mount and the foundation structure. This paper focuses on the modelling of the foundation mobility. The foundation is divided into two parts: a global receiver construction and a local seating structure. The input mobility of the foundation is calculated by coupling the transfer mobilities of the seating with the driving point mobilities of the receiver. The receiver usually has sufficient modal overlap so that it can be modelled as an infinite plate. Local seating structures with simple geometries can be approximated by basic constructions like beams, plates or boxes, for which analytical expressions are available. When the seating becomes more ‘exotic’, the transfer mobilities of the seating can be calculated using an FEM code. The mobilities predicted with this hybrid numerical-analytical method show a very good agreement with the measurements for a box-like seating on a plate. Also the results for a more realistic machinery foundation on a ship’s double bottom structure yield a good agreement. Session B6-2, 13:20 h Paper number 441 (Invited, Oral) Christian Cabos, Carsten Worms and Jürgen Jokat Germanischer Lloyd, P.O.B. 111606, 20416 Hamburg, Germany I-INCE Class 54 APPLICATION OF AN ENERGY FINITE ELEMENT METHOD TO THE PREDICTION OF STRUCTURE BORNE SOUND PROPAGATION IN SHIPS An energy finite element method has been developed at Germanischer Lloyd which is capable of predicting the flow of structure borne sound energy in complex ship structures. Experiences from the re-use of existing finite element models (built up for strength and vibration computations) for this method are reported and post-processing capabilities are shown. An overview is given about the verification of the method with full scale measurements. Session B6-2, 13:40 h Paper number 549 (Invited, Oral) A. De Bruijn and C.A.F. De Jong Binnensingel 43a, 7411 PL Deventer, The Netherlands I-INCE Class 54 UNDERWATER RADIATION OF A PLATE - COVERED WITH A COMPLIANT LAYER - BY A POINT FORCE The radiation underwater from a plate – excited by a single force – has been the subject of an extensive range of literature. A remarkable outcome was that for thin plates the sound radiation appears rather independent of the plate mass. Thicker plates at higher frequencies show more interesting features, since the excitation can be performed by either soft or quasi-rigid indenters, resulting into different radiation characteristics. Another interesting feature is the structure, in which the plate consists of more layers, either compliant or stiff, mounted onto a steel plate. The sound radiation can be greatly modified by the inclusion of layers. A third aspect is the nearby field due to a force excitation. This analysis can be useful considering the reciprocal excitation by a point source 188 Internoise 2001 Abstracts close to a plate structure, so important for the understanding of propeller-hull interaction. In the presentation an overview will given on the various theoretical aspects and its numerical implications, especially in view of shipboard noise control. Session B6-2, 14:00 h Paper number 252 (Contributed, Oral) Hanshin Seol, Byungsuk Jung and Soogab Lee I-INCE Class 54 Underwater Acoustics Research Center (UARC), School of Mechanical and Aerospace Engineering, Seoul National University, Seoul, 151-742, Korea NUMERICAL ANALYSIS OF NON-CAVITATING UNDERWATER PROPELLER NOISE Noise generated by a propeller is critical to underwater detection, and often related to the survivability of the vessels especially for the military purpose. Compared with the extensive amount of literature concerning cavitation noise of propellers, the non-cavitation noise of propellers from underwater propellers has hardly attempted to assess so far. The non-cavitation noise of underwater propeller is considered numerically in this study. The main purpose is to analyze non-cavitation noise from underwater propellers in various operating conditions with different configurations. Noise prediction is performed using time-domain Acoustic Analogy, Boundary Integral Equation Method (BIEM), and Computational Aeroacoustics (CAA). We use a time-domain acoustic analogy governed by the Ffowcs Williams-Hawkings equation. Blade surfaces are divided into rectangular panel sources, radiating noise at different retarded times. Hence, propeller configurations as well as various conditions can better considered with this method. A boundary integral equation method for the prediction of marine ducted propeller non-cavitation noise is also considered. With the BEM method sound deflection and scattering effect on the duct can be considered. The governing equations are based on the assumptions that all acoustic processes are linear. A scattering approach is applied in which the acoustic pressure field is split into known incident and unknown scattered components. Finally, BPF tonal noise is analyzed by Computational Aeroacoustics. We use axis-symmetric linearized Euler equations. The Euler equations in generalized curvilinear coordinate are solved by the DRP finite difference scheme. Optimized Adams-Basheford scheme is used for time integral method. Through this study, we can analyze dominant noise source of marine propeller and provide a basis for proper noise control strategies. Session B6-2, 14:20 h Stefan Weyna Paper number 705 (Contributed, Oral) I-INCE Class 54 Technical University of Szczecin, Faculty of Marine Technology, Al. Piastow 41, 71-065 Szczecin, Poland THE EXPERIMENTAL STUDY OF FLOW ENERGY RADIATED BY A SHIP PARTITIONS One of the major difficulties of acoustic measurements in ship spaces is the complex wave distribution in the acoustic field inside the crew accommodation. The field distribution is affected by both the small dimensions and shapes of the spaces, and the specific sources of noise influencing the acoustic climate of ship living and working quarters. Generally speaking, a large contribution to the total noise of the ship's interior spaces is made by structure-borne noise generated by vibrating cabin partitions with 'acoustics leaks' (door and windows). However, in many sections of the ship, airborne noise may be encountered which gets into the ship's interior. The acoustic processes occur in a wide frequency range and can cause operational harm in technological and health terms. In the context of the Internoise 2001 Abstracts 189 technological aspect of the problem, our interest is focused on vibration effects resulting from the type of construction and technique of ship installation fixing. The research aims at explaining the mechanism of transference of structureborne sound into the surroundings using a graphical presentation of the distribution of space vector of intensity (or power) in the acoustic field created inside cabins in the vicinity of accommodation ship partitions and even in full 3D space. The solution of the problem is vital for the protection of accommodations against noise and creating suitable acoustic climate in enclosures. The paper presents an application of the sound intensity (SI) technique to assess the effects of vibrations on a ship's mechanism and it's influences on the noise energy radiation by ship partitions. Numerous examples illustrate the application of the SI measurement for practical problems at the noise abatement on ships. Results of research are shown in the form of a two and three-dimensional flow map of active intensity vectors, graphically interpreted for the heterogeneous cabin partitions. The ability of conducting tests for the near-field area, with external noises paves the way to the development of tests in real conditions. The wave properties described in the field of source, working in their natural environment allow one to analyze the field qualitatively as well as quantitatively, i.e. to evaluate its energetic distribution and visualize the wave distribution in the tested area. One of the main advantages of marine vibroacoustic control by means of sound intensity analysis is that is enables one to identify the airborne and structure-borne energy transmitted through the different parts of the cabin partition, and to find the flanking transmission paths and the vibroacoustics bridges as an acoustics weak area on the considered partitions. This acoustic hod points and flanking transmission paths in description of energy transportation in real structures is not possible to find with classical measurement method (acoustical pressure) and take very important role in engineering problems of noise abatement in small enclosure. The measurement technique as well as the way of graphical presentation can enrich the knowledge of the mechanism of acoustic energy flux through the real ships partitions. POSTERS Session B6-2, 14:40 h Paper number 144 (Contributed, Poster) I-INCE Class 54 Posterboard BP-35 Wednesday, Author present 15:20 – 16:20 h Young-Ho Park, Suk-Yoon Hong, Ho-Won Lee and Hyun-Gwon Kil Department of Naval Architecture and Ocean Engineering, Seoul National University VIBRATION POWER FLOW ANALYSIS OF SHIP STRUCTURES USING SEA PARAMETERS Power Flow Analysis (PFA) has developed by alternative method overcoming the difficulties of SEA for analysis of structural/acoustic systems in the medium to high frequency ranges recently. This paper proposes the new analytical method of vibration in the medium to high frequency ranges using PFA and SEA concepts. The core part of this method is the applications of coupling loss factor (CLF) in SEA instead of power transmission, reflection coefficients in boundary condition in PFA. This method shows very promising results compared to classical PFA for the various damping loss factors and wide ranges of frequencies. Besides this paper presents the applicable method in Power Flow Finite Element Method (PFFEM) by forming the new joint element matrix with CLF to analyze the various plate structures in shape. The analytical results of ship structures show a good agreement with those of PFFEM using classical coefficients. 190 Internoise 2001 Abstracts Session B6-2, 14:40 h Paper number 275 (Contributed, Poster) Posterboard BP-36 Wednesday, Author present 15:20 – 16:20 h Byoung-Ho Yoo and Dong-Hae Kim A STUDY ON NOISE REDUCTION FOR A CHEMICAL CARRIER I-INCE Class 54 Hyundai Maritime Research Institute, Hyundai Heavy Ind. Co., Ltd, Ulsan, Republic Of Korea Generally, because the size of chemical tanker is relatively small compared to other commercial ships, deckhouse is located closer to the aft end of the ship. Therefore, it is difficult to reduce the noise levels of the cabins induced by the main engine and propeller. The background of this paper is high noise level experienced on the LPG/NH3 carrier during the sea trials. The noise levels onboard exceeded the maximum noise limits in the specification. Although several countermeasures were applied to the vessel, it was not enough to satisfy the specification in some cabins where floating floor was installed by the prediction results at the design stage. This paper presents the applied noise control measures and the measurement results. The measurements of the noise levels, propeller pressure pulses, vibration level and the sound intensity were carried out in order to determine the main noise sources, transmission paths and radiating surfaces of the main problem areas. The propeller and some main engine components dominated the noise in the aft part of the accommodation. The main contribution to the noise level was found to be the higher order harmonics of the propeller blade passing frequency in the range of 50-80Hz. From these studies, the noise control procedure for the LPG carrier was established from the early design stage to the delivery. Session B6-3: Building acoustics Chaired by A. Harold Marshall, Gerrit Vermeir Session B6-3, 15:00 h Paper number 258 (Contributed, Oral) Victor Desarnaulds, António P.O. Carvalho and Gilbert Monay E.P.F. Lausanne (EPFL), 1015 Lausanne, Switzerland THE EFFECT OF OCCUPANCY IN THE SPEECH INTELLIGIBILITY IN CHURCHES I-INCE Class 25 The speech intelligibility by the use of objective parameters is usually evaluated in unoccupied rooms due to practical considerations. However, under normal conditions, the room occupancy can increase or decrease the values of speech intelligibility by the effect of the new sound absorption present in the room or by the change in S/N ratio. Measurements were done in four churches with and without occupancy (mean volume of 6314 m3 and mean occupancy of 155 persons). The results show that occupancy induces a mean increase for the speech intelligibility of dSTI (STI_occup. - STI_unoccup.) = 0.050 with the use of a public address system (PA) and 0.027 without a PA. This increase is caused mainly by the reduction of the room reverberation time, due to the increased sound absorption and by the reduction of the additional RT induced by the PA (using sound speakers with strong directivity oriented to the congregation area)). The variations in the speech intelligibility values with occupancy could be predicted in churches (mean error dSTI< 0.005) by theoretical and empirical formulations (DSTI = 0.175*ln(RT_occup./RT_unocc.) - 0.007 with PA, and 0.0304*ln(RT_occup./RT_unocc.) + 0.018 without PA (R2 = 0.94)). The decrease in the S/N ratio (due to the background sound level augmentation by the presence of people and by sound level decrease with the distance in the congregation area) or poor directivity and orientation of the sound sources could reduce the speech Internoise 2001 Abstracts 191 intelligibility gain in particular cases. Other effect induced by the congregation, such as the variations of humidity and temperature, are generally negligible for the speech intelligibility. Session B6-3, 15:20 h Paper number 537 (Contributed, Oral) L. Nijs, D. Van Berlo and M. Van der Voorden I-INCE Class 25 Delft University of Technology, Faculty of Architecture, Building Physics Group, Berlageweg 1, 2628 BG Delft, The Netherlands ARCHITECTURAL GUIDELINES TO IMPROVE THE ACOUSTICAL QUALITY IN ROOMS FOR MENTALLY CHALLENGED PEOPLE A research project is going on to improve (a.o.) the "acoustical quality" of rooms for mentally handicapped people, where acoustical quality is defined in terms of (early) reverberation time, SPL, S/N-ratio, C50, U50 and STI. A separate paper about the technical aspects of measurements in existing situations will be presented at ICA 2001. Here, our purpose is to give some design rules for architects as derived from computer model calculations. Different requirements are found between "normal" living rooms and rooms for mentally handicapped. - Mentally handicapped people often suffer from hearing defects as well. For instance, a reverberation time as low as 0.4 s is preferred; - For cleaning reasons rooms for mentally handicapped are scantily furnished; - For the same reason the total absorbing surface is kept low; - More people live within rooms for mentally handicapped, which leads to "multi-source" situations with extra disturbance. S/N-ratios as high as 20 dB are required. Only the total amount of absorbing surface counts in Sabine's law, but computer model calculations show a reasonable effect of proper positioning. It is, for instance, useless to add an absorbing floor to an absorbing ceiling. To decrease the reverberation time absorbing materials on the ceiling are most commonly used. However, to meet the requirements for the reverberation time, extra absorbing surfaces are required on vertical surfaces. S/N-requirements of 20 dB are, in general, not found in living rooms. Shielding elements of considerable size may help. Also non-shoe-box shaped (L-shaped, U-shaped, Z-shaped) rooms show better values. Session B6-3, 15:40 h Paper number 437 (Contributed, Oral) Jong-Hwa Lee and Jeong-Guon Ih Korea Advanced Institute Science and Technology, Taejon, Korea I-INCE Class 33 A PREDICTION METHOD FOR THE SOUND TRANSMISSION LOSS OF FINITE DOUBLE PARTITIONS AT LOW FREQUENCIES The statistical energy analysis (SEA) seems to be applicable satisfactorily to the prediction of sound transmission loss through double partitions. Although the use of SEA technique is valid for the frequency bands where fairly high modal densities are guaranteed, the low frequency performance for practical applications cannot be evaluated successfully due to the inherent restriction of SEA. In this paper, a method for calculating the low-frequency sound transmission loss of finite double partitions is presented. The method is based on the classical modal expansion method, assuming that the cavity depth is quite small and the boundary around the cavity between partitions is acoustically rigid. It is shown that the sound transmission at low-frequency range is dependent on the partition area, which is very similar to the case of single partitions. Comparisons between calculated and measured results show good agreements, in particular below the mass-spring-mass resonance frequency, fo. 192 Internoise 2001 Abstracts Session B6-3, 16:00 h Paper number 580 (Contributed, Oral) Tetsuya Sakuma and Takuya Oshima I-INCE Class 33 Institute of Environmental Studies, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan 113-0033 APPLICATION OF A VIBRO-ACOUSTIC METHOD TO PREDICTION OF SOUND INSULATION PERFORMANCE OF BUILDING ELEMENTS A numerical method of vibro-acoustic coupling analysis is applied to predict sound insulation performance of building members, such as walls and windows. In order to evaluate field incidence transmission loss, a three-dimensional sound transmission model is assumed where a partition is mounted in a infinite rigid baffle with plane wave incidence of discretized directions. As for the partition, simple vibration models of membrane and of elastic plate are tested to confirm the validity of the method, and in addition, a model of membrane with additional weights is tested to clarify the insulation characteristics of the unique material. The vibration field is discretized by the FEM for each model, and the sound field is discretized by the BEM based on the boundary integral equation with respect to the sound pressure difference between the two surfaces of the partition. Both the discretized systems are coupled with considering the permeability of the partition. Numerical results for membranes and for plates are validated in terms of the mass law, the natural modes, and the coincidence effect and so on. As for membranes with additional weights, high insulation performance at low frequencies is demonstrated with detailed results of a parametric study on weight geometry. Session B6-3, 16:20 h Paper number 89 (Contributed, Oral) M.D.C. Magalhaes and N.S. Ferguson University of Southampton, Southampton, UK LOW FREQUENCY AIRBORNE SOUND TRANSMISSION I-INCE Class 51 This study focuses on the improvement in the understanding of airborne noise transmission phenomenon in buildings, at low frequencies, under conditions of different architectural configurations. The effects of low-frequency noise have been of a particular concern because many kinds of structures have been inefficient in attenuating low-frequency noise, compared to other frequency components. Likewise the Sound Reduction Index is normally measured under diffuse field conditions and these calculations show that, at low frequencies, this is not the case. An acousticstructural coupled room-plate-room model, which allows variations in room size and shape, and partition properties to be considered, has been formulated and solved. Results of a parametric study show the effect of several configurations for a flexible partition (different geometric size and position in the common wall) and connected rooms (size and absorbing walls) on the Sound Reduction Index by averaging over 1/3 octave bands. Finally, a general discussion, based on the findings of the results obtained, is presented. Internoise 2001 Abstracts 193 Session B6-3, 16:40 h Paper number 240 (Contributed, Oral) Zerhan Karabiber and Emine Çelik Yildiz Technical University, Arch. Fac. 80750, Istanbul, Turkey A SURVEY ON THE AUDITORY COMFORT AT A LIBRARY I-INCE Class 51 A research project has been designed to evaluate interactively the different aspects affecting human comfort in an architectural space. One of the basic parts of the research project is the evaluation of the auditory comfort. A university library has been selected as the subject of the research. The library building which shows unusual settlement and design peculiarities, is an interesting case from acoustical view point as well, because of the closeness to a heavy traffic road and architectural properties. This study deals with the investigation and the evaluation of the library building from auditory comfort. In the acoustical evaluation of the library, noise control and room acoustics criteria have been considered taking into account location, design, detailing and implementation characteristics of the building. Measurements have been repeated throughout different seasons of the year to demonstrate the effects of differences that may occur at outdoor noise. A questionnaire survey developed for the purpose of general evaluation of the building was carried out as a part of the research. The study clearly showed that the library building needs an important renovation in order to improve users comfort. On the other hand, the degree of the influence of different aspects effecting human comfort in a building still is a field of discussion. Each building have special requirements originating basically from its function, design and location to achieve comfort conditions. This paper will present the results of work done to show the acoustics performance of the library and an attempt to realise a procedure to evaluate a building as a whole. Session B6-3, 17:00 h Valtteri Hongisto Paper number 303 (Contributed, Oral) I-INCE Class 51 Institute of Occupational Health, Lemminkäisenkatu 14-18 B, FIN-20520 Turku, Finland FLANKING TRANSMISSION THROUGH A FLOATING FLOOR COVERING Floating floor coverings on concrete slabs are very common in modern buildings. It is shown in this experimental field study that such a floor structure can lead to strong flanking transmission between adjacent dwellings comprizing identical floor structures. The flanking is strongest at the resonance frequency of the mass-spring system formed by the floor covering panel, the concrete slab and the cavity foam in between. In this case study, the reduction in Rw was 9 dB because of flanking. Means to reduce strong flanking at the resonance are presented. It is probable that similar resonance transmission is very general because floor coverings used in this case study are very usual. However, more similar research is needed. This is important to avoid systematic under-dimensioning of structures in buildings. Present models for flanking transmission do not take double panel resonances into account. Thus, the modelling methods via double structures should be developed. 194 Internoise 2001 Abstracts Session B6-3, 17:20 h Paper number 750 (Contributed, Oral) F. Augusztinovicz, A.B. Nagy and I. Dombi I-INCE Class 51 Budapest University of Technology and Economics, Dept. Telecommunications, H-1111 Budapest, Sztoczek u. 2, Hungary NUMERICAL PREDICTION OF INTERIOR NOISE LEVELS IN BUILDINGS, GENERATED BY GROUND VIBRATIONS The design of sound insulation of buildings is relatively straightforward if the dominant noise component is airborne. Standards and calculation methods are well established for the dimensioning of buildings to ensure appropriate airborne sound transmission loss. Nevertheless, the problem is much more complex if the building is essentially excited by structure-borne noise such as ground vibrations arising from nearby road traffic and/or railway tracks. Ground vibrations are usually of low frequency, in which case the generally applied assumption of diffuse sound field is not justified and usual design methods are inadequate. The aim of the paper is to explore the applicability of numerical methods, used in other areas of noise control with success, for the analysis and design of noise control in buildings. The structural vibrations of the building can be analysed by means of standard finite element methods, the interior noise can be calculated from a boundary element approach. The method, known as the (weekly or strongly) coupled FE/BE method, is routinely used in various industries, provided that the excitation of the structure is known from measurements or obvious assumptions, mostly in terms of input force. This is however not feasible for buildings, where the input is not accessible for direct measurements. The presented FE/BE method, adopted for building analyses is based on a trial calculation of the structure by using hypothetical force distribution. The obtained building vibration is compared to real-life measurements, the model validated, updated and re-scaled. The obtained surface vibration of the building is used as input for a subsequent BE calculation. Testing and validation of the method on the example of a recent construction work in Budapest is in progress. Session B6-3, 17:40 h Paper number 91 (Contributed, Oral) Sander J. Van Wijngaarden and Jan A. Verhage TNO Human Factors, PO Box 23, 3769 ZG Soesterberg, The Netherlands I-INCE Class 61 THE INFLUENCE OF FAN AND TRAFFIC NOISE ON SPEECH INTELLIGIBILITY IN DUTCH TRAFFIC TUNNELS Traffic tunnels are generally harsh acoustic environments. Materials used to cover walls, ceilings and road surface are not primarily selected on the basis of their acoustic qualities, but rather according to various safety-related criteria. As a result, long reverberation times are observed, which have an adverse effect on speech intelligibility. Nevertheless, in case of an emergency a public address system is used to reach the public inside a tunnel. In absence of noise, speech intelligibility when using such systems usually complies with a generally accepted minimum standard (Speech Transmission Index STI>0.35). Two sources of noise reduce STI-values to unacceptably low values: noise due to moving traffic and noise produced by high-power fans (necessary to expel smoke from tunnels in case of fire). A computational procedure was developed, based on ray-tracing techniques, to predict the Speech Transmission Index. The necessary input parameters are geometric and acoustic absorption characteristics of the tunnel, the specifications of the public address system and the emission details of the noise sources. The resulting speech intelligibility predictions are found to be very useful in the Internoise 2001 Abstracts 195 design stage of the tunnel (and of the PA-system). This is illustrated by means of predictions for the newly built Benelux tunnel. By systematically varying several design parameters, and computing the effects on speech intelligibility in various sections of the tunnel, a cost-effective compromise between different parts of the tunnel design (fans, PA-system) is found. POSTERS Session B6-3, 18:00 h Paper number 29 (Contributed, Poster) Posterboard CP-8 Thursday, Author present 9:40 – 10:40 h Marc Rehfeld Sekurit Saint Gobain Glass CDI, France DEVELOPMENT OF A THIN LIGHT DOUBLE GLAZING WITH IMPROVED ACOUSTIC INSULATION PROPERTIES I-INCE Class 33 The most usual way to do acoustic rehabilitation of facades is changing the existing windows. One of the main problems encountered is the weight of the new windows, specially in old buildings without lifts. Up to now, the only solutions were the use of double glazings comprising at least one thick glass, as 10+4 or 10+6 with 6 to 16 mm space. A new type of double glazing has been developed, using special laminate and thin glass. An important rule concerning interest of laminates in double glazings is that the thickness of the laminate must be smaller or equal to the thickness of the monolithic glass. This has been put into evidence by modelisation, and will be explained. So, the only way to solve the problem was to decrease the thickness of the laminate, that is to use laminate with very thin glass. The final product is 22-1+4. 22-1 means 2x2 mm glass panes assembled with special acoustic interlayer. The acoustic performance of this product is equal to usual acoustic 10+4, and it fits in the same windows and the use is the same that standard thermal 4+4. Results in different conditions, explanations of the principles and description of the components will be given. Some additional specific problems have been studied, as maximum dimensions and conditions of use, because mechanical properties and wind resistance of thin glass are smaller than those of ordinary 4 mm glass. Charts will be presented. Session B6-3, 18:00 h Paper number 108 (Contributed, Poster) Posterboard CP-9 Thursday, Author present 9:40 – 10:40 h M. Asselineau Peutz & Associates acoustical consultants, 34 rue de Paradis, F75010 Paris, France I-INCE Class 33 CONCRETE VERSUS GYPSUMBOARD: TWO TECHNIQUES FOR A SAME OBJECTIVE The construction of large public venues such as cinemas and performance halls has been done for a long time using concrete walls. However, due to a quicker construction process and greater flexibility, gypsumboard walls are now frequently used for such venues, together with lighter structural elements. Following a short reminder of the usual acoustical objectives, and their translation in terms of wall construction, the concrete construction scheme is compared to the lighter gypsumboard construction. While this comparison is mainly centered on the acoustical performances, some side aspects (such as cost, speed, and safety) are also considered. One of the main conclusions is that while gypsumboard 196 Internoise 2001 Abstracts construction can help complete a project faster than concrete construction, it requires from the workers a degree of skill that is not often to be found with the average contractors. Session B6-3, 18:00 h Paper number 284 (Contributed, Poster) Posterboard CP-10 Thursday, Author present 9:40 – 10:40 h Xuetao Zhang SOUND INSULATION OF MULTI-LEAF HOMOGENEOUS PANELS AND DOORS I-INCE Class 33 SP Swedish National Testing and Research Institute, Box 857, S-501 15 Borås, Sweden For sound insulation problems of multi-leaf panels the most distinct difficulty is to determine the coincidence frequency. For the problems there are two different interactions concerned. Shear interactions will affect the total bending stiffness. Elastic interactions in the direction of the normal will transmit vibrations. The total bending stiffness of the structures should include the effect of the shear interactions. In turn, the coincidence frequency of the structures becomes a function of frequency. This characteristic is one of key points for suitably predicting TL values of the structures. Sound insulation of single plates, steel 1 mm and 2 mm, chipboard 12 mm and 19.5 mm, of geometry (Lx, Ly)=(1 m, 2m), different combinations of them and, some samples of real doors have systematically been investigated under laboratory conditions. A theoretical procedure for evaluating TL of the structures has been proposed. The model predictions can fit the measured data quite well. Sewell’s and Cremer’s formulae for TL of single plates, Fahy’s double wall model for infinite homogeneous plates, the scheme for introducing boundary conditions and the formula for the total bending stiffness of the structures are the basic parts of the model. Session B6-3, 18:20 h Paper number 395 (Contributed, Poster) Posterboard CP-11 Thursday, Author present 9:40 – 10:40 h C. Kalkman and J.H.N. Buijs Adviesbureau Peutz & Associes, Postbus 66, 6585 ZH MOOK, The Netherlands I-INCE Class 51 NOISE LEVELS IN APARTMENT BLOCKS CAUSED BY LIFTS: WHAT CAN BE DONE IN ORDER TO REDUCE COMPLAINTS In the current building practice often lifts are used, of which the engine is mounted inside the lift's shaft. This frequently leads to situations in which the lift engine is mounted on a wall, directly behind a livingroom or bedroom. It is usually only a matter of time before situations like this lead to complaints. Because of the new generation of lifts described before and because of a lower complaint threshold, noise levels in apartment blocks caused by lifts generate more and more complaints from the inhabitants of the apartments. Based on a significant number of measurements the conclusion must be that very many lifts do not comply with the current (Dutch) Building Code. The measurement also reveal however that even if the characteristic noise levels, caused by a lift, comply with the Building Code, often complaints still occur. In the majority of the investigated lifts the measured noise levels in the apartments are structure borne sound levels. Because of operational and safety reasons the possibilities of reducing noise levels by use of vibration isolators are limited. Therefore a number of manufacturers have recently developed low noise lift engines, which have also been measured. The specific low noise features of these new engines as well as the results of several measurements will be discussed in this paper. Furthermore some design conditions will be presented which, in the design Internoise 2001 Abstracts 197 stage of an apartment block project, may help reduce the possible number of complaints in the completed state. Session B6-3, 18:20 h Paper number 433 (Contributed, Poster) Posterboard CP-12 Thursday, Author present 9:40 – 10:40 h Herbert Muellner and Ilja Plotizin I-INCE Class 51 State Institute for testing and research in Heat and Sound Technology, Vienna, Austria HOW SECONDARY" DETAILS OF THE CONSTRUCTION AND WORKMANSHIP EFFECT THE AIRBORNE SOUND INSULATION OF PLASTERBOARD WALLS" Considerable differences between the values of airborne sound insulation determined by measurements (according ISO 140-3) in the laboratory (according 140-1) in former years and measurements carried out in buildings (according ISO 140-4) recently initiated extensive studies to find out what in fact is causing those differences. The examinations had to find out the range of effects on the sound insulation in dependence of the variation of the ‘secondary’ wall-components like the metal-profile-studs, especially their shape, the kind of screwing, the sealing of the joint between plasterboard wall and massive flanking wall or massive ceiling, etc. The measurements showed that the results partly deviated considerably. Hardly noticeable changes of the c-stud’s shape caused different results. Not to neglect the distance and the position of the screws. The way of completion of the joints between test-specimen and aperture-border is also a limiting quantity which determines the results eminently. Even whether acryl or silicon as sealing-material is chosen to seal the joints between test-specimen and aperture-border shows influence to the curve of the sound reduction index in dependence of the frequency, especially if the effect of aging is considered. However, repeated measurements on the identical test specimens yielded a very good repeatability value. This fact confirmed the evidence that test facility, test equipment and test method can not cause those remarkable differences. The presumption was suggested that the deviations must also be essentially caused by workmanship and not only by the changes of the quality of the used components. Nevertheless the examinations showed that apart from the test-specimen’s limiting quantities the effect of the test facilities and the influence of the test method should also be taken into consideration especially in the low frequency range and in the upper frequency range. Due to a kind of flanking transmission of the bare aperture-border in the upper frequency range the result can be affected if the wall yields high values of the sound reduction index in that frequency range. This influence depends on the material of the aperture-border, as former examinations clearly showed. Which necessary action should be taken according to these findings? Should the standard values be adapted or should the workmanship be improved by good skilled workers and by new instructions which take into account the changed conditions? These questions have to be discussed very carefully. 