Heating Monumental Churches by niusheng11

VIEWS: 141 PAGES: 240

									   Heating Monumental Churches
Indoor Climate and Preservation of Cultural Heritage




                      PROEFSCHRIFT


          ter verkrijging van de graad van doctor aan de
       Technische Universiteit Eindhoven, op gezag van de
     Rector Magnificus, prof.dr. R.A. van Santen, voor een
                 commissie aangewezen door het
     College voor Promoties in het openbaar te verdedigen op
           donderdag 19 december 2002 om 16.00 uur




                              door



              Henricus Lambertus (Henk) Schellen
                       geboren te Weert
Dit proefschrift is goedgekeurd door de promotoren:
prof.dr.ir. M.H. de Wit
en
prof.ir. P.H.H. Leijendeckers

Copromotor:
dr. B.A.H.G. Jütte




Cover: Infrared thermograph of St. Matinus’ Weert


Schellen, Henk
Heating monumental churches : indoor climate and preservation of cultural
heritage / by Henricus Lambertus (Henk) Schellen. – Eindhoven : Technische
Universiteit Eindhoven, 2002.
Proefschrift.
ISBN 90-386-1556-6
NUR: 955
Trefwoorden: monumentenzorg kerkgebouwen / warmtecomfort / bouwfysica /
verwarmingsinstallaties-computersimulatie
Subject headings: building conservation / churches / heating systems – computer
simulation / thermal comfort
This research was partially funded by the Netherlands Institute for Cultural Heritage
(ICN) and the Netherlands Department for Conservation (RDMZ)
The climate data were obtained from the National Research and Information Center
for Climate, Climatic Change and Seismology in the Netherlands (KNMI)
CONTENTS

Voorwoord

Samenvatting

Summary

1   Introduction                                                        1-1
    1.1 Status of research                                              1-2
    1.2 Problem statement                                              1-13
    1.3 Objectives of the present work                                 1-13
    1.4 Methodology                                                    1-14
    1.5 Outline of the present work                                    1-14
2   Numerical modeling                                                 2-18
    2.1 Introduction                                                   2-18
    2.2 Interzonal hygrothermal modeling of a church                   2-19
    2.3 Indoor air humidity effects of an open infrared gas heating    2-31
    2.4 Direct infrared radiation in a church                          2-34
    2.5 Hygrothermal load of elements and materials                    2-40
    2.6 Mechanical response of elements and materials                  2-43
    2.7 Intrazonal airflows in a church                                2-48
3   Experimental assessment                                            3-55
    3.1 Introduction                                                   3-55
    3.2 Instrumentation for field measurements                         3-55
    3.3 Laboratory measurements                                        3-60
4   Case studies                                                       4-77
    4.1 Introduction                                                   4-77
    4.2 Heating versus no heating                                      4-77
    4.3 Heating and dewpoint difference                                4-78
    4.4 Heating and relative humidity changes                          4-80
    4.5 Wood shrinkage and swelling due to heating                     4-80
    4.6 Heating effects on monumental organs                           4-81
    4.7 Energy consumption                                             4-82
    4.8 Evaluation of heating systems                                  4-85
5   Closure                                                            5-93
    5.1 Introduction                                                   5-93
    5.2 An upgraded performance array for church heating               5-93
    5.3 Upgraded performance arrays for different heating systems      5-99
    5.4 Qualification of heating systems                              5-101
    5.5 Checklist for choice and design of a proper heating system    5-106
     5.6 Results and conclusions                                           5-110
     5.7 Discussion and recommendations for further research               5-115
6    Literature                                                            6-117
7    Nomenclature                                                          7-125
8    Appendix A: case studies                                              8-129
     8.1 Damage to a monumental organ in the Waalse Kerk in Delft          8-129
     8.2 A modified air heating system in the Sype Kerk in Loosdrecht      8-149
     8.3 A new air heating system in St. Liduïna’s Basilica in Schiedam    8-155
     8.4 Contamination of the ceiling in St. Martinus’ Church in Weert     8-163
     8.5 Contaminated frescoes in St. Gerlachus’ church in Houthem         8-172
     8.6 Planned infrared gas heating in the Grote Kerk in Dordrecht       8-181
     8.7 Condensation and gas heating in the R.K. Kerk in Bemmel           8-187
     8.8 Infrared gas heating in the Grote Kerk in Alkmaar                 8-192
9    Appendix B: Basic data sets of cases                                  9-199
     9.1 Waalse Kerk (Walloon Church) in Delft                             9-199
     9.2 Sype Kerk (Sype Church) in Loosdrecht                             9-203
     9.3 St. Liduïna Basiliek (St. Liduïna’s Basilica) in Schiedam         9-207
     9.4 St. Martinus Kerk (St. Martinus’ Church) in Weert                 9-211
     9.5 St. Gerlachus Kerk (St. Gerlachus’ Church) in Houthem             9-215
     9.6 R.K. Kerk (Roman Catholic Church) in Bemmel                       9-219
     9.7 Grote or St. Laurenskerk (St. Lawrence Church) in Alkmaar         9-223
10   Curriculum vitae                                                     10-227
VOORWOORD

Precies twintig jaar na mijn afstuderen bij FAGO (december 1982) rond ik bij dezelfde
groep mijn promotiestudie af. In 1987 deed ik mijn eerste ervaring met monumentale
kerken op. Ik heb heel prettige herinneringen aan mijn samenwerking met mijn
toenmalige baas prof. Vorenkamp aan verwarmingsproblemen in een aantal kerken in
Nederland. Toen al bleek snel dat multidisciplinaire samenwerking onontbeerlijk was
voor het oplossen van complexe bouwfysische problemen met monumentale
gebouwen. In die tijd leerde ik Ton Jütte kennen. Als medewerker van het toenmalige
Centraal Laboratorium voor Onderzoek van Voorwerpen van Kunst en Wetenschap
(het huidige Instituut Collectie Nederland) had hij regelmatig contacten met
Vorenkamp en mij. Hij heeft me warm gemaakt voor monumenten en geïntroduceerd
in de wereld eromheen. Via een opdracht tot literatuuronderzoek naar de
verwarmingsproblematiek van kerken, heeft Ton mij tot deze opdracht gebracht. In
opdracht van Instituut Collectie Nederland en de Rijksdienst voor de
Monumentenzorg deed ik uiteindelijk een driejarig onderzoek naar kerkverwarming.
De resultaten van beide studies heb ik bijeengebracht in deze dissertatie. Ook daarbij
heb ik heel veel hulp gehad van Ton, als copromotor, waarvoor ik hem veel dank
verschuldigd ben.
Ik kan al meer dan twintig jaar terugvallen op de fysische kennis en ondersteuning van
Martin de Wit. Ik ben trots en blij hem mijn eerste promotor te mogen noemen. Zijn
fysische kennis en inzicht heb ik nog niet geëvenaard gezien. Van hem blijf ik leren.
Pierre Leijendeckers, mijn tweede promotor, was onmiddellijk bereid mijn werk te
ondersteunen. Hij is voor mij hét voorbeeld van een hoogleraar installatietechniek.
Ik dank verder Nico Hendriks, Frans van Herwijnen, prof. Asselbergs, Hans Cauberg,
Dario Camuffo en Hugo Hens voor hun deelname in mijn promotiecommissie.
Ik heb veel geluk gehad mijn kamer te kunnen delen met collega Jos van Schijndel. Op
Jos doe je nooit vergeefs een beroep. Zijn kennis van - en vaardigheden met -
simulatiegereedschappen als Matlab, Femlab en FlexPDE hebben mij geholpen dit
werk een stuk verder te brengen. Na jaren van samenwerking, gemeenschappelijke
lunches, congressen en uitjes is Jos méér dan een collega.
In St. Oedenrode wonen meer fijne mensen. Marcel van Aarle, lang geleden via een
HTS afstudeeropdracht bij ons afgestudeerd en nu medewerker, is mijn steun en
toeverlaat bij het meten. Zijn accuratesse compenseert de mijne. Samen met Jos zullen
wij nog veel overleg moeten plegen...
De kwaliteit van metingen staat of valt verder bij de technische ondersteuning en
voorbereiding van de metingen. Wim van de Ven heeft mij meer dan twintig jaar
zonder aarzelingen en met veel vriendschap meegeholpen. Na Wim’s pensionering
heb ik Wout van Bommel als een fijne opvolger ervaren. Guus Theuws was altijd weer
bereid er in de werkplaats iets moois van te maken.
Ik wil verder Jan Diepens bedanken voor zijn hulp met Radiance en, samen met
Harrie Smulders, voor het onderhoud van de computersystemen. Ferry Bakker en het
bestuur van FAGO dank ik voor het scheppen van de bestuurlijke randvoorwaarden
voor dit werk. Ik heb verder nooit vergeefs een beroep gedaan op andere collega’s van
FAGO. Ik dank ook hen die tijdelijk mijn taken in het projectwerk hebben
waargenomen. De dames van het secretariaat, tenslotte, zijn en blijven bepalend voor
de goede sfeer.
Onderzoek en onderwijs combineren, dat is voor mij het mooie van een universiteit.
Ook ík ben voor een belangrijk deel gevormd door mijn samenwerking met studenten.
Ik hoop dat dat ook voor hen geldt. Heel veel afstudeerders en T8/T9 studenten
hebben een bijdrage aan dit werk geleverd: Jeroen Vugt, Marieke Krijnen , Marieke
Nijland, Janny Stevens, Marc Stappers, Nicole van Hout, Tom ter Bekke, Dionne
Neilen, Marco Martens, Jennifer Sanders, Maaike Mensink en anderen, bedankt.

Als je de weg weet, zijn de mogelijkheden van een universiteit als de onze schier
oneindig. Dat lukt echter alleen met medewerking van mensen. Buiten mijn
capaciteitsgroep heb ik veel ondersteuning gehad van Constructief Ontwerpen en met
name van Eric Wijen. Van Natuurkunde wil ik Leo Pel en Klaas Kopinga bedanken
voor hun ondersteuning van de NMR metingen en medebegeleiding van
afstudeerders. Jan Carmeliet uit Leuven (en nu ook aan de TUE) bedank ik voor zijn
ondersteuning van mijn houtexercities. Marcel Loomans van TNO Bouw wil ik
bedanken voor zijn CFD ondersteuning.

Maatschappelijke betrokkenheid en dienstverlening horen bij een universiteit. Daar
krijg je ook veel voor terug. Ik wil alle betrokken kerkautoriteiten als de bisdommen,
kerkbesturen en stichtingen, kerkvoogdijen, pastores, kosters en anderen danken voor
hun inbreng van kennis, hun betrokkenheid en gastvrijheid. Zonder hen was ik
geeneens aan dit werk begonnen. Een bijzondere plaats neemt de St. Martinuskerk in
Weert in. Dat zie je dan ook aan de omslag van mijn boekje. Geheel bij toeval (?)
raakte ik betrokken bij de problematiek van deze kerk in mijn geboorteplaats Weert.
Ik ben daar nog niet weg...

Bijna twintig jaar aan de TUE, maar al vijfentwintig jaar samen met José! Zij heeft het
grootste deel van mijn vorming meegemaakt en mede bepaald. Daar ben ik heel
gelukkig mee.
Pap, Mam, Aloys, Fred, Bas, Lisje, Anke en Bjorn: ook jullie bedankt voor jullie
interesse, ondersteuning en geduld met mij en mijn beslommeringen.
SAMENVATTING

De meeste oude, monumentale kerken onderscheiden zich qua constructie en
bouwfysische eigenschappen sterk van vergelijkbare hedendaagse gebouwen. De
beschikbare constructieve materialen waren hout, gebakken steen en natuursteen.
Door de grote hoogten en overspanningen zijn buitenwanden en kolommen meestal
meterdik, massief in steen uitgevoerd. Oorspronkelijk was in deze gebouwen geen
verwarming voorzien. Door de eeuwen heen werd het klimaat in deze gebouwen
daarom voornamelijk bepaald door het buitenklimaat. Door de massieve wanden, het
grote volume, relatief kleine ramen en een vaak beperkte ventilatie was het
binnenklimaat veel stabieler dan het buitenklimaat. Er was bijvoorbeeld nauwelijks een
verschil tussen de dag- en nachttemperatuur. In de zomer was het er koeler dan
buiten, in de winter warmer. Door de grote warmte- en vochtcapaciteit bestond er
zowel qua lucht- als oppervlaktetemperaturen en relatieve vochtigheden een
gelijkmatig klimaat. De monumentale inrichting van deze gebouwen heeft in de
meeste gevallen de eeuwen redelijk doorstaan; blijkbaar was het klimaat in de
gebouwen niet zo ongunstig voor het behoud van de inrichting.
Een groot aantal van deze gebouwen is later van een verwarmingssysteem voorzien,
variërend van (hete) luchtverwarming, vloerverwarming, (infrarood)
stralingsverwarming, radiatoren en convectoren verwarming, lokale en
bankverwarming. Echter na strenge winters als die van 1962-1963 werd een groot
aantal ernstige schadegevallen aan kerkorgels geconstateerd. Dat verschijnsel stond
niet op zichzelf. Ook in het verleden was al vastgesteld dat perioden van langdurige
lage vochtigheden, als gevolg van verwarming en aanhoudende strenge vorst, funest
zijn voor monumentale orgels en andere waardevolle interieurdelen. Door al te hoge
luchttemperaturen en bijbehorende lage relatieve vochtigheden ontstonden bovendien
ook in andere hygroscopische materialen door zwel- en krimpverschijnselen schades.
Voorbeelden daarvan zijn houten inrichtingen als altaren, preekstoelen en panelen. In
de jaren ‘60 en begin ‘70 kwam men daarom tot het inzicht, dat met name
kerkverwarmingsinstallaties een rol speelden bij de schades en dat de bedrijfsvoering
ervan daarom met enige terughoudendheid gehanteerd diende te worden.
Door de snelle veranderingen in de maatschappij zijn ook de eisen van kerkgangers
sterk veranderd. Door het hoge thermische comfortniveau in moderne gebouwen
werd dit comfort ook verwacht in kerken. De laatste 30 jaar kenmerken zich
bovendien door een veranderend kerkbezoek. Kerkbezoek vindt vaak plaats vanuit
vervoer per auto zonder overjas. Het verwarmingssysteem moet dan voor een hogere
binnentemperatuur zorgen. Daarnaast neemt de ontkerkelijking toe en voor vele
kerken wordt, mede om financiële redenen, naar een alternatieve bestemming gezocht.
De bijbehorende kleding zal ook sterk verschillen van winterkleding. Culturele
evenementen worden bijvoorbeeld in avondkleding bezocht, waarbij de verwarming
van het gebouw het thermisch comfort moet garanderen. Voorbeelden van het
alternatief gebruik van kerken zijn tentoonstellingen, diners, concerten en
muziekevenementen, TV opnamen en tentamens. Het oorspronkelijke gebruik van het
gebouw met zijn inrichting raakt dus vaak steeds meer buiten het gezichtsveld. Veel
kerken zijn bovendien aan groot onderhoud toe. Daarbij staat de
verwarmingsinstallatie opnieuw ter discussie.
In deze studie zijn de belangrijkste kerkverwarmingssystemen in Nederland en
daarbuiten onderscheiden. Aan de hand van een uitgebreide literatuurstudie,
laboratoriumonderzoek en een aantal bemeten casussen is verder een verband gelegd
tussen typische kerkverwarmingssystemen en een mogelijk negatief effect op het
monument en de inrichting. Daarbij komen aan de orde uitdrogingseffecten en
gerelateerde vervorming van houten interieurdelen en andere monumentale
interieurdelen onder koude winteromstandigheden; oppervlaktecondensatie en hoge
relatieve vochtigheden ten gevolge van direct gestookte verwarmingssystemen en lage
oppervlaktetemperaturen van wanden en beglazingen; vervuiling van wanden en
monumentale gewelven met hun schilderingen door het gebruik van kaarsen en
wierook, in combinatie met bijvoorbeeld door vloerverwarming gegenereerde
luchtstromingen. Daarnaast komen thermisch comfort en energetische aspecten van
het verwarmen van kerken aan de orde.
In de studie zijn prestatie-eisen voor verschillende kerkverwarmingssystemen
geformuleerd en is aangetoond hoe deze vooraf via simulatie met (gedeeltelijk nieuw
ontwikkelde) computermodellen geschat kunnen worden. Daarbij is de voor kerken
typische invoer voor deze computermodellen bepaald. Vervolgens wordt aangegeven
hoe de prestaties van de verwarmingsystemen achteraf via (gedeeltelijk nieuw
ontwikkelde) meetmethodieken getoetst kunnen worden. Ten behoeve van een
weloverwogen keuze van kerkverwarmingssystemen zijn de verschillende voor- en
nadelen van verschillende kerkverwarmingssystemen gerubriceerd. Tenslotte is een
checklist opgenomen die bij de keuze en het ontwerp van een
kerkverwarmingssysteem van dienst kan zijn.
SUMMARY

The structure of old monumental churches differs a lot from contemporary buildings.
The structural materials were wood, bricks and stone. In order to construct high
buildings with huge spans, thick massive walls and many massive columns were
needed. Originally these buildings had no heating and for centuries the indoor climate
of these buildings was mainly determined by the outdoor climate. Because of the
massive walls, the large indoor air volume, the relatively small windows and most
often relatively limited natural ventilation, the indoor climate was much more stable
than the outdoor climate. There was hardly a difference between the day and night
temperature. In summer the indoor air climate was cool compared to outdoors, in
winter indoor conditions were warmer. As a consequence of the large heat and
moisture capacity the indoor air and surface temperatures and the indoor air and
surface humidity led to a reasonably even climate under our external climate
conditions. The monumental interior of this kind of building stood up to the climate
elements for centuries. Apparently the indoor climate in this kind of building was not
very unfavorable to the monumental interior.
In a large number of these buildings heating systems were installed, varying from
warm air heating to floor heating, (infrared) radiant heating, radiator panel and
convector heating, local and pew heating. From literature there is evidence that severe
winters, like the winter of 1962 to 1963 in Western Europe, caused serious damage to
church organs and other valuable church interior parts (Knol 1971). Further research
showed that these were no isolated cases. Long periods of very low relative humidity,
characteristic of the combination of heating and severe frost, appear to be correlated
to damage to church organs and other parts of church interiors. Literature documents
damage to pulpits, altars, wainscoting, and paintings like wall, ceiling and panel
paintings. In the nineteen sixties and seventies the progressive understanding of these
phenomena gave the insight to limit the unrestricted use of heating systems in
churches.
Due to rapid changes in society the demands of the churchgoers changed in the last
decades. As a consequence of the high thermal comfort in modern day buildings, this
was also required in churches. Another major change was attending the services
without winter coats due to the use of cars by churchgoers. A heating system in the
church had to compensate for the evident lack of thermal comfort.
In the last 30 years church going decreased dramatically in a major part of Western
Europe. Church buildings therefore had to be used for alternative purposes in order
to provide for extra financial income. In the Netherlands, monumental churches are
nowadays used for concerts and other musical events, exhibitions, dinners,
examinations and several other events. People are dressed appropriately for these
events, e.g. evening dress, but not in accordance with the indoor climate in a church.
Thus, nowadays, the original use of the church is changing. Where a lot of churches
are in need of a major restoration or renovation, adapting the heating system becomes
an important point of study.
In this study a distinction was made between the most important heating systems in
the Netherlands and abroad. Making use of a literature study, laboratory research and
several case studies in Dutch churches, common patterns and relations between
typical church heating systems and their effects on the deterioration of monumental
churches were identified. Drying out of the monumental wooden furniture, like
organs, altars and other organic materials and the related shrinkage and damage to the
materials under cold winter conditions; Surface condensation due to low surface
temperatures on walls and stained and protective glazing, in combination with high air
humidity due to excessive moisture sources like open-air infrared gas heating;
Indirectly related problems are contamination due to pollution sources like soot from
candles in relation to relatively large airflows, e.g. generated by floor heating; Related
building physical aspects of church heating are annual energy consumption and
thermal comfort problems due to relatively large airflows and low surface
temperatures of walls and glazing.
Performance requirements for church heating systems with respect to preservation,
energy requirements, thermal comfort and aesthetics were formulated. When
performance requirements are known, prognoses have to be made for the most
suitable design of a heating system in a particular monumental church. Simulation
models and tools and their application for monumental churches were proposed.
Furthermore the church characteristic input for these models was determined. A final
check is the measurement in situ to evaluate and to prove that the heating system
meets the performance criteria. Methods of measurement and some new ways to
interpret them are proposed. For the choice and design of a heating system
advantages and drawbacks of different systems are summarized. Finally a checklist to
structure the choice and design of a heating system is included.
1 INTRODUCTION

The structure of old monumental buildings differs a lot from contemporary buildings.
The structural materials were wood, bricks and stone. In order to construct high
buildings with huge spans, thick massive walls and many massive columns were
needed. Originally these buildings had no heating and for centuries the indoor climate
of these buildings was mainly determined by the outdoor climate. Because of the
massive walls, the large indoor air volume, the relatively small windows and most
often relatively limited natural ventilation, the indoor climate was much more stable
than the outdoor climate. There was hardly a difference between the day and night
temperature. In summer the indoor air climate was cool compared to outdoors, in
winter indoor conditions were warmer. As a consequence of the large heat and
moisture capacity the indoor air and surface temperatures and the indoor air humidity
led to a reasonably even climate under our external climate conditions. The
monumental interior of this kind of building stood up to the climate elements for
centuries. Apparently the indoor climate in this kind of building was not very
unfavorable to the monumental interior.
Due to rapid changes in society the demands of the churchgoers changed in the last
decades. As a consequence of the high thermal comfort in modern day buildings, this
was also required in churches. Another major change was attending the services
without winter coats due to the use of cars by churchgoers. A heating system in the
church had to compensate for the evident lack of thermal comfort.
Therefore in a large number of these buildings heating systems were installed, varying
from warm air heating to central convector and radiator panel heating, (infrared)
radiant heating, floor heating and local and pew heating nowadays. From literature
there is evidence that severe winters, like the winter of 1962 to 1963 in Western
Europe, caused serious damage to church organs and other valuable church interior
parts (Knol 1971). Further research showed that these were no isolated cases. Long
periods of very low relative humidity, characteristic of the combination of heating and
severe frost, appear to be correlated to damage to church organs and other parts of
church interiors. Literature documents damage to pulpits, altars, wainscoting, and
paintings like wall, ceiling and panel paintings. In the nineteen sixties and seventies the
progressive understanding of these phenomena gave the insight to limit the
unrestricted use of heating systems in churches.
Heating of massive monumental buildings like churches, fortresses or castles thus may
have great building physical impact. This may lead to the deterioration of the
monumental building itself and its monumental interior.
In the last 30 years church going decreased dramatically in a major part of Western
Europe. Church buildings therefore had to be used for alternative purposes in order
to provide for extra financial income. In the Netherlands, monumental churches are
nowadays used for concerts and other musical events, exhibitions, dinners,


                                            1-1
examinations and several other events. People are dressed appropriately for these
events, e.g. evening dress, but not in accordance with the indoor climate in a church.
Thus, nowadays, the original use of the church is changing. Where a lot of churches
are in need of a major restoration or renovation, adapting the heating system becomes
an important issue.
When we look at a total number of registered monuments in the Netherlands
(46,775), monumental churches (3,229) constitute 7 % of the total. When we extract
the monumental dwelling houses (30,955) from the total, churches are even 49 % of
the remaining monumental buildings (RDMZ 2000/2). In the rest of Western Europe
circumstances may be comparable. As a result, the importance of the subject turns out
to be evident.

1.1   STATUS OF RESEARCH
Next a chronological, global description of the most important literature on church
heating is given. This overview is based on a literature study (Schellen 1998/2). A
general overview of different heating systems is taken from the study and is adapted to
identify physically different operating systems in the following case studies.
Subsequently the heating systems will be related to damage to building and interiors,
as described in literature.

1.1.1     Most important literature on church heating
The Dutch report on church heating and monumental organs, ‘Kerkverwarming en
Kerkorgels’ (Knol 1971) was written in 1971. Knol gave recommendations to church
officials on how to avoid damage to church organs and valuable church interiors.
Nowadays a number of recommendations from this report are still state of the art in
the Netherlands (IBRK 1999).
In 1972 a similar text ‘Richtlinien für die Beheizung von Kirchen’ (guidelines for the
heating of churches, Mainz 1972) was introduced at a conference in Mainz, intended
for church building officials, architects and referees.
‘Heating your Church’ is a small reference booklet from the Church of England,
written by William Bordass to assist in the decision-making process for heating
systems in churches (Bordass, 1983).
Vorenkamp et al. described the choice of a heating system in a monumental church in
the Netherlands. Heat capacity, indoor airflow, moisture, CO2 and thermal comfort
were dealt with (Vorenkamp ea. 1989).
A more recent German study (Künzel 1991) was undertaken at the Fraunhofer
Institut für Bauphysik in Holzkirchen. About 100 churches throughout Germany were
evaluated experimentally.
Among other things, results from the German study were incorporated in a literature
survey on heating of historical buildings (Arendt 1993). Out of 116 objects, 32 objects
were more thoroughly investigated experimentally. Measurements in situ on air
temperatures and stratification, relative humidity and surface conditions, air velocities
and moisture content of wood have been documented in his book.



                                           1-2
Brian Marks, who is an Energy Consultant of the Church of Scotland, wrote a
practical book on energy consumption and savings in Scottish churches (Marks 1994).
Climate conditions for conservation of monumental objects are described in ‘The
Museum Environment’ of Garry Thomson (Thomson 1978) and ‘Passieve
Conservering; Klimaat en Licht’ (Preventive conservation, climate and light, Jütte
1994).
This kind of guidelines can also be found in the Italian guidelines for Cultural
Heritage: environmental conditions for conservations (UNI 10969).
Camuffo studied the indoor climate of the Sistine Chapel. The influence of direct
sunlight on frescoes and contamination of surfaces by different deposition
mechanisms were evaluated in his study (Camuffo 1995).
Oidtmann et al. investigated the protection of stained glass in churches (Oidtmann
1994). Computer models were developed for the prediction of climate conditions in
churches. Measurements in situ and in laboratory were used to validate these models.

1.1.2 Heating systems
In an early work from Germany, Pfeil distinguished the following heating systems for
large monumental churches: local heating surfaces, floor- and pew heating systems
and hot air heating systems (Pfeil 1975). Arendt distinguished three basic heating
systems in large historical buildings in Germany: pew heating, floor heating and air
heating (Arendt 1993). In Great Britain the most important literature source for the
determination and description of different heating systems in churches is (Bordass
1983, 1996). He classified heating systems in heat emitters and the distribution system.
As heat emitters he described radiant heaters (short, medium and long wave and low
temperature radiant heaters), ceiling, floor and wall heating, convective heating
(natural and forced convectors, warm air units, solid fuel stoves, oil, gas and electric
stoves, radiators, plain or finned tubes), direct fresh air heating, local heating and pew
heating. As distributing systems he distinguished between wet systems (hot water
systems), district or group heating and warm air systems.
In this thesis the following distinction will be made: warm air heating, floor heating,
(infrared) radiant heating, radiator panel heating, convector heating and local pew
heating, or combinations of such.

Warm air heating
In air heating systems the heat carrier is air. Warm air can be brought into a church by
a natural, thermally driven way (e.g. convector heating) but is mostly forced in by
ventilator devices. From 1950 on, many churches have been heated by warm air
systems with natural air circulation. The boiler was fired with oil and later on with
natural gas. Since 1960 air heating systems with forced ventilator devices have been
applied. A distinction can be made between direct and indirect air heating systems: the
heat exchanger is heated directly by a gas-burning device, or indirectly by hot water (or
steam). In the Netherlands the most common system is a system with one or two
discharge grilles in the floor or at a height of about 3 meters and a return grille in the




                                           1-3
floor. More complex systems consist of air distribution systems by air ducts in the
floor or elsewhere.

Floor heating
In the late seventies of the twentieth century floor heating became popular as the main
heating system in well-insulated dwellings. Contrary to monumental buildings the well
insulated new building constructions of that time allowed application of floor heating
without supplementary heating systems. Furthermore, the ‘invisible’ heating system
was aesthetically sound and made no noise. The degree of thermal comfort was high: a
high level of mean radiant temperature, a high thermal (feet) contact temperature,
even with stone floor systems, and little thermal stratification. Due to these
advantages it was also introduced in modern (church) buildings, having the same
degree of insulation and limited heights. As a result of its success, floor-heating
systems have also been applied in monumental churches. Due to the lack of insulation
the performance of floor heating in these buildings is very different.

(Infrared) radiant heating
Since 1980 pure radiant heating by infrared heating devices like gas infrared heating
and electrical infrared heating has been promoted. Radiant heaters can be
distinguished according to their wavelength: short wave (white heat), medium wave
(red heat), long wave (black heat) and low temperature (normal hot water
temperatures) (Bordass 1983). The shorter the wavelength, the more compact the
source (size ratios for the four classifications being typically 1:20:300:4000) and the
greater the proportion of radiation to the total heat output.
Typical short wave radiant heaters are electric quartz lamp heating devices consisting
of tubular quartz lamps with external parabolic reflectors. The red to yellow glow
characterizes medium wave radiant heaters. The electric versions consist of coiled
elements enclosed in silica tubes with a reflector behind. They are relatively low in
output and are most often used for local heating. The gas-fired units with ceramic
radiant elements have a large radiant flux density output and they are mounted at
higher elevations and can cover a large area of radiation. The exhaust gasses can be
brought directly into the church (open gas burning) or removed by an exhaust duct.
Typical temperatures of 300 oC do not make long wave radiant heaters glow. In the
gas-fired units a fan draws the combustion gases from the gas-burning device through
a tube mounted in front of a reflector. Heat from the tube is radiated and convected
into the church. Electrical black heating devices are not suitable for churches because
of their low radiation characteristics.
The hot water single panel radiator is an example of a low temperature radiant heating
device. It is operated at hot water temperatures of about 45 to 75 oC. Wall, floor and
ceiling heating also are examples of low temperature heating systems, operated only a
few degrees above room temperature.




                                          1-4
Radiator panel heating
Radiators range from traditional cast-iron to the present pressed steel types. They take
the following main forms: column types, with a battery of vertical fins, single or
multiple panels with flat or corrugated plates and radiant convector types: flat panel
type radiators with profiled convective elements added. Because of the affected
esthetical view radiator panel heating is not popular in monumental churches with
substantial heating demands. They are mainly used in small monumental churches
(less than 200 persons).

Convector heating
Old convectors frequently used plain cast iron pipes and later on finned or grilled
tubes. Nowadays more compact convectors consist of finned radiators. The principle
of a convector is air that enters the base of the convector unit and emerges by
buoyancy at the top or front of it. Forced convectors and warm air units consist of
fans, passing air over a heat exchanger. They may be heated directly by electricity or a
burner, or indirectly by hot water or steam. For a given output, forced convectors are
smaller than natural convectors. They also give more control over temperature,
velocity and distribution of the warm air. In a well-designed system this can help to
reduce draughts and stratification.

Local (pew) heating
Pew heating is the heating of the local area of the pews by convector or pipe heating,
direct electrical heating of the backs of the pews or bottom seats and heating by local
radiant panels.

1.1.3 Heating system related damage
There is a lot of literature about damage to church buildings and their interior that is
related to heating systems. Damage to walls, ceiling and paintings and detriments to
stained glass are the most described ones. A number of damage to monumental
organs and church interiors has been ascribed to the heating system. Contamination
by soot and dust in relation to heating systems like floor heating is described in
literature too.

Walls, ceiling and paintings
Schlieder describes wall paintings and their flake off, due to the drying out of walls
(Schlieder 1967). Typical wall contaminations like dark spots on walls above heating
elements and near in- and outlet grilles are related to heating systems (Beck 1981),
(Wegner 1972), (Müller), (Arendt 1976), (Schmidt-Thomsen 1972), (Groeger 1979). In
(Zehnder 1986) the deterioration of wall paintings by salt attack in the Züricher
Grossmünster was described. The salt attack proved to be related to the change of
relative air humidity by heating the church. Arnold et al. showed the deterioration of
wall paintings in the monastery of Müstar to be the result of the central heating system
installed in 1951 (Arnold 1991). He showed the swelling and shrinking of objects to be
related to changes in relative air humidity. In a laboratory experiment he proved the



                                           1-5
correlation between salt attack and heating regime. Arnold asserted that stationary
heating is the worst heating system in connection with salt attack.

Stained glass
Korn related the deterioration of stained glass to the hygroscopic wall effects of warm
air heating. The hygroscopically extracted moisture from walls condensed on the
stained glass causing glass corrosion (Korn 1971). According to Frenzel the most
important factors for the deterioration of medieval stained glass are the chemical
composition, the surface texture and the availability of water due to condensation
(Frenzel 1981). An ideal indoor climate for glass paintings has been described to be in
the range of 40 to 60 % RH (Mayer1981). Oidtmann et al. concluded from physical
measurements and model simulation studies that church heating can have a positive
effect on the conservation of stained glass and glass paintings. Internally ventilated
protective glazing further improves the conservation of stained glass (Oidtmann
1995).

Monumental organs
Damage to monumental church organs was described in a number of articles.
Schlieder related the introduction of forced ventilation devices and central heating
systems in churches to the damage to church organs. In a research on 52 churches he
found temperature differences between church organs and inlet supply air in summer
of 30 to 36 K and in winter of 22 to 47 K (Schlieder 1967).
(Stadtmüller 1972), (Badertscher 1965) en (Supper 1967) reported cases of damage to
monumental organs over the foregoing 20 years. They concluded the combination of
dry air and fast heating to be the main cause of the damage. Damage consisted of
cracks and deformation of separate organ parts. The air transport parts proved to be
the most vulnerable ones. The authors introduced safe relative humidity intervals to
be: Stadtmüller 60-70%, Supper 60%, Schlieder 50-80%, Badertscher 45-65%.
To reduce large temperature differences during heating up the church Badertscher
introduced the concept of a primary temperature of 6 to 8 oC and suggested a
temperature limit of 15 oC (Badertscher 1965). Furthermore he introduced the
concept of local humidification of the organ. To prevent the occurrence of dramatic
low RH during the winter season he suggested reducing the outdoor air part in re-
circulation.
To reduce excessive differences in relative humidity Gossens suggested to heat only
re-circulated air and to restrict the air temperature to 15 oC (Gossens 1977).

Monumental church interiors
Arendt described damage to visible interior parts. He pointed out the different
dilatation behavior of parts due to thermal and relative humidity changes (Arendt
1993). Hilpert mentioned fast changes to be of greatest importance. Thicker elements
and different hygroscopic materials led to dissimilar moisture behavior.
Schaible investigated the moisture influence on paintings hanging on walls. He
mentioned the condensation risk on walls and paintings hanging to them. Hanging the



                                          1-6
painting free from the wall or insulating it by a protective backing plate can reduce the
condensation risk. Schlieder pointed out the damage to doors and wooden interiors.
The most important type of damage he reported were cracks (Schlieder 1967).

Contamination by soot and dust
Arendt and Müller reported dark silhouettes on indoor surfaces, related to the
underlying construction elements.
Bordass mentioned dark shades and their relation with local heating systems (Bordass
1983).
Contamination due to candle soot has been described by Gossens. He measured the
soot production of stearine candles and showed soot productions of 265 to 400 g for
a yearly candle consumption of 10,000 candles. Furthermore he related deposition of
dust to heating intervals and condensation on walls. Contamination rates were related
to the use of churches, maintenance of floors and hygroscopic properties of walls,
plasters and paintings. The mean dust content in churches he reported was 1 to 2
mg/m3 for air at rest to 10 mg/m3 during sweeping the floor.
On the other hand Schmidt-Thomsen referred to churches having walls that had not
been contaminated for 200 years.
Baumann argued to lower air temperatures as far as possible to prevent wall
contamination. According to him higher air temperatures lead to thermally driven
airflows near cold walls that induce dry deposition of dust particles.
Camuffo described the contamination of the frescos in the Sistine chapel due to
deposition processes. He made a distinction between depositions due to Brownian
movement, thermophoreses, moisture gradients near walls, inert deposition,
gravitational deposition and electrophoreses. Visitors induced the worst indoor
climate. Convective vertical air transport led to vertical dust movements.

Directly heating system related damage
Gossens reported enforced wall contamination due to floor heating. He correlated this
to decreasing wall surface temperatures as a function of height. According to Gossens
a combination of floor and air heating could overcome this problem. This
combination of heating systems namely increases wall surface temperatures at higher
elevations.
Arendt in his literature research reported very few authors to be able to relate
instances of damage directly to heating systems. Arendt himself related wall and
ceiling contamination to radiator and convector heating. Arendt, too, related increased
wall surface contamination as a function of height to floor heating systems. Thermally
generated airflows by floor and pew heating led to contamination of ceilings. Pew
heating systems caused evaporation of moisture from footwear and clothing. This
moisture was reported to cause enforced contamination at ceilings and other surfaces.
Thermally irregular resistances, like wooden beams in vaults, often are a cause of
irregular surface contamination.
In case of floor heating Supper advised to clean floors daily (Supper 1967). Groeger
indicated a relationship between floor waxes and adherence of soot and dust. An



                                           1-7
invisible thin film of adherent floor wax substances has been reported on wall and
ceiling surfaces (Groeger 1979).

Heating versus non-heating
In the report ‘Bauphysikalische Untersuchungen in unbeheizten und beheizten
Gebäuden alter Bauart’ (Künzel 1991) a comparison was made between heated and
non-heated monumental buildings. The most important conclusion from this research
was that the heating of monuments in principle had a positive effect on the
preservation of the building and its interior. However, the restricting condition was
that heating should not be at the level of thermal comfort for the continuing presence
of people. A heating according to the preservation of the building was defined as
     - Relatively small fluctuations of the air temperature during a year period;
         avoidance of temperatures near the freezing point;
     - Relatively small fluctuations of air relative humidity during a year period;
         avoidance of high humidity levels;
     - Reduction of the condensation endangerment, especially in the springtime.
This kind of heating would reduce the hygrothermal load of materials and
construction. In general Künzel suggested to allow larger fluctuations during a long
period and smaller for a short period. From practice he suggested 10 % RH
fluctuations during a day and 30 % RH during a period of a year, between about 50
and 80 % RH.

Guidelines for conservation of cultural heritage
The Italian Regulation UNI 10969 gives the following guidelines for conservation of
cultural heritage.
Deterioration is a process of cumulative and progressive nature, non-linear and
irreversible. In the long run, materials reduce their adaptability and microclimate
changes may become critical. The same causes may provoke different synergism and
effects as a response to the type of object and its past history. The past history of each
object has the major relevance as the microclimate may induce the material to undergo
irreversible changes as a response to the external forcing and as a function of its
physical and chemical characteristics. As far as the conservation is concerned, the
specific and interacting problems of each individual object are much more relevant
than knowing average properties of materials and establishing hypothetical well-being
areas. A comparison with field truth is essential to verify laboratory findings and
establish their limit. This is because laboratory tests are very simple, e.g. performed on
totally homogeneous wood samples having a precise cut, small size, and exposed to
some well defined RH cycles. In the field the situation is complex: e.g. artworks made
with a combination of boards, planks and tablets bound together along surfaces
having different cuts, made of different wood types, and exposed to variable RH
cycles. Some materials are not sensitive to humidity changes (e.g. gold); others require
to be preserved in a dry atmosphere (e.g. ceramics impregnated with hygroscopic salts
or heavily oxidized metals); others require some precise RH intervals (e.g. collections
of hydrated minerals). Finally, some other materials are highly hygroscopic and are


                                           1-8
very sensitive to RH changes (e.g. wood, ivory, animal glue, parchment, paper). These
hygroscopic materials require much care for preservation.
Some materials, originally not very sensitive to variability in T and RH, in the course
of the time may undergo internal transformations and micro fractures and become
highly vulnerable, as it often occurs with archaeological remains.
In some cases it is possible to establish, on the ground of experience and laboratory
tests, some better intervals for the conservation of specific materials. For instance,
elevated RH levels may be dangerous: in the presence of acid or oxidizing agents,
moisture may accelerate the deterioration rate of cellulose, the biodecay and the
physical deterioration. For this reason it is advisable to preserve paper in a moderately
dry environment.
In the case of objects constituted of one or more materials with complex behavior, or
which have been conditioned by its past history (e.g. wood), it is necessary to proceed
as follows.

Guidelines for conservation
   1. Artworks must be preserved in the same microclimate in which it has been
        kept for a long time if this microclimate has been proved to be not harmful.
   2. The microclimate conditions can be improved by attenuating or eliminating
        changes, e.g. diurnal cycles, fluctuations, gradients.
   3. If the original microclimate has to be changed, a specific study must be
        carried out to evaluate the adaptation of the artwork to the new conditions,
        taking into account the past conditions and the response of the object.
   4. In the case we need to keep in a new environment artworks that have been
        recently made, or that had environmental past conditions unknown, the
        conservation microclimate must be studied after the physical and chemical
        characteristics of the artwork.
   5. In the case we need to change the microclimate in which an artwork has been
        preserved, the change must be performed at a very slow rate in order to allow
        a very gradual adaptation to the new environmental conditions. It is also
        necessary to continually verify whether the artwork can adapt to the new
        conditions without been damaged.
   6. In the case of displacement of artworks for restoration, traveling exhibitions
        or so, the original microclimate must be preserved as accurately as possible.
        This is the case both for transport and for storage.
   7. In the case of photo-sensitive surfaces, we have to remember that the damage
        is cumulative, and artworks must be illuminated with a light filtered of the IR
        and UV components, at the lowest acceptable intensity, and only for the time
        strictly necessary for being seen.




                                           1-9
1.1.4 Recommendations to prevent damage related to heating systems
In the 1960-ties the understanding grew that heating monumental churches to high
comfort levels could have large negative effects on the building and the monumental
interior. As stated before the loss of monumental organs, completely destroyed by
dramatically low relative air humidities, was described. This led to a number of studies
in the 1970-ties. These studies resulted in guidelines on church heating and were
meant for church authorities and architects. They were based on the intended use of
the churches for liturgical services. These recommendations are summarized in
Table 1-2.
The most important recommendations are based on Dutch (Knol 1971) and German
(Mainz 1972) research. In a lot of cases the monumental organ is seen as the most
critical interior part in a monumental church. The most important parameters in the
table are the indoor air temperature θi and the relative humidity RHorgan in relation to
the preservation of a monumental organ. For most parameters in the table lower and
upper limits are introduced. The lower limit of relative humidity is based on
preventing dehydration of the wooden interior parts of the organ, the upper limit on
preventing mould grow on organic interior parts of the organ like leather and wood.
Most literature sources indicate a lower limit of relative humidity of 45 to 50 % and an
upper limit of 65 to 75 %. Excessive fluctuations should be prevented: the heating rate
of a church should be limited to 1 to 2 K per hour. Conflicting demands may occur: a
recommended higher level of relative air humidity for the organ may lead to surface
condensation, e.g. on stained glass.
To prevent dramatically low relative humidities under cold winter conditions the
indoor air temperature should be limited to 15 to 16 oC. In this respect the indoor
clothing was assumed to consist of winter coats. In case of extreme outdoor frost
conditions in combination with objects of value the recommended indoor air
temperature was lowered to 12 oC (Knol 1971). For reasons of thermal comfort,
indoor air velocities below 0.10 to 0.15 m/s were recommended.
In case of warm air heating the air supply temperature (θsupply) should be limited to a
maximum of 25 oC above indoor air temperature in order to prevent too large air
temperature stratification. Mean air supply velocities (vsupply) should be limited to 2
m/s to avoid thermal draught.1
In case of floor heating the surface temperature of the floor should be limited to 25 to
28 oC to avoid too large vertical airflows and, for thermal comfort limitations, too
high foot temperatures.
In more recent studies of the 1990-ties (Arendt 1993) and (Künzel 1991) an extensive
research on church heating has been described. These studies are based on about 100
churches, from very small village churches to large basilica. A number of striking
matters appeared. With church heating operation the decline of the air relative

1   Later on in this study we will show this recommendation to be wrong: the air supply velocity will be based
     on the desired length of throw. There is no direct relation between the air supply velocity and draught at
     arbitrary places in a room



                                                     1-10
humidity as expected from psychrometric charts was not observed in their studies.2
They explained this by the hygroscopic behavior of walls and furniture. Their
conclusion was that the interior parts would probably not be influenced by fast
fluctuations of air temperature and humidity. Furthermore Künzel’s study showed
graphs of surface and dew-point difference temperatures for heated and unheated
churches. To prevent condensation and mould growth on cold surfaces of walls and
ceilings a lower primary air temperature of 6 to 8 oC should not be underspent. In this
respect the position, favored in historical circles, ‘there is no better heating than no
heating’ has been left. As values from experience for non-critical daily fluctuations of
temperature and relative humidity fluctuations 10 % RH and 30 % RH for yearly
fluctuations have been mentioned, with limits of 50 to 80 % RH (Künzel 1991).
Measurements by Arendt hardly showed expected large air temperature stratification.
He explained this by the occurrence of large airflows. Therefore Arendt doubted the
benefits of hanging ceiling ventilator devices.




2   In our study of warm air heating in the Waalse Kerk in Delft we will show the opposite



                                                     1-11
1.1.5 Actual array of performance requirements for church heating
The above-mentioned criteria for preservation of monumental churches were
summarized in the table below.

                θi    u                 θprimary RHorgan θorgan θsupply       usupply ∆θ/∆t θfloor
                [oC] [m/s]              [oC] [%]         [oC] [oC]            [m/s] [K/h] [oC]
Schlieder       <=15                    7-10 50-80                                    <2
Möhl            10-12
Mayer           7-10 < 0.1                   40-60
Gossens         12-15                   8-10 50-60                                    1.5-2
Stadtmüller                                  60-70
Badertscher     <=15                    6-8  45-65
Supper          17-19                   5-7  < 60
Bordass                  < 0.15         <=10 45-65                                    < 1.5 < 25
DIN 1946                 0.1<v<0.3
Arendt          <=12                    <=9
Schmidt-        12-15                   8                                             < 1.5
Thomson
Knol            <=15                             < 75        <=12 <θi+25   < 1.5
Mainz           12-15                                             <=45 <=2 < 1.5 < 25
Vorenkamp       < 18                    8                                  <1
Künzel          12-16                   5-8      50-80                     fast
Hennings        15    < 0.15            8        55-75                           < 15
Table 1-1: Recommendations from international literature for the preservation of
monumental churches and interior

Legend
θi         Indoor air temperature        [°C]      θorgan     Air temperature near the organ         [°C]
u          Indoor air velocity           [m/s]     θsupply    Supply air temperature                 [°C]
θprimary   Primary temperature           [°C]      usupply    Supply air velocity                    [m/s]
RHorgan    Relative humidity near the    [%]       ∆θ/∆t      Heating rate                           [K/h]
           organ                                   θfloor     Floor surface temperature              [°C]




                                                  1-12
1.2   PROBLEM STATEMENT
The main problem in this research encountered was the identification of common
patterns and relations between typical church heating systems and their effects on the
deterioration of monumental churches. This should lead to the identification of typical
building physical problems directly related with heating:
Drying out of the monumental wooden furniture, like organs, altars and other organic
materials and the related shrinkage and damage to the materials under cold winter
conditions;
Problems related to the drying of brick constructions, e.g. salt damage to the plaster,
damage to wall and ceiling paintings;
Surface condensation due to low surface temperatures on walls and stained and
protective glazing, in combination with high air humidity due to excessive moisture
sources like open-air infrared gas heating;
Indirectly related problems are contamination due to pollution sources like soot from
candles in relation to relatively large airflows;
Infestation by mould and wood boring insects;
Related building physical aspects of church heating are annual energy consumption and
thermal comfort problems due to relatively large airflows and low surface temperatures of
walls and glazing.

