IEC Centenary Challenge Template

Document Sample
IEC Centenary Challenge Template Powered By Docstoc
					         CENTENARY CHALLENGE




IEC Centenary Challenge ID Number:        00042-KN406

Title:          Standardising Mesopic Vision Conditions and incidence on
                Light Sources Science and Technology

Lead-author:     Georges Zissis

Email:           georges.zissis@cpat.ups-tlse.fr

Co-authors:      Stuart Mucklejohn

Institution:     Université Toulouse 3 - Paul Sabatier

Email:           stuart.mucklejohn@btinternet.com

Date:           18-08-2006



Please ensure that you enter the IEC Centenary Challenge ID Number in the
document footer. Once completed please submit your paper in Word format by
email ich@iec.ch (to Mrs. Iana Choumiatchkina, IEC Central Office)




ID Number: 00042-KN406

Page 1 of 17
    Standardising Mesopic Vision Conditions and
        Incidence on Light Sources Science and
                                 Technology


1      Posing the problem
Standards affect every part of our life. From the quality of air and water, to the
assurance that products and services are safe and effective for use, there are
hundreds of standards helping to improve our everyday life. Standards play an
important role in ensuring that products, services and systems meet our needs;
for example standards should guarantee that electrical products, like lighting
systems, are energy-efficient and safe to use. In principle, our modern world
should fit together like a jigsaw puzzle thanks to standardization. Standards that
have a role in protecting the public‟s safety and health may become mandatory
through inclusion in laws and regulations, such as national, European or
International codes. However, there still exists a number of domains where
standards are cruelly missing and this is a serious handicap that slows down, or
in some cases stops, new product development and all associated business.

In this paper we will discuss, first, how missing standards in the domain of Human
Mesopic Vision affects the development of innovative light sources especially for
urban lighting systems that enhance security and quality of life in urban areas as
well as achieve important energy economies and contribute to sustainable
development. Then, in a second time, we will see how intense collaborative
research can help to overcome the problem. The discussion will be illustrated by
using the results obtained recently in the frame of a large European-funded
research project under FP5-Energie program targeting the development of
intelligent urban lighting schemes. But first of all let us consider the definition of
“Mesopic Vision”.

One of the extraordinary feats of human vision is our ability to see well in lighting
that ranges from moonlight to bright sunlight: almost a thousand-million-fold
change in light level. It is no wonder that the eye remains unsurpassed by any

ID Number: 00042-KN406

Page 2 of 17
single physical detector of light in yielding useful vision over a range that spans
nine units on a logarithmic scale. Operation over such a massive range, however,
is not without compromise. It is achieved through both rapid and slow processes
of adaptation and the involvement of two classes of photoreceptor; cone
photoreceptors that operate at higher light levels and rod photoreceptors that
operate at very low light levels. Mesopic vision describes the transition region
from rod vision (scotopic) to cone vision (photopic), where signals from both rods
and cones contribute to the visual response. Mesopic vision covers approximately
four log units and encompasses a range of light levels often found in occupational
environments.

For this discussion, the perceived brightness of an illuminated surface is
considered to be represented by the luminance, i.e. the light coming off the
surface. The units of luminance are candela per square metre (cd m -2). Human
vision can be characterised by three luminance ranges. Photopic conditions (cone
vision) refer to luminance values greater than 10 cd m -2, e.g. daylight. Scotopic
conditions (rod vision) refer to values less than 0.01 cd m -2, e.g. starlight. The
range between 0.01 and 10 cd m -2 is known as the mesopic region, e.g.
moonlight.

Since Purkinje‟s [1] observations of the changes in relative brightness of red and
blue objects as the ambient illumination decreased during twilight, the visual
changes that occur in crepuscular conditions has attracted a large number of
researchers. Many investigators have concentrated on the measurement of
spectral sensitivity in the mesopic region and the quest to produce an adequate
system of mesopic photometry. The problem is not trivial: Palmer [2, 3, 4] was the
first to put forward a model for mesopic luminous efficiency based on a linear
combination of the photopic and scotopic photometric functions available at the
time. However, for some physiological reasons that we will not develop in this
paper, mesopic lighting and vision research is faced with difficulties due to the
non-linearity and non-additivity of visual perception in mesopic range.