198 Internoise 2001 Abstracts Session B7-1: Aircraft noise sources Chaired by Harry Brouwer Session B7-1, 9:40 h Paper number 82 (Contributed, Oral) S.N. Baranov and L.S. Kuravsky I-INCE Class 13 Problem Laboratory of Mathematical Modeling attached to the Computer Center of Russian Academy of Sciences, c/o “Rusavia”, 6 Leningradskoye Shosse, 125299 Moscow, Russia CHARACTERISTICS OF ACOUSTIC LOAD: THREE METHODS OF INDIRECT ESTIMATION Information concerning parameters of pressure fluctuations over aircraft panels is very important in designing and testing modern aircraft. However their traditional acoustic measurements are frequently failed as aggressive high-level environment is frequently out of the range of available transducers. That is why the method giving the opportunity to avoid direct measurements of pressure fluctuations and estimating their parameters via the information obtained with cheaper transducers would be very valuable. Under consideration are the approaches that predict load characteristics using the data measured with strain-gauges or accelerometers. Presented are three different ways of solution: (1) confirmatory factor analysis based on the examination of covariance matrix structure, (2) the approach based on the estimation of optimal frequency response matrices, (3) selection of neural networks. Typical solution consists of two stages. On the first stage(identification), parameters of a structure model are selected to obtain the best correspondence between expected and observed structure response for a set of given test loads. On the second stage (prediction), these parameters are fixed to get predicted characteristics of an unknown load. Covariance matrices (method 1), matrix spectral characteristics (method 2) and power spectral densities (method 3) are used to represent structure responses. Selected are the load parameters that optimize: a maximum likelihood fit function (method 1), matrix norm of differences for a set of pairs ‘input-output’ (method 2), or neural network training error (method 3). In case of software of equivalent efficiency, the approach based on analysis of covariance matrices is more suitable when total frequency-independent characteristics are computed. Estimation of optimal frequency response matrices is more preferable to study frequency-dependent spectral characteristics. However the best prediction was achieved with the aid of radial basis function neural networks. Session B7-1, 10:00 h Paper number 212 (Contributed, Oral) V.V. Koudriavtsev and A.V. Safronov TSNIIMASH, 4, Pionerskaya str., Korolev, Moscow region, Russia, 141070 I-INCE Class 13 NOISE GENERATION AT SUPERSONIC JET INTERACTION WITH INCLINED DEFLECTOR An acoustic environment generated by main engine jets acoustic radiation and jet interaction with launch pad is one of principal sources of structural vibration and internal noise during launch vehicle lift-off. Launch vehicle engine jets interaction with a launch pad in general is a very complicated gas dynamics problem and there is no reliable theoretical technique for its calculation. Much more it concerns theoretical calculations of acoustic field generated by such interaction. The paper deals with the analysis and generalization of experimental data on unsteady surface pressures and outer broadband acoustic field generated at supersonic jet impingement on a inclined surface - typical case of engine jet interaction with launch pad. It is shown that at relatively small distances between nozzle Internoise 2001 Abstracts 199 exit and deflector the region of direct jet impingement (region of strong interaction) is a main source of very intensive acoustic radiation upstream. Based on test data analysis empirical dependencies for intensity, spectrum and directivity of this source is derived. At least four independent regions of noise generation for the system "jet + inclined deflector" exist: region of undisturbed jet between nozzle exit and deflector, strong interaction region (region of direct jet impingement), jet spreading over deflector surface and the deflector as a reflector of acoustic radiation. The contribution of noise generation regions into the integral acoustic field produced by a system "jet + deflector" varies with a distance between nozzle exit and deflector. The semi-empirical technique for predicting a broadband acoustic field generated during such interaction, which uses a superposition of contributions from different independent noise generation regions, is discussed. Variation of acoustic power of the system "jet + deflector", with a distance between nozzle exit and deflector is analyzed. Session B7-1, 10:20 h Paper number 343 (Contributed, Oral) L. Koop, K. Ehrenfried, A. Dillmann and U. Michel TU-Berlin, HF1, Straße des 17. Juni 135, D-10623 Berlin, Germany AIRFRAME-NOISE REDUCTION BY BLOWING I-INCE Class 13 One dominant source of airframe noise is situated at the trailing-edge flap. At the side edge of the flap, a shear layer detaches and rolls up. Oscillation of this edge vortex structure has been proposed by various authors as one possible mechanism of sound production, which is frequently referred to as flap-side-edge-noise in literature. The objective of the present study is to reduce this noise by blowing a jet of air into the vortical structure. The basic idea behind this concept is to displace or destroy the vortex structure, reducing the surface pressure fluctuations and thus the amplitude of the radiated sound. In order to investigate the underlying physical mechanisms in detail, particle image velocimetry (PIV) and sound pressure measurements were performed on an swept constant chord halfmodel. Flap- and slat angles, angles of incidence and blowing rates were varied within the experiments. The PIV measurements yield a rather complicated vortical structure, which can be almost completely dispersed by blowing, thus resulting in a significant reduction of the sound pressure level. Session B7-1, 10:40 h Paper number 401 (Contributed, Oral) Yoshikuni Umeda, Ryuji Ishii and Yoshida Honmachi I-INCE Class 13 Dept. of Aero- & Astronautics, Graduate school of Eng., Kyoto Univ., Yoshida Honmachi, Sakyo-ku, Kyoto, 606-8501, Japan ON THE SOUND SOURCE OF SCREECH TONE IN THE HELICAL OSCILLATION MODE RADIATED FROM CIRCULAR SUPERSONIC JET It is well known that the supersonic jets radiate very strong sound waves called screech tone. Powell showed that this tone is generated by a feedback loop consisting of the downstream-convecting coherent vortical structure around the jet and the upstream-propagating sound waves in the ambient. In our previous investigation, four oscillation modes of screech tones radiated from circular jets were established by the flow visualization technique and the acoustical phase measurements by using two microphones. The result showed that axisymmetric, axisymmetric, lateral, helical, and lateral oscillation modes appeared with increasing the pressure ratio of the jets. In the present experiment, the generation mechanism of the screech tone in the helical oscillation mode was investigated by using schlieren instantaneous photographs. The experimental result showed that the screech tone in the 200 Internoise 2001 Abstracts helical mode was generated by a moving Mach cone with moving point sound source at supersonic speed around a circular orbit in a plane perpendicular to the jet axis. Session B7-1, 11:00 h Paper number 528 (Contributed, Oral) L. Stechow and M. Herr I-INCE Class 13 German Aerospace Center(DLR), Institute of Design Aerodynamics, Lilienthalplatz 7, 38108 Braunschweig, Germany EFFECT OF LANDING FLAP EDGE BLOWING ON SIDE-EDGE-NOISE RADIATION Noise originating from flow over the airframe of large commercial aircraft has gained importance due to the success in high bypass engine noise reduction. Among other sources of airframe noise the edges of lifting surfaces - landing flaps in particular - constitute potential source areas as a result of strong vortices being formed and shed from such side-edges. Aiming at a basic understanding of the related noise generating mechanisms extensive measurements were performed to determine the vortex steady and unsteady aerodynamic characteristics and their correlation with the ensuing farfield noise signatures. Velocity field measurements were performed to localize the flap side-edge vortex, while acoustic farfield data were determined both by means of an acoustic mirror and employing the microphone-array technique, based on the algorithm of acoustic nearfield holography to in addition calculate the acoustic energy flow in the very acoustic near field (the source area). This data was used to establish the correlation between the local unsteady aerodynamics and the radiated acoustic energy and thus describe the relevant noise source mechanism. As one measure to alter the vortex/flap surface interaction process, the effects on both vortex characteristics and radiated side-edge noise was investigated by injection of air through the side-edge plane. Side-edge blowing tends to destabilize the vortex flow. These experiments were performed on a generic wing-flap configuration in the Aeroacoustic Windtunnel Braunschweig. Session B7-1, 11:20 h Paper number 141 (Contributed, Oral) I-INCE Class 76 B.M. Efimtsov, A.Ya. Zverev, A.O. Andersson and S.V. Kravchenko Central Aerohydrodynamics Institute, Russia PREDICTION OF AIRPLANE INTERIOR NOISE DUE TO FLOW OVER SMALL STEPS – PART 1. RESONANT SOUND TRANSMISSION Flow over small steps on the exterior of an airplane causes aerodynamic pressure fluctuations that are up to 30 dB greater than those under a turbulent boundary layer on a surface without pressure gradients. Although of small extent, regions with such excitation may contribute substantially to the interior noise in airplanes, particularly if well-radiating structures like windows are excited. The aim of this investigation is the prediction of interior noise of airplanes resulting from flow over such steps, common on airplane exteriors. This requires derivation of the relations governing the vibration and acoustic radiation of thin-walled structures in the case of excitation by such pressure-fluctuation fields. Existing interior-noise prediction procedures in the framework of statistical-energy analysis are modified to reflect the peculiarities of these fields, which differ from those under a turbulent boundary layer in terms of correlation scales and non-uniformity scale. This paper is limited to the transmission due to resonant excitation of the structure. A later paper will describe the non-resonant contribution to the radiated field. This work is the result of cooperation between TsAGI and Boeing. Internoise 2001 Abstracts 201 Session B7-2: Detection and identification of noise sources Chaired by Yuri I. Bobrovnitskii Session B7-2, 11:40 h Paper number 726 (Contributed, Oral) J.H. Wang and W.C. Chen IDENTIFICATION OF THE NOISE SOURCES IN HIGH SPEED CD-ROM I-INCE Class 12 Dept. of Power Mechanical Engineering, National Tsing Hua University, Hsinchu, TAIWAN The rotating speed of CD-ROM becomes faster and faster in the past years. Nowadays, the typical rotating speed of a CD-ROM in the market is higher than 9000rpm. It is relative easy to design a spindle motor with a rotating speed higher than 9000rpm. However, if a disk is attached to the spindle motor, the high rotating speed may cause serious noise and vibration problems, especially the noise problem. The main reasons for these problems are due to the fact that many components used in the CD-ROM are not re-designed for the high speed, for instance, the rotating disk, the loading mechanism generally are the same without any modification. In this work, the noise sources of a high speed CD-ROM are identified by experimental method. The results show that the noise is mainly due to aerodynamic sources, partially due to the structure borne sources. The main aerodynamic sources include: (1) the cavity noise due to the interaction of the high speed airflow with the small holes in the loading magazine; (2) the turbulent noise due to the obstruction in the airflow, like some ribs in the loading magazine; (3) the turbulent noise caused by the clamper; (4) trailing edge noise of the rotating disk. Except the trailing edge noise, experimental results show that the above mentioned noise sources can be significantly reduced by some geometry modifications of the CD-ROM components. POSTERS Session B7-2, 12:00 h Paper number 122 (Contributed, Poster) Posterboard BP-37 Wednesday, Author present 13:20 – 14:20 h Juan Lladó, Beatriz Sánchez, Javier Errasti Mechanical Engineering Department, CPS; University of Zaragoza (Spain) NOISE SOURCES PRODUCED BY THE PUMP OF A WASHING-MACHINE I-INCE Class 11 In this work, the existing noise sources when the pump of a washing-machine is operating are experimentally investigated by measurement of the sound pressure level in a semi-anechoic chamber according to the international standard ISO 3744. Using a real-time frequency analyser, the measurements were carried out at the steady state working conditions of the pump, during the airwater phase. To identify the sources and transmission paths of the noise, different modifications of the washing-machine cabinet have been tested. It has been determined that the main contribution to the total noise measured when the pump is working is due to the structural vibration of the cabinet produced through its joining with the pump. 202 Internoise 2001 Abstracts Session B7-2, 12:00 h Paper number 378 (Contributed, Poster) I-INCE Class 72 Posterboard BP-38 Wednesday, Author present 13:20 – 14:20 h Yasumasa Kimura, Toshimitsu Tanaka, Hajime Nakashima and Hirofumi Watanabe Kobe Steel Ltd. Takatsukadai 1-5-5, Nishi-ku, Kobe Hyogo, 651-2271, Japan DEVELOPMENT OF MEASURING METHOD FOR ACOUSTIC CHARACTERISTICS BY USING 3MICROPHONE METHOD An experiment method to analyze the transfer matrix of engine intake systems is proposed in this report .By using our 3-mirophones Method, the four terminal constants of each sub-system was obtained. And those of the whole system were superimposed by combination of sub-systems. Based on experimental results, these method have been proved to be sufficiently reasonable for analyzing acoustic characteristics of engine intake systems. Session B7-2, 12:00 h Paper number 496 (Contributed, Poster) Posterboard BP-39 Wednesday, Author present 13:20 – 14:20 h Wieslaw Wszolek, Ryszard Tadeusiewicz and Andrzej Chyla University of Mining and Metallurgy, Al.Mickiewicza 30, 30-059 KRAKÓW, Poland I-INCE Class 74 RECOGNITION OF SELECTED HELICOPTER TYPES BASED ON TEH GENERATED ACOUSTIC SIGNAL WITH APPLICATION OF ARTIFICIAL INTELLIGENCE METHODS The tasks related to the analysis and recognition of acoustic signals characterizing selected technological objects (vehicles), which should be recognized according to these signals, exhibit specific features, because for such tasks it is very difficult to find the proper rules of signal analysis or the proper algorithm of its recognition. The noise of the flying helicopter is characterized by a short duration period, and time dependent signal reception conditions (spectrum shift resulting from the Doppler effect, sometimes high level of external disturbances etc.) The acoustic and spectral structure of the flying helicopter can be very different, therefore the typical methods of recognition can be very unreliable. At the same time the attempt to reveal the signal's characteristic components (by application of the artificial intelligence techniques), contained in appropriate combinations of its spectral and temporal features, can be useful for its reliable identification by understanding the signal structure, not by its mechanical allocation to a certain group. In the paper the results are shown, which have been obtained by using the technique mentioned above and by employing the learning neural networks in the tasks of recognition of selected helicopter types based on their generated acoustic signal. Internoise 2001 Abstracts 203 Session B7-3: Aircraft noise modeling and monitoring Chaired by Greg Fleming and Michel Vallet Session B7-3, 13:20 h Paper number 273 (Invited, Oral) R. Buetikofer and G. Thomann EMPA, Acoustics and Noise Abatement, CH-8600 Duebendorf, Switzerland VALIDATION OF FLULA, A TIME-STEP MODEL FOR AIRCRAFT NOISE CALCULATIONS I-INCE Class 76 Noise zones around airports impose restrictions on land use. Therefore the noise calculation program used to define those noise zones should be as accurate as feasible. The only way to verify that a program produces correct results is to compare the results with measurements. The aircraft noise simulation program FLULA was developed in Switzerland in the past twenty years, including measurements of the direction dependent sound emission characteristics of 30 aircraft types. A recent international round robin test pointed out some discrepancies in the results calculated with various programs. Therefore a validation of our program became urgent to consolidate previous good agreement with measurements. The presentation first introduces the simulation model FLULA and its unique data base of "in-flight" directivity characteristics. The main section deals with the validation based on measurements around airport Zurich. There were about 450 departures of various aircraft measured at 5 locations up to 10 km away from start of roll. Using the radar information of the flight path, the level-time curves were calculated with FLULA for each measured flight. Average deviations of the SEL between calculations and measurements were smaller than 1 dB for well-controlled situations. The main parameters of variation were (i) the individual power setting of the aircraft and (ii) the accurate modelling of climb profiles. Session B7-3, 13:40 h Gregg G. Fleming Paper number 290 (Invited, Oral) I-INCE Class 76 U.S. Department of Transportation, Research and Special Programs Administration LATERAL ATTENUATION OF AIRCRAFT SOUND LEVELS OVER AN ACOUSTICALLY HARD WATER SURFACE: LOGAN AIRPORT STUDY The National Aeronautics and Space Administration (NASA), Langley Research Center (LaRC), sponsored the Acoustics Facility at the United States Department of Transportation’s John A. Volpe National Transportation Systems Center (Volpe Center) and the Massachusetts Institute of Technology (MIT) to conduct a noise measurement study at Logan International Airport in Boston, Massachusetts, during the summer of 1999 to examine the applicability of currently available mathematical models of lateral attenuation. Analysis of the data collected revealed that lateral attenuation is a function of aircraft geometry. Lateral attenuation for aircraft with tail-mounted engines was found to agree with the published literature, as well as that included in existing aircraft noise models. Lateral attenuation for aircraft with wing-mounted engines was found to be less than documented in the literature. This lower lateral attenuation for aircraft with wing-mounted engines results in a general under-prediction of side-line noise in the existing noise models. 204 Internoise 2001 Abstracts Session B7-3, 14:00 h John M. Gulding Paper number 465 (Invited, Oral) I-INCE Class 76 Federal Aviation Administration, Office of Environment and Energy, AEE-100, 800 Independence Avenue, S.W. Washington, DC 20591, USA SUMMARY OF RESEARCH AND DEVELOPMENT INITIATIVES FOR THE FEDERAL AVIATION ADMINISTRATION'S INTEGRATED NOISE MODEL The Integrated Noise Model (INM) serves as an implementation of the noise modeling guidance specified in SAE-AIR-1845.It also includes noise-modeling routines for areas currently not covered in SAE-1845.The INM is distributed to over 600 organizations worldwide and is used extensively to assist policy makers in assessing the cost and benefit of competing aviation noise mitigation strategies. Specialized research versions have been distributed to groups investigating updates to SAE-AIR-1845. This paper describes INM development goals and priorities including the research and implementation plans for the issues being investigated by Aerospatiale, Boeing, NASA, VOLPE, MIT and the UK/CAA. It will address differences between the existing guidance documents (SAE-AIR-1845, ECAC 29, etc) and the actual model implementation and will discuss efforts to harmonize guidance documents and noise models. The paper will review initiatives taken to promote noise model data development with aircraft operators and manufactures including the commercial transport, business jet, general aviation and helicopter industry. The paper will address issues of maintaining a fielded noise model system. This will include an overview of the worldwide users of the noise model, their noise metric and processing requirements and will briefly address policy metrics that are being reviewed for possible inclusion into noise models such as INM. It will also include the development and distribution of specialized research versions and co-research initiatives with other institutions. Session B7-3, 14:20 h Kenneth J. Plotkin Paper number 503 (Invited, Oral) I-INCE Class 76 Wyle Laboratories, 2001 Jefferson Davis Highway, Suite 701, Arlington, VA 22202, USA THE ROLE OF AIRCRAFT NOISE SIMULATION MODELS Traditional aircraft noise models are integrated, that is, they work from a database of sound exposure level (SEL) from complete flyovers and are directed toward time-integrated noise metrics. A more elaborate modeling approach is simulation, where the actual time history of noise is computed at each receptor. During the 1990s, simulation model NMSIM was developed as a development adjunct to the integrated model NOISEMAP, for the purpose of testing new algorithms for propagation of sound over terrain. NMSIM was subsequently used for examining the effect of wind and temperature gradients on noise footprints, and for study of inner details (such as noise fractions) of integrated models. While simulation models are considered by some to be overkill for environmental analysis, it was soon realized that they did have practical application. The new Rotorcraft Noise Model (RNM) is a simulation model because the complex nature of helicopter noise does not lend itself to simplification into integrated form. Even for relatively simple fixed wing aircraft noise analysis, NMSIM has proven itself to be useful for detailed demonstration of the noise associated with individual events. It has also been of direct use in environmental analysis of special situations, such as audibility of aircraft in noise sensitive areas. This paper reviews the structure of simulation models, and their application for both support of integrated models and direct noise analysis. Internoise 2001 Abstracts 205 Session B7-3, 14:40 h Paper number 574 (Invited, Oral) Ben H. Sharp, Fabio Grandi and John B. Ollerhead Wyle Laboratories, 2001 Jefferson Davis Hwy., Suite 701, Arlington, VA 22202, USA I-INCE Class 76 THE APPLICATION OF AIRCRAFT EQUIVALENCY IN THE GENERATION OF AIRCRAFT NOISE CONTOURS The development of the Model for Assessing Global Exposure to the Noise of Transport Aircraft (MAGENTA), used on behalf of ICAO to estimate the global benefits of aircraft noise mitigation measures, requires calculation of noise contours for 185 airports. To complete the evaluation of just one scenario for these airports using the Integrated Noise Model (INM) would take many days using even the fastest desk computers. In order to evaluate a large number of scenarios it was necessary to reduce this run time significantly. As run time is roughly proportional to the number of different aircraft types modeled, the approach taken was to reduce this to an absolute minimum. Each aircraft is represented by a mixture of those four 'surrogate' aircraft adjusted to match the shape and size of its noise footprint as generated by the INM. By this means, the run time on a high-speed PC to produce noise contours for a single large airport with multiple flight tracks can be reduced from several hours down to a few minutes, while maintaining accuracy in shape and area. This paper will present the details and accuracy of the equivalency procedure and describe situations where it can be applied. Session B7-3, 15:00 h Paper number 731 (Invited, Oral) Darren P. Rhodes and John B. Ollerhead I-INCE Class 76 Environmental Research and Consultancy Department, Civil Aviation Authority, CAA House, 45-59 Kingsway, London WC2B 6TE, UK AIRCRAFT NOISE MODEL VALIDATION Aircraft noise models are now used widely by a large number of countries. Models vary in detail depending upon the needs of their users. Increasingly, models are becoming more sophisticated as additional needs develop. Traditionally noise models have been used to depict long-term historical noise exposure. However, they are often used to assess noise mitigation options and other ‘what if’ type scenarios. These applications have meant that a number of noise models have been subject to close scrutiny by interested parties, spurring further validation and model development studies. This paper describes the methods and procedures used in validation of the UK Aircraft Noise Contour Model, ANCON version 2. This model conforms to international guidance defined in ICAO Circular 205 and SAE Aerospace Information Report 1845. Aircraft noise prediction in the vicinity of an airport is a complex process dependent on a large number of independent and dependent factors. The paper describes how data is collected from various sources and discusses the key factors affecting measured noise levels around an airport. The processes used to validate the various parts of the noise model are discussed. Emphasis is placed on both accurate estimation of source noise emission and propagation from source to receiver. Throughout the paper, the various techniques and methods are illustrated using data from a recent UK study of approach noise at London Heathrow airport. The deficiencies with present noise modelling techniques such as lateral attenuation are discussed and possible options for future development are considered. 206 Internoise 2001 Abstracts Session B7-3, 15:20 h Paper number 613 (Contributed, Oral) John Van der Heijden and Fokke Van der Ploeg I-INCE Class 56 Adviesbureau Peutz & Associés B.V., P.O. Box 66, NL-6585 ZH, MOOK, The Netherlands RECOGNITION AND QUANTIFICATION OF AIRCRAFT NOISE EVENTS INSIDE DWELLINGS To examine the connection between aircraft noise and a certain parameter inside dwellings (e.g. sleep disturbance) it is necessary to determine whether an aircraft passage occurs and to value such an event. Monitoring the soundlevel is necessary for that purpose. As partner in a large survey our company is responsible for the acoustical aspect for which we monitor the soundlevel inside a number of dwellings on different locations in the vicinity of Amsterdam Airport Schiphol. In a pilot study, done by another research institute, the recognition and the quantification of the aircraft noise events was done on a real-time basis by which a threshold level had to be exceeded, with minimal duration and a certain exceedance hysteresis. The disadvantage of this method was that in situations with a high background level or noise disturbance (passing cars, snoring, overlapping aircraft events) events were missed or were taken into account incorrectly. Therefore a method has been developed by which the soundlevel is registered continuously, so afterwards a correct analysis is possible. By linking the sound data to radar observation data from the airport Flight track And Noise Monitoring System (FANOMOS) a good estimate can be made on what moment an aircraft passes. The method is automated in such a way that for every aircraft event maximum and SEL-values are determined. Based on different criteria the event is evaluated whether it has to be taken into account or not. In comparison with the pilot setup the percentage of correctly determined aircraft events is highly increased. Session B7-3, 15:40 h Paper number 413 (Contributed, Oral) Y. Takakuwa, S. Kodani and M. Ohta I-INCE Class 72 Hiroshima Kokusai Gakuin Univ., 6-20-1 Nakano, Aki-ku, Hiroshima, 739-0321 Japan AN ESTIMATION METHOD OF LOW-FREQUENCY AIRCRAFT NOISE UNDER THE CONTAMINATION OF WIND NOISE BY USING THE WIND SPEED INFORMATION This paper describes experimental study on a simple and practical estimation method of low-frequency aircraft noise contaminated by wind noise. Because, it is often difficult to remove low-frequency component of the wind noise by using only usual types of windscreens, some estimation methods of the aircraft noise are required in the windy conditions of outdoor measurement. In this study, a simple estimation method by using the combination of two kinds of information on the wind noise are proposed from the viewpoints of practical use. The two kinds information are (1) the wind speed near a microphone and (2) the infrasonic components of observed wind noise itself. First, the experimental verification of the cross-correlations between the wind noise and the information (1) and (2) are shown. The SPL and the power spectrum of the wind noise are expressed in the form of conditional averages on knowing above two kinds of information. Next, the simple estimation method based on the combination of two conditional averages is shown. Finally, in order to show the effectiveness of the proposed method, some simulation experiments are carried out by using aircraft noises and the wind noise separately observed at a park near an airport and a building roof. The results show the effectiveness of the proposed method. Internoise 2001 Abstracts 207 Session B7-3, 16:00 h Keith Adams Paper number 513 (Contributed, Oral) I-INCE Class 74 Lochard Pty Ltd, 69 Kooyong Road, Caulfield North 3161, Australia SUCCESS AND FAILURE ANALYSIS OF NEURAL NETWORK IDENTIFICATION OF AIRCRAFT NOISE Neural network methods implemented in standard noise monitoring terminals provide a very useful tool in the monitoring and management of aircraft noise. In mixed noise environments it is important to compute correctly the noise loads separately due to aircraft and to other noise sources. The neural network technique results in high rates of correct assignment. But this fact makes the reasons for the small number of failures an interesting topic for research. In this paper we study noise samples whose third-octave and specific loudness spectra are nearly identical but which arise from different types of noise source. An examination of the original (raw) signals leads to a study of the characterisation of additional features which must be added to the input of the neural network to reduce the identification error. The results of simulation and improved training will be presented. Session B7-3, 16:20 h Paper number 97 (Contributed, Oral) L. Montrone, F.J.M. Wubben, A.J.C. Roeloffs and P. Vogel I-INCE Class 76 National Aerospace Laboratory NLR – P.O. Box 90502 – 1006 BM – Amsterdam, The Netherlands FEASIBILITY OF COUPLING OF AN AIRPORT CAPACITY MODEL TO AIRPORT NOISE MODELS In the last few years the demand from Amsterdam Airport Schiphol (AAS) and governmental policy makers for an integrated airport analysis concept (including environmental and capacity issues) is constantly increasing in order to fulfil future strategically decision making. Therefore the NLR has founded the ‘Airport Scenario and Analysis Package’ (ASAP). One of the work-packages of ASAP is a study connecting the fast time simulation modeller TAAM (used for airport capacity and punctuality studies) with two different noise models (the legal Dutch ENVIRA model and the US INM). Anticipating probable European standardisation in airport noise models, the desire exists to compare the present Dutch noise model with the IMN model. First step in this research has been the realisation of the coupling between the TAAM model and both noise models. Then, noise calculations are performed with both noise models based on the same traffic forecast scenario for AAS. Finally, Lden noise contours are compared and analysed. The paper provides a description of the two coupling processes developed and the presentation and analysis of the final results. Session B7-3, 16:40 h Paper number 125 (Contributed, Oral) J.-F. Rondeau and Ph. Jean CSTB, 24 rue Joseph Fourier 38400 St Martin d’Hères, France A NUMERICAL MODEL FOR NOISE PREDICTION NEAR AIRPORTS I-INCE Class 76 Several numerical tools have been developed at CSTB for the prediction of noise near airports. Depending on the main physical aspects considered, various models can be employed. The propagation of sound can be analysed by means of ray-tracing algorithms. Effects such as diffraction can be introduced (computer program MITHRA). The coupling with INM data bases has also been developed. Meteorological effects can be considered by means of various approaches such as FFP (Fast Field Program), GFPE (Parabolic Equation). Complex Green functions which include 208 Internoise 2001 Abstracts meteorological aspects can be used in BEM models. Noise predictions are usually made at points near the facades. The GRIM method, developed at CSTB, has been applied to the assessment of noise inside buildings. It consists into coupling a ray tracing algorithm - in the form of a complex outdoor Green function- with an integral representation. Due to the favoured transmission of sound at grazing incidences, the introduction of the sound transmission through windows is essential for the correct assessment of airport noise. As a consequence it becomes possible to study effects such reflections and diffraction from buildings nearby. Optimised building shapes are rendered possible and may take into account balcony or projected roofs effects. Session B7-3, 17:00 h Paper number 268 (Contributed, Oral) F.J.M. Wubben and S.P. Galis I-INCE Class 76 National Aerospace Laboratory NLR, P.O. Box 90502, 1006 BM Amsterdam, The Netherlands AN AIRCRAFT NOISE EXPOSURE FORECAST MODEL BASED ON ACTUAL MEASUREMENT DATA AT NOISE MONITORING SITES With the realisation of a fifth runway at Amsterdam Airport Schiphol (AAS) in the year 2003, the Dutch government has decided to introduce a new system of noise regulations at the same time. One important requirement is that the system on the long term (2005+), complies with noise monitoring and control based on noise measurements. The new noise regulations use a limited number of noise monitoring sites located in residential areas around the airport. The actual measured noise exposure at each monitoring site should not exceed a predefined calculated limit. During the establishment of the noise control limits, it is necessary to consider the differences between noise calculations and noise measurements. In order to eliminate the mentioned differences, a forecast model is studied, based on measured data of each specific monitoring site. By clustering the measured data of individual flights as function of aircraft type (or category), runway, route and flight procedure, energetically averaged noise levels are obtained. From traffic forecasts the number of flights for the same cluster of parameters can be extracted. By combining the traffic forecast and the averaged measured noise levels, a consistent noise exposure forecast is obtained for each control location (monitoring site). The main advantage of this forecast model is the elimination of differences between calculations and measurements. A disadvantage of the model is that sufficient measurements must be acquired, before stable averaged measured noise levels are obtained for each cluster of parameters. The paper describes the forecast model in more detail. Session B7-3, 17:20 h Paper number 271 (Contributed, Oral) Lorenzo Lombardi and Simone Relandini I-INCE Class 76 ENEA, Univ. of Rome “La Sapienza”, Via Anguillarese, 301 – 00060 S. Maria di Galeria (Roma) – Italy STATISTICAL ANALYSIS OF AIRPORT NOISE: AN ITALIAN CASE STUDY The proposed work will regard the problem of the indetermination of the SEL (Single Event Level) produced by single aircraft noise events and the repercussions involved about the forecast of aeroportual noise indicators by means of calculation models. An experimental study will be presented: measured and calculated data will be compared and subjected to statistic analysis. The experimental data are the results of a set of measurements regarding one of the principal Italian airports and concern the SEL measurements of every take-off and landing events in different days and in different Internoise 2001 Abstracts 209 observation points. The computated data will be instead calculated through ECAC-CEAC Doc. 29 procedures. This study will offer an useful contribution to the planning of the aeroportual traffic in order to contain the acoustic pollution in the aeroportual surroundings, offering the possibility of statistically calculate the expected number of outranges of the law-imposed limits caused by statistic fluctuations. A statistic type model will be proposed; it could be used for the forecast both of the noise level and of his variability by means of the predetermination of the relative standard deviation. POSTERS Session B7-3, 17:40 h Paper number 295 (Contributed, Poster) Posterboard CP-13 Thursday, Author present 9:40 – 10:40 h Rein C. Muchall I-INCE Class 72 OMEGAM, Environmental Research Institute of Amsterdam, P.O. Box 94685, 1090 GR Amsterdam, The Netherlands EXPERIENCES WITH AN INTELLIGENT MEASUREMENT SYSTEM FOR MONITORING AIRCRAFT NOISE WITHOUT RADAR ASSISTANCE At the end of the eighties, the City of Amsterdam was confronted with a great number of calculated noise maps of the Amsterdam Airport Schiphol and its surroundings. There were considerable differences between these noise maps, which caused doubts about the reliability of the Dutch calculation method (using the FAA aircraft noise database). Therefore the decision was made to directly measure the actual yearly average noise level. The costs of monitoring systems, which were available on the world market exceeded the budget of the environmental department of Amsterdam and moreover, many systems were dependent on the radar system of the airport. Therefore OMEGAM was asked to develop a low-cost monitoring system, operating independently of the airport. A stand-alone noise monitoring system was then developed, consisting of a standard noise level meter, a wind speed meter, a standard PC with data acquisition card and later a radio scanner. The software, which is an essential part of the system, controls the measuring process, guards the measuring conditions, performs the aircraft noise recognition and calculates the various noise climates. Finally, a test procedure was developed in order to fine-tune the recognition software and validate the measurements. These tests provide data about the accuracy of this system under different weather conditions and signal-noise ratios. The results over a 10 year period from 1991 -2001 showed a significant difference between the Dutch noise calculations and measured noise levels. Session B7-3, 17:40 h Paper number 517 (Contributed, Poster) I-INCE Class 13 Posterboard CP-14 Thursday, Author present 9:40 – 10:40 h Oleksander Zaporozhets, VadimTokarev and Keith Attenborough National Aviation University (NAU), 1, Cosmonaut Komarov Prospect, Kyiv, Ukraine,03058 BASIC RECOMMENDATIONS FOR ASSESSING NOISE FROM GROUND SOURCES AT AIRPORTS Take-offs, landings, taxiing of the aircrafts, engine testing are main on-groundnoise sources at the airports. In the absence of barriers, the main influence on propagation from an them is ground effect - 210 Internoise 2001 Abstracts the influence of interference between direct and reflected sound waves. Few methods of predicting ground effect are compared with each other and with some data for aircraft engine noise at run-up. All techniques allow consideration of the different types of ground cover, including mixed covering, that are usually found inside and outside airports. Also it is possible to consider the influence of various types of source components including monopole, dipole or quadrupole. The results were used for recommendations of noise calculation in airports. Session B7-3, 17:40 h Paper number 600 (Contributed, Poster) I-INCE Class 52 Posterboard CP-15 Thursday, Author present 9:40 – 10:40 h Luigi Maffei, Saverio F. Del Gatto, Vincenzo Filomena, Gianluca Pezzullo, Rosario Roman and Daniela Siano DISPAMA – Second University of Naples, Borgo S. Lorenzo, 81031 Aversa, Italy NOISE MEASUREMENTS OF MILITARY AIRCRAFT STANDARD FLIGHT OPERATIONS Complaints of people living in the neighbourhood of military airports stimulated the interest of the Italian Air Force in the evaluation of the noise pollution caused by military aircraft operations. CIRA (Italian Aerospace Research Center) carried out suitable measurements aimed at the realization of a noise database for the Italian Air Force aircrafts (Milnoise). The database shall be useful for scheduling ordinary flight operations to minimize the noise impact on inhabited areas. Until now more than 5000 measurements have been carried out during 400 flight operations near three Italian military airports. The measurements concerned three military aircrafts usually utilized by Italian Air Force. The aim of this work is to present the features and capability of the Milnoise Database and Procedure to predict the noise pollution around an airport. The software INM is considered as a tool to predict with a good accuracy the average aircraft noise due to several aircraft standard-operations. It’s the experience of the authors (Prediction of the noise pollution caused by military aircraft flyovers - L. Maffei, V. Filomena, R. Romano, R. Capriotti - Internoise2000 Congress) that the above-mentioned software allows an acceptable accuracy of the aircraft noise in terms of SEL also for a single overflight operation, if detailed NPD curve are used. Instead the prediction accuracy was found worse for take-off and landing operations at locations observated at a long distance from the runway. Attempts were carried out to improve the INM software prediction accuracy also for single take-off and/or landing aircraft operations far away from the runway. Session B8-1: Noise barrier attenuation Chaired by Judith L. Rochat and Courtney Burroughs Session B8-1, 9:40 h Paper number 604 (Invited, Oral) Courtney B. Burroughs and Anthony R. Bontomase NOISE PROPAGATION PATHS FOR HIGHWAY NOISE BARRIERS I-INCE Class 31 Graduate Program in Acoustics, The Pennsylvania State University, State College PA 16804 USA There are three potential paths for noise propagation from the traffic side of a highway noise barrier to the other side of the barrier; 1) transmission through the barrier, 2) diffraction at the edges of the barrier, and 3) scattering from atmospheric turbulence above the barrier. A wood barrier was built . To separate propagation paths, measurements were conducted with an array of microphones steered Internoise 2001 Abstracts 211 toward the surface of the barrier, to the top edge of the barrier, and above the barrier. Also, measurements were made with the ground on the receiver side of the barrier partially covered with reflecting and absorptive surfaces. Finally, the vibration response of the barrier was measured and used to estimate transmission through the barrier. Results from these measurements are discussed. Session B8-1, 10:00 h Paper number 230 (Invited, Oral) I-INCE Class 31 Roger L. Wayson, John M. MacDonald, Wayne Arner, Chris Corbisher, Patricia Tom, D.S.R.K. Srinivas and Brian Kim University of Central Florida, Civil & Environmental Engineering Department, P.O. Box 162450, Orlando, FL 32816-2450 NOISE BARRIER MEASUREMENT, MODELING AND EVALUATION AT MULTIPLE SITES IN FLORIDA The primary goals of this project were to answer three specific questions: Are the barriers providing adequate protection for highway neighbors? Are the noise barriers that have been built in Florida performing as predicted? Is the new FHWA Traffic Noise Model significantly better than the previous model, STAMINA2.0/2.1? To answer these questions, and fulfil the goals of this project, specific data were needed. This data was of two types: field data and computed data. The field data included measured sound levels (both 1/3 octave and A-weighted), specific site geometrics, traffic counts, traffic speeds, weather data, and field notes of unusual occurrences. The computed data was overall trend analysis of the collected data, prediction of sound levels at specific positions and comparative analysis. This paper provides a brief overview of the data collected and a discussion of the measured results when compared to modeled results. Important points are then discussed in relation to the three questions stated above. Conclusions are then presented on the effectiveness and benefits of the noise barriers. Important findings included actual zone of protection, differences in modeling methodologies and the importance of ground effects. Session B8-1, 10:20 h Lloyd A. Herman Paper number 238 (Invited, Oral) I-INCE Class 31 Ohio University, 141 Stocker Center, Athens, OH, 45701,USA INVESTIGATION OF TRAFFIC NOISE BARRIER EFFECTIVENESS FOR DISTANT RECEIVERS In the United States a number of state transportation agencies have reported complaints of increased noise levels for residents living in the vicinity of highways after traffic noise barrier construction. Referred to as the distant receiver problem, complaining residents have generally been located at distances greater than the typical source-to-receiver distances considered for traffic noise abatement. The completion of the largest traffic noise barrier project in the state of Ohio brought a mixed response from the public, with some residents claiming that traffic noise levels had been reduced, while others contended that noise levels had increased. An extensive public opinion survey concluded that the most pervasive complaint, the perception of increased traffic noise levels due to noise barrier construction, was cited most often by residents living in the range of 180m to 540m from the roadway. Further, those residents protected by parallel barrier configurations were more likely to perceive increased levels than those protected by single barriers. No noise measurements were conducted prior to noise barrier construction. Therefore, the investigation used both field measurements and noise models to identify the contributions of a number of mechanisms, which affected noise propagation. 