1.3   OBJECTIVES OF THE PRESENT WORK
An important objective of this study was the determination of performance requirements
for church heating systems with respect to preservation, energy requirements, thermal
comfort and aesthetics. Because of the large differences in church building construction,
geometry, glazing, usage, location and monumental value of building and interior no
‘generally ideal’ heating system may be identified. Rather one of the possible systems for
each particular church can be acknowledged. When performance requirements are
known, prognoses have to be made for the most suitable design of a heating system in a
particular monumental church. The aim is the identification of required simulation
models and tools and their application for particular types of buildings. Furthermore the
characteristic input for this kind of model should be determined. A final check is a
measurement in situ to evaluate and prove that the heating system meets the performance
criteria. The identification of measurement systems and their application for this kind of
building is required.
The questions thus to be answered were:
     Which characteristics of the building and the heating system must be known to make
     a proper choice for the design of a heating system?
     What are the performance criteria for a heating system?
     How can the long-term behavior of the indoor climate be predicted?
     What are the effects of the indoor climate on the monumental interior?
     What control parameters and control strategies of the system are needed to meet the
     performance demands?
     How can a realized system be evaluated experimentally?



                                          1-13
1.4   METHODOLOGY
The results of the preceding literature study led to Table 1-2, which gives an overview of
preliminary assumptions regarding the supposed characteristics of the identified heating
systems at the start of this research. For each assumption a verifying experiment or test
and a simulation study has been carried out in this research. This led to the framework
for this thesis. In a 5 point scale from −− to ++ is indicated if the assumed characteristic
has been demonstrated in this thesis. A mark indicates that no experimental results or
simulation results have been obtained yet.
The method of research was the following. For most of the heating systems mentioned,
case studies have been evaluated by means of measurements and the measurement results
have been compared with results from computer simulation studies. From these
evaluation studies typical building physical characteristics have been derived and an
attempt has been made to generalize the characteristics of the heating systems. In this
way modern simulation tools can predict the building physical behavior of the church
building, its interaction with the heating system and the effects on its interior. These
predictions have been validated by experiments in laboratory and in situ.
Based on this research tools were developed to predict the impact of a heating system on
a monumental building and to improve the system design by means of simulation studies.

1.5   OUTLINE OF THE PRESENT WORK
Chapter 2 starts with an introduction to computational simulation tools. Thermal and
hygric simulation models are introduced first. A distinction has been made between
simplified and more detailed thermal and hygric simulation tools. Fairly accurate tools like
WaVo (Wit 2000) were used to identify the building physical behavior of the entire
church as a response to interior occupation, heating and exterior climate. Other tools like
WUFI, Matlab and FlexPDE were used to determine in more detail the 1D, 2D and 3D
thermal, hygric and mechanical behavior of the building materials and the monumental
interior. 3D radiant calculations have been performed by Radiance. The model Radiance,
originally developed for studies on lighting and lighting systems, was used to perform the
direct thermal radiant studies. Fluent, finally, was used for studies on indoor airflows in
relation to different heating systems.
In chapter 3 tools to identify and investigate the building physical behavior of monumental
churches are introduced. This chapter deals with the experimental work, both in situ as in
the laboratory. This chapter describes the measurement techniques, used in situ, to
determine the building physical behavior of churches. Measurements on air and surface
temperature, air and near surface humidity, airflows, ventilation and infiltration are dealt
with. Furthermore, experimental laboratory research on parts is described. NMR
measurements on moisture content changes of wood, due to changes in air temperature
and relative humidity, have been performed. In addition tests on the moisture content
related deformation were carried out. Soot production of candles and incense was studied
by laboratory experiments.
The most important part of this study is the evaluation of different heating systems,
measured in situ. Chapter 4 describes the evaluation studies of 8 different monumental
churches in The Netherlands. The churches were selected within the framework of their


                                            1-14
heating systems and related problems. It was not possible to get a full discrimination on
heating systems. Some churches e.g. have combinations of heating systems. On the other
hand some heating systems, like radiator panel heating, are hardly found exclusively in
(large) monumental churches. A full coverage of heating systems therefore was not
possible for this study.
From the results of this research a further elaboration of recommendations to prevent
heating system related damage in monumental churches has been derived in chapter 5.
Together with recommendations, intended for choosing an appropriate heating system
these conclude this work.




                                          1-15
                   Characteristics,        Test,                  Results3          Model,            Results3
                   supposed                experiment                               simulation
Floor              Thermally               Air velocity u         ++                Fluent,           ++
heating            driven flow             = f(∆θ)                                  Airpak
(Section                                   Smoke tests            0
8.4, 8.5)          Contamination           Indoor air             −                 Fluent,
                   of surfaces due         pollution                                Matlab
                   to deposition           Surface dust           +
                                           analysis
                                           Candle soot            +
                                           production
                   Large time              Heating up             ++                WaVo              ++
                   constant,               and cooling
                   no fine-tuning          down
                   Vertical temp           Infrared               +                 Fluent
                   gradient on
                   walls
Warm air           Hot air, low            Air temp θi =          +                 Fluent            +
heating            velocity leads to       f(h)
(Section           stratification          Smoke test             ++
8.1, 8.2,          Sound air               Air velocities u       +                 Fluent            0
8.3)               distribution by         = f(x,y)
                   mixing due to
                   high turbulence,        Smoke tests            0
                   high velocities
                   Sound air               Air velocities u           4             Fluent            +
                   distribution by         = f(x,y)
                   sound
                   distribution of         Smoke tests
                   inlets
                   Air exhausts are        Air velocities u       +                 Fluent            +
                   not important           = f(x,y)
                   for air
                   distribution            Smoke tests            +
Gas-fired          Relative high           Dewpoint               ++                WaVo              ++
infrared           water vapor             measurement
radiant            production
heating
(Section
8.6, 8.7,
8.8)

3   In a 5 point scale from −− to ++ is indicated if the assumed characteristic has been demonstrated in this thesis
4   No church has been evaluated in situ yet



                                                         1-16
                   Condensation             Difference             ++                WaVo          ++
                   on cold surfaces         surface and
                                            dewpoint
                                            temperature
                   Relative high            Measurement            ++                Radiance      ++
                   radiant                  of surface
                   intensities              temperature

                   Shrinking and            Deformation            ++                FlexPDE       ++
                   swelling of              tests
                   wooden interior          Moisture               +
                   parts                    content
Pew                Thermal                  Temperature            −                 Fluent            5
heating            draught in pews          and velocities
(Section                                    in pew
1.1)               Uncontrolled             Air velocities u       −                 Fluent        −
                   air flows                = f(x,y)
                                            Smoke tests            0
Radiator           Uncontrolled             Temperature            +                 Fluent        −
panel              air flows                and velocities
heating
(Section                                    Smoke tests            +
1.1)
Convector          Comparable to            Temperature                              Fluent
heating            low-velocity air         and velocities
(Section           heating                  Smoke tests
1.1)
General            Primary                  RH and                 ++                WaVo          ++
(Section           temperature              surface
8.1)               prevents mould           temperatures
                   growth and
                   condensation
                   Fast temp/RH             Dynamic                +                 FlexPDE       +
                   variations lead          deformation/
                   to damage of             moisture
                   wooden and               content tests
                   other organic            of wood
                   parts                    Tension tests          +

Table 1-2: Preliminary aspects of research


5   These simulations have started in April 2002 in a EU research; results are not yet available



                                                         1-17
2 NUMERICAL MODELING


2.1   INTRODUCTION
The knowledge of the short and long term thermal and hygric indoor climate of a
monumental church is of great interest for preservation conditions (Thomson 1978, Jütte
1994) and for thermal comfort conditions (Fanger 1970, 1988). Furthermore for thermal
comfort and preservation (contamination of surfaces) the indoor airflows are of
importance. These conditions can be measured in an existing building with an existing
heating system. For the design of a heating system that meets the performance criteria
simulation tools that predict the behavior are indispensable.
In this chapter the main theory and models behind the computational calculations are
described. The model ‘WaVo’ (Wit 2000) has been used for the modeling of the indoor
temperature and humidity climate and for the calculation of the energy use and
capacitance of a heating system. ‘Radiance’ (Larson 1998) was used to calculate the
radiant effects of infrared radiant heating. To calculate 1 dimensional (1-D) temperature
and moisture effects in materials ‘Wufi’ (Künzel 1994) was used. ‘FlexPDE’ (Backstrom
1994) was applied to solve the partial differential equations regarding 2- and 3
dimensional (2-D and 3-D) temperature, moisture and stress and strain calculations in
materials. The CFD program ‘Fluent’ (Fluent 1995) finally was used to predict the indoor
airflow.
To facilitate the design process some formulas from literature are assimilated. The most
commonly used formula to calculate the heating capacity for churches is presented.
Formulas to calculate the moisture effects of open gas burning are assimilated. The direct
radiant effects of the infrared radiant heaters can be estimated from radiant formulas.
Further more well known indoor airflow formulas for simple air supply and jet
calculations are presented. Formulas to facilitate the calculation of airflow near cold
surfaces are added.
In chapter 4 and appendix A the evaluation of the different heating systems is
demonstrated, according to these computational calculations and formulas. Chapter 5
finally describes the use of these computational calculations and formulas in the early and
later design stage.




                                           2-18
2.2   INTERZONAL HYGROTHERMAL                MODELING OF A CHURCH


2.2.1 Objectives
To be able to predict the preservation of church and interior parts and the thermal
comfort conditions the knowledge of short and long-term indoor climate parameters for
the church is of great interest. Air and surface temperatures, as well as relative humidity
should be known as a function of time. The knowledge of the required heating system
capacity is needed for heating system selection. Furthermore it is of importance to
determine the energy requirements, for example over a one-year period, as a function of
heating system and intermittent usage of it.

2.2.2 Model requirements
Because churches do not consist of a large number of rooms, only a limited number of
zones are needed to describe most churches. And although the geometry of a lot of
monumental churches is complex, non-insulated, uniform and thick walls, floors, ceilings
and construction elements most often lead to reasonably uniform surface temperatures.
Therefore the external constructions may be treated one-dimensionally regarding the
indoor climate and energy loss. Because of the thickness of the constructions and the
related phase shifting of temperature and humidity, the thick walls, floor and vault
however should be modeled accurately. From measurements (Arendt 1993) it was found
that the temperature stratification in most heated churches was small. Apparently the
mixing of the air due to low ceiling and wall surface temperatures is nearly complete in
most monumental churches. In a first approximation the indoor air temperature and
absolute humidity may be treated as uniform. Furthermore it was decided to start with a
model that has no need for an input that contains all details. This has the advantage that
the complex geometry of the church does not have to be put into the model. The
longwave radiation is equally distributed over the wall, using the integrating sphere
approach. This leads to a model with two temperature nodes and one humidity node for
the indoor climate. This approach was used before (Wit 1987) and with simple geometries
the results were very similar to the ones found with a more sophisticated model. Later on
temperature stratification and draught effects may be modeled more accurately using
CFD. To allow for the calculation of different heating systems and controls the model
should be flexible and transparent regarding its implementation.

2.2.3 Method
An existing computer simulation model WaVo (Wit 2000) was adapted for thick walls,
different heating and humidifying systems and controls to describe the temperature and
humidity behavior of a monumental church. The model originates from two early
simulation models: the simplified multizone thermal simulation model ELAN (Wit, 1988)
and the second order model AHUM for the prediction of indoor air humidity (Wit,
1990). The geometrical complexity, together with the uncertainties regarding material
properties, dimensions and construction assemblies, air infiltration, outdoor rain exposure
etc. make a real prediction of the indoor climate almost an impossible task. A calibration
or fine-tuning of the model with measurements in a church is unavoidable. A small


                                            2-19
number of calibration parameters, the fine-tuning 'knobs', is an advantage and the risk of
dependant parameters is smaller. This is an argument to keep the model simple. With the
calibrated model changes of the indoor climate by a heating system then can reasonably
be predicted with a model that has essentially the same physics.

2.2.4    Description

Thermal room model
The thermal network of the room model is given in Figure 2-1. The room model basically
consists of 2 nodes: an air temperature node (right) for the description of the ventilation
heat losses and an environmental temperature node (left) for the transmission heat losses.
                        ΣΦ xy                                 ΣΦ ab

                                             A th c(1+h c/h r)

                                        θx                       θa



                (1+h c/h r)Φ r                                        Ca          Φ c-(h c/h r)Φ r


Figure 2-1: Thermal network for the room

The heat balance of the air node leads to the following differential equation:
     dθ a                h
Ca        = ΣΦ ab + Φ c − c Φ r + L xa (θ x − θ a )                                                   2-1
      dt                 hr

The heat balance of the environmental temperature node gives:
                   hc
0 = ΣΦ xy + (1 +      )Φ r − L xa (θ x − θ a )                                                        2-2
                   hr

In the figure and equation the symbols represent the following quantities:
       Ca        =        heat capacity of the air                                                   [J/K]
       θa        =        air temperature of the room                                                [oC]
       ΣΦab      =        heat flow by ventilation and interzonal airflows                           [W]
       Φr        =        the radiant part of the heat sources                                       [W]
       Φc        =        convective part of the heat sources                                        [W]
       ∑ Φ xy =
         y
                          total transmission heat flow through the                                   [W]

                             constructions of the zone
                                                                           hc
        Lxa        =         coupling coefficient = A t h c (1 +              )                      [W/K]
                                                                           hr



                                                   2-20
        θx             =          ‘environmental temperature’                        [oC]
        hr             =          mean surface heat transfer coefficient for         [W/m2K]
                                  radiation
        hcv            =          mean surface heat transfer coefficient for         [W/m2K]
                                  convection
        At             =          the total inner surface area of the zone           [m2]

The heat sources consist of heat supply, casual gains and solar heat gain. The solar gain is
corrected for the short-wave radiation falling from the interior side on the windows. The
ventilation flows are given by

∑Φ     ab   = L a ( θ e −θ a ) + ∑ j L ab ( θ b − θ a )                               2-3

where
        La             =          ventilation heat loss coefficient                  [W/K]
        θe             =          external air temperature                           [oC]
        θb             =          temperature of other zone b                        [oC]
                                  (a is the zone considered)
        Lab            =          ventilation heat loss coefficient for air flow     [W/K]
                                                                       ⋅
                                  from zone b to zone a (Lab = ρ a c p V )

The temperature θy is the 'environmental temperature' of the adjacent room, or in case of
an external wall, the equivalent temperature as an outdoor reference temperature:
θ y = θ e + aE sol / h e − εL at / h e                                                2-4

where
        Esol           =          total solar irradiance of the surface considered   [W/m2]
        a              =          absorptivity of the surface                        [-]
        Lat            =          (σTe4 - atmospheric radiation)                     [W/m2]
                                  view factor surface->sky)
        ε              =          emissivity of the surface                          [-]
        he             =          total external surface coefficient                 [W/m2K]




                                                          2-21
Hygric room model
The hygric room model is quite analogues to the thermal wall model.
                                  Σ G xy                          Σ G ab

                                                  p va



                                           C ft           C va    ΣGp


Figure 2-2: The hygric network for a room

The moisture storage in the air volume (right node) is far more important than the heat
storage in the air in the thermal case. Also the moisture storage in furniture (left node)
cannot be neglected, as wood and other interior furniture are very hygroscopic. As the
moisture storage in the material is determined by the relative humidity instead of the
absolute humidity, the moisture flow to the furniture in Figure 2-2 is defined as:
       d t
g=        C f p va                                                                          2-5
       dt

The furniture storage capacity is:
                       p sat (θ ref )
C ft = C ft (θ ref )                                                                        2-6
                       p sat (θ a )

The equation for the vapor pressure of the indoor air thus is:
       dp va dC ft p va
C va        +           = ∑ G ab + G p + ∑ G xy                                             2-7
        dt     dt                        y



Where:
     pva                  =             vapor pressure of the air                          [Pa]
     psat                 =             satured vapor pressure                             [Pa]
     Cva                  =             storage capacitance of the air: (0.62)10-5Ca/cp    [kg/Pa]
      C ft (θ ref )       =             storage capacitance of the furnishing (at 20 oC)   [kg/Pa]
     Gp                   =             production and humidification (-drying)            [kg/s]
     ΣGxy                 =             moisture flow from walls                           [kg/s]
     Gab                  =             moisture flow by air exchange                      [kg/s]

∑G      ab
             = L va (p e − p va ) + ∑ L vab ( p vb − p va )                                 2-8




                                                         2-22
where
        Lvab        =        (0.62)10-5 Lab/cp                                  [m.s]
        Lva         =        (0.62)10-5 La/cp                                   [m.s]


Thermal wall model
                                          wall ...



                                   ∆R1           ∆R1         ∆R1      A/h i
               θy                                                             θx

                    wall 1
                                          C1            C1         C 1 /2




                                           wall ...


Figure 2-3: Thermal network for the walls

The total transmission heat flow ΣΦxy is calculated with the heat diffusion equations of
the multilayer walls. The heat diffusion equation for a homogeneous slab, subject to one-
dimensional heat flow, is given by:
k ∂ 2 θ ∂θ
       =                                                                            2-9
ρc ∂x 2 ∂t

where
         k          =        thermal conductivity                                  [W/mK]
         ρ          =        mass density                                          [kg/m3]
         c          =        specific heat                                         [J/kgK]

The ELAN model (Wit 1987) uses a simplified approach for the influence of walls on the
thermal climate. This is reasonably accurate for normal well-insulated walls but not for
the massive walls of a church. In this WaVo church model the diffusion equations for
heat and moisture transfer in the walls are modeled with a finite difference scheme and
solved with an implicit method (degree of impliciteness close to 0.75). The time step is
one hour and the place step of each layer for the heat transfer calculations is determined
by the Fourier number of the layer (Fo~1).




                                                 2-23
Hygric wall model

                                                               wall ...



                                      ∆ Z 1/7      4∆ Z 1/7        2∆ Z 1/7        ∆ Z 1/7         A/β i
             pe                                                                                            pva


                  wall 1


                            ∆ C tv1       11∆ C tv1 /14   6∆ C tv1 /14    3∆ C tv1 /14   ∆ C tv1 /14



                                                                 wall ...

Figure 2-4: Hygric network for the walls

The total moisture flow from walls ΣGxy is calculated with the vapor diffusion equation
of the multilayer walls. The moisture transfer through a homogeneous slab, subject to
one-dimensional vapor flow, is given by:
δ a ∂ 2 p v ∂ p v / p sat
           =                                                                                                 2-10
µξ ∂x 2          ∂t

where
        pv            =         vapor pressure                                                               [Pa]
        δa            =         vapor permeability of air (1.8 10-10)                                        [s]
        µ             =         vapor resistance number                                                      [-]
        ξ             =         specific moisture capacity (=dw/dφ)                                          [kg/m3]

By linearization of the hygroscopic curve (ξ is constant) and by considering the vapor
flow only, this wall model is not very accurate for high and very low humidities (RH <
30%, RH > 80%). The advantage, however, is that less sophisticated material properties
are needed. The effect of this inaccuracy on the indoor climate is expected to be very
small as the hygric storage of the air volume is more important (large volume to church
ratio in churches).
For isothermal transport the moisture diffusion equation is analogous to the heat
                                                      k
diffusion equation, with the thermal diffusivity a =     replaced by the hygric diffusivity
                                                      ρc
        δ p
 D v = a sat and θ replaced by pv.
         µξ




                                                             2-24
The effective penetration depth for vapor transfer is much smaller than for heat transfer.
The effective thickness indicates how far a cyclic disturbance at the surface has
penetrated into the material. At a distance of three times the effective thickness the
alitude of the cyclic temperature or vapor pressure is reduced to 5 % of the original value.
                                               at 0
The effective thickness is defined by d * =         for the cyclic heat transfer and
                                                π
         Dvt0
d* =           for the cyclic vapor transfer. In these formulas ‘a’ represents the thermal
           π
diffusivity and Dv the water vapor diffusivity. In the hygric model therefore smaller
spatial steps are needed near the surface. Close to the surface the space discretization for
heat and vapor transfer is separated. As calculation time would be very long by the small
steps the interpolation is only applied to the layers near the surface. For vapor transfer
the first thermal layer is divided into a number of sub-layers. To avoid the inaccuracies
associated with sudden changes in the grid-point interval, a graded grid is used (Laan
1994). In the wall model the first layer therefore is split into 3 in such a way that from
surface to the inside each step is twice the step until the same step-size is reached as for
temperature. The electric analogon of the room and wall model is given in Figure 2-2.
The hygric material capacitances in this figure are temperature dependant and are
analogous to the furniture capacitances.

2.2.5 Heating systems and controls
At this time floor heating and air heating are modeled in WaVo. Thermostatic and
hygrostatic controls of the indoor climate are modeled. The heating and cooling down
rate dθ/dt can be controlled too.

2.2.6 Input data
The global input data of the WaVo model consist of the following elements.

Geometrical and material data
A description of volume of zones and area, thermal and hygric composition, slope and
orientation of walls, glazing, floor, ceiling and roof is needed in this input section. The
geometrical, material and dimensional data can be derived from church plans,
measurements and observations in situ. Physical material properties may be derived from
the WaVo material database or otherwise.

Source terms
Source terms of heating capacity, internal casual heat gains and internal moisture
production are used as input into WaVo. The estimated capacity of the heating system
and the humidifying capacity required may be calculated by WaVo, or a limited given
heating capacity, e.g. an existing heating capacity or a calculated heating capacity from so
called church formulas, may be introduced.




                                              2-25
Heating capacity
Up till now the heating capacity required for church heating was mostly calculated by so
called church formulas. Pfeil gives a historical survey of the development of formulas for
the calculation of heating capacity for churches (Pfeil 1975). Heat losses by transmission,
ventilation and/or infiltration and heating up of construction parts are common parts in
the formulas. Pfeil discriminates between three situations:
Instationary heating (1): the church is heated infrequently and is heated over a short time.
Stationary heating (2): the church is kept at a constant temperature.
A combination of 1 and 2: the church is kept to a constant temperature (3a), a primary
temperature, e.g. 8 oC, and is heated shortly before and during services to a comfort
temperature of about 15 oC (3b).

            θa




                  2
                             3b

                                                                primary temperature
                  3a
                               1




                             z0 (3b)                                     z0
                                               z0 (1)

Figure 2-5: Heating strategies

Three parts dominate the heat loss for an instationary heating regime: the heat loss
through non-accumulating surfaces like windows and doors, the heat loss due to
accumulating surfaces like walls, floor, ceiling and pillars and finally heating up the air.
In case of a stationary heating, standard heat loss calculations according to DIN 4701 are
recommended. Additional factors for heating up the building are used within DIN 4701.
For the instationary heating capacity calculations Pfeil advises to use the calculation
methods of (Gröber et al 1935) or (Krischer et al 1957), because the results of these
calculations are comparable. These calculation methods originate from an analytical
derivation of heating up a slab of infinite thickness. If an infinite thick wall is subject to a
heat flow q, the resulting temperature difference between air and wall is given by:
             1  2 t               
θ i − θ s = q +       = q 1 + 2 t                                              2-11
             h  πλρc     h b π
                                  




                                              2-26
For a non-insulated monumental church building a factor az0 thus can be introduced,
which accounts for the pre-heating time z0 and the heat penetration:
                  1
a z0 =                                                                             2-12
             1 2 z0
              +
             h b π

Where
         h            =       total indoor heat transfer coefficient               [W/m2K]
         z0           =       heating up time                                      [s]
         b            =       thermal effusivity = kρc                             [J/(m2Ks1/2)]

The heating power (or heating capacity) is calculated from the heating loss:

Φ = ∑ (A s a z 0,s (θ i − θ 0 )) + ∑ (A w U w (θ i − θ e )) + ρc p V(θ i − θ e )
                                                                   ⋅
                                                                                   2-13
         s                        w


Where
         As           =       heat accumulating surrounding surfaces               [m2]
                              (walls, ceiling, floor, pillars)
         Aw           =       window and door surfaces                             [m2]
         Uw           =       U-value of window or door                            [W/m2K]
         θi           =       Indoor air temperature                               [oC]
         θe           =       Outdoor air temperature                              [oC]
         θ0           =       Primary or initial indoor air temperature            [oC]

Moisture production sources
Internal moisture production sources are people’s breathing and other moisture source
terms, e.g. due to the burning of gas.

Human production sources
People produce carbon dioxide and vapor. A resting person produces about 0,5 m3 air
exhaust per hour with a CO2 content of about 70 g/m3. The CO2 production therefore is
about 35 g CO2 per hour. The human production of vapor is by respiration and by
perspiration. In this research the perspiration term is assumed to be twice the respiration
term. The moisture loss by respiration depends on the exhale rate and the moisture
content of the indoor air:
              .
G = ρ a V ∆x                                                                       2-14

and
         G            =       moisture production                                  [kg/h]
         ρa           =       density of the air                                   [kg/m3]



                                                   2-27
        .
       V          =        exhale rate                                           [m3/h]
       xi         =        indoor air water content                              [kg/kg]
       ∆x         =        difference in water content of the
                           exhale and indoor air                                 [kg/kg]

The exhale air temperature is about 37 oC with relative humidity of 100 %. The saturated
moisture content therefore is 41.1 g/kg.
The vapor production therefore is approximated by 2(0.025 – 0.6.xi) kg per person per
hour.

Moisture production sources by ‘open gas heating’
Infrared radiant gas heating systems are often so called ‘open air’ gas heating systems and in
the Netherlands natural Groningen G-gas is used. The exhaust combustion gasses are
released directly into the room. The most important pollution producing sources are
moisture and CO2.

Natural gas production sources
‘Groningen’ natural G-gas consists of the following components:

Component                        Formula                          Volume percentage [vol%]
Methane                          CH4                              81.28
Nitrogen                         N2                               14.27
Ethane                           C2H6                             2.82
Carbon dioxide                   CO2                              0.94
Propane                          C3H8                             0.40
Butane                           C4H10                            0.14
Pentane                          C5H12                            0.03
Oxygen                           O2                               0.01
Table 2-1: Constituents of the Groningen G-gas

The most important constituents of the Groningen gas therefore are methane and
nitrogen. To simplify the gas combustion reaction equation the gas composition was
limited to methane only. The combustion equation of methane can be given as:
CH4 + 2O2 ⇒ CO2 + 2H2O                                                            2-15

Full combustion of methane with a theoretical full amount of oxygen therefore leads to
the production of carbon dioxide and water (vapor).
In Table 2-2 the mol air weights of the gasses are summarized and the hourly use and
production of gasses per kW installed capacity, based on a heating capacity of 9.3 kW per 1
m3/hr used gas, has been calculated.




                                             2-28
Formula                       Mol air weight [g]             Production/Consumption
                                                             [kg/(kW.hr)]
CH4                           16                             0.077
2O2                           64                             0.310
CO2                           44                             0.213
2H2O                          36                             0.174
Table 2-2: Mol air weight and hourly production of gasses per kW installed heat capacity

Control strategies
In WaVo control strategies for temperature and humidity control can be given as a
function of time.

2.2.7 Limitations of the model
Because of the integrated sphere approach only a coarse calculation of radiant effects is
possible. If there is an important radiant exchange by surfaces, e.g. due to large surface
temperature differences, these radiant effects are not included.
The model makes use of only one air temperature node in a zone. Temperature
stratification and its effects on heat losses at surfaces therefore are not included. This may
lead to more serious errors than the simplified radiation model.
Due to the one-dimensional approach of heat flow corner and other thermal bridge
effects are not included.
The surface coefficients for convection and radiation are constant. In fact these
coefficients depend on the total flow pattern in a building and on local temperature
differences.
WaVo modeled the moisture effects, resulting from the moisture production of infrared
radiant gas heating. In the present version of WaVo, different irradiant fluxes on separate
walls cannot be modeled. The effect of radiant heating on critical surfaces therefore was
studied using Radiance (see section 2.4). To study heat and moisture processes in a wall
more closely the output of the simulated indoor climate by WaVo is used as boundary
input conditions for Matlab and FlexPDE models for 1D (original WUFI), 2D and 3D
heat and moisture transfer. Furthermore the output from WaVo was used as boundary
conditions for the airflow calculations by Fluent.

2.2.8 Validation
A comparison of WaVo calculation results and measurements e.g. took place in the
Bemmel case. Results of calculated air temperatures and relative humidities, calculated
wall temperatures and relative humidity at the indoor wall and glazing surfaces were
compared with measurements.




                                            2-29
                             20                                   1

                                                                 0.8
                             15




                T air [oC]




                                                   RH air [-]
                                                                 0.6
                             10
                                                                 0.4
                             5                                              Simulation
                                                                 0.2
                                                                            Measusurement
                             0                                    0
                                  14   16   18                         14    16       18


                             20                                   1

                                                                 0.8
                             15




                                                   RH wall [-]
                T wall oC




                                                                 0.6
                             10
                                                                 0.4
                             5
                                                                 0.2

                             0                                    0
                                  14   16   18                         14    16       18



Figure 2-6: Comparison of WaVo calculations and measurements in the Dordrecht case
Air temperature (left above), relative humidity (right above), surface temperature outdoor
wall (down left), relative humidity near outdoor wall (down right) as a function of time of
day (see section 8.7.5)

As a consequence of the hourly weather files, at that time (1999) WaVo could only work
with a time discretisation of an hour. In the figures this limitation is clearly visible. Later
on the time steps were free to choose. For times within the hour the weather data were
linearly interpolated.




                                                 2-30
2.3     INDOOR AIR HUMIDITY EFFECTS OF AN OPEN INFRARED GAS HEATING
For a first impression of the effects of an open infrared gas heating on the indoor air
humidity, a first order approximation of the CO2 and vapor concentration in the air can
be made. This approximation can be used as design help for the consequences of an open
gas infrared heating system for a church.
For demonstration purposes of the condensation effects on cold surfaces of an open air
gas heating, infrared thermography, in combination with air humidity measurements, can
be useful. A transformation of an infrared determined thermograph in a ‘hygrograph’ will
be demonstrated.

2.3.1 First order approximation
To describe the CO2 and vapor concentration of the indoor air we can make a first order
approximation. If we assume a full mixing of the CO2 or vapor into the air and if we
neglect moisture adsorption or desorption on the walls in the vapor case, a first order
differential equation can be extracted from a mass balance of the room.
The mass balance can be written as incoming (CO2 or vapor) + (CO2 or vapor) produced
in a zone = outgoing (CO2 or vapor) + (CO2 or vapor) stored in the zonal air:

             IN + PRODUCTION = OUT + STORAGE

This leads to the first order differential equation
             dC i   .
Gp = V            + V (C i − C e )                                         2-16
              dt

with
        Gp            = (CO2 or vapor) source production                  [kg/s]
        V             = indoor air volume                                 [m3]
        Ci            = (CO2 or vapor) indoor concentration               [kg/m3]
        Ce            = (CO2 or vapor) outdoor concentration              [kg/m3]
         .
        V             = air infiltration or ventilation rate              [m3/s]

This results in the following exponential equation:
                                                          .

                                       Gp             −
                                                          V
                                                            t
 C i = C i ( 0) + ( C e − C i ( 0) +    .
                                            )(1 − e       V
                                                                )          2-17
                                       V

In section 8.6.5 the use of this equation is demonstrated as design help for the
consequences of an open gas infrared heating system for the church in Dordrecht.




                                                                2-31
2.3.2 Hygrothermal surface behavior by infrared thermography
For mould germination e.g. on an indoor surface the water activity at the surface is an
important quantity. In steady state or slowly changing conditions, the water activity is the
relative air humidity ϕ at the surface (Hens ea. 1991). It can be calculated from the vapor
pressure pv near the surface and the saturation pressure psat at that surface:
         pv
ϕ=                                                                             2-18
     p sat (θ si )

We found that there are only small spatial differences between the vapor pressures pv
measured. The air therefore is well mixed and differences from relative humidities
measured arise from differences in the saturation pressures, e.g. near the surfaces. The
vapor pressure pv near the surface thus may be taken from indoor air relative humidity
measurements or dew point measurements.
The saturation pressure at the indoor surface is related to the indoor surface temperature
θsi and can be derived from infrared thermal images of the indoor surface, as for
temperatures θ between –20 and 60 oC, psat is in good approximation

            θ ≤0 o C
                         θ

p sat (θ ) = 611e 272.44+ θ
                    22.44

                                                                               2-19

            θ >0 o C
                         θ

p sat (θ ) = 611e 234.18+ θ
                    17.08


where θ is the temperature in oC.




  Figure 2-7: Thermograph                           Figure 2-8: ‘Hygrograph’




                                            2-32
       Each pixel in an infrared thermograph represents an infrared measured surface
       temperature. A Matlab routine has been written to calculate the relative humidity at each
       pixel in the thermograph, from the saturation pressure of the measured surface
       temperature at that pixel and the measured vapor pressure from equation 2.44. The result
       is, what will be called, a surface hygrograph: a two-dimensional representation of the
       relative humidity close to the surface.

       2.3.3 Validation
       In the next figures a comparison was made between indoor glass surface temperatures in
       the Bemmel case, measured by infrared thermography (Tsurf) and thermistors (Tsensor).
       The absolute humidity near the surfaces was calculated from a dewpoint measurement
       (Tdew). The figures show the infrared measured temperatures to almost equal the
       thermistor measured surface temperatures. As a result the calculated relative humidity
       near the surfaces also agreed.
                           Surface temperatures leaded windows                                                      Relative humidity near surface leaded windows
            12                                                                                         1
                                                                      Tsurf1                                                                                        RHsurf1
                                                                      Tsurf2                         0.95                                                           RHsurf2
            11                                                        Tsensor
                                                                      Tdew                            0.9

            10                                                                                       0.85

                                                                                                      0.8
            9
Temp [oC]




                                                                                            RH [-]




                                                                                                     0.75
            8
                                                                                                      0.7


            7                                                                                        0.65

                                                                                                      0.6
            6
                                                                                                     0.55

            5                                                                                         0.5
            13.5   14   14.5   15     15.5     16     16.5       17   17.5      18                      13.5   14   14.5     15     15.5     16     16.5     17     17.5      18
                                        time [h]                                                                                      time [h]



       Figure 2-9: Measured surface temperatures and dewpoint (left) and near surface relative
       humidity (right) (see section 8.7.5)




                                                                                     2-33
2.4    DIRECT INFRARED RADIATION IN A CHURCH

2.4.1 Objectives
Gas infrared heaters are used to compensate for a lack of thermal comfort due to low air
temperatures in a church. Principally they are dimensioned to give an operative temperature
comparable to thermal comfort conditions at living height. In a church the installation height
of the heaters most often is determined by the availability of drawbars and the area to be
heated. As a result the radiant heaters are often installed at a considerable height. Radiation
on surfaces, closer to the heaters than the living height, causes higher radiant flux densities
on these surfaces. The surface temperatures of these irradiated surfaces increase, which may
lead to dehydration of the objects.

2.4.2 Model requirements
To calculate the radiant effects of the gas infrared heaters on indoor furniture, wall and
pillar surfaces accurately, the geometry of the church and the radiant characteristics of the
heaters are most important. Therefore a model is needed to account for the direct radiant
effects of geometrical complex radiating heaters on geometrical complex configurations
of obstructing building elements.

2.4.3 Method
Existing geometric light models like Radiance well account for the accurate simulation of
emission from light sources and include specular and directional-diffuse reflection for
visible light. Radiance further supports the implementation of complex geometries. It
employs a so-called light-backwards ray-tracing method. Light is followed along
geometric rays from the point of measurement (the view point or virtual photometer)
into the scene and back to the light source. The result is mathematically equivalent to
following light forward, but the process is generally more efficient because most of the
light, leaving a source, generally never reaches the point of interest.

2.4.4 Description
Because infrared- and light radiation are both examples of electromagnetic radiation, the
light ray-tracing model Radiance can be used for both simulations too. Because most
building material surfaces are close to black regarding infrared radiation (the emission
factors are nearly 1), only the direct component of radiation was accounted for.
The analogies for light- and thermal infrared radiation are:


Light radiation                                  Infrared radiation
Light power [lumen]                              Heat power [watt]
Illuminance [lumen/m2]                           Irradiance [watt/m2]
Luminance [lumens/steradian.m2]                  Radiance [watt/steradian.m2]

Table 2-3: Analogies for light- and thermal infrared radiation



                                             2-34
2.4.5 Input data
To calculate the radiant effects of the heaters the radiant characteristics of the heaters have to
be described.

Radiant heaters
For radiation in a half room, without thermal air absorption, the radiant flux density E,
on a plane perpendicular to the direction of the source as a function of the distance to the
source, can be calculated from the inverse square law:
             n
E 2  r1 
   =                                                                              2-20
E 1  r2 
     

Where
        r=        distance to the radiant source                                   [m]
        E=        radiant flux density                                             [W/m2]
        n=        factor, dependant on source;                                     [-]
                  0 for infinite plane sources, 1 for infinite line sources
                  and 2 for point sources

For the gas infrared sources most frequently used the factor n is mostly in the range 1.95 to 2
(Kämpf 1994). The radiant flux density distribution of the heaters has to be be taken from
measurements, e.g. of (Kämpf 1994). The radiant flux density distribution of the heaters
was determined by measurements with a radiometer in a half room. The distance of the
radiometer to the center of the heater was described to be 2753 mm. An example of the
radiant flux density distribution of one of the small heaters in the Dordrecht case (10 kW,
type 2 of GoGas) is given in Figure 2-10:




Figure 2-10: Radiant flux density distribution of a small heater (10 kW, type 2 of GoGas)



                                               2-35
The cosine law describes the cosine curve in the figure:
E s ,β = E s , n ⋅ cos(β)                                                             2-21

Where
        Es,β           =          Radiant flux density in direction with angle β to
                                  normal                                              [W/m2]
        Es,n           =          Radiant flux density in normal direction            [W/m2]
                                                           dA



                                                 β2
                                                      β1
                                       E s,β2




                                        E s,β1



                                                            E s,n



Figure 2-11: Radiant flux density distribution as a function of radiant direction

To convert the radial radiant flux density distribution to a horizontal surface
equation 2-22 can be used:
                              n
           R ⋅ cos(α ) 
Es = ER ⋅               ⋅ cos(α)                                                    2-22
                H      

The radiant efficiency ηs of the heaters is defined as the fraction of radiant emitted power
and total gas heat power:

       Qs        ∑ E ∆A  s
ηs =         =    A
                                                                                      2-23
       Q gas          Q gas

Where
        Qs             =          radiant emitted power of heater                     [W]
        Qgas           =          total gas heat power                                [W]
        Es             =          measured radiant flux density in a half sphere      [W/m2]
        ER             =          radiant heat flux at a distance R normal on the
                                  direction                                           [W/m2]
        ∆A             =          measurement related surface                         [m2]
        A              =          surface of the half sphere                          [m2]



                                                       2-36
                                                            R
                                                    α
                                           ER




                       Es                               H



Figure 2-12: Conversion of radial radiant flux density distribution to a horizontal plane


In the Radiance calculations for the Dordrecht case, the radiant efficiency of the radiant
heaters used has been approximated by ηs = 0.6.
Radiant heaters are mostly dimensioned, making use of the air temperature experienced,
the resultant radiant indoor temperature θres. It is derived from:
φ = h c A c (θ i − θ s ) + εA r (E s + σTr4 − σTs4 )                              2-24


                                    εE            
φ = h c A c (θ i − θ s ) + A r h r  s + θ r − θ s 
                                    h             
                                    r                                           2-25
  = (h c A c + A r h r )(θ res − θ s )


          h c A c θi + A r h r θr      εA r E s
θ res =                           +
             h cAc + Arh r          h cAc + Arh r
                                                                                  2-26
                     εA r E s
     = θ res 0   +
                   hcAc + Arh r

if θr ≈ θi then
                    εA r E s
θ res = θ i +                                                                     2-27
                 h cAc + h rAr




                                                    2-37
Where
        ε              =   emissivity of person’s surface                      [-]
        hc             =   convective heat transfer coefficient                [W/m2K]
        Ac             =   person’s convective surface area                    [m2]
        hr             =   radiant heat transfer coefficient                   [W/m2K]
        Ar             =   person’s radiant surface area                       [m2]
        Es             =   radiant heat flux density                           [W/m2]
        θs             =   person’s surface temperature                        [K]
        θr             =   mean radiant temperature of environment
                           without heater                                      [K]
            1
When A r ≈     A cv , h cv ≈ 2 and h r ≈ 5 is substituted in equation (2-27) the resultant
             2
temperature θres can be calculated from
θ res ≈ θ i + 0.1E s                                                            2-28

Figure 2-13 has been taken from (Kämpf 1994). From this figure the linear factor can be
calculated to be approximately 0.072.




Figure 2-13: Resultant radiant indoor temperature θr (Tr) according to (Kämpf 1994)

For a given desired resultant radiant indoor temperature θr and a design air temperature θi
the desired radiant flux density Es at living floor level can be taken from Figure 2-13 or
can be calculated from equation 2-28. Given the installation height of the gas infrared
heaters (mostly determined by the height of the drawbars) and the radiant distribution of
the heater the maximum radiation at floor level can be calculated from equation 2-20.
Equation 2-22 can be used to calculate the floor level radiant flux density at different
angles.



                                            2-38
2.4.6 Limitations of the model
In the model only the direct radiation effects are included. The effects of heating up the
materials and therefore changing temperatures and changing radiation are not included.
Interreflection effects also are neglected.




                                           2-39
2.5      HYGROTHERMAL LOAD OF ELEMENTS AND MATERIALS

2.5.1 Objectives
To predict the effects of the changing in- and outdoor climate on monumental
construction and interior parts, a more detailed description and calculation of thermal and
hygric behavior of these parts is needed.

2.5.2 Model requirements
Because of the more detailed character and the directional properties of materials 2- and
3 D calculations are needed to make a more detailed description of the thermal and hygric
behavior of construction and interior. Furthermore these behaviors are a function of time
due to changing climate conditions and these time dependant effects need to be
calculated. The equations, which describe thermal and hygric transport, are strictly
coupled and need to be solved accordingly.

2.5.3 Method
The output of the, by WaVo, calculated and/or measured indoor climate as a result of
outdoor climate, heating system and use of the church will be used as boundary
conditions for the more detailed modeling of construction and interior parts. To perform
these more detailed calculations on the thermal and hygric behavior of thick walls and
monumental interior parts a thermal and hygric description model has been adapted from
(Künzel 1994), which again was adopted from (Kiessl 1983), and has been implemented
in Matlab (1-D), Femlab (2-D) and FlexPDE (2- and 3-D).

2.5.4      Description

Governing differential equations
For the coupled heat- and moisture transport in porous materials the governing
differential equations can be written (Künzel 1994)

Heat equation:
      ∂θ        →
ρc       = −∇ ⋅ q + S h                                                       2-29
      ∂t

Moisture equation:
    ∂ϕ        →
ξ      = −∇ ⋅ g + S w                                                         2-30
    ∂t

where
      dw
ξ=                                                                            2-31
      dϕ


                                           2-40
The heat flux is given by:

→
q = − k∇θ − h v δ p ∇(ϕ ⋅ p sat (θ))                                                 2-32


The moisture flux is given by:

→
g = − D ϕ ∇ϕ − δ p ∇(ϕ ⋅ p sat (θ))                                                  2-33

where
         w          =         water content                                          [kg/m3]
         θ,T        =         temperature                                            [oC,K]
         φ          =         relative humidity                                      [-]
         t          =         time                                                   [s]
         k          =         thermal conductivity                                   [W/mK]
         Dφ         =         liquid diffusivity                                     [kg/ms]
         δp         =         vapor permeability                                     [kg/msPa]
         hv         =         latent heat of evaporation                             [J/kg]
         psat       =         saturation pressure                                    [Pa]
         Sh         =         heating source term                                    [W/m3]
         Sw         =         moisture source term                                   [kg/m3s]

2.5.5 Input data
The material properties in the calculations were measured or taken from WUFI. The
boundary indoor conditions were measured in situ or calculated by WaVo. The outdoor
climate was taken from measurements in situ or measurements from a nearby weather
station of the KNMI. 6

2.5.6 Limitations of the model
Water transport by temperature gradients is neglected in the model of Künzel
(Grünewald 1997). Salt transport is not included in the model. Air and transportation of
air are not integrated in the model. The phase separation of vapor and liquid flow is in a
way arbitrary. The coupling of heat and moisture is not strong. No hysteresis is included.

2.5.7 Validation
The results of the hygrothermal measurements on wood from section 3.3.2 were
compared with the hygrothermal modeling. Figure 2-14 shows the hygrothermal
response, as calculated with the model, related to the free deformation of the cylindrical
wood sample measured in the NMR measurements.


6   KNMI is the national research and information center for climate, climatic change and seismology
     in the Netherlands



                                                2-41
                                        tangential deformation sample 50                                                                      tangential deformation sample 50
                     0                                                                                                  0.6
                                                                           tan meas                                                                                              tan meas
                                                                           tan sim beta 1E-3                                                                                     tan sim beta 1E-8
                   -0.2                                                    tan sim beta 1E-8                            0.5


                   -0.4                                                                                                 0.4
deformation [mm]




                                                                                                     deformation [mm]
                   -0.6                                                                                                 0.3


                   -0.8                                                                                                 0.2


                    -1                                                                                                  0.1


                   -1.2                                                                                                   0


                   -1.4                                                                                                 -0.1
                          0   20   40     60      80      100     120        140     160       180                             0   50   100     150     200     250     300        350     400       450
                                                   time [hr]
                                                                                                                                                         time [hr]



Figure 2-15: Comparison of simulation and measurement of free deformation results due
to changing RH at a wooden cylinder of 25 mm. Stepwise changing RH from nearly 0 to
90 %RH (left) and sine curve changing RH from nearly 0 to 90 %RH with a period of 24
and 12 hours (right) (see section 3.3.2)




                                                                                                 2-42
2.6     MECHANICAL RESPONSE OF ELEMENTS AND MATERIALS

2.6.1 Objectives
The changing indoor climate of a church may lead to dimensional changes of the
(organic) indoor construction and interior parts. These dimensional changes result from
changing material temperature and moisture content. To predict the deformation of
monumental interior parts like monumental organs, the temperature and moisture effects
of the changing indoor climate must be known. In case of restriction of deformation or
non-uniform moisture distribution, the heat and moisture effects lead to internal stresses
in the material, which ultimately may cause damage to it.

2.6.2 Model requirements
A model is needed to calculate stresses and strains in relation to changes in temperature
and moisture content of wood.

2.6.3 Method
In a first approximation the deformation of wood will be considered to be elastic and
orthotropic. Partial differential equations (PDE) for the linear stress and strain
description by Hooke’s Law will be coupled to PDE’s for heat and moisture description.
The general procedure will be first to solve for the displacements, then compute strains
by differentiation and finally extract stresses by Hooke’s law. The calculations have been
done using FlexPDE.