All in all, at present there is still no internationally accepted standard for the
measurement of luminous flux that reflects the spectral sensitivity of the eye in
the mesopic range. It is, however, generally agreed that a system of mesopic

ID Number: 00042-KN406

Page 3 of 17
photometry is needed to ensure that lighting in mesopic environments meets
safety requirements and is efficient.


2      Missing standards and urban lighting
More recently the focus has turned towards the measurement of mesopic spectral
sensitivity using criteria that relate more closely to levels of performance achieved
in visual tasks undertaken in the real world. Particular emphasis has been placed
on visual tasks that relate to road lighting. For instance, night-time driving is the
most commonly encountered situation that involves mesopic vision.

“Road lighting encompasses the lighting of all types of highways and public
thoroughfares, assisting traffic safety and ease of passage for all users. It also
has a wider social role, helping to reduce crime and the fear of crime, and can
contribute to commercial and social use at night of town centres and tourist
locations”.

The above paragraph is part of the introduction to the 2003 published British
Standard on Road Lighting [5]. It shows clearly that road lighting, and more
generally, outdoor lighting system designers have many different aspects to
consider. Safety and the perception of safety, for example, are paramount for
pedestrians. Lighting for drivers has a number of claimed benefits including the
number of night-time road traffic accidents, improving the visual comfort of drivers
and the potential to increase the night-time traffic capacity of a given road.
Running costs and aesthetic considerations, however, also have to be taken into
account for many installations. Outdoor lighting encompasses many other
applications, e.g. sports stadium lighting, illuminating historic buildings,
highlighting architectural features and pedestrian areas.

However, perhaps it is time to ask whether the current method of specification of
road lighting delivers maximum benefit to all classes of users for minimum cost in
terms of energy consumed.

The performance of light sources is often compared by reference to their efficacy,
i.e. output measured in lumens per watt of input power. However, the definition of
the lumen is derived from the spectral luminous efficiency function for photopic
vision only:
ID Number: 00042-KN406

Page 4 of 17
The eye‟s sensitivity is described by standards published by the Commission
Internationale de l‟Eclairage (CIE) for photopic and scotopic conditions. Under
photopic conditions, the sensitivity of the human eye peaks at 555 nm. As the
luminance decreases, the peak of the sensitivity shifts towards lower
wavelengths. The peak sensitivity under scotopic conditions is at 507 nm. These
data are known as the spectral luminous efficiency functions or the V() curves.
There is not an equivalent standard for the mesopic region.


2.1      What is the incidence of missing standards?
There are few direct incidences of this situation:

      Generally, luminance levels for roadway lighting fall under the mesopic
       range. Currently the properties of all light sources are quantified using
       human photopic response. Under mesopic conditions, this will be clearly
       inadequate as measure of the real performance of the source. This forbids
       to lighting industry, which is aware of the problem, to display the real
       characteristics of novel lamps. For example a 150W metal halide
       discharge lamp with a ceramic arctube displays an efficacy of 90-95 lm/W,
       but the same lamp attains more than 150 “mesopic-lumen” per Watt, but
       the “mesopic-lumen” is not defined because the equivalent V m() for
       mesopic conditions is not standardised…

      Another key feature of the energy saving characteristics of innovative
       urban lighting systems is the ability to tune the output of the light source to
       match the mesopic response of the human eye and to confirm its
       acceptability under roadway and outdoor lighting schemes. Once again all
       established standards assume photopic conditions and this seriously jams
       (or, in worse cases depending on national standards, forbids) the
       deployment of new technologies, like dimmable metal halide discharge
       lamps, just by classifying the system as “non-compliant” to the (although
       inexistent) standards…

      Last, but not least, due to the absence of standards, it is today impossible
       to express with certitude anything about colour-rendering characteristics of
       the new lamps under dimming operation that corresponds fully to mesopic
ID Number: 00042-KN406

Page 5 of 17
       conditions. However, city engineers are aware that high colour rendering
       rhymes strongly with higher security for pedestrians and drivers,
       embellishment of the urban areas, increased attractiveness of commercial
       and tourist venues. To the other side elected persons in the city councils
       know that better light in the city rhymes with satisfied citizens, happy
       merchants, increased number of tourists, etc… Lack of standards
       concerning colour rendering under mesopic conditions prevents the
       lighting industry from convincing city engineers, as well as specialised
       enterprises in illumination, that the novel technologies offer all the above-
       cited advantages…

What damage for several economic sectors!