212 Internoise 2001 Abstracts The study concluded that the perception of the increased traffic noise levels due to noise barrier construction was valid for some locations. It was found that the inter-relationships between barrier attenuation and ground attenuation plus the effect of reflections between parallel barriers were the most probable cause for the perception of increased noise levels for distant receivers. Session B8-1, 10:40 h Paper number 477 (Invited, Oral) Douglas E. Barrett and Eric R. Seavey I-INCE Class 31 Harris Miller Miller & Hanson Inc., 15 New England Executive Park, Burlington, MA 01803 USA COMPARISON OF STAMINA2.0/OPTIMA TO TNM RESULTS AND EFFECTS ON NOISE BARRIER ANALYSIS Harris Miller Miller & Hanson Inc. conducted evaluations for nine noise barrier areas using both traffic noise prediction models currently accepted by the U.S. Federal Highway Administration (FHWA) and compared the results. FHWA currently is transitioning from STAMINA2.0/OPTIMA to the FHWA Traffic Noise Model (TNM), and will require use of TNM on all projects begun after December 31, 2001.During the transition period, use of either traffic noise prediction model is permitted by FHWA. The comparisons were conducted using identical geometries and traffic data. In all of the no-barrier cases, sound levels computed by the two traffic noise prediction models were in close agreement to each other (generally within one to two decibels).In all of the with-barrier cases, TNM predicted greater average insertion loss than did STAMINA2.0/OPTIMA (which uses barrier mathematics based upon Maekawa).STAMINA2.0/OPTIMA, however, typically predicted greater additional incremental benefit with increased barrier height than did TNM. Analysis of the models' output indicated that noise generated by traffic on unshielded frontage roads enhanced this result. The comparisons also indicated that the choice of noise prediction model could significantly affect the costs associated with noise mitigation due both to significant differences in recommended barrier heights and whether or not a particular barrier would be recommended as acoustically feasible and economically reasonable. Session B8-1, 11:00 h Judith L. Rochat Paper number 506 (Invited, Oral) I-INCE Class 31 U.S. Department of Transportation, Volpe Center Acoustics Facility, 55 Broadway, DTS-34, Cambridge, MA 02142, USA OBSERVATIONS OF HIGHWAY TRAFFIC NOISE MEASUREMENTS BEHIND BARRIERS AND COMPARISONS TO FHWA'S TRAFFIC NOISE MODEL In 1998, the United States Federal Highway Administration (FHWA) released a new tool for highway traffic noise prediction and noise barrier design, the Traffic Noise Model (TNM). In order to assess the accuracy and make recommendations on the use of TNM for the FHWA, the Volpe Center Acoustics Facility performed numerous roadside measurements, obtaining over 100 hours of traffic noise data from highways around the country. A majority of the measurement sites included noise barriers protecting homes, schools, or recreational parks in the area. These barriers were either berms or walls constructed of various materials with varying heights and configurations. For each site, acoustical, meteorological, and traffic data were collected simultaneously throughout the measurement period. Spectrum analyzers were used to collect 1/3-octave band A-weighted equivalent sound levels, and the microphones were deployed at distances from 50 to 300 feet behind the barrier and at two heights off Internoise 2001 Abstracts 213 the ground, the number of microphones used being site dependent; a reference microphone was also deployed. Preliminary results indicate that these barriers are providing substantial noise reduction to the protected area, with attenuation over distance varying depending on the noise barrier configuration. Results also indicate that TNM is adequately modeling these typical sites; the calculated sound levels are generally within 2 dB of the measured levels. Also, the effects of wind are seen to influence the accuracy of the model; since TNM does not account for wind, the model is, in general, over- and under-predicting during different wind conditions. Session B8-1, 11:20 h Paper number 682 (Invited, Oral) E. Walerian, R. Janczur, M. Czechowicz and W. Barwicz I-INCE Class 31 Institute of Fundamental Technological Research, Polish Academy of Sciences, 00-049 Warsaw, Poland EFFECTIVENESS OF SCREENS DEPENDING ON AN URBAN STRUCTURE ARRANGEMENT Effectiveness of screens in an urban area depends first of all on the fact whether the screen and the protected building form parallel reflecting surfaces. To present the quantitative differences resulting from building arrangement, the simulation program PROP7 is applied. The screen is assumed parallel to the road and two building configurations are tested. First, buildings are placed parallel to the road, next, they are inclined to the road axis. As diffraction accounts for the screen efficiency, the source position in relation to the diffracting wedge is the decisive factor. In the case of a road as noise source this requires enough accurate modeling by elementary point sources. The PROP7 program fulfils the requirements. The PROP7 program has been developed to forecast the equivalent sound level in urban areas. For buildings modeled as shoeboxes and plane screens represented by panels of limited dimensions, wave interaction during propagation process includes reflections from the ground and building walls up to an arbitrary order, and single and double diffraction at wedges. The acoustical energy emitted by a road as a noise source is related to the sound exposure level of an individual vehicles pass-by. The road geometrical parameters are: a number of lanes and their positions. A vehicle movement is replaced by a set of its discrete positions along the route. Thus, when vehicles are divided into classes, for each class the equivalent point source of a given height above ground, power spectrum and directivity characteristic is assumed. A vehicle stream on a lane is characterized by a number of classes and, for each class, by an average speed and flow rate. Session B8-1, 11:40 h G.R. Watts Paper number 686 (Invited, Oral) I-INCE Class 31 Transport Research Laboratory, Crowthorne, Berkshire, RG45 6AU, United Kingdom SOUND TRANSMISSION THROUGH TIMBER NOISE BARRIERS - A SURVEY OF IN-SITU PERFORMANCE Some concern has been expressed over the possible significant leakage of sound through timber noise barriers that are frequently specified in the UK. Inspection of some installations indicated clear visible gaps between and splits in the main timber planks. These gaps could be due to poor workmanship or the result of the ageing process where timbers have warped, shrunk or split. An in-situ method was required to assess the acoustic performance of this type of barrier. A novel sound intensity technique that had previously been validated against a standard laboratory test method (ISO 140/3) was used to measure sound transmission using the normal traffic noise as the source of sound. A survey was 214 Internoise 2001 Abstracts undertaken covering a range of typical barrier types and including examples of different ages of the more common types of construction. This paper describes the tests that have been carried out and gives results in terms of sound transmission loss at different frequencies with an indicator of overall performance based on the single number rating of airborne sound insulation defined in EN1793-2. The observed effects of gaps on the acoustic performance is discussed. Session B8-1, 12:00 h Paper number 697 (Invited, Oral) K. Fujiwara and T. Nanba I-INCE Class 31 Department of Acoustic Design, Kyushu Institute of Design, Shiobaru 4-9-1, Minami Fukuoka 8158540 Japan APPLICATION OF A NOISE BARRIER TO HIGH RISE APARTMENT HOUSES Noise barrier is now widely utilizing for protecting noise from vehicles on the expressway. Although the noise barrier is quite effective at the receiving location lower than the barrier edge, that is, in the shadow region, it has quite less effect in the bright region. In the urban area, it often happens to build the expressway just in front of high-rise apartment houses or to build the high-rise apartment houses quite near to the expressway. The upper part of the high-rise apartment houses is out of the shadow region of the noise barrier, and usual noise barrier has no effect in such a region. Therefore an effective noise shielding system has been required for a log time. This paper presents the effective noise shielding system, which includes noise barrier and barrier edge devices, for the high rise apartment houses. Usual apartment house has balcony with podium and the floor of the balcony of the upper story is the eaves of the lower story. The noise shielding system is quite simple and to cover the lower surface of balcony with absorptive material and install absorptive or soft obstacles at the podium edge. Following the experimental study about 15dB noise reduction in all over frequency range compared with in the case without any absorptive treatment. In the paper the optimum arrangement of these materials will be discussed. Session B8-1, 13:20 h Paper number 180 (Contributed, Oral) S.P. Lim, H.P. Lee and W. Shao I-INCE Class 31 Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117576 PERFORMANCE OF NOISE BARRIERS OF DIFFERENT TOP EDGE DESIGNS Semi-empirical methods based on ray tracing and geometrical acoustics are frequently used in acoustical analysis. However, these methods are usually applicable to noise barriers with simple configurations and are unable to treat complex diffraction and multiple scattering effects. For the latter problems, the Boundary Element Method (BEM) appears to be a viable approach to compute the acoustical performance of such noise barriers. In this paper, the BEM is used to predict the acoustical performance of single vertical noise barriers, double vertical noise barriers, barriers with T-shaped top edge, barriers with Big-cap top edge as well as barriers with multiple top edges. All of these barriers have similar acoustical geometry. Comparisons of their effectiveness of noise reduction at different receivers are presented. Numerical results show that the double vertical barrier can exhibit much better performance than the single vertical barrier. The acoustical performances of other kinds of barriers are varied at different frequencies. Contour maps to indicate the sound pressure distribution behind these barriers are also presented. Internoise 2001 Abstracts 215 Session B8-1, 13:40 h S.A. Storeheier Paper number 504 (Contributed, Oral) I-INCE Class 31 SINTEF Telecom and Informatics - Acoustics, N-7465 Trondheim, Norway PREDICTION OF SOUND REFLECTION FROM VERTICALLY ERECTED SURFACES - A PRACTICAL APPROACH The treatment of sound reflection from vertically erected surfaces enters into most community noise prediction methods. Especially in city environments sound reflections can be of great importance. Often, sound reflections are dealt with by applying the principle of image sources and sound rays. Some problems may arise in practical implementations. The algorithms should be capable of handling the frequency dependence of the sound reflection. In addition, the effects of partial reflection that occur when reflection points hit on the edge of the surface or even outside it, should be taken into consideration. A simple heuristic method based on Fresnel zone application is proposed. It seems to work well for reflections inside the surface, and give reasonable estimates for reflections just outside the surface. The theory will be outlined, and results from this approach and other methods will be compared. Session B8-1, 14:00 h Paper number 592 (Contributed, Oral) I-INCE Class 31 T. Van Renterghem, D. Botteldooren, D. Gabriels and W. Cornelis Department of Information Technology, The Ghent University, St. Pietersnieuwstraat 41, 9000 Gent, Belgium MODIFYING WIND PROFILES TO IMPROVE THE EFFICIENCY OF NOISE BARRIERS IN WIND It is well known that the insertion loss of noise barriers is influenced by refraction of sound by wind. Several authors have analyzed this effect both numerically and in wind tunnel experiments. The question arises naturally whether it is possible to influence the wind-factor and to what extend the performance of the barrier could eventually be increased. This paper reports on a wind tunnel experiment constructed to try to answer this question. The experiment focuses on the insertion loss close to the barrier (10-50m). A model at scale 1/20 is constructed. Both single barrier and double barrier situations are considered. Measured insertion losses in the absence of the barrier are of the order of 10 dB. Wind velocities up to 10 m/s at a height of 10 m are used in the experiments. Windscreens that are roughly of the same height as the barrier are used to modify wind profiles close to the noise barrier. Measurement results are supplemented by numerical simulations for more detailed analyses. The positive effect on barrier performance is of the order of a few dB. Although this effect may seem small at first sight, it is obtained at virtually no additional cost and therefor is of practical interest. 216 Internoise 2001 Abstracts POSTERS Session B8-1, 14:20 h Paper number 306 (Contributed, Poster) Posterboard BP-40 Wednesday, Author present 15:00 – 16:00 h Pontus J. Thorsson Department of Applied Acoustics, Chalmers, S-412 96 Gothenburg, Sweden OPTIMISATION OF THE SURFACE IMPEDANCE OF NOISE BARRIERS PLACED ON BOTH SIDES OF A ROAD I-INCE Class 31 Noise barriers are a common tool for traffic noise abatement. Extensive work has therefore been performed on this subject to form accurate prediction models and to optimise noise barrier performance. In previous work the equivalent sources method was applied as a useful tool for optimisation of noise barrier insertion loss. However, only barriers at one roadside have been optimised, which in general is a simplified situation. At many locations barriers are needed on both sides of a road. It has been shown in several studies that reflections between barriers located on opposite sides of a road can decrease the insertion loss. This effect is dependent on the shape and the surface treatment of the barriers, and the effect increases with increasing ratio between barrier height and separation distance. Special attention should thus be paid to such barrier sites when optimising the surface impedance of the barriers. For practical reasons the impedance should also be smooth over the barrier surface and show only a weak dependency on the frequency. In this paper the surface impedance of two parallel barriers located on opposite sides of a road is optimised to give maximum insertion loss at multiple frequencies simultaneously, keeping the surface impedance constant inside the frequency band of interest. The ground is modelled as a flat plane with inhomogeneous locally reacting impedance. The resulting insertion loss is studied and compared to the insertion loss achieved from rigid barriers with equal geometry to the optimised barriers. Session B8-1, 14:20 h Paper number 494 (Contributed, Poster) Posterboard BP-41 Wednesday, Author present 15:00 – 16:00 h Carlo Baistrocchi and Luciano Rocco DPMPE, Università di Firenze, Via G. C. Vanini 30, 50129 Firenze, Italy I-INCE Class 31 A CALCULATION MODEL OF HEIGHT AND LENGTH OF TRAFFIC NOISE BARRIERS The paper describes a model of calculation developed by the authors to determine height and length of roadway and railway traffic noise barriers, according to the International Standard ISO 9613-2/96. Solutions to problems which are left open by the Standard are proposed. By the method the road is broken into a series of straight segments of finite length for analysis. The model allows the prediction of the contribution of each segment to the level at the receiver and, consequently, the effectiveness of the single segment of noise barrier, so that an evaluation of costs and benefits connected with the changing of the dimensions of the barrier is made possible. The model allows the separate evaluation of the different attenuation terms when the relevant input data are changed. The paper is completed by an example of application and by the comparison of the results with the results obtained with traditional methods. Internoise 2001 Abstracts 217 Session B8-2: Military weapon noise mitigation Chaired by Frank van den Berg and Keith Attenborough Session B8-2, 14:40 h Paper number 312 (Invited, Oral) Frank H.A. Van den Berg TNO-TPD, Stieltjesweg 1, PO Box 155, 2600 AD Delft, The Netherlands COUPLING OF A METEOROLOGICAL MODEL WITH A SOUND PROPAGATION MODEL I-INCE Class 24 By order of the Dutch Ministry of Defense research has been carried out in order to predict sound levels around shooting ranges under complex meteorological situations. These situations occur where shooting ranges are situated near large water surfaces. Due to large differences between the surface temperature of water and land, wind and temperature gradients will change along the sound propagation path. In some cases wind directions above water can be opposite to those above land, the so-called sea-breeze. The most commonly used models can’t handle these complex situations: Often only one effective sound speed profile is used to describe the meteorological situation for the whole propagation path. In this study a coupling has been made between a meteorological model and a sound propagation model. In the meteorological model a detailed treatment of soil and vegetation has been built in and also the tide is taken into account. The sound propagation model is based on the wellknown PE-method. The combination of the two models gave promising results, some numerical results will be presented. Both models are two-dimensional. In future they will be extended to three dimensions. Session B8-2, 15:00 h K.-W. Hirsch Paper number 309 (Invited, Oral) I-INCE Class 13 Institut für Lärmschutz, Arnheimer Straße 107, D-40489 Düsseldorf, Germany AN ENERGY MODEL FOR PROJECTILE SOUND Supersonic projectiles generate an acoustical shock wave along their trajectory. This projectile sound is only audible in the Mach area. The geometry of this area depends on the projectile speed relative to the speed of sound and on the decrease of the projectile speed along the trajectory. At some distance from the projectile, the shape of the waveform is the typical N-wave. In 1952 and 1953, Witham published two papers on the prediction of the sound pressure of projectile sound including the nonlinear effects. The pressure prediction depends on the diameter, length and shape of the projectile and on the local Mach number. As a consequence of non-linearity, the spectral energy content is not constant but depends on distance. For large area, multiple source noise contour maps, this model leads to long calculation times and - due to some limitations in the model - generates prediction errors in those cases where the projectile speed becomes subsonic along its trajectory. Therefore, an energy model for projectile sound was developed to overcome these problems. This energy model assumes that the source of the projectile sound is the local loss of kinetic energy. A fraction of that energy loss is radiated as sound energy into the direction determined by the local Mach number. For distances far enough to apply linear acoustics, this model predicts the free field sound exposure level and a constant time duration of the N-wave. The paper introduces this model and compares the result to the nonlinear pressure model. 218 Internoise 2001 Abstracts Session B8-2, 15:20 h Paper number 51 (Invited, Oral) Olga Umnova, Keith Attenborough and Alan Cummings I-INCE Class 35 Department of Engineering, University of Hull, Cottingham road, Hull, England, HU6 7RX HIGH AMPLITUDE PULSE PROPAGATION AND REFLECTION FROM A RIGID POROUS LAYER High amplitude sound propagation in rigid-framed and elastic porous materials has been studied theoretically for both harmonic plane waves and acoustic pulses. Previously, it has been shown that the conditions for shock formation in the waves propagating through media with Forchheimer’s nonlinearity differ significantly from those in media with linear absorption. Harmonic signals with frequency lower than a critical value (or pulses with duration higher than a critical value) propagate without distortion irrespective of their amplitude. In the work reported here, equations describing acoustic pulse propagation in rigid-framed porous materials have been solved analytically by the method of characteristics taking into account the weak nonlinearity of the hydrodynamic equations and the equation of state. The dependence of dc flow resistivity on the amplitude of the particle velocity has been combined with Johnson’s expression for complex flow resistivity to give a model for high amplitude harmonic wave propagation in (rigid) porous media. Coupled equations for the amplitude and phase of fast and slow modes propagating in elastic porous materials have been derived at high amplitude and for the case of Forcheimer nonlinearity. A method for their approximate analytical solution is suggested. Nonlinear acoustic admittance has been calculated as a function of sound amplitude for a wide range of sound frequencies and compared with data. The effect of frame properties on the absorption characteristics of the material has been estimated. Session B8-2, 15:40 h Paper number 81 (Invited, Oral) I-INCE Class 35 Edwin Standley, Olga Umnova, Keith Attenborough and Alan Cummings Department of Engineering, The University of Hull, Hull HU6 7RX, England ABSORPTION CHARACTERISTICS OF POROUS MATERIALS FOR HIGH INTENSITY SOUND A 2m long shock tube with 3mm thick steel wall and containing a 50cm long transparent section has been constructed. A shock wave produced at one end of the tube is recorded by means of data acquisition card linked to pressure transducers at different locations along the tube. The absorption characteristics of various materials have been deduced from the recorded pressure peaks. The displacements of the porous materials have been measured using a rectangular section of flexible wire subjected to a strong magnetic field. The voltage fluctuations resulting from dynamic compression of the porous material are proportional to the speed of the surface. To measure the non-linear impedance of the porous materials the shock tube is used as an impedance tube. Phase speeds, transmission loss and absorption coefficients are obtained in addition to impedance measurements. Transmission loss measurements have been carried out with various air gaps between porous material and the back wall. Session B8-2, 16:00 h Paper number 310 (Invited, Oral) K.-W. Hirsch and E. Buchta Institut für Lärmschutz, Arnheimer Straße 107, D-40489 Düsseldorf, Germany ABSORPTION OF BLAST SOUND CLOSE TO THE SOURCE I-INCE Class 35 The firing of weapons or demolitions for training purposes normally takes place at dedicated sites. Commonly used, pre-defined training scenarios result in repetitive blast emissions. In many cases, Internoise 2001 Abstracts 219 these emissions are dominated by reflections of the blast sound from surfaces close to the weapon or from obstacles that are present at the site during firing. The ground is one of these reflecting surfaces. As a consequence, sound at receiving sites can be dominated by the reflected sound at the source. Therefore, noise abatement measures applied to reflecting surfaces close to the muzzle of a gun or close to a demolition can very effectively reduced the sound at receiver sites. Close to a blast source, the acoustical levels are very high and the rules of linear acoustics do not necessarily apply for absorption phenomena. In order to develop noise abatement measures close to a blast source, tests were conducted using small demolitions (50 g) and muzzle blasts of a pistol shooting blanks. This paper reports on tests that investigated the use of mineral and steel wool for blast sound absorption. One major result is that the benefits of absorbing materials near to the source are limited by the significant reflections of the blast at the surface of the absorbing material. To reduce this effect, a pile of layered absorbers was tested with the density gradually increasing from layer to layer. Session B8-2, 16:20 h Paper number 410 (Invited, Oral) E. Vedy, E. Salomons and Jan V.D. Heijden TNO-TPD, Stieltjesweg 1, P.O. Box 155, 2600 AD Delft, The Netherlands NUMERICAL SIMULATIONS OF STRONG SHOCKS I-INCE Class 35 A computational study of pressure waves generated by large guns is presented. We use aero-acoustic simulations of the time dependent flow field of a muzzle blast wave, in the region inside and near the muzzle of the gun. The long-term objective of this research is to achieve mitigation of the pressure waves by means of absorbing (ground) surfaces, a great issue for the reduction of sound from military shooting ranges. The method and simulations presented here are based on a finite volume flux corrected transport algorithm, designed to solve the unsteady, compressible Euler equations. The flux correction technique allows an accurate reproduction of strong flow discontinuities, such as shocks, without producing oscillations, nor over damping the solution. As a preliminary evaluation simulation we consider an axisymmetric configuration and use simplified initial conditions: outside the gun, the temperature, pressure and velocity are set to the normal values for the air at rest, while, inside the gun, they are set to, respectively, 18MPa, 1000 m/s and 1650 K. The results of this simulation, in agreement with available data, show that the method is well suited for this type of computations, in terms of cpu time and accuracy. We will present other results for more complex geometries (muzzle, barriers), as well as a coupling of the present method with the NPE method, to compute the propagation of the sound waves in the further range of the gun. Developments in that sense (three dimensional curvilinear extension) have been initiated. Session B8-2, 16:40 h Donald G. Albert Paper number 596 (Contributed, Oral) I-INCE Class 24 USA Cold Regions Research and Engineering Laboratory, 72 Lyme Road, Hanover, NH 03755 USA BLAST NOISE MITIGATION BY GROUND CONDITIONS - PRELIMINARY ANALYSIS Experimental measurements were conducted to determine the effect of ground conditions near the source on blast noise levels. The measurements were conducted at two locations, one with temperate soil conditions and one with a seasonal snow cover. Pressure sensors were used to record the waveforms produced by the detonation of 0.6 kg charges of C4 explosive at distances between 10 m and 110 m. The measured blast waveforms were elongated and the peak amplitudes were significantly 220 Internoise 2001 Abstracts reduced at all distances when a snow cover was present. Theoretical waveforms were calculated and agree with the measured data. The results show that the interaction of the blast wave with the ground near the explosion is an important factor in the blast noise levels received at long propagation distances. This finding shows that modification or control of the ground properties near military artillery training locations is a potential method of mitigating the noise produced by these activities. Session B8-3: Numerical modeling Chaired by Paul Sas Session B8-3, 17:00 h Paper number 536 (Invited, Oral) B. Van Hal and W. Desmet and D. Vandepitte and P. Sas K.U. Leuven, dept. Mechanical Engineering, division PMA, Belgium A COUPLED FINITE ELEMENT - WAVE BASED APPROACH FOR THE STEADY-STATE DYNAMIC ANALYSIS OF ACOUSTIC SYSTEMS I-INCE Class 76 The finite element (FE) method is widely used for the steady-state dynamic analysis of coupled vibroacoustic (VA) systems. There is almost no limitation with respect to (wrt) the geometry of such systems. However, the FE method is limited to the low frequency range, due to the high number of elements required for an accurate description of the spatial variations of the field variables. An alternative method is the wave based (WB) method developed by Desmet. This method exhibits better convergence properties wrt the number of degrees of freedom, such that it can be applied for higher frequencies than the FE method. The reason for these high convergence rates is the use of approximation functions, which unlike the FE polynomial approximation functions satisfy the governing partial differential equations. However, the good convergence properties lead only to higher computational efficiency for VA systems of moderate geometrical complexity. In order to obtain accurate higher frequency predictions for a VA system of more complex geometry, a coupling between the FE method and the WB method is proposed. The resulting hybrid model consists of a small WB submodel for the subsystems of simple geometry and a large, detailed FE submodel for the subsystems of complex geometries. This new approach is validated for a VA system of simple geometry to show its advantage over the FE method in terms of its accuracy at high frequencies. Session B8-3, 17:20 h Paper number 587 (Invited, Oral) Svante Finnveden, Fredrik Birgersson and Neil S. Ferguson MWL, Vehicle Engineering, KTH, SE-100 44 Stockholm, Sweden I-INCE Class 76 MODELLING WITH THE SPECTRAL FEM FOR TURBULENCE EXCITATION OF CYLINDRICAL PIPES The Dynamic Stiffness Method (DSM) and equally the spectral Finite Element Method (FEM) use exponential functions as basis functions, so that exact solutions to the homogeneous equations of motion may be used. This restricts analysis to elements where the excitation is at the ends. This study models distributed excitation that has an exponential dependence. This requires that the particular solution is included in the set of basis functions. Waveguides of increasing complexity are studied as well as some different descriptions of the excitation. A numerical implementation for various beam Internoise 2001 Abstracts 221 elements with distributed excitation is presented. A Corcosmodel of Turbulent Boundary Layer excitation is considered, using both modal analysis and spectral FEM/DSM, for comparison. Session B8-3, 17:40 h Paper number 352 (Contributed, Oral) J.L. Horner and R. Lyons MODAL SCATTERING FROM RECTANGULAR ORIFICES I-INCE Class 23 Aeronautical & Automotive Engineering, Loughborough University, Loughborough, LE11 3TU, UK It is a common problem in engineering to determine the reflected and scattered sound fields resulting from an incident wave, of arbitrary angle, impinging on a rectangular orifice in a rigid baffle of finite thickness. Previous investigations have used Fourier Transforms to establish the amplitudes of these fields. This study concentrates on the coupling of the modes in the scattered field. The work in this paper is a continuation and logical progression of previous studies on modal coupling in the scattered fields from circular orifices. The objective is to determine if a modal approach may be taken to approximate the scattered field from such an orifice. Results are shown in the form of the modal contribution to the amplitude of the scattered field. Simple, empirically based, expressions are used to determine the maximum modal contribution to the field, based only on knowledge of the mode number and the orifice dimensions. Session B8-3, 18:00 h Paper number 511 (Contributed, Oral) Takuya Oshima, Tetsuya Sakuma and Kotaroh Hirate Niigata University, 8050 Igarashi-Ninocho, Niigata City, Niigata Pref., Japan I-INCE Class 23 NUMERICAL STUDY ON EFFECT OF INFLOW WIND CHARACTERISTICS TO GENERATION OF AEROACOUSTIC NOISES FROM BALUSTERS OF BUILDINGS Aeroacoustic noises radiated from exterior subsidiaries of buildings like balustrades or louvers in windy environments have been recognized as a problem for inhabitants recent years. Although many wind-tunnel and field experimental studies have been conducted to address the problem, systematic and effective methods have not still established. One of the reasons is that it is difficult to reproduce precisely the identical conditions in a experiment, because characteristics of the generated aeroacousitc noises are highly sensitive to minor fluctuations in inflow wind conditions, such as wind direction and intensity of turbulence. Therefore, it is considered that computational methods should be introduced which can control the conditions exactly. In the present paper, a numerical study is conducted to investigate the effects of inflow wind conditions to characteristics of generated aeroacoustic noise from exterior balusters of buildings. Numerical techniques applied to the present study include the incompressible LES with standard Smagorinsky subgrid-scale stress model and a discritized sound calculation equation based on Lighthill-Curle formulation. Balusters are modeled as infinite rows of square cylinders by applying periodic boundary conditions on the both sides of the computational domain. The wind directions are chosen to be from zero to 30 degrees, and inflow turbulence is introduced by artificially generating random fluctuation which roughly agree with the turbulent kinetic energy in real-world inflow winds. The results will be shown in the forms of time histories of the generated sound, their frequency-domain representations, coherency function of the pressure distribution on a baluster, and cross-correlation functions between balusters. 222 Internoise 2001 Abstracts Session B8-3, 18:20 h Paper number 119 (Contributed, Oral) I-INCE Class 72 Svend Gade, Nis B. Møller, Henrik Herlufsen, Rune Brincker and Palle Andersen Brüel & Kjær Sound and Vibration Measurement A/S, Skodsborgvej 307, DK-2850 Nærum, Denmark MODAL TESTING OF MECHANICAL STRUCTURES SUBJECT TO OPERATIONAL EXCITATION FORCES Operational Modal Analysis also known as Output Only Modal Analysis has in the recent years been used for extracting modal parameters of civil engineering structures and is now becoming popular for mechanical structures. The advantage of the method is that no artificial excitation need to be applied to the structure or force signals to be measured. All the parameter estimation is based upon the response signals, thereby minimising the work of preparation for the test. This test case is a controlled lab setup enabling different parameter estimation methods techniques to be used and compared to the Operational Modal Analysis. For Operational Modal Analysis two different estimation techniques are used: a non-parametric technique based on Frequency Domain Decomposition (FDD), and a parametric technique working on the raw data in time domain, a data driven Stochastic Subspace Identification (SSI) algorithm. These are compared to other methods such as traditional Modal Analysis. As an example some preliminary measurements on a Volvo S60 passenger car are presented. POSTER Session B8-3 Paper number 677 (Contributed, Poster) Posterboard CP-16 Thursday, Author present 9:40 – 10:40 h Pierre Gélat and Richard Barham I-INCE Class 76 National Physical Laboratory, Centre for Mechanical and Acoustical Metrology, Queens Road, Teddington TW11 0LW, United Kingdom FINITE ELEMENT MODEL OF THE IEC 60318-1 ARTIFICIAL EAR An ear simulator is a standardised device used for the calibration and characterisation of audiometers, earphones, and telecommunications and audio equipment. Two types of ear simulator are widely used: the artificial ear standardised in IEC 60318-1 and the acoustic coupler standardised in IEC 60318-3. The application of these devices to the calibration of new earphone types would be assisted by a better understanding of how an earphone interacts with the ear simulator. Although the artificial ear can be successfully modelled at low frequencies using a lumped parameter approach, the underlying assumptions become less valid at high frequencies. Also, the transducer contained in the earphone and the diaphragm of the microphone used in the artificial ear may no longer move as a rigid bodies. In order to overcome these limitations associated with lumped parameter models, a full threedimensional finite element model of the artificial ear, earphone and microphone has been developed and its response predicted over the audio range. A lumped parameter model was used to compare the results yielded by the model at lower frequencies whereas a practical measurement method provided guidelines as to how the artificial ear behaved at higher frequencies. Internoise 2001 Abstracts 223 Thursday August 30 Plenary Lecture 8:20 – 9:20 h Ulf Sandberg Paper number 741 I-INCE Class 52 Swedish National Road and Transport Research Institute, SE-581 95 Linköping, Sweden TYRE/ROAD NOISE - MYTHS AND REALITIES At this conference, the subject of tyre/road noise is on the agenda to a greater extent than ever before. This paper aims at giving an insight into the past, present and future of tyre/road noise emission in the society as well as its control. This will be made by means of exploring some myths and realities related to the subject. The following myths will be explored: 1. Tyre/road noise has become a concern only during the last decades, say from the 70's. It will be shown that already long ago tyre/road noise was an important issue. 2. Tyre/road noise is an important part of vehicle noise at speeds above 50 km/h (70 for trucks). The truth is that nowadays tyre/road noise dominates during almost all types of driving for cars and at least down to 50 km/h for trucks (in the EU). 3. Different road surfaces may give a large variation in noise levels. True, the variation is very large, albeit the most common surfaces are close together on the noise scale. 4. Different tyres emit similar noise levels. This is not true, the variation is large if sufficiently many tyres are included in the data set. 5. Tyre/road noise is very broadband nowadays. This is both true and not true - in the paper it will be shown that current tyres emit noise very much concentrated within one or at most two octaves. 6. We know everything about the speed influence. It will be shown that there are strange relations between speed-related factors. 7. The width of the tyre is a very influential factor. The paper will present a noise-width relation covering the range from "tiny" bicycle tyres to large truck tyres - a relation surprisingly valid for all these different types. 8. Winter tyres are much more noisy than summer tyres. This is a myth based on the past - today winter tyres are the "quiet" tyres. 9. We can get more quiet tyres only if sacrificing safety. Recent results show that this there is no tradeoff between low noise emission and high safety; also rolling resistance will be covered. 10. It is expensive to reduce tyre/road noise. Some examples will be shown to illustrate that for example a porous road surface may be cost efficient even if it becomes clogged rather soon. A "ultralow-noise" surface may become more cost effective than noise barriers. Internalization of noise costs and noise-related taxes on tyres is an issue in some Nordic countries. 11. Manufacturers have done a lot to reduce vehicle and tyre/road noise. Yes, correct in many respects; but it seems that the noise levels have increased rather than decreased despite this (possibly the trend may have been broken in the past 5 years). 12. Tyre/road noise will be much reduced by the introduction of noise emission limits. The current EU proposal will be almost totally ineffective. The paper will conclude that reduction of tyre/road noise is one of the most interesting challenges of the coming decades. 