2.6.4    Description




Figure 2-16: General state of stresses (Hibbeler 1997)

For a 3-D volume element in a continuous body in equilibrium the differential equations
of mechanical equilibrium in Cartesian coordinates are described by (Lekhnitskii, 1963).
In a first approximation wood can be considered as an orthotropic material. Due to the
strength in the axial direction the displacements in this direction can be neglected. The
third dimension, however, cannot be decoupled in a simple way by ‘insulation’ or by
considering the cross section of a long bar. According to Poison’s ratio, stresses in the xy



                                            2-43
plane generally produce strain in the z direction as well. In elasticity we choose for
vanishing z-stress σz as an assumption: no external force opposes deformations along the
third axis (Backstrom 1994). Therefore plain stress is assumed: σz = 0.
                                y




Figure 2-17: Plane stress (Hibbeler 1997)

The components of stresses in a continuous plane in equilibrium under the action of
surface (without external body forces) satisfy two differential equations of equilibrium.
These equations expressed in Cartesian coordinates have the form:

∂σ x ∂τ xy
    +      =0                                                                  2-34
 ∂x   ∂y

∂τ yx       ∂σ y
        +          =0                                                          2-35
 ∂x         ∂y

where
        σx,y        =           the normal components of stress               [N/m2]
        τxy, τyx    =           the shear components of stress associated     [N/m2]
                                with two axes

The relation between stress and strain is described by the generalized Hooke’s law and for
an anisotropic material it can be written as:
          1                µ xy         
                       −            0 
 εx   x                                σ  α
            E                Ey
          µ yx                           x   x           κx 
                          1                                
 εy  = −                          0  σ y  +  α y ∆θ +  κ y ∆w       2-36
γ   Ex                   Ey              
                                            τ        0         0
 xy                               1  xy                
          0
         
                            0            
                                   G xy 
                                         


                                                 2-44
where
        εx, εy       =         the normal strain components                    [-]
        γxy          =         the shear strain component associated with
                               two axes                                        [-]
        µxy, µyx     =         Poisson’s ratio                                 [-]
        Ex, Ey       =         moduli of elasticity or Young’s moduli          [N/m2]
        Gxy          =         the shear modulus                               [N/m2]
        αx, αy       =         the linear thermal expansivity                  [m/m.K]
        ∆θ           =         a temperature increment                         [K]
        w            =         moisture content                                [kg/m3]
        κx, κy       =         the linear relative deformation (shrinkage or
                               swelling) due to moisture content changing [m/m(kg/m3)]

In case of small displacements of a continuous body, we can write
       ∂u                ∂v                   ∂v ∂u
εx =      ,       εy =        and    γ xy =     +                                               2-37
       ∂x                ∂y                   ∂x ∂y

Furthermore the matrix must be symmetrical, therefore
µ yx       µ xy
       =                                                                                        2-38
Ex         Ey

For the elastic description of an orthotropic material, 4 independent elastic material
properties must be known: Ex, Ey, Gxy and one of µxy or µyx. Because as not all of these
material properties were known, a further approximation was made:
                                            E
µ=µxy=µyx, E=Ex=Ey so G xy =                      .
                                         2(1 + µ)
In fact, the material was treated as an isotropic continuous body, with orthotropic
thermal and hygric deformation properties.

Inverting the two-dimensional matrix 2-36 and substituting the expressions for the
stresses in the equations of equilibrium 2-34 and 2-35, we obtain the PDEs:

 ∂  E  ∂u         ∂v                                      ∂  E  ∂v ∂u  
         2 
                 + µ − (α x + µα y )∆θ − ( κ x + µκ y )∆w   +                +  =0
                                                                                            
∂x  1 − µ  ∂x
                   ∂y                                      
                                                            ∂y  2(1 + µ)  ∂x ∂y  
                                                                                           
 ∂  E  ∂v ∂u   ∂  E  ∂v                    ∂u                                       
              +  +                 +µ          − (α y + µα x )∆θ − ( κ y + µκ x )∆w   = 0
∂x  2(1 + µ)  ∂x ∂y   ∂y  1 − µ 2  ∂y
                                              ∂x                                       
                                                                                            


                                                                                                2-39




                                                     2-45
2.6.5 Input data
The input of indoor changing climate conditions can be derived from indoor
measurements or WaVo calculated output data. Material data for thermal and hygric
elasticity may be derived from measurements (Stappers 2000, Hout 2001, Krijnen 2002)
or material property handbooks.

2.6.6 Limitations of the model
The model described above was based on linear elastic behavior of materials. Wood is a
difficult material to simplify: it is anisotropic and has a typical non-linear behavior.
Relaxation for example is one of the typical non-linear qualities, which should be included
in the modeling. The non-linearity is important, especially when it comes to describe the
state before cracking. The linear, quasi-orthotropic approach therefore is only a first
coarse start of modeling. From modeling side of view, the 2D limitation is not difficult to
improve to 3D. The material properties needed are the real problem.

2.6.7 Validation
First the quality of the simulation results will be tested. The results of the displacement
measurements of chapter 3 at the wooden cube of 50*50*50 mm3 will be compared to
the strain simulation results. The chosen boundary conditions are the suddenly changing
relative humidity conditions from 85 to 60 %RH, which led to the shrinking of the cube.
The vapor diffusivity, which was used, was the determined Dw=5.9⋅10-10 m2/s from NMR
measurements.
                                                 Shrinkage of a sample of 50*50*50 mm^3
                                    0
                                                                                          tangential sim
                                                                                          radial sim
                                  -0.1                                                    tangential meas
                                                                                          radial meas
                                                                                          axial meas
                                  -0.2


                                  -0.3


                                  -0.4
               Deformation [mm]




                                  -0.5


                                  -0.6


                                  -0.7


                                  -0.8


                                  -0.9


                                   -1
                                         0   5     10             15              20      25                30
                                                              Time [days]




Figure 2-18: Comparison of measured and simulated shrinkage of a cube of 50*50*50
mm3 at changing relative humidity from 85 to 60 %RH (see section 3.3.2)




                                                             2-46
When the results of measurements and simulations are compared, the agreement of
measured and calculated displacements is obvious: the shrinking deformation limit from
the measurements was used to determine the linear assumed shrinking as a function of
moisture content (see section 3.3.2). This assumption seems to be justified by the quality
of the results. The diffusivity used was determined by the NMR measurements.
The model therefore will later be used to give an estimation of the response effects of
changing sample dimensions and boundary conditions. The implementation of more
realistic three-dimensional, full orthotropic simulations also depends on the knowledge of
three-dimensional characteristic material data and is not available at this moment. More
results will be presented in a master study, which has started in March 2002.
The 2-D plane quasi-orthotropic simulation study was aimed at the stress effect of
changing boundary conditions from a sudden block shaped change of 85 to 25 %RH for
a wooden organ pipe and an organ wind drawer. The results are presented in the Walloon
Church case in appendix A (see section 8.1.9).




                                           2-47
2.7    INTRAZONAL AIRFLOWS IN A CHURCH

2.7.1 Objectives
The knowledge and prediction of indoor airflow is most important for preservation and
thermal comfort conditions. Thermal stratification, due to inappropriate air inlet
conditions, may lead to high air temperature and related dramatic low relative humidity
near monumental objects, situated in higher places, like organs. Airflows due to
buoyancy, e.g. caused by floor heating, may result in contamination by dry deposition on
(cold) surfaces. Cold surfaces of great height such as high (stained glass) windows result
in convective cooling of the air and may cause draught due to increased vertical
downward velocities.

2.7.2 Model requirements
It is difficult to predict indoor airflow calculation for large spaces with great heights, large
cold surfaces and several infiltration gaps. Airflows in monumental churches are
dominated by the forced air supply from the inlet of the heating system in case of air
heating. Hot surfaces from heating systems like floor and radiator panel heating and
thermal draught from cold surfaces like ceilings, walls and windows determine the
airflows for an important part. The three dimensional geometry of the room and
boundary conditions near inlet and walls, ceiling and floor are of great importance for the
generated air flow characteristics. The required model should account for these
conditions.

2.7.3 Method
The study of indoor airflows in rooms using numerical techniques has been going on for
nearly thirty years. The range of indoor airflow simulations varies from laminar to
turbulent, from one- to two- and three-dimensional, steady and transient and buoyancy-
affected flows. Numerical calculation techniques may be used to predict air velocity,
temperature and concentration distributions in a room, but validation of the computed
results by corresponding measurements is always required (Chen 1988). Therefore
Loomans (Loomans 1998) presented a research on the measurement and calculation of
indoor airflow. He used Fluent Structured (Fluent 1995) to perform the numerical
calculations. In his research he showed that the logarithmic wall function is not valid for
the type of boundary layer flows that appear indoors. He suggested to improve the wall
heat transfer characteristics by imposing convection heat transfer data. He preferred the
low-Reynolds variant of the RNG-k-ε for the simulation of indoor airflow and showed a
non-equidistant Cartesian grid to be capable of predicting a reliable flow pattern for the
normally rectangular indoor space.
In this research Looman’s suggestions have been followed and again Fluent Structured
has been used to predict the effects of changes in heating strategies, presuming an initial
case, which has been compared with some validating measurements. Due to the complex
nature of airflows in large monumental buildings with great height, non-insulated walls
and vaults and most often large areas of window planes, the validation can only be coarse.



                                              2-48
2.7.4        Description

Governing equations
The differential equations for flow in a room consist of the continuity equation, the
momentum equation and the energy equation. For low Reynolds numbers the equations
can be simplified.
Continuity equation:
    →
∇⋅u = 0                                                                         2-40

Momentum equation:
        →
   ∂u         →      →           →       T − T0
ρ0    + ρ 0 ( u⋅ ∇ ) u = µ 0 ∇ 2 u − ρ 0        g − ∇P                          2-41
   ∂t                                      T0

Energy equation:
          ∂T               →
ρ0 c p0      + ρ 0 c p 0 ( u ⋅ ∇ ) T = ∇ ⋅ ( k 0 ∇T )                           2-42
          ∂t

Where
            →
            u          =          velocity vector                               [m/s]
            P          =          the static pressure                           [Pa]
            µ0         =          the dynamic viscosity                         [kg/ms]
            k0         =          thermal conductivity                          [W/m.K]
            cp0        =          constant pressure (isobaric) specific heat    [J/kg.K]
            t          =          time                                          [s]
            ρ0         =          the density                                   [kg/m3]
            g          =          the gravitational acceleration                [m/s2]

For the prediction of the turbulent flow characteristics the k-ε turbulence model for the
mean turbulent flow was used. It calculates the statistical characteristics of the turbulent
motion by averaging the flow equations over a time scale much larger than that of the
turbulent motion.

2.7.5 Input data
In the design stage a first estimation of the order of magnitude of forced air inlet
conditions should be known as boundary conditions for the CFD model. From literature
a number of empirical relations is known for some simple air supply configurations
(Regenscheit 1976), (Katz 1974) and (Hanel 1994). Furthermore Kriegel gave an
indication of thermal draught conditions for isolated cases like thermal draught from
high, cold walls (Kriegel 1973).




                                                        2-49
2.7.6 Limitations of the model
The CFD model, which was used, is one of the leading models in the world. The
boundary conditions, however, are most determinative for the result. In this kind of
practical applications they are not well known. The wall temperatures, for example, were
assumed to be uniform over the surface. The heat transfer coefficient was also assumed
to be known and constant. The geometries were simplified to very simple geometries.
The end result therefore is also of very limited accuracy. The model however was used,
and is capable, for parameter changing, comparative studies.

2.7.7                   Validation
                                         Temperature stratification                                                      Temperature at
               16                                                                                         70             h i ht 4 9 )
                                  fan:         on                 Tin:        70.3                                                               CFD
                                                                                                                                                 measurement
               14
                                                                                                          60

               12

                                                                                                          50
               10                                                                                     C]
                                                                                                      o
                                                                                                      te
  height [m]




                                                                                                      mp
               8                                                                                      er 40
                                                                                                      atu
               6                                                                                      re
                                                                                measurement           [ 30
                                                                                CFD
               4                                                                difference

                                                                                                          20
               2
                                  vrijdag 15 januari 1999            1720.01
                                                                  time:
               0                                                                                          10
                    0    5   10          15         20      25           30          35   40                   0   0.5     1         1.5     2   2.5       3
                                              temperature [ oC]                                                                   time [h]



Figure 2-19: Comparison of measured and calculated results for temperature stratification
and time evaluation of air temperature (see section 8.1.9)

The actual results measured in the Delft case, compared with the Fluent simulation
results, show differences in the curve of stratification. At low and high heights the
agreement is rather good, at half height there is a difference of about 3 K. The trend in
time of the measured and simulated results is reasonable. Absolute differences however
vary from 0 to 5 K.

2.7.8 Empirical relations
For some simple geometries and configurations the following empirical relations exist:

Simple wall air supply
In most Dutch churches the air heating system consists of a warm air system with a
simple air supply wall grille in combination with an air extraction grille, e.g. in the floor.
To prevent thermal draught near people the air inlet most often is placed at an elevated
height above people, typically 2.5 to 3 metres above floor level. In case of a large
monumental church the air supply can be thought of as a non-isothermal air jet in a half-
room, which can be described by the air inlet angle α and the number of Archimedes. It
describes the quotient of the buoyancy and the kinetic energy.




                                                                                               2-50
       g ⋅ ∆θ 0 ⋅ D h
Ar =                                                                          2-43
           Ti ⋅ u 0
                  2




Where
        Ar          =     Archimedes number                                   [-]
        g           =     gravitational acceleration                          [m/s2]
        ∆θ0         =     temperature difference air-inlet and indoor air     [K]
        Dh          =     hydraulic diameter                                  [m]
        Ti          =     indoor air temperature                              [K]
        u0          =     air supply velocity at the inlet                    [m/s]

Regenscheit formulated a universal formula to calculate the course of the jet for heating
and cooling systems (Recknagel 1982):
                                                   3
y   x                              x       
  =                            D ⋅ cos(α ) 
      ⋅ tan(α) ± 0.33 ⋅ m ⋅ Ar                                              2-44
Dh Dh                          h           

Where
        x           =     x-coordinate jet                                    [m]
        y           =     y-coordinate jet                                    [m]
        α           =     angle of inlet air velocity to horizontal plane     [rad]
        m           =     mixing number = 0.12 to 0.2 for round jets          [-]

For the velocity in the centre of the jet he formulated:
ux x0         b   h      b
   =            =                                                             2-45
u0   x        h m⋅x      h

Where
        ux          =     velocity at horizontal distance from the inlet      [m/s]
        b           =     width of inlet                                      [m]
        h           =     height of inlet                                     [m]
        x           =     horizontal distance from the inlet                  [m]
        x0          =     length of kernel = h/m                              [m]

The formula is valid starting at
x 1 b
 = ⋅                                                                          2-46
h m h

In case of grilles, h is corrected to h=h/µr, where µ is a contraction factor and r is the
ratio of free surface in relation to the total surface of the grille (Recknagel 1982).
For the decrease of temperature in relation to the distance, the following can be written:




                                            2-51
∆θ x 3 x 0         b 3 h                 b
    =               =                                                                   2-47
∆θ 0 4 x           h 4 m⋅x               h

Where
        ∆θ0          =               temperature difference between air inlet- and      [K]
                                     indoor air temperature

        ∆θx          =               temperature difference between temperature at      [K]
                                     distance x from inlet and indoor air temperature

Katz formulated for the course of the air jet the following (Katz 1974):
                                 3
                             −
y = 0.17 ⋅ Ar ⋅ x 3 ⋅ D h        2                                                      2-48

For a rectangular wall air inlet supply, the hydraulic diameter is calculated from:

D h = 1.27 ⋅ 5
                 (a ⋅ b )3                                                              2-49
                  a+b


Thermal airflows at cold (and hot) surfaces
To obtain an indication of the magnitude of vertical airflows, due to cold or warm
surfaces, the work of Kriegel is important. He did an empirical research on upward and
downward directed airflows at warm and cold surfaces.
In large rooms the Rayleigh number Ra >> 1010 and therefore the airflows are always
turbulent (Kriegel 1973). If the room is large and the airflow can be assumed to be
independent of the rest of the surfaces and the heating system, the airflows due to
cooling or heating at vertical surfaces can be calculated from Kriegel’s formulas.
The co-ordinate system has been defined by the following figure:
                                                 y

                                             x

                                                                      u




                                             u

                                                                      x

                                                      g                    y

                                      θw < θl                    θw > θl

Figure 2-20: Co-ordinate system for relatively cold (left) and warm (right) surfaces




                                                      2-52
The massflow in the boundary layer can be calculated from:
 ⋅
M = 0.536 ⋅ ρ ⋅ ν ⋅ b ⋅ Gr 0.35                                                 2-50

where
        ρ          =         the mean air density determined by
                             the mean air temperature                         [kg/m3]
        ν          =         kinematic viscosity                              [m2/s]
        b          =         the width of the wall section                    [m]
        Gr         =         Grashof number                                   [-]
       ∆T g ⋅ L3
Gr =      ⋅ 2                                                                   2-51
       T∞   ν

where
        ∆T         =         temperature difference between air and wall      [K]
                             temperature

          ∆T = T∞ − Tw

        T∞         =         air temperature                                  [K]
        Tw         =         wall temperature                                 [K]
        L          =         characteristic length, i.e. height of the wall   [m]
        g          =         gravitational acceleration                       [m/s2]
        u          =         vertical air velocity                            [m/s]

The Reynolds number is defined by:
        ux
Re =                                                                            2-52
        ν

The relation with the Grashof number is
Re x = 1.25 ⋅ Gr 0.45                                                           2-53

The maximum vertical air velocity at different heights x can be derived from:
          Re x ν
u max =                                                                         2-54
           x

The width of the boundary layer can be calculated from:
               ⋅
           M
δ=                                                                              2-55
   0.386 ⋅ ρ ⋅ b ⋅ u max



                                                2-53
The mean temperature in the boundary layer is given by:
Tm = T∞ + (Tw − T∞ ) ⋅ 0.247 ⋅ Gr −0.017                                     2-56

The convective heat transferred at the wall can be calculated from:
             ⋅
Q c = c p ⋅ M⋅ (T∞ − Tm )                                                    2-57

The mean convective heat transfer coefficient is determined by:
                Qc
hm =                                                                         2-58
       L ⋅ b ⋅ (T∞ − Tw )

The vertical velocity at a distance y from the wall is calculated from:


                      1
                                                                 
                 y  10  y  
                                             2

u ( y) = u max   ⋅ 1 −  ⋅ 1.575 ⋅  y  − 2.628 ⋅ y + 1.538 
                                                                           2-59
                δ   δ            δ             δ        
                                                                 
               

The dimensionless air temperature at a distance y from the wall is calculated from:
                 1
             y 7
θ( y) = 1 −                                                                2-60
            δ

Where
        y            =      distance from wall                               [m]
        x            =      distance along the wall from the beginning of
                            the flow, downward in case of cooling, upward
                            in case of heating                               [m]
        cp           =      specific heat at constant temperature of air     [J/kgK]




                                            2-54
3 EXPERIMENTAL ASSESSMENT


3.1     INTRODUCTION
At the start of this study the literature research led to Table 1-2, which gives an
overview of preliminary assumptions regarding supposed characteristics of heating
systems. In the objectives of this work it was stated that these assumptions led to the
framework of this thesis. The most important part of the thesis therefore was
appointed to be the evaluation of heating systems commonly used in Dutch churches.
For that reason a measurement program to evaluate these heating systems was
formulated. This program was adapted to the specific questions for each evaluation
case to be answered. In this chapter the measurements –in situ- for the evaluation of
these case studies are described.
To predict the building physical behavior of churches in a more general sense and to
predict the effects of changes, simulation models were used. Where it was necessary
they were developed for, or adapted to, specific problems. An important objective of
the measurements therefore was also to create the possibilities to compare the results
of the test cases with simulation results. Simulation models need specific input data.
An example of this kind of input is the infiltration rate of a church.
To predict the effects of changing indoor climate on specific indoor interior parts like
wooden organ or altar parts, combined heat-, moisture-, stress- and strain-calculations
had to be performed. Some of the input information consists of material data, which
can be derived from well-known building physical material properties. Other data,
however, are not well known or have to be considered to be very specific. For this
reason some laboratory experiments were done. Examples are laboratory work on
specific data of wood properties and on the soot production of candles. To ascertain
the quality of the outcome of these calculations some additional measurements on
these combined effects were performed too.

3.2     INSTRUMENTATION FOR FIELD MEASUREMENTS

3.2.1    Temperature and relative humidity

Types of sensors
For the experimental work in situ Pt100 sensors (4 wire) were used for general
(shielded) air temperature and surface temperature measurements. The data were
acquired with Grant datalog systems with 4 wire electrical resistance measurement
possibilities. Typical accuracies for the 4 wire Pt100 temperature devices are 0.3 K
(DIN 43760/class B). The dimensions of the sensors are typically in the order of ∅ 10
mm.



                                          3-55
For most air temperature and relative humidity measurements Escort combined air
temperature and relative humidity sensors have been used. They were used in
combination with separate surface temperature sensors, making use of thermistor
temperature measurement sensors and electrical capacity relative humidity
measurement sensors. The units have integrated programmable datalog facilities, with
typical capacities of 2000 to 4000 readings. Typical accuracies of the temperature
measurement are 0.3 K and 3 % RH for the relative humidity measurements.
Hanwell radio frequency transmitters were used with a radio-log receiver to monitor
and record air temperature and relative humidity on great heights, without the need
for wires between each sensor and the receiver. The accuracy of the thermistor
measurement was 0.5 F, for the capacitive RH measurement 2 % RH.
The temperature and relative humidity measurement devices have been compared to
calibrating temperature and dewpoint temperature devices in a climate cabinet of
Weiss. The Weiss cabinet was programmed to process a temperature trajectory from
10 to 30 oC and a relative humidity trajectory from 30 to 90 % RH. Typical
accuracies for the calibrating reference temperature Pt100 sensors are 0.1 K and 0.2 K
for the Mitchell dew point device. The measurement devices were rejected if the
accuracy indicated in the technical specifications of the devices did not meet the
comparison accuracy measurement.

Infrared thermography
A Varioscan Thermography System was used for the infrared thermographic
measurements. The 2011 model used is nitrogen cooled and operates in the
wavelength range from 8-12 µm. The optics has been made of germanium. The
camera operates on the principle of object scanning through a 2-dimensional
reflecting scanner. The horizontal scanner scans in lines, with 300 pixels each being
recorded at a frequency of 270 Hz. The vertical scanner builds up the complete image
from 200 vertical lines. The image refresh frequency of the system is 1.25 Hz, the
image refreshing rate therefore is 0.8 s. The focal length of the optics is 35 mm, its
aperture is F:1.
After every scanned image a reference source (chopper) is moved into the optical path
and its temperature is measured by a thermocouple. On-line temperature
measurement is made possible by comparing the radiation intensity of object and
chopper. The temperature resolution of the measurement is < 0.1 K and the
geometrical resolution is 1.5 mrad. The factory-calibrated absolute accuracy of
temperature measurement at ambient temperature of 20 oC is ± (1.5 K + 1.5 % of
measuring range, being –10..100 oC). The unit has an individual data set of
calibration data and settings for temperature calculation from the dealer with typical
accuracies of ±1 K in the measuring range of –10..100 oC.
The radiation received by the camera in the scanning phase consists of the
characteristic radiation of the measured object, the reflected environment radiation
and the characteristic radiation of the air path. Total radiation thus is calculated from
the apparent temperature of the object, and from this the shares of environment
reflection and air path are subtracted (Jenoptik):




                                          3-56
Φ r = ε o τ p Φ (To , λ ) + (1 − ε o )τ p Φ (TA , λ ) + (1 − τ p )Φ (T∞ , λ ) + Φ u    3-1

where
        Φr           =         radiation heat flow received by camera                 [W]
        εo           =         emission factor of object                              [-]
        τp           =         transmission factor of air path                        [-]
        To           =         temperature of object                                  [K]
        TA           =         ambient temperature                                    [K]
        T∞           =         air temperature                                        [K]
        λ            =         wavelength                                             [m]
        Φu           =         control variable for offset and gain of the image
                               signal amplifier                                       [W]

During the measurements of non-metal objects the default values of εo=0.90 and τp=0
were used. The ambient temperature was set to the air temperature or black bulb
temperature (if measured).

3.2.2 Air velocity
For the air velocity measurements Dantec hot sphere anemometers (Low Velocity
Transducer 54R10; Dantec 1984) have been used. To allow fast sampling frequencies
(10 Hz) a special combination with the 54N21 3-channel input Module (Dantec) and
an A/D converter (DATAshuttle DS-12-8-GP; Strawberry Tree 1995) was applied.
The measurement devices were calibrated in the laboratory in a glass tube calibrating
device (Loomans 1998). Typical accuracy of the air velocity measurements has been
determined to be ±0.025 m/s at velocities between 0.05 and 0.50 m/s.

3.2.3 Ventilation and infiltration rate
For churches ventilation and infiltration rates are not easy to predict theoretically.
They mainly depend on the air leakage through openings like cracks and joints in
(stained glass) glazing, doors, vault and construction parts. The only way is to measure
the ventilation rate in different churches and to derive a relation with typical
characteristics of the churches. A measurement set-up was developed to get an
impression of the air exchange rate.
If there is a well mixing of a tracer gas inside the room, the bulk air exchange rate of a
room can be estimated by the concentration decay method. To determine the
homogeneousness of the mixing, a multiplexing device has been developed to take
samples at different locations in the church (at different locations in a horizontal plane
and at different vertical heights). A Bruel & Kjaer gas-monitoring device (B & K
1302) determined the concentration of samples of air. Typical processing and
sampling time intervals were 2 minutes. The mixing of the air in time was considered
to be satisfactory if the concentrations measured at different locations coincided
graphically in a graph of concentrations measured as a function of time. A typical
graph is represented in Figure 3-1. The figure shows the concentration SF6 at 6
different monitoring points. The gas was released near measuring point 1 in the choir.
After about 0.5 to 1 hour the mixing was complete.


                                                    3-57
                              2.5
                                                                                  1.   Choir
                                                                                  2.   Nave middle
                                                                                  3.   Organ
                                2
                                                                                  4.   Outdoor
                                                                                  5.   Pillar
                                                                                  6.   Nave ceiling
                              1.5
                  SF6 [ppm]




                                1



                              0.5



                                0



                              -0.5
                                     0   0.5       1     1.5    2       2.5   3          3.5      4   4.5
                                                                 time [h]




Figure 3-1: Concentration SF6 measured in the Hervormde Kerk in Beusichem

The mixing in all our measured churches proved to be satisfactory. Monitoring the
concentration of a tracer gas SF6 in the churches therefore resulted in the
determination of the bulk air exchange rate.
The mass balance for a room leads to the following differential equation:

    dC i     ⋅
V        = − V Ci                                                                                            3-2
     dt

where
        V                        =             volume of the room                                           [m3]
        Ci                       =             concentration tracer gas as a function of time               [kg/m3]
         ⋅
        V                        =             ventilation or infiltration rate                             [m3/s]

If one assumes a constant ventilating flow rate the indoor decay of concentration
tracer gas after interrupting the source can be written as:
                        .
                      V
                  −     t
C i = C i,0 ⋅ e       V
                                                                                                             3-3

where
        Ci,0                     =             concentration tracer gas at time t=0                         [kg/m3]




                                                                3-58
                            ⋅
                           V
The air exchange rate n =      can be determined by curve fitting the exponential
                           V
decay.
From a limited number of measurements (about 12) the following was concluded:
churches with an airtight vault, like plastered stone vaults, in most cases have a very
low air exchange rate. Values of 0.08 to 0.12 have been measured. Churches with a
wooden vault, consisting of boards, have a much larger air exchange rate. Values
ranging from 0.5 to 0.75 have been measured.




                                          3-59
3.3   LABORATORY MEASUREMENTS

3.3.1 Source of contamination
To determine the individual soot emission contribution of different candles and
incense a laboratory set-up to measure soot production of candles and incense has
been developed (Stevens, 1999). Filtered air from the laboratory is brought into a glass
tube, height 560 mm and width 210 mm, by a ventilator device, which is controlled by
a voltage supply. In the glass tube a candle is burned and the air flows upwards along
the candle into a filtering device. The filtering device is provided with an inflammable
fibreglass filter mat (manufactured by Whatman, type 1820, diameter 150 mm). To
determine the soot capture of the glass fibre a second glass fibre has been placed in
serial to the first one.
The filter mat is weighed before and after the candle burning using a Mettler HK 160
analytical balance (accuracy 0.1 mg). Before and after the test a Mettler precision
balance PC 4400 has been used to determine the candle weight (accuracy 0.1 g). To
control the air velocity an omni directional hot sphere anemometer (Disa low velocity
flow analyser 54N50) is positioned in the glass tube during the adjustment of the
ventilator device. Typical mean air velocities for the bulk flow were adjusted to 0.10
m/s at the flame position. To control a draught turbulent horizontal airflow for candle
flickering an additional computer ventilator device (Sunon DC12V, 2.6 W) generated a
controlled horizontal disturbance of the airflow. The mean air velocity at the flame
has been determined to be 0.37 m/s, with a turbulence intensity of 0.17.




Figure 3-2: Glass tube with ventilator device, anemometer and computing device

The glass tube was used to determine the soot production of candles and incense in
the case of St. Martinus Weert (see section 8.4).




                                         3-60
3.3.2 Experiments on wood
The adsorption of moisture makes hygroscopic objects swell and desorption leads to
shrinkage. During changing size the shape of the objects may also change. Objects in
churches like pews, pulpits and altars are often made of different pieces of materials,
mostly wood, which have been joined together. These different pieces rarely respond
to moisture changes in the same way: they have different time characteristics and
often have different directional characteristics. Wood, bone and ivory e.g. swell much
more across the grain than along it (Thomson 1978). Thomson describes the most
sensitive category for musea to be panel paintings, veneered furniture, musical
instruments and wooden objects from the tropics.
To determine the effects of indoor climate changes on interior elements of wood
some experiments were performed on the mechanical deformation of wood (Hout,
2001). The experiments were performed for the hygroscopic range only. To describe
the hygroscopical performance of wood, a number of characteristic material data of
wood are needed: the vapor diffusion resistance and the sorption isotherm to express
the dynamical relation between moisture content and indoor climate, the relation
between moisture content and deformation, and the stress and strain relationship.

Vapor diffusion resistance
In a standard laboratory wet-cup/dry-cup set-up with saturated aqueous salt solutions
in a desiccator, the vapor diffusion resistance of beech wood has been measured in
three directions: tangential, radial and axial. The measurements were performed in a
climate room at isothermal (20 oC) conditions for two trajectories of relative humidity:
from 85 to 33 %RH (mean 59 %RH) and from 85 to 53 %RH (mean 69 %RH). The
results are summarized in Table 3-1.
Mean relative       Thickness                    Mean vapor diffusion resistance
humidity            d [mm]                                  µmean [-]
RH [%]                                 tangential        radial           axial
59                  10.0               52 ± 3            27 ± 2           7±1
59                  15.0               71 ± 9            37 ± 2           7±1
69                  10.0               55 ± 6            35 ± 1           9±3
69                  15.0               79 ± 5            67 ± 13          9±1
Table 3-1: Vapor diffusion resistance of beech wood (based on 3 samples of
measurements)


Sorption isotherm
In the same climate room the sorption isotherm of beech wood was determined for
an adsorption trajectory from 33 %RH to 97 %RH and the reverse desorption
trajectory, making use of saturated aqueous salt solutions in a desiccator. For each
direction (tangential, radial and axial) the sorption isotherm was determined using 3
different test samples. The results are summarized in Figure 3-3. As can be seen from
the figure the different sorption isotherm curves nearly coincide.


                                          3-61
                                                              Equilibrium moisture content (EMC) of beech
                                           25




                                           20


               Moisture content EMC [m%]



                                           15




                                           10




                                           5




                                           0
                                                0   10   20   30        40        50        60        70    80   90   100
                                                                       Relative Humidity [RH%]




Figure 3-3: Equilibrium Moisture Content (EMC) of beech


Free deformation
The hygric deformation of wood due to a sudden change in relative humidity first was
tested on two beech cubic test samples of 50*50*50 mm3. For the adsorption test the
samples were initially isothermally (20 oC) conditioned at 33 %RH and hygroscopically
wetted to 60 %RH and for the desorption test they were conditioned at 85 and dried
to 60 %RH. The free hygric shrinkage and swelling was measured as a function of
time in three directions by displacement measurement gauges with an accuracy of
0.003 mm.




Figure 3-4: measurement set-up for shrinkage and swelling tests on a 50*50*50 mm3
sample

The graphical results are reported in Figure 3-5.


                                                                             3-62
                                                           Shrinkage of a sample of 50*50*50 mm^3                                                                                                      Swelling of a sample of 50*50*50 mm^3
                                       0                                                                                                                        1
                                                                                                                tangential                                                                                                                                       tangential
                     -0.1                                                                                       radial                                         0.9                                                                                               radial
                                                                                                                axial                                                                                                                                            axial
                     -0.2                                                                                                                                      0.8

                     -0.3                                                                                                                                      0.7
  Deformation [mm]




                                                                                                                                            Deformation [mm]
                     -0.4                                                                                                                                      0.6

                     -0.5                                                                                                                                      0.5

                     -0.6                                                                                                                                      0.4

                     -0.7                                                                                                                                      0.3

                     -0.8                                                                                                                                      0.2

                     -0.9                                                                                                                                      0.1

                                  -1                                                                                                                            0
                                           0           5          10            15                  20          25           30                                      0                        2   4        6        8       10      12          14         16       18         20
                                                                            Time [days]                                                                                                                                 Time [days]



Figure 3-5: Shrinkage (left) and swelling (right) of a 50*50*50 mm3 sample. Left: RH
change from 85 to 60 RH%. Right: RH change from 33 to 60 %RH.

From the figures it is seen that the largest deformation is in tangential direction. The
radial deformation is about 1.5 to 2 times smaller than the tangential deformation. The
axial deformation can be neglected. According to Camuffo, for practical purposes, the
relationship between deformation and equilibrium moisture content may be assumed
to vary linear (Camuffo 1998). The results from the wood deformation tests therefore
were used to determine the deformation as a function of equilibrium moisture
content. A linear relation was assumed, derived from the shrinkage deformation test
of the cubic 50*50*50 mm3 beech sample. The relationships are graphed in Figure 3-6.
                                                                             Deformation of beech                                                                                                                       Equilibrium deformation of beech
                                           2                                                                                                                                           8
                                                                                                                       tangential                                                                                                                                              tangential
                                                                                                                       radial                                                                                                                                                  radial
                                                                                                                                                                                       6
                                           0


                                                                                                                                                                                       4
                                        -2


                                                                                                                                                                                       2
                     Deformation [%]




                                                                                                                                                                     Deformation [%]




                                        -4

                                                                                                                                                                                       0

                                        -6
                                                                                                                                                                                       -2


                                        -8
                                                                                                                                                                                       -4


                                       -10
                                                                                                                                                                                       -6



                                       -12                                                                                                                                             -8
                                               0   2          4            6            8           10     12         14            16                                                   30       40           50              60             70            80            90                100
                                                                       Equilibrium Moisture Content [m%]                                                                                                                     Relative Humidity [%]




Figure 3-6: Wood contraction of wood as a function of EMC (left) and as a function
of RH (right)

The thermal and hygric deformation of beech wood in three dimensions was also
determined with an Electronic Speckle Pattern Interferometer (ESPI). The measuring
principle is as follows. The surface of a specimen is illuminated from two positions
with coherent laser light in various phases of testing.




                                                                                                                                         3-63
                  Figure 3-7: Measuring principle of an ESPI system

The speckle interferometry uses the interference characteristics of electromagnetic
waves. It is based on the fact that an optically rough surface appears granulated when
it is lit by a coherent light (laser). This phenomenon is called a speckle and the
individual grains are termed speckle.
After each load step, the reflected light is captured with a CCD camera. First a speckle
pattern is found, which includes the deformation information of each point of the
object measured. By subtracting speckle patterns from various phases of testing,
interference fringes are formed. The number of fringes and their widths are a measure
for the displacements of the illuminated area.
3D-Displacement and strain distribution can be determined with a resolution of 10
nm or 1 µm/m.




Figure 3-8: Measuring set-up of the ESPI system for determining the deformation of
beech wood

Automatic evaluation of the measurement by real-time subtraction and phase-shifting
algorithms is possible. Changing the polarity of the laser, displacements in X, Y and Z
direction can be obtained. To compare speckle patterns and to establish strains from
displacements measured special computer software is available.
Thermal and hygric deformation tests were performed on three tangential and four
axial sawn test samples of l*w*h=45*45*10 mm3. Figure 3-8 shows the measuring
set-up of the ESPI system. Because of the sensitivity for vibrations the measuring set-


                                          3-64
up consists of a heavy, stable metal table. The ESPI system was placed in the climate
room at 20 oC and 60 %RH.

Three types of measurement were performed:
   − Hygric deformation measurements at constant air temperatures and changing
        relative humidity;
   − Thermal deformation measurements at constant relative humidity and
        changing air temperatures;
   − Zero measurements at constant air temperature and relative humidity to test
        the stability of the ESPI system and the climate room.




Figure 3-9: Horizontal displacement of a beech test sample of l*w*h=45*45*10 mm3.

The results of these measurements are graphically represented in Figure 3-9. Figure
(left) shows the surface deformation at the beginning of the drying process. The
bottom of the sample is placed on a stable reference table and initially is drying less
than the other sides of the sample. At the end of the drying time the deformation of
the sample is nearly complete (right). The figures can be interpreted in the following
way: the point in the figure left above corresponds with the horizontal deformation of
the upper right corner of the sample, the upper right point in the figure with the upper
left corner of the sample. The camera looks at the front of the square of l*w=45*45
mm2. The next figures show the swelling deformation in tangential, radial and axial
direction due to a change in relative humidity from 33 to 60 %RH.




                                         3-65
                                       Humidifying swelling deformation                                                                               Humidifying swelling deformation velocity
                     140                                                                                                             1
                                                                               tangential                                                                                                              tangential
                                                                               radial                                               0.9                                                                radial
                     120                                                       axial                                                                                                                   axial
                                                                                                                                    0.8




                                                                                                    deformation velocity [µm/min]
                     100                                                                                                            0.7
  deformation [µm]




                                                                                                                                    0.6
                     80
                                                                                                                                    0.5
                     60
                                                                                                                                    0.4

                     40                                                                                                             0.3

                                                                                                                                    0.2
                     20
                                                                                                                                    0.1

                      0                                                                                                              0
                           0   2   4   6       8      10       12   14    16      18        20                                            0   2   4        6      8      10       12    14        16      18        20
                                                   time [hr]                                                                                                          time [hr]



Figure 3-10: Swelling humidifying deformation (left) and deformation velocity (right)
due to a change in relative humidity from 33 to 60 %RH

Figure 3-11 shows the shrinking deformation due to drying from 85 to 60 %RH in
tangential and axial direction.
                                         Drying shrinking deformation                                                                                   Drying shrinking deformation velocity
                     300                                                                                                             4
                                                                               tangential                                                                                                              tangential
                                                                               axial                                                                                                                   axial
                                                                                                                                    3.5
                     250

                                                                                                                                     3
                                                                                                    deformation velocity [µm/min]




                     200
                                                                                                                                    2.5
  deformation [µm]




                     150                                                                                                             2


                                                                                                                                    1.5
                     100

                                                                                                                                     1

                     50
                                                                                                                                    0.5


                      0                                                                                                              0
                           0   2   4   6       8      10       12   14    16      18        20                                            0   2   4        6      8      10       12    14        16      18        20
                                                   time [hr]                                                                                                          time [hr]



Figure 3-11: Shrinking drying deformation (left) and deformation velocity (right) due
to a change in relative humidity from 85 to 60 %RH.

A comparison between thermal and hygric shrinkage is given in Figure 3-12.
                                         Drying shrinking deformation                                                                                   Drying shrinking deformation velocity
                     300                                                                                                             4
                                                                                hygrical                                                                                                                hygrical
                                                                                thermal                                                                                                                 thermal
                                                                                                                                    3.5
                     250

                                                                                                                                     3
                                                                                                    deformation velocity [µm/min]




                     200
                                                                                                                                    2.5
  deformation [µm]




                     150                                                                                                             2


                                                                                                                                    1.5
                     100

                                                                                                                                     1

                     50
                                                                                                                                    0.5


                      0                                                                                                              0
                           0   2   4   6       8      10       12   14    16      18        20                                            0   2   4        6      8      10       12    14        16      18        20
                                                   time [hr]                                                                                                          time [hr]



Figure 3-12: Comparison between hygric (85 to 60 %RH) shrinkage due to drying and
thermal shrinkage (30 to 20 oC)


                                                                                                 3-66
The thermal deformation in this range is about 15 % of the hygric shrinkage.
Furthermore thermal deformation is more than 10 times faster, compared to hygric
deformation velocity.

Moisture content
In the study of the moisture transport in wood it is very important to measure
moisture profiles as a function of time. Nuclear magnetic resonance (NMR) is a
powerful technique to measure the time evolution of moisture profiles with a high
spatial resolution. However, many technological porous materials contain large
amounts of paramagnetic ions, which complicates the NMR measurements. Therefore
the moisture transport in porous media is studied using specially developed NMR
scanners. The dynamically changing moisture content of wood due to changing
climatic conditions was therefore measured in an NMR set-up at the Centre for
Material Research with Magnetic Resonance, Eindhoven University of Technology.
The method is extensively described by (Pel, 1995). In an NMR experiment the
magnetic moments of hydrogen nuclei are manipulated by suitably chosen
electromagnetic alternating radio frequency (RF) fields. In a pulsed NMR experiment
the orientation of the moments of the spins in a static magnetic field is manipulated
by short electromagnetic pulses at the resonance frequency, bringing the system in an
excited state. The amplitude of the resulting so-called spin-echo signal is proportional
to the number of (hydrogen) nuclei excited by the radio frequency field. Because of
the resonance condition for nuclei the method can be made sensitive to hydrogen only
and therefore the spin-echo signal is a measure for the moisture content.
The NMR measurement set-up consists of a cylindrical coil inserted between two
magnets. The cylindrical samples can be inserted in the cylindrical coil. In this scanner
it is possible to measure quantitatively the moisture profiles with a spatial resolution of
0.5 mm in two dimensions. To enable the exposure of the sample to temperature and
humidity conditioned air, a cylindrical wooden sample with a diameter of 25 mm and a
length of 200 mm is placed in a cylindrical tube with an inner diameter of 50 mm and
a length of 1500 mm. Conditioned air is prepared by mixing dry and moist air. It is
brought into the tube and is blown along the cylindrical surface of the wooden
sample. To smooth the airflow along the cylinder a cylindrical synthetic sample with a
pointed front is placed in front of the cylindrical wooden sample. Temperature and
relative humidity of the air are measured at the in- and outlet of the perspex tube. The
radial and tangential deformation of the wooden sample is measured by resistance
strain gages, which are connected to clips over the diameter of the cylinder. The
typical accuracy of this measurement is in the order of 1 µm. Figure 3-13 shows the
NMR-setup and the prepared wooden sample.




                                           3-67
Figure 3-13: NMR set-up (left) and prepared beech wooden sample (right)

The schematic drawing in Figure 3-14 shows the vertical section of the perspex tube
and the sample within.




Figure 3-14: Schematic section of the perspex tube


The NMR spin-echo signal was related to the moisture content of wood using
moisture prepared samples. The mean of the determined moisture profile from the
NMR-signal was associated to the mean moisture content of the sample. Figure 3-15
shows the calibration curve.




                                        3-68
                                                                          Calibration curve of beech wood
                                                  0.4



                                         0.35



                                                  0.3



                                         0.25
              Mean NMR-signal [V]




                                                  0.2



                                         0.15



                                                  0.1



                                         0.05



                                                   0
                                                        6        8   10         12              14          16        18            20
                                                                              Moisture content [m%]




Figure 3-15: calibration curve of beech wood

The NMR set-up we used is not able to detect moisture contents less than 7 mass%.
Within this limitation the drying and humidifying process of wood was monitored.
Figure 3-17 shows the drying process of an initially at 85 RH% conditioned axial sawn
wooden sample, which was exposed to absolutely dry air.

                                                  20
                                                                                                                           0 hr
                                                                                                                           2 hr
                                                                                                                           4 hr
                                                  18                                                                       6 hr
                                                                                                                           8 hr
                                                                                                                           10 hr
                                                                                                                           12 hr
                                                  16                                                                       14 hr
                                                                                                                           16 hr
                                                                                                                           18 hr
                                                                                                                           20 hr
                          Moisture content [m%]




                                                                                                                           50 hr
                                                  14
                                                                                                                           100 hr
                                                                                                                           200 hr


                                                  12




                                                  10




                                                   8




                                                   6
                                                            14       16           18              20             22        24
                                                                                  Position [mm]



Figure 3-16: Drying cylinder of beech; moisture profiles over half of diameter




                                                                                  3-69
                                      20
                                                                                     12.5 mm
                                                                                     2 mm
                                                                                     0 mm
                                      18




                                      16
              Moisture content [m%]




                                      14




                                      12




                                      10




                                      8




                                      6
                                           0   50   100       150        200   250             300
                                                          Time [hours]




Figure 3-17: Drying cylinder of beech; time evolution at three different depths to the
surface




                                                          3-70
Moisture diffusivity
The NMR measured moisture profiles are used to determine the moisture diffusivity
of wood, making use of partial differential equations to describe the isothermal
moisture transport in wood. The moisture diffusivity of porous materials is dependent
on the moisture content of the material. In a first approximation, however, the
moisture diffusivity will be considered to be a constant. The moisture profiles from
the drying experiment are used as a reference. Femlab is used to simulate the moisture
profiles using an initial guess for the moisture diffusivity. The guess for the diffusivity
is updated until the optimal value for it is reached, i.e. when the sum of the squares of
the deviations is minimal. The geometry of the model was the 2D circular section of
the cylindrical wooden sample with diameter 0.025 m.
The Femlab PDE model used to describe the moisture evaluation in wood is the
following:
∂w
   = ∇ ⋅ ( D v ∇w )                                                                                                        3-4
∂t

The initial value for the moisture content at t=0 was the moisture profile first
measured. The moisture content at the surface was considered to be w=0 for t>0.
Because the NMR signal for moisture contents below 7 mass% returned values for the
moisture content of 7 mass%, these lower values were not accounted for.
                                            Optimal diffusivity: 5.8884e-010 m2.s -1

                                                                                       Simulated profile
                                                                                       Measured profile
                              740                                                           t=2 hr

                                                                              t=8 hr

                              720
   Moisture content [kg/m3]




                                                                    t=14 hr
                              700



                              680                               t=16 hr



                              660



                                                                                                           -0.0125   0.0         0.0125
                              640
                                    -0.01   -0.005             0              0.005              0.01
                                                         Position [m]


Figure 3-18: Fitting the diffusivity from simulated moisture profiles to measured
moisture profiles

The result of the fitting of the measured moisture profile to the simulated profile is
indicated at the top of Figure 3-18: Dv=5.9⋅10-10 m2/s.




                                                                                   3-71
This vapor diffusivity was compared with results from previously measured material
properties, i.e. the vapor diffusion resistance µ and the tangent of the sorption
isotherm ζ. The measured interval for µ was about 7<µ<79; the tangent of the
sorption isotherm was about ξ=140. For this range the diffusion coefficient Dv was in
the range of 3.8⋅10-9 to 4.3⋅10-8. The difference may be the result of the hygroscopic
range of the measurements (30 <RH%<95) of material properties and the range of
the drying process (0<RH%<75).

The calculated diffusion coefficient Dv from NMR results was used for a comparing
simulation study of free deformation measurement results of drying and humidifying,
stepwise and according to a sine curve varying boundary conditions at the cylindrical
wooden sample. The results were fitted using a varying surface vapor transfer
coefficient β.
The results for a stepwise varying relative humidity at the surface of the wooden
cylinder of the NMR measurements, described before in chapter 3, are given in Figure
3-19. The simulated deformation was assumed to vary linearly with the moisture
content. The asymptotic deformation values from the measurements were the input
values for this linear relation.