2.2      What gains are expected if new standards are adopted?
First of all, highly efficient light sources are one route to help governments reduce
the power consumption of lighting systems and hence reduce the emissions of
greenhouse gases in line with the Kyoto agreements. Today more than 2,100
TWh of electrical energy per annum are used worldwide for lighting, which
corresponds to 12-15% of the global electricity production. Urban/road lighting
corresponds to 8% of the above-cited figure [6]. The following figure gives more
details about the world energy consumption for lighting.




               Figure 1: Energy consumption for lighting, sector-by-sector [6]

Further lamp optimisation, coupled with optimisation of the ballast characteristics
and optical properties of the fixture, together with potential gains from sources
specifically designed for human visual perception under mesopic lighting

ID Number: 00042-KN406

Page 6 of 17
conditions has the potential to dramatically reduce energy usage in roadway
lighting. Such systems would simultaneously improve the visual quality, energy
efficiency and effectiveness of lighting schemes. For Europe, energy savings of
more than 20 TWh could be expected. This energy saving corresponds to a
reduction of over 10 million tons of CO2 or the output from several big power
stations. The resulting reduction in greenhouse gas emissions would be
equivalent to approximately 4 per cent of the total EU commitment to the Kyoto
agreement.

Second, according to an OECD report [7], the lighting industry is one of the four
fastest growing sectors worldwide (growth rate in the order of 1% per annum).
Today, and since several years now, the most common products used for urban
lighting are in “saturation”: In many parts of Europe the low-pressure sodium lamp
(LPS) is used extensively for roadway lighting. LPS lamps are easily recognised
by their distinctive orange coloured light and long, narrow shape. Although the
LPS lamp (covered by the international standard IEC 60192) is a very efficient
light source it has very poor colour rendering properties because of the
monochromatic output. As more research is carried out there is growing evidence
that good colour rendering provides increased safety for roadway lighting. This
product can be considered as “decadent” and high effort is employed in order to
limit the use of LPS. The most widespread light source for roadway use in Europe
is the high-pressure sodium (HPS) lamp covered by IEC 60662. These lamps
offer many attractions over LPS lamps but still produce a noticeably orange light
that many users dislike, and as we will see later in this paper, HPS under
dimming conditions degrades seriously the security conditions in the road. The
lighting industry continually searches for new technologies that could replace (at
least partially) the existing products. These new technologies should be more
efficient, better adapted to the modern demand and, of course, guaranteeing a
higher growth rate for the industry itself. In recent years white light sources, such
as metal halide discharge lamps (covered by IEC 61167), have begun to be used
for road and street lighting not only to improve the aesthetic quality of installations
but also to quantify improvements in road safety where light sources with good
colour rendering properties are used. The recent emergence of metal halide
discharge lamps with ceramic arctubes has led to a new generation of highly

ID Number: 00042-KN406

Page 7 of 17
efficient light sources with high colour rendering indices, which are highly suited
to outdoor lighting applications. This latest generation of metal halide discharge
lamps offer additional benefits to the earlier lamps, which have arctubes
fabricated from fused silica. The most important of these benefits for roadway
lighting are higher efficacy and longer life.

Last, but not least, cities and citizens may draw substantial benefits from new
technologies. In the first example we should state the fact that cities may achieve
substantial economies on their urban lighting budget. A concrete case will be
stated in the following paragraph, it concerns a pilot system installed in a
medium-size city in southwest France (Albi capital city of Tarn, 70 000
inhabitants). In that case, more than 40% of economies have been accrued within
one year. On the other hand the road-users feel in better security conditions when
strolling or driving in the city. A user-satisfaction study in Albi has shown that
more than 90% of the implied citizen sample considers that the new lighting
scheme represents a significant advance in the quality of life (better visibility,
better colours, less glare) in the city. In addition almost 100% of them encourage
the city authorities to continue with such experiments and regret (after informing
them about the present situation) that missing standards could put the brakes on
such advancements.