224 Internoise 2001 Abstracts Session C1-1: Measurement of tyre/road noise Chaired by Jerry Ejsmont Session C1-1, 9:40 h Cristo J. Padmos Paper number 50 (Invited, Oral) I-INCE Class 72 Road and Hydraulic Engineering Division, Post Office Box 5044, 2600 GA Delft, The Netherlands THE ROEMER AS AN INSTRUMENT FOR CONTROL OF NOISE PRODUCTION OF TIRES ON HIGHWAYS Just on this time a discussion is going on if the proposal for an ISO-standard for the Close Proximity Method (CPX) will be accepted and published as a draft for formal voting. The result of this discussion is not sure at the moment of preparing this abstract. In the Netherlands porous pavements are used for a decrease of the noise emissions and to realise lower noise barriers. Specially the new generation of low noise pavements with reductions up to 7 dB, compared with dense asphalt concrete (DAC), have to be tested not only direct after application but during its life time as well. An accepted draft or not, in the Netherlands there is a need for a control method. The Statistical Pass By Method (SPB) is not well suited as an instrument because you need more measurements for approval control method that is statistically justified. The ROEMER is developed for this type of control and the measurements are based on proposed ISO-standard, with some additional requirements to improve the results; the proposed improvements for the ROEMER will be described in the paper. By the time of the conference the results are available to compare results of different locations on the same road type and the result of measurements by different teams. Session C1-1, 10:00 h Paper number 17 (Invited, Oral) Panu Sainio and Ismo Halén Helsinki University of Technology, P.O.Box 4300, 02015 HUT, Finland I-INCE Class 72 NOISE MEASUREMENT TRAILER HUT NOTRA - MEANS FOR MEASURING NOISE DURING EVOLUTION OF ROAD SURFACE Reduction of noise emission has became more important issue among automotive industry. There are also possibilities to affect traffic noise by means of road construction and maintenance. Reduction of noise means new challenges and at the same time gives new business opportunities. There are methods for absorbing noise emitted by traffic and also various amount of means for reducing amount of emission already by the source of noise. One major source is noise generated in tyre-road interaction. These are interesting challenges for automotive engineers and there is need for cooperation between automotive industry and national road administrations. Several standard measurements have been developed for measuring noise caused by vehicles and tyres. Road administration authorities have paid more attention on road pavement construction and its affect on the noise emitted. Laboratory of Automotive Engineering in Helsinki University of Technology has engineered, implemented and utilized a HUT NOTRA (Helsinki University of Technology, Noise Trailer) for measuring both pavements and tyres with CPX-method. This is part of three year national survey of Evolution of Road Surface and Noise from Tyre-Road Interaction. HUT NOTRA is road legal trailer and can be used under normal traffic conditions on normal roads. There are 10 different test pavements in Finland to be monitored over seasons. Some of these pavements are experimental. Some preliminary results about Internoise 2001 Abstracts 225 different road surface materials and their noise behavior development under normal traffic load are already available. Session C1-1, 10:20 h Ryszard Wozniak Paper number 163 (Invited, Oral) I-INCE Class 72 Technical University of Gdansk, Mechanical Faculty, ul. Narutowicza 11/12, PL 80-952 Gdansk, Poland MEASUREMENT OF TYRE/ROAD NOISE IN LONGITUDINAL SLIP CONDITIONS Most measurements of tyre/road noise are nowadays performed at free rolling conditions (without longitudinal slip). On the other hand, it is known that considerable longitudinal or lateral forces acting on the tyre increase tyre/road noise to the great extent. It is not an easy task to reliably measure tyre/road noise at slip conditions. Generally it is possible to make such measurements on ordinary vehicle (with great problem to obtain stabile slip) or on special trailers. At the Technical University of Gdansk a special trailer has been constructed which makes possible the measurements of tire/road noise when large (at about ±10%) longitudinal slip is applied to the tested tire. The slip can be changed continuously and it can be forced by special driving system using principle of ‘circulating power’. The tested tire is enclosed in a semi-anechoic chamber. Microphones are placed in the near field. The paper will present the trailer as well as the first results of road measurements. The investigations were taken on three different road asphalt surfaces, at three speeds (30, 50 and 70 km/h) and longitudinal slip at ranging from -10% to +10%. Session C1-1, 10:40 h Jerzy A. Ejsmont Paper number 105 (Invited, Oral) I-INCE Class 72 Technical University of Gdansk, ul. Narutowicza 11/12 PL 80 952 Gdansk, Poland CERTIFICATION OF VEHICLES DESIGNED TO PERFORM CLOSE PROXIMITY TESTS OF TYRE/ROAD NOISE The ability to measure noise produced by any of the vehicle components is a foundation for reduction of vehicle noise. Since one of the major vehicle noise components is tyre/road noise, it is essential to establish trustworthy and practical methods of tyre/road noise measurements. There are two basic approaches to the problem of the tyre/road noise measurements. In one approach measurements are performed with the microphone(s) located on the roadside (for example Coast-By and Trailer CoastBy methods) and in another they are moving together with the test vehicle (Close Proximity Method "CPX" formerly known as Trailer Method). In case of the CPX method it is important to assure that construction and performance of the vehicles used to carry out the measurements will be "neutral" for the tyre/road noise and at the same time good measuring conditions will be provided. The most essential and critical requirements are then that the background noise will be appropriately reduced and that the reflections inside an enclosure around the test setup (if used) are negligible or under control. The paper deals with the problem of background noise during CPX measurements, distortions of the result due to imperfections in test vehicles designs and proposed approval procedures for the CPX vehicles. The results of experiments evaluating two test trailers built at the Technical University of Gdansk will be presented. 226 Internoise 2001 Abstracts Session C1-1, 11:00 h Paper number 336 (Invited, Oral) S.M. Phillips and P.G. Abbott FACTORS AFFECTING STATISTICAL PASS-BY MEASUREMENTS I-INCE Class 72 Transport Research Laboratory, Crowthorne, Berkshire, RG45 6AU. United Kingdom The Statistical pass-by (SPB) method is widely recognised as the preferred method for characterising the influence of a road surface on traffic noise. The measurement procedure has been used by TRL for nearly thirty years and was developed to provide the road surface correction for the UK prediction method, Calculation of Road Traffic Noise (CRTN). To meet the requirements of the proposed European Directive on Environmental Noise, the SPB method is expected to play a significant role in determining vehicle noise source terms used as input to a European harmonised road traffic noise prediction method. It is likely that the SPB method may also need to be extended to include frequency characteristics as well as overall vehicle noise levels. There are concerns, however, that changes in vehicle and tyre characteristics may undermine the reproducibility and long term stability of SPB values. As part of a recent study for the Highways Agency, TRL carried out a substantial measurement survey to investigate some of these factors. Over twenty thousand vehicle passages were measured at five separate sites in which light vehicles were classified by tyre size and heavy vehicles by axle configuration. The subsequent analysis determined the variation in noise levels from traffic streams consisting of different mixes of these vehicle types. This paper describes this work and the results from a further study that examined the influence of temperature and wet weather on noise levels. Session C1-1, 11:20 h Paper number 257 (Invited, Oral) Wladyslaw Gardziejczyk I-INCE Class 72 Technical University of Bialystok, Building Engineering Institute, ul. Wiejska 45e, 15-373 Bialystok, Poland ANALYSIS OF TYRE/ROAD NOISE RESULTS ON BITUMINOUS AND BLOCK PAVEMENTS SURFACES OBTAINED BY SPG METHOD Tyre/road noise level depends on tyre tread pattern and on the technical and operational conditions of the road surface. The Statistical Pass-By method (SPB) described in ISO Standard is recognized as one of two basic methods suitable for evaluation of noise generated by passing cars. Most of the noise registered by the SPB method originates from tyre/road noise interaction. Thus, the results obtained by SPB method may be used to evaluate road surfaces in respect to their influence on tyre/road noise and on noise propagation over the surface. The paper will present results of SPB measurements carried out in Poland on bituminous and block pavements surfaces both for passenger cars and trucks. In addition a few results of Controlled Pass-By (CPB) tests for selected passenger cars (Fiat Sienna, Renault Laguna and Ford Mondeo) on wet and dry surfaces for rolling speeds from 50 to 110 km/h in steps of 10 km/h will be presented. The paper will also address influence of some local particularities (like traffic composition, vehicle condition) on SPB results. Internoise 2001 Abstracts 227 Session C1-1, 11:40 h Paper number 183 (Invited, Oral) Yasuo Oshino and Tetsuo Mikami I-INCE Class 72 Japan Automobile Research Institute, 2530 Karima, Tsukuba city, Ibaraki, 305-0822 JAPAN STUDY OF ROAD SURFACE INDICES FOR THE ASSESSMENT OF TIRE/ROAD NOISE Since the tire/road noise much changes by the difference of road surface properties, it is necessary to examine the relationship between them in detail and to find the optimum road surface indices for the assessment of tire/road noise. For this purpose, two kinds of investigations were carried out in this study. Firstly, field measurements of tire/road noise were performed on the roads in Europe and in Japan by the close-proximity method. As a result, it was found that there is no significant difference between the measurement results on the roads in Europe and in Japan. Secondly, the relationship between the road surface macrotexture and tire/road noise was examined by the field measurements on various types of roads in Japan. In this study, the macrotexture was measured by using a laser profilometer and the value of MPD (mean profile depth) and the profile level of each road surface were calculated from the measurement results. As the result, it was found that the correlation between the tire/road noise and MPD value is low for overall level but it is significant for the frequency components from 200 Hz to 630 Hz. Session C1-1, 12:00 h Paper number 405 (Contributed, Oral) P.A. Morgan, G.R. Watts and S.M. Phillips I-INCE Class 72 Transport Research Laboratory, Crowthorne, Berkshire, RG45 6AU, United Kingdom TRIALS OF A MOBILE MLS TECHNIQUE FOR CHARACTERISING ROAD SURFACE ABSORPTION A test based upon Maximum Length Sequence (MLS) methods has been developed for the in-situ measurement of the sound absorption of porous road surfaces. This test is detailed in a draft ISO standard and allows a relatively large area of road surface to be examined over a wide frequency range without the need to extract core samples. In addition, because the signal to noise ratio is very high, valid roadside measurements in the presence of traffic can be realised. Currently these tests are carried out using static MLS equipment, however it would be advantageous to have a mobile method. This is because the requirement to close a road to traffic in order to gather large numbers of samples largely precludes measurements being carried out. Therefore TRL has examined the feasibility of conducting such measurements from a mobile platform. This paper describes initial trials carried out on a number of test surfaces using a specially designed trailer with which both static and moving measurements were made. The static measurements are related to the mobile measurements and both sets are compared with standard impedance tube measurements of core samples taken from the test surfaces. Measurements were taken on a range of surfaces including both absorptive and reflective surfaces such as porous asphalt and brushed concrete and speeds ranged up to 25km/h. The paper also examines the potential use of a similar system for the in-situ testing of noise barrier performance. 228 Internoise 2001 Abstracts Session C1-2: Absorbing materials and surfaces Chaired by Z. Engel Session C1-2, 13:20 h Paper number 370 (Contributed, Oral) Kazunori Kimura and Kohei Yamamoto I-INCE Class 31 Kobayasi Institute of Physical Research, 3-20-41 Higashi-motomachi, Kokubunji, Tokyo, 185-0022 Japan THE REQUIRED SAMPLE SIZE IN MEASURING OBLIQUE INCIDENCE ABSORPTION COEFFICIENT In order to decrease the sound reflected from the underside of the upper decks of elevated roads with a double deck structure, highly absorptive materials have been attached to areas of reflection. The method of measuring oblique incidence absorption coefficient was proposed for optimum Aoshima’s time-stretched pulse (OATSP) in a semi-anechoic room. The required size has been determined to be 4m-5m. This is twice the size of that required in the method of random incidence sound absorption coefficient. There are possibilities of measuring with a smaller sample area. For example: if the area of attached absorptive panels is small in site, if the highly absorptive materials are developed by trial and error in a laboratory, or if a pre-production prototype is measured for judging the acoustical performance. The minimum required size is determined by examining specimens ranging from the minimum size (one piece) to maximum size, the full size of the measuring room. The measurements are made on glass fiber and ceramics. It is shown that the minimum required size is 15m2 or more for glass fiber and 20m2 or more for ceramics for the present geometrical arrangement of the source and the receiver. Session C1-2, 13:40 h Paper number 222 (Contributed, Oral) Stéphane Gasser and Florin Paun I-INCE Class 35 Office National d’Etudes et de Recherches Aérospatiales, 92322 Châtillon Cedex, France CLOSE-PACKED METALLIC HOLLW BEADS AS AN ABSORPTIVE MATERIAL FOR AERONAUTICAL PURPOSES Noise control has become a major issue for aeronautics; a way of damping the noise produced in a turbojet while keeping low weight is to use absorptive light materials as structural exhauster components in place of current ones. Such a material can be made of metallic hollow beads piled up and sintered in close-packing. Absorption measures are currently made to compare the frequencyabsorption coefficient behaviour of the real material with the behaviour given by analytical models (cf. Stinson and Champoux, J. Acoust. Soc. Am. 92, 1120 (1992)) at room temperature. As the material will experience a large temperature gradient (500 K to 1300 K) when used as an exhauster component, we are also investigating sound absorption in genuine conditions to figure out whether a multi-layered model is enough to explain the results. Internoise 2001 Abstracts 229 Session C1-2, 14:00 h Paper number 485 (Contributed, Oral) I-INCE Class 35 Miguel A. Picard Lopez, Pedro E. Solana Quiros and Juan V. Arizo Serrulla Depart. de Física Aplicada, High Technical School of Industrial Engineering, Polytechnic University of Valencia, Camino de Vera, 14. D5 B. 46022. Valencia. SPAIN DISCONTINOUS STRATIFIED CONFIGURATION OF FIBROUS MATERIALS The physical discontinuities among materials as those that take place between tires and pavement, or among the different materials of a multilayered composite for architecture constructions, they represent a singular problem in the study of the sound propagation. In particular the simple variation of the density between two layers of fibrous material, it supposes a considerable variation of the porosity and other physical properties in the point of union, for that is not easy the calculation of the acoustic behaviour. When a discontinuity exists in the density of two layers of porous fibrous materials i.e. glass fibre, rockwool and foams, the sound absorption and transmission cannot be obtained by means of the well-known formulas, in fact erroneous results are obtained when the layers have the same physical characteristics. The discontinuity that supposes the union of two layers produces an additional energy dissipation that affects to the theoretical calculations of acoustic properties, difficult to evaluate. Indeed, the permeability suffers an abrupt variation, it affects to the so much conditions of continuity of pressure like of the speed of the particle in the interface. There are different forms of focusing the problem. In this article this discontinuity is evaluated, leaving of the physical and acoustic properties of the samples located before and after the same one. With it is possible to evaluate with more precision the acoustic properties of the discontinuous sample. Session C1-2, 14:20 h Paper number 573 (Contributed, Oral) Yoshihito Kobayashi, Shigeo Hase and Shinji Yamada I-INCE Class 35 Asahi Glass Sound Proof System, co., ltd., 3-14-9, Yushima, Bunkyo, Tokyo, 113-0034 Japan Experimental study on sound absorption of low frequency range with molding chairs in large space room It is one of the important performances to secure speech intelligibility in the large space rooms such as a pool, a gymnasium, a convention center, and an airport. But it is difficult to absorb sound inside these rooms, particularly in a pool. When being treated with sound absorption materials inside the pool, it often happens that reverberation time becomes very long due to the deficiency of the sound absorption power in low frequency range. There are many molding seats in the competitive pool, and the seat that itself became resonator to have the much sound absorption power of the low frequency range, was developed. As the result of examining the form of the seat, it have been achieved that sound absorption power per 1 seat in 125Hz range is 0.15 áu with the experiment in reverberation room. 230 Internoise 2001 Abstracts POSTER Session C1-2 Paper number 136 (Contributed, Poster) Posterboard CP-17 Thursday, Author present 15:00 – 16:00 h Chung P. Park, Michel Brucker and Jean-Francois Koenig Dow Deutschland GmbH & Co. OHG ACOUSTICAL POLYOLEFIN FOAM SUITABLE FOR BOTH THERMAL AND SOUND INSULATION I-INCE Class 35 Conventional acoustical materials are porous mineral fibers and thermoset polymeric foams. Both types of materials are susceptible to moisture, which they often encounter in real-life uses. Mineral fibers cause skin irritation and present handling difficulty. Polyurethane and melamine resin foams tend to undergo hydrolysis during a prolonged exposure to moisture. In addition, these thermoset foams are not recyclable. Recently, we developed a macrocellular polyolefin foam that absorbs sound well even though the cells are substantially closed-celled. With its outstanding moisture resistance and ready recyclability, the foam material finds numerous applications. The material needs an improvement in the thermal insulation property for use in some demanding applications. Large cells do not block heat transmission by radiation as well as small cells. We have successfully developed an acoustical material that is effective in blocking both heat and sound transmission in addition to being resistant to moisture. The material is an open-cell polypropylene foam with relatively large pores and small cell size. The foam is optionally perforated. Ordinarily, achieving such acoustical foam by the direct extrusion process is difficult since cell opening interferes with foam expansion. The difficulty has been overcome by an optimum selection of the resin, process conditions, and foam and cell structures. The foam product comes in a coalesced strand structure with a relatively low airflow resistance in the extrusion direction. Some of the foams are very flexible with densities as low as 15 kg/m3, while others are semi-rigid. The foams can be exposed to temperatures as high as 160 ˚C under no load condition. With their advantageous properties, the acoustical polyolefin foam products are anticipated to find applications in automotive, appliances and industrial uses as well as in buildings. Session C2-1: Active noise control at the ear (continued) Chaired by Boaz Rafaely Session C2-1, 9:40 h Paper number 406 (Invited, Oral) Paul Kellett, Soo H. James Defence Evaluation and Research Agency, UK HEARING PROTECTION FOR UK MILITARY AIRCREW I-INCE Class 36 Cockpit noise levels in UK military aircraft are steadily increasing and hence greater levels and more ingenious methods of noise reduction are required if aircrew hearing damage risk is to be contained and communications are to remain intelligible during operations. Personal hearing protection in the form of flight helmets, headsets and ear insert devices provide an optimal solution for the majority of noise induced problems - being both cost and weight effective. This paper outlines the problem areas Internoise 2001 Abstracts 231 in the military cockpit and looks at the effectiveness of newer personal protection technologies. A history of the development of Active Noise Reduction at DERA from the early analogue systems and hybrid analogue/digital controllers to fully digital implementations is provided, along with results from both experimental and in-service operational trials. Issues associated with the integration of ANR with other helmet and cockpit systems are also discussed. Predictions of noise levels in the next generation of military aircraft suggest that even more effective solutions will be needed in the future. Results of research into combinations of active noise reduction systems are discussed, as well as the potential limitations to conventional hearing protection devices. Session C2-1, 10:00 h Paper number 436 (Invited, Oral) Christian Carme and Jean-Laurent Peube TechnoFirst, 48 avenue des Templiers, 13676 Aubagne Cedex - FRANCE A DIGITAL SOLUTION FOR AN ANR HEADSET IS: THE FEEDFORBACK I-INCE Class 36 TechnoFirst® compagny started by developing an ANR headset named NoiseMaster™. This standard product is based on an analogue control. Few years ago we have decided to improve this product by reducing some drawbacks as the spillover and to improve also the bandwidth and the level of noise reduction. These improvements were respectively presented in Active 1997 and in InterNoise 1998. Finally the digital ANR solution used for this headset was officially introduced: the FeedForBack™, as a new control principle in Active 99. This new control combines analogue and digital control and improves the two standard solution I mean the feedback and the feed forward. We will describe more in details in this paper what are the benefit to use this technology instead of a regular analogue feedback control or a digital feedback or even a feed forward control. Session C2-1, 10:20 h Paper number 36 (Contributed, Oral) V. Zimpfer, P. Hamery and K. Buck I-INCE Class 36 French German Research Institute of Saint Louis, 5 rue du général Cassagnou, BP 34, 68301 Saint Louis Cedex, France ACTIVE EARPLUG WITH A DIGITAL CLOSED LOOP REGULATION This paper describes the new ANR ( Active Noise Reduction) earplug device using a digital filter in closed loop regulation. Compared to an earmuff, an earplug offers a better passive attenuation at the low frequencies and allows active attenuation at higher frequencies (up to about 2 kHz). Compared to an analog filter, a digital filter allows a fast and easy adaptation to different noise environments and to different users. An electroacoustic device which yields specific requirements for this application (dimension, output pressure...)and a digital filter with a dynamic range of 60 dB are proposed this paper. The loudspeaker consists of a cylindrical piezoelectric ceramic loaded with a thin layer of air. This laterally radiating loudspeaker provides a high volume flow compared to its dimension. To obtain a causal filter compatible, with the short acoustic delay, the filters have been implemented as IIR (Infinite Impulse Response) type filters. The problem with the IIR filter is the propagation of the roundoff noise on the output signal. To minimize the error of the output signal, a new algorithm has been created. Hence this algorithm allows to increase the dynamic range of the filter type. The paper presents experimental results of the active attenuation with the digital ANR earplug. This active attenuation is very satisfying and allows to have better performances than those with any analog ANR hearing protectors. 232 Internoise 2001 Abstracts POSTER Session C2-1 Paper number 584 (Contributed, Poster) Posterboard CP-18 Thursday, Author present 15:00 – 16:00 h Carel Ostendorf HEADPHONE WITH ANTI-NOISE USED FOR SOUND MEASUREMENTS I-INCE Class 36 Cauberg-Huygen Raadgevende Ingenieurs, P.O. Box 480, 6200 AL Maastricht, The Netherlands When performing sound measurements in order to check the soundlevel of for instance a factory, the height of the microphone is very often 5 meters. The soundlevel at this height can be different from the soundlevel on the ground where the technician is standing. Thus, what you hear on the ground is not necessarily the noise you are measuring on 5 meters height. The use of a headphone to listen to the measured signal can solve this problem. Background noise can make it hard to hear the measured sound. The use of a headphone with anti-noise can improve the hearing of the measured signal because the background noise is at least partial taken away. This paper presents this improvement as experienced in different situations. Session C2-2: Practical aspects of active control Chaired by Shiro Ise and Geoff Leventhal Session C2-2, 10:40 h Shiro Ise Paper number 683 (Tutorial, Oral) I-INCE Class 38 Department of Engineering, University of Kyoto, Yoshida-honcho, Sakyo-ku, Kyoto 606-8501, JAPAN PRACTICAL ASPECT OF THE ACTIVE NOISE CONTROL AND ITS POSSIBILITY In 1990's when the significant growth of the digital technique and the cost reduction of the DSP hardware realized, there was a "boom" of the ANC study. Then, we have found that it is not easy to apply the ANC system to the practical tool as a commercial device. At present, shouldn't we clarify the practical problem of ANC? In this paper, we try to find the practical problem of ANC and discuss on the research direction of ANC. Further, a method of estimating the cost and the effect of the ANC in the three dimensional sound field is proposed. Session C2-2, 11:20 h Paper number 215 (Invited, Oral) Jean Laurent Peube and Christian Carme TECHNOFIRST S.A., 48 Avenue des Templiers - 13 676 AUBAGNE - FRANCE OSCILLATING BODY IN A FLOW AS AN AEROACOUSTIC SOURCE I-INCE Class 21 Pressure fluctuations on a wall due to a flow are the source of a noise emission. So, moving a body in a flow, we can get an aeroacoustic source of which time characteristics are in direct relation with the kinematics data of movement. For example, it is easy to produce a rotating oscillation around an axis Internoise 2001 Abstracts 233 with some electrical device. This aeroacoustic source is, to the first order, an acoustic dipole of which direction depends on aerodynamic properties of flow around this body: nearly perpendicular to the flow direction for a profiled wing with an incidence so as to have a strong lift, this dipole will be parallel to this direction for a body with a high drag. Acoustic power of such an aeroacoustic source is proportional to the square of the mean velocity for a given amplitude of vibration. This kind of acoustic source has some interest particularly in flow with velocity high enough, in which loudspeakers cannot be easily settled. It may be used also as a secondary source for active noise control in ventilation ducts or exhausts of thermal engines. Some examples of configuration are discussed with some experimental results. Session C2-2, 11:40 h Paper number 129 (Contributed, Oral) E. Friot, A. Roure and M. Winninger I-INCE Class 38 CNRS-Laboratoire de M´ecanique et d’Acoustique, 31 Chemin Joseph Aiguier, 13402 Marseille Cedex 20, France A SIMPLIFIED REMOTE MICROPHONE TECHNIQUE FOR ACTIVE NOISE CONTROL AT VIRTUAL ERROR SENSORS Recently the Remote Microphone Technique has been implemented in aircrafts to reduce turbofan radiation as well as interior noise at some locations where no error sensors can be put. However, this technique requires much more real-time computations than the usual X-LMS control algorithm; in the multichannel case it was applied to stationary pure-tone noises only. In this paper an algorithm is introduced for broadband control of noise at virtual sensor locations. It combines a model of the transfer matrix from real to virtual sensors with the secondary Frequency Response Functions into a single set of FIR filters. Eventually the algorithm requires roughly the same amount of real-time computations as the multichannel X-LMS algorithm. Numerical simulations of various noise control devices including this Simplified Remote Microphone Technique are given in the paper. Experimental results of multichannel noise control inside a large cavity are also given. Session C2-2, 12:00 h Paper number 434 (Contributed, Oral) I-INCE Class 38 Tsuyoshi Usagawa, Yuichi Ogata, Yoshitaka Nishimura and Masanao Ebata Kumamoto University, Graduate School of Science and Technology, 2–39–1 Kurokami, Kumamoto, 860–8555, Japan MULTI-CHANNEL ACTIVE NOISE CONTROL SYSTEM FOR PERIODIC NOISE USING DXHS ALGORITHM DXHS (Delayed-X Harmonic Synthesizer) algorithm is proposed by authors in order to control the periodic noise. This algorithm does not need any reference signal, and it can track the fundamental frequency of target noise. In this paper, multichannel DXHS algorithm is discussed by both analytical way and computer simulation. Because DXHS algorithm is designed to control the specific harmonic noise, the discussion about the cross-terms on multichannel ANC system can be simplified into an inverse problem for each frequency component. Especially, if the frequency of control system is the same for all channel, it can be simplified very much. The optimum solution can not be obtained in general, however the quasi-optimum solution, or practical solution, for an ANC system can be obtained. As an actual demonstration system, an active noise control system for ambulance siren is implemented as a real-time system. Attenuation obtained by ANC stereo headset is more than 14dB 234 Internoise 2001 Abstracts and the system is stable whether headset is worn or not. Also this ANC system selectively control sound of siren and it does not affect other signal such as speech. Session C2-2, 13:20 h Paper number 52 (Contributed, Oral) P. Hamery and K. Buck I-INCE Class 72 French German Research Institute of Saint Louis, 5 rue du général Cassagnou, BP 34, 68301 Saint Louis Cedex, France EVALUATION OF ACTIVE NOISE REDUCTION HEARING PROTECTORS WITH OBJECTIVE METHODS The evaluation of passive protection devices is usually made with subjective methods. The REAT (Real Ear At Threshold) method (ISO) is normalized and widely used all over the world. It gives a good appreciation of the protection possibilities (Insertion Loss) as long as the device is linear. However, if the protective devices have active parts, this method, using the threshold of hearing, may become too imprecise. The reason for this is that any active device will produce electronic noise that will mask the threshold of hearing, and so lead to imprecise measurements. With increasing level of background noise, this effect will be more pronounced. With objective evaluation methods, like MIRE (Microphone In Real Ear) or using an ATF (Artificial Test Fixture, Artificial Head), the noise level may be chosen in a way, that the background noise has no influence on the measurement of the Insertion Loss(IL). However, the evaluation of ANR (Active Noise Reducing) hearing protectors should not be limited to the measurement of the IL. Other parameters, like stability, behavior at overload and background noise have to be taken into account to fully describe such a system. These parameters are especially important in military environments, where high level continuous and impulse noises may be present. The presentation will propose different procedures for the objective evaluation of ANR hearing protectors taking into account the specificity of the military acoustical environment. Session C2-2, 13:40 h Paper number 304 (Contributed, Oral) Edin Begic and M.O. Tokhi I-INCE Class 74 Department of Automatic Control and Systems Engineering, The University of Sheffield, Sheffield, United Kingdom VOICE ECHO CANCELLATION - A CRITICAL OVERVIEW This paper presents an overview of echo cancellation techniques and discusses issues and concerned associated with the developed techniques. A description of the phenomenon of echo and its causes in telephone systems are described. A comparative assessment of finite-impulse response (FIR) and infinite-impulse response (IIR) filter structures in the identification of echo path is provided. The emphasis in this study is between time domain least-mean squares (LMS) and recursive least-squares (RLS) based set of commonly used algorithms. The excitation signals utilised include Gaussian white noise, female speech and male speech. The performance indices utilised in the comparative are those as recommended in ITU G167. Certain aspects of the ITU G167 recommendation are discussed and corresponding suggestions made. The overview, additionally, includes frequency domain cancellers, subband cancellers, affine projection algorithm, non-linear cancellers as well as double talk detectors (DTD) and time delay estimation (TDE). Internoise 2001 Abstracts 235 Session C2-2, 14:00 h Paper number 499 (Contributed, Oral) M. Trimpop and D. Krahé Institut für Lärmschutz, Arnheimer Straße 107, D-40489 Düsseldorf, Germany I-INCE Class 79 EVALUATION OF A SYSTEM FOR ACTIVE SOUND REDUCTION ON THE BASIS OF HEARING TESTS AND PSYCHO-ACOUSTICAL MEASURING METHODS Noise reduction systems are usually described by a measure of damping, response time and other characteristic values. The system examined in this paper was already presented in detail on the basis of such values in the last years. This system analyses a given primary sound field at the border of a defined closed area using the generalised Kirchhoff integral. It generates a secondary sound field by sources located in the proximity of the border, that cancels the primary field in the closed area by superposition. However, measured values like the average damping can only slightly express the improvement heard by a human. Therefore different sound signals (artificial signals and real life signals) were used as primary sources in a simulation of the system. The secondary fields and the superposed fields were calculated with different parameters and in different hearing positions to evaluate the effects by hearing and more hearing related measuring methods. The results will be presented and compared with the former ones. This should make clear, which kind of optimisation based on objective rules leads to best result in hearing. POSTERS Session C2-2, 14:20 h Paper number 244 (Contributed, Poster) I-INCE Class 74 Posterboard CP-19 Thursday, Author present 15:00 – 16:00 h Kenji Muto, Guoyue Chen, Kikuo Asai, Yugi Sugimoto, Kiyohiro Yuki and Kazuo Shimada Tokyo Metropolitan College of Aeronautical Engineering, 8-52-1 Minamisenju, Arakawa-ku, Tokyo, 116-0003, Japan AN ADAPTIVE ALGORITHM WITH VARIABLE STEP GAIN FOR MULTI-CHANNEL ECHO CANCELLER The sound effect is very important in a remote education using a teleconference system. Especially, it is indispensable that the echo is estimated fast so as not to influence the conversation. We study of the presence communication system using virtual environmental system with the multi point connection. A problem is to improve the speech quality of the echo canceller of the multi-channel. In case of a multi-channel teleconference system, one channel signal becomes the noise for the other channel signal. Therefore, when the voices are given to the multi-channel at the same time, the estimation of the echo becomes difficult. In this report, we describe multi-channel echo canceller constructed using the adaptive algorithm and the FIR filter. We have proposed the adaptive algorithm in which step gain (filter coefficients are corrected using this gain) is sequentially corrected for the fast estimation of the FIR filter coefficients. We solve this problem of multi-channel echo canceller by our proposed algorithm. 