                                                        tangential deformation sample 50
                                     0
                                                                                           tan meas
                                                                                           tan sim beta 1E-3
                                   -0.2                                                    tan sim beta 1E-8


                                   -0.4
                deformation [mm]




                                   -0.6


                                   -0.8


                                    -1


                                   -1.2


                                   -1.4
                                          0   20   40     60      80      100     120        140     160       180
                                                                   time [hr]


Figure 3-19 Comparison of simulation and measurement results of stepwise changing
RH from nearly 0 to 90 %RH at the surface of a cylinder of beech

When we compare the results of the tests and the simulation the influence of the
vapor transfer coefficient clearly is important. In a future measurement set-up this
coefficient should be measured. Furthermore there is a remarkable irregularity in the
measurement results. Extrapolating the curve without this irregularity would plead for
a value of β = 1E-8 for the vapor transfer coefficient to be used.




                                                                   3-72
The results for a sine curve varying relative humidity at the surface of the wooden
cylinder of the NMR measurements, using this vapor transfer coefficient, are given in
Figure 3-20. The quality of the results is very promising: The curves of measurement
and simulation nearly coincide.
                                                         tangential deformation sample 50
                                   0.6
                                                                                            tan meas
                                                                                            tan sim beta 1E-8
                                   0.5


                                   0.4
                deformation [mm]




                                   0.3


                                   0.2


                                   0.1


                                     0


                                   -0.1
                                          0   50   100     150     200     250     300        350     400       450
                                                                    time [hr]


Figure 3-20: Comparison of simulation and measurement results of sine curve
changing RH from nearly 0 to 90 %RH at the surface of a cylinder of beech with a
period of 24 and 12 hours




                                                                    3-73
Modulus of elasticity
In the relation between stress and strain the modulus of elasticity E is the most
important material property. In co-operation with the Catholic University of Leuven
the modulus of elasticity of beech wood was determined as a function of the moisture
content in the three characteristic wood directions.
Nine small cylinders of wood, three in each direction, with a diameter of 25 mm and
a height of 50 mm, were cut out of a timber. The wooden samples initially were
brought to the following three moisture contents: laboratory equilibrium moisture
content (about 20 oC, 50 % RH), capillary saturated moisture content and fully
immersed. They were weighed before and directly after the test.
In a servo hydraulic compression and tensile stress facility the wooden cylinders were
compressed. The force applied on top of the cylinder and the strain, measured by
strain gauges connected to the samples over the height of the cylinder, were
registered as a function of time. The tangent of the stress strain diagram afterwards
determined the modulus of elasticity. Figure 3-21 shows the test facility.




Figure 3-21: Compression and tensile test facility at the Catholic University of Leuven

Figure 3-23 shows the results of the test for one (tangential) direction and three
moisture contents.
                                   wo1acap1, E = 378.0695 MPa                                                      wo1bwet1, E = 166.893 MPa
                  4                                                                               3


                 3.5
                                                                                                 2.5

                  3

                                                                                                  2
                 2.5
  stress [MPa]




                                                                                  stress [MPa]




                  2                                                                              1.5


                 1.5
                                                                                                  1

                  1

                                                                                                 0.5
                 0.5


                  0                                                                               0
                       0   0.005   0.01       0.015      0.02   0.025   0.03                           0   0.005       0.01                0.015   0.02   0.025
                                            strain [-]                                                                        strain [-]



Figure 3-22: Stress strain diagrams for capillary saturated (left) and immersed wetted
(right) beech samples


                                                                               3-74
                                  wood1a1, E = 1558.456 MPa                                                               wo1cdry1, E = 1576.3328 MPa
                 12                                                                                14


                                                                                                   12
                 10


                                                                                                   10
                 8
  stress [MPa]




                                                                                    stress [MPa]
                                                                                                   8
                 6
                                                                                                   6

                 4
                                                                                                   4


                 2
                                                                                                   2


                 0                                                                                 0
                      0   0.005       0.01                0.015   0.02   0.025                          0   0.002 0.004 0.006 0.008     0.01 0.012 0.014 0.016 0.018   0.02
                                             strain [-]                                                                               strain [-]



Figure 3-23: Stress strain diagrams for equilibrium moisture content (left) and dry
samples (right)

The results of all tests are summarized Table 3-2 and are graphically represented in
Figure 3-24.
The results clearly indicate a decline of the modulus of elasticity at higher moisture
contents. In the hygroscopic range an indication of a decline in strength is present.
Further compression tests in the hygroscopic range are needed to confirm this
dependency. For the simulation study two different constant values of modulus of
elasticity will be used in the hygroscopic range for the three directions: tangential and
radial Emod=103 MPa and axial Emod=104 MPa.
                                  Emodulus measurement
                                               mass [g]       w [m%]          Emod [Mpa]
Tangential         1a            init          19.0           13.1            1558
                                 cap           29.4           75.0            378
                   1b            init          19.7           13.2            917
                                 wet           29.5           69.5            167
                   1c            init          19.0           8.0             625
                                 dry           17.6           0.0             1576
Axial              2a            init          18.3           11.6            17285
                                 cap           28.5           73.8            8279
                   2b            init          18.9           12.5            12319
                                 wet           27.8           65.5            167
                   2c            init          18.7           7.5             13914
                                 dry           17.4           0.0             19064
Radial             3a            init          19.5           12.1            2164
                                 cap           30.0           72.4            630
                   3b            init          19.6           12.0            1644
                                 wet           29.4           68.0            242
                   3c            init          19.8           8.2             1684
                                 dry           18.3           0.0             1058
Table 3-2: Results of measurements of modulus of elasticity



                                                                                 3-75
                                   4
                               x 10     Modulus of elasticity as a function of moisture content
                          2
                                                                                            Tangential
                         1.8                                                                Axial
                                                                                            Radial
                         1.6

                         1.4

                         1.2
            Emod [MPa]




                          1

                         0.8

                         0.6

                         0.4

                         0.2

                          0
                               0       10      20      30        40       50         60           70     80
                                                       Moisture content [m%]

Figure 3-24: Modulus of elasticity as a function of moisture content




                                                              3-76
4 CASE STUDIES


4.1   INTRODUCTION
In appendix A, 8 different case studies of measurements in Dutch monumental
churches are described extensively. They span the main modes of church heating, i.e.
warm air heating, floor heating and open gas infrared heating. To keep the text
readable, only the results and conclusions from the case studies are discussed below.
The full text is in appendix A. The more general results and conclusions will first be
evaluated.

4.2   HEATING VERSUS NO HEATING
For St. Martinus Church in Weert a simulation study was done for the effect of
heating in comparison to no-heating. The results are presented in Figure 4-1 and
Figure 4-2.
The long-term behavior of calculated indoor temperatures and outdoor temperatures
were compared, as well as the indoor vapor pressure and the outdoor vapor pressure.
In the figures the mean daily values are correlated. The left figures show the
correlation of the mean daily indoor and outdoor temperatures. The indoor climate
lags behind the outdoor climate. During the spring (II) the indoor temperature
remains lower than the outdoor climate. In the autumn (IV), however, the indoor air
temperature is clearly higher than the outdoor temperature.

                                    Indoor Ti vs outdoor Te                                 Indoor Pi vs outdoor Pe
                        35                                                     2500
                                                                                                                      y=x
                                                                                                                      I
                        30                                                                                            II
                                                                                                                      III
                                                                                                                      IV
                        25                                                     2000


                        20


                        15                                                     1500
                                                                     Pi [Pa]
              Ti [oC]




                        10


                         5                                                     1000


                         0


                         -5                                                    500


                        -10


                        -15                                                      0
                              -10     0      10        20     30                      0   500   1000   1500      2000       2500
                                           Te [oC]                                                Pe [Pa]



Figure 4-1: Effects of no heating in St. Martinus Weert


                                                                   4-77
In Figure 4-2 it is clearly to be seen that the indoor church temperature is
thermostrated stationary during the heating season and was kept close to 15 oC. The
relation between the indoor and outdoor vapor pressures is less clear. During the
heating season (I, IV) there is a slight increase of the vapor pressure, due to
desorption of moisture from walls and ceiling.

                                     Indoor Ti vs outdoor Te                                 Indoor Pi vs outdoor Pe
                         35                                                     2500
                                                                                                                       y=x
                                                                                                                       I
                         30                                                                                            II
                                                                                                                       III
                                                                                                                       IV
                         25                                                     2000


                         20


                         15                                                     1500




                                                                      Pi [Pa]
               Ti [oC]




                         10


                          5                                                     1000


                          0


                          -5                                                    500


                         -10


                         -15                                                      0
                               -10     0      10        20     30                      0   500   1000   1500      2000       2500
                                            Te [oC]                                                Pe [Pa]



Figure 4-2: Effects of stationairy heating St. Martinus Weert

4.3   HEATING AND DEWPOINT DIFFERENCE
Due to the lagging effect of the indoor climate on the outdoor climate, surface
condensation and high relative humidities near cold indoor surfaces may occur, e.g.
during spring. An effective way to reduce the condensation risk is heating the church
to a primary temperature level, thus raising the surface temperatures to a higher level.
During heating, however, the absolute humidity of the church also will increase
slightly, due to desorption of moisture at the walls and ceiling. The difference between
the surface temperature and the dewpoint temperature, the so-called dewpoint
difference, is a measure for this condensation risk. For St. Martinus in Weert, a
simulation study was done to show the effects of different primary temperature levels.
The results are presented in the next figures.
Figure 4-3 shows the calculated (mean) indoor wall surface temperatures of the church
when the church would not be heated. Those temperatures were compared to the
situation when the church is floor heated to the actual stationairy air temperature level
of 15 oC. For those situations the dewpoint temperature is calculated too. The effect
of the heating on the dewpoint is that it will slightly increase: the absolute humidity
will increase due to desorption of moisture from walls and ceiling.




                                                                    4-78
                                                            Surface temperatures versus dewpoint
                                    25
                                                                                                       Tdew no-heating
                                                                                                       Tsurf no-heating
                                                                                                       Tdew 15
                                    20                                                                  Tsurf 15




              Tsurf, Tdew [oC]      15




                                    10




                                     5




                                     0




                                     -5




                                    -10
                                          0   1000   2000   3000       4000     5000         6000   7000      8000
                                                                          Hours



Figure 4-3: Comparison of surface temperatures and dewpoint temperatures for no-
heating St. Martinus and heating it to a primary level of 15 oC

Figure 4-4 shows the effects of changing the primary temperature level from no-
heating to a primary temperature level of 5, 10 and 15 oC. The results are presented as
a dewpoint difference. It is clearly to be seen that increasing the primary temperature
level will increase the dewpoint difference too. Condensation risks during spring and
winter season therefore will be decreased effectively.

                                                                    Dewpoint difference
                                    18
                                                                                                            No heating
                                                                                                            Ti = 5
                                    16                                                                      Ti = 10
                                                                                                            Ti = 15

                                    14


                                    12


                                    10
                  Tsurf-Tdew [oC]




                                     8


                                     6


                                     4


                                     2


                                     0


                                     -2
                                          0   1000   2000   3000       4000     5000         6000   7000      8000
                                                                          Hours



Figure 4-4: Effect of a primary heating on the dewpoint difference



                                                                       4-79
4.4            HEATING AND RELATIVE HUMIDITY CHANGES
From the literature study (Schellen 1998/2) it was known that air heating e.g. might
cause severe problems for monumental organs and other monumental objects in the
interior of a church. High air inlet temperatures e.g. cause large thermal stratification
and thus lead to high air temperature at elevated levels, where in most cases
monumental organs are to be found. A high air temperature involves a low relative
humidity. Dramatic low relative humidity values and related drying out and shrinkage
of the organic parts may therefore be the result. Cracks and other indications of
shrinkage in wooden cabinets of the organs and other wooden interior parts
supported this theory.
From the indoor air conditions measured during a year in the Walloon Church a
typical Sunday service (January 17th 2000) was extracted. The figure below indicates air
temperatures and relative humidities in front of the organ at a height of 15 metres, just
beneath the vault.
                             Temperature                                                    Relative humidity
          35                                                               60


          30                                                               50


          25                                                               40
                                                                  RH [%]
 θ [ C]
    o




          20                                                               30


          15                                                               20


          10                                                               10
               00:00 04:00 08:00 12:00 16:00 20:00 24:00                        00:00 04:00 08:00 12:00 16:00 20:00 24:00
                                 time                                                             time

Figure 4-5: Air temperatures (left) and relative humidities (right) measured in front of
the organ in the Walloon Church in Delft at a height of 15 meters


4.5            WOOD SHRINKAGE AND SWELLING DUE TO HEATING
When the ambient relative humidity falls, the equilibrium moisture content (EMC) of
wood (and other organic materials) drops and the wood shrinks with important
resulting deformations. Vice-versa the wood will swell with increasing relative
humidity. For practical purposes, the relationship between deformation and
equilibrium moisture content may be assumed to vary linear (Camuffo 1998).
The results from the wood deformation tests from chapter 3.3.2 were used to predict
the deformation and resulting internal stresses of two wooden organ parts: a wooden
organ pipe and a wind drawer. For the deformation as a function of equilibrium
moisture content a linear relation was assumed, derived from the shrinkage
deformation test of the cubic 50*50*50 mm3 beech sample (see section 3.3.2). The
relationships are graphed in Figure 4-6.




                                                           4-80
                                           Deformation of beech                                                                            Equilibrium deformation of beech
                    2                                                                                                     8
                                                                               tangential                                                                                               tangential
                                                                               radial                                                                                                   radial
                                                                                                                          6
                    0


                                                                                                                          4
                    -2


                                                                                                                          2
 Deformation [%]




                                                                                                        Deformation [%]
                    -4

                                                                                                                          0

                    -6
                                                                                                                          -2


                    -8
                                                                                                                          -4


                   -10
                                                                                                                          -6



                   -12                                                                                                    -8
                         0   2   4       6            8           10     12   14            16                              30   40   50          60             70           80   90                100
                                     Equilibrium Moisture Content [m%]                                                                          Relative Humidity [%]




Figure 4-6: Wood contraction of wood as a function of EMC (left) and as a function
of RH (right)

4.6                          HEATING EFFECTS ON MONUMENTAL ORGANS

4.6.1 Damage from wood deformation
A damage analysis was done to show that the low relative humidity levels in Delft
resulted in shrinkage of the wooden parts of the monumental organ and thus led to
related cracks in the wooden parts. (Sanders et. al. 2000). Section 8.1.8 describes the
results of this damage analysis. The photographs below show the shrinking damage of
a wooden organ pipe (left) and a wind drawer (right).




Some of the wooden organ pipes were cracked in the corner and showed stretching
cracks over the length of the pipe, due to the presence of wooden tuning caps in the
pipe. The caps square wood direction did not strike with the square pipe wood
direction (was assumed perpendicular), and thus blocked the deformation of the
surrounding wooden pipe. Cracks in the wind drawer were the result of the shrinking
in one direction and blockage of it by other wooden parts.
Section 8.1.9 describes simulations on the changing moisture content of both
examples, due to changing relative humidity effects and the related deformation and
stresses. The simulations demonstrated the dramatically humidity effects on the
monumental organ.


                                                                                                 4-81
                                                    x 10
                                                        7              Stresses σ x
                                              4.5


                                               4


                                              3.5


                                               3
              deformation stress σ x [N/m2]




                                              2.5



                                               2


                                              1.5


                                               1


                                              0.5



                                               0
                                                    0       10   20   30              40   50   60   70
                                                                           ∆RH [%]



Figure 4-7: Maximum calculated stresses in the x-direction of a wooden organ pipe,
100*100*5, due to block shape changing of relative humidity ∆RH at the surface (see
section 8.1.9)

The stresses turned out to be in the order of magnitude of maximum allowed stresses
parallel to the fibers (ft;0;rep=14 N/mm2), but exceeded those in the direction
perpendicular to the fibers (ft;90;rep = 0.4 N/mm2) of hard wood. Therefore cracks
in the wooden organ pipes (and also in the wind drawer) parallel to the fibers could be
explained by changes in relative humidity. In the model, however, no relaxation was
involved. In practice the situation might be less critical than upper graph suggests.

4.6.2 Mistuning of organs due to heating
Prof. Leijendeckers (TUE) examined the mistuning of church organs in relation to the
heating of churches. In his calculations he varied the air temperature from –10 to 20
oC. He calculated the lengthening of metal organ pipes, due to thermal deformation

and the effects of changing temperatures on sound velocity in air. The effects of the
lengthening of the pipes on its frequency were negligible: an inaudible change of 0.01
Hz at 15 Hz and a ditto change of 7 Hz at 16 kHz. The temperature effects on the
sound velocity, however, turned out to be the cause of the mistuning: audible changes
of 0,8 Hz at 15 Hz and 876 Hz at 16 kHz were the result of it. The conclusion is that
organs should be tuned at the temperature they are used. Furthermore the organ
should be used at air temperatures, which are reasonably constant. Air temperature
stratification over the height of the organ pipes should be less than 1 K.

4.7   ENERGY CONSUMPTION
WaVo was used for a simulation study to examine the effect of several parameters on
the energy consumption and the heating capacity. In order to compare the churches


                                                                       4-82
with each other the churches have been simulated on the basis of the same standard
input. This standard input was based on the situation in which all churches would be
equipped with the same heating system, e.g. a warm air heating system. Each time one
input parameter was varied in comparison to the standard input. Parameters that have
been varied were: the primary temperature that was maintained in the church
continuously, the comfort temperature that was desired during the service, the
ventilation rate and the heating rate of the church. The influence of additional
protective glazing and heat insulation of the vaults on the energy consumption of the
building was examined. The full results were reported in (Neilen 2002).

Basic input
The churches were simulated on the basis of the same standard input. The standard
values of the input parameters are represented in the second column of Table 4-1:
Input parameters. The third column represents the new values of the parameters that
have been varied in the simulations. For each simulation only one input parameter was
varied in comparison to the standard input.

Parameter                   standard value                       variable value
θprimary[°C]            8                            0                  6         10
θcomfort [°C]           18
Heating rate [K/h]      100 (no restriction)         1                  2
Ventilation rate        measured value               0.2                0.9
[1/h]
Uglass [W/m2K]                                       5.2               3.5
Uvault [W/m2K]          from drawing                 1/(Rl+1)
Church usage            once a week on
                        Sunday from 12-14h
Table 4-1: Input parameters


Church                        Ventilation      Uvault      Heating capacity   Energy
                              rate             [W/m2K]     [kW]               consumption
                              [1/h]                                           [*104 kWh]
Grote (St. Laurens) Kerk,     0.5              3.3         900                33.7
Alkmaar
H. Donatuskerk, Bemmel         0.6             2.4         270                10.6
N.H. Kerk, Beusichem           0.6             2.5         120                4.6
Waalse Kerk, Delft             0.2             3.5         75                 2.1
H. Gerlachuskerk, Houthem 0.2                  2.0         140                3.9
Sypekerk, Loosdrecht           0.9             2.5         100                4.2
H. Liduïnakerk, Schiedam       0.5             2.3         280                10.1
St. Martinuskerk, Weert        0.1             1.2         340                9.9
Table 4-2: Simulation results standard input




                                            4-83
Because the influence of additional glazing and heat insulation of the vaults had to be
examined, it was assumed that these were not present in the churches in the standard
input.
In order to do not change the characteristic features of the building, the ventilation
rate measured in the churches was used for the standard input.
The simulation results of the standard input are represented in Table 4-2. The results
of the alternative simulations were compared to these values.




Figure 4-8: Influence of lowering the primary temperature

Figure 4-8 represents the influence of lowering the primary temperature in the
churches on the heating capacity and the energy consumption. The figure shows that
lowering the primary temperature in most of the examined churches results in an
energy saving of 30 to 50%. The explanation is obvious: less energy is needed to
maintain a lower primary temperature. On the other hand more energy is needed to
heat the air in the church from the primary temperature to the comfort temperature,
which is desired during the services. But the total results in an energy saving.
Lowering the primary temperature, however, results in an increase of heating capacity,
if the air in the church has to be heated from primary to comfort temperature in the
same period of time as before. Another possibility is to maintain the available heating
capacity and start heating earlier. For a particular church an optimizing study can be
done.




                                          4-84
4.8     EVALUATION OF HEATING SYSTEMS

4.8.1    Warm air heating

Introduction
The early types of warm air heating systems used gravity circulation by natural
buoyancy. Nowadays warm air heating with forced-fan assisted circulation is standard
and is the most common heating system in Dutch churches (Kimmenade 2000).
Forced air systems use an air heat exchanger and a fan. In direct air heating systems
the gas (or oil) burner directly heats up the air in a heat exchanger system, while the
indirect systems use hot water air heating coils, supplied from a boiler plant. In the
Netherlands the system commonly used consists of one or two air inlet grilles in a wall
or floor near the heater system and one or more air return grilles in the floor next to
the inlet system. In more sophisticated systems the air is distributed throughout the
church by a duct heating system. In this way the air is carefully diffused throughout
the church to ensure a fast warm-up, good temperature distribution and avoidance of
draughts and stratification.

Problem definition
If air is heated in an air heating system without moisture control, the relative air
humidity will decrease. During winter conditions, like frost, the absolute air humidity
can be very low in a church and the heated air relative humidity may decrease to
values that are dangerous for monumental organs and other wooden interior parts of
great value. Furthermore, if this air is introduced into the church above a certain air
inlet temperature level, stratification is almost inevitable unless air distribution is very
carefully considered (Bordass 1996). As a consequence the air temperature at a certain
height can be several degrees higher than the set-up temperature at comfort level for
the church visitors and the relative humidity drops to more endangering values.

Objectives
The objective was to detect differences in air heating and to explain their physical
behavior. Especially the thermal stratification and its relation to the air inlet conditions
were of particular interest. The effects of the distribution of air inlets and air outlets
were part of the investigation. Furthermore the (rapid) changes of relative humidity
due to this kind of heating and their effects on monumental interior parts were
considered.

Case studies
In the Waalse Kerk Delft (Walloon Church in Delft) the monumental organ had to be
restored due to severe relative humidity conditions. The indoor air conditions were
measured and recorded throughout a heating season before the restoration of the
organ and before the heating system was adapted. The adaptation recommendations
were derived from this study. Afterwards the indoor air conditions were measured
throughout a new winter heating season.



                                            4-85
In the Sype Kerk (Sype Church) in Loosdrecht, a heating system firm improved the air
heating sytem, consisting of two inlet grilles and one outlet grille, and this system was
evaluated by indoor air measurements.
In Schiedam St. Liduïna’s basilica got a new air heating system, with one central air
inlet grille and a duct system in the floor with several outlet grilles near the wall. For a
year the heating system was evaluated and a simulation study took place in order to
improve indoor air conditions.

Conclusions
The original warm air heating system in the Walloon Church in Delft led to a large
thermal stratification (up to 15K over the height of the church). Very high air
temperatures (up to 30 to 35 oC) were measured at the height of the monumental
organ. These were the result of a combination of too high air inlet temperatures and
too low air inlet velocities. Dramatically low relative humidities near the monumental
organ resulted from these conditions. The organ was in such a bad condition, due to
excessive shrinkage of the wooden parts, that restoration was necessary.
An improved air heating system was proposed. The inlet conditions were based on a
combination of low inlet air temperatures (up to 40 oC) and high air velocities (up to 4
m/s), leading to an Archimedes number less than 0.05. Afterwards the air
stratification proved to be less than 2 K over the total height of the church. Initially
the air heating system was switched off at a lower limit of the relative humidity near
the organ of 40 %RH. The result of this was that the church could not be used at
outdoor temperatures below about 0 oC. Therefore an air humidifying system was
proposed, simulated and implemented. Condensation on cold surfaces like glass
surfaces was predicted.
In the Sype Kerk in Loosdrecht the modified air heating system nearly fulfilled its
requirements. Modification of the air inlet grilles apparently led to high air induction at
these grilles and air stratification measurements indicated that the air temperature
differences over the total height of the church were less than 2 K. The Archimedes
number was in the order of magnitude of 0.05.
The heating rate was controlled to a value of less than 2 K/h. A primary temperature
of 10 oC was established. Furthermore the heating system was controlled by a
hygrostatic device, which limited relative humidity to an upper limit of 80 %RH.
However, there was no lower limit control. Therefore the relative humidity within the
church reached lower values of less than 45 %RH for about 10 % of the measuring
period of half a year. Lowest values measured were around 35 %RH. Here it was
advised to lower air temperatures in the church during low outdoor temperatures
from 20 oC to 18 oC and to limit lowest relative humidities by an hygrostatic device to
40 %RH.
The air heating system in the St. Liduïna Basilica did not perform well. The air
temperatures were controlled to 20 oC at living height, but the air stratification turned
out to be larger than 5 K over the height of the church. Measurement of the air inlet
conditions showed during low rate running a combination of too high inlet air
temperatures (up to 70 oC) and low velocities (up to 2 m/s), resulting in Archimedes
numbers of up to 0.4. A higher velocity rate improved the inlet air conditions to



                                           4-86
Archimedes numbers down to 0.07, but resulted in too high air velocity values (up to
30 cm/s) in the comfort zone.
A CFD simulation study showed that it was difficult to prevent air stratification in this
church using a single central hot air inlet: the geometry of the church was complex
and due to the large volume and heat loss area a rather high heating capacity was
needed. Furthermore it was concluded that the large number of air outlet grilles was a
waste; a single outlet had probably led to the same result. A well-distributed air inlet
system, combined with one outlet would improve the temperature and humidity
climate considerably: temperature stratification in this variant was expected to be less
than 2 K over the height of the church. Interchanging the air inlet and outlet systems
would probably improve the system to a large extend.

4.8.2   Floor heating

Introduction
Floor heating became most popular as a main heating system in well-insulated
dwellings in the late seventies of the twentieth century. The improved insulation of
building constructions allowed the application of floor heating without additional
installation of supplementary heating devices. Furthermore, the invisible heating
system was practical and esthetically sound. The degree of thermal comfort was high:
a high level of mean radiant temperature, a high thermal (feet) contact temperature,
even with stone floors, and a small degree of thermal stratification. Due to this
popularity floor heating has also been introduced successfully in modern (church)
buildings, with the same degree of insulation and limited heights. As a result of their
success, floor-heating systems have also been applied for the heating of monumental
churches.

Problem definition
Floor heating systems in modern buildings, with limited heights and well insulated
walls, floors and vaults are mostly very effective. Usually their capacity is sufficient
and only in a limited number of cases some kind of additional heating system, like
radiators or convectors, is needed to give a very high level of thermal comfort. In
monumental churches the circumstances, however, essentially differ from those in
these modern buildings. In general the walls, vault and floors are not insulated and the
heights of the rooms are more often considerably greater. Furthermore the area of the
floor is relatively small as compared to the heating loss area of walls and vault.
Therefore often the heating capacity does not meet the heating capacity required. As a
result the floor temperature is usually increased trying to reach a reasonable degree of
thermal comfort. This can often lead to excessive thermally driven airflows,
accelerated by the cooling of air near cold surfaces like ceilings, walls and stained glass
windows with great heights. A reverse effect often has been reached: due to the
increased air velocities the total degree of thermal comfort is lowered and complaints
on thermal draught are the most common result of it. Furthermore the monumental
floors often consist of very thick stone slabs like gravestones. The heating pipes of the
floor heating system are placed underneath on a thermal insulation layer. The result of



                                           4-87
it is a very inert thermal system, with heating up time constants of several hours to
days.

Case studies
The floor heating system in St. Martinus’ church in Weert was evaluated during two
years of measurements. Contamination of the vault paintings was one of the
investigated research items.
St. Gerlachus’ church in Houthem has beautiful frescoes on wall and ceiling. After a
major restoration and renovation and the introduction of a floor- and pew heating
system the ceiling soon became contaminated. Measurements were made in order to
determine the origin of the problem.
In the Hervormde Kerk (Dutch Reformed Church) in Beusichem floor heating was
combined with pew heating and radiators. The achieved thermal comfort was not
acceptable and measurements as well as simulations indicated it to be very difficult to
reach thermal comfort with the installed heating systems.

Conclusions
At the start of the research in St. Martinus’ in Weert the control of the floor heating
system was improved and the control of the air temperature during the heating season
was adjusted from 20 oC to 15 oC. At the same time protective glazing with external
ventilation was installed from the outside of the stained glass windows.
A study of the near surface relative humidity of the ceiling showed no critical surface
humidities. Furthermore surface samples of the pollution on the ceiling showed no
traces of fungi or algae.
Measurements proved a direct relation between floor temperatures, air temperatures
reached and resulting airflows. The contamination of the ceiling was clearly a result of
thermally induced vertical airflows and pollution due to soot production by candle and
incense burning. Due to the impuls of the vertical airflow and the temperature
difference between air and surface of the ceiling inertial and thermophoretic
deposition were expected to be the deposition mechanisms (Camuffo 1995). A soot
production analysis of the candles and incense used showed wax lights to produce
only a fraction of soot as compared with candle burning. Therefore reduction of the
airflows, improvement of the ceiling surface temperature, substitution of the candles
by wax lights and increase of the ventilation rate would reduce the soot contamination
to a fraction of the original.
Additionally the implemented measures of temperature reduction and the installation
of protective glazing led to energy loss reductions of up to 35 % as compared to
earlier years.
The air conditions near the monumental organ most of the time (more than 95% of
time) satisfied the recommendations from literature (Jütte 1994).
The results of the WaVo simulations were of the same magnitude (within 5 to 10%) as
the results of the measurements.
The contamination of the ceiling in St. Gerlachus’ resulted from thermally driven
airflows and the transported dirt and soot. The airflows could effectively be reduced
by lowering the temperatures of the pew convectors, e.g. from 40 to 30 oC.



                                          4-88
The difference in contamination between the original and the, after a fire, renewed
part of the ceiling of St. Gerlachus’ in Houthem clearly seems to be the result of the
differences in indoor surface temperatures of the ceiling. Infrared thermal imaging of
the surface temperatures led to the same pattern in surface temperatures as the visual
appearance of the contamination. Even the visual pattern of wooden construction
beams in the contamination is the same as in surface temperatures. Obviously lower
surface temperatures result in more severe contamination. The deposition of particles
from a laminar, natural convection boundary layer flow adjacent to a cool surface
occurs due to a combination of thermophoretic drift and Brownian motion. The
temperature difference between air and surface is responsible for the resulting force of
dirt particles towards the cold surface; this phenomenon is known as thermophoresis.
The solution to the contamination problem therefore is reducing this temperature
difference by increasing and equalizing the surface temperatures.
The origin of the lower surface temperatures proved to be the thermal resistance of
the vault construction. Improving the thermal resistance of the vault or roof
construction will therefore reduce the contamination.
Apart from that it was recommended to take the offering of wax lights outside the
church to reduce the source term of the contamination.

4.8.3   Infrared gas heating

Introduction
In a number of case studies the building physical behavior of gas-infrared heating
systems was the subject of investigation. In 1996 the board of the Grote Kerk in
Dordrecht considered to install an open air gas infrared heating system in their church
as a supplemental local heating system in addition to the local floor heating system of
the west transept of the church. At that time only the intended layout and capacity of
the infrared gas radiant heaters were known. A simulation study and some exploratory
measurements in the church showed the dramatic effect of the large moisture and
CO2 production to be expected. Furthermore the simulated radiation heating effect on
the monumental wooden pulpit resulted in great concern. To verify the numerical
simulations two further case studies were made. In a catholic church in Bemmel an
infrared gas heating system caused great condensation problems on the stained glass
windows, the granite of the altar, the stone benches in the niches of the choir as well
as an unacceptable rise in temperature of irradiated paintings on the wall. In the Grote
or Lawrence church in Alkmaar an experimental set-up of an infrared gas heating
system has been tested on moisture and radiation effects.


Problem definition
The following building physical related problems may be identified with gas infrared
heating:
Surface condensation due to low surface temperatures on walls and (stained) and
protective glazing in combination with a high (absolute) air humidity, due to excessive




                                         4-89
moisture production by open air infrared gas heating; in this respect shadowed
surfaces are most critical.
Problems related to the drying out of wooden constructions by direct radiant heating;
drying out of the monumental wooden furniture, like organs, altars and other organic
materials and the related shrinkage and damage to these materials.
Problems related to the drying out of stone constructions by direct radiant heating,
e.g. damage to the plaster, damage to wall and mural paintings; salt migration caused
by drying effects etc.
Thermal comfort problems due to relatively large airflows, which are generated by
local, heated surfaces and are accelerated due to the cooling down of the air caused by
the lower surface temperatures of walls and glazing.

Objectives
The objectives of this study were the identification and description of the most
important ‘open air’ gas infrared heating systems and the determination of their
building physical impact on the buildings and their interiors. In three case studies these
heating systems were evaluated, both experimentally as well as by modeling and
simulation.

Case studies
In this section the results of the three case studies on ‘open air’ gas infrared heating
systems will be presented.
In the Grote or Onze Lieve Vrouwekerk in Dordrecht prior to this study an
installation dealer designed an infrared gas heating system. The literature study
(Schellen 1998/2) indicated problems to be expected with this kind of open air gas
burning system. Therefore a simulation study was performed on the moisture
conditions to be expected in this church and a prediction was made, based on
simulation, of the radiant effects on the monumental interior parts.
The Rooms Katholieke Kerk (Roman Catholic Church) in Bemmel proved to have a
lot of moisture problems due to an open air gas infrared heating system. A
measurement and simulation study was performed to determine the cause of the
problems and to fit theory and practice.
The Grote of Laurenskerk (Grote or Lawrence church) in Alkmaar hired an open gas
infrared heating system. Cultural heritage advisors expected moisture and radiation
problems for the building as well as the monumental interior. An extensive
measurement setup was realized by the Netherlands Energy Research Foundation
(ECN), together with The Netherlands Institute for Cultural Heritage (ICN) and The
University of Technology Eindhoven (TUE).

Conclusions
Gas infrared radiation heating systems without exhaust air extraction proved to be
harmful heating systems for monumental buildings. The gas burning process leads to
serious moisture and carbon dioxide production. For every 100 kW heating capacity
installed, about 17 liters of water in vapor condition is released into the air and
brought into the church.



                                           4-90
In the Grote Kerk in Dordrecht an installed heating capacity of 375 kW was planned.
The infiltration rate of the church proved to be small (0.08 h-1) and the calculated dew
point would therefore be increased by several degrees centigrade. During heating no
condensation would occur on the direct radiated walls but rather high relative
humidities would occur near surfaces in the shadow of the heaters. Due to the rather
low infiltration rate directly after termination of the heating absolute humidity would
remain nearly the same in the church interior. The air temperatures and surface
temperatures however would decrease in short time and high near surface humidities
and even condensation would occur most often.
The idea to ventilate the church via the attic proved to be very dangerous. Frequently
occurring condensation on the inner surface of the roof would be the result of it.
A simulation study of the moisture content of the directly radiated pulpit showed the
water content to decrease dramatically in a short time. Surface temperatures of 40 oC
were reached at the soundboard of the monumental pulpit. Near surface relative
humidities of about 20 %RH were the result of it. When we compared these rather
low surface relative humidities with shrinkage diagrams of tangential oak wood, we
expected wood contraction percentages of up to 5 % to be reached.
Furthermore the relatively quick changes of surface temperatures and accompanying
relative surface humidities did not meet the standards for museum properties (52 ± 3
%RH, θmin=2 oC, θmax=25 oC, maximum change in twenty-four hours 3 K, (Jütte
1994)). In principle, however, a church is not a museum; therefore these requirements
were adapted for churches (Jütte 2000). The heating system did not reach these
requirements either.
Measurements in Bemmel proved the calculations for Dordrecht to be fail-safe: a two-
hour operating period of the gas infrared heating devices led to most serious surface
condensation on stained glass windows, walls and floor. In this church thermal images
were translated into hygro-thermal images: ‘hygrograms’. These images showed the
most critical condensation surfaces to be lying in the shadow of the infrared heating
devices. It was clear from these images that the most critical time for surface
condensation on cold surfaces is just after stopping heating: the temperature decreases
in a short time and if the infiltration rate is low the absolute humidity is nearly
maximum. In the church of Bemmel the combination of a large moisture production
source and a relatively minor infiltration rate again were a serious threat to the
monumental building and its interior.
Very extensive measurements during two winter periods in the Grote or Laurenskerk
in Alkmaar led to the following conclusions. The infiltration rate proved to be high:
values of 0.4 to 1.2 h-1 were measured. The existing floor heating system prevented
surface temperatures to drop to too low values. As a result of this excessive
ventilation and heated surfaces by floor heating, condensation on cold surfaces, other
than glazing, was not detected. Wooden beams in cold walls, glass windows and cold
walls proved to be critical places. High surface temperatures of pillars (up to 35 oC),
pulpit top (up to 25 oC) and pews (up to 35 oC) due to directly heating by radiation
were recorded. It was concluded that the gas heating system should be modulated and
controlled. The moisture content of wooden beams, surface temperatures of directly



                                         4-91
heated monumental elements and wall surface condensation should be monitored and
used to control the system during the utilization of the gas infrared system.




                                      4-92
5 CLOSURE


5.1   INTRODUCTION
In this chapter the outcome of the research will be applied to facilitate the design
process and give guidelines for the selection and design of heating systems. First the
recommendations from literature to prevent heating system related damage will be
updated, substituted and completed according to the results of this study. This will
result in demands and requirements for the different heating systems. The outcome of
literature and this study indicate several advantages and disadvantages of the different
heating systems that will be summarized to facilitate the choice for a heating system. A
checklist will be presented for the choice and design of a proper heating system.
Results, conclusions, discussion and recommendations for further research will end
this final chapter.

5.2   AN UPGRADED PERFORMANCE ARRAY FOR CHURCH HEATING

5.2.1 Introduction
When we look at Table 1-1 from chapter 1 with recommendations from literature for
the preservation of monumental churches and interior, we can summarize these as
follows.
Property                    Symbol       Unity Lower value         Upper value
Indoor air temperature      θi           °C                        12..19
Primary temperature         θprimary     °C       5                8..10
Relative humidity           RH           %         45..50          60..75
Daily change RH             ∆RHday       %                         10
Yearly change RH            ∆RHyear      %                         30
Heating rate                ∆θ/∆t        K/h                       1..2
Indoor air velocity         u            m/s                       0.1..0.3
Supply air temperature      θsupply      °C                        θi + 25
                                                                   45
Supply air velocity         uinlet       m/s                       2
Floor surface temperature θfloor         °C                        25..28
Table 5-1: Summarized recommendations from international literature for the
preservation of monumental churches and interior

For the different heating systems, studied in this work, with the present understanding
this table can be substituted and completed.




                                         5-93
5.2.2 Indoor air temperature
The values in Table 5-1 for the acceptable indoor air temperature are based on
different assumptions: the lower values are based on assumptions on thermal comfort
and the upper values are based on allowable low relative humidities to be reached.
Thermal comfort depends on conditions related to the visitor (clothing and
metabolism) and indoor conditions (air temperature, mean radiant temperature,
temperature asymmetry, air humidity, air velocity and air turbulence). In monumental
churches optimal thermal comfort is rare: due to great heights of walls and glazing and
relative low surface temperatures turbulent airflows in churches cannot be prevented.
Winter clothing or thermal radiation (by floor or other higher temperature sources)
may compensate for a lack of thermal comfort due to lower air temperatures. An
upper limit of 15 oC therefore is used for Roman Catholic churches (Knoll 1971) and
(Mainz 1972) and is based on the wearing of winter coats during services. Other
thermal comfort predictions may be based on models for thermal comfort calculations
(Loomans 1998).
After the severe winters of the 1960-ties a lot of problems with monumental organs
were observed. The upper value of 12 °C therefore originates from these extreme
winter conditions, i.e. longer frost periods. Nowadays, however, it is technically
possible (and relatively easy) to limit the allowable lowest relative humidity due to
heating by a hygrostatic device. In this (hygric) respect it is better to have limitations in
the table on lower relative humidities, guarded in the church by a hygrostatic device.

5.2.3 Primary temperature
The concept of the primary temperature was introduced for some reasons: to prevent
surface temperatures to drop below dew point temperature, to accelerate heating up
times and to improve thermal comfort due to higher surface temperatures.
Low surface temperatures may lead to high surface relative humidities, as was shown
in paragraph 2.3.2. When the surface temperatures are about 10 oC, the difference
between surface temperature and dew point temperature is about 4.2 K for surface
relative humidities of 75 % and 1.6 K at RHsurf = 90 %. To prevent long term higher
surface humidities of above 75 % RH it is therefore recommended to keep the dew
point difference larger than 3 to 4 K. Long time (one year) measurements in a
particular church thus may give an indication of the absolute humidity and dew point
in that church. For the churches in the case studies a primary temperature based on
this criterion has been calculated and is given in Table 5-3.
Measurements in infrared gas heated churches (Koene, 2000), (Heimeriks, 1989)7
showed that one essential condition to prevent surface condensation problems with
this kind of heating is to heat up the surface temperatures by some kind of primary
heating.
It is therefore recommended to keep the primary temperature above about 8 to 10 oC.
To prevent church interior wooden parts to be exposed to sudden changes in relative
humidity it is recommended to limit heating rates to 1 or 2 K/h. Heating up times

7   Heimeriks did not mention it in the summary of his report explicitly, but the Eusebius church in Arnhem
     had floor heating and thus the surface temperatures were not as low as otherwise they would have been



                                                    5-94
therefore will be very long, when the difference between comfort and primary
temperatures is too large.
Thermal comfort also depends on mean radiant and thermal asymmetry temperatures.
If the indoor air temperature is maintained at the primary temperature, indoor wall
surfaces will approximately be maintained at that level too. This will have a positive
effect on mean radiant and thermal asymmetry temperatures.
The maintenance of a primary temperature involves the use of energy. Figure 4-8
summarizes the yearly extra energy to maintain a primary temperature level.

5.2.4 Relative humidity
The indoor air relative humidity conditions for monumental churches should be
related to the most critical and valuable interior parts. Most often these are
monumental organs and/or wooden interior parts like monumental pulpits, altars and
pews, which are objects of inestimable value. Where these objects are exposed to
indoor conditions, relative humidity should be limited. A lower value of 40 to 45 %
RH is critical when it comes to shrinkage problems. A temporary upper limit of 70 to
75 is critical for fungi attack.
To reduce the hygrothermal load of materials and construction it is suggested to allow
larger fluctuations during a long period and smaller for a short period. From practice
Künzel (Künzel 1991) suggested 10 % RH fluctuations during a day and 30 % RH
fluctuations during a period of a year, between about 50 and 80 % RH.

5.2.5 Heating rate
Of these objects the monumental organs seem to be most critical for changing indoor
conditions: they consist of very fragile to more robust wooden and other organic
parts, which react from very fast (small and fragile parts) to very slowly (wooden
construction parts) to changing indoor conditions. To protect these objects from
internal stress, changing conditions should lead to equally changing conditions in both
types of objects. Therefore the largest time constant determines the rate of changing
conditions. Furthermore during instationary heating airflows will be generated, which
may lead to contamination of surfaces. A heating (and cooling) rate of 1 to 2 K/h
proved to be a safe indoor temperature changing rate (Knoll 1971).

5.2.6 Thermal stratification
In case of warm air heating systems thermal stratification should be limited to 1 to 2
K for the total height of the church. This will protect the monumental organs against
too high temperatures; a thermostatic device on occupation height mostly controls
indoor air temperature. Furthermore the temperature difference over the length of an
organ pipe should not exceed 1 K. For most churches the thermal stratification
should therefore be limited to 0.1 to 0.2 K/m, measured over the height of the
church.




                                         5-95
5.2.7 Indoor air velocity and turbulence
For thermal comfort reasons the indoor air velocity should be limited to 0.1 to 0.15
m/s. The maximum turbulence intensity level at occupation level then can be
calculated from the indoor air temperature (Fanger 1988).

5.2.8 Number of air inlet and extraction grilles
Where the behavior of the airflow in a room is considered, this is fully determined by
the situation and number of the air inlet grilles and is hardly influenced by the number
and place of the air extraction grilles. Figure 5-1 shows the principle dimensionless
velocity fraction at the radius of an air inlet grille and an air extraction grille (Fitzner
1996). An air inlet opening generates an air jet with an air velocity at the center of the
jet, which remains nearly the same at a distance of about 5 times the diameter of the
opening. On the other hand the air from an air extraction aperture loses its velocity
completely (about 4 % remains) at the same distance from the grille. That is the
reason why a candle cannot be blown out by suction.
                              1
                                                                              ux
                    u x/u 0                        D

                                                                          x



                          0.5
                                                             Inlet




                                      Extraction

                              0
                                  0        5           x/D           10


Figure 5-1: dimensionless air velocity at the centre of an air inlet and air extraction
grille

Therefore the number of inlet grilles, not extraction grilles, is essential for the airflow
in a church. As a rule of thumb (Fitzner 1986) gives an indication of the number of
grilles for mixed air handling. Horizontal and vertical sections of a room can be
divided into square sections (Figure 5-2). Every square section then needs an air inlet
grille. The number of squares thus determines the minimum necessary number of inlet
grilles. The number of extraction grilles can be determined to be about one fifth of the
number of inlet grilles. Figure 5-2 is taken from (Fitzner 1996) and shows an example.




                                               5-96
                         Example: Floor area 10.5m x 14 m




                                                                                     8
height




                                                                                5
                                                                          3.5
                                                                    2.7
                                        14



Figure 5-2: rule of thumb for the determination of the number of air inlet grilles in a
room


Room height           Number of squares       Number of air inlet Number of air
                                              grilles               extraction grilles
2.7                   5x4                     20                    4
3.5                   4x3                     12                    2
5                     2x3                     6                     1-2
8                     1x2                     2                     1
Table 5-2: example taken from (Fitzner 1996) to demonstrate the calculation of
number of air inlet and extraction grilles in accordance to Figure 5-2

5.2.9 Supply air temperature
In the work of (Knol 1971) and (Mainz 1972) the supply air temperature for warm air
heating systems is recommended to be limited to a temperature difference between
supply and indoor air of 25 K or a temperature of 45 oC. It is not mentioned explicitly,
but it is expected that this recommendation resulted from limitations on thermal
stratification. In this respect it seems to be better to limit the Archimedes Number Ar
to a maximum number, because thermal stratification not only depends on the
temperature difference between supply and indoor air, but is also determined by the
air supply velocity. If the Archimedes Number Ar is limited to approximately 0.05 it is
possible to limit the thermal stratification to about 0.1 K/m. Furthermore the supply
air should not directly reach the monumental organ. A limitation on the length of
throw should be given, much smaller than the distance between inlet air supply and
object.

5.2.10 Supply air velocity
The maximum of supply air velocity depends on the length of throw and the
minimum depends on the maximum Archimedes number. The length of throw should


                                                       5-97
be limited to a maximum of about 2/3 times the length to an air reachable object of
art, like a monumental organ, mostly being at the back end of the church.

5.2.11 Floor surface temperature
When the church is heated by floor heating, the floor surface temperature maximum
allowed could be based on two assumptions: generated airflows and thermal feet
comfort.
The difference between floor surface temperature and air temperature leads to
considerable airflows. This may lead to thermal discomfort and contamination by soot
and dust. In this respect it is difficult to control heating on airflows or air velocities. A
control based on the above mentioned temperature differences, or a maximum air
temperature allowed under winter conditions should be considered. For thermal
comfort reasons near the feet (to prevent swollen and sweaty feet) the upper floor
surface temperatures should be limited to a maximum of 29 oC (Loomans 1998). The
allowable lower floor surface temperatures depend on the thermal feet contact
temperature and may be improved by the contact floor material. For stone floors the
lower floor temperature is about 24, for wooden floors this temperature limit is about
16 oC.

5.2.12 Relative humidity near surfaces
In principle relative humidity near surfaces should not exceed the limits, which are
mentioned for indoor air relative humidity. Due to the lower surface temperature of
cold walls and glazing the relative humidity near these surfaces increases. For short
periods, e.g. less than an hour, higher values up to 90 %RH may be accepted.