3      How European research can point-out missing
standards
In this paragraph we will see how intense collaborative research can help to
point-out clearly the problem and put standard-making organisms under
(additional) “pressure”. The discussion will be illustrated by using the results
obtained recently in the frame of a large European-funded research project under
FP5-Energie program targeting the development of intelligent urban lighting
schemes.

The collaborative project, „NumeLiTe‟ (NNE5-2001-282), brought together, for a
3-year period, 11 partners from industry and academia in six European countries
and was established to harness expertise in all areas of component design to
generate a set of interacting tools to facilitate total system design. The final goal
ID Number: 00042-KN406

Page 8 of 17
of the project was to demonstrate the new road lighting system in the city of Albi.
This historic city is famous for its Saint Cecile cathedral (Unesco world-historical
heritage) and collection of Toulouse Lautrec paintings.

The research carried-out in the frame of NumeLiTe led to the creation of more
efficient, ceramic enveloped metal halide lamps with 97 lm/W radiant efficiency,
and more than 90 Colour Rendering Index, supporting dimming without important
degradation of their characteristics. The new lamps present a 16,000 h lifespan
when similar existing technologies, i.e. metal halide discharge lamps with silica
envelopes cannot reach beyond 10,000 h. New appropriate reflectors ensured
that the maximum quantity of light is directed onto the street and avoid skies light
pollution. New electronic ballast with dimming capabilities and communicating
through DALI (Digital Addressed Lighting Interface) protocol has been also
developed. Centralised control allows a better and less expensive lighting
scheme for municipalities.

The trial installation (demonstrator) of 131 lighting points illuminating 6 streets
downtown, was completed during the summer of 2004. Prior to the
commissioning of this installation, off-road trials were be carried out under
carefully controlled conditions on a test track in the UK. Since the end of the
project, the demonstrator has been used to collect data on reliability,
maintenance costs, energy costs and public perception.

The key-future demonstrated by NumeLiTe was the possibility to adapt the new
light sources to the human mesopic vision. This allows to decrease input power
during less busy hours with out degrading visibility conditions or lamp lifespan.
The following illustrates how light source efficacy is affected under mesopic vision
operating conditions.

In the mesopic range the effectiveness of the light source is critically dependent
on the spectral output. Depending on the energy distribution, the source
effectiveness may either rise or fall. To simplify, figure 2 illustrates two theoretical
spectral power distributions.




ID Number: 00042-KN406

Page 9 of 17
                                                                   Spectrum of 2-l ine lamp

                                              20




                     Rad iat ed power ( W )
                                              15


                                              10


                                              5


                                              0
                                                   400             500                      600            700
                                                                           w avelengt h (nm )



                                                                   Spectrum of 3-l ine lamp

                                              20
                     Rad iat ed power ( W )




                                              15


                                              10


                                              5


                                              0
                                                   400             500                      600            700
                                                                           w a ve lent h (nm )




                                                   Figure 2: Illustrative spectral power distributions

The 2-line spectrum has energy emitted at the two wavelengths corresponding to
emissions from the thallium line at 535 nm and the sodium lines at 590 nm. The
3-line spectrum has, in addition, radiation corresponding to emission from the
indium line at 450 nm. The total power radiated in each spectrum is 30 W. Table
1 below summarises the calculated colour coordinates (x & y), correlated colour
temperature (CCT) and photometric performance of these spectral power
distributions, using existing standards under photopic conditions.

 Table 1: Calculated colour and photometric properties of theoretical spectral power distributions


                                                               x              y                  CCT    Photopic lumens

         2-line spectrum                                    0.481         0.513                  3100        16900
         3-line spectrum                                    0.340         0.450                  5300        14800



Figure 3 shows these two spectra in relation to the scotopic, mesopic and
photopic eye sensitivity curves (all normalised to unity).