236 Internoise 2001 Abstracts Session C2-2, 14:20 h Paper number 363 (Contributed, Poster) I-INCE Class 38 Posterboard CP-20 Thursday, Author present 15:00 – 16:00 h Yoha Hwang, Jong Min Lee, Kyung-Mok Kim, Seung-Jong Kim and Yong-Bok Lee Korea Institute of Science & Technology, Korea NEW ACTIVE MUFFLER SYSTEM USING VARIABLE LENGTH BYPASS EXHAUST PIPES AND A VERY LOW BACK PRESSURE MUFFLER A unique muffler system is proposed and its noise reduction performance and great potential for better engine power and fuel consumption is demonstrated with an engine experiment. The proposed muffler system uses transmission path difference of divided exhaust pipes to reduce major exhaust noise component. In this system the exhaust pipe is divided into two parts and joined again downstream. Exhaust noise is reduced by destructive interference when two divided noise join again with transmission paths' difference which is half of the wavelength of a main noise frequency. One divided pipe has a sliding mechanism to vary its length to adjust to rpm change. The system also has a simple structure low back pressure muffler to remove remaining wideband noise. It was tested with a real engine and result shows comparable performance with passive muffler system thereby proving its potential resulting from low back pressure design. Session C2-2, 14:20 h Paper number 588 (Contributed, Poster) Posterboard CP-21 Thursday, Author present 15:00 – 16:00 h Thomas Sors and Stephen Elliot Industrial Research Limited, PO Box 2225, Auckland, New Zealand I-INCE Class 38 MULTIPLE SINGLE CHANNEL CONTROL SYSTEMS FOR ACTIVE STRUCTURAL ACOUSTIC CONTROL A method of controlling the vibrations and sound radiation from a plate by means of multiple single channel feedback control systems is introduced. Each control system consists of an accelerometer, a simple feedback controller and, either a point force input or piezoceramic actuator. The effects of using these simple control systems on both the vibration levels of the plate and the sound radiation are analysed. The results are introduced for point force inputs, which would result in true collocated control systems, but the paper concentrates on the more practical arrangement using piezoceramic actuators. The number of control loops is also varied to observe the effects on control performance. The technique holds promise for the future as the price of transducers falls and as microelectromechanical systems (MEMS) are developed. Internoise 2001 Abstracts 237 Session C2-2, 14:20 h Paper number 638 (Contributed, Poster) I-INCE Class 79 Posterboard CP-22 Thursday, Author present 15:00 – 16:00 h Maria De Diego, Alberto Gonzalez, Gema Pinero, Miguel Ferrer and J.J. GarciaBonito Dept. of Comunicaciones. Universidad Politecnica de Valencia. Spain SUBJECTIVE EVALUATION OF ACTIVE NOISE CONTROL TECHNIQUES APPLIED TO ENGINE NOISE The final objective of active noise control inside a vehicle should be to improve the comfort sensation. In this work it is analyzed the subjective sensation caused by actively controlled interior car noise and also to get some valuable information in order to predict its behaviour in general cases. Two strategies of noise evaluation are used: a jury test and a predictive comfort model. Different engine noise signals are presented to a group of 30 volunteers in order to evaluate how the comfort level have been modified. The prediction model of comfort is based in four psychoacoustical descriptors (loudness, sharpness, tonality and roughness). The local active noise control system is tested in a listening room with two secondary sources and two error sensors mounted on the headrest of a typical car seat. Recordings are made using a Head Acoustics mannequin positioned in the seat with two calibrated microphones at the ear-canals. Results show that not all the psychoacustical parameters are directly controlled by ANC and in some cases this can affect to satisfactorily predict the pleasantness of engine noise. Jury test validates ANC performance. It can be concluded that ANC techniques can be considered as a useful method to minimize noise levels. Session C2-3: Occupational noise Chaired by P. Novak Session C2-3, 14:40 h Paper number 251 (Contributed, Oral) Jukka Starck, Esko Toppila and Ilmari Pyykkö I-INCE Class 62 Finnish Institute of Occupational Health, Topeliuksenkatu 41, 00170 Helsinki, Finland HOW TO IMPROVE THE EFFICIENCY OF HEARING CONSERVATION PROGRAM? Hearing conservation programs (HCP) focus on occupational noise in work life and also to a wide range of environmental noise which all should be included in the evaluation of the total noise exposure. The exposure of the inner ear to noise is also dependent on the use of hearing protectors. The highest risk exists in occupations in which workers are exposed to impulsive noise. Properly selected HPDs should decrease the noise levels at ear below the risk limit in industrial environments. However, according to our experience the usage rate of hearing protectors in industry is often low varying from 55% in paper mill to 90% in forestry work. Contributing factors like smoking, use of painkiller, high blood pressure and cholesterol explain the large variation in the hearing loss. Moreover there seems to be a dependency of the age. In risk analysis the contributing factors partly masked the effects of noise in the development of hearing loss (HL). Factors independently but causally related to age, were important in the development of NIHL among workers exposed to noise levels below 98 dB(A). The inner ear in older subjects seems to be more vulnerable to noise than those in younger ones. HCP actions can be improved if we can improve the risk assessment and the validity of exposure evaluation. The urgent task is to develop expert programs and models, which include all 238 Internoise 2001 Abstracts risk factors i.e. individual susceptibility and life time exposure evaluation including leisure time and military service exposure. Session C2-3, 15:00 h Paper number 585 (Contributed, Oral) Franci L.H. Vanweert and Theo M. De Graaf I-INCE Class 62 Cauberg-Huygen R.I. BV, P.O. Box 480, NL-6200 AL MAASTRICHT, The Netherlands HARMFUL NOISE IN THE WORKPLACE OF THE DUTCH PAPER, CARDBOARD AND CORRUGATED CARDBOARD INDUSTRY: AN INVENTORY OF THE CURRENT SITUATION AND THE OPTIONS TO REDUCE NOISE The Dutch paper, cardboard- and corrugated cardboard producing companies and – processing companies are planning to come to an convention with the Ministry for Social Affairs and Employment about several job risks including the risk of harmful noise. The purpose of the convention is to make agreements that acuminate the actual circumstances of the companies in order to optimise improve the noise circumstances in the workplaces. With the preparing of this convention there has been research with a bipartite purpose. At fist the current situation of harmful noise is investigated in the hole sector. This inventory relates to approximately 160 companies in which approximately 20.000 people work. In the second part of the research all the possible and reasonable measures will be dated and the state of technology will be determined. The measures research is based on practical experience within companies and supplies of machines and materials, study of literature and internet-screening. Session C2-3, 15:20 h André L'Espérance Soft dB inc. , PQ, Canada Paper number 575 (Contributed, Oral) I-INCE Class 75 SABINE'S HIGH ORDER METHOD FOR ACOUSTICAL PREDICTION IN ROOMS Actual prediction models are sometimes too approximate or too complex to be useful to noise specialists or engineers who want to evaluate the effect of acoustical treatments in workspaces. To fill this gap, a new method based on Sabine’s method has been developed. This method consists in determining the contributions of the direct field and of the first orders of reflections to estimate the sound pressure at the receivers locations. The image theory is used to estimate the first reflections on each wall of the room. The residual energy that has not been considered in the first order contributions is then distributed on the receivers according to the energy distribution at order n. This computation approach allows to obtain the noise map of the room in a few seconds or minutes. The results obtained with the new model will be compared to results obtained with the ray-tracing model and also to experimental results. To make this acoustical prediction model easy-to-use, a user-friendly graphical interface under Windows environment has been developed. It allows to define or modify the geometry and data that are necessary for modeling. Effect of absorbent panels on the ceiling or on the wall as well as the effect of acoustical barriers may be considered. Special modules have also been included in the model to compute the noise exposure of workers depending on their exposition time at different locations in the rooms. The users of the new model may thus analyze the effect of various treatments not only on the sound level in the room, but also on the reduction of the noise exposure. Internoise 2001 Abstracts 239 Session C2-3, 15:40 h T. Ward Paper number 87 (Contributed, Oral) I-INCE Class 88 Health and Safety Executive, Magdalen House, Stanley Precinct, Bootle, Merseyside, L20 3QZ, UK COMPLIANCE WITH NOISE AT WORK REGULATIONS - A TARGETED ENFORCEMENT APPROACH In the UK, 1.3 million workers are exposed to noise above levels which could damage their hearing, and 170,000 people consider that they suffer hearing loss or other ear conditions caused by their work. Therefore noise at work is still a priority issue for the Health and Safety Executive (HSE), the body responsible for ensuring that risks to people's health and safety from work activities are properly controlled. Health and safety inspectors continue to be active in seeking and enforcing compliance with legislation to protect workers’ hearing. During 2000/01 HSE is undertaking a ‘Single Issue Inspection’ programme on noise. Inspectors will visit 5,000 workplaces in industries where noise is likely to be an issue, targeting their inspection on ensuring compliance with the Noise at Work Regulations. The main areas addressed will be employers’ duties to assess noise exposure, to provide, maintain and ensure use of noise control equipment and hearing protection, and to give adequate information, instruction and training to employees. Where employers are not meeting their obligations, a mixture of formal enforcement and advice will be used. Follow-up visits will be made to ascertain progress on the main issues. The primary result of the programme will be a reduction in the number of workers exposed to a risk of hearing loss. But important information will be gathered on the effectiveness of a targeted approach. Analysis of the level of compliance at the follow-up will provide information on the effectiveness of formal enforcement when compared with an advisory approach. Session C3-1: Hearing loss by leisure activities Chaired by Yo-iti Suzuki and Willy Passchier-Vermeer Session C3-1, 9:40 h Paper number 40 (Invited, Oral) I-INCE Class 62 Hiroaki Takinami, Katsufumi Kuroda, Shoji Aono and Koichi Takagi RION Co., Ltd.: 3-20-41 Higashimotomachi, Kokubunji, Tokyo, 185-8533 JAPAN ESTIMATION OF TTS DUE TO AMUSEMENT SOUNDS BY USING TTS METER Recently there has been increasing concern about the effects of sounds to which we are exposed in amusement activities. In order to evaluate the hearing-effect, we have developed a system which predicts TTS (temporary threshold shift) in real-time operation at actual sound induced fields. The system consists of a combination of a sound level meter with one-third-octave band filters and a multipurpose personal computer. In this investigation, TTS values of amusement sounds were estimated by using this system (TTS meter). Twenty kinds of amusement sounds were considered as sound sources; game arcade (4), pachinko parlor (2), bowling alley (3), karaoke saloon (2), speedboat-racing pool (2), football stadium (2) and baseball stadium (5). The result of the estimation on TTS was that some sounds could produce so large TTS that those hearing-effects should not be ignored. Especially, one of pachinko parlors and two of baseball stadiums could produce larger than 20 dB of TTS at the test frequency of 4 kHz. 240 Internoise 2001 Abstracts Session C3-1, 10:00 h Paper number 140 (Invited, Oral) I-INCE Class 62 M.R. Serra, E.C. Biassoni, U. Richter, A.J. Carignani, S. Joekes, M.R. Yacci, G. Minoldo, S. Abraham, G. Franco and A.R. Pollet Centro de Investigación y Transferencia en Acústica (CINTRA) - Universidad Tecnológica Nacional FRC, Mtro. M. López esq. Cruz Roja Argentina, 5016 Córdoba, Argentina HIGH-LEVEL MUSIC EXPOSURE AND HEARING EFFECTS IN ADOLESCENTS: THIRD YEAR OF A LONG-TERM STUDY In our country, Argentina, a high percentage of young people are rejected by hearing loss in the preoccupational check up, without having physiological antecedents. Most of that youth is exposed to high level music during their recreational activities by attending discotheques and life concerts or by using walkman or diskman. The literature reveals different opinions about possible harmful hearing effects of music. This paper deals with the first three years of an interdisciplinary long-term study with adolescents about music exposition and auditory behaviour in conventional and extended high frequencies ranges, taking into account: a) physical and psychosocial variables, b) sound immission levels in the discotheques more visited by the adolescents and by using walkman or diskman. It is developed applying national and international standards. Two groups of adolescents - males and females - are examined along four years of their high school level, from 14 to 17 years old. The results of the three first years of study show some differences between boys and girls, revealing: a) changes in the recreational habits, b) increase in high level music exposition, c) hearing threshold shift in relation to higher music exposition, d) high sound immission levels in discotheques and through personal music devices. One year more of study let will us to define the behaviour of the variables studied. Session C3-1, 10:20 h Paper number 145 (Invited, Oral) I-INCE Class 62 Esko Toppila, Heli Laitinen, Pekka Olkinuora, Kaarina Kuisma and Katja Perälä Finnish Institute of Occupational Health, Topeliuksenkatu 41, 00250 Helsinki, Finland DEVELOPMENT OF HEARING CONSERVATION PROGRAM FOR FINNISH NATIONAL OPERA Purpose of the study was to develop a hearing conservation program for the whole opera personnel. Methods: The exposure measurements were performed among the orchestra, singers, dancers, choir, and technical personnel. The needs and the attitudes towards hearing conservation were questioned. Also a technical inspection on the premises was performed. Results: For technical personnel the radio communication system could cause a considerable exposure. The orchestra string players were exposed more during group rehearsals and performances than during personal training. For drummers and horn players the situation was the other round. Singers and choir members were exposed during personal rehearsals (100 dB) more than during performances. For dancers the exposure was always below 85 dB(A). According to the questionnaire the opera personnel was worried about their hearing but did still not use hearing protectors. People had difficulties in getting used to them. Conclusion: Based on the results a hearing conservation program was developed. It consisted of information about hearing loss, training programme how to get used with hearing protectors and continuous inspections as a method to avoid unnecessary exposure and to improve the working conditions. Internoise 2001 Abstracts 241 Session C3-1, 10:40 h Martin Liedtke Paper number 169 (Invited, Oral) I-INCE Class 62 Berufsgenossenschaftliches Institut für Arbeitssicherheit - BIA, Alte Heerstraße 111, 53754 Sankt Augustin, Germany DETERMINATION OF SOUND IMMISSIONS FROM SOURCES PLACED CLOSE TO THE EARS SUCH AS HEAD- AND EARPHONES When determining sound immissions from head- or earphones used to reproduce music or speech the problem arises, that procedures usually used for the determination of sound immissions cannot be applied: The sound pressure levels measured at the position of the exposed person but with the person absent do not adequately represent the sound exposure. Therefore miniature or probe microphones inserted in the ears of humans (Microphone in Real Ears, MIRE-technique) or manikins (maninikintechnique) are used. But sound pressure levels measured in the ear canal of exposed humans or exposed manikins do not bring about the assessment of the exposure by means of well established criteria. This assessment is enabled by the conversion of the measurement result into a corresponding free-field or diffuse-field level. This level is virtual - but only such a level can be compared to levels given by regulations. For comparable results obtained by measurement and conversion working group 6 of ISO Technical Committee 43 "Acoustics" elaborated drafts of standards for "Determination of sound immissions from sound sources placed close to the ears": - DIS 11904 - 1: 2000: "Technique using microphones in real ears" and - CD 11904 - 2: 2000: "Technique using a manikin". These standards may be applied to equipment tests, determination of noise exposure at the work place or other exposure from sound sources close to the ears. Session C3-1, 11:00 h Paper number 492 (Invited, Oral) I-INCE Class 62 Stephen E. Keith, Stephen H.P. Bly, Vincent Chiu and Robert G. Hussey Consumer and Clinical Radiation Protection Bureau, Health Canada, 775 Brookfield Rd. 6301B, Ottawa, Ontario, K1A 1C1, Canada SOUND LEVELS FROM HEADPHONE/PORTABLE COMPACT DISC PLAYER SYSTEMS III. Due to the large population of youths and children using personal stereo players, there is potential for noise induced hearing loss and tinnitus, even if the risk is small. Information is needed on the factors affecting sound levels from these devices to help determine the most appropriate strategies for reducing the risk. This study extends two previously published studies from our laboratory on the range of sound exposures available from players sold in Canada. These earlier studies showed that, even at maximum volume settings, large variations in sound level arose from a variety of factors, including: (i) type of CD player, (ii) type of headphone, (iii) how well the headphone sealed against the ear, and (iv) program material. New data is presented on the effects on 1/12 octave band and Aweighted sound levels due to the use of: (i) soft artificial pinnas on the measurement device(a Bruel and Kjaer head and torso simulator), (ii) insert headphones and (iii) program material that consisted of the top ten albums of 1998, and the ten albums containing the top ten singles of 1998. The maximum equivalent free field sound levels ranged between approximately 65 and 115 dB(A). The results of this study lend further support to the conclusion of previous studies from this laboratory. To reduce the risk of noise-induced hearing loss and tinnitus, users need to be educated about subjectively based criteria for limiting noise exposure. 242 Internoise 2001 Abstracts Session C3-1, 11:20 h Paper number 535 (Invited, Oral) Takashi Kawamura, Masayuki Misaki and Tadashi Abe I-INCE Class 62 Multimedia Development Center, Audio Technology Group, 1006 Kadoma, Kadoma City, Osaka, 5718501 JAPAN PORTABLE AND PERSONAL AUDIO PLAYERS, AND SOUND ENVIRONMENTS OF THEIR CUSTOMERS The personal and portable audio players using various media such as compact cassette, CD and MD have been widely used especially among young persons. Those devices offer music sound listeners through headphones. Since those peoples have usually enjoyed music in the outdoor, they have faced the problem of quite poor sound environment such as the big noise of underground. As a result, the listening music level through headphones often have to be considerably increased to overcome the outcoming noise. Various unique technologies have been developed to offer far comfortable listening conditions of portable audio device. For instance they are open-aired type head units, bass booster, Automatic Level Controller and anti-sound-leak head units. The impact to human listening ability caused by such high level sound listening was surveyed. First using several different models of portable audio player, those maximum capable sound levels had been measured. It had been confirmed that the maximum level is well controlled not to exceed 100 dBA, which is the criteria settled by the French Government Authority for Healthy. The further survey upon TTS (Temporary Threshold Shifting Åc.. temporary shift of hearing ability threshold due to excess sound level) had been reported. It had shown some degree of the impediment in hearing (mainly TTS), but not indicated as the serious permanent shift. It can be concluded that the widely popular portable audio player might cause some degree of TTS while the headphones are used, but it must be free of permanent impediment in hearing ability. Session C3-1, 11:40 h Paper number 551 (Invited, Oral) I-INCE Class 62 Heli Laitinen, Yôiti Suzuki , Kentarô Iwata, Erkki Björk, Pasi Myyryläinen and Jaana Jokitulppo Finnish Institute of Occupational Health, Topeliuksenkatu 41 a A, 00250, Helsinki, Finland COMPARISON BETWEEN NOISY LEISURE ACTIVITIES OF JAPANESE AND FINNISH YOUNG PEOPLE Relation between noise-induced hearing loss (NIHL) risk, problems with hearing, and leisure time activities among young people has been researched quite extensively. Most of the studies agree that the leisure noise exposure, and hearing problems as well, have increased, and that young people should be educated about the risks involved. University of Kuopio in co-operation with Finnish Federation of the Hard of Hearing has made a questionnaire about leisure time noise exposure including music, hearing problems and usage of hearing protectors to 1069 (60 % male and 40 % female) Finnish young people of ages between 13 and 25. The correspondents were students of junior high school, senior high school, vocational school, polytechnic, or university. The present study focuses how NIHL risk, problems with hearing, and leisure time activities depend on cultural background. Therefore, the same questionnaire was translated into Japanese so that comparable data can be obtained. The questionnaire is distributed to corresponding age distribution and school types of Japan. These two data pools are compared and thus the study gives us cross-cultural information on young people's noisy leisure time activities, hearing problems, and usage of hearing protection in two Internoise 2001 Abstracts 243 different cultural surroundings. Preliminary examination has suggested that the consciousness of the risk of leisure activities to hearing are different between the two countries. Session C3-1, 12:00 h Paper number 124 (Contributed, Oral) Jaana Jokitulppo and Erkki Björk I-INCE Class 62 Finnish Institute of Occupational Health, P.O. Box 486, FIN-33101 Tampere, Finland ESTIMATED LEISURE-TIME NOISE EXPOSURE - A FINNISH QUESTIONNAIRE SURVEY Leisure-time noise is an environmental problem that is often uncontrolled by regulations. However studies from several countries indicate an increased incidence of hearing damage among teenagers as a result of recreational exposure to noise. Information is lacking, nevertheless, on the overall leisuretime noise exposure of the general public. Therefore, a Finnish research project was initiated to gather data on the total weekly exposure of the Finnish population to leisure-time noise. The time devoted to noisy leisure-time activities, the self-evaluated loudness of the activities, and hearing symptoms (i.e., tinnitus, pain in the ear and hearing loss due to noise exposure) were studied. A questionnaire survey on weekly leisure-time activities among different age groups of the Finnish population was used for this purpose, and no measurements of sound levels or hearing loss were attempted. The SPSS program was used for the analysis of the data. Weekly noise levels of over 85 dBA were estimated for 9% of the subjects aged 25-55 years (n = 1323) and 51% of those aged 12-16 (n = 405) years. The incidence of hearing symptoms seemed to correlate with age and the weekly dose of leisure-time noise. The noisiest leisure-time activities were going to nightclubs, using home tools, playing in a band, shooting firearms and attending motor sport events. It was concluded that people who have several noisy leisure-time activities and spend a great deal of time in these activities, and also possibly have a noisy job may be at risk of hearing loss. Session C3-2: Effects of low frequency noise Chaired by Kerstin Persson Waye Session C3-2, 13:20 h Paper number 484 (Contributed, Oral) I-INCE Class 63 Kerstin Persson Waye, Angela Clow, Frank Hucklebridge, Phil Evans and Ragnar Rylander Department of Environmental medicine, Göteborg University, Box 414, 405 30 Göteborg, Sweden EFFECTS OF NIGHT TIME LOW FREQUENCY NOISE AND TRAFFIC NOISE ON CORTISOL RESPONSE TO AWAKENING Background: While sleep disturbance due to traffic noise is well documented, much less is known of sleep disturbance due to low frequency noise (<200 Hz).The cortisol response to awakening is a reliable index of adrenacortical activity. The normal pattern of cortisol displays a peak shortly after wake up. As noise may disturb the normal pattern of sleep, it was of interest to evaluate if the normal pattern of cortisol was influenced by noise during sleep. The aim was to evaluate effects of traffic noise (TN) and low frequency noise (LFN) on sleep as measured by a change of the saliva cortisol pattern after wake up and subjective evaluations. Methods: 12 male subjects slept during 5 consecutive nights in a noise laboratory. After two quiet habituation nights, followed an exposure night with either LFN or TN, night 4 was a quiet reference night (RN) and night 5 either LFN or TN. The noise levels 244 Internoise 2001 Abstracts were 40 Laeq for LFN and 35 Laeq, 50 dBmax for TN. Saliva samples for cortisol were taken directly after wake up and at 15 minute intervals during the first 45 minutes. Questionnaires were answered in the morning and evening. Results: Preliminary analyses show a normal pattern of cortisol after the RN. The pattern of cortisol after TN was somewhat attenuated at 15 to 45 minutes. After LFN the normal peak at 30 minutes was significantly attenuated. The main subjective finding was a longer time to fall asleep during the LFN. The implications of the results will be discussed at the conference. Session C3-2, 13:40 h Paper number 299 (Contributed, Oral) I-INCE Class 63 Johanna Bengtsson, Kerstin Persson Waye, Ragnar Rylander and Anders Kjellberg Dep of Environmental Medicine, Göteborg University, Box 414, 405 30 Gothenburg, Sweden EFFECTS ON TIREDNESS AND MOTIVATION AFTER LOW FREQUENCY EXPOSURE DURING WORK Many working environments contain noises dominated by the frequency range 20 to 200 Hz, low frequency noise. The noise is emitted from ventilation/air condition systems but also by the lower attenuation by the walls, floor and ceiling. Symptoms reported in connection with annoyance due to low frequency noise and which could reduce a person's working capacity are fatigue, headache and irritation. Previous study (Persson Waye et al, 2000) used demanding tasks and the importance to work as fast and correctly as possible were emphasised. This study aimed to study the possible interference of low frequency noise and tiredness, motivation, annoyance and performance. Thirtyeight test persons, categorised as high-sensitive to low frequency noise, worked in a noise environment for 4 hours with five different performance tasks, of routine-type character, and were instructed to work in their own pace as if they were to work with the tasks for a whole working day. The noise environment were either dominated by low frequencies or dominated by middle frequencies, both at a level of 45 dB LAeq. The effects were evaluated as changes in subjective reactions using questionnaires and changes in performance. Preliminary results showed that the subjects reported a higher degree of annoyance and work impairment when working in low frequency noise. They also reported a higher increase of tiredness and irritation. Preliminary results further give some indications on reduced efficiency in performance over time, during exposure to low frequency noise. The results will further be discussed during the conference. Session C3-2, 14:00 h Paper number 379 (Contributed, Oral) Yukio Takahashi and Yoshiharu Yonekawa I-INCE Class 49 National Institute of Industrial Health, 6-21-1, Nagao, Tama-ku, Kawasaki, 214-8585, Japan RELATIONSHIP BETWEEN VIBRATORY SENSATION AND THE HUMAN BODY VIBRATION INDUCED BY LOW FREQUENCY NOISE The human body vibration induced by low frequency noise (the 'noise-induced vibration') was measured on the body surface. The relationship between the body surface vibration and vibratory sensation was examined and it was found that the acceleration levels of the vibration measured on the chest and the abdomen were sufficiently correlated with the vibratory sensation on each corresponding position. It was also found that the correlation was independent of frequency. These findings suggested that the vibratory sensations on the chest and the abdomen were perceived by detecting the mechanical vibration on the body surface. With respect to the acceleration level of the vibration Internoise 2001 Abstracts 245 measured on the head, on the other hand, close correlation with the vibratory sensation on the head was not found. In this case, the frequency-dependence in the correlation was more sufficient than the cases of the chest and the abdomen. This finding suggested that the mechanism to perceive the vibratory sensation on the head was different from ones on the chest and the abdomen. Session C3-2, 14:20 h Paper number 77 (Contributed, Oral) Frits (G.P.) Van den Berg I-INCE Class 62 Science Shop for Physics, University of Groningen, Nijenborgh 4, 9747 AG Groningen, the Netherlands TINNITUS AS A CAUSE OF LOW FREQUENCY NOISE COMPLAINTS Low Frequency Noise (LFN) exposure in dwellings may cause severe health problems, especially problems related to sleep. Although a multitude of possible LFN sources have been proposed, actual sources cannot always be identified. In a routine procedure we investigate the sound as recorded by complainants themselves at a time they positively stated the annoying LF sound was audible. In several cases the sound levels at all low frequencies (< 200 Hz) were so low that it was improbable or even impossible that LF sounds could be heard at all. In other cases there was -in principle- audible sound, but without an apparent relation to the source of complaints. In these cases the source of complaints seems not to be a sound in the dwelling, but a sound originating within the complainant. This sound, most likely caused by blood flow or tinnitus ('ringing' or 'buzzing' in ears), apparently sounds so much like a "real" sound that complainants find it hard or impossible to believe that it is not coming from some machine. The fact that it is not recognised as tinnitus may be the reason that this type of tinnitus (i.e. at low frequencies) has received little attention, and practical investigations have focused on finding real sound sources. A complication may be that tinnitus perhaps is influenced by real sounds. The fact that LFN complaints can be caused by two very different kinds of sources -real machines or tinnitus- should be considered when trying to solve complaints. Session C3-2, 14:40 h Paper number 615 (Contributed, Oral) Gert Lassche and Ferry Koopmans I-INCE Class 69 Adviesbureau Peutz & Associés B.V., P.O. Box 66, NL-6585 ZH MOOK, The Netherlands LOW FREQUENCY NOISE CRITERIA: FOUNDATION, APPLICABILITY AND CONSEQUENCES. In recent years there is an increasing attention to nuisance attributed to low frequency noise. In order to decide if there is in practice a low frequency noise problem and moreover to decide if noisereducing measures are necessary, there should be an objective criterion. Key question is whether it is possible to formulate a criterion that deals in an appropriate way with both the interests of the disturbed persons as well as the interests of the possible cause of the low frequency noise. There are some criteria that try to deal with the low frequency noise problems but each criterion has its own characteristics and imperfections. Using some real cases as illustration the available criteria will be evaluated. Doing so, the requirements of an adequate assessment criterion will be discussed. One of the basic assumptions is the low frequency hearing threshold. Due to the fact that not many experimental data on hearing thresholds in the low frequency area are available, most of the mentioned thresholds are based on experimental data in the normal audible frequency area. The perception of low frequency noise however differs from the perception of ‘normal sound’ and therefore the ‘rules’ of normal sound will not automatically apply for low frequency noise. It is for instance difficult to 246 Internoise 2001 Abstracts distinguish between the perception of airborne low frequency noise and vibrations. Another difficulty of low frequency noise is that the perception is very individual. This makes a general criterion hardly possible. POSTER Session C3-2 Paper number 152 (Contributed, Poster) I-INCE Class 62 Posterboard CP-23 Thursday, Author present 15:00 – 16:00 h Rauno Pääkkönen, Ilona Valtanen, Jouko Ruotsalainen and Jaana Jokitulppo Tampere Regional Institute of Occupational Health, P.O.Box 486, FIN-33101 Tampere Finland LOW-FREQUENCY NOISE IN A PAPER MILL Persons working in a paper mill complained of non-specific symptoms in association with the noise properties of the mill. Infrasound or low-frequency noise was thought to be the source. The literature contains only a few reports of similar circumstances, and therefore it was decided to measure the noise at the paper mill and survey workers. The level of low-frequency noise was very high and in many areas it was more than 100 dB in the frequency range of 100-300 Hz. The noisiest machines were the sucking pumps, cone grinders, compressors, and vertical pumps in the cellar of the paper mill. The highest sound pressure level was 126 dB at a frequency of 125 Hz, measured near the sucking pumps. The interviews covered 21 persons who answered the questionnaire. The commonest symptoms were hearing disorders (57%), sleep disorders (38%), headache (29%), arrhythmia (29%), equilibrium disorders (24%) and musculoskeletal disorders (19%). Tinnitus can indicate a lack of hearing protection, and, especially in the low-frequency range, hearing protection is not so effective as it should be. The cause of the arrhythmia, headaches and equilibrium disorders can be body resonance entering the chest and internal organ areas during exposure to low-frequency noise. Headache can be initiated by many other causes also, for example, heat, improper ergonomic conditions and so on. In conclusion, the connection between noise properties and symptoms remains somewhat open. Session C4-1: Industrial noise Chaired by J.Salvador Santiago Session C4-1, 9:40 h Paper number 179 (Contributed, Oral) Cheolung Cheong, Jewook Ryu and Soogab Lee I-INCE Class 14 Center for Environmental Noise & Vibration Research, School of Mechanical and Aerospace Engineering, Seoul National University, Seoul, 151-742, Korea DEVELOPMENT OF NOISE PREDICTION PROGRAM FOR HEAT RECOVERY STEAM GENERATOR It is usually the contractual responsibility of HRSG (Heat Recovery Steam Generator) supplier to limit combustion turbine exhaust noise at cogeneration sites. Thus, it is necessary for the supplier to predict the noise level from HRSG at the stage of preliminary design. HRSG is usually composed of inlet duct, main casing, outlet duct and stack. To satisfy the noise limit level, additional equipments are Internoise 2001 Abstracts 247 sometimes required; duct shroud, silencer. We develop algorithms for predicting the noise emission from all HRSG units. The main algorithms are as follows; the transmission loss of duct wall, the insertion loss of tube bundles and duct silencer, the attenuation through outdoor sound propagation and so on. For the convenience of user, we develop the window version program, named NP-HRSG (Noise Prediction for Heat Recovery Steam Generator). To evaluate the accuracy of this program, predicted noise levels from a real HRSG model are compared with measured data. Through this comparison, we observe that the maximum difference of overall SPL (Sound Pressure Level) between the predicted and measured is just about 3dB. Session C4-1, 10:00 h Jan-Willem Bulder Paper number 586 (Contributed, Oral) I-INCE Class 53 Cauberg-Huygen Consulting Engineers, PO Box 9222, 3007 AE ROTTERDAM, The Netherlands FROM ZONE-MANAGEMENT TO AREA-MANAGEMENT In the Netherlands there are a number of industrial estates where big noise pollutants are accepted. These estates are especially nominated in accordance with the law for noise pollution. In addition to this, there are noise zones that are described in the council-plans for the area surrounding each industrial estate. Purpose of these zones is to separate housing and industrial areas. On one hand the zone protects the industrial area, as only a certain number of houses are allowed in this zone. On the other hand the zone protects the housing area, as the level of noise produced by industry in this zone must not exceed the maximum described level. Every industrial area has a zonekeeper. Whenever a new activity on or a new industry is brought into the area the zone keeper will determine if the noise levels that come with these activities can fit in with the already existing noise levels. If they cause the total noise level to exceed the maximum allowed noise level, no permit will be given out by the governing body, which acts according to the law for Environmental management. The zookeeper can also determine the best place for new noise pollutants to establish their new position on the estate. There are also industrial areas that don’t have a noise zone as described but yet produce to much noise. It is recommended for those estates to develop a noise control plan like the existing zoneplans. Plans like this are not described by law but they do provide some advantages: the available noise space can be divided. In a better way, cumulative target values can be described so to protect the surrounding housing estates, whenever permits are given out for new activities on the estate, it is possible to determine if they fit in regarding the noise level they produce. For a noise control plan to have any effect, both the governing body and the industry must recognise it and act according to it. To establish this everyone involved (the governing body, zonekeeper, industry and houseowners) will have to be informed on the matter, in order to have an open discussion about it. Session C4-1, 10:20 h J. Richards Paper number 762 (Contributed, Oral) I-INCE Class 53 M.W. Kellogg Ltd., Greenford, Middlesex. UK NOISE DESIGN ENGINEERING OF A LARGE INDUSTRIAL PLANT - MEASURED EQUIPMENT NOISE LEVELS A number of petrochemical plants have been constructed in Norway over the past 20 years. These plants are complex in design, consisting of a large number of pipes and many pieces of noise generating equipment. Such plants require careful design with regard to noise to ensure that they fulfil 248 Internoise 2001 Abstracts the requirements for neighbourhood noise, as well as for the working environment. A night time limit of 40dBA at the nearest dwelling is often the governing factor in Noise Control Design. Modifications and extensions to existing plants can provide special problems as the noise quota is often already fully utilised. This paper gives details of work undertaken during the design and construction of a large extension to an existing facility at the KÂrst¯ Refinery on the West Coast of Norway. It presents the results of significant testing work undertaken during this project, with particular attention being paid to both the source and attenuation of piping noise. Session C4-1, 10:40 h Paper number 443 (Contributed, Oral) I-INCE Class 55 Eduardo Bauzer Medeiros, Maria Lucia Machado Duarte and Jorge B. Pereira Bittencourt Departamento de Engenharia Mecânica da Universidade Federal de Minas Gerais, Av. Antônio Carlos 6627, 31270-901, Belo Horizonte, MG, BRAZIL CONTROLLING NOISE IN ELECTRICAL SUBSTATIONS Outdoor plant noise control design and construction, OR 76, or 30 abstract = In recent times unplanned urban occupation has caused very serious environmental problems, including those caused by the acoustic emission from electrical substations. At the same time, with stricter noise regulations and with increased power supply requirements, there is now a growing demand for noise control solutions for these substations. Even though research on substation noise control has been under way for nearly forty years, there is still a strong need for efficient and cost effective noise control solutions. This work describes some of the research effort being carried out at the Federal University of Minas Gerais, UFMG at the local electricity power company CEMIG. A description of the initial acoustic evaluation of a substation inside an urban environment is provided, together with the corresponding acoustic modelling procedure. Experimental and computational results already obtained from actual passive noise control techniques are presented. Active noise control modelling, mainly by structural means is also described. Session C4-1, 11:00 h Paper number 614 (Contributed, Oral) Guido Van Leemput and Guust Stevens NOISE EMISSION OF POWERSTATIONS. BIOMASS AS AN ALTERNATIVE FUEL I-INCE Class 56 Adviesbureau Peutz & Associés B.V., P.O.Box 66, 6585 ZH MOOK, The Netherlands Along with the saving of energy, the reduction of greenhouse-gasses (like carbondioxide) is an environmental issue nowadays. This also applies to (electricity) powerstations. In order to meet national and european regulations for minimizing impacts on the environment and also in order to meet the growing demand for "green electricity", the electricity production companies use biomass as a fuel on an ever-increasing scale. Currently the most important biomass-fuels are wood-chips, municipal waste, poultry-litter and other animal wastes. They are used today in a wide variety of processes. There are four primary types of biopower systems: direct-fired, cofired, gasification and modular systems. Each of these techniques require specific components which can have its influence on the noise-emission of the powerplant. In this reading the specific acoustical characteristics of these techniques are being discussed. Research shows that especially the (road) transport of the biomass can affect the total noise emission of the powerplant. Biomass (like wood-chips, municipal waste and animal waste) is mostly being transported by road while "conventional" fuels like coal and natural gas Internoise 2001 Abstracts 249 are mostly being transported by ship and through pipelines. Compared to the transport of coal, the transport of biomass requires many transport-movements per unit of produced energy. The influence of the transport of gas (mostly underground) on the total noise emission of the powerplant can be considered to be even less then the influence of the transport of coal. The conclusion can be drawn that the total noise-emission of biomass-fired powerplants will be larger (in some degree) then comparable conventional (coal- of gas-fired) powerplants, particularly due to the transport of the biomass. POSTERS Session C4-1, 11:20 h Paper number 31 (Contributed, Poster) I-INCE Class 52 Posterboard CP-24 Thursday, Author present 13:20 – 14:20 h S. Curcuruto, C. Fabozzi, A. De Leo, R. Betti, P. Amadio, A. Guarnieri, A. Iacoponi, G. Inversini, G. Licitra, F. Micozzi, W. Piromalli, E. Rambelli , M. Schirone and D. Sepulcri ANPA (National Agency for Environmental Protection), Via V. Brancati 48, 00144 – Rome, Italy THE NOISE PRODUCED BY HARBOUR INFRASTRUCTURES The Italian legislation regarding acoustic pollution is slowly going to be completed with the issue of all decrees in application of the framework law of 26th October n. 447/95. One of this decrees refers to the noise pollution produced by harbour infrastructures. There is few national and international information regarding this argument, above all because such infrastructures are highly complex, characterised by many acoustic different variables, represented by the single activities that are present in the infrastructure during the daily work. For this study eight Italian harbours were chosen, seven marine and one lacustrine. For the choice of the harbours several criteria have been used based on importance, territorial characteristic (e.g. characteristic of the harbour area, distance from inhabited areas), presence of other activities (commercial, tourist and industrial). The study was realised by ANPA in collaboration with some Regional Agency for the Environment-ARPA. The first step of the work was the location on adequate cartography of the different noise sources present in the harbours and of the mainly exposed receivers. In a second time some measurements of noise were executed outside the harbour, in correspondence of the perimeter and at different distance along determined directrix, and near the receivers mainly exposed. From the results emerges that harbour noise, during the diurnal period, doesn’t exceed the valour of 70 dBA. During the night noise levels are often lower than 60 dBA. Session C4-1, 11:20 h Paper number 481 (Contributed, Poster) Posterboard CP-25 Thursday, Author present 13:20 – 14:20 h Tadeusz Wszolek University of Mining and Metallurgy, Al.Mickiewicza 30, 30-059 Krakow, Poland I-INCE Class 76 PREDICTION METHOD FOR LONG-TERM CORONA NOISE LEVELS IN UHV TRANSMISSION