                                            5-98
5.3    UPGRADED PERFORMANCE ARRAYS FOR DIFFERENT HEATING
       SYSTEMS

                                 Warm air heating
Property                      Symbol    Unity Lower value          Upper value
Indoor air comfort            θi        °C        15               20
temperature
Primary temperature           θprimary     °C     5                10
Relative humidity mean        RHmean       %      45               75
Yearly change RH              ∆RHyear      %                       30
Relative humidity short       RHshort      %      40*              90
term
Daily change RH               ∆RHday       %                       10
Heating rate                  ∆θ/∆t        K/h                     2
Indoor air velocity comfort   u            m/s                     0.15
area
Temperature stratification    ∆θ/∆h        K/m                     0.1
                              ∆θmax        K                       2**
Supply air temperature        θsupply      °C                      θi + 25
Length of throw               lmax         m                       2/3 lobject
Supply air velocity           usupply      m/s    Ar < 0.05        From lmax
Number of air inlet grilles   nin                 From Figure
                                                  5-2
Number of air extraction   nout                   nin/5
grilles
Floor surface temperature  θfloor         °C                       25..28
Table 5-3: Summarized recommendations from this research for the preservation of
monumental churches and their interior in case of warm air heating

Notes
* Limited by a hygrostatic device
** Over height of church




                                         5-99
                                   Floor heating
Property                      Symbol     Unity Lower value               Upper value
Indoor air comfort            θi         °C                              15***
temperature
Primary temperature           θprimary      °C        5                  10
Relative humidity mean        RHmean        %         45                 75
Yearly change RH              ∆RHyear       %                            30
Relative humidity short       RHshort       %         40*                90
term
Daily change RH               ∆RHday        %                            10
Indoor air velocity           u             m/s                          0.15
comfort area
Floor surface temperature     θfloor        °C                           25..28**
Table 5-4: Summarized recommendations from this research for the preservation of
monumental churches and their interior in case of floor heating

Notes
* Limited by a hygrostatic device
** Limited to avoid too large temperature driven air flows
*** Comfort is reached by a combination of radiant and air temperature



                               (Infrared) Radiant heating
Property                       Symbol       Unity Lower value           Upper value
Indoor air comfort             θi           °C       10                 15
temperature
Primary temperature         θprimary      °C         8**                10
Relative humidity mean      RHmean        %          45                 75
Yearly change RH            ∆RHyear       %                             30
Relative humidity short     RHshort       %          40*                90
term
Daily change RH             ∆RHday        %                             10
Heating rate surfaces       ∆θ/∆t         K/h                           2
Indoor air velocity comfort u             m/s                           0.15
area
Surface temperatures        θsurf         °C                            25..30
Table 5-5: Summarized recommendations from this research for the preservation of
monumental churches and their interior in case of (infrared) radiant heating

Notes
* Limited by a hygrostatic device
** Limited by floor heating or air heating or otherwise, but not by infrared gas heating




                                         5-100
5.4     QUALIFICATION OF HEATING SYSTEMS
To choose an appropriate heating system for a specific monumental church it is
necessary to know the advantages and drawbacks of particular systems. The following
section gives an overview of heating systems and their positive or negative
characteristics. These are summarized in Table 5-6.

5.4.1    Warm air heating

General
       Advantages
       Energy consumption: Air heating systems are fast heating up systems. In this
       way their energy use may be very efficient.
       Control: The control of air heating systems is fast responding.
       Preservation: Air handling is possible by means of moisture suppletion or
       extraction. Proper air filtration can avoid bringing dust into recirculation.
       Ventilation: a combination with ventilation is relatively easy.
       General: no space consumption in room.
       Drawbacks
       Energy consumption: In principle the whole church is heated. This is not the
       most energy efficient system.
       Thermal comfort: Without primary heating surface temperatures are low. A
       primary temperature setting may compensate for this lack of radiant
       temperature comfort. A wooden floor beneath the pews can prevent low feet
       temperatures.
       Preservation: Air heating systems are fast heating up systems. They have to be
       tempered to avoid too fast heating up the air and consequently lowering the
       relative humidity.

Well-distributed air-inlet systems
        Advantages
        Well-distributed air-inlet systems throughout the church lead to a
        homogeneous air temperature distribution. A combination of low
        temperature and low velocity air inlets prevent too large temperature
        stratification and draught. Heat can be transported by air ducts in the floor or
        by hot water tubes to wall or floor ventilator-convector devices.
        Drawbacks
        Thermal comfort: lower air temperature experienced due to air velocity.
        Space consumption: Air duct systems in a church floor for distributed air-
        inlets need large cross section areas and large lengths. To prevent excessive
        heat losses to the underground the air ducts have to be insulated. These are
        expensive duct systems.
        Construction work: Furthermore the connection to the air heating plant may
        cause considerable construction work throughout the church.
        Valuable cultural underground archives may be destroyed.


                                         5-101
        Ventilator convector devices in the floor are an alternative. These devices are
        subject to dust. Filtering of the air is of great importance. Air inlet systems
        near people may lead to draught. A noise treatment of the ventilator device is
        of importance.

Non-distributed air-inlet systems
      These systems consist of one, two or several air inlet systems in the wall(s) or
      floor, depending on the geometry and dimensions of the church.
      Advantages
      The grilles are mostly arranged in such a way that air duct distances to the
      heating plant are minimised. Most often wall grills are mounted in the wall
      between plant and church room. If designed well a very uniform air
      temperature distribution may be reached. A low Archimedes number can
      achieve low thermal stratification.
      Drawbacks
      Depending on the Archimedes number the wall inlets may cause large
      stratification due to too high air temperatures or too low air velocities. Non-
      uniform air temperature and velocity profiles may result from it. The air inlet
      systems may cause draught due to high turbulence and air velocities near
      people. High velocity air inlet grills in floors may lead to local discomfort.
      High air velocities and turbulence may cause dust to be brought into
      circulation.
      The airflow direction and length of throw should be designed to avoid
      directly blowing warm air onto monumental furniture.
      Attention: Depending on the noise treatment the air inlet system and ducts
      may cause noise.

5.4.2   Floor heating
        Advantages
        The heating system is not noticeable in the church room.
        Thermal comfort: Floor heating leads to comfortable feet contact
        temperatures. It increases mean radiant temperatures and therefore the
        reachable comfort level.
        Preservation: Due to radiant fluxes to walls and ceiling it increases these
        surface temperatures and reduces condensation and high relative humidity
        values near these surfaces.
        Drawbacks
        Energy consumption: Time constants of the floor heating system may vary
        from hours to days, depending on the thickness and material of the covering
        floor deck. A stationary heating operation may be the only possible choice.
        Prohibitive energy consumption may be the result of it.
        Thermal comfort: The heating system may cause considerable airflows,
        depending on the difference of floor and air temperatures and the lay out area
        of the floor heating system. This may lead to discomfort due to draught. This
        is especially true for partially heated floors.



                                        5-102
        Preservation: the airflows generated will cause dust and dirt deposition on
        ceiling and wall surfaces. Reduction of the temperature differences mentioned
        may reduce this effect.
        If the floor has a historical, monumental value, installation of a floor heating
        may be of considerable risk. This may also be true for an underground archive
        present.

5.4.3   Infrared heating
        Advantages
        Energy consumption: Infrared heating devices may give a relatively high
        comfort level due to a relative high thermal radiant flux and only slightly
        heating up the air. The systems therefore are energy efficient because they
        directly deliver heat at the place where it is needed. Only those parts of the
        churches where people are sitting have to be heated. Heating up times are
        very fast.
        General: the system installation costs are low.
        Drawbacks
        Thermal comfort: Asymmetric heating may cause thermal comfort
        complaints. Higher head temperatures feel uncomfortable. Higher carbon
        dioxide levels will cause indisposition of people. Local heating up of surfaces
        may cause draught. People and parts of bodies remaining in the ‘shadow’ of
        the heaters will feel cold.
        Preservation: The radiant heating beam may heat up (monumental) interior
        parts like pillars, pews and pulpits very fast. This may cause thermal and
        hygric deformation of the materials. Furthermore open gas burning devices
        produce carbon dioxide and water vapor due to gas burning. Higher water
        vapor levels can cause condensation and high relative humidity problems near
        (unheated) indoor surfaces during heating up and cooling down periods. Only
        exhaust air extraction can solve this problem. The elements itself may disturb
        the esthetics of the monumental interior.
        General: the infrared heating devices and especially the red glare may disturb
        the monumental and architectural value. The gas burned heaters may cause
        considerable noise.

5.4.4   Radiator heating and radiant panels
        Advantages
        Radiator heating and radiant panels are heating systems that can be installed
        in a relatively simple way.
        Drawbacks
        These systems are only applicable in relatively small churches (< 200 seats). In
        larger spaces they may cause relatively large convective airflows. Heating up
        times may be considerable.
        General: the radiators may have a large impact on the monumental interior.




                                        5-103
5.4.5   Local (pew) heating
        Local pew heating most often consists of (foot) heating pipes, radiant heating
        panels or seat heating.
        Advantages
        Energy consumption: The systems are relatively energy efficient because they
        deliver heat directly to people. Due to this fact the heat capacity needed is
        relatively low.
        Thermal comfort: Radiant panels can compensate for low radiant surface
        temperatures. Heating up the feet area can prevent low feet temperatures.
        Drawbacks
        Thermal comfort: Due to the local heating, convective airflows are generated
        within the pews, which are a result of the heating up of the relative cool air
        entering the pews. This often leads to draught at the lower parts of the body.




                                        5-104
                                                           Heating systems
                               Warm air               Floor Infrared Radiator Convec- Local
                                                            radiant panel     tor     pew
                               Few      Distributed
                               inlets   inlets
Energy use   Heating rate      ++       ++            --     ++       0      0        ++
             Local heating     --       --            --     +        0      -        ++
Heating      Needed            0        0             +      +        -      --       ++
capacity     Producable        ++       ++            -      +        --     -        -
Thermal      Generated         -        +             0      -        -      0        -
comfort      airflows
             Radiant           -        -             +      ++       +      -        +
             temperature
             Radiant           0        0             +      -        0      0        +
             asymmetry
             Floor             08       08            ++     +        08     08       ++
             temperature
             Stratification    +9       +             +      +        +      +        +
                               -10
Preservation Contamination     +11      +11           --     -13      -13    -13      ++
                               -12      -12
           Changing RH         +14      +14           +      --15     -      -        +
           Condensation16      +        +             ++     --17     +      +        ++
Monumental Construction        +        --            --     0        0      -        0
impact     work
           Esthetical impact +          0             ++18 --         --     0        0
Costs      Installation      +          --            --   ++         0      0        0
     Table 5-6: Qualification of heating systems




     8 In case of wooden floor
     9 In case of low Archimedes number
     10 In case of high Archimedes number

     11 In case of air filtration

     12 Near air extraction grilles

     13 Above heater

     14 For low heating rates

     15 For directly heated objects

     16 ++ Low condensation risk

     17 Without air extraction
     18 If the monumental floor is not disturbed




                                              5-105
5.5     CHECKLIST FOR CHOICE AND DESIGN OF A PROPER HEATING
        SYSTEM
      1. Inventory
             a. Building
                      i. Geometry
                     ii. Volume
                    iii. Area and construction of construction parts
                             1. Floor
                             2. Walls
                             3. Ceiling
                             4. Roof
                             5. Glazing
             b. Heating system
                      i. Type of heating
                     ii. Heating capacity
                    iii. Control
                             1. Thermostatic device
                             2. RH control
                    iv. Condition of heating system
                     v. Space utilization
             c. Use of building
                      i. Church usage
                             1. Number of churchgoers
                             2. Number and duration of services
                     ii. Other use
             d. Ventilation and infiltration
                      i. Open or closed ceiling
                     ii. Recirculation or mixed heating
                    iii. Ventilation possibilities
             e. Interior of preservation
                      i. Monumental organ
                     ii. Wooden interior parts
                             1. Altar
                             2. Pews
                             3. Pulpit
                             4. Confession chair
                             5. Choir fence
                             6. Escutcheons
                    iii. Paintings
                             1. Ceiling
                             2. Wall
                             3. Canvas paintings
                    iv. Monumental floor
                             1. Tombstone


                                        5-106
                         2. Underground archive
                v. Stained glass
2. Existing heating system
       a. Check on meeting requirements using Table 5-3, Table 5-4 or Table
           5-5
                 i. Preservation
                         1. Thermal stratification
                         2. Relative humidity
                         3. Damage interior parts
                ii. Thermal comfort
                         1. General satisfaction
                         2. Draught
                         3. Asymmetry
                         4. Feet
3. Choice of heating system
       a. Formulate performance requirements according to Table 5-1
                 i. Choose appropriate limitation controls
       b. Replacement of old system
                 i. Check of meeting requirements Table 5-3, Table 5-4
                            or Table 5-5
       c. New heating system
                 i. Look at qualification of heating systems in Table 5-6
                ii. Calculate heating capacity using equation 2-8
               iii. Warm air heating?
                         1. Distributed air inlets?
                                 a. Calculate minimal number of inlets
                                     necessary using Figure 5-2
                                 b. Use ISSO_33 for global dimensioning air
                                     ducts and velocities
                                 c. Use ISSO_33 for global dimensioning filter
                                     sections
                                 d. Use ISSO_33 for global dimensioning
                                     ventilator device
                                 e. Use ISSO_33 for global dimensioning type
                                     boiler and capacity
                                 f. Check construction and possibilities floor
                                     work and underground archive
                                 g. Air flow calculation according to section 2.7
                                 h. Check thermal comfort at locations near air
                                     inlets using comfort analysing instrument
                         2. Concentrated inlet(s)
                                 a. Calculate minimal number of inlets
                                     necessary using Figure 5-2
                                 b. Check air inlet conditions




                                   5-107
                            i. Check possibilities of inlet
                               locations
                           ii. Calculate length of throw using
                               formula 2-51
                          iii. Choose direction of inlet flow not
                               directional with monumental
                               valuable parts
                          iv. Calculate Archimedes number
                               from formula 2-48
                 c. Use ISSO_33 for global dimensioning filter
                     sections
                 d. Use ISSO_33 for global dimensioning
                     ventilator device
                 e. Use ISSO_33 for global dimensioning type
                     boiler and capacity
iv. Floor heating?
        1. Monumental floor?
        2. Underground archive present?
        3. Check heating capacity and floor area
                 a. Use ISSO_33 for global floor heating
                     capacity
                 b. Use WaVo from section 2.2 or another
                     computer simulation program like ESPR or
                     (ISSO_10 1985) for calculation of
                            i. Capacity
                           ii. Heating performance
        4. Be aware of ceiling and or wall paintings and
            deposition
                 a. Limit temperature difference between floor
                     surface and air as much as possible; lower
                     air temperatures according to Table 5-4
                 b. Use calculations according to section 2.7 to
                     predict air flows
        5. Think of an additional (air) heating and use floor
            heating as primary heating
 v. Infrared radiant heating?
        1. Check direct radiant heating according to section 2.4
                 a. Check effects on monumental parts
                     according to section 2.4
        2. No air exhaust extraction
                 a. Check infiltration rate by measurement
                     according to section 3.2.3
                 b. Check condensation at cold surfaces using
                     moisture calculation with WaVo or another




                   5-108
                                               (hygric) computer simulation program
                                               according to section 2.2
                                  3. Check esthetical acceptance
                          vi. Radiator panel heating?
                                  1. Check number of seats less than 200
                                  2. Check esthetical acceptance
                                           a. Choose panels according to meet the
                                               heating capacity, calculated by formula 2-8.
                                               Use ISSO_33 for global dimensioning panel
                                               surface.
                                  3. Check on air flows and thermal comfort according
                                      to section 2.7, using Fluent, Airpak or another CFD
                                      program
                                  4. Check thermal comfort at locations near radiator
                                      panels
                         vii. Convector heating?
                                  1. Calculate minimal number of convectors necessary
                                      according to heating capacity, calculated by formula
                                      2-8. Use ISSO_33 for global dimensioning
                                      convectors.
                                  2. Check construction and possibilities floor work
                                  3. Air flow calculation by Fluent, Airpak or another
                                      CFD program, according to section 2.7
                                  4. Check thermal comfort at locations near convectors
                         viii. Pew heating?19
                                  1. Choose system according to meet thermal comfort
                                           a. Check feet temperature
                                           b. Check on draught
                                                     i. Air flow calculation by Fluent,
                                                        Airpak or another CFD program,
                                                        according to section 2.7
                                                    ii. Thermal comfort prediction
                                           c. Thermal model measurement in church or
                                               climate room comparable to (Loomans
                                               1998)




19   In April 2002 an EU research has started with the development of a pew heating system as the main
     object of study



                                                 5-109
5.6   RESULTS AND CONCLUSIONS

5.6.1 Case studies
For the experimental part the first choice was made to investigate a number of case
studies, representative for the most commonly encountered heating systems in the
Netherlands: warm air heating, floor heating and open gas infrared heating. The
general formulated results and conclusions, based on these field studies, were the
following:

Heating of churches
   - In principle artworks must be preserved in the same microclimate in which
       they have been kept for a long time. The microclimate conditions can be
       improved by attenuating or eliminating changes, e.g. diurnal cycles,
       fluctuations, gradients (Italian regulation UNI 10969, see section 1.1.3).
   - Moderate heating of churches therefore has a positive effect on the building
       itself and the preservation of the interior. Furthermore the dewpoint
       difference will increase and the risk for condensation on cold surfaces will
       decrease (see section 4.2). Heating of churches thus is an effective way to
       prevent churches to have too high air and near-surface relative humidity
       levels (see section 8.2.7).
   - Heating during cold winter days, however, can lead to a decrease of relative
       humidity, which is dangerous for the indoor monumental interior; it can lead
       to hygric shrinkage and cracking of wooden parts (see section 8.1.8).
   - As opposed to earlier times it is now possible (and needed) to have
       sophisticated control possibilities: computerized controls and monitoring
       make it possible to guard over the preservation criteria in church and to
       automatically adapt the heating to the posed demands.
   - Due to thermal stratification, monumental interior parts at higher positions,
       like monumental organs, are endangered more (see section 8.1.7).
   - A hygrostatic control is necessary for the preservation of valuable interior
       parts; it must switch off the heating system at low relative humidity levels of
       40 % RH and can switch the system on at higher relative humidity levels of
       75 % RH (see section 8.1.10, 8.2.7).
   - Limited air humidification is a possible solution for too low relative
       humidities during cold winter conditions. Setting a higher primary
       temperature might prevent condensation on cold indoor surfaces (see section
       8.1.9).
   - The natural ventilation of churches ranges from very low air exchange rates to
       high, depending mostly on the air permeability of the ceiling. Churches with
       wooden ceilings with air gaps had a much higher ventilation rate (n ≈ 0.4 to
       0.75) than churches with a stony, airtight ceiling (n ≈ 0.05 to 0.15) (see
       sections 8.2.5, 8.3.6, 8.4.6, 8.5.6, 8.6.4, 8.7.5, 8.8.6).



                                        5-110
    -   Protective glazing increases radiant temperatures and reduces downward
        draught airflows and thus improves thermal comfort. It reduces energy losses
        and risk for indoor surface condensation on the glazing (see section 8.3.7).

Warm air heating
   - For the temperature stratification the air inlet conditions are most critical. The
      choice of a smaller Archimedes number (decreasing air inlet temperature,
      increasing air velocity) is effective in improving mixing and thus reducing
      temperature stratification (see section 8.1.9, 8.2.5, 8.3.7).
   - Ventilating devices, hanging from the ceiling and blowing downward, do not
      improve the temperature stratification above the propeller blades; the sucking
      effect of the propeller blade is negligible and warm air might keep sticking to
      the ceiling (see section 8.1.6).
   - A distribution of air inlet grilles has a positive influence on the air
      temperature and velocity distribution and leads to a uniform air temperature
      (see section 8.3.7).
   - Air extraction grilles cannot influence the air distribution in a room (see
      section 8.3.7).

Floor heating
   - The time constant of a floor heating system can be in the order of magnitude
       of more than ten hours. It depends on the construction and insulation of the
       construction floor, the position of heating pipes, the thickness of stone floor
       (see section 8.4.7).
   - The temperature difference between floor and air causes thermally upward
       driven airflows. Increasing this temperature difference increases the air
       velocity (see section 8.4.6).
   - The radiant effect of floor heating leads to increased wall and ceiling
       temperatures. Wall temperatures decrease with height. Pews intercept this
       radiation (see section 8.4.9).
   - The increased airflows lead to contamination of wall and ceiling (see sections
       8.4.6 and 8.5.6).
   - Lower surface temperatures of the ceiling result in an increased
       contamination. Differences in the thermal resistance of the vault can be the
       origin of differences in contamination. It is concluded that thermophoresis
       might be an important mechanism of contamination (see section 8.5.6).

Open gas infrared heating
   - For every 100 kW installed capacity about 17 kg of vapor production is
      brought into a church each hour (see section 8.6.3).
   - Removing vapor by extracting air via the attic is a dangerous air extraction
      strategy: condensation will frequently occur at the cold inner roof surfaces
      (see section 8.6.5).
   - Condensation on cold surfaces like glazing and walls is almost inevitable (see
      section 8.7.5).


                                        5-111
    -   A possible solution to prevent early condensation on cold indoor surfaces is
        to heat up those surfaces with some kind of primary heating, other than
        infrared gas heating. Floor heating is one possible solution for this problem:
        the radiant heating effect is effective for heating up walls and ceiling (see
        section 8.8.7).
    -   For every 100 kW installed capacity about 21 kg of CO2 production is
        brought into a church each hour (see section 8.6.3); furthermore acids from
        rest gas constituents are the result of open air burning gasses.
    -   Direct heated surfaces heat up fast. Especially parts close to the radiant
        heaters are endangered. Dramatically low relative humidities at the surface will
        be the result of it. As a consequence drying out of wooden parts and related
        shrinkage will occur (see section 8.6.5).

5.6.2 Laboratory research
An important part of the experimental research was the laboratory work. This work
consisted of experiments on the soot production of candles and the material response
of wood on fluctuating relative humidity.

Soot production of candles and incense
Altar and offering candles and incense are important soot emitters. Due to the large
number of candles and duration of burning, the offering candles, however, may be
responsible for the largest contribution to the contamination (see section 8.1.6).

Experiments on wood
The material characteristics of beech in three main wood directions radial, tangential
and axial were measured. The ratio of vapor diffusion resistances were about
tangential : radial : axial = 10 : 5 : 1.
The sorption isotherm of beech was determined for a moisture adsorption trajectory
from 33 %RH to 97 %RH. For 9 different beech samples, spanning the three main
directions of wood, nearly full coinciding moisture isotherms were found.
Compared to hygric deformation, thermal deformations are negligible. From three
directions hygric tangential deformation is most important. Hygric radial deformation
is about 1.5 to 2 times smaller than the tangential. Hygric axial deformation is
negligible.
Compression tests on small wooden cylinders result in a clear decline of the modulus
of elasticity at higher moisture contents. In the hygroscopic range only a slight
indication of a decline in strength is present.
The response of wooden materials to stepwise and cyclic fluctuating indoor climate
conditions can be experimentally determined using NMR. The lower limit of moisture
content which could be measured was about 7 mass%. From the measurements a
mean moisture diffusivity for the hygroscopic range can be determined. During the
tests the related changing dimensions can be measured using strain gauges. Stepwise
and cyclic changing RH tests on wooden parts can be carried out. The moisture
content and the related deformation can be measured. For some well-defined




                                        5-112
geometries and boundary conditions thus experimental results can be obtained for
validating purposes of the numerical models.

5.6.3 Numerical modeling
An important part of the thesis is the part on numerical modeling. It demonstrates
that models nowadays are capable to predict the effects of heating of churches in an
early design state.
The model WaVo turned out to be a suitable tool for the hygrothermal modeling of a
church. It accounts for the thick massive walls. The model can be used to calculate
heat capacity, energy losses, air and surface temperatures and relative humidities. The
external climate can be derived from a weather station. For the ventilation rate
measured data are necessary. The accuracy of the model was checked with some
experimental results of the case studies. The results can be accurate within 5 to 10 %.
Indoor air humidity effects of open gas infrared heating can be predicted. For a first
impression of the effects of CO2 and vapor production a first order model can be
used. Later on WaVo can be used to model the indoor air humidity effects of open
gas infrared heating. In this way the moisture adsorbing effects of walls and interior
can be included.
The direct radiation effects of open gas infrared heating on a monumental interior can
be modeled with the raytracer computer simulation program Radiance. In a first
approximation only the direct radiant effects of the heaters can be modeled.
The changing moisture content of interior parts of wood due to changing relative
humidity of the indoor climate can be modeled with a hygrothermal 2-D model, based
e.g. on WUFI. The model, however, can only be used in the hygroscopic range. For
the input of the model the results of the laboratory measurements can be used. The
stepwise and cyclic changing moisture content of wood can be represented well.
The linear elastic, mechanical response of materials can be modeled. The partial
differential equations for a linear elastic approach of the mechanical strain and stresses
can be derived. The PDE’s can be modeled with FlexPDE. The stresses in two
wooden organ parts, due to a sudden change in relative humidity at the boundary,
were derived by simulation. Due to the hygrothermal load the maximum allowed
stresses were reached. The cracks in parts of a monumental organ thus could be
explained. The main limitation of the model, however, was the linear elastic approach.
The CFD model Fluent can be used for the calculation of airflows in churches. Due
to the complexity of geometry and partly unknown boundary conditions the accuracy
of the results may be limited. The prediction of tendencies in time is reasonable. On
the basis of measurements boundaries for the model can be determined and the
model then can be used for extrapolation in variant studies. The effects of an air jet
and the distribution of air inlet grilles with warm air heating, thermally generated air
flows with floor heating, thermal draught effects at cold walls have been modeled.

5.6.4 Performance requirements for church heating
From the literature study an actual array of performance requirements for church
heating was obtained. The foregoing work resulted in an upgrade and completion of
these performance requirements. For the three most important church heating



                                          5-113
systems, warm air heating, floor heating and open gas infrared heating, the arrays are
presented (see section 5.3).

5.6.5 Qualification of heating systems
A qualification table of heating systems was obtained for choosing an appropriate
heating system for a monumental church. The analysis thus results in advantages and
drawbacks of different heating systems from different points of view: energy usage,
heating capacity, thermal comfort level, preservation, monumental impact and costs
(see section 5.4).

5.6.6 Checklist for choice and design of a proper heating system
The thesis ends with a checklist, which can be used as a guideline for the choice and
design of a proper heating system (see section 5.5).




                                         5-114
5.7    DISCUSSION AND RECOMMENDATIONS FOR FURTHER RESEARCH

5.7.1 Case studies
The case studies in 8 different churches turned out to be the most valuable source for
information for this thesis. Some chronological progression in the field studies,
however, was present. As a result of that not all case studies had the same treatment
of study. The studies turned out to be valuable sources for building physics related
problems. A continuing field research program on building physical related problems
in monumental buildings therefore is proposed.

5.7.2 Heating system research
The most important heating systems for Dutch churches were incorporated: warm air
heating, floor heating and open gas infrared heating. Other heating systems like
radiator and convector heating, other infrared heating systems and pew heating were
not included. From these systems pew heating seems to be the most promising. In a
European Project on Cultural Heritage, started April 2002, a new pew heating system
will be developed and analyzed experimentally and by simulation. Part of the study is a
research for the possibilities of (local) air-conditioning: de- and humidification of air.
Further research on the improvement of computerized controls, monotoring and
protection mechanisms is proposed.

5.7.3 Research on wood
The laboratory experiments on wood were done for young beech. It was available in
the laboratory at the start of the research, but altered oak would have been a better
choice: it is more representative for the interior parts of old churches.
Relaxation test were not included in the present study. Measuring forces and stresses
due to hygrothermal load were not included in the present study. During the
hygroscopic measurements the vapor transfer coefficient was not known; in future
measurement it should be determined.
The model on wood was based on linear elastic behavior of materials. Wood is a
difficult material to simplify: it is anisotropic and has a typical non-linear behavior.
Relaxation for example is one of the typical non-linear qualities, which should be
included in the modeling. The non-linearity is important, especially when it comes to
describe the state before cracking. The linear, orthotropic approach therefore is only a
coarse start of modeling. Furthermore material data are needed for the orthotropic
modeling.
In March 2002 work on this omission already started in a master study.

5.7.4 Contamination of indoor surfaces
The contamination of indoor surfaces, due to deposition, was only described in the
fieldwork. The deposition mechanisms encountered were only mentioned, but not
studied. An experimental and modeling approach of the deposition and contamination
is subject for further study. In St. Martinus in Weert the discoloring of the ceiling as a
function of time is subject of one of our studies at the moment. In a round opening in


                                          5-115
the ceiling a sample tray is mounted. The discoloring and physical conditions of
samples and ceiling, subject to contamination, will be monitored for three years.

5.7.5 Degradation of plaster
At the start of this research, during the exploration of problems related with heating
systems in churches, degradation of plasters due to wall moisture transfer and salt
attack at interior surfaces was encountered in some churches. A relation with heating
systems seems evident. Therefore it was mentioned in the problem statement of this
thesis. In this study, however, until now no research on this topic was done. It is
recommended to do a research on the 2-D modeling of the building envelop, taking
into account capillary water uptake, e.g. at foundation level and water, vapor and salt
transport in walls.

5.7.6 Infestation by wood boring insects
In the Netherlands infestation by wood boring insects is an important degradation
problem in monumental buildings. One interesting topic of research is the coherency
of heating, especially stationary heating, and the occurrence of insect plagues in
buildings.
Another topic is a possible cure of the problem. One commonly used treatment in the
Netherlands and abroad is a so-called heat treatment. Especially attics are treated with
hot air to eliminate boring insects in wooden construction elements. The
hygrothermal and deformation effects on wood were subject of a master study
(Krijnen 2002). More research is necessary.

5.7.7 Ceiling insulation
Insulation of vaults is a very energy efficient measure in churches. Due to the lowering
of the attic temperature during wintertime, it also may have a positive effect to
prevent nestle of insects. The insulation material itself, however, may be an attractive
place for insects and rodents. In the Netherlands authorities on cultural heritage have
an aversion to this insulating measure. In a number of monumental buildings
insulation of vaults seemed to be the cause of most serious deterioration of
construction, such as the decay of wooden beams at the supporting area in walls. A
preliminary study indicated that summer condensation and thermal bridging effects
could have been the cause of it (Martens 2001). This research should be validated
experimentally.




                                         5-116
6 LITERATURE

Adan, O.C.G., 1994: On the Fungal Defacement of Interior Finishes. Thesis
   Eindhoven University of Technology.
Arendt, C., 1976: Kirchenheizung und Denkmalschutz. Heizung-Lüftung-
   Haustechnik. Heft 12, Seite 435-441
Arendt, C., 1993: Raumklima in grossen historischen Räumen: Heizungsart,
   Heizungsweise, Schadensentwicklung, Schadensverhinderung. Köln: R.Müller
Arendt, Claus, 1981: Kirchenheizung. Db 4/81
Arendt, Claus, 1995: Temperierung von Kirchen durch Solarenergienutzung aus dem
   Dachraum – Möglichkeiten und Grenzen. DAB 6/95
Arkel, Adriaan van, 1992: Verwarmingsinstallaties in oude kerken, in verband met het
   cultuurgoed en met meerdere gebruiksdoeleinden. Simulatie van
   verwarmingsinstallaties in de Grote- of St. Lebuinuskerk te Deventer.
   Afstudeerverslag Technische Universiteit Eindhoven (1992)
Arnold, Andreas, 1988: Natuurwetenschappen en Monumentenzorg. Centraal
   Laboratorium voor Onderzoek van Voorwerpen van Kunst en Wetenschap.
   Amsterdam
Arnold, A.; Zehnder K.; Küng, A; Emmenegger, O., 1991: Wandmalereizerfall, Salze
   und Raumklima in der Klosterkirche von Müstair. Zeitschrift für Kunsttechnologie
   und Konservierung. 5/91
Backstrom, Gunnar, 1994: Fields of Physics on the PC by Finite Element Analysis.
   Chartwell Bratt Ltd. Sweden.
Backstrom, Gunnar, 1998: Deformation and Vibration by Finite Element Analysis.
   Sudentlitteratur Lund, Sweden.
Backstrom, Gunnar, 1999: Fluid Dynamics by Finite Element Analysis.
   Sudentlitteratur Lund, Sweden.
Badertscher, H., 1965: Kirchenheizung und Luftbefeuchtung, ihr Einfluss auf die
   Orgel in den Kirchen. Firmenschrift der PLASCON AG, Basel
Baumann, Ernst, 1995: Kirchenrenovationen. Schweizer Ingenieur und Architekt. 37,
   808-812
Beck, W.; Koller, M., 1981: Probleme der Heizung in historischen Bauwerke
   Österreichs. Restauratorenblätter Band 5
Bekke, T. ter, 2001: Vochttransport in Monumentaal Metselwerk. Afstudeerverslag
   TU Eindhoven. FAGO 01.09.W.
Bordass, William, 1983/1996: Heating your Church. CIO Publishing London
Bowman, Mark, 1987: Holy Radiance. Building Services. April 1987
Camuffo, D.; Bernardi, A., 1995: The Microclimate of the Sistine Chapel. European
   Cultural Heritage Newsletter on Research. 2, 18.
Camuffo, D., 1998: Microclimate for Cultural Heritage. Developments in
   Atmospheric Science 23. Elsevier Amsterdam.



                                       6-117
Chen, Qingyan, 1988: Indoor Airflow, Air Quality and Energy Consumption of
   Buildings. Dissertation Delft University of Technology.
Costerus, A., 1973: Kerkverwarming. Afstudeerverslag Technische Universiteit Delft.
Du Ry van Heest Holle, Imke, 2001: Het Binnenklimaat van Monumentale Kerken;
   De Waalse Kerk te Delft. Afstudeerverslag Technische Universiteit Delft, Faculteit
   Bouwkunde, Vakgroep Bouwtechnologie, Sector Bouwfysica.
Eck, N.A.; Drolenga, J.C., 1997: Klooster te Wittem. Klimaatregeling.
   Projectverslagen TU Eindhoven
Ende, G., 1966. Die elektrische Speicherheizung in Schulen und Kirchen
   HLH Heizung Lüftung Haustechnik. 9
Ensink, Dipl.-Ing. Hermann, 1994: Neue Wege in der Beheizung von Kirchen und
   Gebäuden alter Bauart. TAB 6
Fanger, P.O., 1970: Thermal Comfort, thesis, Technical University of Denmark,
   Lyngby, Denmark
Fanger, P.O.; Melikov, A.K.; Hanzawa, H.; Ring, J., 1988: Air Turbulence and
   Sensation of Draught. Energy and Buildings. 12, 21-39
Fitzner, K., 1986: Luftführung in klimatisierten Sälen. Ki Klima-Kälte- Heizung 3,93-
   98
Fitzner, K., 1996: Energetisch günstige Abluft-Kanäle und –Öffnungen. HLH
   Heizung Lüftung Haustechnik. 10, 52-56
Franz, D., 1966: Elektro-Speicherheizungen in kirchlichen Anlagen. HLH Heizung
   Lüftung Haustechnik. 9
Frenzel, G., 1981: Probleme der Restaurierung, Konservierung und prophylaktischen
   Sicherung mittelalterlicher Glasmalereien. Kunstspiegel, 3. Jahrgang, 1981, Heft
   Nr. 3, Seite 173-209
Fluent Inc., 1995: Fluent User’s Guide, version 4.3, Lebanon, USA
Gossens, Ing. Heinz, 1977: Heizung, Lüftung und Klimatisierung von Kirchen.
   Sanitär und Heizungstechnik, 720-726
Gröber, H.; Sieler, W., 1935: Wärmebedarfsbestimmung von Kirchen. Beihefte zum
   Gesundheitsingenieur, Reihe 1, Heft 35
Groeger, H., 1979: Schmutzablagerungen in der Katholischen Pfarrkirche Schwanfeld,
   Gutachten vom 8. 3. 1979
Grunewald, J., 1997: Diffusiver und konvektiver Stoff- und Energietransport in
   kapillarporösen Baustoffen. Dissertation Technischen Universität Dresden
Hanel, Bernd M., 1994: Raumluftströmung. C.F. Müller Verlag GmbH, Heidelberg
Hamelinck, ir. M.J.H.; Vorenkamp, prof. Ir. J., 1993: Bethlehemkerk te Zwolle.
   Technische Universiteit Eindhoven, vakgroep FAGO
Heimeriks, J.H.M., 1989: Stralingsverwarming in Kerkgebouwen, VEG-Gasinstituut
Hennings, F., 1966: Probleme der Kirchenheizung. HLH Heizung Lüftung
   Haustechnik. 9
Hens, H., 1991: Reports International Energy Agency IEA Annex XIV, 1..4
   Condensation and Energy
Hibbeler, R.C., 1997: Mechanics of Materials. Third Edition. Prentice-Hall Inc. New
   Jersey




                                       6-118
Hollender, Joost; Hornikx, M., 2001: Klimaatbeheersing Rooms-Katholieke Kerk
    Heilige Maria Hemelvaart De Meern. Project verslag TU Eindhoven FAGO
    01.18.W.
Hout, Nicole H.A.M., 2001: Hygrisch en Mechanisch Gedrag van Hout bij
    Veranderende Vochtomstandigheden. FAGO-report 01.14.M, May 2001
Huynh, C.K.; Savolainen, H.; Vu-Duc, T.; Guillemin, M.; Iselin, F., 1991: Impact of
    Thermal Proofing of a Church on its Indoor Air Quality: the Combustion of
    Candles and Incense as a Source of Pollution. The Science of the Total
    Environment 102, 241 251
IBRK, 1983: Interdiocesane Bouw- en Restauratie Commissie van de R.-K.
    Kerkprovincie in Nederland. Zuiniger Energiegebruik in Kerkgebouwen
IBRK, 1999: Interdiocesane Bouw- en Restauratie Commissie. Het Beheersen van het
    Energieverbruik van Kerkgebouwen
ISO_7730, 1984: Moderate thermal environments – Determination of the PMV and
    PPD indices and specification of the conditions for thermal comfort. International
    Standard ISO 7730. International Organization for Standardization.
ISSO_10, 1985: Vloerverwarming. Publikatie 10, 2e druk, ISSO, Stichting Instituut
    voor Studie en Stimulering van Onderzoek op het Gebied van Verwarming en
    Luchtbehandeling
ISSO_33, 1996: Kengetallen en Vuistregels bij het Ontwerp van Gebouwinstallaties.
    Publikatie 33. Sichting ISSO-Rotterdam.
Jenoptik, 19xx: Handbook Varioscan. Jenoptik, Jena
Jütte, B.A.H.G., 1994: Passieve Conservering; Klimaat en Licht. Centraal
    Laboratorium voor Onderzoek van Voorwerpen van Kunst en Wetenschap
Jütte, B.A.H.G.; Schellen, H.L., 2000: Concept ‘Toetsingscriteria voor de verwarming
    van de Grote Sint Laurenskerk te Alkmaar middels infrarood (gas-)
    stralingsverwarming’. Intern rapport TUE
Kaan, H.F.; Ligthart, F.A., 1997: Grote Sint Laurenskerk te Alkmaar. Rapport
    Energieonderzoek Centrum Nederland. ECN-C—97-084
Kämpf, Andreas, 1994: Energetische und Physiologische Untersuchungen bei der
    Verwendung von Gasinfrarotstrahlern im Vergleich zu konkurierenden
    Heizsystemen für die Beheizung großer Räume. Dissertation Ruhr-Universität
    Bochum
Kanne, 1966: Neuartige Elektro-Warmluftheizungs- und Lüftungsanlage in der Kirche
    Brünen. HLH Heizung Lüftung Haustechnik. 9
Kiessl, K, 1983: Kapillarer und dampfförmiger Feuchtetransport in mehrschichtigen
    Bauteilen – Rechnerische Erfassung und bauphysikalische Anwendung,
    Dissertation, Universität-Gesamthochschule Essen
Kimmenade van, 2000: List of heating systems in Dutch churches, as collected by van
    Kimmenade, Mark air heating systems; available in Microsoft Excel format
Knol, H. e.a., 1971: Kerkverwarming en Orgels. Rapport in opdracht van de
    Vereniging van Kerkvoogdijen in de Nederlandse Hervormde Kerk
Koene, F.G.H.; Ligthart, F.A.T.M.; Schellen, H.L.; Jütte, B.A.H.G.; Kragt, W.;
    Stappers, M.H.L., 2001: Stralingsbranders in de Grote Sint Laurenskerk te




                                        6-119
   Alkmaar. Hoe gaan gebruik en behoud van de kerk samen. ECN rapport ECN-
   C—01-125
Kopinga K. and Pel, L., 1994: One dimensional scanning of moisture in porous
   materials with NMR, Rev. Sci. Instrum. 65, 3673-3681.
Korn, Ulf-Dietrich, 1971: Ursachen und Symptome des Zerfalls mittelalterlicher
   Glasgemälde? Deutsche Kunst und Denkmalpflege, 1
Korsgaard, Vagn, 1990: Energibesparelser I Kirker. Laboratoriet for Varmeisolering.
   Meddelelse. 210. Danmarks Tekniske Hojskole
Kriegel, Bernd, 1973: Fallströmungen vor Abkühlungsflächen in Gebäuden und
   mögliche Schutzmassnahmen
Krijnen, M.J., 2002: Warmtebehandeling ter bestrijding van houtaantastende insecten.
   De effecten op het gedrag van hout. FAGO-rapport 02.05.W
Krischer, O.; Kast, W., 1957: Zur Frage des Wärmebedarfs beim Anheizen selten
   beheizter Gebäude. Gesundheits-Ingenieur. 21/22, 321-352
Krus, Dipl.-Ing. Martin, 1995: Feuchtetransport- und Speicherkoeffizienten poröser
   mineralischer Baustoffe. Theoretische Grundlagen und neue Messtechniken.
   Dissertation Universität Stuttgart
Künzel, Dr.-Ing.; Holz, Dipl.-Ing. D., 1991: Richtiges Heizen in historischen
   Gebäuden. VDI Berichte, 896
Künzel, Dr.-Ing. H.; Holz, Dipl.-Ing. D., 1991: Bauphysikalische Untersuchungen in
   unbeheizten und beheizten Gebäuden alter Bauart.Bericht aus dem Fraunhofer-
   Institut für Bauphysik, FB-32/1991
Künzel, Dr.-Ing. H.M., 1994: Verfahren zur Eind- und Zweidimensionalen
   Berechnung des gekoppelten Wärme- und Feuchtetransports in Bauteilen mit
   einfachen Kennwerten. Dissertation Universität Stuttgart
Laan, M.J. van der, 1994: A Model for the Combined Heat and Moisture Transfer in
   Building Constructions. Report 94-CBO-R0331 Centrum Bouwonderzoek TNO-
   TUE Eindhoven, April 1994
Larson, Greg Ward; Shakespeare, Rob A., 1998: Rendering with Radiance; The Art
   and Science of Lighting Visualization. Morgan Kaufmann Publishers.
Leijendeckers, P.H.H., 1995: Leerboek Klimaatregeling. Technische Universiteit
   Eindhoven, Faculteit Bouwkunde.
Lekhnitskii, S.G.,1963: Theory of Elasticity of an Anisotropic Elastic Body. Holden-
   Day, Inc. San Francisco
Ligthart, F.A.T.M.; Koene, F.G.H., 2000: Het functioneren van stralingsbranders in de
   Grote Sint Laurenskerk te Alkmaar. Voorafdruk rapport ECN, ECN-I—99-001
Loomans, Marcel, 1998: The Measurement and Simulation of Indoor Air Flow.
   Dissertation Eindhoven University of Technology
Mainz, 1972: Diözesan-Baumeister und Baureferenten. Richtlinien für die Beheizung
   von Kirchen. Tagung der Diözesan-Baumeister und Baureferenten. Mainz. Das
   Münster, 4, 257-271
Martens M., 2001: Onderzoek naar het na-isoleren van de Sint Martinuskerk te Weert.
   T8 Projektverslag TU Eindhoven, Capaciteitsgroep FAGO.




                                       6-120
Mayer, G., 1981: Konservierung und Restaurierung mittelalterlicher Glasmalereien.
   Ein Plädoyer für Sicherung und gegen zu weit gehende Restaurierung.
   Kunstspiegel, 3. Jahrgang, 1981, Heft Nr. 3, Seite 245-250
Müller, C.: Auswirkungen von Kirchenheizungen auf Innenräume und Inventar.
   Jahrbuch der bayerischen Denkmahlpflege, Band 24, Seite 399-409
Neilen D., 2002: Monumentale kerken met verwarming. Een bouwfysische
   vergelijking aan de hand van kentallen. Afstudeerverslag TU Eindhoven. FAGO-
   rapport 02.07.W.
Nijland M., 1999: Problematiek Roetdepositie in de St. Martinuskerk te Weert. T8
   Projektverslag TU Eindhoven, Capaciteitsgroep FAGO. December 1999
Norgaard, Jens; Madsen, Thomas Lund, 1986: Kirkevarmeanlag. Laboratoriet for
   Varmeisolering. Meddelelse. 181. Danmarks Tekniske Hojskole
Oidtmann, S.J.C.; Schellen, H.L., 1992: Schutzverglasung vor historischen Gemälden
   – eine theoretische Betrachtung. Bauphysik 14
Oidtmann, S.J.C., 1994: Die Schutzverglasung – eine wirksame Schutzmassnahme
   gegen die Korrosion an wertvollen Glasmalereien. Dissertation Technische
   Universität Eindhoven
Pel, Leo, 1995: Moisture Transport in Porous Building Materials. Dissertation
   Eindhoven, University of Technology
Peyer, Werner, 1994: Programmierbare Heizungsregelung für Kirchen. HeizungKlima.
   3, 56-59
Pfeil, dr.-ing. Axel, 1975/1: Kirchenheizung und Denkmalschutz. Bauverlag GMBG –
   Wiesbaden und Berlin
Pfeil, dr.-ing. Axel, 1975/2: Untersuchungen über das Raumklima sowie den Wärme-
   und Feuchtehaushalt in Kirchen alter Bauart und deren Einflüsse beim Einbau von
   Kirchenheizungssystemen. Dissertation Fakultät für Bauwesen. Technischen
   Universität Carolo-Wilhelmina zu Braunschweig
Preissler, Dipl. Ing. H.A., 1988: Statt Ausstellungsstück gebrauchsfähiges Kulturgut.
   HLH Heizung Lüftung Haustechnik. 6.
RDMZ, 2000/1: Rijksdienst voor de Monumentenzorg. Signaleringskaart ten behoeve
   van informatie en inventarisatie van schades aan monumentale orgels als gevolg
   van verwarming. RDMZ Zeist
RDMZ, 2000/2: Rijksdienst voor de Monumentenzorg. Jaarverslag 2000. RDMZ
   Zeist
Recknagel, Sprengel, 1981: Taschenbuch für Heizung und Klimatechnik. R.
   Oldenbourg München Wien
Rutten, Paul, 1996: Geïntegreerd Ontwerpen van Gebouw en Klimaatinstallaties.
   Dictaat Technische Universiteit Eindhoven, Faculteit Buwkunde.
Sanders, Jennifer; Mensink, Maaike, 2000: Schade aan Monumentale Orgels.
   Reinwardt Academie, Beheer en Behoud.
Schellen, H.L.; Schot, B.; Oidtmann, S.J.C., 1994: Surface Condensation on leaded
   Windows and the Indoor Climate of a Church. European Simulation
   Multiconference. Barcelona
Schellen, H.L.; Stevens, J., 1998/1: Rapportage Metingen Eerste Half Jaar 1998 in de
   St. Martinuskerk Weert. TNO-rapport 98-CBO-R1306 SNH/BNI.