ID Number: 00042-KN406

Page 10 of 17
                                                                                  1.2                                                                                                                       4.500
                                                                                                                                                                    p h o t o p ic re s p o n s e V
                                                                                                                                                                                                            4.000
                                                                                                                                                                    s c o p t ic re s p o n s e
                                                                                  1.0




                                                                                                                                                                                                                    sp e c tr a l i n te n si ty (W / n m )
                                                                                                                                                                    m e s o p ic re s p o n s e             3.500




                                               r e l a ti v e se n si ti v i ty
                                                                                  0.8                                                                               2 -lin e s p e c t ru m                 3.000

                                                                                                                                                                                                            2.500
                                                                                  0.6
                                                                                                                                                                                                            2.000

                                                                                  0.4                                                                                                                       1.500

                                                                                                                                                                                                            1.000
                                                                                  0.2
                                                                                                                                                                                                            0.500

                                                                                  0.0                                                                                                                       0.000
                                                                                        350     400      450    500             550               600         650               700                   750

                                                                                                                      w a v e l e n g th (n m )




                                                                                  1.2                                                                                                                       3.000
                                                                                                                                                                     p h o t o p ic re s p o n s e

                                                                                  1.0                                                                                s c o t o p ic re s p o n s e          2.500




                                                                                                                                                                                                                    sp e c tr a l i n te n si ty (W / n m )
                                                                                                                                                                     m e s o p ic re s p o n s e
                                               r e l a ti v e se n si ti v i ty




                                                                                  0.8                                                                                3 -lin e s p e c t ru m                2.000



                                                                                  0.6                                                                                                                       1.500



                                                                                  0.4                                                                                                                       1.000



                                                                                  0.2                                                                                                                       0.500



                                                                                  0.0                                                                                                                       0.000
                                                                                        350     400      450    500             550               600         650               700                   750

                                                                                                                      w a v e l e n g th (n m )




 Figure 3: Relationship between scotopic, mesopic and photopic response curves and illustrative
                                                                                                         Spectral Power Distributions

The mesopic curve employed here is derived by the method described by He et
al. [8] at a luminance of 0.5 cd m-2.

We have used the term „mesopic-lumens‟ in this work to represent the sum of the
product of the spectrum energy and the mesopic response curve at each
wavelength. Figure 4 shows the calculated mesopic lumens as a function of
luminance for the two spectra.

                                                                                                    Mesopic lumens as a function of luminance

                                 25000
     calculated mesopic lumens




                                 20000


                                 15000
                                                                                                                                                                                                                                                              3 - line sp ect r um
                                                                                                                                                                                                                                                              2 - line sp ect r um
                                 10000


                                 5000


                                    0
                                         0.0                                                  0.5                1.0                                    1.5                                       2.0
                                                                                                          lum inance cd.m -2




                                                                                        Figure 4: Mesopic lumens as a function of luminance
ID Number: 00042-KN406

Page 11 of 17
Clearly as the luminance reduces, the mesopic lumens will decrease in the case
of the 2-line spectrum and increase in the case of the 3-line spectrum. It is this
effect which gives rise to the opportunity to design spectral power distributions
which show an increase in lighting effectiveness as light levels are reduced from
the photopic region This increase in effective efficacy at low luminance, such as
used in roadway lighting, may then be taken as an energy saving instead of an
increase in the „effective‟ light level.

This principle applies to real lamps. As shown on Figure 5, reducing the power
input of a 150W metal halide discharge lamp with a ceramic arctube (C_MHL) will
enhance important variations of the photometric properties. Colour temperature
(CCT) will increase and Colour Rendering Index (CRI) will decrease.

                                                      CRI and CCT vs lamp power
                           7000                                                                                     100



                           6000
                                                                                                                    90
                                                                                                    CRI
                                                                                                    CCT [K]
                           5000
                                                                                                                    80


                           4000




                                                                                                                          CRI(a. u.)
                  CCT(K)




                                                                                                                    70

                           3000


                                                                                                                    60
                           2000


                                                                                                                    50
                           1000



                              0                                                                                     40
                                  50   60   70   80    90      100       110      120   130   140             150
                                                             Power(W)




   Figure 5: Influence of input power to the Correlated Colour Temperature (CCT) and Colour
                                       Rendering Index (CRI) of 150 W C-MHL

From these observations, we could conclude that dimming from 150W to 70W
(corresponding to 53% of power reduction) would involve an important colour
variation with a shift to bluish colours. This variation seems to be beneficial for
visibility under mesopic vision conditions because under photopic conditions, the
sensitivity of the human eye peaks at 555 nm and then 1W radiate power at this
wavelength corresponds, according to the standards, to 683 lm. As the luminance
decreases, the peak of the sensitivity shifts towards lower wavelengths. The peak
sensitivity under scotopic conditions is at 507 nm (more bluish colour) but then
1W of input power would correspond to 1700 “scotopic-lumens” (also not
normalized unfortunately…).