LINES The existing acoustic models of the UHV power lines are mostly empirical models, and their application is actually reduced to calculations of levels of corona noise generated in bad weather 250 Internoise 2001 Abstracts conditions. However for evaluation of long-term levels of the noise it is also necessary to know the noise level in fair weather. For damages and contaminations of the conductor surface the corona-noise emitted in the fair weather conditions is increased, what can considerably affect the long-term levels. The basic difficulty in the modeling of corona effects is the problem of evaluation of the critical electric field, above which the corona phenomena are observed. There are known empirical formulas, which determine the values of the above parameter for bad weather, but in the present work the neural networks technique has been applied for estimation of that value for fair weather conditions. The neural model used in the work has been applied for the estimation. In the practical realization the final model has been constructed in such a way, that to the previously established model for bad weather conditions, an additional segment has been added, taking into account the corona processes in fair weather. The results of the model calculations have been compared with the experimental results. Session C4-2: Impulsive noise: effects, prediction and control Chaired by Aad Houtsma Session C4-2, 11:40 h Paper number 435 (Contributed, Oral) Beno Groothoff and David Grantham I-INCE Class 36 Occupational Health Unit, Division of Workplace Health and Safety, DIR, Brisbane, Queensland IMPLEMENTATION OF ABRASIVE BLASTING INDUSTRY CODE OF PRACTICE: WILL IT BE PRACTICE OR IDEALS ONLY? Blasting practices have been demonstrated to create some of the most extreme risks to health of workers from inhaled abrasive blasting dusts and toxic material substrates, as well as from noise. Proper worker protection for this occupation is amongst the most difficult of all industrial processes, and also one of the most difficult to verify. Workplace health and safety obligations, including those from Hazardous Substances and Noise Regulations now make risk control a complex issue. The ability to comply with the new Abrasive Blasting Industry Code of Practice, particularly on hazardous substance and noise obligations after one year, has been reviewed by means of a snapshot audit conducted in all regions throughout Queensland. This audit on 49 workplaces concentrated on four issues; type of abrasive blasting medium, respiratory protection, noise exposure and protection and primary equipment safety during blasting. Formalised risk assessments and their controls, as well as training are not yet common features in this industry and overall knowledge about the levels of protection against noise and airborne contaminants is relatively low. There is a distinct lack of information of in situ testing of the performance of protective helmets with respect to noise and respiratory protection during the blast processes. Additional enforcement activities may encourage this industry to maintain OHS performance to the good standard of the industry code. Internoise 2001 Abstracts 251 Session C4-2, 12:00 h Paper number 168 (Contributed, Oral) Samir N.Y. Gerges, Felipe Vergara and Robert Birch I-INCE Class 49 Federal University of Santa Catarina , Department of Mechanical Engineering, Laboratory of Vibrations and Acoustics, P. Box 476, Florianópolis, SC, Brazil IMPULSIVE NOISES PRODUCED IN SEMI-ANECHOIC, REVERBERANT AND FREE FIELD USING SHOCK WAVE GENERATOR There is a paucity of experimental data relating to impulsive and impact noise that can be accessed and analysed for the purpose of developing international standards. The difficulties in the generation of a broad database are related with the large number of transient type noises present in the industrial and military environments. For example, in the industries a typical impact noise can be considered to have a variable peak sound pressure in excess of 130 dB that may be imposed on continuous noise background. Military weapon systems are capable of producing high level discharges of sound energy with peak sound pressure levels in excess of 160 dB, either in reverberant or nonreverberant conditions that may be attributed to the formation and propagation of shock waves. This paper details a system of generating pulses that uses a mechanism of rapid decompression of compressed gas inside a shock tube where the physical parameters of an ideal Friedlander pulse can be controlled. Other methods to obtained pulses in the free field and also in the reverberant and semi-anechoic chambers, of the Laboratory of Vibrations and Acoustics of UFSC, that are examined and presented in the time and frequency domain. The spectral differences of the type A and B duration are highlighted and the importance of this spectra with respect to potential cause of auditory damage is an important objective of recognising the frequency distribution of large amounts of sound energy. Session C4-2, 13:20 h G. Kerry and C. Lomax Paper number 281 (Contributed, Oral) I-INCE Class 62 University of Salford, United Kingdom PREDICTING THE ACOUSTIC REFLEX RESPONSE TO TRANSIENT NOISE SOURCES EXPERIENCED IN THE COMMUNITY When considering the potentially adverse effects of fast transient noise sources, the onset rate is an important parameter that must be taken into account. A draft ANSI standard has gone so far as to include an adjustment for onset rate. However, the standard is based on community response and does not ‘address other effects of noise such as sleep disturbance or health effects’. A crucial factor when considering the effect of a transient on the human ear is the response of the acoustic reflex, which may provide protection to the inner ear. It has been shown that the acoustic reflex attenuation is a function of stimulus level and onset rate. Current assessments of onset rate measure the gradient of a fixed portion of the waveform, not necessarily representative of real waveforms that tend to be complex with gradients that vary during the rise of the event. A new definition of onset rate has been proposed which would take into account the whole onset portion of a fast transient above the acoustic reflex threshold, using the ultra short LAeq time history. With some refinements, this method can also be used to predict the acoustic reflex attenuation, supplying a more thorough assessment of the effect of transients on the human ear. In this paper, best practice for calculating the ‘Ultra Short LAeq Onset Rate’ is discussed and illustrated with predictions of the response of the acoustic reflex to a range of transient noise sources . 252 Internoise 2001 Abstracts Session C4-2, 13:40 h Paper number 458 (Contributed, Oral) Torben Holm Pedersen DELTA, Building 356, Akademivej, DK-2800 Kgs. Lyngby, Denmark I-INCE Class 69 OBJECTIVE METHOD FOR MEASURING THE PROMINENCE OF IMPULSIVE SOUNDS AND FOR ADJUSTMENT OF LAeq Noise with prominent impulses is more annoying than continuous types of noise with the same equivalent sound pressure level. According to ISO 1996-2:1987/Amd.1:1998 no descriptor exists which can define the presence of impulsive sound. This paper describes an objective method for the measurement of the prominence of impulsive sounds and for the adjustment of LAeq. The method aims at predicting the prominence of impulsive sounds in correspondence with average subjective judgements. The method provides the possibility of operating with different adjustments for different categories of noise sources. The method consists of the following elements: 1) Definition of an impulse; 2) Measurement of level difference and onset rate; 3) Calculation of the prominence P of the impulses; 4) Calculation of a graduated adjustment KI to LAeq. The method is developed on the basis of a literature study and a correlation between a listening test and a large number of possible objective measures. Four Nordic laboratories have made a Round Robin test of the method. The mean standard deviation on the adjustment KI was 0.6 dB. Session C4-2, 14:00 h Paper number 256 (Contributed, Oral) Tae-seok Yoon and Soogab Lee I-INCE Class 76 Center for Environmental Noise & Vibration Research (CENVR), School of Mechanical and Aerospace Engineering Seoul National University, Seoul, 151-742, Korea NEW METHODS FOR THE PREDICTION OF THE MICRO-PRESSURE WAVE BY A HIGH-SPEED TRAIN ENTERING A TUNNEL The compression wave generated by a high-speed train emerges from the exit portal of the tunnel, causing an impulsive noise called micro-pressure wave. There are various countermeasures to reduce the compression wave and micro-pressure wave. In this study, we adopted a new method for the prediction of the sonic boom noise, considering the effect of the nose shape on the resultant noise. Euler equations are first solved, after which the linear Kirchhoff formulation is used for the prediction of acoustic far-field from the flow field data. Euler equations are first solved, after which the linear Kirchhoff formulation is used for the prediction of resultant the micro-pressure wave. The computational prediction and experimentally measured data, compared in this paper, show a good agreement with each other. And the effect of the shape of slanted entry portal on the compression wave and the micro pressure wave is also investigated by numerical simulation. Internoise 2001 Abstracts 253 Session C5-1: Noise mapping Chaired by John Hinton and Christian Popp Session C5-1, 9:40 h Alan Bloomfield Paper number 57 (Invited, Oral) I-INCE Class 56 Department of the Environment Transport and the Regions, Zone 4/G17 123 Victoria Street, London SW1E 6DE OPTIONS FOR NOISE MAPPING IN ENGLAND The proposed EC Directive on Environmental Noise would require the preparation of noise maps for road traffic, railways, airports and industry in major agglomerations and for major roads, railways and airports outside the agglomerations. This paper discusses the options for organising noise mapping in the UK context. It outlines the relevant local and national political structures, as well as those of the aviation, road transport and railway industries as they affect noise mapping. Possible alternatives for the noise mapping project and the technical approaches likely to be adopted in the UK are discussed. Session C5-1, 10:00 h Judith Lang Paper number 234 (Invited, Oral) I-INCE Class 56 A-1090 Vienna Latschkagasse 4, Austria NEW AUSTRIAN GUIDELINES ON THE PRESENTATION OF AIRCRAFT NOISE PROTECTION ZONES AROUND AIRPORTS AND AIRFIELDS In Austria no legislation on aircraft noise around airports and airfields exists. However ÷AL-guideline 24 was edited in 1986 dealing with noise around airports. About 3 years ago a working group started to revise the guideline 24-1 on airports and work out a guideline 24-2 on airfields. As basis a decision had to be taken on the quantity to be chosen to describe the noise immission and on the method to calculate it and on the source data to be used. Further the relevant limit values with respect to land use planning had to be selected. As the A-weighted equivalent sound level LA,eq, for day and night separately, had been used until now, it was decided to use LA,eq also for the new guideline, however using the new noise indicator Lden developed in WG 1 established by the European Commission and to calculate Lday, Levening, Lnight, Lden and the envelope of these. The comparison of the source data and the methods of calculation used in different countries and the final decision will be discussed. For small airfields the use of LA,eq is not sufficient, therefore zones for the maximum level LA,max have to be calculated additionally. Source data have been established. A set of zones for LA,max and LA,eq has been calculated and added to the guideline in an Annex. These can be used for small airfields with only few or without further calculations to define the relevant noise protection zones. Session C5-1, 10:20 h Søren Rasmussen Paper number 416 (Invited, Oral) I-INCE Class 56 COWI, Odensevej 95, Post-box 444, DK-5260 Odense S, Denmark DANISH RAILWAY NOISE PROGRAMME MAPPING AND CONTROL STRATEGY Since 1985 the Danish National Railway Agency has run a noise control programme. The aim is to install noise control measures at all houses exposed to more than 65 dB(A)-24 hours along the main railway lines. The programme is expected to be terminated within the next 10 years. The main control 254 Internoise 2001 Abstracts measures are either a special designed noise barrier mounted close to the track in areas with dense noise impact or grants between 50 and 90% to improve the sound insulation of facades. The annual budget is around DKK 10-20 Million and the average cost on control measures has been DKK 40. 100.000 per house. The achieved reduction has been between 5 and 15 dB(A). The noise abatement programme is based on priority schemes ensuring that houses with the highest noise impact are given priority. The mapping, the cost estimates and the principles of priority setting will be presented along with key figures on costs and reductions. Session C5-1, 10:40 h Jeroen C. Borst Paper number 471 (Invited, Oral) I-INCE Class 56 TNO-PG, P.O. Box 2215, 2301 CE Leiden, The Netherlands THE ROLE OF URBIS' NOISE AND NOISE EFFECTS MAPS IN LOCAL POLICY An important aspect of the EU noise policy is mapping of noise and noise effects and the formulation of noise action plans. In the Netherlands, due to the new policy on noise (MIG), the municipalities will be responsible for the formulation of a local noise policy. An instrument for the assessment of the noise and noise effects in a municipality or a region, can form the basis of setting realistic goals. TNO has developed a GIS based instrument, Urbis, which makes it possible to assess the noise and noise effects in a city or region as a whole, with a high level of detail at relatively low cost. The noise and noise effects maps are based on existing information, wherever possible. Information on activities (e.g. road traffic data), emissions (e.g. rail noise emissions) or immissions (e.g. aircraft noise contours) are used, combined with information on the area (e.g. building configurations). Recently, Urbis has been applied in several (parts of) municipalities. Results of these surveys in the form of detailed noise maps and diagrams will be presented. Also, the implementation of this information in the municipal policy is discussed. Session C5-1, 11:00 h Paper number 529 (Invited, Oral) Christine Bourbon, Phillippe Noël, Rolf Dieter Mummenthey Brussels Institute for the Management of Environment, Gulledelle 100, Bruxelles, 1200, BELGIUM I-INCE Class 56 BRUSSELS LIFE PROJECT: NOISE MAPPING AS A TOOL FOR MANAGEMENT AND PLANNING ROAD TRAFFIC NOISE IN URBAN AREA Carried out for a city of one million of inhabitants on a 160 km≤ area, including 220.000 buildings and 2.000 km of streets, the Brussels noise mapping required a wide work of collection of data's and the collaboration of various departments of the regional administrations but also expertises on GIS and noise calculation software. The first results are already used by district administrations, transport,... to solve "hot spots", to test new traffic management solutions,... As tool of communication, the maps have to be understandable by politicians and citizens. It's the reason why it has been decided not only to produce noise maps with dB(A) indications but to cross land use maps, population maps with noise maps. The results are more useful to establish priorities of actions and to communicate on a non technical way. In the main frame of the European Noise Directive (END), such project takes into account since the beginning the content of the END (Indicators Lden& Lnight, periods, calculation methods,...). The challenges are real: - To adapt the methodology of work to the requirements of stakeholders; - To maintain the collaboration between the actors and users; - To ensure a good Internoise 2001 Abstracts 255 communication and an optimal level of information to citizens; - To transfer the methodology of the project to other European cities. Session C5-1, 11:20 h Paper number 568 (Invited, Oral) Hardy Stapelfeldt and Andrew Jellyman Wilhelm-Brand-Str. 7, D 44141 Dortmund, Germany NOISE MAPPING IN LARGE URBAN AREAS I-INCE Class 56 The technical aspects of large urban noise maps are discussed in three parts: 1. Strategies for the postprocessing of GIS data including the closing of open polygons, defining the heights of buildings, including the use of airborne laser scanning (based on a pilot study in the German town of Bonn) and other techniques to improve the usability of available data; 2. Improving the speed of calculation by the simplification of geometry and/or changing calculation parameters, such as the handling of reflections, and determining the influence on the subsequent results by the use of statistical analysis using the data from Düsseldorf and Birmingham; 3. Noise exposure analysis utilising geo-referenced address data as used in a pilot study in Birmingham, and the influence of various methods of determining façade noise levels including examples using large scale noise maps. Session C5-1, 11:40 h Christian Popp Paper number 660 (Invited, Oral) I-INCE Class 56 LÄRMKONTOR GmbH, Große Bergstraße 213-217, D-22767 Hamburg, Germany THE STEPS FROM A NOISE MAPPING SYSTEM TO AN ACTION PLAN AND ITS EVALUATION Noise mapping means more than drawing coloured pictures, and the development of an action plan is more than preparing a list of different noise mitigation measures. The presentation will give an overview over the necessary working steps of a noise abatement process. It will start with the elaboration of classical noise maps (including sensitivity and conflict maps as well as tables and diagrams with numbers of exposed residents depending on the land-use) for different sources and it will continue with the development and the evaluation of city-wide conceptions for the implementation of noise mitigation measures. For every working step will be shown how the starting point and the hereof based results look like. Furthermore it will be explained in detail what has to be done in the different working steps, which regulations have to be considered, which documents have to be used and which data are necessary to produce acceptable results. In addition to this the competent authorities involved in that process will be listed and it will be described what the advantages of the different working steps for the municipalities are. Especially for the actions plans possibilities for priority settings (e.g. differentiated after streets or districts) will be presented. Finally the evaluation of the different measures (e.g. based on costs, cost-benefit-ratios) and their contribution to the whole noise mitigation conception will be shown. 256 Internoise 2001 Abstracts Session C5-1, 12:00 h Paper number 729 (Invited, Oral) Edgar Wetzel and Catherine Serve Wölfel Measurement Systems Software, B-4700 Eupen, Belgium I-INCE Class 56 HOW TO ASSURE QUALITY OF NOISE MAPPING SOFTWARE? TWO APPROACHES EXPLAINED Noise Mapping has gained additional momentum due to the European Directive initiative: The declared aim is to achieve greater harmonisation amongst Member States. Although the text achieves a fair harmonisation and standardisation of prediction calculation methods, indicators and mapping methodology, no attention is brought to the software tools that will have to be used. Nevertheless, the quality of these software tools, their usability, reliability and precision will be of tremendous importance if comparability of calculated results amongst Member States should be assured. The paper will first point out why quality assurance for Noise Mapping Software is needed and then give an overview of quality assurance techniques. In addition French and German normalisation efforts will be briefly explained and compared. Session C5-1, 13:20 h Paper number 738 (Invited, Oral) Michael Fillery and Thomas Stephan University of Derby, Derby, U.K. I-INCE Class 56 NOISE MAPPING OF MOTOR RACING NOISE FROM ALTERNATIVE TRACK LAYOUTS AT DONINGTON PARK In the autumn of 2000, Donington Park Motor Racing Circuit was preparing plans to upgrade the circuit to comply with modern international standards. The planned upgrade would involve substantial alterations to the track layout and to the ancillary buildings around the track. This paper describes the methods used to prepare maps of the noise impact of track use upon surrounding communities. Noise maps were prepared for the existing track layout and for the proposed new track layout for different types of vehicle races. A differential map comparing the two track layouts revealed that the noise impact on most noise sensitive location would remain unchanged as a result of the new track layout. The noise prediction programme was also used to model the pass-by noise levels at distant locations and to model possible changes in pit lane noise arising from relocation of the pit and paddock. Session C5-1, 13:40 h Paper number 67 (Contributed, Oral) Wolfgang Probst and Bernd Huber ACCON GmbH, Gräfelfinger Str. 133A, D-81375 Munich, Germany I-INCE Class 56 A COMPARISON OF DIFFERENT TECHNIQUES FOR THE CALCULATION OF NOISE MAPS OF CITIES Noise mapping is an important tool for the establishment of widely accepted noise abatement strategies in Europe. Many mapping projects have shown that traffic noise is the source that is responsible for the major part of noise annoyance in cities. Different procedures for calculating the noise emission of roads and the sound propagation in built-up areas are used in different countries. These procedures are discussed taking into account the complexity of typical townscapes and the requirement to handle all this in a formalized and reproducable way by means of software tools. Relations between possible depth of data acquisition and accuracy of the results have an important influence on the ranking of models. Internoise 2001 Abstracts 257 Session C5-1, 14:00 h David T. Dubbink Paper number 239 (Contributed, Oral) I-INCE Class 56 Noise Management Institute, 864 Osos Street, Suite D, San Luis Obispo, California, 93401, USA. PROVIDING COMMUNITY NOISE INFORMATION WITH INTERACTIVE MAPS Community concerns about environmental noise are usually a combination of quite localized, individual concerns. People want to know how they will be impacted by a proposal. They also want to know how the noise burden they experience compares to that experienced by others. Interactive geographic maps and sound simulation systems now make it technically possible to answer such localized questions. The author has reported on his experiences using interactive sound information technologies (D. Dubbink; InterNoise 2000, Noise Control Engineering Journal, Vol 41/Number 3, 1993). Mapping issues are currently under intense discussion because of the mapping initiatives of the EC Directive Related to the Assessment and Management of Environmental Noise. The government of Australia has also supported innovative work in mapping airport noise (D. Southgate; InterNoise 2000). The question is how to best organize and present interactive map information. Issues include the enduring problem of finding the best way to describe and depict community noise. The new presentation technologies are highly effective but they are quite demanding of support resources. The paper explores several innovative approaches and discusses their comparative advantages for providing community information and support for decision-making. Session C5-1, 14:20 h Paper number 278 (Contributed, Oral) P. Paffen and J.J.A. Van Leeuwen I-INCE Class 56 Ministry of Transport, Public Works and Water Management, Directorate-General of Public Works and Water Management, AVV Transport Research Centre, Rotterdam A NOISE MAPPING SYSTEM FOR HIGHWAYS IN THE NETHERLANDS A noise map is a tool for developing a policy about quality of life. Silence 2.0 is a new application for noise calculations on a national or regional scale. The application will calculate noise contours as well as the amount of surface above a certain noise level and the number of people annoyed. The application uses a large number of databases from different origin. Information of hectometre signs, traffic flow, road surface, noise barriers is added to geographical information of roads. Other information as geographical information from residential areas, industrial areas, nature areas and the so-called ‘silent areas’ is also combined to get the complete model for noise calculations. Data is collected for the complete network, which consists of 3000 km of highways, and also for the last 10 years. The noise levels will be calculated in discrete points and interpolations between these points will give noise contours. The calculation core is the DGMR GeoNoise with the Dutch calculation method for traffic noise. The noise contours are combined to information on number of inhabitants of different areas so the number of people annoyed by noise will be determined. The presentation will give information about the gathering of data, the calculation method, the method of noise mapping, and some first results about the number of people annoyed around the Dutch network. 258 Internoise 2001 Abstracts Session C5-1, 14:40 h Paper number 391 (Contributed, Oral) L. Dekoninck and D. Botteldooren A TRAINNOISE-GIS FOR STRATEGIC PLANNING I-INCE Class 56 Ghent University - INTEC (ACOUSTICS), St- Pietersnieuwsstraat 41, 9000 Gent - Belgium Commonly the goal of noise mapping is to predict noise levels on a local scale as accurately as possible. This involves taking into account every building and geographic feature in detail. For strategic planning, this procedure shows several shortcomings. Including that much detail makes the job tedious and expensive. Moreover details on new infrastructure may not be known. The indicators used for strategic planning are noise annoyance and exposed population (Lden). This involves locating the population geographically with sufficient accuracy. A detailed land use grid was used to map people to buildings. A GIS (geographic information system) based methodology was developed to solve these problems. Railway tracks were digitised and split up into segments with length between 1 and 5 km. Train traffic on these segments is known and resulting noise emission can be calculated. The GIS is used to retrieve parameters that reflect the presence of barriers, hard and soft ground, closed rows of buildings and dispersed buildings, vegetation and average altitude. These propagation parameters are used to calculate the attenuation (ISO 9613) spatially averaged along the axes of the segment of the railway track and at a specific distance. Using this technique, scenarios for changing traffic or a certain noise policy can easily be evaluated without consulting the GIS, if the infrastructure does not change. Railway segments causing high numbers of annoyed people can be identified. This fast calculating Trainnoise GIS for Strategic Planning is implemented for Flanders, Belgium. POSTERS Session C5-1, 15:00 h Paper number 32 (Contributed, Poster) Posterboard CP-26 Thursday, Author present 15:20 – 16:00 h N. Fürst CERTU - 9, rue Juliette Récamier 69456 LYON Cedex 06 – France DESCRIPTION OF A METHOD FOR CONSTRUCTION OF NOISE OBSERVATORY CONCERNING SURFACE TRANSPORTS INFRASTRUCTURES ON ALL THE FRENCH TERRITORY I-INCE Class 56 Noise French law is intended to limit noise due to new infrastructures, to protect new buildings near existing infrastructures. Moreover it deals with black spots noise(1)non covered by preceding measurements. The problem thus consists in developing a census method of black spots noise and follow-up of their treatment. To determine black spots noise, to program and to follow acoustic processing, the adopted method proposes to create a noise observatory along roads of whole French territory. This observatory is based on the sound classification of surface transports infrastructures, which provides their distribution in five categories of noise emissions, established according to their geometrical characteristics and traffic. Black spots noise are listed graphically while crossing, strongly noise exposed areas, in day or night, according to acoustic propagation formulas using noise emissions of classification and sensitive built spaces (dwellings, schools, hospital). The common areas thus visualised correspond to black spots noise. The method is supplemented with a dedicated tool, made up of a geographical information system and a relational data base, where black spots noise are located Internoise 2001 Abstracts 259 on common geographical reference frame for France and associated each to descriptive information (exposed buildings and population, actions planed, actions carried out). This methodology is intended to be used to program actions plan for black spots noise on whole French territory. It must thus be simple to implement. It does not replace detailed operational studies, but allows to constitute on a large scale an observatory for the follow-up of national programmes of noise abatement. Session C5-1, 15:00 h Paper number 285 (Contributed, Poster) Posterboard CP-27 Thursday, Author present 15:20 – 16:00 h Maria Bite, István Dombi and Pál Bite Institute for Transport Sciences, H-1119, Budapest, Thán K. u. 3., Hungary TRAFFIC NOISE SITUATION ALONG HUNGARY'S MAIN ROADS I-INCE Class 56 In accordance with the EU guidelines, noise measurements and traffic counting were made along Hungary’s arterial roads. When evaluating the results, a new system became necessary to understand and judge the severity of the noise pollution. To assess the situation at a newly planned area we came with a number that we gained from calculating the exceeded noise quantity and weighted the number of houses effected by this higher noise pollution. As a result of this method the priority of interventions can be set up objectively. Using these measurements noise maps were compiled along those main roads. Moreover, we saved the results in a kind of database, in which with the help of a map the user can find out the date measured, and information concerning the present noise measures, such as noise abt. barrier or passive noise control. In our presentation we show the noise situation in Hungary, our method to prioritise the interventions, the noise maps along the main roads and the database we have come up with. Session C5-1, 15:00 h Paper number 349 (Contributed, Poster) I-INCE Class 56 Posterboard CP-28 Thursday, Author present 15:20 – 16:00 h Alice Elizabeth González, Bach. Andrés Jorysz and Esteban Gaja Díaz Departamento de Ingeniería Ambiental – IMFIA – Facultad de Ingeniería de la Universidad de la República (Uruguay) A PROPOSAL OF A METHODOLOGY TO BUILD A ROAD NOISE MAP AND ITS APPLICATION TO MONTEVIDEO CITY (URUGUAY) The detailed methodology for recording field information to build road noise maps is not prescribed by any standard. This fact introduces additional limitations in order to compare results from different investigation teams and sometimes even to make the presented information credible. This paper presents a methodology for collecting field information to build a road noise map, which was first developed to build the noise map of Montevideo City (Uruguay). The major part of this work was included in the Ph.D. thesis in Environmental Engineering of the author. There are two main proposals: - The determination of the optimum sampling time, based on the statistical processing of field information. Recorded information of twelve hour measurements in different points of the city was processed. The equivalence among different hours during the day and different days of the week was demonstrated. The optimum sampling time for describing the day's noise equivalent levels was determined; - The selection of the points to consider on a road, using multivariate analysis in the processing of field information. A preliminary measurement campaign is necessary in order to collect information of every block. Cluster analysis allow to choose the points of measurement that represent 260 Internoise 2001 Abstracts the noise reality along the street. Both methodologies are explained in the paper. An application case in Montevideo is presented to illustrate them. Session C6-1: Structure borne sound Chaired by R. Bernhard and J. Cuschieri Session C6-1, 9:40 h Gunnar Lundmark Paper number 128 (Contributed, Oral) I-INCE Class 42 Langelandsgatan 27, S-164 43 Kista, Sweden SKATING ON THIN ICE - AND THE ACOUSTICS OF INFINITE PLATES A tone is radiated when people are skating on thin, black ice. The phenomenon is well known to the experienced skater. In fact he estimates the thickness of the ice by listening to the tone. It is easy for most people to hear the frequency shift when the thickness of the ice changes. A person with "golden ears", absolute pitch, can estimate the thickness of the ice with an uncertainty of 5% just by listening. Measurement results of the frequency of the tone and the thickness of the ice are presented together with sound recordings. The basic physics, the coincidence frequency of mass loaded infinite plates, which explains this phenomenon is discussed. Infinite plates are very common in the acoustic world and ice on a large lake is probably the best example of an infinite plate in the real world. A high, intense tone indicates thin, dangerous ice. Is it just by chance that the sensitivity of the ear increases when the thickness of the ice decreases, and becomes very high when the ice tends to crack under human weight? Session C6-1, 10:00 h Paper number 734 (Contributed, Oral) N. Joly and J.C. Pascal I-INCE Class 42 Laboratoire d'Acoustique de l'Université du Mans (UMR 6613), Université du Maine, Av. O. Messiaen, 72085 Le Mans cedex 9, France APPROXIMATIONS INTRODUCED BY THE USE OF PLATE KINEMATIC MODELS TO DETERMINE ENERGY FLOW AND ENERGY DENSITIES The study of the vibrations of a plate in the low-frequency domain usually is performed using a displacement formulation ; this is the case using the finite element method. All elastic variables (strain, stress) can be derived from this displacement field. Then the energy flow within the structure (structural intensity) and strain energy density can be determined, and this is the procedure used to study energy transfer in structures for low frequencies. Plate studies use displacement fields based on cinematic models (Kirchhoff, Reissner / Mindlin), which include or not shear force or rotatory inertia and give more or less refined solutions. The approximations introduced by such models are well known concerning the solution accuracy (wave number, frequency) and the field (displacement, stress). The aim of this paper is to discuss the approximations introduced by different cinematic models when used to express energy quantities (structural intensity, strain and kinetic energy density) in comparing them to exact elasticity theory solutions. Different plate waves occurring in low- and medium-frequency range (bending, quasi-longitudinal, shear horizontal) are presented. It is found that, although the displacement and the stress profile are well described in the first order, some simple Internoise 2001 Abstracts 261 cinematic models neglect some term of the in-plane thickness-averaged structural intensity and are unable to describe energy transfer in the thickness of the plate. Session C6-1, 10:20 h Paper number 773 (Contributed, Oral) N. Kojima, M. Mikami, M. Yamaguchi and S. Kumura I-INCE Class 42 Department of Mechanical Engineering, Yamaguchi University, Tokiwadai 2-16-1, Ube, Yamaguchi 755-8611, Japan EFFECTS OF EXTENSIONAL MOTION ON FLEXURAL VIBRATION ENERGY FLOW IN SHELL For reducing vibration and noise from mechanical structures effectively, it is important to identify the excitation sources and vibration transmission paths. Vibration Intensity (VI) technique, in which vibration energy flow is expressed as a vector quantity, is one of the effective ways for achieving this purpose. The vibration energy flow is recognized as the active component of vibration intensity. This vibration energy is carried by the flexural (out-plane) and the extensional (in-plane) vibration motions. In a flat plate, the flexural and the extensional vibration motions are uncoupled. In that case, therefore, it is useful to detect the flexural vibration motion alone for estimating the noise radiated from the plate. In a shell, however, these two components of vibration motion are coupled due to the curved structure. It is important to detect the extensional vibration as well for measuring the flexural vibration intensity. The energy flow in a shell will be thus characterized by introducing the additional understandings of curvature contribution. In the present paper,vibration motions and vibration intensities in a shell structure were studied through finite element method (FEM) analysis and measurement in a shell. Effects of extensional vibration motion on the analysis of vibration energy transmission in a shell were discussed. Session C6-1, 10:40 h Paper number 194 (Contributed, Oral) Pascale Neple, Ali El Hafidi and Claude Lesueur EADS AIRBUS SA, 316 Route de Bayonne, 31060 Toulouse Cedex 03, France I-INCE Class 43 ON SOUND TRANSMISSION THROUGH A TRUNCATED ISOTROPIC CONICAL SHELL: EXPERIMENTAL AND PRELIMINARY NUMERICAL RESULTS The present paper deals with the vibroacoustic behaviour of a truncated isotropic conical shell excited by a diffuse sound field for the purpose of understanding airborne sound transmission through aircraft cockpit. Experimentation has been carried out on a truncated conical shell. The lower and upper ring frequencies of this structure are fR2 = 1800 Hz and fR1 = 4500 Hz, its critical frequency is fc = 12000 Hz. fR2 and fR1 correspond to the ring frequencies of cylindrical shells with radius equal to the larger and the smaller radii of the cone. Numerical results have been obtained on the same structure with Ideas Vibroacoustics software. Noise Reduction (NR) has been measured in the 100 Hz - 5000 Hz range, that is, between 0.05*fR2 and 1.11*fR1. Three major results have been observed : - Below fR2 a membrane behaviour dominates with minima due to cavity modes. The eight first cavity modes (between 0.1*fR2 and 0.3*fR2) have been determined experimentally and compared successfully to the numerical results; - Between fR2 and 0.7*fR1, the NR reaches a plateau; - Above 0.7*fR1, the transmission is characterized by a mass law tendency behaviour. Whereas the membrane and mass law behaviours are well known for sound transmission through a cylindrical shell, the plateau is a new result for sound transmission through the cone. Note that the mass law emerges below fR1. It is then 262 Internoise 2001 Abstracts concluded that sound transmission through a truncated conical shell is dictated by the larger section of the truncated cone, i.e. the less rigid part, below the emergence of the mass law. Session C6-1, 11:00 h Kei Andow Paper number 248 (Contributed, Oral) I-INCE Class 43 Kajima Technical Research Institute, 182-0036 Tobitakyu Chofu Tokyo, Japan COMPARISON WITH SCALE MODEL MEASUREMENTS FOR STUDY OF STRUCTURE-BORNE SOUND MODEL MADE BY POLYVINYL CHLORIDE We study on the structure-borne sound (S.B.S) propagation use by the small-scale model made by vinyl chloride. On similar law, 1/20 small-scale model is analogous to third octave frequency. We have a chance to measure the quarter cut and three floors mock-up model. The S.B.S propagation character was measured using this model. The comparison between the real size and the scale model is very analogous to the wide band frequency. This method shows very effective at the S.B.S propagation. Session C6-1, 11:20 h Paper number 727 (Contributed, Oral) H. Nishino and M. Ohlrich I-INCE Class 43 Nagasaki R & D Center, Mitsubishi Heavy Industries, LTD., 5-717-1, Fukahorimachi, Nagasaki, Japan APPLICATION OF WAVE INTENSITY ANALYSIS FOR PREDICTING MID-FREQUENCY VIBRATION TRANSMISSION IN EXTENDED PLATE STRUCTURES The frequency range for which prediction methods can be applied for vibration analysis of built-up structures can be divided into three regions: low, medium and high frequencies. The conventional finite element method is successfully used at low frequencies, that is, for problems where the structural wavelengths are relatively large, whereas the statistical energy analysis (SEA) is applicable at high frequencies - or for short wavelengths. The mid-frequency range is currently difficult to estimate by means of these conventional analyses. However, the wave intensity analysis (WIA) newly developed by Langley offers an improved solution to the established SEA for predicting vibration levels in the mid-frequency range. The wave intensity analysis is applicable to non-diffuse wave fields, which by assumption is not accommodated in SEA. In this paper, the wave intensity technique is applied for predicting the vibratory energy distribution in six-section plate structures. Both bending waves and in-plane waves are included in the analysis and the predicted results are compared with experiments based on random excitation by point forces. It is concluded that the wave intensity analysis is suitable for predicting the vibration response in the mid-frequency range and that the inclusion of in-plane waves has a significant influence on the predicted response of plates that are distant from the excited plate. Internoise 2001 Abstracts 263 Session C6-1, 11:40 h Jacek Cieslik Paper number 749 (Contributed, Oral) I-INCE Class 43 University of Mining and Metallurgy, Dept. of Robotics and Machine Dynamics, Al. Mickiewicza 30, 30-059 Krakow, Poland STRUCTURE SURFACE INTENSITY IN APPLICATION TO THE TRANSFER OF VIBRATIONAL ENERGY ANALYSIS Paper presents the method of application of structure surface intensity to the investigation of vibrational energy transfer in solid structures. The new formulation of the method regarding the wave approach is discussed allowing the analysis of curved bars. Besides, there are presented the results of experiments. Virtual experiment results with modal approach are compared with results of measurement done on real structure - the railway coach element. The conclusions give the some considerable remarks on possibility of use presented method in medium frequency range and comparison with other methods used nowadays in research of energy transfer. Session C6-1, 12:00 h Paper number 771 (Contributed, Oral) F.S. Sui, M.N. Ichchou and L. Jezequel I-INCE Class 49 Laboratoire de Tribologie et Dynamique des Systèmes, Ecole Centrale de Lyon, 36, Avenue Guy de Collongue, 69131, BP 163, Ecully, FRANCE STUDY OF TRANSIENT ENERGY OF COUPLED SYSTEMS USING A LOCAL ENERGY TECHNIQUE The transient energy flow between two coupled rods subject to an impulse is studied by a newly developed energy method called as transient local energy approach (TLEA). For the purpose of validation, the transient energy predicted by TLEA discretized equation was compared with that from transient statistical energy analysis (TSEA) and an exact energy result. The numerical simulations show that the TLEA can give the more precise and reliable prediction of transient energy than TSEA. Session C6-1, 13:20 h Paper number 761 (Contributed, Oral) Colin Fox and Keith Ballagh I-INCE Class 51 Mathematics Department, The University of Auckland, PB 92019, Auckland, New Zealand IN SITU MEASUREMENT OF POWER FLOW AND MECHANICAL PROPERTIES OF VIBRATING TIMBER STRUCTURES Determining vibrational energy flow in a structure requires knowledge of local mechanical properties as well as measurement of the vibrational motion. By using more measurements than are required to measure the motion, we model-fit to find the local mechanical properties and hence determine energy flow. In this paper we discuss the selection of optimal transducer placement for greatest measurement accuracy across a range of frequencies, and give examples of measuring directional bending-wave energy in a wood beam where the dynamic mass and rigidity are unknown a-priori. 