                                       6-121
Schellen, H.L., 1998/2: Onderzoek naar Klimaatbeheersing in Monumentale
   Gebouwen. Literatuuronderzoek. TNO-rapport 98-CBO-R1307 SNH/BNI
Schellen, H.L.; Diepens, J.F.L.; Schijndel, A.W.M. van; Vugts, J.; Schopmeijer,
   M.E.D., 1999/1: Bouwfysische Aspecten van de Toepassing van gas-
   infraroodinstallaties in de Grote Kerk te Dordrecht. TNO-rapport 99-CBO-R006
   SNH/BNI.
Schellen, H.L.; Stevens, J., 1999/2: Rapportage Metingen Eerste Jaar 1998 in de St.
   Martinuskerk Weert. TNO-rapport 99-CBO-R14 SNH/BNI.
Schellen, H.L.; Wit, M.H. de., 1999/3: Heat- and Moisture Modeling of a
   Monumental, Massive Building. 5th Symposium on Building Physics in the Nordic
   Countries. Gothenburg, Sweden.
Schellen, H.L.; Schijndel, A.W.M. van, 2000/1: Temperatuur- en vochtbelasting van
   kerk en interieur door gas-infrarood-stralers in de Grote Sint Laurenskerk in
   Alkmaar. Intern rapport FAGO, 00.02.W, Technische Universiteit Eindhoven.
Schellen, H.L., 2000/2: Klimaatbeheersing in Monumentale Kerken. Katern
   Praktijkboek Instandhouding Monumenten. SDU Uitgevers, Den Haag.
Schellen, H.L., 2000/3: Building Physical Impact of Heating Systems on Monumental
   Buildings and their Interior. International Building Physics Conference 2000.
   Eindhoven, The Netherlands.
Schellen, H.L.; Aarle, M.A.P. van, 2000/4: Eindrapportage van het Onderzoek naar de
   Vervuiling van Gewelfschilderingen in de St. Martinuskerk Weert. TNO-rapport
   2000-CBO-R027 SNH/DNA.
Schellen, H.L.; Aarle, M.A.P. van, 2001/1: Rapportage van de Binnenklimaatmetingen
   in de Sypekerk te Nieuw-Loosdrecht. TUE-rapport 01.01.W. Eindhoven.
Schellen, H.L., 2001/2: Prestatie-eisen en simulatieberekeningen voor de
   klimaatbeheersing van de St. Maartenskerk te Zaltbommel. TUE-rapport 01.02.W.
   Eindhoven.
Schellen, H.L.; Diepens, J.F.L.; Schijndel, A.W.M. van, 2001/3: Building Physical
   Impact of Infrared Gas Heating on Monumental Churches and their Interior. VII
   Polish Conference of Science and Technology. Building Physics in Theory and
   Practice. Lodz, Poland.
Schellen, H.L.; Hout, N.H.A.M.; Stappers, M.H.L., 2001/4: Warmtebehandeling van
   door insecten aangetast hout. Een vergelijking van twee in Nederland gebruikte
   methoden. TUE-rapport 01.08.W. Eindhoven.
Schellen, H.L., 2001/5: Eindrapportage van het onderzoek naar de vervuiling van
   gewelfschilderingen in de St. Gerlachuskerk te Houthem. TUE rapport 01.21.W
Schellen, H.L., 2001/6: Eindrapportage van het onderzoek naar de klimaatbeheersing
   in de St. Liduïna Basiliek in Schiedam. TUE rapport 01.20.W
Schellen, H.L., 2001/7: Glas-in-lood en voorzetbeglazing in relatie tot het
   binnenklimaat in kerken. Studiedag WTA Nederland-Vlaanderen: glas in lood,
   restauratie en conservatie
 Schijndel, A.W.M. and Wit, M.H. de., 1999: A Building Physics Toolbox in Matlab.
   5th Symposium on Building Physics in the Nordic Countries. Gothenburg,
   Sweden.




                                      6-122
Schlieder, Hans, 1969: Betrachtungen über Kirchenheizungen und Heizungsschäden.
   Nachrichtenblatt der Denkmalpflege in Baden Würtemberg. 4
Schlieder, Hans, 1967: Schäden an historischgen Kirchenorgeln und anderen
   Kunstwerken. Deutsche Kunst und Denkmalpflege. 1.
Schmidt-Thomsen, 1972: Denkmalpflege und Kirchenheizung
Schmidt, Dipl.-Ing. Karl Heinz, 1982: Beheizung von Kirchen. DBZ Bauerneuerung.
   9, 1227-1230.
Stadtmüller, P.A., 1972: Orgel und Heizung. Kirchenheizung, Bericht und Ergebnis
   der Jahrestagung der Diözesanbaumeister und Baureferenten vom 24. 5.-27.5.1972
   in Mainz. Sonderausdruck Das Münster, 25. Jahrgang, Heft ½, 1973, Seite 2-3
Stangier, R.; Rüegg, T.; Stöckli, B.; Lommel A., 1995: Luftwechsel und
   Lüftungswärmeverlust der Evangelisch-Reformierten Kirche von Ebnat (SG).
   Untersuchungsbericht EMPA F+E 144
Stappers, M.H.L., 2000: De Waalse Kerk te Delft. Een onderzoek naar het behoud
   van een monumentaal orgel. Afstudeerrapport TU Eindhoven, Capaciteitsgroep
   FAGO. FAGO-rapport 00.03.W.
Stappers, M.H.L.; Hout, N. van; Schellen, H.L., 2000: Warmtebehandeling tegen
   Houtaantastende Insecten. Fact Sheet Rijksdienst voor de Monumentenzorg.
Stevens J., 1999: Klimaatbeheersing in de St. Martinuskerk te Weert; onderzoek naar
   vervuiling van gewelfschilderingen. Afstudeerrapport TU Eindhoven,
   capaciteitsgroep FAGO.
Stevens J.; Schellen, H.L., 2000: Klimaatbeheersing in de St. Martinuskerk te Weert;
   onderzoek naar vervuiling van gewelfschilderingen. Bouwfysica, Nederlands
   Vlaamse Bouwfysica Vereniging, april 2000.
Stoy, B., 1966: Diskussionsbeitrag zum Thema “Elektrische Beheizung von Schulen
   und Kirchen”. HLH Heizung Lüftung Haustechnik. 9
Supper, W., 1967: Kirchenheizung und Orgel. Deutsche Kunst- und Denkmahlpflege,
   25. Jahrgang, 1967. Heft 1, Seite 39-45
Tassau, Savvas; Xiang, Weizhong; Kolokotroni, Maria, 2001: Evaluation of Heating
   Systems in Churches. Final Report. Department of Mechanical Engineering.
   Brunel University.
Thomson, Garry, 1978: The Museum Environment. Butterwort Publishers Inc.
Toussaint W., 1964: Elektrische Fussbodenspeicherheizung im Kölner Dom. Conti
   Elektro Berichte, April/Juni 1964
Vorenkamp J. , Schellen H.L. en Pernot C., 1989: Rapportage Grote of Lebuïnuskerk
   te Deventer. Intern rapport FAGO, nr. 89.09.K
Vorenkamp, J., 1993: Eindrapport Betlehemkerk te Zwolle. Spraakverstaanbaarheid,
   zoutuitbloei, verwarmingsinstallatie, vochtregeling, boktor. FAGO 93.90.W
Vugts J., 1999: Onderzoek naar het klimaat in de R.K. Kerk te Bemmel. Project
   verslag FAGO, Technische Universiteit Eindhoven.
Wallot, Paul, 1993: Die Heiz- und Raumlufttechnik im historischem
   Teichtagsgebäude. Die Bauverwaltung, 4,154-156.
Wegner, M., 1972: Untersuchungen zur Beheizung von Kirchen. Abschlussbericht,
   Forschungsstelle für Energiewirtschaft, München.




                                       6-123
Weik, Prof. Dr. H., 1986: Zur Frage der Beheizung grosser Kirchen durch Sonnen-
   und Umweltenergie. Haustechnische Rundschau, 7-8
Wit, M.H. de, 1987: ELAN- A Computer Model for Building Energy Design
   Building and Environment 23, 285-289
Wit, M.H. de; Donze, G.J., 1990: A Second Order Model for the Prediction of Indoor
   Air Humidity. NOVEM contract, FAGO-W/E, 90.08.W, April 1990, pp. 106.
Wit, M.H. de, 2000: WaVo, a simulation tool for the thermal and hygric performance
   of a building. FAGO internal report. Eindhoven University of Technology.
Zanten, M.M. van, 1994: Gids voor behoud en beheer van kerkelijk kunstbezit: een
   praktische handleiding. Stichting Kerkelijk Kunstbezit in Nederland, Den Haag




                                      6-124
7 NOMENCLATURE

                                k
a       thermal diffusivity (      )                                    [m2/s]
                                ρc
a       absorptivity                                                    [-]
A       surface area                                                    [m2]
b       thermal effusivity
        [J/(m2Ks1/2)]
b       width                                                           [m]
C       concentration                                                   [kg/m3]
c       specific heat                                                   [J/kgK]
C       heat capacity                                                   [J/K]
cp      constant pressure specific heat                                 [J/kgK]
d       thickness                                                       [m]
D       diameter                                                        [m]
Dv      water vapor diffusivity                                         [m2/s]
E       radiant flux density                                            [W/m2]
Emod    modulus of elasticity                                           [N/m2]
F       force                                                           [N]
g       gravitational acceleration                                      [m/s2]
G       production rate                                                 [kg/s]
h       heat transfer coefficient                                       [W/m2K]
h       height                                                          [m]
H       specific enthalpy                                               [J/kg]
Iclo    thermal resistance from skin to outer surface of clothed body   [clo]
k       thermal conductivity                                            [W/mK]
l       length                                                          [m]
L       heat loss coefficient                                           [W/K]
M       mixing number                                                   [-]
M       rate of metabolic heat production                               [W/m2]
    ⋅
M       mass flow rate                                                  [kg/s]
n       air exchange rate                                               [h-1]
n       number of samples                                               [-]
Nu      Nusselt number                                                  [-]
p       vapor pressure                                                  [Pa]
psat    saturation pressure                                             [Pa]
Pr      Prandtl number                                                  [-]
PD      percentage of dissatisfied due to draught                       [%]
PMV     predicted mean vote                                             [-]
PPD     predicted percentage of dissatisfied                            [%]
q       heat flux                                                       [W/m2]


                                        7-125
qc        convective heat flux                                         [W/m2]
Q         heat load                                                    [W]
r         radius                                                       [m]
Re        Reynolds number                                              [-]
RH        relative humidity                                            [%]
t         time                                                         [s]
T         temperature                                                  [K]
∆T        temperature difference                                       [K]
TI        turbulence intensity                                         [-]
u         flow velocity                                                [m/s]
U         overall heat transfer coefficient                            [W/m2K]
v         velocity component                                           [m/s]
V         volume                                                       [m3]
 ⋅
V         air flow rate                                                [m3/s]
w         velocity component                                           [m/s]
w         width                                                        [m/s]
w         mass content of water                                        [kg/m3]
x, y, z   cartesian coordinates in three directions                    [m]
x         water content                                                [kg/kg]
zo        heating up time                                              [s]


Greek symbols
α      linear thermal expansivity                                      [m/mK]
α         angle                                                        [rad]
β         volumetric thermal expansian coefficient                     [K-1]
βv        vapor transfer coefficient                                   [s/m]
δa        water vapor permeability                                     [s]
∆         difference
ε         emission factor                                              [-]
ε         normal strain component                                      [-]
Φ         heat flow                                                    [W]
γ         the shear strain components associated with the three axes   [-]
η         efficiency                                                   [-]
κ         linear relative deformation
          [m/m(kg/m3)]
λ         wavelength                                                   [m]
µ         vapor diffusion resistance number                            [-]
µ         Poisson’s ratio                                              [-]
µ         the dynamic viscosity ( νρ )                                 [kg/ms]
                                µ
ν         kinematic viscosity ( )                                      [m2/s]
                                ρ



                                              7-126
θ       temperature                                    [oC]
ρ       mass density                                   [kg/m3]
σ       normal component of stress                     [Pa]
σ       standard deviation                             [-]
τ       time constant                                  [s]
τ       shear component of stress                      [Pa]
τp      transmission factor air path                   [-]
ϕ       relative humidity                              [-]
ξ       tangent hygroscopic curve                      [kg/m3]

Subscripts
A      ambient
c      convective
d      diameter
e      outdoor, external
h      hydraulic
i      indoor
i, j   coordinate
k      conductive
m      mean
mrt    mean radiant temperature
o      object
op     operative
r      radiant
s      surface
sat     saturation
t       turbulant
tot     total
v       vapor
w       wall
x       environmental
0       initial
0       at inlet

Operators
∇      gradient operator (also written as grad)        [m-1]
∇⋅     divergence operator (also written as div)       [m-1]
∇2     Laplacian operator (also written as div grad)   [m-2]




                                         7-127
Dimensionless numbers
       ∆T g ⋅ L
Ar =     ⋅                      Archimedes number
       T u2


       a⋅t
Fo =                               Fourier number
       L2

       ∆T g ⋅ L3
Gr =     ⋅ 2                       Grashof number
       T   ν

       h⋅L
Nu =                               Nusselt number
        k

       ν
Pr =                               Prandtl number
       a

Ra = Gr ⋅ Pr                      Rayleigh number

       u⋅L
Re =                              Reynolds number
        ν




                        7-128
8 APPENDIX A: CASE STUDIES


8.1   DAMAGE TO A MONUMENTAL ORGAN IN THE WAALSE KERK IN
      DELFT

8.1.1 Description and history of the church
The original chapel of St. Agatha’s Convent was built between 1430 and 1480. William
of Orange frequently attended the French spoken services from 1573 to 1584. The
name ‘Prinsenhofcomplex’ is a result of this.
In 1585 the church was handed over to the Walloon Community. Because the length
of the church was not appropriate for Protestant services, one half of the church was
separated. Nowadays the Walloon Church in Delft is still one half of the original
church building: the other part is a museum: the Prinsenhof Museum. The
Prinsenhofcomplex as a whole was restored from 1950 to 1962. The monumental
Bätz/Witte organ in the church dates from 1869 and was last restored in 2000. The
monumental interior parts consist of a monumental pulpit, monumental pews,
monumental escutcheons and an important burial monument.




8.1.2 Building data
The outdoor walls of the church are made of masonry and have a mean thickness of
0.75 m. The walls on the indoor side are plastered and whitewashed. The glazing in
the outdoor walls is made of plain leaded glass. The floor in the church with an area
of 210 m2 consists of natural stone and gravestones. The vault is made of boards of
soft wood and has a lot of cracks. The roof is made of construction wood, covered
with slates. The volume of the church is about 3000 m3.



                                        8-129
8.1.3 Heating system
The heating system in the Walloon Church in Delft was originally a direct hot air
system Mark Fohn type 115RH of Mark Heating Systems. The heat capacity of the
heating system was 115/77 kW (high/low). The hot air was brought into the church
by a re-circulation system with an airflow rate of 7400/5000 m3/h (high/low). The
inlet consisted of a single grille at about 3 m above floor level. The outlet was a single
grille in the floor. The air was re-circulated for 100 %. The thermostatic device in the
church was located near the pulpit at a height of about 3 meters.

8.1.4 Formulation of the problem
In the Walloon Church in Delft the monumental organ was in bad shape as a result of
the hot air heating and a major restoration was inevitable. The most important target
of the study was to determine the main cause of the deterioration of the organ and to
save the organ from demolition in the future.

8.1.5 Analysis
From the literature study (Schellen 1998/2) it was known that air heating might cause
severe problems for monumental organs and other monumental objects in the interior
of a church. The assumption was made that high air inlet temperatures cause large
thermal stratification and thus lead to high air temperature at elevated levels, where in
most cases the monumental organs are to be found. A high air temperature involves a
low relative humidity. Dramatic low relative humidity values and related drying out
and shrinkage of the organic parts are therefore the result. Cracks and other
indications of shrinkage in the wooden cabinet of the organ supported this theory.
The analysis consisted of a damage analysis on the monumental organ (Sanders ea,
2000), building physical measurements (Stappers, 2000) and a simulation study of the
church in situ (Neilen, 2002) and laboratory study on the behavior of wooden parts
under changing climatic conditions (Hout, 2001).

8.1.6 Measurements
The measurements in Delft can be grouped into short time or one-day measurements,
and long-term or one-year measurements.

Smoke analysis
During the one-day measurements a smoke analysis of the indoor airflow pattern took
place. Smoke was generated by evaporation of paraffin oil and introduced at the re-
circulation outlet and thus, by re-circulation, introduced in the church by the air inlet.
The flow pattern of the smoke can be described as follows: smoke entered the church
at the air inlet grille in a vertical east oriented wall with a certain horizontal velocity.
Due to its temperature difference with the surrounding church air temperature, the
inlet air had an upward directed thermal driven velocity component and directly
moved to the ceiling, sticking close to the ceiling and moved to the opposite west wall
of the church. Ventilating devices, hanging from the ceiling in front of the organ and
blowing downward, could not improve the temperature stratification above the



                                           8-130
propeller blades; the sucking effect of the propeller blade was negligible and warm air
kept sticking to the ceiling. The smoke finally cooled down at this wall and was
redirected downwards, directly moving to the monumental Bätz/Witte organ cabinet
at the upper floor organ department. From here the air moved down into the church
and followed its way to the air exhaust grille in the floor, located a few metres before
the east wall with the air inlet grille.




Figure 8-1: Smoke pattern in the Walloon Church


Inlet air conditions
The inlet air conditions were measured at the inlet grille. Inlet air temperature was
measured continuously during the long-term measurement period and inlet air velocity
measurements took place during the one-day measurement period.
                         Air temperature near inlet                                            Air velocity near inlet
         80                                                                    4

         70
                                                                               3
         60

         50                                                                    2
                                                                     u [m/s]
θ [ C]
   o




         40
                                                                               1
         30

         20
                                                                               0
         10
         00:00   04:00    08:00 12:00     16:00   20:00   24:00                12:50   13:50      14:50      15:50       16:50   17:50
                                 t [h]                                                                 t [h]
Figure 8-2: Air temperature and velocity near inlet

Indoor air conditions
To determine the temperature stratification, the air temperatures and relative
humidities were measured at four different height levels in front of the organ. From



                                                                  8-131
the data measured a typical period was extracted. The figures below indicate air
temperatures and relative humidities measured at four different heights, 15 m being
the total height of the church.
                               HANWELL T h=5.2m                              HANWELL T h=8.5m
                     30                                             30

                     25                                             25
            θ [ C]




                                                          θ [ C]
                     20                                             20
               o




                                                             o
                     15                                             15

                     10                                             10
                          0     20    40     60     80                   0    20    40    60     80
                                    time                                          time
                               HANWELL T h=11.7m                             HANWELL T h=15m
                     30                                             35

                                                                    30
                     25
                                                                    25
            θ [ C]




                                                          θ [ C]

                     20
               o




                                                             o




                                                                    20
                     15
                                                                    15

                     10                                             10
                          0     20     40   60      80                   0    20     40   60     80
                                     time                                          time


Figure 8-3: Air temperatures measured at different heights

                              HANWELL RH h=5.2m                              HANWELL RH h=8.5m
                     70                                             70

                     60                                             60

                     50                                             50
            RH [%]




                                                           RH [%]




                     40                                             40

                     30                                             30

                     20                                             20
                          0     20    40    60     80                    0         500           1000
                                    time                                          time
                              HANWELL RH h=11.7m                             HANWELL RH h=15m
                     70                                             80

                     60
                                                                    60
                     50
            RH [%]




                                                           RH [%]




                                                                    40
                     40
                                                                    20
                     30

                     20                                              0
                          0           500          1000                  0          500          1000
                                     time                                          time

Figure 8-4: Relative air humidities measured at different heights




                                                     8-132
From the indoor air conditions measured during a year a typical Sunday service
(January 17th 2000) was extracted. The figure below indicates air temperatures and
relative humidities in front of the organ at a height of 15 metres, just beneath the
vault.
                                  Temperature                                                       Relative humidity
           35                                                                 60


           30                                                                 50


           25                                                                 40




                                                                     RH [%]
  θ [ C]
     o




           20                                                                 30


           15                                                                 20


           10                                                                 10
                  00:00 04:00 08:00 12:00 16:00 20:00 24:00                          00:00 04:00 08:00 12:00 16:00 20:00 24:00
                                    time                                                               time

Figure 8-5: Air temperatures and relative humidities measured in front of the organ at
a height of 15 meters


Temperature and relative humidity stratification
From the data of the temperatures at different heights the temperature stratification
can be extracted. The figure below gives a typical representation.


                          Temperature stratification                                                   Relative humidity stratification
        16                                                                    16
                                  station 11                                                           station 11
        14                                                                    14


        12                        station 10                                  12                       station 10

        10                                                                    10

                                  station 9                                                            station 9
h [m]




                                                                     h [m]




           8                                                                  8


           6                                                                  6
                                  station 8                                                            station 8

           4                                                                  4


           2                                                                  2
                 zondag 17 januari 1999                  1
                                                     time: 200                      zondag 17 januari 1999               time:1200
           0                                                                  0
                                                                               10           15          20            25           30     35   40
            10          15         20           25          30
                                                                                                                    RH [%]
                                      o
                                   θ [ C]

Figure 8-6: Typical temperature and relative humidity stratification during a service




                                                                 8-133
8.1.7   Conclusions actual situation
     −  The air heating with high air inlet temperatures led to high temperature
        stratification and thus to very low relative humidity in the upper part of the
        church.
    − Ventilator devices, hanging at the ceiling of the church and mixing the air
        below the propellor blade, could not prevent the warm air to stick to the
        ceiling and move above the propellor blades to the organ.
    − The actual indoor climate conditions at the monumental organ were
        compared with the recommendations from literature to protect it from
        deterioration. It was clear that the relative humidity dropped to very
        threatening low levels of 20 to 25 % RH. This must have had a great impact
        on the wooden parts and cabinet.
    − A damage analysis was needed to show that these low relative humidity levels
        resulted in shrinkage of the wooden parts of the monumental organ and thus
        led to related cracks in the wooden parts.
Therefore it was decided to carry out a damage analysis for the monumental organ
(Sanders et. al. 2000).




                                        8-134
8.1.8 Damage analysis
During the restoration of the organ at the restoration workplace of Gebroeders Reil
BV, the inner parts of the organ were analyzed on the occurrence of damage by the
Reinwardt Academy, a Dutch school for museum science based in Amsterdam.
The bellows were severely damaged; severe cracks occurring within the wooden parts
of the bellows; their sheepskin parts were torn.




Several wooden ducts of the organ were cracked. The bends of it were covered with
sheepskin and here leakages occurred.
The bank of keys, the clavier, had a problem with ‘hanging’ keys due to the shrinkage
or swelling of parts of the playing mechanism.




The wind drawer, the heart of an organ, was damaged most severely. At the glued
parts of the wood open joints occurred. At the wind holes of the drawer cracks
separated the wooden parts. This kind of damage is most disastrous for an organ: the
leakage of the wind drawer will have the effect that other organ pipes will sound
during playing of the separate pipes.




                                        8-135
The slider showed shrinkage damage leading to leakage and wind sag.




The wooden organ pipes had cracks at the tuning caps due to warping and shrinking
of the wooden edges.

Furthermore the organ was out of tune due to displacements of the tuning caps.

At the wooden cabinet of the organ in the Walloon Church the following damage has
been observed:




                                       8-136
The joints of the wooden panels in the wooden frames showed shrinkage damage.
One of the panels was hanging loose in the frame.




Within the wooden cabinet large cracks occurred.

The door of the cabinet was stuck due to warping.
The lacquer of the wooden cabinet flaked off (in different parts).




Ornaments and decorations showed cracks in relatively thin parts.




The wooden floor of the cabinet showed large cracks.


                                         8-137
Creep, probably in combination with high temperatures and rapid temperature
changes and quality of metal, collapsed the frontal metal pipes.

A result of the damage analysis was the development of an observation chart, which
can be used for an early inventory of damage at monumental organs (RDMZ 2000/1).




                                      8-138
8.1.9   Simulations

Organ stress analysis
The 2D strain and stresses model of section 8.1.9 was used to reproduce the strains
and stresses in a wooden organ pipe and a wooden wind drawer. A block shape
change of a sudden drop in relative humidity was simulated. The drop in RH was
simulated from 85 to 25 % RH. The first simulations took place with a wooden organ
pipe with cross square dimensions of 100 mm * 100 mm * 5 mm. The cross section
was taken at the place of the tuning cap. The wood direction of the tuning cap was
assumed to be perpendicular to the wood direction of the pipe. Due to the drop of
humidity at the outline of the pipe, the wooden pipe will shrink, which is limited by
the tuning cap. Stresses will be the result.
Figure 8-7 and Figure 8-8 show the moisture content of the square section, one hour
after the drop in RH.




Figure 8-7: Moisture content in a wooden organ pipe, four hours after a drop in RH
from 85 to 25 %RH




                                                Organ pipe and tuning cap

                                        8-139
Figure 8-8: Moisture content in a wooden organ pipe, four hours after a drop in RH
from 85 to 25 %RH

Figure 8-9 shows the change of shape of the square section.




Figure 8-9: Change of shape of the square section (relative magnitude changes 10*)


                                        8-140
Figure 8-10 shows the stresses in the x-direction σx.




Figure 8-10: Stresses σx

Figure 8-11 shows a square section of the wind drawer and its deformation. In this
simulation study the effects of a sudden RH drop at the inner and outer surface of the
drawer were calculated. It was assumed that the wind drawer was fastened at the outer
half height of the drawer. The tension in the upper and lower part again was in the
order of magnitude of 107 N/m2 (10 N/mm2).

When we look at the order of magnitude of the stresses in both simulation studies, σ
≈ 107 N/m2 (10 N/mm2), and compare these results with the maximum allowed
stresses of hard wood parallel to the fibers (ft;0;rep=14 N/mm2), and in the direction
perpendicular to the fibers (ft;90;rep = 0.4 N/mm2) cracks will probably occur parallel
to the fibers. Therefore cracks in the wooden organ pipes could be explained by
changes in relative humidity. In the model, however, no relaxation was involved. In
practice the situation might be less critical.




                                         8-141
Figure 8-11: Calculated deformation of a wind drawer.




Figure 8-12: Stresses σx in the x-direction of a wind drawer




                                         8-142
Actual airflow situation
In the next study, airflow simulation results are compared with measurements in the
actual situation, in order to estimate the accuracy of the CFD simulations. The
boundary conditions for the simulation study were extracted from the measured inlet
air conditions (inlet air temperatures and velocities as a function of time and
turbulence intensity) and wall, floor and ceiling surface temperatures as a function of
time. The geometry of the church was simplified to a global 3D model of it. The
following figures indicate the flow pattern during the heating of the church.




Figure 8-13: Heating up the church; interval 5’, θin=70 oC, θi=20 oC, uin=2.4 m/s,
Ar=0.29, Re=2.4*105




                                         8-143
Comparison between simulated and measured results
In Figure 8-14 the measured and calculated temperature stratification is derived.

                                        Temperature stratification                                                  Temperature at
               16                                                                                    70             h i ht 4 9 )
                                 fan:         on                 Tin:        70.3                                                           CFD
                                                                                                                                            measurement
               14
                                                                                                     60

               12

                                                                                                     50
               10                                                                                C]
                                                                                                 o
                                                                                                 te
  height [m]




                                                                                                 mp
               8                                                                                 er 40
                                                                                                 atu
               6                                                                                 re
                                                                               measurement       [ 30
                                                                               CFD
               4                                                               difference

                                                                                                     20
               2
                                 vrijdag 15 januari 1999            1720.01
                                                                 time:
               0                                                                                     10
                    0   5   10          15         20      25           30          35   40               0   0.5     1         1.5     2   2.5       3
                                             temperature [ oC]                                                               time [h]



Figure 8-14: Comparison of measured and calculated results

The actual results measured, compared with the simulation results, show differences in
the curve of stratification. At low and high heights the agreement is rather good, at
half height there is a difference of about 3 K. The trend in time of the measured and
simulated results is reasonable. Absolute differences however vary from 0 to 5 K.

Future situation
To prevent damage in the future, a simulation study was undertaken to show the
effects of different heating strategies. To predict the airflow pattern in the church
under different inlet air temperature and velocity conditions of the church air heating
system, a further CFD simulation study with Fluent took place.
The inlet air conditions have been derived from literature recommendations (Schellen
1998/2): the inlet air temperature was reduced from 70 oC to 40 oC (Table 5-3). Two
variants have been studied: an air velocity of 2.4 m/s and an according airflow of 4200
m3/s (Ar=0.17) and an air velocity of 4.1 m/s and an according airflow of 6700 m3/s
(Ar=0.05). The primary (initial) air temperature was 8 oC.




                                                                                         8-144
Figure 8-15: Heating up the church: interval 5’, θin=40 oC, θi=12 oC, uin=2.4 m/s, Ar=0.17




Figure 8-16: Heating up the church: interval 5’, θin=40 oC, θi=16 oC, uin=4.1 m/s, Ar=0.05




                                        8-145
    Comparison of variants
    In the figures below the calculated air temperatures at two different heights of 4.9 and
    15.2 metres are shown.
            Comparison of air temperatures at a height of 4.9m                        Comparison of air temperatures at a height of 15.2m
          40                                                                         40
                                             CFD basis                                                                  CFD basis
          35                                 CFD variant 1                           35                                 CFD variant 1
                                             CFD variant 2                                                              CFD variant 2
          30                                                                         30

          25                                                                         25
θ [ C]




                                                                            θ [ C]
          20                                                                         20
   o




                                                                               o
          15                                                                         15

          10                                                                         10

           5                                                                             5

           0                                                                             0
                0        0.5          1         1.5         2                                  0       0.5              1            1.5            2
                                    t [h]                                                                             t [h]

    Figure 8-17: Comparison of air temperatures at different heights


    In the following figures the comparison between air velocities for the middle of the
    church and the expected air stratification near the organ are shown.
            Comparison of air velocitiy in the middle of the church                                                                  fc o
                                                                                                             T e m pe ratu re s tra tii ati n
          0.5                                                                                 16
                                                CFD basis
                                                CFD variant 1                                 14
          0.4                                   CFD variant 2
                                                                                              12

                                                                                              10
          0.3
U [m/s]




                                                                                     h [m ]




                                                                                               8

          0.2                                                                                  6

                                                                                               4
          0.1                                                                                                                          C F D basi  s
                                                                                               2                                                  a
                                                                                                                                       C F D va ri n t 1
                                                                                                                                                  a
                                                                                                                                       C F D va ri n t 2
           0                                                                                   0
                0         0.5          1         1.5            2                                  0    10                20             30             40
                                                                                                                           o
                                     t [h]                                                                               θ[ C]

    Figure 8-18: Comparison of air velocities in the middle of the church and comparison
    of expected stratification

    The next figures show the trajectory of the air jet for different numbers of
    Archimedes as they have been calculated from (Katz 1974).




                                                                    8-146
Figure 8-19: Comparison of different Archimedes numbers, resp. Ar=0.29 (left
above), Ar=0.17 (right above), Ar=0.05 (left below), Ar=0.01..0.5 (right below)

8.1.10 Conclusion
In comparison to e.g. the measured air temperature stratification of floor heating, the
temperature stratification with this kind of air heating is very large. Reducing the
Archimedes number can reduce the temperature stratification. This can be accomplished
by reducing the temperature difference between inlet air and indoor air to e.g. 25 K and
to increase the inlet air velocity. The latter is needed to keep up the heating capacity.
To predict the effects of changing air temperature, velocity and the heating rate the
dynamic air heating can be modelled by CFD: Computational Fluent Dynamics.
The hot air heating system in the Walloon Church in Delft led to a large thermal
stratification (up to 15K over the height of the church). Very high air temperatures
(up to 30 to 35 oC) have been measured at the height of the monumental organ. These
were the result of a combination of too high air inlet temperatures and too low air
inlet velocities. Dramatic low relative humidities near the monumental organ resulted
from these conditions. The organ proved to be in a very bad state, due to excessive
shrinkage damage of the wooden parts.
A new air heating system was proposed. The new inlet conditions were based on a
combination of low inlet air temperatures (up to 40 oC) and high air velocities (up to 4
m/s), leading to an Archimedes number less than 0.05. Afterwards the air
stratification proved to be less than 2 K over the total height of the church. Initially


                                         8-147
the air heating system was switched off at a lower limit of the relative humidity near
the organ of 40 %RH. The result of this was that the church could not be used at
outdoor temperatures below about 0 oC. Therefore an air humidifying system was
proposed, simulated and installed. Some condensation on cold surfaces, like glass
surfaces, was predicted.




                                         8-148
8.2   A MODIFIED AIR HEATING SYSTEM IN THE SYPE KERK IN
      LOOSDRECHT


8.2.1 Description and history of the church
The Sypekerk in (Nieuw-)Loosdrecht originally dates from about 1400 and probably
was built first as a baptismal chapel. The tower was built in 1450 with walls of about a
meter thickness. The church has 3 naves. Before the restoration of the church in 1972,
the entrance of the church was at the north and south façade. Nowadays the original
entrance, situated at the west façade of the church, has been restored. The height of
the windows in the choir is larger than the height of the church windows. The choir
itself is a little elevated, compared to the rest of the church. During the restoration a
wooden partition between the choir and the church was removed and the original
space resplendence was recovered.
The original organ dating from 1727 was completely restored during the restoration.




8.2.2 Building data
The outdoor walls of the church are made of masonry and have a mean thickness of
0.85 m. In the church 4 pillars of masonry with a diameter of 1.2 m and a height of 6.5
m are located. The walls on the indoor side are plastered and latex painted. The
glazing in the outdoor walls is made of plain leaded glass. The concrete floor in the
church, with an area of 365 m2, has a floor heating system and is insulated with a 100
mm thick PS insulation, covered by cement. Natural stone and gravestones with a
thickness of 0.06 m cover the floor. The vault is made of boards of wood and has a
lot of cracks. The roof is made of construction wood, covered with slates. The volume
of the church is about 3300 m3.




                                         8-149
8.2.3 Heating system
The original directly heated warm air heating system was replaced by an indirect
heating system, based on an air handling unit. The warm air is brought into the church
by two grilles, located at the choir. A church heating firm modified the grilles for a
better mixing and air distribution in the church. The air handling unit is an Airbox 800
type, manufactured by Rosenberg. The heating is based on a recirculation of 6 times
the volume of the church per hour. The air flow rate can be changed from 2500 to
5000 m3/h. The heating capacity is 60 kW. The recirculated air is filtered in a filter
section. The church has a floor heating system in the middle nave. The floor heating is
divided in 6 groups with a total area of 100 m2. The floor is kept at a constant surface
temperature of about 18 oC.

8.2.4 Formulation of the problem
The quality of the indirectly heated warm air heating system, based on an air handling
unit, was checked considering its effect on the monumental organ and the thermal
comfort in the church.

8.2.5 Measurements
During two days the intermittent heating and cooling down of the church was
monitored extensively. A smoke analysis of the air brought into the church was
performed. Thermal comfort was monitored using a thermal comfort analyser of
Bruel & Kjaer (type 1212). Air temperatures and surface temperatures were monitored
at different places in the church. Dantec anemometers monitored the horizontal air
velocity distribution. An indication of the air infiltration rate was obtained by tracer
gas analysis with SF6 and a Bruel & Kjaer gas monitoring device. The temperature and
relative humidity stratification was monitored over the height of the church by a
Hanwell Architect wireless monitoring system.
After this two-day measuring period, the indoor climate of the church was monitored
for 4 months from February to May 2000. Air temperature and relative humidity were
monitored at different locations in the church in horizontal and vertical sections. The
air inlet temperature of the heating system was monitored too. Furthermore the
surface temperatures of north and south wall, floor and glazing were monitored.

8.2.6   Results

Outdoor conditions
The outdoor conditions were monitored for 4 months and are represented in Figure
8-20.




                                         8-150
                                                                                           outdoor temperature
                                             30


                                             20
                                  θ e [°C]

                                             10


                                               0
                                                40                 60              80            100               120               140                160               180

                                                                                         outdoor relative humidity
                                         100

                                             80

                                             60
                                RH [%]




                                             40

                                             20

                                               0
                                                40                 60              80        100       120         140                                  160               180
                                                                             time [days]: 15-02-2000 is start date (day 46)


Figure 8-20: Outdoor conditions


Indoor conditions
The indoor air temperature first was regulated at 18 oC; after that it was adjusted to 20
oC. Figure 8-21 shows the indoor air temperatures measured and the effects on the

relative humidity at the monumental organ.
                                              air temperature at the organ                                                                           organ conditions
                     30
                                                                                                                   15
      θ organ [°C]




                     20
                                                                                                       % of time




                                                                                                                   10

                     10
                                                                                                                   5


                     0                                                                                             0
                      40   60            80          100        120          140   160     180                          0        5          10           15           20              25        30
                                                                                                                                             air temperature organ [°C]
                                              relative humidity at the organ
                 100                                                                                               20
                     80
                                                                                                                   15
  RH organ [%]




                                                                                                       % of time




                     60
                                                                                                                   10
                     40

                     20                                                                                            5

                     0                                                                                             0
                      40   60         80        100       120         140          160     180                          0   10       20    30         40     50      60     70   80        90   100
                                time [days]: 15-02-2000 is start date (day 46)                                                                  relative humidity organ [%]



Figure 8-21: Indoor conditions at the monumental organ


Heating rate
The heating rate of the heating system was calculated from the indoor air temperature
measured. Figure 8-22 indicates the heating rate for one-hour intervals. The hourly
fluctuations in relative humidity due to the fluctuations in temperature are represented
too.


                                                                                                  8-151
                                                          hourly fluctuations in air temperature at the organ
                                            4

                                            2




                             dθ/dt [K/hr]
                                            0

                                            -2

                                            -4
                                              40    60            80         100        120          140        160     180

                                                        hourly fluctuations in air relative humidity at the organ
                                            10

                                            5
                        dRH/dt [%/hr]




                                            0

                                            -5

                                        -10

                                        -15
                                           40       60            80        100       120         140           160     180
                                                            time [days]: 15-02-2000 is start date (day 46)


Figure 8-22: Hourly fluctuations of air temperature and relative humidity


Thermal stratification
Thermal stratification over the height of the church is indicated at different times in
Figure 8-23.
                                                        Temperature stratification in time
                              11
                                                                                                                      t = 0 hr
                              10                                                                                      t = 4 hr
                                                                                                                      t = 8 rr
                                        9                                                                             t = 12 hr
                                                                                                                      tmax = 29 hr
                                        8

                                        7
               Height [m]




                                        6

                                        5

                                        4

                                        3

                                        2

                                        1

                                        0
                                         10        12         14      16        18              20         22
                                                              Air temperature [°C]

Figure 8-23: Temperature stratification as a function of time




                                                                             8-152
Air velocities
The air velocity measured as a function of time is represented in Figure 8-24.
                                            air velocity about 1 m from inlet                                                       air velocity in middle of church
                    0.4                                                                                         0.4


            0.35                                                                                               0.35


                    0.3                                                                                         0.3


            0.25                                                                                               0.25
 v [m/s]




                                                                                                     v [m/s]
                    0.2                                                                                         0.2


            0.15                                                                                               0.15


                    0.1                                                                                         0.1


            0.05                                                                                               0.05


                     0                                                                                           0
                          0   0.5     1      1.5        2        2.5     3      3.5     4   4.5                       0   0.5   1   1.5        2        2.5     3      3.5   4   4.5
                                             time [hr], start at 16-02-2000                                                         time [hr], start at 16-02-2000



Figure 8-24: Air velocities at living height near the inlet (left) and in the middle of the
church (right)


Air inlet temperature
                                                 air inlet temperature
                    60



                    50



                    40
     θ inlet [°C]




                    30



                    20



                    10



                     0
                      40        60         80        100       120         140        160   180
                                     time [days]: 15-02-2000 is start date (day 46)



Figure 8-25: Air temperatures at the inlet




                                                                                                  8-153
Thermal comfort
The obtained thermal comfort was calculated according ISO_7730, 1984 for
combinations of air velocities measured and expected clothing levels.
                                                  Predicted Mean Vote

                                                                                    v=0.05,clo=1
                                                                                    v=0.05,clo=1.5
                           0.5                                                      v=0.05,clo=2
                                                                                    v=0.1,clo=1
                                                                                    v=0.1,clo=1.5
                                                                                    v=0.1,clo=2

                           -0.5
                 PMV [-]




                                  0   5        10             15           20          25            30
                                          time [hr], start at 15:06 hr 15-02-2000




Figure 8-26: Calculated predicted mean vote for different air velocities and clothing
values


Air infiltration rate
The air infiltration rate as calculated from the tracer gas measurements turned out to
be in the order of magnitude of 0.8 to 1.0 h-1.

8.2.7 Conclusions
In the Sype Kerk in Loosdrecht the existing, modified, air-heating system nearly
fulfilled its requirements. Modification of the air inlet grilles apparently led to high air
induction at these grilles and air stratification measurements showed that the air
temperature differences over the total height of the church were less than 2 K. The
Archimedes number proved to be in the order of magnitude of 0.05.
The heating rate was controlled to a value of less than 2 K/h. A primary temperature
of 10 oC was established. Furthermore the heating system was controlled by a
hygrostatic device, limiting relative humidity to an upper limit of 80 %RH. However,
there was no lower limit control. Therefore the relative humidity within the church
reached values of less than 45 %RH during about 10 % of the measuring period of
over six months. Relative humidity values dropping down to 35 %RH were measured.
It was advised to lower air temperatures in the church during low outdoor
temperatures from 20 oC to 18 oC and to limit lowest relative humidities by a
hygrostatic device to 40 %RH.



                                                       8-154
8.3   A NEW AIR HEATING SYSTEM IN ST. LIDUÏNA’S BASILICA IN
      SCHIEDAM

8.3.1 Description and history of the church
In 1969 there were two Catholic churches in Schiedam: the parish church of St. Jan de
Doper (St. John the Baptist) at the Lange Haven from 1882 and the parish church of
Onze Lieve Vrouw Visitatie at the Nieuwe Haven, dating from 1869. This church was
renamed to St. Liduïna and and its original name went to a new church in the western
part of Schiedam. Because of high maintenance costs St. Liduïna’s church was
demolished in 1969. The name St. Liduïna was given to the church at the Singel of
Schiedam, St. Liduïna’s Basilica or Onze Lieve Vrouw Rozenkrans.
St. Liduïna’s Basilica is a neogothic cross basilica. The nave has 12 small and 4 large
pillars of polished marble from the cave of Belvoye. It has a vault of brick. The nave is
lightened by a number of sharp arched windows with traceries of natural stone. The
chancel is a polygon and the cross part of the church has circular windows. The slim
tower is in front of the nave and has a height of 60 meters. The building has a length
of 51 meters, a width of 31 meters and the church has a height of 22 meters.




8.3.2 Building data
The outdoor walls of the church are made of masonry and natural stone and have a
mean thickness of 0.85 m. In the church 12 pillars of marble with a diameter of 0.8 m
and a height of 5 m and 4 pillars with a diameter of 1.0 m and a height of 7.7 m are


                                         8-155
located. The walls on the indoor side are not finished with plaster. The glazing in the
outdoor walls is made of stained glass with external protective glazing. The floor in
the church has an area of 850 m2 and is made of natural stone with a thickness of 0.06
meters. The vault is made of masonry. The roof is made of construction wood,
covered with slates. The volume of the church is about 10,000 m3.

8.3.3 Heating system
The heating system consists of a warm air heating system with a heating capacity of
300 kW. The air is brought into the church by a central grille of 2 by 0.75 m2 at a
height of about 3 meters and is extracted by grilles at floor level near the outer walls.
The total number of extraction grilles is about 25.

8.3.4 Formulation of the problem
As part of a restoration of the church the above-mentioned new warm air heating
system replaced the original warm air heating system. The monumental organ was
being restored and the climatic conservation conditions should therefore be safe
before re-installing the organ.

8.3.5 Analysis
In the design of this heating system in the St. Liduïna Basilica it was assumed a most
elementary design mistake has been made. It was shown before (Fitzner 1996) that the
airflow in a room is fully determined by the spatial distribution and number of the air
inlet grilles and is hardly influenced by the number and place of the air extraction
grilles, as was shown before in section 5.2.8. Therefore the number of air inlet grilles is
important for a homogeneous air distribution, not the number of air extraction grilles.
In St. Liduïna’s Basilica the number of air inlet grilles turned out to be 1, the number
of air extraction grilles was about 25, a waste of money as was assumed. An airflow
simulation study was performed to prove this assumption.
Furthermore the installed warm air heating system was assumed to behave like the
original air heating system in the Walloon Church in Delft.

8.3.6   Measurements

Outdoor conditions
The outdoor conditions were monitored for about 1 year. Furthermore meteorological
data from the airport of Rotterdam were obtained from KNMI.

Indoor conditions
The indoor air temperature during services was regulated at 18 oC. Figure 8-27 shows
the indoor air temperatures measured and the effects on the relative humidity at the
monumental organ and at the position of the thermostatic device.




                                          8-156
                                             air temperature and relative humidity at organ and thermostatic device
                               30
                                                                                                             Organ
                               25                                                                            Thermostat




                 θ air [°C]
                               20

                               15

                               10

                                       5
                                       -20         0         20         40       60       80         100       120        140
                                                                     time [days] 0=>1 jan 2000
                               80
                                                                                                             Organ
                               60                                                                            Thermostat
                 RH [%]




                               40

                               20

                                       0
                                       -20         0         20         40       60       80         100       120        140
                                                                     time [days] 0=>1 jan 2000


Figure 8-27: Indoor conditions at the monumental organ and the thermostatic device


Heating rate
The heating rate of the heating system was calculated from the indoor air temperature
measured. Figure 8-28 indicates the heating rate for one-hour intervals. The hourly
fluctuations in relative humidity due to the fluctuations in temperature are represented
too. In Figure 8-29 the daily fluctuations in temperature and relative humidity are
shown.
                                                           hourly fluctuations in air temperature at the organ
                                        10


                                         5
                         dθ/dt [K/hr]




                                         0


                                        -5
                                         -20           0        20        40        60        80       100         120     140

                                                           hourly fluctuations in relative humidity at the organ
                                        10

                                         5
                       dRH/dt [%/hr]




                                         0

                                        -5

                                       -10
                                         -20           0        20        40       60       80         100         120     140
                                                                       time [days] 0=>1 jan 2000


Figure 8-28: Hourly fluctuations of air temperature and relative humidity




                                                                            8-157
                                                                                  daily fluctuations in air temperature at the organ
                                                                15




                                               dT/dt [K/day]
                                                                10


                                                                 5


                                                                 0
                                                                 -20          0           20           40                60           80         100              120                140

                                                                                  daily fluctuations in relative humidity at the organ
                                                                40

                                                                30
                                               dRH/dt [%/day]




                                                                20

                                                                10

                                                                 0
                                                                 -20          0           20         40       60       80                        100              120                140
                                                                                                  time [days] 0=>1 jan 2000




Figure 8-29: Daily fluctuations of air temperature and relative humidity


Air inlet conditions
The air velocity and air temperature measured at the inlet as a function of time is
represented in Figure 8-30
                                  air velocity at the inlet (TI=0.11 to 0.17)                                                                          Inlet air temperature
                   6                                                                                                     80


                                                                                                                         70
                   5

                                                                                                                         60

                   4
                                                                                                                         50
   v inlet [m/s]




                                                                                                             θ in [°C]




                   3                                                                                                     40


                                                                                                                         30
                   2

                                                                                                                         20

                   1
                                                                                                                         10


                   0                                                                                                     0
                       0   2000        4000         6000               8000       10000        12000                          338   340    342   344       346     348         350    352   354   356
                                               time [seconds]                                                                                              time [days]




Figure 8-30: Air velocity (left) and air temperature at the inlet (right)


Air velocities
In the figures below the air velocities at four different measurement points in the
church is represented

Air infiltration rate
The air infiltration rate as calculated from the tracer gas measurements turned out to
be in the order of magnitude of 0.5 to 0.6 h-1.



