25TH SESSION OF THE CIE - SAN DIEGO - 25 JUNE - 2 JULY 2003

PROCEEDINGS Volume 1 Part 2

COMMISSION NTERNAJIONALE DE L'ECLAIRAGE INTERNATIONAL COMMISSION ON ILLUMINATION INTERNATIONALE KLEUCHTUNGSKOMMISSION m

CIE 152:2003 ISBN 3901 90621 5

Ezrati: Solar protection glazing and Muséum lighting

SOLAR PROTECTION GLAZING AND MUSEUM LIGHTING Jean-Jacques Ezra t i , Eliyahu Ne'eman

A B S T R A C T

Daylight is widely considered as the most suitable light source for viewing works of art. However, if not properly controlled, it présents potentially harmful effects to their conservation, in particular by the energy of short wavelengths - the ultraviolet radiation. Damage quite as important, can be caused by the infra-red radiation content of daylight which may bring about undesirable rises and variations in température of the objects. The potential damage inside the exhibition space when glazing and protection films are clear is obviously unacceptable. But the situation can be différent when suitable solar protection glazing and films are applied to the daylight admitting openings. The présent study has started with a colorimetric modeling, and has been followed by validation through real expérimentation. The study demonstrates that the proposed installation of solar protection tinted glazing does not inhibit the good viewing of coloured objects.

Keywords: Solar protection, Muséum lighting, Famsworth-Munsell 100 Hue test.

R E S U M E

La lumière du jour est idéale pour la vision des œuvres d'art, elle représente néanmoins un danger pour leur conservation matérielle notamment par les effets énergétiques néfastes des courtes longueurs d'onde notamment des ultraviolets. Mais des dommages tout aussi importants peuvent être causés par les infrarouges, eux aussi très présents dans la lumière du jour et par les élévations et les variations de température. Cela est le cas dans les espaces d'exposition où les vitrages et films de protection ultraviolet sont incolores. Cette situation sera nettement différente si on utilise des vitrages ou des films de protection solaire. La présente étude démontrera à partir d'une modélisation colorimétrique, validée par une expérimentation réelle, que la pose d'un vitrage de protection solaire, serait-il teinté, n'altère pas la vision des oeuvres peintes.

Mots clefs : protection solaire, éclairage muséographique, test Farnsworth-Munsell 100 Hue

Z U S A M M E N F A S S U N G

Obwohl das Tageslicht fur die Betrachtung von Kunstwerken idéal ist, stellt es trotzdem eine Gefahr fur ihre matérielle Erhaltung, insbesondere durch die unheilvollen Auswirkungen der stratlung im Bereich der Energie der kurzen Wellenlângen wie der ultravioletten, dar. Das stadig gegenwartige Infrarotlicht im Tageslicht stellt durch die Temperaturerhôhungen und die variationen eine ebenso grope Gefahr dar. Die vorgestellter Untersuchungen beweisen, wie durch theoretische kolorimetrische Modelle, Expérimente validiert werden, die zeigen, dass das Anbringen von einer eventuell getônten Sonnenschutzverglasung, die Betrachtung der Malereien nicht beeintrâchtigt.

INTRODUCTION

Many muséum curators consider daylight as the most suitable light source for viewing works of art. However, conservator-restorers see it as the most harmful source for their conservation. Both are right but... daylight as a source works very well for the viewing of colours. The harmful effects for conservation originate from three parts of solar optical radiation: • The high level of ultraviolet radiation. • High level of visible light. • The very high intensity solar infra-red radiation. Consequently, we have to take the necessary measures to eliminate or at least reduce the harmful conséquences of thèse radiations. It is suggested here that the solar protection bronze tinted glazing, or a clear glass with an appropriate film can be used as an adéquate solar protection. However curators will certainly consider such a solution as totally unacceptable because the colours seem changed! In our view this is nevertheless not such a critical problem. We suggest that what matters in the satisfaction of the viewing coloured artwork, is not the colours themselves but the différence between one colour and the other.