264 Internoise 2001 Abstracts Session C6-1, 13:40 h Paper number 138 (Contributed, Oral) Olivier A. Bareille and Louis Jezequel I-INCE Class 75 LTDS (CNRS UMR-5513) – Ecole Centrale de Lyon, 36 Avenue Guy de Collongue, 69131 Ecully Cedex, France VIBRO-ACOUSTICS COUPLING EQUATIONS: WAVE COMPATIBILITY CONDITION Dealing with vibro-acoustics leads us to the choice of an adapted and cost-effective method. At the crossroad of boundary integral methods and of modal synthesis approaches, the method developed in this paper masters the increase of unknown terms when the study is carried to the medium and high frequency range. Then it allows us to widen the field of applications for the previously quoted methods. It is therefore an alternative for the study of mid-frequency range problems. The primal and dual fields used to define the behavior of each substructure are described through a set of generalized boundary values. Hence the propagation of waves as well as local stationary responses are represented by their images along the edges of the sub domains. Interactions are derived at these locations through the focus of Wave Compatibility Conditions. The example of coupling between mono-dimensional and bi-dimensional structures has been studied previously. However, back then the main was to highlight how the coupling equations can be written in a straightforward manner. Here, we go much further in the formalism and the description of problems with a set of easily interpretable coupling equations. Results of radiating beam and the propagation in semi-infinite media are exposed. Session C6-1, 14:00 h Alexandr Yu. Dushin Paper number 772 (Contributed, Oral) I-INCE Class 76 Moscow State Engineering Physics Institute, Kashirskoe shosse, 31, Moscow, 115409, Russia NEW FINITE ELEMENTS FOR MIDDLE FREQUENCY VIBRATION ANALYSIS OF AXISYMMETRICAL SHELLS A lot of noisy subjects in different areas of technique are the complicated axisymmetrical shells. To analyze its vibration often use the numerical methods, for example, finite element method (FEM). FEM is very convenience for the low frequency range of vibration, but the increasing of excited forces frequencies follows the grown of number the elements in model. The present investigation is devoted to developing of new types of shell finite elements with special shape functions based on the analytical solutions of moving equations. These solutions are the exponents with complex argument for cylindrical shells and the Bessel functions for the circular plates. Using the traditional FEM methodologyitis possible to get effective dynamic stiffness matrices, which can replace in finite element mesh up to some dozens traditional elements. To avoid the numerical difficulties by the calculations of exponential terms, the special form of analytical solution of moving equations is used. In fact, the knots of such finite elements can be placed only in the points, where the shell changes its geometry form. The results of comparison with analytical solutions and modal analysis of complicated shell structure are given. Internoise 2001 Abstracts 265 Session C6-1, 14:20 h Paper number 766 (Contributed, Oral) Odobescu Grigore Liviu I-INCE Class 14 Institute of Solid Mechanics, Constantin Mille 15 , sector 1, RO-70701 Bucharest; Romania METHOD OF CHARACTERIZATION FOR THE ADAPTATION DIPOLE GENERATORPIEZOELECTRIC TRANSDUCER The ultrasonic systems, which generate high energy vibrations in the frequency range [20-40] kHz are supplied from electronic generators which work in commutation. The commutation working regime assure a very high efficiency for electronic generator, but generates a very large spectrum of frequency. The piezoelectric transducer, for maximum efficiency work must be supplied with a sinusoidal tension which have a frequency at close quarter of the transducer series frequency. Because of the parameters complexity which intervene, the work with maximum efficiency can be realized only in very precise conditions attached of the value frequency, the accuracy of the shape command tension and the charge value which can be very variable in function of the work medium where it is transmitted the ultrasound mechanical vibrations . The theoretical and experimental analysis of generated spectrum accomplished in this paper leads to very useful conclusions in order to choose a proper adaptation dipole between generator and acoustic charge. It is presented a new adaptation dipole which ensures a high efficiency and stability. Also it is presented the working diagrams. This analysis enables the determination of efficiency for whole ultrasonic system and it found same possibilities for it raising. POSTERS Session C6-1, 14:40 h Paper number 107 (Contributed, Poster) Posterboard CP-29 Thursday, Author present 15:00 – 16:00 h Wojciech P. Pdzanek Jr., Zbigniew Engel and Witold Rdzanek Pedagogical University, Department of Acoustics, 35-310Rzeszów, Poland AN EXPERIMENTAL VERIFICATION OF SOME ANALYTICAL CONSIDERATIONS ON SOUND RADIATION PHENOMENON OF A CLAMPED ANNULAR PLATE I-INCE Class 42 The paper presents an experimental verification of an earlier purely theoretical asymptotic approach to solve the phenomenon of sound radiation of a clamped annular plate. The Kelvin-Voigt linear theory of a visco-elastic plate is employed. The question is how good agreement of the employed theory and some experimental results is. A short description of analytical methods used is given and the experiment system scheme, located in an an-echoic chamber, is shown. The comparison of analytical results with those from experiment is performed and discussed. 266 Internoise 2001 Abstracts Session C6-1, 14:40 h Paper number 333 (Contributed, Poster) Posterboard CP-30 Thursday, Author present 15:00 – 16:00 h Ivan V. Andronov ON INFINITELY LONG LIVING VIBRATIONS IN MECHANICAL CONSTRUCTIONS I-INCE Class 42 St.Petersburg University, 1-1 Ulianovskaya Str., 198904 St.Petersburg – Petrodvorets, Russia Mechanical constructions containing plates, shells and beams are widely used. Such elements may cause resonances to appear. In this paper it is shown that the resonances in thin plates may be with infinitely small attenuation. Examples of eigen vibrations are presented. These are the flexure vibrations in an isolated infinite plate in a vicinity of a circular hole with specially chosen mass and momentum loading on its edge. The wave field in this case exponentially decreases in all directions. If the plate is fluid loaded, such eigen vibrations become resonances with very weak attenuation. The other example is the shear vibrations in Timoshenko plate of the same geometry. Shear eigen vibrations are possible even if the plate is fluid loaded. Sufficient conditions for the resonances of flexure and shear types not to be presented in the plate are found. Session C6-1, 14:40 h Paper number 709 (Contributed, Poster) Posterboard CP-31 Thursday, Author present 15:00 – 16:00 h Yiu-kee Kam and Johnny C.Y. Wong I-INCE Class 88 Environmental Protection Department, The Government of The Hong Kong Special Administrative Region, Hong Kong CONSTRUCTION NOISE V. COMMUNITY'S DESIRE - A PARTNERSHIP APPROACH The Quarry Bay Congestion Relief Works of the Mass Transit Railway Corporation Limited in Hong Kong consists of the construction of a 6.3 m diameter, 2 km long twin-tube tunnel parallel to an existing underground railway tunnel under a densely populated area. The construction noise assessment in the environmental impact assessment report of the project indicated that with the incorporation of suitable environmental mitigation measures, the planned tunnel boring work roundthe-clock would not cause any noise impact on the residents along the alignment of the new tunnel. However, during the construction stage commencing in late 1998, the noise from the construction work has generated numerous complaints from the residents nearby, particularly during the evening and night-time periods. This paper will describe the concerted effort to resolve the issue through collaboration between the project client, the contractors and the noise control authority. The experience demonstrates to the construction industry of the need to strike a balance between maintaining a tight construction schedule of their development projects and meeting with the community’s desire for a tranquil environment. Internoise 2001 Abstracts 267 Session C6-1, 14:40 h Paper number 701 (Contributed, Poster) I-INCE Class 43 Posterboard CP-32 Thursday, Author present 15:00 – 16:00 h E. Berestovitsky, U. Gladilin, V. Korchanov, A. Pavlov and V. Sirotnikov Concern “Aurora”, st. Karbiysheva 15, 196024 St. Petersburg, Russia LOW LEVEL OF VIBRATION, AS CRITERION OF QUALITY OF DESIGNING HYDRAULIC DEVICES The scientific methodical bases of designing low noise level of hydraulic devices allowing to carry out an optimum choice a flowing part and to determine the basic parameters are considered. The method of estimation the quality of designing hydraulic devices consisting from two stages (approximate and final) is developed. Approximate method allows to carry out an expert estimation of alternative variants of flowing parts at design time. The method is based on the available information on results of tested analogues, which allows to determine frequency-dependent coefficient. It describes transfer function of the device on the basis of the power approach. The specified approach gives good result in a zone of frequencies 1000-10000 Hz and allows to calculate levels vibration of devices (flowing parts) and to choose best (with less level of vibration) variant. At the second stage the acoustic account of the chosen variant of a flowing part of the device is specified. The non-stationary mode of operations of the hydrodevice causing change of the charge of a liquid in pipelines of hydrosystem is taken into account. Excitation of vibration of the device and pipelines generated by action of wave processes, which can result in resonances arising due to auto fluctuations or pulse component of pressure, developed by the pump is also considered. The "wave" pulsation of pressure raise low-frequency fluctuations, which are necessary for minimizing while designing the hydraulic device, hydrosystem and control system. The method is confirmed by experimental researches and allows to choose and to optimize the basic parameters of hydrodevices at design time. Session C7-1: Acoustical diagnosis methods Chaired by Andras Illenyi Session C7-1, 9:40 h Paper number 11 (Contributed, Oral) Adam Charchalis and Andrzej Grzadziela I-INCE Class 44 Polish Naval Academy, Gdynia, Institute of Construction and Propulsion of Vessels, Poland DIAGNOSING OF NAVL GAS TURBINE ROTORS WITH THE USE OF VIBROACOUSTIC PARAMETERS In the paper results are presented of vibroacoustic research on balance control of gas turbine rotors and assessment of their permissible operation times. 1241 RE missile corvettes, which are fitted with COGAG gas turbine propulsion systems, are subject to a permanent basic diagnostic system. The investigations were based on the following assumption: If technical state degradation of gas turbine rotor sets is a function of their operation time (at a load spectrum assumed constant) then it is possible to select from the recorded vibration signal spectrum such parameters whose changes can be unambiguously assigned to the operation time. Application of the proposed approach makes managing the engine's operation time much more rational, especially at its end. The proposed approach is noninvasive and does not require taking the ships out of service. Realisation of investigations of the kind 268 Internoise 2001 Abstracts makes it possible to collect data for a database of the future monitoring system of ships, expected to improve their operational features. Session C7-1, 10:00 h Andrzej Grzadziela Paper number 24 (Contributed, Oral) I-INCE Class 44 Polish Naval Academy, Gdynia, Institute of Construction and Propulsion of Vessels DIAGNOSING THE SHAFTING ALIGNMENT IN THE COGAG PROPULSION SYSTEM BY MEANS OF VIBRATION MEASUREMENT The application of vibro-acoustic analysis methods has been presented in this paper. The described conception concerns evaluative process of the centring state in a transmission shafts within powered, marine gas turbine system as a function of ships displacement. Some structural components of the gas turbine unit and reduction gearbox have been selected for the analysis. Some results of the vibro-tests have been presented as well. The tests had been worked out during sea trials. This element of BDS system is accepted and used in the ships, which are powered by the COGAG power plant. The presented method is the typical database for the monitoring system of shaft alignment off-line and online type. This described idea is based on the researches of the corvette power plants. Session C7-1, 10:20 h Paper number 764 (Contributed, Oral) Batko Wojciech and Mikulski Andrzej I-INCE Class 74 University of Mining and Metallurgy, Dept. of Mechanics & Vibroacoust., Al.Mickiewicza 30, 30-059 Krakow, Poland APPLICATION OF A WAVELET ANALYSIS TO DIAGNOSTIC OF ROPE PULLEY ROLLING BEARINGS IN MINE WINDING GEARS The paper presents possibilities created by application of a wavelet analysis in diagnostic of pulley rolling bearings in mine winding gears. Examples of application of new diagnostic signal analysis for bearings technical state evaluation with presence of nonstationary disturbances are described. Analysis results and assessment of their usability in diagnosis of rope pulley rolling bearings in winding gears are included. Session C7-1, 10:40 h Paper number 204 (Contributed, Oral) I-INCE Class 75 Shuichi Sakamoto, Ryoichi Ichimiya, Kensaku Yanagimoto and Taiki Sato Niigata University, Faculty of Engineering, 2-8050, Ikarashi, Niigata, 950-2181, Japan METHOD OF DETECTING NUMBER OF CLOTHS USING SOUND (NON-CONTACT MEASUREMENT BY USING ACOUSTIC CHARACTERISTICS OF CLOTHS) This paper deals with the development of a new method for measuring the number of cloths without contact. It is necessary to detect one or plural number of sheets to avoid mishandling. This detecting method utilizes the variation of the phase of the transfer function of entire acoustic system. The incident impedance of cloths varies with the number of cloths. Characteristic acoustic impedance and complex wave number define peculiar and most fundamental characteristics of the material. It is useful for the detection of the number of cloths to know characteristic impedance of cloths. So that, we tried to measure characteristic impedance and wave number. The measured impedance of cloths by the transfer function method was introduced into theoretical analysis. The results of the experiments and theoretical analysis show fairly good coincidence. Internoise 2001 Abstracts 269 POSTERS Session C7-1, 11:00 h Paper number 65 (Contributed, Poster) Posterboard CP-33 Thursday, Author present 13:20 – 14:20 h Tomsz Boczar Technical University of Opole 45-271 Opole, ul. Mikolajczyka 5, Poland FREQUENCY ANALYSIS OF DISTURBING SIGNALS GENERATED BY CORONA DISCHARGES IN OVERHEAD POWER LINES I-INCE Class 74 This paper is concerned with problems connected with the analysis of acoustic emission (AE) signals generated by corona discharges that can occur during partial discharge measurements using AE method, in industrial conditions. It presents a system for corona discharge generation in laboratory conditions and the apparatus used that makes the measurement and analysis of AE signals possible. It describes a way of carrying out frequency analysis of the measured AE pulses, and the results are in the form of time and frequency spectra runs. The numerical values of descriptors characterizing the amplitude spectra and energy density of AE pulses generated by corona discharges are also presented. This paper also presents the results of measurements and frequency analysis of AE signals generated by corona discharges appearing in overhead power lines. It presents a comparative analysis of the obtained results for model setups in laboratory conditions and for corona discharges in overhead lines. The summary part determines the influence of the measured AE signals generated by corona discharges on the results of the frequency analysis of AE pulses generated by partial discharges in the insulation of electrical power devices. Session C7-1, 11:00 h Paper number 66 (Contributed, Poster) Posterboard CP-34 Thursday, Author present 13:20 – 14:20 h Tomasz Boczar Technical University of Opole 45-271 Opole, ul. Mikolajczyka 5, Poland I-INCE Class 72 FREQUENCY ANALYSIS OF INTERFERING SIGNALS OCCURING IN POWER TRANSFORMERS DURING PARTIAL DISCHARGES MEASUREMENTS This paper is concerned with problems connected with diagnostic measurements of electric power device insulation carried out using the acoustic emission (AE) method. It presents the results of measurements and frequency analysis of interfering signals of an acoustic character, whose source is a power transformer. It describes the sources of interfering signals occurring in a power transformer during its regular work, and it presents their influence on frequency runs of AE pulses generated by partial discharges occurring in its insulation. As part of the frequency analysis of the measured AE signals that was carried out, the runs of amplitude spectra and energy density were designated and the numerical values of the descriptors that characterize them calculated. The summary defines the influence of the measured acoustic disturbances connected with the work of the transformer on AE pulses generated by partial discharges occurring in its insulation. 270 Internoise 2001 Abstracts Session C7-1, 11:00 h Paper number 387 (Contributed, Poster) Posterboard CP-35 Thursday, Author present 13:20 – 14:20 h Masayuki Yokoi, Mikio Nakai and Toshio Takahagi I-INCE Class 75 Junior College, Osaka Sangyo University, 3-1-1 Nakagaido, Daito, Osaka 574-8530,Japan ACOUSTICAL DIAGNOSIS FOR THE LOOSENESS OF WHEEL FIXING NUTS OF AUTOMOBILE BY HAMMERING At the regular inspection of automobile, an inspector strikes wheel fixing nuts with a hammer and detects the looseness of nuts by listening to impact sound. The purpose of this study is to develop the inspection technique that detects the looseness of wheel fixing nuts easily and quantitatively. The experiments were carried out using both an actual passenger car and a model which simulated the wheel. In the experiments using the actual passenger car we strike the wheel fixing nuts with a hammer and search for the variation of sound frequencies owing to the looseness of nuts. The sound pressure waveform was analyzed by using power spectrum, wavelet transform and short-time Fourier transform. We can quantitatively determine the relationship between frequency component characterized by impact sound and torque of fixing nuts in the model experiment. The obtained results are summarized: (1) When the nuts are loosened, impact sound changes from clear sound to thick sound and their frequency change. These frequencies are close to the axial natural frequencies of wheel; (2) Wavelet transform and short-time Fourier transform analysis are available for analysis of impact sound. Session C7-2: Instrumentation and measurement techniques Chaired by Hiro Takinami and Per Bruel Session C7-2, 11:20 h Peter Hedegaard Paper number 610 (Invited, Oral) I-INCE Class 71 Brüel & Kjær, Sound & Vibration Measurement A/S, Skodsborgvej 307, DK-2850 Nærum Denmark SOUND LEVEL METER TIME-WEIGHTING. A HISTORICAL-TECHNICAL REVIEW Since the first IEC standard for sound level meters (IEC 123) was published in 1961, important development has taken place in the detector-indicator part. This development not only reflects the increased technical possibilities, but also the desire to ensure continuity for the purpose of comparison of previous measurement results. This paper is a historical-technical discussion of methods and circuits used for signal processing since 1961, together with their advantages and disadvantages. For sounds with fluctuating levels the problems regarding the reading of an analog indicator and the corresponding problems regarding relevant sampling for digital indication are also discussed. Internoise 2001 Abstracts 271 Session C7-2, 11:40 h Alan H. Marsh Paper number 633 (Invited, Oral) I-INCE Class 71 16072 Santa Barbara Lane, Huntington Beach, CA 92649, USA IEC 61672, THE NEW INTERNATIONAL STANDARD FOR SOUND LEVEL METERS In 2001, the new International Standard for sound level meters, IEC 61672, was in the final stages of development by IEC Technical Committee 29 for electroacoustics. The standard, when published, will replace the 1979 document, IEC 651 (or IEC 60651 as now re-designated by IEC) for conventional sound level meters. It will also replace the 1985 document, IEC 804 (or IEC 60804), for integratingaveraging sound level meters and integrating sound level meters. IEC 61672 will be issued in Parts with technical specifications in Part 1 and procedures for pattern evaluation (or type approval) tests in Part 2.On 2000 October 31, the National Committee members of TC 29 voted approval for the last Committee Draft for Part 1 to proceed to the stage of Final Draft International Standard. Part 1 should be published by IEC in 2002, at the latest, followed shortly thereafter by Part 2.This paper reviews the major technical differences between the new IEC 61672 standard and its predecessor standards for sound level meters that measure exponential-time-weighted sound levels, time-average sound levels, and sound exposure levels. Plans will be discussed for development of the additional Parts of IEC 61672. Session C7-2, 12:00 h Paper number 210 (Contributed, Oral) W. Krebber, A. Bronkhorst, T. Fedtke and P. Giua HEAD acoustics GmbH, Herzogenrath, Germany NEW METHODS FOR NOISE EXPOSURE METERS - THE DOSE PROJECT I-INCE Class 78 The individual noise exposure in everyday live can be measured by personal noise exposure meters. Conventional devices normally pick up the sound on the shoulder. For hearing loss related measurements as well as for studies on noise annoyance, however, the sound pressure in the ear canal is only relevant input. Within the European project DOSE a new portable, light-weight noise exposure meter has been developed using binaural sound pick-up in the ear. Thus, a lot of important individual influences can be taken into consideration now: (a) the perturbation of the sound field by the body; (b) directional effects of incoming sounds, especially relevant for impulsive noises; (c) the shielding of the ears by headphones and/or hearing protectors; (d) sound generation by headphones. The influence of the wearer’s own voice on the measurement results is reduced using a new voice subtraction algorithm. Compared to the conventional noise exposure meter according to IEC 1252 some additional parameters are calculated in order to achieve a better evaluation of impulsive noises. The project has been supported by the European Commission within the SMT (Standards, Measurement and Testing) programme. Current standardization activities are presented. 272 Internoise 2001 Abstracts Session C7-2, 13:20 h Paper number 365 (Invited, Oral) I-INCE Class 71 Koichi Taya, Hajime Suzuki, Kazunori Kimura, Ichiro Yamada, Kenji Fukushima, Toshiaki Koshiroi and Hideaki Marukawa Kobayasi Institute of Physical Research, 3-20-41 Higashi-motomachi Kokubunji, Tokyo 185-0022, JAPAN DEVELOPMENT OF AN ACOUSTIC LEVEL METER FOR FINE-GRAIN SILO This paper describes a new acoustical method of measuring the volume of fine-powder of grain stored in a silo. Ultrasonic level meters are generally used for such measurements, in which we estimate the volume of material by measuring the residual hollow volume in the silo, i.e., by measuring the depth from the ceiling of the silo down to the top surface of the stored material, using an echo signal detection technique. But, this method may be not successful in case of dust material such as fine powder of grain, because the powder floats up into the air and scatters ultrasonic sound signal. We, therefore, developed a depth measuring system using an audible sound signal. The longer is the wavelength of sound, the less the sound is scattered. However, using an audible sound has brought us other difficulties; deterioration of S/N ratio due to the background machinery noise in the silo, disturbance by reflection of sound at the inner surface of the silo and so on. To solve these difficulties, we devised how to install the sound source (a directional loudspeaker) and microphones. We also applied a time stretched pulse technique as well as Hilbert Transform technique to make it easier to detect echo signal from the surface of the material. The new method is superior to the conventional method because of its ability to obtain a stable and precise estimation. Session C7-2, 13:40 h Paper number 374 (Contributed, Oral) I-INCE Class 72 Lorenzo M. Muñiz Mendoza, Sergio R. Rojas Ramírez, Alfredo A. Elías Juárez and Salvador Echeverría Villagómez Centro Nacional de Metrología. División de Vibraciones y Acústica, km 4,5 carretera a los Cués, Municipio del Marqués, Qro. México, CP 76241 ANALYSIS OF ERROR SOURCES FOR SOUND PRESSURE LEVEL DETERMINATION IN A PISTONPHONE COUPLER A pistonphone is a very stable sound pressure level (SPL) source with high stability on frequency. A pistonphone and a LS1P microphone are used for secondary calibration of microphones by the substitution method with an expanded uncertainty from 0,06 dB up to 0,14 dB at 250 Hz. Results from measurements and theoretic evaluations of SPL with pistonphone coupler of 20 cm3, 30 cm3 y 40 cm3 are presented. Ambient pressure, coupler volume effects on the SPL determination and the ambient pressure influence on the microphone sensitivity using two different methods are also studied. In addition, some preliminary results on modeling and simulation of the dynamical behavior for the air volume inside coupler cavities are presented. Principal error sources on a calibration by the substitution method are ambient pressure correction, uncertainty of the reference microphone’s sensitivity and volume corrections. Influence of the above error sources is taken into account on the associated uncertainty budget for a microphone calibration. Internoise 2001 Abstracts 273 Session C7-2, 14:00 h Paper number 79 (Invited, Oral) Ole-Herman Bjor, Jørund Enger and Bjørn Winsvold Norsonic As, PO Box 24, N-3421 Lierskogen, Norway SOUND INTENSITY FOR IDENTIFICATION OF AIRCRAFT NOISE I-INCE Class 72 The paper describes an outdoor noise-monitoring system where the information from a threedimensional sound intensity probe is used to distinguish aircraft noise from different other noise sources. The method is particularly useful if there is one dominating source at a time. The system is used to calculate the amount of noise related to aircraft activity in an environment where other noise sources are present. Directional and spectral information is processed in a neural network to obtain automatic identification of the type of noise. Session C7-2, 14:20 h Paper number 567 (Invited, Oral) I-INCE Class 72 Y. Nakajima, K. Fukushima, A. Yokota, H. Tsukioka, K. Makino, H. Ochiai and I. Yamada Rion Co., Ltd, 3-20-41, Higashi-motomachi, Kokubunji, Tokyo 185-8533 JAPAN UNATTENDED MONITORING AND SOURCE IDENTIFICATION OF LOW FREQUENCY IMPULSE SOUND This paper describes an experience in developing an unattended monitoring system of low-frequency impulse sound due to artillery firing. In Japan, environmental effect of artillery sound is evaluated using LCE and LCeq, and our concern was concentrated on the reduction of wind-noise and on the detection of impulse events. However it is difficult to distinguish impulse sound from wind-noise automatically and to detect it immersed in wind-noise with ordinary windscreen. We designed a new type of windscreen that is available for long-term unattended monitoring. Using this windscreen, we investigated how to detect events of low frequency impulse sound. The new windscreen enables to reduce wind-noise. We have developed the monitoring system that can detect impulse sound automatically and that can measure LCE, LCeq and LAeq simultaneously. The parameters that is set to the monitoring system are composed of time duration of impulse sound and time history of LCeq. We also examined how long the threshold of duration time is suitable as well as how long averaging time is adequate to get LCeq so as to detect impulse sound successfully. We could know the effect impulse sound on environmental noise by comparing LCE, LCeq and LAeq mutually. The paper says also about the design of the new windscreen as a trial product and about a result of long-term outdoor experiment using this device and the monitoring system. Session C7-2, 14:40 h Paper number 720 (Contributed, Oral) M.P.M. Luykx and M.L.S. Vercammen Adviesbureau Peutz & Associés, Postbus 66, 6585 ZH Mook, the Netherlands REFLECTIONS IN ANECHOIC ROOMS I-INCE Class 73 Anechoic rooms have to fulfil anechoic conditions to ensure that measurements are taken in a (almost) perfect free field. The practical use of the room requires the presence of potentially reflecting elements that disturb the decaying free field. ISO 3745 prescribes a qualification method for the anechoic room with continuously moving microphone. However many anechoic rooms are actually qualified using discrete measuring points. Measurements were made in several rooms with continuously moving 274 Internoise 2001 Abstracts microphone and maximum length sequence measurements. A room was designed with minimum reflections to fulfil the requirements of ISO 3745. It was found that the described method in ISO 3745 is very sensitive to low level reflections. The measured rooms show significant deviations from the allowable differences between measured and theoretical free field sound levels. The anechoic room that was designed did fulfil the requirements. Although most anechoic rooms will be sufficiently fit for their purpose, it is expected that many of these rooms do not comply to the qualification method in ISO 3745. The discrete point’s method however is insufficient for qualifying anechoic rooms. POSTERS Session C7-2, 15:00 h Paper number 137 (Contributed, Poster) Posterboard CP-36 Thursday, Author present 15:20 – 16:00 h Gunnar Lundmark Lundmark Akustik & Vibration, Langelandsgatan 27, S-164 43 Kista, Sweden ACUVIB - A TOOLBOX FOR ACOUSTIC AND VIBRATION ENGINEERING I-INCE Class 75 Engineers that are working with acoustic and vibration (as myself) now and then use formulas and information from books such as Beranek "Noise and Vibration Control", Blevins "Formulas for Natural Frequency and Mode Shapes", B&K "Acoustic Noise Measurements", "Mechanical Vibration and Shock Measurements", Handbooks and International Standards. It is time consuming to find the relevant formulas and keying in the formulas and numbers on calculator when they are needed. There is also an obvious risk for mistakes when keying in data on a hand-held calculator. The aim has been to collect information and formulas in a series of computer programs for practical use. The programs have been developed in 32-bit Visual Basic for Windows 95/98/NT. The AcuVib program consists of a Main-program (a menu), and more than 50 Sub-programs. Examples of Sub-programs are: dB converter, A-C weighting of spectra, sound power - sound pressure calculations for outdoor and indoor conditions, barrier, composite transmission loss (mass law), reverberation time, resilient skin, Helmholtz resonators, cut-on frequency for pipes, expansion chambers, Aeolian tone, accelerationvelocity-displacement lin-log relationship for vibration, vibration isolation, resonance frequencies of dual systems, damping relationship, resonance frequency of plates and beams, shock isolation, unbalance force, fan noise, firing frequency of combustion engines, blow off noise, dynamic data of solid materials, point mobility, critical frequency, bending wavelengths etc. Background information and examples of engineering problems where the programs have been used are presented. Session C7-2, 15:00 h Paper number 327 (Contributed, Poster) Posterboard CP-37 Thursday, Author present 15:20 – 16:00 h M. Ruiz, J.M. Lopez and M. Recuero I-INCE Class 78 Dpto. Sistemas Electrónicos y de Control, (INSIA) Universidad Politécnica de Madrid, Spain A CLINICAL AUDIOMETER IMPLEMENTATION BASED ON DIGITAL SIGNAL ANALYZER AND A LABVIEW APPLICATION In this paper it will be presented a new hardware and software architecture for the implementation of clinical audiometers. Concretely it will be showed: - A hardware prototype based in a digital signal Internoise 2001 Abstracts 275 analyser and a custom signal low noise conditioner for audiometric transducers; - The experimental results obtained using artificial ear and artificial mastoid; - A new software model for local and remote testing and diagnostic of audiometric trails. Session C7-2, 15:00 h Paper number 491 (Contributed, Poster) Posterboard CP-38 Thursday, Author present 15:20 – 16:00 h Arturo O. Santillan and Morten Lydolf I-INCE Class 73 Department of Acoustics, Aalborg University, Fredrik Bajers Vej 7-B4, Dk-9220 Aalborg Øst, Denmark USE OF FIR FILTERS TO EXTEND THE FREQUENCY RANGE IN A TEST CHAMBER FOR LOWFREQUENCY NOISE AND INFRASOUND EXPOSURE For the study of noise and its effects on human beings, recorded signals are often played back in a test chamber under controlled conditions in laboratory experiments. Obviously, the effects of the loudspeakers and the enclosure on the reproduction of these signals are desired to be significantly reduced. Currently a test chamber is being arranged to extend the frequency range where it can be used. The aim is to obtain a flat frequency response in a continuous three-dimensional region, which should include the volume around the subject and be large enough to allow for his/her movements. The implementation of a preliminary system of sound equalization is presented in this paper with a general description of the test chamber. For this chamber of 2.7x2.5x2.4 cubic meters, a practically flat frequency response was originally obtained from 0.05 Hz to approximately 30 Hz in a pressure field condition. For the implemented sound equalization system, a free field approximation has been followed, and FIR filters have been used to modify the input signal before it is fed to the loudspeakers. The results show that the frequency range with flat frequency response can be extended to approximately 180 Hz using the FIR filters. Moreover, the resulting zone of equalization occupies a considerable part of the complete volume of the chamber. Session C7-2, 15:00 h Paper number 589 (Contributed, Poster) Posterboard CP-39 Thursday, Author present 15:20 – 16:00 h R. Raangs, W.F. Druyvesteyn and H-E. De Bree University of Twente, P.O. box 217, 7500AE Enschede, The Netherlands I-INCE Class 71 A NOVEL TWO-DIMENSIONAL SOUND PARTICLE VELOCITY PROBE FOR SOURCE LOCALIZATION AND FREE FIELD MEASUREMENTS IN A DIFFUSE FIELD. Sound can be described as variations in sound pressure and sound particle velocity. Sound pressure is most commonly used for measuring acoustical parameters of rooms and sound sources. Sound velocity however yields more acoustical information about rooms and sources simply because the velocity is a vector quantity, whereas the pressure is only a scalar. Another important aspect is the fact that the diffuse field is not measured with a velocity sensor the same way a microphone would, but only partially because of the directivity. Combinations of more than one velocity sensor can further increase the ratio between direct and diffuse sound, and some combinations can even be proven to contain no diffuse sound at all in case of an ideal diffuse sound field. We used two velocity sensors so that we are able to measure two independent sound velocity vectors that are projected on a surface. These two signals can be used to obtain more information about position of the source of the incoming sound signal, and further the two signals yield information about the direct and indirect sound. By 276 Internoise 2001 Abstracts using the cross-correlation of the two signals, the signal to noise ratio is also improved largely compared with only one sensor. With use of this technique we are able to measure direct sound field radiated from an acoustical source in a reverberant room, the position of the sound source, and acoustical properties of the room. The room should however approximate an ideal diffuse sound field; but this property can be examined with the use of two velocity sensors. Session C7-2, 15:00 h Paper number 774 (Contributed, Poster) I-INCE Class 61 Posterboard VIP All days, Author present Thursday 15:20 – 16:00 h Per V. Brüel Brüel Acoustics, Denmark IS A-WEIGHTING OF NOISE CORRECT? It is explained how the A-weighting does not have the increased amplification for frequencies around 2-3 kHz. Why we use A-weighting for all levels is still a mystery. The consequence is that lower frequencies are underestimated. By using D-weighting it is possible to simulate Zwicker for continuous noise in the levels from 50 to 100 dB(A). Session C8-1: Nearfield Acoustic Holography Chaired by Julian Maynard and Augustinovicz Session C8-1, 9:40 h J.D. Maynard Paper number 96 (Tutorial, Oral) I-INCE Class 72 Penn State University, 16802University Park, Pa, USA NEARFIELD ACOUSTIC HOLOGRAPHY: A REVIEW Nearfield Acoustic Holography (NAH) has undergone significant development since its original conception. A review of the basic concepts and new developments will be presented in this lecture. Concepts include the role of level surfaces of separable coordinate systems, hologram data processing concerns, evanescent wave and uniqueness problems, regularization and the utility of singular-valuedecomposition, and extensions for treating arbitrarily shaped source and hologram surfaces. The last topic includes the boundary element method, the equivalent source method, and eigenfunction expansions. It is intended that this lecture will provide background material for other talks on NAH and related subjects. Session C8-1, 10:20 h Paper number 445 (Invited, Oral) H. Houjoh and S. Ohshima I-INCE Class 72 Precision and Intelligence Laboratory, Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama 226-8503, Japan SOME PROBLEMS OF NEAR FIELD HOLOGRAPHIC MEASUREMENT WITH A LIMIT SIZED HANDY 2-D ARRAY FOR AN ON-SITE MACHINE NOISE ASSESSMENT Holographic imaging is effective for sound source localization of a machine. However, it is difficult to bring it into a laboratory room such as an anechoic chamber because the noise problem comes out at an individual operating condition on the site. Thus it is worthy to bring an array measurement Internoise 2001 Abstracts 277 instrument to the site where it suffers disturbance due to either reflection or other noise sources. Nevertheless, if the measurement is done real time, like video, we can guess what is happening by “looking” into the object as well as by trial and error treatment at a time even though the image is a little vague. For this purpose, authors had proposed a 2-D array with a limited size of about 500mm square, and with a limited number of microphones of 8 by 8, which can be handled easily near the object machine. The proposal is based on the idea that if the array is closely placed in front of the object, the sound of interest becomes strong enough, and hence the effect of disturbance can become relatively small. Under the condition mentioned above, the effect of array insertion into the field on the measured sound pressure is examined and discussed. And the validity of NAH method is also investigated with a comparison to the conventional holography. Finally the sound identifying methodology for a complex shaped machine is proposed by using a simplified source-and-obstacle model which represents the sound diffraction. Session C8-1, 10:40 h Paper number 541 (Invited, Oral) Andreas P. Schuhmacher and Per Christian Hansen SOUND SOURCE RECONSTRUCTION USING INVERSE BEM I-INCE Class 72 Brüel & Kjær Sound and Vibration Measurements A/S, Skodsborgvej 307, DK-2850 Nærum, Denmark We describe a computational system for reconstruction of sound sources from data collected in the near field of the source. The system is based on a boundary element model that expresses the acoustic field by means of boundary vibration variables. By solving the associated inverse problem, we are able to determine the vibrations from the acoustic field data. Tikhonov regularization is imposed in order to stabilize the solution, and the regularization parameter is chosen by means of the L-curve criterion, which is based on a plot of the regularized solution's norm versus the norm of the corresponding residual. We demonstrate that our algorithm is able to reconstruct sound sources with good spatial resolution, and we show that the L-curve criterion is superior to generalized cross-validation in the presence of correlated noise. Finally, we outline how to detect and handle irregular frequencies in an efficient manner. The project is a collaboration with Brüel & Kjaer Sound and Vibration, Denmark, where most of the actual work has taken place. More details and images can be found at http://www.imm.dtu.dk/~pch/Projekter/acoustic.html Session C8-1, 11:00 h Gee-Pinn James Too Paper number 599 (Invited, Oral) I-INCE Class 72 Department of Naval Architectural and Marine Engineering, National Cheng Kung University, Tainan, Taiwan 70101, R.O.C. APPLICATION OF SOURCE-DISTRIBUTED METHOD FOR ACOUSTIC HOLOGRAPHY Source-distributed methods were developed earlier by several researchers who used different names for the methods and different source allocations in their approaches. However, the basic theorem is basically the same in these methods. In the present study, the applications of source-distributed methods are illustrated for the estimations for radiation and scattering sound fields. In addition, it gives good estimations for exterior sound fields as well as for interior sound fields. Furthermore, the method gives good acoustic holograms in the far field via a near-field measurement. Even with some random errors in the near-field measurement, one can still obtain a fair estimation in the far field [work supported by NSC of ROC]. 