                                                                                                            8-158
                              air velocity at position 2                                                   air velocity at position 8
            1                                                                            1

           0.8                                                                          0.8

           0.6                                                                          0.6
v2 [m/s]




                                                                             v8 [m/s]
           0.4                                                                          0.4

           0.2                                                                          0.2

            0                                                                            0
                 0   2000   4000         6000         8000   10000   12000                    0   2000   4000         6000         8000   10000   12000
                                   time [seconds]                                                               time [seconds]
                              air velocity at position 7                                                   air velocity at position 4
            1                                                                            1

           0.8                                                                          0.8

           0.6                                                                          0.6
v7 [m/s]




                                                                             v4 [m/s]
           0.4                                                                          0.4

           0.2                                                                          0.2

            0                                                                            0
                 0   2000   4000        6000          8000   10000   12000                    0   2000   4000        6000          8000   10000   12000
                                   time [seconds]                                                               time [seconds]



       Figure 8-31: Air velocities at positions 2,4,7 and 8 (Schellen 2001/6)


       8.3.7 Simulations
       The results of the measurements showed that re-installing the monumental organ
       under the current indoor climate conditions could cause severe damage. Relative
       humidity at the monumental organ dropped to severe values of 20 %RH. For that
       reason a simulation study with Fluent was done to indicate a possible solution. Four
       different simulation variants were studied:
       the present situation with measured inlet air conditions (θin,max=70 oC)
       the present situation with limited inlet air temperatures to θin,max <=45 oC and adjusted
       air inlet rates to equal the heating capacity
       exchange of air inlet and air extraction grilles to improve air distribution
       variant 2 with inlet grille at floor level

       Present situation with inlet air conditions measured (θin,max=70 oC)




       Figure 8-32: Present situation with inlet air conditions measured (θin,max=70 oC); air
       temperature in 2 vertical sections (left), temperature stratification in front of the organ
       (right)




                                                                      8-159
Present situation with limited inlet air temperatures to θin,max <=45 oC




Figure 8-33: Present situation with limited inlet air temperatures to θin,max <=45 oC; air
temperature in 2 vertical sections (left), temperature stratification in front of the organ
(right)


Exchange of air inlet and air extraction grilles




Figure 8-34: Exchange of air inlet and air extraction grilles; air temperature in 2
vertical sections (left), temperature stratification in front of the organ (right)


Inlet grille at the floor level




Figure 8-35: Inlet grille at the floor level; ; air temperature in 2 vertical sections (left),
temperature stratification in front of the organ (right)


                                             8-160
Results
In this church the current air inlet conditions of one air inlet grille in combination
with high inlet air temperatures of 70 oC and air inlet velocities of 2 m/s (low air flow
rate) resulted in an Ar number of about Ar=0.4 (variant 1). An upward directed air
flow was the result of it, which led to a rather strong air temperature stratification in
front of the organ of about 5 to 6 oC over the height of the church. At the wall above
the inlet grille, where panel paintings of the legend of St. Liduïna are hanging, air
temperatures of about 25 oC were expected. The air velocities turned out to vary from
10 (in the middle of the church) to 20 cm/s (near the inlet) on living height, slightly
beyond comfort level.
Increasing the air flow rate to 4 to 5 m/s (high air flow rate) in combination with a
decrease of the inlet air temperature to 45 oC led to an Ar number of Ar=0.07 (variant
2). The result of it was a slight improvement of the air stratification in front of the
organ to about 3 to 4 oC over the height of the church. Wall temperatures above the
air inlet grille were reduced to about 23 oC. Air velocities in the living zone however
were increased to about 30 cm/s, which is beyond the comfort range.
The simulation study of the exchange of air inlet grille to air extraction grille, in
combination with the exchange of air extraction grilles to air inlet grilles (variant 3)
led, as expected, to well distributed air temperatures and velocities in the church, with
temperature stratification as low as 1 to 2 oC over the height of the church. In front of
the wall paintings, air temperatures of about 20 oC were expected. Air velocities were
reduced to about 10 cm/s in the living zone.
Replacing the single air inlet from wall to floor (variant 4) did not improve the
expected temperature stratification, nor did it improve the conditions for the panel
paintings.

8.3.8 Conclusions
The new air heating system in St. Liduïna’s Basilica did not perform well. The air
temperatures were controlled to 20 oC at living height, but the air stratification turned
out to be larger than 5 K over the height of the church. During low re-circulation
operating rate, measurement of the air inlet conditions showed a combination of too
high inlet air temperatures (up to 70 oC) and lower velocities (up to 2 m/s), resulting
in Archimedes numbers of up to 0.4. The air humidity conditions measured at the
organ were dramatic: low RH values of 20 %RH were reached. A higher velocity rate
improved the inlet air conditions to Archimedes numbers down to 0.07, but also
resulted in too high air velocity values (up to 30 cm/s) in the comfort zone.
A CFD simulation study showed that it was difficult to prevent air stratification in this
church using a single central hot air inlet: the geometry of the church was complex
and due to the large volume and heat loss area a rather high heating capacity was
needed. Furthermore it was concluded that the large number of air outlet grilles was a
waste of money; a single outlet would probably have led to the same result. A well-
distributed air inlet system, combined with one outlet would improve the temperature
and humidity climate considerably: temperature stratification in this variant was




                                         8-161
expected to be less than 2 K over the height of the church. Exchanging of the air inlet
and outlet system would probably improve the system to a large extent.




                                         8-162
8.4 CONTAMINATION OF THE CEILING IN ST. MARTINUS’ CHURCH IN
WEERT

8.4.1 Description and history of the church
St. Martinus’ church was built in a late gothic style. Together with St. Michael’s church
in Zwolle, it is the only church in the Netherlands with three naves of the same
height, each having its own roof.
The eastern part of the church was constructed in 1456. The western part dates from
1500. In the horizontal section the difference of the supporting buttress can be seen.
Most gothic churches, dating from the Middle Ages, have ceiling paintings. They were
often plastered or painted after a plague. At the major restoration of the church in
1975 the ceiling paintings were discovered and restored too.




8.4.2 Building data
The outdoor walls of the church are made of masonry and have a mean thickness of
1.1 m. The indoor sides of the walls are plastered and whitewashed. In the church 16
pillars of natural stone are located. The glazing in the outdoor walls is made of stained
glass. In 1997 the stained glass was protected from the outside with protective glazing.
The concrete floor in the church, with an area of 1260 m2, has a floor heating system
and is insulated with a 50 mm thick PS insulation, covered by cement. Natural stone
and gravestones, with a thickness of about 0.05 m, cover the floor. The vault is made
of masonry bricks and has a thickness of about 300 mm. It is plastered from both
sides. The indoor side is painted. The roof is made of construction wood, covered
with 5 mm slates. The volume of the church is about 18,850 m3.

8.4.3 Heating system
The heating system of St. Martinus’ church consists of a central heating boiler, with a
heat capacity of 127 kW. The floor heating pipes have been divided into three groups
and each group has a water temperature sensor in the floor, limiting the water


                                         8-163
temperature to a maximum of 55 oC. Originally the church was heated to a level of
about 20 oC. Since the beginning of our research, the system has been controlled
electronically. A thermostatic device connected to the floor heating, stationary
controls the air temperature in the church to 15 oC.

8.4.4 Formulation of the problem
The church was restored in the years 1974-1980 and during this time beautiful
medieval vault paintings were detected and restored. Originally an air heating system
heated the St. Martinus Church in Weert. In 1984 a floor heating system replaced this
former heating system. After this major renovation a number of problems arose:
Within a few years after this main restoration the ceiling rapidly contaminated. Due to
the floor heating system and massive granite floor the church could only be heated
stationary. This resulted in a relatively large thermal driven vertical airflow, causing
thermal discomfort. The air temperature was therefore raised to overcome this
discomfort, thus causing the 16th century painted ceiling to contaminate more rapidly.
Furthermore due to the high floor and air temperatures, the energy costs were
substantial.

8.4.5 Analysis
From literature (Schellen 1998/2) it was known that floor heating may lead to
thermally driven air-flows. When a contamination source is present, there is a high risk
of contamination of surfaces like ceiling and walls. Furthermore the infiltration rate is
important for the value of the contaminant concentration. Therefore a research was
started to investigate the sources of contamination, the infiltration rate and the effects
of floor heating on the thermal driven airflows (Stevens, 1999).
The contamination on the ceiling was expected to be the result of the soot deposition,
originating from burning candles and incense, rather than the result of fungi
contamination. The presupposition in the research was that heating the floor, with an
increasing temperature difference between floor surface and air, led to an increased
thermally driven airflow. Furthermore this increased airflow was expected to lead to
an increased deposition and to thermal discomfort.
The stationary heating of the church was the other focus of attention. One of the
questions was how much time it would take to heat the church with the existing floor
heating system. Was it possible to develop some heating strategy to improve the
interior climate conditions as well as to diminish the energy consumption, without loss
of thermal comfort?

8.4.6 Measurements
During two years the following long-term measurements were performed.
Measurements with combined air temperature and relative humidity sensors were
taken each half hour at the monumental pulpit, organ, in the attic above the vault and
outdoors. With the same time-interval surface temperatures of walls, floor and ceiling
were measured during this period. Furthermore, with an interval of 5 minutes,
measurements of surface temperatures of heated floor, water temperatures and air-



                                          8-164
velocities were made. Apart from these measurements some short-time measurements
were performed. Infrared surface thermographic images were taken outdoors and
indoors of ceiling, wall surfaces and floor surface. The infiltration rate was measured
and a chemical analysis of the indoor air and of the deposition on the surface of the
ceiling took place.

8.4.7 Simulations
WaVo (Wit 2000) was used to predict the effects of some measures: the installation of
protective glazing and the reduction of the air temperature and their effects on energy
savings and humidity conditions near the monumental organ and the painted ceilings.
First, heating up the floor was two-dimensionally simulated by the use of PDEASE
(Backstrom 1994). The calculated surface heat flow was then used in the thermal
WaVo network and was introduced as a one-dimensional heat flow in the network of
the floor.

8.4.8   Results

Floor heating system
In Figure 8-36 a comparison of the measured and calculated dynamic response of the
floor heating is shown. From the figure one can see that it takes about three days to
heat the floor to a stationary level.

                                                  40



                                                  35
             water-en floor temperatuur [deg C]




                                                  30



                                                  25



                                                  20


                                                                                                    Twa-meas
                                                  15
                                                                                                    Twa-sim
                                                                                                    Tsurf-sim
                                                                                                    Tsurf-meas
                                                  10
                                                   -10   0   10   20      30      40      50   60        70      80
                                                                  hours from 2-12-98 10:10h


Figure 8-36: Comparison of FlexPDE calculated and measured temperatures of
floor heating water and floor surface temperatures




                                                                         8-165
Air temperature and relative humidity
The long-term behavior of measured and calculated air temperatures and relative
humidities in the church is shown in Figure 8-37. The agreement of both air
temperature and relative humidity is rather good, except for the hours from about
5000 to 5500. In reality the gas burning device broke down, and this was not
accounted for in the simulation.


                             30

                             25
              Tair [deg C]




                             20

                             15

                             10

                              5
                                   0   1000   2000     3000   4000   5000          6000


                              1

                             0.8

                             0.6
             RH [-]




                             0.4

                             0.2                                            Meas
                                                                            Sim
                              0
                                   0   1000   2000    3000    4000   5000          6000
                                                      hours


Figure 8-37: The long term (one-year) simulated and measured air temperature and
humidity in the St. Martinus Church in Weert.


Climatic conditions near the monumental interior
The climatic conditions near the pulpit and the monumental organ were measured.
Figure 8-38 shows bars of air temperatures and relative humidities measured, for the
heating and non-heating period, during a year. From the figures left it is clear that
during the heating period the air temperature is mostly in the range of 14 to 15 oC.
The electronic thermostatic floor heating control functions rather well. From the
figures left it is to be seen that the relative humidity during the heating period most of
the time remains between 40 to 70 %RH. The results are within the advised intervals
of Table 1-1.




                                                     8-166
                                                        data at organ: heating period                                              data at organ: non-heating period
                                60                                                                                       60

                                50                                                                                       50

                                40                                                                                       40
              % of time




                                                                                                             % of time
                                30                                                                                       30

                                20                                                                                       20


                                10                                                                                       10

                                   0                                                                                     0
                                       5           10           15        20             25     30                            5   10         15         20         25         30
                                                            airtemperature [oC]                                                          air temperature [oC]


                                                        data at organ: heating period                                              data at organ: non-heating period
                                40                                                                                       40

                                35                                                                                       35

                                30                                                                                       30

                                25                                                                                       25
              % of time




                                                                                                             % of time
                                20                                                                                       20

                                15                                                                                       15

                                10                                                                                       10

                                   5                                                                                     5

                                   0                                                                                     0
                                       0           20          40            60          80     100                           0   20        40            60       80         100
                                                                    RH [%]                                                                       RH [%]




Figure 8-38: Climatic conditions near the organ during heating (left) and non-heating
period (right)

                                                    air temperature, surface temperature floor and air velocity at pulpit
                                       30

                                       25
                temperature [oC]




                                                        **********************                ***
                                       20

                                       15

                                       10                                                     Tfloor
                                                                                              Tair
                                           5
                                               0          50            100              150    200     250      300                        350            400          450
                                                                                        number of days after 25-11-97
                                   0.6
           air velocity [m/s]




                                                                                                      non heating period
                                   0.4


                                   0.2
                                                                             ***              ***

                                           0
                                               0          50            100              150    200     250      300                        350            400          450
                                                                                        number of days after 25-11-97


Figure 8-39: Floor surface temperature, air temperature and air velocity during 18
months of measurements (*** indicates the floor temperature sensor to be unattached
to the floor)




                                                                                                     8-167
Air velocities
Figure 8-39 shows measured surface and air temperatures in relation to measured air
velocities. It can be clearly seen that the air velocity has a relation with the heating
period. From these measurements the air velocity is derived, as a function of the
difference between floor surface and air temperature, and is represented in Figure
8-40 (left). There is a clear indication that the air velocity increases with an increased
temperature difference. In the right figure, the velocities measured were divided in
classes of temperature differences. The mean values of the velocities were thus
represented in their temperature difference classes. The result is a clear velocity
increase, as a function of temperature difference.

                                                                             air velocity as a function of temperature difference Tfloor-Tair
                                                                    0.19

                                                                    0.18

                                                                    0.17

                                                                    0.16
                                               air velocity [m/s]


                                                                    0.15

                                                                    0.14

                                                                    0.13

                                                                    0.12

                                                                    0.11

                                                                     0.1

                                                                    0.09
                                                                        -2       0             2             4            6            8        10
                                                                                                     Tfloor-Tair [oC]



Figure 8-40: Air velocity as a function of temperature difference between surface floor
and air; measured values (left), mean values (right)


Ventilation and infiltration rate
The natural ventilation and infiltration rate was measured in St. Martinus’ church,
according to section 3.2.3. These values were determined to be very small (n=0.08 h-
1), when the main portal doors were closed, and on average (n=0.33 h-1), when these

doors were open.

Use of energy
The annual use of heating energy in St. Martinus’ church was recorded for some years.
WaVo predicted the energy effects of the introduction of the new temperature setting
and the installation of protective glazing within an accuracy of 5 to 10 percent, as
compared to the calculated energy savings by the gas bills (35 %). In Table 8-1, the
independently predicted results of some of the potential measures are given. The
installation of protective glazing by itself has a lowering effect of about 15 % on the
energy costs. Independently lowering the air temperature from the original 20 oC, to
the current value of 15 oC, had a predicted energy effect of about 41 %. Thermal
insulation of the vault would have an energy saving effect of about 29 %. Further
research takes place to look at the building physical effects of insulation of the vault
(Martens 2001).


                                           8-168
 Variables                            Simulated gas    Simulated gas          Percentage
                                      consumption      consumption            [%]
                                      [m3]             with correction [m3]
 With protective glazing (*)          31,700           28,700                 100
 Without protective glazing           37,000           33,500                 117
 Infiltration rate n=0.15 all day (*) 31,700           28,700                 100
 Infiltration rate n=1 during day 76,900               69,600                 243
 Infiltration rate n=1 during         36,500           33,100                 115
 afternoon
 Air temperature 15 oC (*)            31,700           28,700                 100
 Air temperature 17.5 oC              41,700           37,700                 131
 Air temperature 20 oC                54,000           48,800                 170
 Without insulating vault (*)         31,700           28,700                 100
 With insulation vault                22,700           20,500                 71
Table 8-1: Predicted simulation results of a floor heating system (*) Current situation


Contamination of the ceiling

Source of contamination
In a study in Geneva (Huynh 1991) an analysis of dust led to the conclusion that the
main contamination in that church was due to burning candles and incense. From the
bills of purchase of candles and incense in Weert, it is known that a large number of
candles and a lot of incense have been burned over the last 10 years. To determine the
source of contamination unambiguously, a chemical analysis of the air and dust in St.
Martinus’ church was undertaken (Stevens 1999). During a service the air in the
church was sampled by absorption of PAH’s (polycyclic aromatic hydrocarbons) in
glass tubes on a polymer. A gas chromatic analysis of the samples did not show an
increased PAH concentration, in comparison with air samples taken from a laboratory
setting. Therefore an additional dust analysis was performed. The concentration of
PAH’s in dust was relatively high and comparable with the concentrations as
measured in Geneva. From the analysis it was concluded that soot of candles and
incense was the main source of the contamination on the ceiling.

Results of measurement of soot production
The soot production was measured in the glass tube soot-measuring device of section
3.3.1 (Nijland, 1999). The figure below gives a qualitative impression of the
comparison of soot production of the different candles used. The figure shows the
contamination of the glass fibre filters, due to the burning of the candles, represented
above them.




                                             8-169
  Figure 8-41: Qualitative comparison of the contamination of the glass fibre filters in
  the experiments; the figure shows the contamination on burning the shown candles


                                                                                                              production in mg soot for each burning hour
                               burning hours each candle

                                                                                           20,0                      17,5
250                                                                                        18,0
                                                                                216
                                                                                                      15,2
                                                                                           16,0
200
                                                                                           14,0
                                                                                           12,0
150
                                                                                           10,0
100                                                                                         8,0
                              52                               52                           6,0
 50                                                                                         4,0
                                              18                                                                                     0,8             0,5
             2                                                                              2,0                                                                      0,1
  0                                                                                         0,0
       offering candle   altar candle    offering light   offering light   novene candle          offering candle altar candle   offering light   offering light novene candle
                                             small            large                                                                  small            large

                 production in mg soot by each candle (mean values)


1200                         1097

1000
 800

 600

 400
 200
             30                                2               11               15
   0
       offering candle    altar candle   offering light   offering light   novene candle
                                             small            large




  Figure 8-42: Quantitative comparison of the candle burning time (left above), soot
  production per hour (right above) and soot production per candle (left below)




                                                                                      8-170
From the figures above it can be seen that the original offering candles and altar
candles are the main polluting candles. The offering candles have been substituted by
offering lights.

8.4.9 Conclusions
At the start of the research in the Martinus Kerk in Weert the control of the floor
heating system was improved and air temperatures were reduced from 20 oC to 15 oC.
At the same time the stained glass windows got externally ventilated protective glazing
on the outside.
A study of relative humidity at the ceiling showed no critical surface humidities: the
contamination of the ceiling could not be the result of fungal defacement on the
surface.
Measurements proved a clear relation between floor temperatures, air temperatures
reached and resulting airflows. The contamination of the ceiling was clearly a result of
thermally induced vertical airflows and contamination due to soot production by
candle and incense burning. Due to the impuls of the vertical airflow and the
temperature difference between air and surface of the ceiling, inertial and
thermophoretic deposition were expected to be the deposition mechanisms (Camuffo
1995). A soot production research of the candles and incense used showed wax lights
to produce only a fraction of soot, as compared with candle burning. Therefore
reduction of the airflows, increasing of the ceiling surface temperature, substitution of
the candles by wax lights and improvement of the ventilation rate would reduce the
soot contamination to a fraction of the original.
Furthermore the implemented measures of temperature reduction and the mounting
of protective glazing led to energy loss reductions of up to 35 % as compared to, and
corrected for, earlier years. These values were taken from the gas bills.
The air conditions near the monumental organ most of the time (> 95%) satisfied the
recommendations from Table 5-4.
The accuracy of the results of the WaVo simulations were within 5 to 10 % of the
results of the measurements.
Infrared pictures of walls, ceiling and interior parts showed that the radiant effect of
the floor heating led to increased wall and ceiling temperatures. Wall temperatures
decreased with height. Pews intercepted this radiation
.




                                         8-171
8.5    CONTAMINATED FRESCOES IN ST. GERLACHUS’ CHURCH IN
       HOUTHEM

8.5.1 Description and history of the church
St. Gerlachus’ church in Houthem was built around 1720. Johan Adam Schöpf
painted the walls and ceiling in fresco technique in 1751. It is the only church in the
Netherlands with original frescoes. During a major restoration in 1971, the frescoes
were restored.
The Binvignat organ dates from 1784 and was restored in 1989. Another important
monumental interior part is the tomb of St. Gerlachus.




8.5.2 Building data
The outdoor walls of the church are made of masonry and marl and have a mean
thickness of 1.3 m. The indoor sides of the walls are plastered and fresco painted. The
glazing in the outdoor walls is made of stained glass and has an outdoor protective
glazing. The concrete floor in the church with an area of 475 m2, partially has a floor
heating system and is insulated. Natural stone covers the floor. The vault is made of
plastered twigs and is fresco painted. The roof is made of construction wood, covered
with slates. The volume of the church is about 6,680 m3.

8.5.3 Heating system
St. Gerlachus’ Church has a floor heating system, combined with pew heating. The
heating system has a heating capacity of 224 kW. The floor heating has a water
temperature of about 30 oC, the pew heating is driven with a water temperature of
about 50 oC.

8.5.4 Formulation of the problem
During the last restoration of this church in 1971, original frescoes of Johan Adam
Schöpf were uncovered and restored. Apart from that, historical recondition work on
the paintings of Hermans from 1808 became visible again. The organ of Binvignat


                                          8-172
from 1784 was restored too. During this major restoration a floor heating system was
installed, together with a pew heating system, consisting of convector heating under
the seats. To protect the stained glass, to improve thermal comfort and to reduce
energy losses, the windows got externally ventilated, protective glazing at the outside.
After this major renovation and restoration the fresco vaults were contaminated in
short time. Around 1800 a part of the vault crashed down. It was restored afterwards.
Nowadays the break between the original and restored part of the vault is still visible.

8.5.5 Analysis
From a scaffold it was concluded that the contamination consisted of dirt and
probably soot deposition. Furthermore a discolored pattern of wooden beam of the
vault construction was to be seen in the contamination pattern. From literature
thermal deposition due to thermophoreses, together with thermal air flows due to
floor and pew heating were expected to be the most important causes of
contamination (Schellen 1998/2).

8.5.6 Measurements
The measurements in the church were aimed at recording the in- and outdoor climate
conditions and determination of the vault surface temperature and near surface
humidity conditions (Schellen, 2001/5). Air temperature and relative humidity
measurements were performed at the chancel, near the tomb of St. Gerlach from
1785, at the organ and in the attic above the vault construction, under the roof
construction. To get an indication of the surface temperature distribution of the
ceiling, thermal infrared surface imaging was performed. Surface temperature sensors
were used to monitor the thermal conditions of the vault, as a function of time,
underneath the ceiling, at different light and dark discolored contamination sections.
Corresponding surface measurements were performed above the vault construction in
the attic. To determine the thermal resistance of the vault construction, heat flow
transducers on the vault construction monitored the heat flux passing it. The floor-
and pew heating was monitored by floor and convector surface temperature
measurements. To correlate floor and pew heating with thermally induced airflows, air
velocity measurements were carried out in two places. To get an indication of
infiltration and ventilation, two infiltration rate measurements took place. Apart from
that, a chemical analysis of the dirt contamination on the ceiling was made and soot
production measurements of the wax lights were carried through.

8.5.7   Results

Surface temperatures floor and pew heating system
The surface temperatures of the floor and the pew heating convectors are represented
in Figure 8-43. The water temperature of the pew heating convectors turned out to be
controlled at 40 oC. The controlling of the floor heating resulted in a floor surface
temperature of about 18 oC.



                                         8-173
                                              surface temperature bench heating = f(time)
                             50
                                                                                                                           8
  surface temperature [°C]




                             40
                                                                                                                           6




                                                                                                               % of time
                             30                                                                                            4

                             20                                                                                            2

                             10                                                                                            0
                              280   300      320     340      360      380      400      420   440   460                    10      15            20       25       30      35        40              45    50
                                                                                                                                                  surface temperature bench heating [°C]
                                               surface temperature floor heating = f(time)
                             20
  surface temperature [°C]




                             18                                                                                        30




                                                                                                           % of time
                             16                                                                                        20
                             14
                                                                                                                       10
                             12

                             10                                                                                            0
                              280   300       320      340     360      380   400     420     440    460                       0         5              10          15            20             25         30
                                          time [days]: 24-10-2000 start measurement (day 298)                                                      surface temperature floor heating [°C]



Figure 8-43: Surface temperatures of pew heating convector and floor


Indoor air temperature and relative humidity
The indoor air temperatures and relative humidities were measured in three places in
the church: at the altar, at the tomb of St. Gerlachus and near the monumental organ.
Figure 8-44 shows the air temperature and humidity measurements near the altar.
                                                     altar air temperature = f(time)
                             20
                                                                                                                       30
  air temperature [°C]




                             15
                                                                                                           % of time




                                                                                                                       20


                             10                                                                                        10


                              5                                                                                            0
                              280   300       320     340      360      380      400     420   440   460                       0         5              10            15          20             25         30
                                                                                                                                                         altar air temperature [°C]
                                                     altar relative humidity = f(time)
                             80                                                                                        25
  relative humidity [%]




                             70                                                                                        20
                                                                                                           % of time




                             60                                                                                        15

                             50                                                                                        10

                             40                                                                                            5

                             30                                                                                            0
                              280   300       320      340     360      380   400     420     440    460                       0   10        20      30       40        50     60     70    80         90   100
                                          time [days]: 24-10-2000 start measurement (day 298)                                                             altar relative humidity [%]




Figure 8-44: Air temperature and relative humidity at altar


Climatic conditions near the monumental organ
The climatic conditions near the organ did not differ much from the conditions near
the altar. The temperature differences between monumental organ and at altar level
did not differ more than 1 to 2 K. In Figure 8-45 the results of these measurements
are shown. As was expected from this kind of heating, there turned out to be no
important temperature stratification.




                                                                                                       8-174
                                                                       organ air temperature =
         20                                                            f(ti )                                                                30
     air
     te
     mp 15                                                                                                                                   20




                                                                                                                                 % of time
     er
     atu
     re 10                                                                                                                                   10
     [°C
     ]
          5                                                                                                                                     0
          280                                         300      320      340     360      380      400     420       440   460                       0             5              10           15           20              25           30
                                                                                                                                                                                  organ air temperature [°C]
                                                                      organ relative humidity =
                                       80                             f(ti )
     rel                                                                                                                                     20
     ati                               70
     ve                                                                                                                                      15




                                                                                                                                 % of time
     hu                                60
     mi                                                                                                                                      10
                                       50
     dit
     y                                 40                                                                                                       5
     [%
                                       30                                                                                                       0
                                        280           300       320      340    360    380     400    420           440   460                       0    10           20      30       40       50     60      70     80        90      100
                                                            time [days]: 24-10-2000 start measurement (day                                                                         organ relative humidity [%]
                                                            298)


Figure 8-45: Climatic conditions near the monumental organ


Fluctuations
From the measurements above, the hourly and daily fluctuations have been calculated
and plotted in the next figure.
                                                              hourly fluctuations of air temperature at the organ
                                               4
                                                                                                                                                60
          dT/dt organ [K/h]




                                               2                                                                                                50
                                                                                                                                    % of time




                                                                                                                                                40
                                               0
                                                                                                                                                30

                                               -2                                                                                               20
                                                                                                                                                10
                                               -4
                                                280   300    320       340     360        380     400      420   440      460                       -2    -1.5             -1       -0.5         0      0.5        1      1.5            2
                                                        time [days]: 24-10-2000 is start measurement (dag 298)                                                        hourly fluctuations of air temperature at the organ
                                                           hourly fluctuations of relative humidity at the organ
                                         10

                                                                                                                                                30
  dRH/dt [%RH/h]




                                               5
                                                                                                                                    % of time




                                               0                                                                                                20

                                               -5                                                                                               10

                                    -10                                                                                                         0
                                      280             300       320     340      360      380     400      420      440   460                   -10      -8            -6       -4      -2       0      2        4       6          8    10
                                                                                                                                                                      hourly fluctuations of relative humidity at the organ




Figure 8-46: Hourly fluctuations of temperature and relative humidity at the organ


                                                              daily fluctuations of air temperature at the organ
                                               5
                  Max. T difference [°C/day]




                                                                                                                                                25
                                               4
                                                                                                                                                20
                                                                                                                                    % of time




                                               3
                                                                                                                                                15
                                               2
                                                                                                                                                10
                                               1                                                                                                5
                                               0                                                                                                0
                                               280    300    320       340     360       380     400      420   440       460                        0        1             2          3          4       5         6           7        8
                                                        time [days]: 24-10-2000 is start measurement (dag 298)                                                         daily fluctuations of air temperature at the organ
                                                           daily fluctuations of relative humidity at the organ
     Max. RH difference [%RH/day]




                                         15
                                                                                                                                                10

                                                                                                                                                8
                                         10
                                                                                                                                    % of time




                                                                                                                                                6

                                               5                                                                                                4

                                                                                                                                                2

                                               0                                                                                                0
                                               280    300       320     340      360      380     400      420      440   460                        0                             5                         10                          15
                                                                                                                                                                      daily fluctuations of relative humidity at the organ




Figure 8-47: Daily fluctuations of air temperature and relative humidity at the organ




                                                                                                                                8-175
Air velocities
Figure 2-1 shows the air velocities measured near the organ as a function of time.

                                                                    Air velocity at organ = f(time)
                                       0.4



                               0.35



                                       0.3


                               0.25



                                       0.2
             v [m/s]




                               0.15



                                       0.1



                               0.05


                                        0



                           -0.05
                               280            300   320        340        360          380      400        420    440           460
                                                      time [days]: 24-10-2000 start of measurement (day 298)




Figure 8-48: Air velocities measured near the organ

Figure 8-49 gives an overview of temperatures measured in the church, at the pew
heating pipes and of the floor surface.
                                                                            Temperatures
                                       20
                                                                                                                        altar
                                                                                                                        tomb
                                       15                                                                               organ
                       Tair [°C]




                                       10


                                        5
                                        280   300   320       340          360         380            400   420   440           460

                                       50
                       Tbenchheating




                                       40

                                       30

                                       20

                                       10
                                        280   300   320       340          360         380            400   420   440           460

                                       20
                       Tfloor




                                       15




                                       10
                                        280   300   320        340        360          380      400        420    440           460
                                                      time [days]: 24-10-2000 start of measurement (day 298)



Figure 8-49: overview of measured temperatures



                                                                           8-176
Next figures indicate the relation between the operation of the heating system and the
resulting temperature driven air velocities.
                                                                                                                          Temperature difference: bench - air (at organ)
                                                                            40


                                                                            30
                                                       Tbench - Tair [°C]



                                                                            20

                                                                            10


                                                                             0


                                                                            -10
                                                                              280             300             320        340        360          380      400        420                               440      460
                                                                                                                time [days]: 24-10-2000 start of measurement (day 298)


                                                                            0.2


                                                                       0.15
                                               air velocity [m/s]




                                                                            0.1


                                                                       0.05


                                                                             0
                                                                              -5                    0               5                    10          15                   20          25              30         35
                                                                                                                                              Tbench - Tair [°C]



Figure 8-50: air velocity near the organ as a function of the temperature difference
between the pew heating pipes and the indoor air

Ventilation and infiltration rate
The ventilation and infiltration rate were measured again using the B&K gas analyzer.
The results of the measurements are shown below. The infiltration rate turned out to
be about 0.2 1/h.
                               4
                           x 10                                     B&K gas measurements                                                                                         SF6fit : n= 0.20964 [1/h]
                       3                                                                                                                                      5
           CO2 [ppm]




                       2
                                                                                                                                                             4.5
                       1
                                                                                                                                                              4
                       0
                           0       0.2   0.4   0.6                          0.8       1       1.2       1.4   1.6       1.8          2
                                                                                                                                 4
                 200                                                                                                          x 10                           3.5
  SF6 [ppm]




                                                                                                                                                 SF6 [ppm]




                 100                                                                                                                                          3


                       0                                                                                                                                     2.5
                           0   4 0.2     0.4   0.6                          0.8       1       1.2       1.4   1.6       1.8          2
                           x 10
                 1.5                                                                                                             4
                                                                                                                              x 10                            2
   H2O [mg/m3]




                       1                                                                                                                                     1.5


                 0.5                                                                                                                                          1
                           0       0.2   0.4   0.6                          0.8       1       1.2       1.4   1.6       1.8          2                             0   2000    4000          6000       8000   10000   12000
                                                                                   time [s]                                      4                                                         time [s]
                                                                                                                              x 10



Figure 8-51: Results of B&K gas measurements


Ceiling conditions
The relative humidity near the indoor surface of the restored vault turned out to be
the most critical, because of the lower indoor surface temperature. It was calculated
from the indoor surface temperature and the indoor absolute humidity.


                                                                                                                                               8-177
                                               Relative Humidity near the indoor surface of the restored vault
                                        100

                                        90

                                        80

                                        70




                     RH surface [%RH]
                                        60

                                        50

                                        40

                                        30

                                        20

                                        10

                                         0
                                         400       410         420          430          440          450        460
                                                                          Time [h]




Figure 8-52: Relative humidity near the indoor surface of the restored vault


Infrared thermal images
In the next figures a comparison of the optical contamination and the infrared
determined surface temperatures is made.




Figure 8-53: Photographs (left) and infrared thermal images (right) of the
contaminated ceiling; lower pictures: upper part of pictures shows restored, lower part
shows original vault



                                                                       8-178
Thermal resistance of the vault construction
The thermal resistance of the vault construction has been determined from the heat
flux measurements through, and the temperature difference across, the vault
construction. The momentarily obtained values of the thermal resistance of the
original and restored part of the vault construction are compared in Figure 8-54. The
moving average values of these heat resistances, are shown in Figure 8-55.
                                                    thermal resistance original vault                                                                                                    thermal resistance restored vault
                                0.5                                                                                                                                  0.5

                               0.45                                                                                                                                 0.45

                                0.4                                                                                                                                  0.4
 thermal resistance [m2*K/W]




                                                                                                                                       thermal resitance [m2*K/W]
                               0.35                                                                                                                                 0.35

                                0.3                                                                                                                                  0.3

                               0.25                                                                                                                                 0.25

                                0.2                                                                                                                                  0.2

                               0.15                                                                                                                                 0.15

                                0.1                                                                                                                                  0.1

                               0.05                                                                                                                                 0.05

                                 0                                                                                                                                    0
                                 405   410   415   420     425    430    435    440                           445   450    455                                        405   410   415   420     425    430    435    440     445   450   455
                                                   time [days]: 08-02-2001 is day 405                                                                                                   time [days]: 08-02-2001 is day 405




Figure 8-54: Momentarily determined thermal resistances of original and restored
vault

                                                                                                     cumulative mean determined thermal resistance original and restored vault = f(time)
                                                                                                    0.32
                                                                                                                                                                        restored
                                                                                                                                                                        original
                                                                                                     0.3
                                                                       thermal resitance [m2*K/W]




                                                                                                    0.28



                                                                                                    0.26



                                                                                                    0.24



                                                                                                    0.22



                                                                                                     0.2
                                                                                                             410     415     420     425     430    435     440                                445       450
                                                                                                                             time [days]: 08-02-2001 is day 405



Figure 8-55: Moving average determined thermal resistances of original (upper part)
and restored vault (lower part)


Conclusions
The contamination of the ceiling in the St. Gerlachuskerk resulted from thermal
driven airflows and the transported dirt and soot. The airflows could be reduced
effectively by lowering the temperatures of the pew convectors, e.g. from 40 to 30 oC.
The difference in contamination of the original and renewed vault of the St. Gerlachus
Kerk in Houthem clearly seemed to be the result of the differences in surface
temperatures. Infrared thermal imaging of the surface temperatures led to the same


                                                                                                                                     8-179
pattern in surface temperatures as the visual appearance of the contamination. Even
the visual pattern of wooden construction beams in the contamination was the same
as in surface temperatures. Obviously, lower surface temperatures resulted in more
severe contamination. The deposition of particles from a laminar, natural convection
boundery layer flow, adjacent to a cool surface, occurred due to a combination of
thermophoretic drift and Brownian motion (Camuffo 1995). The temperature
difference between air and surface was responsible for the resulting force of dirt
particles towards the cold surface; this phenomenon is known as thermophorese. The
solution to the contamination problem therefore was reducing this temperature
difference, e.g. by increasing the surface temperatures.
The origin of the lower surface temperatures proved to be the thermal resistance of
the vault construction. Improving the thermal resistance of the vault or roof
construction therefore would reduce the contamination.
Apart from that it was recommended to move the offering of wax lights outside the
church, to reduce the source term of the contamination.




                                       8-180
8.6    PLANNED INFRARED GAS HEATING IN THE GROTE KERK IN
       DORDRECHT

8.6.1 Description and history of the church
The Grote or Onze Lieve Vrouwekerk in Dordrecht is one of the largest medieval
churches in the Netherlands. The oldest part of the present church dates from 1285,
but underlying Romanesque foundations from the 12th century have been found. The
church was built in a so-called Brabant gothic style, characterized by typical soberness.
The church has a stone vault. In the chapels beautiful pictures, wall paintings and
stained glass can be found. The choir pews have beautiful hand curved sculptures in
the wooden panels. The pulpit dates from 1757. The fence of the choir dates from
1743.




8.6.2 Building data
The volume of the church is 50,000 m3.

8.6.3 Heating system
An electrical floor heating under the pews heats part of the church. For this part of
the church a draft layout of an infrared gas heating system was made. It consisted of
13 heaters, 19.8 kW each, with asymmetrical reflectors at a height of 13 m in the nave
and 3 symmetrical 39.6 kW heaters at a height of 18 m in the crossing part. The
capacity of the heaters therefore totals 376.2 kW.

8.6.4 Measurements
To get an indication of the expected initial and boundary conditions, a number of
indicating measurements on indoor air temperature and relative humidity and surface
temperatures were carried out. Furthermore the expected infiltration rate was
determined experimentally.



                                         8-181
Infiltration rate
The infiltration rate was determined by the measurement of the decay curve of the
CO2 concentration, produced by human sources. From a number of measurements
the results were 0.11, 0.07 and 0.06. In the calculations a mean infiltration rate of 0.1
was used.

Indoor temperature and humidity measurements
To get an indication of the indoor temperature and humidity conditions with the
present installed local floor heating system, during a week in October 1998 a number
of measurements was carried through. The results are summarized in Table 8-2 and
Table 8-3.
Position         Serial nr       θmean     θmin     θmax     RHmean RHmin       RHmax

Attic            9829-166        10.3       8.3     12.6          82     72          82

Outdoor          9829-176        11.3       7.8     13.7          78     69          90

Pew              9829-171        13.4       12.6    16.3          67     57          73

Organ            9829-172        13.1       12.3    14.7          71     64          75

Table 8-2: Results of air temperature and relative humidity measurements

Position           Serial nr            θmean              θmin               θmax
South wall         9747-508             13.9               13.9               14.5
Floor              9747-508             14.3               14.2               14.8
Table 8-3: Results of surface temperature measurements


8.6.5 Calculations
The initial and boundary conditions and the values of parameters used in the
calculations are summarized in Table 8-4.
Initial indoor air temperature θI,0                           13 oC
Initial indoor air relative humidity RHi,o                    60 %
Initial indoor air carbon dioxide concentration ci,0          0,6 g/m3
Initial outdoor water vapor ratio xo,0                        5 g/kg
Initial surface temperatures θs,0                             14 oC
Radiant flux                                                  600 W/m2
Infiltration rate ni                                          0.1 h-1
Ventilation rate nv
Volume                                                        50,000 m3
Table 8-4: Initial and boundary conditions and parameter values



                                            8-182
Radiance results
The geometrical setup of the planned infrared radiation heating system is given in
Figure 8-56 as it has been implemented in Radiance (Schellen ea, 1999/1). Figure 8-56
(left) shows the geometry of the church pillars and the radiant heaters. The radiation
has been calculated to heat up the wooden soundboard of the pulpit with a radiation
flux of up to 600 W/m2. Figure 8-56 (right) shows the radiant heat flux calculated by
the ray tracer model Radiance.




Figure 8-56: Geometry of Radiance implementation (left) and calculated irradiance on
the surface of the soundboard of a monumental wooden pulpit (right)


Wufi results
The expected effects of the thermal radiation on the heating up of the sound board of
the wooden pulpit were calculated with the simulation model Wufi. The calculated
radiant heat flux by Radiance was introduced as a radiant source of 600 W/m2 in the
climatic data file of Wufi for an expected weekend use of the church: 4 hours on, 8
hours off, 3 hours on. The initial temperature conditions were determined by the
indicative indoor temperature and humidity measurements near the pews. For the
material properties of the wooden pulpit the material database of Wufi was used:
longitudinally cut oak was the selected wooden material. The figures below indicate
the expected changes in surface temperature and water content for the wooden
soundboard of the pulpit during a weekend use of the church.




                                        8-183
                                   Figure 8-57: WUFI results of moisture content in wooden soundboard of pulpit during
                                   direct heating by gas infrared heaters


                                   1st Order indoor air humidity calculations
                                   In a first approximation, the expected indoor air humidity in the church due to the
                                   vapor production of the infrared gas burning devices was calculated from equation 2-
                                   43. Water absorption and desorption processes at walls, ceiling and floor were
                                   neglected. The results are derived, shown in Figure 8-58, as an absolute water vapor
                                   ratio x [g/kg] and a dewpoint θdew temperature as a function of time.

                                              2 hr operation, no mechanical ventilation                                                             2 hr operation, no mechanical ventilation
                              10                                                                                                       12


                              9

                                                                                                                                       10
                              8


                              7
                                                                                                       Dewpoint temperature Tdwp [C]




                                                                                                                                       8
Water vapour ratio x [g/kg]




                              6


                              5                                                                                                        6


                              4

                                                                                                                                       4
                              3


                              2
                                                                                                                                       2

                              1


                              0                                                                                                        0
                                   0      5             10                    15          20                                                0   5             10                    15          20
                                                              Time [h]                                                                                              Time [h]




                                   Figure 8-58: Water vapor water ratio (left) and dew point temperature (right) due to
                                   indoor gas burning without extraction; V=50,000 m3, n=0.1 h-1, Gp=0.0173 kg/s, 300
                                   persons




                                                                                               8-184
1st Order indoor air carbon dioxide calculations
Equation 2-43 also was used to give an approximation of the expected carbon dioxide
concentrations.
                                                                  2 hr operation, no mechanical ventilation
                                                   5000


                                                   4500


                                                   4000


                                                   3500
                       Concentration CO2 c [ppm]




                                                   3000


                                                   2500


                                                   2000


                                                   1500


                                                   1000


                                                    500


                                                     0
                                                          0   5             10                    15          20
                                                                                  Time [h]




Figure 8-59: CO2 concentration due to indoor gas burning without extraction; V=50,000
m3, n=0.1 h-1, Gp=0.0212 kg/s, 300 persons


WaVo calculations
For the calculations made before, the conditions measured were used as an indication
for initial conditions and boundary conditions. To account for the expected behavior
of the church for yearly and more extreme winter weather conditions, and to account
for more realistic thermal and hygric surface behavior, the computer program WaVo
was used. In this study, WaVo made use of climate condition files, measured over
many years by KNMI in The Bilt.

8.6.6 Conclusions
In the Grote Kerk in Dordrecht an infrared radiation gas heating system was planned
with a total heating capacity of 375 kW. The infiltration rate of the church proved to
be small (~0.1 h-1) and the calculated dew point therefore would be increased by
several degrees centigrade. During heating no condensation would occur on the
directly radiated walls but surfaces in the shadow of the heaters would consequently
have rather high near surface humidities. Due to the rather low infiltration rate directly
after terminating heating the absolute humidity would remain nearly the same. The air
temperatures and surface temperatures however would decrease in a short time and
high relative humidities near cold surfaces and even condensation would often occur.
The intention of the heating firm to ventilate the church via the attic proved to be
very dangerous. This would lead to frequently occurring condensation on the inner
surface of the roof timber and tie beams.
One of the infrared heaters directly heated the monumental pulpit. A simulation study
to predict temperature and moisture content of the directly radiated pulpit was carried


                                                                            8-185
out. Surface temperatures of 40 oC were reached at the top of the pulpit, resulting in
relative humidities near the surface of about 20 %RH. The simulation results showed
that the water content of the wooden pulpit would decrease dramatically in a short
time. Comparison of these rather low surface relative humidities with shrinkage
diagrams of tangential oak wood, led to an expected wood contraction percentage of
up to 5 %.
The relatively quick changes of surface temperatures and accompanying relative
surface humidities did not meet standards for museum properties (52 ± 3 %RH,
θmin=2 oC, θmax=25 oC, maximum change in twenty four hours 3 K, (Jütte 1994)). In
principle, however, a church is not a museum; therefore these requirements were
adapted for churches (Jütte 2000). The heating system did not reach these
requirements either.




                                         8-186
8.7   CONDENSATION AND GAS HEATING IN THE R.K. KERK IN BEMMEL

8.7.1 Description and history of the church
H. Donatus’ church in Bemmel was built on the old foundations of the original
church, which was demolished during the Second World War.




8.7.2 Building data
The outdoor walls of the church are made of masonry and have a mean thickness of
0.7 m. The church has 10 masonry columns. The indoor sides of the walls are
plastered and mural decorative painted. The glazing in the outdoor walls is made of
stained glass and has outdoor protective glazing. The floor in the church has stone
tiles and has an area of 840 m2. The vault is made of brick. The roof is made of
construction wood, covered with slates. The volume of the church is about 9800 m3.

8.7.3 Heating system
The heating system consisted of 17 radiant heaters: 2 heaters of 26.4 kW in the choir,
4 heaters, 13.2 kW each, in the aisles, 10 heaters of 13.2 kW in the nave of the church
and 1 heater of 6.6 kW near the chorale. The 2 heaters in the choir were symmetrical
ones, the rest of the heaters were of an asymmetric type.

8.7.4 Formulation of the problem
In the R.K. church in Bemmel a radiant heating system like the one designed for
Dordrecht had been in use for about 10 years. Because of complaints about
condensation on walls, floor and glazing the decision was made to remove the system
from the church. Just before the final removal of the heaters the opportunity was




                                         8-187
given to do some measurements on the moisture and radiant heating effects of the
system under winter conditions in February 1999 (Vugts 1999).