ID Number: 00042-KN406

Page 12 of 17
A calculation route has been established to assess the performance of output
spectra at any luminance level in terms of its potential efficacy under mesopic
lighting conditions. The following currently available HID sources, some of which
are used for roadway lighting, have been assessed: high-pressure mercury
vapour (HPMV); high-pressure sodium (HPS); low-pressure sodium (LPS); CMH
with CCT 3000K and 4000K. Figure 6, illustrates the results of the calculations
where the light source outputs have been normalised to 1000 photopic lumens.
Based on the above calculations, it can be seen that a broad band metal halide
spectrum at CCT 4000K at a luminance level 0.1 cd.m -2 is approximately 2.5
times as efficient as the spectrum from a high pressure sodium lamp.


                                   1800
                                   1600
         output (mesopic lumens)




                                   1400
                                   1200                                                  LP S
                                   1000                                                  HP MV
                                    800                                                  HP S 150W
                                    600                                                  CMH 4K 150W
                                    400
                                    200
                                      0
                                          0   0,2   0,4     0,6      0,8       1   1,2
                                                     luminance (cd/m2)



  Figure 6: Comparison of mesopic luminous efficiency for some light sources as a function of
                                                          photopic luminance

The conclusion is strait-forward: Under dimming conditions corresponding to
mesopic vision, novel metal halide lamps ensure better visibility conditions than
any other existing technology. In other words, lighting systems based in that
principle would achieve better security and substantial energy savings. Of course
white light generated by these lamps increase the comfort and it is better
perceived by the end-user. All the above-announced points have been tested and
confirmed with NumeLiTe project by means of the Albi based demonstrator.

The new road lighting installation in the city of Albi has given energy savings of 40
to 45% per annum. A strait-forward extrapolation of these energy savings in
western European level shows that gains attain more than 1 MTEP (million
tonnes equivalent petrol) per annum. In addition an end-user satisfaction study


ID Number: 00042-KN406

Page 13 of 17
targeting pedestrians, drivers and merchants confirmed that this system offers a
better quality of life and security to road uses in urban environment.

Another approach to determining the appropriate lighting conditions for road and
street lighting is the measurement of reaction times, for example, as described in
reference [8]. Reaction time measurements under laboratory conditions carried
out as part of the NumeLiTe project will be published elsewhere [9].

A second objective to the project was to provide evidence to support possible
changes to standards and regulations that could further reduce the amount of
energy, and other resources, than are consumed in road and street lighting.

To this point of view, the trial installation (or any similar lighting scheme) has to
satisfy the requirements of the existing standards and regulations. If it can be
demonstrated that the luminance, as measured in photopic terms, can be
reduced without compromising safety for road users and pedestrians, then there
are further opportunities for energy savings.

A key element in changing the regulations for road and street lighting is the
definition of an internationally agreed value for the V()-curve for mesopic
conditions. This will provide a new measure of light source output, similar to the
term „mesopic lumens‟ used in this paper. This will then allow the characteristics
of various light sources to be compared against internationally recognised criteria
applicable to road and street lighting.

Furthermore, urban lighting schemes are subject of various legislations and
standards (except the above stated problems linked to visual requirements):
Lamp holders are covered in IEC 60838. The family of standards under IEC
60598 cover the requirements for luminaries, IEC 60598-2-3 being dedicated to
luminaries for road and street lighting. Safety specification for discharge lamps,
excluding fluorescent types, is given in IEC 62035. The lighting installation must
satisfy the national requirements for cabling and fusing; these will be derived from
IEC standards. The installation must also meet the requirements for
electromagnetic compatibility as defined in IEC 61547. It is interesting to note that
although there are international standards for electronic ballasts for fluorescent
lamps, international standards for ballasts for high intensity discharge lamps are
restricted to electromagnetic types, IEC 60923. However, activities are well
ID Number: 00042-KN406

Page 14 of 17
advanced in the preparation of a proposed standard for electronic ballasts for low
wattage high intensity discharge lamps but almost nothing exists for ballasts that
are adapted to urban lighting schemes.