278 Internoise 2001 Abstracts Session C8-1, 11:20 h Paper number 605 (Invited, Oral) Courtney B. Burroughs The Pennsylvania State University, State College PA 16804 USA I-INCE Class 72 APPLICATIONS OF NEARFIELD ACOUSTICAL HOLOGRAPHY IN THE RESEARCH OF MECHANISMS OF ACOUSTIC RADIATION Three examples of the application of nearfield acoustical holography (NAH) are presented to illustrate the use of NAH in the research of acoustic radiation mechanisms. In the first example, a mathematical simulation of NAH in a wind or water tunnel is used in the reconstruction of the freefield acoustic radiation from simulated measurements inside the tunnel. NAH is then applied in the measurement of the acoustic radiation from a violin. In this example, a bowing machine is used to excite the violin and meaurements conducted over planar surfaces that surround the violin. In the last example, NAH measurements are made in the acoustic field of automobile tires rolling on highway pavement surfaces. Partial coherence with three reference microphones is used to complement NAH spatial localization in the separation of the sources of tire/pavement interaction noise. Session C8-1, 11:40 h Sean F. Wu Paper number 680 (Invited, Oral) I-INCE Class 72 Department of Mechanical Engineering, Wayne State University Detroit, MI 48202, USA RECONSTRUCTION OF SOUND RADIATION FROM A VEHICLE FRONT END This paper presents the Combined Helmholtz Equation Least Squares (CHELS) method for visualizing the radiated acoustic pressure fields from a general, three-dimensional complex vibrating structure in free space. The structure under consideration emulates a full-size four-cylinder engine. To simulate sound radiation from a vibrating structure, harmonic excitations are assumed to act on arbitrarily selected surfaces. The resulting vibration responses are solved by the commercial FEM software IDEAS. Once the normal component of surface velocity distribution is determined, the surface acoustic pressures are calculated using standard boundary element method (BEM) codes. The radiated acoustic pressures over several planar surfaces at certain distances away from the source are calculated by the Helmholtz-Kirchhoff integral formulation. These field pressures are taken as the input to the CHELS formulation to reconstruct acoustic pressures in the entire field, including the source surface. The reconstructed acoustic pressures thus obtained are then compared with the benchmark values. Numerical results demonstrate that the CHELS method can yield satisfactory reconstruction of acoustic radiation from a complex vibrating structure in a cost-effective manner. For the same level of accuracy and spatial resolution, the CHELS method requires much less field acoustic pressure measurements than the Helmholtz integral formulation does. Also, for the same amount of the pressure input, the CHELS method can provide a better reconstruction than the HELS method does. [Work supported by a grant from the U.S. National Science Foundation.] Internoise 2001 Abstracts 279 Session C8-1, 12:00 h Paper number 280 (Contributed, Oral) I-INCE Class 23 Marcelo Bruno S. Magalhaes, Roberto A. Tenenbaum and Moyses Zindeluk Acoustics and Vibration Lab., Federal University of Rio de Janeiro, C.P. 68503, 21945-970, Rio de Janeiro, Brazil APPLICATION OF EQUIVALENT SOURCES METHODS IN ACOUSTICAL HOLOGRAPHY Equivalent source methods, also known as wave superposition or simulated source methods, have long been used in the treatment of acoustic radiation. Heckl, Koppmann and Ochmann, among others, have studied the many variations of such methods. The technique relies upon the replacement of the actual source by a set of simple sources, adequately distributed in its interior. Many works have been published about the advantages and drawbacks of particular distributions, concerning the radiation and/or scattering problem. Less attention has been given, however, to the application of such methods to the inverse radiation problem. The works of Wu et al. and Stepanishen present the Helmholtz Least Squares (HELS) method and the Generalized Internal Density (GISD) method, which are variations of the equivalent source methods with all sources located at one point (HELS) or distributed over an axial ring-source distribution (GISD). Both techniques require the resolution of Fredholm equations of first kind, usually ill-posed. This work presents another variation of equivalent source method, based on the full-field equations developed by Ochmann, which lead to Fredholm equations of second kind, with better numerical properties. This goal is achieved by minimizing the discrepancy between the pressure generated by the set of equivalent sources and that measured on the hologram, using the weighted residual method. The weighting functions are the derivatives with respect to the hologram’s normal of the pressure generated by each of the equivalent sources. The weighted residuals statement results, after discretization, in diagonally dominant matrices, leading to a better numerical behaviour. Session C8-1, 13:20 h Paper number 282 (Contributed, Oral) G. Geentjens and G. Vermeir I-INCE Class 72 Laboratory of Building Physics and Laboratory of Acoustics, Catholic University of Leuven, Celestijnenlaan 200D, B-3001 Heverlee, Belgium APPLICATION OF ACOUSTICAL HOLOGRAPHY TO THE STUDY OF STRUCTURE-SOUND INTERACTION IN BUILDINGS ACOUSTICS One of the major topics in building acoustics is the investigation of air-borne sound insulation and structure-borne sound transmission and radiation in building constructions and of building elements. The methods used for the measurement of sound insulation are described in several national and international standards. These prescribed methods are tools for the control of the acoustical performance of products in the laboratory or of the global field performance on the building site. A more detailed investigation of constructions asks for measuring techniques relating the radiated sound field to the vibrations on the structure. Acoustical holography solves this problem. In fact, following the theoretical developments, it is possible to obtain the sound field as well as the vibration of the radiating object from the measurement data in a single plane. Based on the theory of acoustical holography, M. Villot created a test set-up in which sound is radiated in an acoustically ‘semiinfinitive’ room with rectangular cross-section and perfectly reflecting walls. A scale model of such a test set-up is built at the Laboratory of Building Physics and the Laboratory of Acoustics of the Catholic University of Leuven. The cross section of this set-up is 1.2x1.5 m2. In this paper, the instrumentarium of the set-up is described and the first measurement results are presented. 280 Internoise 2001 Abstracts Session C8-1, 13:40 h Paper number 321 (Contributed, Oral) Chris Talbot and John D. Fieldhouse University of Huddersfield, Queensgate, Huddersfield HD1 3DH, United Kingdom I-INCE Class 72 MATHEMATICAL REPRESENTATION OF DISC BRAKE SQUEAL USING LASER HOLOGRAPHY The problem of disc brake noise or squeal has been investigated by studying real systems using the technique of double pulsed laser holography. By recording three orthogonal high resolution images of a system generating noise, considerable information has been obtained of the mode of vibration of the disc, pads and other components. The method has been refined by using electronic triggering devices which enable a series of time related holographic images to be obtained, corresponding to successive times in the cycle of vibration of the braking system. Using digitised versions of the images it has been possible to interrogate the fringe patterns of the time related holographic interferograms and hence construct a mathematical representation of the vibrating surface of both the disc and pad. This has enabled the authors to construct a three dimensional animation of a real disc brake generating noise. It also makes possible the construction of a Fourier Series type representation of the vibrating surfaces which it is believed will be of importance in developing realistic mathematical models of disc brake squeal. Session C8-1, 14:00 h Paper number 533 (Contributed, Oral) M.R. La Grange and N.B. Roozen Philips Centre for Industrial Technology (CFT), Eindhoven, The Netherlands I-INCE Class 72 THE USAGE OF NEARFIELD ACOUSTIC HOLOGRAPHY" FOR SOUND SOURCE LOCALISATION APPLIED TO SMALL OBJECTS" With the increasing demand for low noise emission of consumer products, in noise control engineering advanced measurement tools are occasionally required. Within Philips CFT Nearfield Acoustic Holography (NAH) is used to assist in the identification of structural as well as air-borne emission sources. Philips wants to apply this technique on relative small objects like razors, CD-ROM drives, et cetera. Therefore a high resolution holography measurement system has been realised that is capable of localising sound sources on small sized objects. With the measurement technique sound sources of approximately 6mm can be localised. The session will elaborate on the results of NAH-reconstructions applied to small sound sources. Session C8-1, 14:20 h Paper number 711 (Contributed, Oral) Jean-Hugh Thomas and Jean-Claude Pascal I-INCE Class 72 Laboratoire d’Acoustique de l’Universit´e du Maine (UMR CNRS 6613), Ecole Nationale Supérieure d’Ingénieurs du Mans, rue O. Messiaen, 72085 Le Mans, France USING WAVELETS TO REDUCE DISTORTION PROBLEMS IN NEAR FIELD ACOUSTIC HOLOGRAPHY Near field acoustic holography is an inverse technique for reconstructing the sound field at the sound source. As with any inverse method, near field acoustic holography presents problems of stability which must be solved by various techniques. For plane holography, the problem is well identified with the evanescent waves in the wavenumber spectrum. The delicate point is, then, exponential amplification of these components in the reconstruction procedure which leads to significant Internoise 2001 Abstracts 281 distortions when highly corrupted by noise. The solution to the problem is to filter the evanescent components of a high wavenumber or to use more selective regularization methods. However, the problem may be exaggerated if the wavenumber spectrum of the sound field collected by the microphone antenna (hologram) is inaccurate. The leakage effect due to the truncation of the sound field by an antenna of finite size and under-sampling are problems that it is difficult to solve in acoustic holography. The consequences of the additional distortions of the wavenumber spectrum affect the evanescent waves in particular. In this article, processing methods based on wavelet packet analysis of the sound field collected by an antenna of microphone are evaluated. It is shown that the methods provide a selective treatment which allow the reduction of the troncature effect before using the space Fourier transform. Examples from simulated cases and from engineering measurements show the advantage of this technique and its importance in practical application of near field acoustic holography. POSTER Session C8-1, 14:20 h Paper number 307 (Contributed, Poster) Posterboard CP-40 Thursday, Author present 15:00 – 16:00 h A. Versluis, C. Azais, J.P. Guilhot and M. Aubes I-INCE Class 72 L.A.M.I. – Université Paul Sabatier TOULOUSE III, Bâtiment 3R1B2 - 118, route de Narbonne 31062 TOULOUSE Cedex France NEARFIELD ACOUSTIC HOLOGRAPHY APPLIED TO ACOUSTIC FIELD RADIATED BY A CYLINDRICAL SOURCE The High frequency supply of a plasma filled lamp induces at any frequencies an acoustic resonance phenomenon inside the lamp. These resonance frequencies are linked to the temperature and the pressure of the gas and also to the geometry of the lamp. Apart from these resonance frequencies, the discharge connects the electrodes as a straight line. At resonance frequencies, we can hear a powerful sound and notice a vibration of the tube, as we can see a distortion of the discharge arc which brush the cylinder wall. This phenomenon induces destruction of the lamp. The aim of this paper is the study of a method to characterise the field radiated by a cylindrical source. We will carry out acoustic measurements near the outside of the cylinder, and backward propagate it to the acoustic source. We propose a method based on the Nearfield Acoustic Holography to solve the inverse problem. In order to validate this method, we have checked off cylinder's resonant frequencies. We have carried out measurements using an accelerometer. We have compared these results with results obtained using Nearfield Acoustic Holography. We can find the vibratory behaviour of the cylinder, although interferences who disturb the image. They will be treated by filtering. 282 Internoise 2001 Abstracts Authors index Aasvang, Gunn Marit, 153 Abbott, P.G, 227 Abe, Tadashi, 243 Abe, Yufuko, 175 Abraham, S, 241 Abramian, Andrei, 120 Abramowitch, Jean-Marc, 109 Abukhader, Sajed, 150 Adachi, Zempachiro, 128 Adams, Keith, 208 Adams, Marleen, 163 Agleitti, G.S, 33 Aguas, Artur, 54 Ahn, Jaekwon, 77 Aimasso, Roberto, 30 Akagi, Masato, 44 Akdag, Nese Yugruk, 108 Akishita, Sadao, 142 Akita, Takeshi, 162 Aknesil, Ayse Erdem, 39 Albert, Donald G, 220 Ali, Sayed A, 18 Allard, J.F, 185 Allevard, Thomas, 71 Almeida, David De, 15 Alves-Pereira, Mariana, 52, 54, 55 Amadio, P, 250 Andersen, Palle, 223 Anderson, Grant S, 103 Andersson, A.O, 201 Andersson, Patrik, 126 Andow, Kei, 263 Andronov, Ivan V, 267 Andrzej, Mikulski, 269 Angelo, Onorati, 116 Antila, Marko, 25, 30 Antoine, H, 142 Aono, Shoji, 240 Aquino, Igor A, 97 Arana, Miguel, 57, 67 Araujo, Amilcar, 52, 55 Argentino, Marcos Antonio, 116 Arizo, Juan V, 20 Arner, Wayne, 212 Asai, Kikuo, 236 Asselineau, M, 196 Atmoko, H, 141 Attenborough, Keith, 210, 219 Aubes, M, 282 Augustin, P, 186 Augusztinovicz, Fülöp, 165, 195 Auweraer, Herman Van der, 73, 79 Avdeev, Ilya V, 93 Azais, C, 282 Babisch, Wolfgang, 151 Bai, Mingsian R, 23 Baistrocchi, Carlo, 217 Bajaj, Anil K, 56 Balant, Anne C, 61 Ballagh, Keith, 264 Baranov, S.N, 183, 199 Bareille, Olivier A, 265 Barham, Richard, 223 Barrett, Douglas E, 103, 213 Barriere, Nicolas, 109 Barrigón, Juan M, 22 Barsikow, Bernd, 174 Bartel, Hans-Holger, 19 Barti, Robert, 161 Bartolini, Ruggero, 121 Barwicz, W, 214 Basner, M, 155 Baumgartner, Thomas, 15, 16 Beckenbauer, Thomas, 126, 128 Bécot, François-Xavier, 108, 126 Beek, Pieter van, 70 Begic, Edin, 235 Bekebrede, Gerard, 89 Bendtsen, Hans, 10, 132 Benedicto, Miguel A. Luzón, 164 Bengtsson, Johanna, 245 Bérengier, Michel, 101, 102, 133 Berestovitsky, E, 268 Berg, Frank H. A. van den, 218 Berg, Frits (G.P.) van den, 246 Berg, M. van den, 64 Berge, Truls, 100 Berglind, N, 152 Berglund, Birgitta, 148, 150 Berkhoff, Arthur P, 136, 141 Berlo, D. van, 192 Berry, B, 156 Bertolini, Claudio, 82 Bertrand, Eric, 119 Besnard, F, 103 Betti, R, 250 Beveridge, Carl, 72 Bhandari, Nimesh, 114 Bianchi, M, 180 Biassoni, E.C, 241 Birch, Robert, 252 Birgersson, Fredrik, 221 Birlik, Gülin, 50 Birta, J.A, 87 Bisio, Giacomo, 121 Bite, Maria, 260 Bite, Pál Zoltán, 165, 260 Internoise 2001 Abstracts 283 Bittencourt, Jorge B. Pereira, 249 Bjor, Ole-Herman, 274 Bjørhus, M, 107 Björk, Erkki, 243, 244 Blanchet, Denis, 95 Blokland, Gijsjan van, 124, 133 Bloomfield, Alan, 254 Bluhm, G, 152 Bly, Stephen H.P, 242 Bobrovnitskii, Yuri I, 69 Boczar, Tomasz, 270 Bohineust, Xavier, 83 Bolton, Stuart, 127 Bonacker, Margit, 11 Bontomase, Anthony R, 211 Boone, Marinus M, 86, 179 Boonen, René, 24 Borchers, I, 142 Borchi, Francesco, 17 Boroditsky, Leo, 187 Borst, Jeroen C, 255 Botteldooren, Dick, 167, 216, 259 Bourbon, Christine, 255 Bouter, P, 90 Bouty, E, 142 Bradley, John S, 42, 87 Branco, Nuno A.A. Castelo, 52, 54, 55 Bree, H-E. de, 276 Brentnall, Sarah L, 36, 154, 156 Brincker, Rune, 223 Bris, Petr, 186 Bronkhorst, A, 272 Brothanek, Marek, 26 Brucker, Michel, 231 Brüel, Per V, 277 Bruijn, A. de, 188 Buchta, E, 219 Buck, K, 232, 235 Buess, H, 155 Buetikofer, R, 204 Buijs, J.H.N, 197 Bujnoch, Ludek, 186 Bulder, Jan-Willem, 248 Bullinger, Monika, 157 Buna, Bela, 174 Burdisso, Ricardo A, 139, 144 Burgess, Marion, 59, 66 Burroughs, Courtney B, 211, 279 Busink, J.J, 92 Buss, Sandra, 129 Cabanellas, Susana, 89 Cabos, Christian, 188 Capdevila, R, 32 Carfagni, Monica, 17 Carignani, A.J, 241 Carme, Christian, 232, 233 Carter, Norma, 47, 160 Carvalho, António P. O, 191 Castelo Branco, Nuno A.A, 54 Catron, F.W, 121 Cazzolato, Ben S, 140 Çelik, Emine, 194 Celma, Javier, 164 Cerchiai, Mauro, 133 Cesini, Gianni, 91 Chan, K.S, 88 Charchalis, Adam, 268 Chen, Guoyue, 236 Chen, Mingkun, 120 Chen, W.C, 202 Cheong, Cheolung, 247 Cheung, Wan-Sup, 53 Chiu, Vincent, 242 Cho, Kyungseok, 78 Choi, Byung-Jae, 53 Choi, Young-Chul, 181 Chou, C.S, 134 Chung, J, 136 Chyla, Andrzej, 203 Cieslik, Jacek, 264 Claesson, Ingvar, 30, 33, 146 Clow, Angela, 244 Colon, Halima, 52 Corbisher, Chris, 212 Cornelis, W, 216 Costa-Félix, Rodrigo P.B, 97 Cremers, L, 82 Cummings, Alan, 113, 219 Curcuruto, S, 250 Czechowicz, M, 214 Dahl, Mattias, 146 Danneskiold-Samsøe, Ulrik, 174 Dassen, A.G.M, 47 Davies, Patricia, 56 Deblauwe, Filip, 163 Defrance, Jérôme, 102, 109 Degen, K.G, 180 Degn, Uffe, 174 Deigaard, Lars, 177 Deijen, J.B, 157 Dekoninck, L, 259 Delzanno, Paolo, 30 Desarnaulds, Victor, 191 Desmet, W, 221 Díaz, Esteban Gaja, 260 Diego, Maria de, 238 Dillman, A, 182 Dillmann, A, 200 Dittrich, M.G, 172 Dobrzynski, W, 91 Dodd, George, 37 Doelman, N.J, 27, 31 Dolmans, J.H.J, 47 Dombi, István, 195, 260 Dongen, Jef E.F. van, 159 Dool, T.C. van den, 105, 179 Dowling, Ann P, 125 Drutowski, Carol, 61 Druyvesteyn, W.F, 276 Duarte, Maria Lucia Machado, 249 Dubbink, David T, 258 Dubois, Danièle, 166 Duhamel, D, 130 284 Internoise 2001 Abstracts Dushin, Alexandr Yu, 265 Dussaugey, Cosette, 62 Ebata, Masanao, 234 Eekhout, Félippe van, 13 Efimtsov, B.M, 183, 201 Eguchi, Kei, 43 Ehrenfried, K, 182, 200 Ejsmont, Jerzy A, 226 Elbers, Frank, 9, 11 Elliot, Stephen, J, 138, 237 Emmen, H.H, 157 Enblom, Roger, 172 Engdahl, Bo, 58, 153 Engel, Zbigniew, 266 Enger, Jørund, 274 Enghardt, Lars, 142, 144 Errasti, Javier, 202 Estorff, Otto von, 80, 126 Eun, Hee-Joon, 53 Evans, Gary W, 157 Evans, Jack B, 109 Evans, Phil, 244 Fabozzi, C, 250 Fadavi, A, 130 Fagerlund, A.C, 121 Fedtke, T, 272 Ferguson, Neil S, 193, 221 Ferrer, Miguel, 238 Fieldhouse, John D, 72, 281 Fillery, Michael, 257 Filomena, Vincenzo, 211 Finegold, Lawrence S, 65 Finne, Per, 177 Finnveden, Svante, 221 Fischer, P.H, 157 Fischer, Raymond, 187 Fisher, Nick, 36 Fleming, Gregg G, 204 Foch, James D, 169 Forssén, J, 147 Fourcade, Patrick, 168 Fox, Colin, 264 Fraanje, P.R, 27 Fragata, Jose I.G, 54 Franco, G, 241 Franssen, A. Ellis M, 41 Friot, E, 234 Fujita, Hajime, 118 Fujiwara, Kyoji, 175, 215 Fukushima, Kenji, 273, 274 Fung, K.Y, 88 Furihata, Kenji, 38, 40, 67 Fürst, N, 259 Gabriel, S.B, 33 Gabriels, D, 216 Gade, Svend, 223 Galis, Stephan P, 86, 209 Garcia, Amando, 57, 67 Garcia-Bonito, J.J, 238 Gardonio, P, 138 Gardziejczyk, Wladyslaw, 227 Garesci, Francesca, 73 Gasser, Stéphane, 229 Gatto, Saverio F. Del, 211 Geentjens, G, 280 Gélat, Pierre, 223 Gerges, Samir N. Y, 252 Gerretsen, Eddy, 128, 130, 160 Giancarlo, Ferrari, 116 Giardi, Hélène, 83 Giua, P, 272 Gjestland, Truls, 147 Gladilin, U, 268 Godfrey, N.S, 133 Gómez, Valentín, 22 Gonneau, E, 28 Gonzalez, Alberto, 238 González, Alice Elizabeth, 260 Graaf, Theo M. de, 239 Graaff, D.F. de, 22 Graf, Regina A.G, 125 Graha, Will R, 125 Grande, Nuno R, 54 Grandi, Fabio, 206 Grange, M.R. La, 281 Granneman, Jan H, 63 Grantham, David, 251 Gregg, Martin, 225 Groothoff, Beno, 251 Grzadziela, Andrzej, 268, 269 Guarnieri, A, 250 Guerrand, Sylvie, 15 Guilhot, J.P, 28, 282 Gulding, John M, 205 Guski, Rainer, 40, 41, 44 Guyader, J.L, 94 Hafidi, Ali El, 262 Haines, Mary M, 36, 154, 156 Hakansson, L, 33 Hal, B. van, 221 Halén, Ismo, 225 Halkyard, C. Roger, 35, 135 Hallmann, S, 41, 44 Hallows, Rob, 111 Hamery, P, 232, 235 Hamet, Jean-François, 100, 125 Hansen, Colin H, 140 Hansen, Per Christian, 278 Hardy, Andrew E.J, 175 Hase, Shigeo, 230 Hashimoto, Takeo, 161 Hatano, Shigeko, 161 Hatfield, Julie, 42, 46, 47, 149, 160 Haven, H.B.G. Ten, 90 Hayashi, A, 68 Hayek, Sabih I, 138 Head, J.A, 156 Hecht, K, 152 Hecq, W, 16 Hedegaard, Peter, 271 Heidebrunn, Frank, 19 Heijden, John van der, 207, 220 Heimerl, E.H. Gerhard, 13 Heinig, Klaus, 145 Hellweg, Robert D. Jr, 61 Hendricx, Wim, 73 Internoise 2001 Abstracts 285 Henriksen, V, 107 Henrioulle, Kris, 139 Heo, J.W, 136 Herlufsen, Henrik, 223 Herman, Lloyd A, 212 Hernandez, J, 28 Herpe, François Van, 80 Herr, M, 201 Heutschi, Kurt, 99 Hidaka, Takayuki, 75 Himmel, Chad N, 109 Hiramatsu, Kozo, 156, 169 Hirao, Yoshihiro, 48 Hirate, Kotaroh, 222 Hirsch, K.-W, 218, 219 Hofmann, Jan, 102 Holmberg, Kjell, 158, 159 Hong, Suk-Yoon, 96, 190 Hongisto, Valtteri, 194 Honianaki, Fotini, 58 Honmachi, Yoshida, 200 Horner, J.L, 111, 222 Houjoh, H, 277 Huang, X, 111 Huber, Bernd, 257 Hübner, G, 119 Hübner, Peter, 174 Hucklebridge, Frank, 244 Hume, Ken, 225 Hussey, Robert G, 242 Huybregts, Cornelius (Neil), 106 Hwang, Woo Seok, 85 Hwang, Yoha, 237 Hygge, Staffan, 156, 157 Hyrynen, J, 77, 79 Iacoponi, A, 250 Iannotti, Antonio, 91 Ichchou, M.N, 264 Ichimiya, Ryoichi, 26, 78, 269 Igarashi, Katsumi, 88 Ih, Jeong-Guon, 114, 192 Ikeda, Toshiyuki, 128 Inaba, Toru, 165 Inoue, K, 75 Inoue, Yasunori, 142 Inversini, G, 250 Irmer, Volker K.P, 62 Ise, Shiro, 26, 233 Ishida, Riei, 51 Ishigami, Akihiro, 44 Ishii, Naoki, 128 Ishii, Ryuji, 200 Ishii, Tatsuya, 143 Ishitake, Tatsuya, 49, 50 Ising, Hartmut, 151 Ito, Takahiro, 26, 78 Iwagase, Tomoyuki, 84 Iwahashi, K, 97 Iwase, Teruo, 101 Iwata, Kentarô, 243 Jabben, Jan, 14, 17, 47 Jacques, Jean R, 63 Jaecker-Cueppers, M, 173 James, Soo H, 231 Janczur, R, 214 Jansens, G, 185 Janssens, M.H.A, 172 Jean, Ph, 208 Jellyman, Andrew, 256 Jen, Ming Une, 117 Jeng, Ming-Chang, 31 Jeon, Wan-Ho, 112 Jeon, Wan-Jo, 122 Jezequel, Louis, 264, 265 Jiggins, M, 156 Jiménez, S, 32 Jiricek, Ondrej, 26 Job, R.F. Soames, 46, 47, 149, 160 Joekes, S, 241 Johansson, Sven, 30, 146 Johnson, M.E, 140 Jokat, Jürgen, 188 Jokitulppo, Jaana, 243, 244, 247 Joly, N, 261 Jonasson, Hans G, 75, 98 Jones, Richard R.K, 175 Jong, Christ A.F. de, 70, 188 Jong, Ronald G. de, 41 Jonkhart, S, 90 Jorysz, Bach. Andrés, 260 Juárez, Alfredo A. Elías, 273 Jung, Byungsuk, 189 Just, W, 142 Kakehi, Mizue, 44 Kalivoda, Manfred T, 171, 174 Kalkman, C, 197 Kam, Yiu-kee, 267 Kamp, Irene van, 41, 157 Kanda, Kazunobu, 43 Kang, D.-J, 68 Kang, Ku Tae, 166 Karabiber, Zerhan, 194 Karczub, D.G, 121 Kårfalk, Göran, 76 Kari, Leif, 184 Karjalainen, Ari, 63, 77, 79 Kataja, Jari, 25 Kawaguchi, Masataka, 44 Kawahara, Kohji, 184 Kawamura, Takashi, 243 Keith, Stephen E, 242 Kellett, Paul, 231 Kennepohl, Fritz, 145 Kerkers, A.J, 12 Kerry, G, 252 Kestell, Colin D, 140 Khan, M. Shafiquzzaman, 162 Kidner, M.R.F, 140 Kihlman, Tor, 147, 150 Kil, Hyun-Gwon, 190 Kim, Brian, 212 Kim, Dong-Hae, 191 Kim, G.J, 131 Kim, Jaeheon, 166 Kim, Joon Hyun, 56 Kim, Kyung-Mok, 237 Kim, Seong Keol, 56 Kim, Seung-Jong, 237 286 Internoise 2001 Abstracts Kim, Seungbo, 93 Kim, Y.-H, 68 Kim, Yang-Hann, 181 Kim, Yong-Joe, 127 Kimura, Kazunori, 229, 273 Kimura, Yasumasa, 203 Kitagawa, Toshiki, 175 Kitahara, Senji, 114 Kjellberg, Anders, 159, 245 Klaeboe, R, 168 Klein, Philippe, 125 Knapp, Yannick, 119 Knight, Sarah B, 109 Kobayasyhi, Yoshihito, 230 Kodani, S, 207 Koenig, Jean-Francois, 231 Koffeman, A.I, 12 Koizumi, Takayuki, 84 Kojima, N, 262 Kollamthodi, S, 132 Koop, L, 182, 200 Koopmans, Ferry, 13, 246 Korchanov, V, 268 Koshiroi, Toshiaki, 97, 273 Koudriavtsev, V.V, 199 Kragh, Jørgen, 98, 132 Krahé, D, 236 Kravchenko, S.V, 201 Krebber, W, 272 Kristiansen, U.R, 107 Krog, N.H, 58 Kropp, Wolfgang, 108, 126, 150 Krüger, Friedrich, 174 Kruize, Hanneke, 41 Ksiazek, Marek A, 52 Kudrna, Monika, 171 Kuijpers, Ard, 124, 128, 133 Kuisma, Kaarina, 241 Kumar, B. Suresh, 84 Kumura, S, 262 Kuno, K, 68 Kuo, Chih Y, 125 Kuravsky, L.S, 199 Kuroda, Katsufumi, 240 Kurra, Selma, 66 Kuypers, Hans, 63 Kwon, Hyu-Sang, 53 L’Espérance, André, 239 Lacore, Jean-Paul, 63 Lai, Jienwen, 23 Laitinen, Heli, 241, 243 Lamotte, Lucille, 164 Lane, Steven A, 93 Lang, Judith, 254 Lang, William W, 64 Langley, R.S, 33 Lansink, Dion F. Oude, 86 Lapcik, L. Jr, 186 Larsen, Lars Ellebjerg, 10, 132 Larsson, Krister, 126 Lassche, Gert, 246 Lauriks, W, 185 Layfield, R.E, 133 Lazarev, L.A, 183 Lebiedowska, Barbara, 149 Lecce, Leonardo, 29 Lechner, Christoph, 59 Lee, H.P, 215 Lee, Ho-Won, 96, 190 Lee, Jeonghan, 77 Lee, Jong Min, 237 Lee, Jong-Hwa, 192 Lee, Joonkeun, 112 Lee, Sang-Kwon, 85 Lee, Soogab, 77, 78, 189, 247, 253 Lee, Y.W, 136 Lee, Yong-Bok, 237 Leemput, Guido van, 249 Leeuwen, J.J.A. van, 258 Lefèvre, H, 103 Lekkala, Jukka, 30 Lelong, Joël, 20, 178 Lemmen, René, 136 Lenders, Aude, 16 Leniowska, Lucyna, 137 Leniowski, Ryszard, 137 Leo, A. De, 250 Lepage, Arnaud, 80 Lercher, Peter, 167 Lesueur, Claude, 262 Letourneaux, Fabien, 166 Li, Deyu, 93 Li, Xiaodong, 112 Li, Ziaodong, 113 Licitra, Gaetano, 133, 250 Liederer, Werner, 129 Liedtke, Martin, 242 Lim, S.P, 215 Lima, Denis de, 116 Lin, Hai-Ping, 135 Lin, Tian Ran, 141 Lin, Yuanpei, 23 Lin, Yung-Chang, 31 Linden, P.J.G. van der, 79 Linet, Virginie, 80 Liu, Ku, 120 Liviu, Odobescu Grigore, 266 Lladó, Juan, 164, 202 Lomax, C, 252 Lombardi, Lorenzo, 209 Lopez, J.M, 275 Lopez, Miguel A. Picard, 230 Losa, Massimo, 133 Lu, Ming-Hung, 117 Lu, Yadong, 112 Luks, N, 155 Lundmark, Gunnar, 261, 275 Lundquist, Pär, 158, 159 Luykx, M.P.M, 274 Lydolf, Morten, 276 Lyons, R, 222 Maass, H, 155 MacDonald, John M, 212 Mace, Brian R, 35, 95, 135 Maeda, Setsuo, 48, 51 Maffei, Luigi, 211 Magalhaes, M.D.C, 193 Magalhaes, Marcelo Bruno S, 280 Mahieux, Bruno, 81 Internoise 2001 Abstracts 287 Maier, R, 142, 143 Makino, K, 274 Malker, H, 176, 178 Mangiante, Gérard, 23 Mansfield, Neil J, 49 Marciniak, Wochiech, 52 Marsh, Alan H, 272 Marsman, Gooitske, 37 Martens, Alexander, 179, 180 Martin, Maria Luisa San, 67 Marukawa, Hideaki, 273 Maschke, C, 152 Maso, Francesco Di, 29 Masoero, Marco, 174 Mast, A, 179 Matoba, Tsunetaka, 49, 50 Matsui, Toshihito, 156 Matsumiya, H, 182 Mattei, Eugenio, 91 Maurin, Michel, 45 Mawet, L, 155 Maxit, L, 94 Maxon, Christopher, 174 Maynard, J.D, 277 McCulloch, C.F, 82 McGuirk, J.J, 111 Medeiros, Eduardo Bauzer, 249 Meier, Andreas, 72 Mellert, Volkert, 129 Meloni, Tommaso, 56, 150 Méndez, Juan A, 22 Mendoza, Lorenzo M. Muñiz, 273 Menge, Christopher W, 103 Mettier, Thomas, 15 Meunier, Yves, 131 Meurers, T, 32 Michel, U, 182, 200 Micozzi, F, 250 Miedema, Henk M.E, 9 Mikami, M, 262 Mikami, Tetsuo, 228 Miller, Nicholas P, 60, 170 Minoldo, G, 241 Minorikawa, Gaku, 122 Misaki, Masayuki, 243 Mishina, Y, 68 Mitani, Yuuichiroh, 142 Miwa, Motoatsu, 181 Miyakita, Takashi, 156 Miyara, Federico, 89 Miyauchi, Toshiya, 181 Moehler, Ulrich, 39 Mohanty, A. R, 84 Molisani, Leonardo R, 139 Møller, Nis B, 223 Monay, Gilbert, 191 Monteiro, Emanuel, 54 Montrone, L, 208 Moorhouse, Andy, 71 Morgan, P.A, 132, 228 Moriya, Motohiro, 122 Morrell, Stephen, 47, 160 Mortensen, Frank Rysgaard, 177 Mosconi, Patricia, 89 Motylewski, Jerzy, 54 Muchall, Rein C, 210 Mueller, E.W, 155 Mueller, U, 155 Muellner, Herbert, 198 Mulder, Era H, 128 Mummenthey, Rolf Dieter, 255 Munn, Jacqueline M, 140 Muto, Kenji, 236 Myyryläinen, Pasi, 243 Mzali, Myriam, 166 Nackenhorst, Udo, 126 Nagai, Kenichiro, 143 Nagakura, Kiyoshi, 175 Nagy, A.B, 195 Naito, T, 97 Nakagawa, Shin-ichi, 76 Nakai, Mikio, 271 Nakajima, Y, 274 Nakasako, Noboru, 76 Nakashima, Hajime, 203 Nakayama, Yoshio, 114 Nanba, T, 215 Necati, Görgün, 30 Neise, Wolfgang, 144 Nelson, James T, 173 Neple, Pascale, 262 Neus, H, 154 Nicolson, Nancy A, 155 Niemann, H, 152 Nii, Y, 182 Nijland, Hand, 14 Nijs, L, 192 Nijsse, G, 31 Nilsson, Mats E, 148 Nilsson, N.A, 176, 178 Nishimura, Yoshitaka, 234 Nishino, H, 263 Nishinuma, Masaharu, 44 Nobile, Matthew A, 61 Nóbrega, Joao, 52 Noël, Matthieu, 94 Noël, Phillippe, 255 Nogueira, Celso Figueiredo, 116 Nordborg, Anders, 179, 180 Nordebo, Sven, 30 Nording, E, 152 Norton, M.P, 121 Nunes, Ronaldo Fernandes, 116 Nutt, David, 153 Nykänen, Hannu, 30 Ochiai, Hiroaki, 110, 274 Oertli, Jakob, 9 Ogata, Saburo, 87 Ogata, Seigo, 175 Ogata, Yuichi, 234 Ogata, Yukie, 175 Ögren, Mikael, 147 Ogura, T, 68 Oh, Geok Lian, 180, 183 Oh, Shi-Hwan, 35 Ohkawa, Shinpei, 115 Ohlrich, Mogens, 69, 263 Ohno, Shinichi, 86 288 Internoise 2001 Abstracts Öhrström, Evy, 148, 150 Ohsawa, Yoshiaki, 184 Ohshima, S, 277 Ohta, M, 207 Ohya, M, 97 Oinoma, Hideshi, 143 Okada, Y, 106 Okamoto, Nobuhisa, 48 Okubo, Nobuyuki, 128 Olafsen, Sigmund, 18 Olkinuora, Pekka, 241 Ollerhead, John B, 206 Ollikainen, Veli-Jukka, 25 Olsson, S, 33 Omiya, M, 68 Oostdijk, John P.J, 63 Orrenius, Ulf, 172 Osada, Yasutaka, 156 Oshima, Takuya, 193, 222 Oshino, Yasuo, 228 Ostendorf, Carel, 233 Pääkkönen, Rauno, 247 Padmos, Cristo J, 225 Paffen, P, 258 Page, G.J, 111 Pais, Fernando, 52, 55 Pallas, Marie-Agnès, 178 Pan, Jie, 92, 141 Pan, Min-Chun, 31 Pang, S.K, 180 Park, Changho, 78 Park, Chul-Min, 114 Park, Chung P, 231 Park, J.M, 136 Park, Kihong, 56 Park, Taeyoung, 78 Park, Young-Ho, 96, 190 Park, Youngjin, 35 Pascal, Jean-Claude, 94, 261, 281 Pasch, Vivian, 89 Paun, Florin, 229 Pavic, Goran, 71 Pavlov, A, 268 Pdzanek, Wojciech P. Jr, 266 Pedersen, Torben Holm, 253 Pedrosa, Ana, 54 Peploe, Peter, 47, 160 Perälä, Katja, 241 Pereira, Antonio de Sousa, 54 Perera, Donna, 36 Pettersson, Henrik, 12 Pettersson, L, 33 Peube, Jean-Laurent, 232, 233 Pezerat, Charles, 71 Pezzullo, Gianluca, 211 Pezzutto, Alberto, 73 Pfister, Sabina, 15 Phillips, S.M, 132, 227, 228 Picard, Miguel A, 20 Picaut, J, 101 Piehler, C, 155 Pietrzyk, Andrej, 125 Pinero, Gema, 238 Pinto, Ricardo, 82 Piromalli, W, 250 Pistek, A, 186 Plath, G, 155 Ploeg, Fokke van der, 90, 207 Plotizin, Ilja, 198 Plotkin, Kenneth J, 205 Plovsing, Birger, 105 Poisson, Franck, 166 Polack, Jean-Dominique, 166 Poll, Ric van, 155 Pollet, A.R, 241 Ponseele, Patrick Van de, 163 Popkov, Vladimir I, 70 Popp, Christian, 11, 19, 256 Potma, C.J.M, 17 Prek, Matjaz, 118 Presle, Gerard, 171 Probst, Wolfgang, 257 Pyykkö, Ilmari, 238 Quiros, Pedro E. Solana, 230 Raangs, R, 276 Rafaely, Boaz, 146 Rall, Juan C, 89 Rambelli, E, 250 Ramirez, Ricardo, 52 Ramírez, Sergio R. Rojas, 273 Rasmussen, Søren, 254 Rdzanek, Witold, 266 Recuero, M, 275 Rehfeld, Marc, 196 Relandini, Simone, 209 Renew, Warren, 66 Renterghem, T. Van, 216 Rey, E, 155 Rezek, Rüdiger, 80 Rhodes, Darren P, 206 Rich, Neil, 153 Richard, Jochen, 11 Richards, J, 248 Richter, U, 241 Rocco, Luciano, 217 Rochat, Judith L, 213 Roeloffs, A.J.C, 208 Rogers, E, 33 Roman, Amaury, 52 Roman, Rosario, 211 Romeu, J, 32 Rondeau, Jean-François, 102, 109, 208 Roo, Foort de, 86, 99, 128 Roosnek, N, 181 Roozen, N.B, 281 Roure, A, 234 Rückert, Ulrich, 13 Ruiz, M, 275 Ruotsalainen, Jouko, 247 Rust, Alfred, 10 Rylander, Ragnar, 244, 245 Ryszard Tadeusiewicz, , 203 Ryu, Jewook, 247 Ryu, Yunseon, 163 Internoise 2001 Abstracts 289 Sá, Armando Bordalo e, 52 Safronov, A.V, 199 Sagemuehl, Dirk, 60 Sahasrabudhe, A.D, 114 Sainio, Panu, 225 Sakamoto, Shinichi, 100 Sakamoto, Shuichi, 26, 269 Sakuma, Tetsuya, 101, 193, 222 Salomé, M, 52 Salomons, E, 220 Salver, R. Le, 21 Samel, A, 155 Samuels, Stephen, 106 Sánchez, Beatriz, 164, 202 Sandberg, Ulf, 224 Santillan, Arturo O, 276 Sas, Paul, 24, 139, 221 Sasaki, Susumu, 165 Sato, Akira, 184 Sato, Fumiaki, 74 Sato, Taiki, 269 Schimming, Peter, 144 Schirmer, Stefan, 80 Schirone, M, 250 Schmidt, Lothar, 60 Schmidt, Manfred, 72 Schoen, Eric D, 130 Scholl, Werner, 74 Schomer, Paul D, 45 Schott, Marie-Odile, 81 Schreckenberg, Dirk, 39 Schuemer, Rudolf, 39, 41, 44 Schuhmacher, Andreas P, 278 Schulte-Fortkamp, B, 170 Schulze, M, 155 Schümann, M, 154 Seavey, Eric R, 213 Sehrndt, Gustav A, 96 Sekine, Hidehisa, 115 Seo, Y.S, 136 Seol, Hanshin, 189 Sepulcri, D, 250 Serra, M.R, 241 Serrulla, Juan V. Arizo, 230 Serve, Catherine, 257 Sezgin, Önder, 50 Shao, W, 215 Sharp, Ben H, 206 Shen, Rongying, 185 Shibayama, Hideo, 170, 181 Shimada, Kazuo, 236 Shinohara, Naoaki, 87 Shrivastava, Ashish, 114 Siano, Daniela, 211 Siddique, M.N.H, 34 Silva, Nancy da, 16 Simon, Frank, 94 Singh, Rajendra, 93 Sirotnikov, V, 268 Sjösten, Per, 30 Skånberg, Annbritt, 148 Smith, Andy, 153 Solana, Pedro E, 20 Sors, Thomas, 237 Southgate, David, 36 Spång, Kjell, 49 Srinivas, D.S.R.K, 212 Staatsen, B.A.M, 157 Stait, R, 133 Standley, Edwin, 219 Stansfeld, Stephen A, 36, 154, 156 Stap, Paul van der, 9 Stapelfeldt, Hardy, 256 Starck, Jukka, 238 Stechow, L, 201 Steimer, Vincent, 100 Stein, George Juraj, 34 Stenschke, Reiner, 129 Stephan, Thomas, 257 Steven, Heinz, 21, 104 Stevens, Guust, 249 Stöcker, U, 91 Støfringsdal, Bård, 147 Storeheier, S.A, 107, 216 Straetemans, Masja, 155 Strasser, Patricia, 82 Sturrus, Lowi, 107 Su, Fu-cheng, 135 Sugai, Michihiro, 88 Sugimoto, Yugi, 236 Sui, F.S, 264 Sum, K.S, 92 Suzuki, Hajime, 273 Suzuki, Shoji, 122 Suzuki, Yôiti, 243 Swart, W.J.R, 17 Tabata, Toru, 43 Tachibana, Hideki, 65, 74, 75, 100 Takagi, Koichi, 240 Takahagi, Toshio, 271 Takahashi, N, 182 Takahashi, Yukio, 245 Takakuwa, Y, 207 Takano, Yasushi, 165 Takao, Hideo, 114 Takeda, Katsumi, 143 Takenaka, Takaaki, 84 Takinami, Hiroaki, 240 Takmori, Susumu, 184 Talbot, Chris, 281 Tamaoki, Gen, 53 Tamura, A, 18 Tanaka, Toshimitsu, 203 Tanifuji, Katsuya, 26 Tanttari, Jukka, 63 Tapken, Ulf, 142, 144 Taraldsen, G, 107 Tardy, L, 28 Tatsuda, K, 106 Taya, Koichi, 273 Taylor, Richard, 47, 160 Tenenbaum, Roberto A, 280 Terao, Michihito, 115 Termeer, Martijn, 70 Terranova, Daniela, 225 Thaden, Rainer, 161 Thite, Anand N, 71 Thomann, G, 204 Thomas, Callum, 225 Thomas, Jean-Hugh, 281 Thompson, David J, 71, 171 290 Internoise 2001 Abstracts Thorsson, Pontus J, 108, 217 Tian, Jing, 112, 113, 120 Tichy, Jiri, 124 Tirado, Guillermo, 52 Tixier, Nicolas, 168 Toi, Takeshi, 128 Tokhi, M.O, 34, 235 Tom, Patricia, 212 Tomilina, T.M, 71 Too, Gee-Pinn James, 278 Toorn, J.D. van der, 105 Toppila, Esko, 238, 241 Tornar, U, 21 Torres, Roberto, 52 Tournour, M, 82 Traub, Paul, 145 Trimpop, M, 236 Tsujiuchi, Nobutaka, 84 Tsukioka, H, 274 Tuna, Joao M.C. Lopo, 52, 55 Uehara, Takashi, 156 Ueno, Fumio, 43 Umeda, Yoshikuni, 200 Umnova, Olga, 219 Uosukainen, Seppo I, 24 Usagawa, Tsuyoshi, 234 Vadim Tokarev, , 210 Vainio, Matti, 14 Vakulenko, Sergei, 120 Välimäki, Vesa, 25 Vallet, Michel, 86 Valtanen, Ilona, 247 Vandepitte, D, 221 Vanweert, Franci L.H, 239 Vasut, Stanislav, 186 Vaure, Antoine Buffel du, 119 Vazquez, Jorge, 89 Vedy, E, 220 Vejvoda, M, 155 Vela, Antonio, 67 Vercammen, M.L.S, 274 Veres, S.M, 32 Vergara, Felipe, 252 Verhaegen, M, 27, 31 Verhage, Jan A, 195 Verheij, Jan, 136 Verkeyn, Andy, 167 Vermeir, G, 185, 280 Vermeulen, Ruud, 136, 188 Versluis, A, 282 Vietzke, Peter, 129 Vílchez, Rosendo, 22 Villagómez, Salvador Echeverría, 273 Villefrance, Louise Rebien, 174 Vipperman, Jeffrey S, 93 Viscard, Massimo, 29 Vivier, E, 21 Vliet, W.J.A. van, 105 Vogel, P, 208 Vogiatzis, Constantinos, 57, 58 Voorden, M. van der, 192 Vorländer, Michael, 73 Vos, Joos, 46 Vos, Paul H. de, 104 Walerian, E, 214 Wandel, Martin, 80 Wang, Hui, 181 Wang, J.H, 202 Wang, Qiang, 185 Ward, T, 240 Watanabe, Hirofumi, 203 Watanabe, Seiji, 26, 78 Watson, A.E, 59 Watts, G.R, 133, 214, 228 Waye, Kerstin Persson, 244, 245 Wayson, Roger L, 212 Weber, Reinhard, 129 Wedemann, Jörg, 179, 180 Wei, Wen, 112 Wenzel, J, 155 Wester, E.C.N, 95 Westerlund, Nils, 146 Wetzel, Edgar, 257 Weyna, Stefan, 189 White, Seth W, 56 Wijngaarden, Sander J. van, 195 Willenbrink, Ludger, 179, 180 Wilson, Sue, 153 Winberg, Mathias, 30, 146 Winninger, M, 234 Winsvold, Bjørn, 274 Wittstock, V, 119 Wojciech, Batko, 269 Won, Jungyun, 138 Wong, C.L, 88 Wong, Johnny C.Y, 267 Worms, Carsten, 188 Wozniak, Ryszard, 226 Wright, S.E, 141 Wszolek, Tadeusz, 250 Wszolek, Wieslaw, 203 Wu, Jian-Da, 135 Wu, Sean F, 279 Wu, Tianxing, 171 Wu, W.D, 134 Wubben, F.J.M, 208, 209 Wullens, Frédéric, 126 Wyckaert, K, 79 Xia, J, 113 Yacci, M.R, 241 Yamada, Ichiro, 87, 273, 274 Yamada, Shinji, 230 Yamaguchi, M, 262 Yamamoto, Kohei, 229 Yamamoto, Takeo, 156 Yamazaki, Koji, 84 Yamazaki, Toru, 86 Yan, Chye Hwang, 183 Yanagimoto, Kensaku, 26, 78, 269 Yanagisawa, Takesaburo, 38, 40, 67 Yanitelli, Marta, 89 Yano, Hiroo, 74 Yano, Takashi, 43 Internoise 2001 Abstracts 291 Yap, C.L, 180 Yasuda, Y, 67 Yasuoka, M, 75 Yeo, Seung Dong, 166 Yeung, Maurice, 88 Yin, Fuxing, 184 Yin, H.P, 130 Yokoi, Masayuki, 271 Yokokura, Yutaro, 187 Yokota, Akinori, 48, 274 Yonekawa, Yoshiharu, 245 Yonekawa, Yoshiharu, 48 Yoo, Byoung-Ho, 191 Yoon, Tae-seok, 253 Yoshihisa, Koich, 101, 106 Yoshimura, Takuya, 53 Yuki, Kiyohiro, 236 Zaporozhets, Oleksander, 210 Zenda, Yasuo, 175 Zhang, Charles Q, 81 Zhang, Xuetao, 197 Zhang, Zhiyong, 185 Zheng, Y, 111 Zhu, Yongbo, 120 Zillmann, J, 142, 143 Zimpfer, V, 232 Zindeluk, Moyses, 280 Zouboff, V, 103 Zverev, A.Ya, 201 292 Internoise 2001 Abstracts