8.7.5 Analysis
The condensation on floor, glazing, walls and natural stone seats were expected to be
the result of the moisture production by indoor gas heating.

Measurements
To get an impression of the moisture and radiant heating effects and to determine the
infiltration characteristics of the church, the following measurements took place
(Vugts, 1999).

Infiltration rate
The infiltration rate was determined by the CO2 decay of the CO2 produced by the
radiant heating system. For that reason a B&K Multigas Monitor was used, together
with a multiplexing system. The infiltration rate was calculated from the results to be
about 0.06 h-1.

Indoor temperature and humidity measurements
To determine the moisture effects of the heating system, 1-minute interval air
temperature and relative humidity measurements took place near the altar, in the nave
of the church and near the organ. Furthermore the B&K multigas monitor was used
for measurements on vapor concentration with a 2 minutes interval. To calculate near
surface moisture effects, surface temperatures were measured by thermistor probes on
walls, glazing, floor and canvas paintings. Furthermore infrared thermography
measurements were made with intervals of 5 minutes.

Measurement results
The figures below show the results of the indoor surface temperature of the glazing,
the calculated dew point and the near surface relative humidity, as it was calculated
from the absolute humidity measured and the thermal infrared imaging measurements.
From the figures it is clear that within a one and a half hour-operation period of the
gas infrared heaters the dewpoint temperature exceeds the surface temperatures of the
stained glass windows. Furthermore the figures show the infrared measured
temperatures to almost equal the thermistor measured surface temperatures.




                                         8-188
                           Surface temperatures leaded windows                                                   Relative humidity near surface leaded windows
            12                                                                                      1
                                                                      Tsurf1                                                                                     RHsurf1
                                                                      Tsurf2                      0.95                                                           RHsurf2
            11                                                        Tsensor
                                                                      Tdew                         0.9

            10                                                                                    0.85

                                                                                                   0.8
            9
Temp [oC]




                                                                                         RH [-]
                                                                                                  0.75
            8
                                                                                                   0.7


            7                                                                                     0.65

                                                                                                   0.6
            6
                                                                                                  0.55

            5                                                                                      0.5
            13.5   14   14.5   15     15.5     16     16.5       17   17.5      18                   13.5   14   14.5     15     15.5     16     16.5     17     17.5      18
                                        time [h]                                                                                   time [h]




      Figure 8-60:     Measured surface temperatures and dewpoint (left) and near surface
      relative humidity (right)

      Figure 8-61 shows the relative humidity near the wall surface in a graphical way. With
      an infrared thermal camera, thermal images were taken of the wall surface in the
      direction of the altar section with intervals of half an hour. A large part of this surface
      is located in the 'shadow' of the heaters. The relative humidity was calculated from the
      absolute humidity and the thermal image surface temperatures, according to section
      2.3.2.




                                                                                 8-189
Near surface humidity: ‘hygrographs’




Figure 8-61: Relative humidity near the surface as calculated from absolute humidity
and thermographic surface temperatures; the shown infrared images were taken with
intervals of half an hour




                                       8-190
Simulation results
To check the quality of results, obtained from calculations before, a simulation run
with WaVo on the Bemmel Church characteristics was carried out. The results of it
were compared with the results measured. A typical graphical representation is in
Figure 8-62.




                           20                                   1

                                                               0.8
                           15
              T air [oC]




                                                 RH air [-]
                                                               0.6
                           10
                                                               0.4
                           5                                              Simulation
                                                               0.2
                                                                          Measusurement
                           0                                    0
                                14   16   18                         14    16       18


                           20                                   1

                                                               0.8
                           15
                                                 RH wall [-]
              T wall oC




                                                               0.6
                           10
                                                               0.4
                           5
                                                               0.2

                           0                                    0
                                14   16   18                         14    16       18



Figure 8-62: Comparison of WaVo calculations and measurements


8.7.6 Conclusions
Measurements in Bemmel proved the calculations for Dordrecht to be fail-safe: a 2-hour
operating period of the gas infrared heating devices led to most serious surface
condensation on stained glass windows, walls and floor. In this church thermal images
were translated into hygro-thermal images: ‘hygrograms’. These images show the most
critical condensation surfaces to be lying in the shadow of the infrared heating devices.
Furthermore it is clear from these images that the most critical time for surface
condensation on cold surfaces is a little while after stopping the heating: the temperature
decreases in a short time and where the infiltration rate is low the absolute humidity
nearly reaches its maximum. In this Bemmel church the combination of a large moisture
production source and a relative small infiltration rate again were a serious threat to the
monumental building and its interior.




                                               8-191
8.8   INFRARED GAS HEATING IN THE GROTE KERK IN ALKMAAR

8.8.1 Description and history of the church
The oldest parts of The Grote or St. Lawrence church in Alkmaar date from the 11th
or 12th century. They date from an early Romanesque church, which was built before
the start of the present church. This later church was built from 1470 to 1512. The
church was originally dedicated to St. Lawrence.
The type of church is a basilica. The church has a nave with two side aisles, a choir
and a north and south transept. The side aisles have a lower height and a vault, made
of stone. The nave has a wooden vault. Therefore the pillars are slimmer and have a
greater height than would have been possible with a stone vault.




8.8.2 Building data
The horizontal section of the church has a shape like a Latin cross. The nave is the
longest part of the church. The dimensions of the church are 85 m by 56 m. The
height of the church is about 35 m.
The outdoor walls have a thickness of about 1.0 m. The leaded glazing has no
protective glazing. The floor of the church has an area of 2180 m2 and is made of
natural stone. The floor bedding beneath is made of concrete, which is insulated and
is partly covered with a floor heating. The vault in the central part of the church is
made of soft wood with an area of about 4200 m2. The volume of the church is about
45,700 m3.




                                        8-192
8.8.3 Heating system
The heating system in the church consists of three gas burned heating systems: a floor
heating system with a capacity of 224 kW, which is able to keep the church at a
primary temperature. During the heating season this floor temperature system is
nearly continuously driven. The second heating system consists of EcoCeramics
infrared radiant heating devices. A total of 240 kW capacity is distributed over 10 units
of 24 kW, each consisting of 4 burning modules. The 4 burning modules are hanging
between two pillars at a height of 8 meters. These gas burning modules can be
adjusted by hand from 1.5 to 6 kW heating power. In the choir and the south transept
20 rented infrared radiant heaters make up the third (temporary) heating system. In
pairs of 2 combined 17.5 kW radiant heaters these are mounted between two pillars at
a height of 8 meters. These radiant heaters cannot be regulated. The infrared radiant
heating system is a so-called ‘open gas’ heating system: the heaters have no gas
exhaust pipe system. The combustion gasses, mainly consisting of carbon dioxide and
moisture, are released into the church.

8.8.4 Formulation of the problem
Apart from church services the church is mostly in use for cultural events. For these
services and events a heating system is needed, giving a reasonable level of thermal
comfort. The floor heating system turned out to be able to heat the church to a
primary temperature level only, far away from thermal comfort at normal winter
clothing level. For that reason the church management rented radiant heating devices
as mentioned before. For the future the management was aiming for a more definite
heating system. Former studies (Kaan 1997) showed there were hardly any alternatives
to heat this church: a combination of a very large volume and a very high infiltration
rate showed prohibitive heating costs to be expected. One of the early ideas was to
split up the church in an upper and lower part by some kind of horizontal foil. Later
on infrared radiant heating was expected to be a more realistic way to heat the church
to comfort level. During a meeting of the church management, the advising institute
Netherlands Energy Research Foundation ECN, municipal authorities, national
heritage authorities of the Netherlands Institute for Cultural Heritage (ICN) and the
Netherlands Department for Conservation (RDMZ) and Eindhoven University of
Technology (TUE), advantages and drawbacks of infrared radiant heating systems
were discussed.

8.8.5 Analysis
The main goals turned out to be the expected level of thermal comfort and the
limitation of the heating costs. The expected drawbacks were the moisture production
and its effects on the monumental interior and church construction, the CO2
production and the direct radiant heating flux and its possible overheating of parts of
the monumental interior. Therefore, it was agreed to start a measurement period in
the heating season to monitor indoor conditions in the church, its effects on the
monumental interior and the construction parts of the church. The measurements
were carried out by ECN (Koene ea, 2001), in close co-operation with the other



                                         8-193
partners (ICN, RDMZ and TUE). The checking criteria were formulated beforehand
by ICN and TUE (Jütte 2000). These are summarized in Table 5-5.

8.8.6   Measurements

Infrared gas heater usage
The times the infrared gas heaters were used is indicated in the pictures. Green bars
indicate the usage of the EcoCeramics-heaters, yellow bars indicate the usage of the
rental heaters.

Outdoor climate conditions
The outdoor air temperature and relative humidity were recorded in the tower at the
north west corner of the church, this being well ventilated by and open to outdoor air.

Indoor climate conditions
In 12 different places in the church air temperatures and relative humidities were
recorded: in the north part of the nave, under and above the vault, in both
monumental organs, at the connection of wooden beam and outdoor construction
walls, in the north and south transept. In the south transept the sensors were directly
heated, to record the effect of the heat on wooden wall panels.
                     25
                                                                                                           outdoor
                                                                                                           triforium N
                     20                                                                                    Eco-bur. on
                                                                                                           rentbur. on

                     15
            T [°C]




                     10



                      5



                      0



                     -5                                                                               0
                     11-dec   25-dec   8-jan   22-jan   5-feb   19-feb   5-mrt   19-mrt   2-apr   16-apr
                                                            date




Figure 8-63: Indoor and outdoor air temperature in relation to the burning of the
infrared heaters

When we look at the figure above the floor heating is able to heat the church to a
lower primary temperature level of 5 oC. The floor heating is necessary to maintain the
surface temperatures of the church to a minimum level, to prevent too early surface
condensation.




                                                                8-194
                    100                                                                                                  outdoor
                                                                                                                         triforium N
                                                                                                                         attic choir
                                                                                                                         day var Trif N
                          80
                                                                                                                         day var attic
                                                                                                                         Eco-bur. on
                                                                                                                         rentbur. on
                          60
         RH [%]




                          40



                          20



                          0                                                                                     0
                          11-dec    25-dec   8-jan       22-jan   5-feb   19-feb   5-mrt   19-mrt   2-apr   16-apr
                                                                      date



Figure 8-64: Indoor and outdoor relative humidity in relation to the burning of the
infrared heaters


                          1800
                                                                                                                             outdoor

                          1600                                                                                               triforium N
                                                                                                                             attic choir
                          1400                                                                                               trif-outdoor
                                                                                                                             Eco-bur. on
                          1200                                                                                               rentbur. on
             P H2O [Pa]




                          1000

                           800

                           600

                           400

                           200

                               0                                                                                    0
                               11-dec   25-dec   8-jan      22-jan   5-feb    19-feb   5-mrt   19-mrt   2-apr   16-apr
                                                                          date




Figure 8-65: Indoor and outdoor vapor pressure and difference

In Figure 8-65 the effect of the heaters on the absolute humidity is clearly to be seen.
During usage of the heaters the vapor pressure increases. The difference between
outdoor and indoor vapor pressure thus indicates the extra air moisture content.




                                                                             8-195
Surface conditions
Surface temperatures of the indoor roof surface, of wooden beam and outdoor wall
connections, of organ parts, of the directly heated pillar and soundboard parts, of
floor and window surfaces were measured and recorded.
                                         30
                                                                                                                                   soundboard
                                                                                                                                   upper
                                         25                                                                                        soundboard
                                                                                                                                   lower
                  T [°C], dT/dt [K/hr]




                                         20                                                                                        dT/dt, lower


                                                                                                                                   dT/dt, upper
                                         15

                                                                                                                                   Eco-bur. on
                                         10
                                                                                                                                   rentbur. on
                                          5


                                          0


                                         -5                                                                                   0
                                         11-dec 25-dec     8-jan    22-jan    5-feb     19-feb   5-mrt   19-mrt   2-apr   16-apr
                                                                                     date




Figure 8-66: Surface temperature conditions at the soundboard

During heating the soundboard e.g. is directly heated by an infrared heater. The
surface temperature increases to more than 25 oC.


                 100%

                                                                                                                                   soundboard upper
                                                                                                                                   soundboard lower
                 80%                                                                                                               Eco-bur. on
                                                                                                                                   rentbur. on


                 60%
        RH [%]




                 40%




                 20%




                       0%                                                                                                     0
                        11-dec                25-dec     8-jan     22-jan    5-feb     19-feb    5-mrt   19-mrt   2-apr   16-apr
                                                                                 date



Figure 8-67: Relative humidity near the soundboard, calculated from indoor absolute
humidity and surface temperature



                                                                                        8-196
Due to the overheating effect of the soundboard the relative humidity near the surface
of it decreases to levels of ca. 30 % RH. These are dangerous levels for preservation.

                             40                                                                                      4
                                                                                                                                          pillar bench
                             35
                                                                                                                                          bench choir
                             30
                                                                                                                                          Eco-bur. on

                             25
                                                                                                                                          rentbur. on
                    T [°C]




                             20                                                                                      2


                             15


                             10


                             5


                             0                                                                                       0
                             11-dec 25-dec     8-jan     22-jan    5-feb     19-feb    5-mrt    19-mrt   2-apr   16-apr
                                                                          date




Figure 8-68: surface temperature conditions at pillar and pew choir

When we look at the pillar and bench surface temperatures in the choir surface
temperatures even take values of more than 35 oC.

                     100
                                                                                                                                            RH Triforium

                                                                                                                     1,5                    conduct. wall

                                                                                                                                            Eco-bur. on
                        80                                                                                                                  rentbur. on

                                                                                                                     1,0
           RH [%]




                                                                                                                           conduct. [V]




                        60
                                                                                                                     0,5




                        40                                                                                           0,0
                         11-dec 25-dec       8-jan     22-jan     5-feb    19-feb     5-mrt    19-mrt    2-apr   16-apr
                                                                      date




Figure 8-69: electrical conductivity at surface wall north




                                                                             8-197
Condensation detection
Condensation detection devices were connected to the northeast indoor surfaces of
windows and outer walls. During some times they indicate surface condensation
during usage of the heaters.


Carbon dioxide and infiltration rate
The infiltration rate of the church was determined by tracer gas measurements with
the B&K gas-monitoring device, in combination with the multiplexer unit, according
to section 3.2.3. Furthermore, the carbon dioxide concentration was measured and the
infiltration rate too was calculated from its decay.

8.8.7 Conclusion
Very extensive measurements during two winter periods in the Grote or Laurenskerk
in Alkmaar led to the following conclusions. The infiltration rate proved to be high:
different values of 0.4 to 1.2 h-1 were measured. Furthermore the existing floor
heating system prevented surface temperatures to drop to too low values. As a result
of this, excessive condensation on cold surfaces has not been detected in this church.
Wooden beams in cold walls, leaded windows and cold walls proved to be critical
spots. High surface temperatures of pillars (up to 35 oC), pulpit top (up to 25 oC) and
pews (up to 35 oC) due to direct radiation heating were recorded. It was concluded
that the gas heating system should be modulated, monitored and controlled. The
moisture content of wooden beams, surface temperatures of directly heated
monumental elements and wall surface condensation should be used as limiting
control indicators during utilization of the gas infrared system.




                                         8-198
9 APPENDIX B: BASIC DATA SETS OF CASES


9.1     WAALSE KERK (WALLOON CHURCH) IN DELFT

Table 9-1: Dimensions

                           Total   Thickness of   Closed   Glazing   Percentage
                           area    construction   facade    [m2]       glass in
                           [m2]        [m]         [m2]              facade [%]
Outdoor walls
North                      234         0.75        179       55         24
North-northeast             29         0.75         19       10         34
East-northeast              29         0.75         29        0          0
East                        29         0.75         20       9          30
East-southeast              29         0.75         29        0          0
South-southeast             29         0.75         19       10         34
South                      234         0.75        156       78         33
Total outdoor wall         745                     585      160         22

Adiabatic walls
       church - museum     133         0.34
       church – vicarage   100         0.76
Total adiabatic walls      233

Vault                      330         0.02

Roof
North                      200        0.145
North-northeast            13         0.145
East-northeast             13         0.145
East                        13        0.145
East-southeast             13         0.145
South-southeast             13        0.145
South                      200        0.145
Total roof area            465
Floor                      210         0.1


Volume                     [m3]
church                     3050

attic                      195




                                     9-199
Table 9-2: Physical properties of materials

                                                 d       k        ρ         c
Building part             Material               [m]     [W/mK]   [kg/m3]   [J/kgK]

outdoor wall              plaster                0.02      0.8      1900      840
                          brick                  0.73      1.3      2100      840

wall between church and   plaster                0.02      0.8      1900      840
vicarage
                          brick                  0.72      1.3      2100      840
                          plaster                0.02      0.8      1900      840

wall between church and   plaster                0.02      0.8      1900      840
museum
                          brick                  0.3       1.3      2100      840
                          plaster                0.02      0.8      1900      840

pillars(organ)            plaster                0.01      0.8      1900      840
                          brick                            1.3      2100      840
                          plaster                0.01      0.8      1900      840

floor                     natural stone          0.10      2.9      2750      840
                          sand                   2.0        3       1650      840

vault                     wood                   0.02      0.14     550       1880


roof                      wooden panelling        0.02     0.14      550      1880
                          insulation             0.05     0.036       35      1470
                          air spalt              0.05     0.023      1.2      1000
                          wood                   0.02      0.14      550      1880
                          slate                  0.005     2.9      2750       840

Windows                   Type                   Spalt            U         ZTA
                                                                  [W/m2K]   [-]
leaded glass              colored glazing                         5.2       0.3




                                             9-200
Table 9-3: Heating system properties

                                                                              Unit
Boiler
number of boilers                                          2
manufacturer                                            Remeha
type                                                    Quinta 45
nominal capacity                                         45 kW                  each
efficiency                                               109 *1                  %

(Warm) Air heating
     supply temperature                                       40                 °C
     inlet grille                                air jet grille with 6 jets
     made, type                                      Solid Air, JGTA
     dimensions (h x w)                                 800 x 1200              mm
     height to floor                                           3                m
     direction jets                                       variable
     flow rate
               - before service                            7500                 m3/h
               - during service                            3750                 m3/h
     air velocity near grille
               - before service                            8                    m/s
               - during service                            4                    m/s
     re-circulation                                      100%
     return grille in floor                           1000 x 1200               mm

Hygrostatic control
     lower level (stop heating)                             40                   %
     upper level (extra heating)                            70                   %

Control devices (position)
     position 1                                    pulpit, 3m height
     position 2                                    organ, 7m height
*1. at a load of 30% and a return water temperature of 30°C




                                       9-201
Table 9-4: Church usage

                                                                                    Unit

Interior:                                                    organ
                                                             wooden pews

Indoor climate
primary temperature                                                  10                    °C
comfort temperature (during service)                                 20                    °C
RH min at organ*2                                                    40                    %
RH max at organ                                                      70                    %

Church usage
Standard service
number of persons                                                    50
additional internal moisture production                                                kg/h
additional internal heat load                                                           W
Special usage (holidays, concerts etc.)                              200
number of persons
additional internal moisture production                                                kg/h
additional internal heat load                                                           W

Natural ventilation
ventilation flow - church                                            488               m3/h
- attic                                                                                m3/h
ventilation rate - church                                            0.16               h-1
- attic                                                                                 h-1
*2. A moisture source with a capacity of 8 kg/h was installed in November 2001

Table 9-5: Energy consumption

Energy consumption                 2000   1999       1998     1997          1996           1995
Gas [m3]                          5,131   8,011      9,548   9,252          9,145          10,703
Electricity [kWh]                                    4,956   4,424          4,194           4,494




                                             9-202
9.2     SYPE KERK (SYPE CHURCH) IN LOOSDRECHT

Table 9-6: Dimensions

                               Total area    Thickness     Closed   Glazing   Percentage
                                 [m2]       construction   facade    [m2]       glass in
                                                [m]         [m2]              facade [%]
Outdoor walls
North                             205           0.87        170       35         17
North-northeast                    25           0.87         19        6         24
East                               95           0.87         79       16         17
South-southeast                    25           0.87         19        6         24
South                             140           0.87        120       20         14
      West                         95           0.87         95        0          0
Total outdoor wall area           585                       502       83         14

Indoor walls
     columns                      29            1.04
     organ                        27            0.03


Adiabatic wall                    233
     church-consistory             65           0.74

Wall between
      church – tower              75            1.3

Church tower
       north                      195           1.3
       east                       120           1.3
       south                      195           1.3
       west                       195           1.3
Total church tower                705

Vault                             450           0.01

Roof
North                             280          0.025
North-northeast                   15           0.025
East                               15          0.025
South-southeast                    15          0.025
South                             280          0.025
Total roof area                   605

Floor
       with floor heating         100           0.35
       without floor heating      265           0.35
Total floor                       365




                                            9-203
Volume                        [m3]
columns                           30
     church (- columns)         3330
     attic                       230


Table 9-7: Physical properties of materials

                                                 d       k        ρ         c
Building part             Material               [m]     [W/mK]   [kg/m3]   [J/kgK]

outdoor wall              plaster                0.02      0.8      1900      840
                          brick                  0.73      1.3      2100      840

wall between church and   plaster                0.02      0.8      1900      840
consistory
                          brick                  0.7       1.3      2100      840
                          plaster                0.02      0.8      1900      840

wall between church and   plaster                0.01      0.8      1900      840
tower
                          brick                  1.3       1.3      2100      840

columns(organ)            plaster                0.02      0.8      1900      840
                          brick                  1.0       1.3      2100      840
                          plaster                0.02      0.8      1900      840

floor                     natural stone          0.05      2.9      2750      840
(with and without floor
heating)
                          cement                 0.05      0.38     1000       840
                          insulation, PS         0.1      0.036      35       1470
                          concrete               0.15      1.04     1900       840
                          sand                   2.0        3       1650       840

vault                     wood                   0.01      0.14     550       1880

roof                      wooden panelling       0.02      0.14      550      1880
                          slate                  0.005     2.9      2750       840

organ plateau             wood                   0.03      0.14     550       1880



Windows                   Type                   Spalt            U         ZTA
                                                                  [W/m2K]   [-]
leaded glass              blank glazing                           5.2       0.7




                                             9-204
Table 9-8: Heating system properties

                                                            Unit

Boiler
     number of boilers                           2
     manufacturer                              Remeha
     type                                      HR 40
     nominal capacity                          40 kW          each
     efficiency                                  88            %

Floor heating
      supply temperature                          40           °C
      return temperature                          20           °C
      flow rate                                              [m3/h]
      number of groups                             6           [-]
      surface total                               100          m2

(Warm) Air heating
     supply temperature                            45          °C
     inlet grille                              2 in walls
     made, type
     dimensions (h x w)                        305 x 405      mm
     height to floor                               3          m
     direction grille
     flow rate
               - before service                  4000         m3/h
               - during service                  2500         m3/h
     air velocity near grille
               - before service                    4          m/s
               - during service                               m/s
     recirculation                              100%

Hygrostatic control
     lower level (stop heating)
     upper level (extra heating)                  70           %

Control devices (position)




                                       9-205
Table 9-9: Church usage

                                                                                    Unit

Interior:                                                  organ
                                                           wooden pews

Indoor climate
primary temperature                                                                        °C
comfort temperature (during service)                                 18                    °C
RH min at organ                                                                            %
RH max at organ                                                      70                    %

Church usage
       Sunday morning                                       10.00 - 11.00hr
       Sunday evening                                       18.30 - 19.30hr
Church usage
Standard service
number of persons                                                    50
       Sunday morning                                               200
       Sunday evening                                              50-60
additional internal moisture production                                                kg/h
additional internal heat load                                                           W
Special usage (holidays, concerts etc.)
number of persons
additional internal moisture production                                                kg/h
additional internal heat load                                                           W

Natural ventilation
ventilation flow - church                                                              m3/h
- attic                                                                                m3/h
ventilation rate - church                                          0.8-1.0              h-1
- attic                                                                                 h-1



Table 9-10: Energy consumption

Energy consumption                 2000   1999      1998    1997             1996          1995
Gas [m3]
Electricity [kWh]




                                            9-206
9.3      ST. LIDUÏNA BASILIEK (ST. LIDUÏNA’S BASILICA) IN SCHIEDAM

Table 9-11: Dimensions

                           Total area   Thickness      Closed   Glazing   Percentage
                             [m2]       of onstruct-   facade    [m2]       glass in
                                          ion [m]       [m2]              facade [%]
Outdoor walls
North                         830          0.85         668       162        20
North-northeast               50           0.85         40        10         20
      North east              50           0.85         37        13         26
East-northeast
East                          380          0.85          315       65        17
East-southeast                50           0.85          40        10        20
      Southeast               35           0.85           27        8        23
South-southeast               50           0.85           40       10        20
South                         750          0.85          605      145        19
      Southwest                15          0.85           11        4        27
      West                    540          0.85          455       85        16
      Northwest               15           0.85           11        4        27
Total outdoor wall area      2815                       2253      562        20

Indoor walls
       colums nave            44            0.7
       columns side nave      27            0.9
Total columns                 71

Adiabatic walls
       church - tower        145            0.85
       church – sacristy      80            0.85
Total adiabatic walls        225
                             233
Vault                        2150           0.22

Roof
North                         500          0.025
North-northeast               10           0.025
       Northeast              15           0.025
East-northeast                10           0.025
East                          170          0.025
East-southeast                10           0.025
       Southeast               15          0.025
South-southeast                10          0.025
South                         500          0.025
       Southwest                5          0.025
       West                   160          0.025
       Northwest               5           0.025
Total roof area              1410




                                        9-207
Floor                              850         0.06

Volume                       [m3]
church                       10325
columns                      55
attic                        1825


Table 9-12: Physical properties of materials

                                                d            k        ρ            c
Building part            Material               [m]          [W/mK]   [kg/m3]      [J/kgK]

outdoor wall             brick                  0.85           1.3          2100         840

columns nave             brick                  0.7            1.3          2100         840

columns side nave        brick                  0.9            1.3          2100         840


floor                    natural stone          0.06           2.9          2750         840
                         concrete               0.15           1.04         1900         840
                         sand                   2.0             3           1650         840

vault                    plaster                0.02           0.8          1900         840
                         brick                  0.2            1.3          2100         840

roof                     wooden panelling        0.02          0.14          550         1880
                         slate                  0.005          2.9          2750          840



Windows                  Type                   spalt                 U            ZTA
                                                                      [W/m2K]      [-]
stained glass            colored glazing
                                                ventilated            3.5          0.3
protective glazing       blank glazing




                                            9-208
Table 9-13: Heating system properties
                                                             Unit

Boiler
     number of boilers                              1
     manufacturer
     type
     nominal capacity                              300          kW
     efficiency                                                  %

(Warm) Air heating
     supply temperature                            70           °C
     inlet grille
     made, type
     dimensions (h x w)                         2000 x 750     mm
     height to floor                                3          m
     flow rate
               - before service                   21600        m3/h
               - during service                   10800        m3/h
     air velocity near grille
               - before service                   4-5          m/s
               - during service                    2           m/s
     re-circulation

Hygrostatic control
     lower level (stop heating)                     -           %
     upper level (extra heating)                    -           %

Control devices (position)




                                        9-209
Table 9-14: Church usage

                                                                                      Unit

Interior:                                                    organ
                                                             wooden wall
                                                             boards
                                                             wooden pews

Indoor climate
     primary temperature                                             10                      °C
     comfort temperature (during service)                            20                      °C

Church usage
     Tuesday, Thursday                                       19:00 – 19:30 h
     Wednesday, Friday                                       8:30 – 9:00 h
     Saturday                                                18:00 – 18:45 h
     Sunday                                                  10:00 – 11:30 h

Church usage
Standard service
number of persons
      Thuesday, Wednesday, Thursday, Friday                        15 - 25
      Saturday                                                      100
      Sunday                                                        225
additional internal moisture production                                                  kg/h
additional internal heat load                                      16,200                 W

Natural ventilation
ventilation flow - church                                                                m3/h
- attic                                                                                  m3/h
ventilation rate - church                                         0.5 – 0.6               h-1
- attic                                                                                   h-1
*2. A moisture source with a capacity of 8 kg/h was installed in November 2001

Table 9-15: Energy consumption

Energy consumption                2000        1999    1998        1997         1996          1995
Gas [m3]                             20,816
Electricity [kWh]




                                              9-210
9.4      ST. MARTINUS KERK (ST. MARTINUS’ CHURCH) IN WEERT

Table 9-16: Dimensions

                          Total area    Thickness     Closed   Glazing     Percentage
                            [m2]            of        facade    [m2]     glass in facade
                                       construction    [m2]                    [%]
                                           [m]
Outdoor walls
North                     695          1.1            500      195       28
North-Northeast           70           1.0            52       18        26
East-Northeast            95           1.0            63       32        34
East                      135          1.1 – 3.1      103      32        24
East-Southeast            95           1.0 – 2.8      63       32        34
South-Southeast           70           1.0 – 2.8      52       18        26
South                     695          1.1 – 3.1      500      195       28
West                      470          1.0            384      86        18
Total outdoor wall area   2325                        1717     608       26

Indoor walls
columns                   151          1.04

Adiabatic walls
church – tower            220          2.8
church – vicarage         90           1.0 – 2.8

Tower
North                     265          2.8
East                      145          2.8
South                     265          2.8
West                      365          2.8
Total tower               1040

Vaults                    1665         0.34

Roof
Noorth                    1345         0.085
North-northeast           80           0.085
East-northeast            80           0.085
East                      80           0.085
East-southeast            80           0.085
South-southeast           80           0.085
South                     1345         0.085
Total roof area           3090

Floor                     1260         0.35




                                       9-211
Volume                               [m3]
columns                               157
church (- columns)                   18693
attic                                5000

Table 9-17: Physical properties of materials

Building part                Material                   d [m]          k     ρ [kg/m3]   c [J/kgK]
                                                                    [W/mK]

outdoor wall                 plaster               0.02             0.8      1900        840
                             brick                 1.0 – 3.1        1.3      2100        840

wall between church and      brick                 1.0 – 2.8        1.3      2100        840
vicarage

wall between church and      brick                 2.8              1.3      2100        840
tower

columns                      plaster               0.02             0.8      1900        840
                             brick                 1.0              1.3      2100        840
                             plaster               0.02             0.8      1900        840

floor                        natural stone         0.05             2.9      2750        840
                             cement                0.1              0.38     1000        840
                             insulation            0.05             0.036    35          1470
                             concrete              0.15             1.04     1900        840
                             sand                  2.0              3        1650        840

vault                        plaster               0.02             0.8      1900        840
                             brick                 0.3              1.3      2100        840
                             plaster               0.02             0.8      1900        840

roof                         wooden panelling      0.02             0.14     550         1880
                             spalt                 0.04             0.023    1.2         1000
                             wood                  0.02             0.14     550         1880
                             slate                 0.005            2.9      2750        840

Windows                   Type                                  spalt           U          ZTA
                                                                             [W/m2K]        [-]
stained glass             colored glass            external ventilated       3.5         0.3
protective glazing        non-colored glazing




                                                9-212
Table 9-18: Heating system properties

                                                                             Unit

Boiler
number of boilers                                  1
type
capacity                                           127 kW                 each boiler
efficiency of the boiler                                                  %

Floor heating
supply temperature                                 50                     °C
return temperature                                 35                     °C
volume rate                                                               [m3/h]
number of groups                                   3                      [-]
total area                                         1050                   m2
T floor max                                        25 – 28                °C

Sensor position
temperature sensor                                 in floor


Table 9-19: Church usage

                                                                     Unit

Interior                                           vault paintings
                                                   organ
                                                   wooden pews

Indoor climate
stationary heating                                 15                °C

Church usage
Standard service
number of persons
additional internal moisture production                              kg/h
additional internal heat load                                        W
Special usage (holidays, concerts, etc.)
number of persons
additional internal moisture production                              kg/h
additional internal heat load                                        W

Natural ventilation
ventilation flowrate - church                      1885              m3/h
- attic                                                              m3/h
ventilation rate - church                          0.1               h-1
- attic                                                              h-1




                                           9-213
Table 9-20: Energy consumption

              Energy            2001           2000   1999   1998   1997
Gas [m3]                    27,237
Electricity [kWh]           57,100




                                       9-214
9.5     ST. GERLACHUS KERK (ST. GERLACHUS’ CHURCH) IN HOUTHEM

Table 9-21: Dimensions

                             Total    Thickness     Closed   Glazing   Percentage
                             area         of        facade    [m2]       glass in
                             [m2]    construction    [m2]              facade [%]
                                         [m]
Outdoor walls
North                         428        1.31         300      128         30
North-northeast               66         1.31         58        8          12
East                           70        1.31          70       0           0
East-southeast                66         1.31         58        8          12
South                         467        1.31         382       85         18
West                          139        1.31         139       0           0
Total outdoor wall area      1240                    1010      230         18

Indoor walls
       columns under organ     5         0.8
       organ plateau          80         0.1
Total indoor walls            85

Adiabatic walls
       church – vicarage      32         1.32
       church – lower part    79         0.76
       church - corridor     252         0.76
Total area indoor walls      363

Vaults
       old construction      300        0.0414
       new construction      300        0.0222
Total vault area             600

Roof
       North                 102         0.03
North-northeast              11          0.03
       East                   14         0.03
South-southeast               11         0.03
       South                 102         0.03
Total roof area              240

Floor
       with floor heating    253         0.31
       with floor heating    222         0.31
Total floor area             475




                                     9-215
Volume                                     [m3]
     columns + organ plateau                12
     church                                6680
     attic                                 1040

Table 9-22: Physical properties of materials

                                                          d        k        ρ         c
Building part                    Material                 [m]      [W/mK]   [kg/m3]   [J/kgK]

outdoor wall                     plaster                  0.02       0.8     1900       840
                                 brick and marl           1.3        1.3     2100       840

wall between church and lower    plaster                  0.02       0.8     1900       840
part
                                 brick                    0.72       1.3     2100       840
                                 plaster                  0.02       0.8     1900       840

wall between church and          plaster                  0.02       0.8     1900       840
vicarage
                                 brick                    1.3        1.3     2100       840
                                 plaster                  0.02       0.8     1900       840

columns under organ              plaster                  0.02       0.8     1900       840
                                 brick                    0.76       1.3     2100       840
                                 plaster                  0.02       0.8     1900       840

floor (with floor heating)       natural stone            0.06       2.9     2750      840
                                 cement                   0.07       0.38    1000       840
                                 insulation               0.06      0.036     35       1470
                                 concrete                 0.12       1.04    1900       840
                                 sand                     2.0         3      1650      840

floor (without floor heating)    granulate                0.13      2.9      2750       840
                                 concrete                 0.18      1.04     1900       840
                                 sand                     2.0        3       1650       840

old vault. plaster on wood and   wood/cement              0.0414    0.12      450      1470
straw

vault new                        wood/cement              0.0222    0.12      450      1470


roof                             wooden panelling         0.025     0.14      550      1880
                                 slates                   0.005     2.9      2750       840

organ floor                      hard wood                0.1       0.17      800      1880




                                                  9-216
Windows                      Type                        Spalt            U [W/m2K]     ZTA
                                                                                        [-]
stained glass                colored glass                   outdoor
                                                                             3.5             0.3
outdoor protective glazing   non-colored glass              ventilated




Table 9-23: Heating system properties

                                                                                      Unit

 Boiler
 number of boilers                                           1
 manufacturer                                             Remeha
 type                                                      Gas 3
 nominal capacity                                           112                        kW
 efficiency                                                  80                         %

 Floor heating
 supply temperature                                        34 *1                        °C
 return temperature                                         30                          °C
 flow rate                                                                            [m3/h]
 number of groups                                                                       [-]
        area each group                                                                 m2

 Pew heating
 manufacturer
 type
 nominal capacity                                                                      kW
 supply temperature                                         55                          °C
 return temperature                                                                     °C
 flow rate                                                                            [m3/h]

 Sensors
      Thermostatic                                   column near pulpit
*1 Measured value is 20°C




                                             9-217
Table 9-24: Church usage

                                                                                  Unit

Interior                                         paintings on ceiling and walls
                                                 organ
                                                 wooden pews, pulpit,
                                                 confession chair, altar
                                                 tomb H. Gerlachus

Indoor climate
primary temperature                                                                °C
comfort temperature (during service)                                               °C

Church usage
     Saturday                                            19.00 – 20.00u
     Sunday                                               9.30 – 10.30u

Church usage
Standard service
number of persons                                                                  -
Saturday                                                        60
Sunday                                                         120
Special usage (holidays, concerts etc.)
number of persons
additional internal moisture production                                           kg/h
additional internal heat load                                                      W

Natural ventilation
ventilation flow - church                                                         m3/h
- attic                                                                           m3/h
ventilation rate - church                                      0.21                h-1
ventilation rate - church                                                          h-1



Table 9-25: Energy consumption

Energy consumption                        2000             1999           1998    1997
Gas [m3]                                     10.171
Electricity [kWh]                            20.819




                                             9-218
9.6      R.K. KERK (ROMAN CATHOLIC CHURCH) IN BEMMEL

Table 9-26: Dimensions

                         Total area    Thickness     Closed   Glazing     Percentage
                           [m2]            of        facade    [m2]     glass in facade
                                      construction    [m2]                    [%]
                                          [m]
Outdoor walls
North                       565           0.7          490       75           13
Northeast                   45            0.7          35        10           22
East                        160           0.7          150       10            6
Southeast                   45            0.7          35        10           22
South                       565           0.7          490       75           13
Southwest                    25           0.7           20        5           20
West                        130           0.7          125        5            4
Northwest                   25            0.7           20        5           20
Attic north                  40           0.7           40        -            -
Attic south                  40           0.7           40        -            -
Total outdoor walls        1640                       1445      195           12

Indoor walls
Columns                     94            0.85

Adiabatic walls
church - tower              120           0.7
church - vicarage           95            0.7

Church tower
North                       145           0.7
East                         70           0.7
South                       145           0.7
West                        190           0.9
Total tower                 550

Vaults                     1090           0.2

Roof
North                      490
Northeast                   15
East                        215
Southeast                    15
South                       490
Southwest                    10
West                        210
Northwest                    10
Total roof                 1455

Floor                       840           0.06




                                      9-219
Volume                                    [m3]
columns                                    80
church (- columns)                       10,600
attic                                    2,350


Table 9-27: Physical properties of materials

Building                                d         k        ρ                c
part             Material               [m]       [W/mK]   [kg/m3]          [J/kgK]

outdoor wall     brick                  0.7       1.3      2100             840

columns          brick                  0.85      1.3      2100             840

floor            natural stone          0.06      2.9      2750             840
                 sand                   2.0       3        1650             840

vault            brick                  0.2       1.3      2100             840

roof             wooden paneling        0.02      0.14     550              1880
                 slats                  0.005     2.9      2750             840

Windows                          Type                      Spalt            U         ZTA
                                                                            [W/m2K]   [-]
stained glass                    colored glass
                                                           non ventilated   3.5       0.3
outdoor protective glazing       non-colored glass




                                                   9-220
Table 9-28: Heating system properties

                                                            Unit

Infrared gas heater type I
made                                    GoGas
type                                    K8424 RN
capacity                                26.4                kW
number                                  2                   -
properties                              symmetric heater

Infrared gas heater type II
made                                    GoGas
type                                    K8412 RAS
capacity                                13.2                kW
number                                  14                  -
properties                              asymmetric heater

Infrared gas heater type III
made                                    GoGas
type                                    K8406 RAS
capacity                                6.6                 kW
number                                  1                   -
properties                              asymmetric heater

Sensors




                                        9-221
Table 9-29: Church usage

                                                                                           Unit

Interior                                                            paintings, organ

Indoor climate
primary temperature                                                 8                      °C
comfort temperature                                                 15                     °C

Church usage
Thursday                                                            45                     min.
Saturday                                                            60                     min.
Sunday                                                              60                     min.

Church usage
Standard service
            mean number of people per week                1997      355
                                                          1998      570
                                                          1999      500                    pers.
                                                          2000      440
                                                          2001      450
            extra internal moisture production (from the heaters)   42                     kg/h
            extra internal heat load                                                       W

Natural ventilation
ventilation flow - church                                           5900                   m3/h
- attic                                                                                    m3/h
ventilation rate - church                                           0.6                    h-1
- attic                                                                                    h-1



Table 9-30: Energy consumption

Energy                                     2000           1999      1998           1997
Gas [m3]                                   6,155          300       6,642          5,057
Electricity [kWh]                          7,534          5,407     ?              ?




                                                  9-222
9.7     GROTE OR ST. LAURENSKERK (ST. LAWRENCE CHURCH) IN
        ALKMAAR

Table 9-31: Dimensions of the Grote Kerk in Alkmaar

                          Total area    Thickness     Closed   Glazing    Percentage
                            [m2]       construction   facade    [m2]        glass in
                                           [m]         [m2]               facade [%]
Outdoor walls
North                     1420         1              1175     425       27
North-northeast           110          1              84       26        24
East-northeast            110          1              84       26        24
East                      838          1              812      26        3
East-southeast            110          1              84       26        24
South-southeast           110          1              84       26        24
South                     1165         1              729      382       34
West                      1060         1              1056     33        3
Total outdoor wall area   5080                        4110     970       19

Indoor walls
pillars at nave           71           1.05
pillars corners nave      47           1.7
pillars at choir          89           1.05
walls between chapels     387          0.45
                          594

Adiabatic walls
church - sexton living    120          0.8
church – room             135          0.8

Vault                     4190         0.06

Roof
North                     1245         0.025
North-northeast           63           0.025
East-northeast            63           0.025
East                      330          0.025
East-southeast            63           0.025
South-southeast           63           0.025
South                     1245         0.025
West                      268          0.025
Total roof area           3340

Floor                     2180         0.37




                                       9-223
Volume                          [m3]
pillars                         248
church (- volume pillars)       45720
attic                           2665

Table 9-32: Physical properties of materials

                                                  d       k        ρ         c
Building part               Material              [m]     [W/mK]   [kg/m3]   [J/kgK]

outdoor wall                brick                 1.0     1.3      2100      840

pillars corner nave         brick                 1.7     1.3      2100      840

pillars nave                brick                 1.05    1.3      2100      840

pillars choir               brick                 1.05    1.3      2100      840

walls between chapels       plaster               0.01    0.8      1900      840
                            brick                 0.43    1.3      2100      840
                            plaster               0.01    0.8      1900      840

floor                       natural stone         0.2     2.9      2750      840
                            cement                0.1     0.38     1000      840
                            insulation            0.05    0.036    35        1470
                            concrete              0.2     1.04     1900      840
                            sand                  2.0     3        1650      840

vault                       soft wood             0.006   0.14     550       1880

roof                        wooden paneling       0.02    0.14     550       1880
                            slate                 0.005   2.9      2750      840

Windows                     Type                  Spalt            U         ZTA
                                                                   [W/m2K]   [-]
stained glass               colored glazing                        5.2       0.3




                                              9-224
Table 9-33: Heating system properties

                                                                               Unit

Boiler
number of boilers                           2
manufacturer                                Reminox
type                                        Gas 2000 16 ECO
nominal capacity                            2*112                              kW
efficiency                                  (HR-boiler)

Floor heating
supply temperature                          60                                 °C
return temperature                          ≤ 40                               °C
flow rate                                                                      m3/h
number of groups                            7                                  -
area each group                             ± 310                              m3

Infrared heaters type I
manufacturer                                EcoCeramics
type                                        Eco Heat Line 4
nominal capacity                            24                                 kW
number                                      10                                 -
properties                                  Each heater has 4 modules each
                                            6kW. Modules controlling from 10
                                            tot 100%

Infrared heaters type II
manufacturer
type
nominal capacity                            17.5                               kW
number                                      20                                 -
properties                                  on/off switchable

Thermostatic device (position)




                                        9-225
Table 9-34: Church usage

                                                                                  Unit

Interior:                                                       organ
                                                                wooden objects
                                                                parchment

Indoor climate
primary temperature                                             5                 °C
comfort temperature (during service)                            15                °C
RH min at organ                                                 40 - 45           %
RH max at organ                                                 60 - 75           %

Church usage

Church usage
Standard service
number of persons
additional internal moisture production                                           kg/h
additional internal heat load                                                     W
Special usage (holidays, concerts etc.)
number of persons
additional internal moisture production                                           kg/h
additional internal heat load                                                     W

Natural ventilation
ventilation rate - church                                                         m3
                        - attic                                                   m3
ventilation rate - church                                       0.5 – 1.2         h-1
                        - attic                                                   h-1



Table 9-35: Energy consumption

Energy consumption                        2000           1999        1998        1997
Gas [m3]
Electricity [kWh]




                                                 9-226
10 CURRICULUM VITAE

Henk Schellen has been a researcher since 1983 and assistant professor since 1986 at
Eindhoven University of Technology. He is an engineer in building physics,
specialized in heat and moisture transfer in buildings. His main expertise is on building
physical measurements and simulation.
In 1982 he developed the first thermal bridge program in the Netherlands and used it,
among other things, for the error estimation of heat flux devices. From 1984 to 1988
he participated in projects on the development of heat flux meters, very accurate
guarded hot plate apparatus and a hot-box device and made simulation calculations on
its expected accuracies. In 1989 he was active on the measurement of moisture in
soils.
Due to his contacts with the Netherlands Institute for Cultural Heritage (ICN), he
became interested in, and worked on, building physical heat and moisture problems in
monumental buildings, like churches and museums.
From 1991 to 1994 Henk had intensive inter-university research contacts with the
Technical University of Wroclaw, Poland and participated during these years in spring
school projects in Karpacz, Poland.
In 1992 he developed simulation models for- and participated in coaching of- a PhD
study on protective glazing for stained glass church windows and participated in a
German Government project (Bundesministerium Forschungs Programm) on the
protection of historical windows. In 1993 he conducted a study on the abundance of
dust mites in dwellings and after that on cockroaches.
In 1994 a 3-year research program was started on an innovative IFD (Industrial
Flexible Durable) way of building in steel: ISB. A 1:1 experimental building has been
investigated on the campus of Eindhoven University of Technology. In 1996 a
participation in a Swiss project on thermal bridges in naturally ventilated building
facades took place and the natural free convection airflows have been simulated with
CFD (Computational Fluid Dynamics).
From 1996 to 1999 he coached a PhD study on the measurement and simulation of
indoor airflow.
In 1998 a four-year research started on the heating of large, massive monumental
buildings, like monumental churches. The research was in cooperation with the
Netherlands Institute for Cultural Heritage (ICN) and the Netherlands Department
for Conservation (RDMZ). Part of the study was in cooperation with the Netherlands
Energy Research Foundation (ECN). Up to now about 10 of the most important
Dutch monumental churches have been investigated on building physical effects of
heating systems like floor heating, infra-red gas heating and air heating. Due to Henk’s
contacts with ICN and RDMZ, he is frequently invited to give his opinion on building
physical and indoor climate problems in Dutch monumental buildings like the Anne
Frank House, the Rembrandt House, the National Museum, Fortress Fort Aan de
Hoek van Holland and some 20 monumental churches. As a result of the experience


                                         10-227
gained on these subjects, he is often invited to give lectures on subjects related with
monumental buildings.
In April 2002 a European project on the development, measurement and simulation
of a new heating system for (monumental) churches has started. Henk is responsible
for the simulation survey and will be the co-promoter of a PhD student.
From 1983 to now Henk has participated in a cooperative research team, formed by
TNO (the Institute of Applied Physics) and TUE (Eindhoven University of
Technology). This resulted in 35 research reports, mostly in Dutch.
A complete survey of publications and lectures is available in an html file
Publications_Henk_Schellen_to_2002.html.




                                         10-228

								
To top