The communication protocol (DALI) introduced by NumeLiTe project for urban
lighting schemes and adopted recently by other manufacturers is covered by
IEC60929 and ANSI C82.11 standards for interior lighting applications only.
Nothing exists for urban lighting. One can be see on that novel and exciting
opportunities for research in close relation with standard making.

It should be underlined here that the adoption of international standards has
another huge benefit for specifiers and purchasers of systems that are
themselves composed of sub-systems that are subject to standards. As have
been illustrated above lighting systems are such an example where many of the
components, e.g. the light source, the light source holder, the ballast, the fitting,
are defined by internationally accepted standards. Thus, the end user has a
choice of supplier for each of the sub-systems knowing that they will be
compatible with the other components. This gives significant advantages in terms
of maintaining competition for the quality, cost and availability between
component suppliers.


4     Are the standards the only way to promote
novel technologies?
To achieve this ambitious objectives, like these discussed in this paper which is a
very specific case, we need to follow an integrated strategy based on scientific
but   widely    accessible   arguments,    training,   socio-economic    knowledge,
standardization, inciting measures and extra legislation. This is the only way to
realise new generations of more reliable and more efficient products. The
following “fishbone” graphics summarises all the above issues.




ID Number: 00042-KN406

Page 15 of 17
                                    Inciting
    New Standards
                                   Measures

                           Extra                 Market
                         Legislation           knowledge
    What next ?




                                                                   Next product
                                                                    generation



                       Technological             Social
                       Development             knowledge

    High Quality
     Research                     Training


Standards occupies a predilection place in this strategy and should, for sure, be
integrated in any reflection from the real beginning, but standards are not “the
panacea” that solves all problems found the way leading from existing (or brand
new) concepts to new generations of products.

In all above it shown that, Standards are as important as patents as a source of
knowledge for applied researchers and experimental developers. However, basic
researchers rely more on scientific publications. In addition, the structure of
standardisation processes, e.g. their speed and flexibility, is perceived as a major
problem for the effective transfer of knowledge from research to standardisation.
Thus an important dimension of the interaction between research and
standardisation is the question of how it takes place within organisations. An
integrated strategy as proposed in the previous paragraph, and as has been
experimented (partially) in the frame of NumeLiTe seems to be well adapted for
overcoming these problems.


5                 References
[1] Purkinje, J. (1823) Beobachtungen und versuche zur physiologie der sinne.
Neue beitra¨ge zur kenntniss des sehens in subjectiver hinsicht. Reimer, Berlin.

[2] Palmer, D. A. (1966) A system of mesopic photometry. Nature 209, 276–281.

ID Number: 00042-KN406

Page 16 of 17
[3] Palmer, D. A. (1967) The definition of a standard observer for mesopic
photometry. Vision Res. 7, 619–628.

[4] Palmer, D. A. (1968) Standard observer for large field photometry at any level.
J. Opt. Soc. Am. 58, 1296–1299.

[5] British Standard (2003) BS 5489-1: Code of Practice for the Design of Road
Lighting, British Standard London.

[6] Mills E. (2002), “Why we‟re here: The $230-billion global lighting energy bill”,
5th Right Light conference (May 2005), Nice, France

[7] OECD Environment Directorate and International Energy Agency (2001):
Information     Paper   COM/ENV/EPOC/IEA/SLT(2001)3,         An   initial   view   on
methodologies for emission baselines: Energy efficiency case.

[8] He Y., M.Rea, A.Bierman and J.Bullough (1997) J. Illumin. Eng. Soc., p.125-
138.

[9] Raynham P., M.H.Girach, S.A.Mucklejohn and B.Preston, to be published




ID Number: 00042-KN406

Page 17 of 17