Ezrati: Soiar protection glazing and Muséum lighting

T H E S T U D Y

1. The colorimetric study

We used the Farnsworth-Munsell 100-Hue test ' to find if discrimination of two adjacent couloured capsules (caps) is similar under the CIE standard illuminant D65 as under the CIE standard illuminant A. We measured the chromatic co-ordinates L*a*b* of every coloured cap, and then calculated the colour différence " AE*ab between one cap and the adjacent one, first with the CIE standard illuminant D65 (Curve 1, Figure 1) and then with the CIE standard illuminant A (Curve 2, Figure 1). Finally we calculated the déviation of the colour différences of the results (Curve 3, Figure 1). Thèse curves demonstrate that for both illuminants (D65 and A), the chromatic co-ordinates for the same cap are quite différent, but the différence AE*ab for the colours remains similar from one cap to another, and it is lower than 2 "'.

Figure 1: The colorimetric measurement -Colour différences between Farnsworth-Munsell caps

and the adjacent ones

7 -i

Capsule numbers

2. The psychometric study The colorimetric study described above can be considered as a modelling of a real case, namely the view of coloured artifacts under daylight filtered by transparent glazing, which obviously transmits the différent wavelengths of the visible spectra, as if they were illuminated by the CIE standard illuminant D65. On the other hand a glazing or a film with sun protection may transmit the light as if it is similar to CIE standard illuminant A. For the présent study we chose a film with a tint similar to that used in sunbathing devices, this film has the same visual effect than a température conversion filter used in photography to convert the daylight spectrum (5700 K) into artificial light spectrum (3200 K). A "ionized sputter" dark bronze* film used for sun protection was chosen in this psychometric experiment. Sixteen employées of the laboratory participated to this study. After spending a suitable time for adaptation, about 3 minutes, they performed the Fernworht-Munsell 100-hue colour vision test according to the normal protocol that is in a time limit, under a normalized light D65 from a colorimetric contrat lightbox v . At the end of the test (the basic test) we counted points following the number of noticed faults. The majority of participants in such tests may make a limited amount of errors due to stress, inattention or fatigue rather than real errors of vision. For a simple inversion counting for 4 points, a margin of error of 8 points during the full test, has been considered as acceptable.

Ezrati: Solar protection glazing and Muséum lighting

About one month later the same persons repeated the same test under light which imitated the sunlight transmitted through a sun protection film. The room in which the test was carried out was lit only by a desk lamp and the lightbox. After a few minutes of adaptation (no more than at the first test) the test was done according to the same protocol as the first time. The results show that the rate of errors is not différent than in the basic test (figure 2, table 1).

Figure 2 : Graph of the Farnsworth - M u n s e l l test faul t points repartition

Human tes tnumbers

Table 1: Results of Farnsworth - Munsell tests

Order# Genre Age D65 différence D65+film 1 M 4 +4 8 2 M 45 0 0 0 3 F 40 8 0 8 4 F 25 8 -8 0 5 F 25 4 0 4 6 F 34 24 -16 8 7 F 25 20 -8 12 8 F 30 12 -8 4 9 F 25 12 0 12 10 M 49 16 0 16 11 M 25 0 0 0 12 F 26 4 0 4 13 M 29 8 +4 12 14 M 29 20 -6 14 15 M 42 16 -4 12 16 F 33 8 -8 0

164 -50 114

Ezrati: Solar protection glazing and Muséum lighting

R E S U L T S 1. Coloured caps from the Farsworth-Munsell 100-Hue have been tested under CIE illuminant D65 source a typical continuous spectrum source, and then under a CIE illuminant A . The différences in AE*ab between two adjacent caps are similar. 2. The same test, carried out by a psychometric method, with human observers has confirmed thèse results.

C O N C L U S I O N This study has revealed that by using solar protection glazing or solar protection films, it is possible to obtain in the muséum good colour perception of coloured art works. However, further field validation of this conclusion is needed.

B IBL IOGRAPHY

CIE 15.2 - 1986 C o l o r i m e t r y CIE 142 - 2001 I m p r o v e m e n t t o i n d u s t r i a l c o l o u r - d i f f e r e n c e évaluation N E ' E M A N , E., CUTTLE , C. (ed), C o n t r o l o f d a m a g e t o muséum o b j e c t s b y o p t i c a l r a d i a t i o n , Report of the CIE technical commitee 3-22, CIE, 2003. HUNT, R. W. G. , M e a s u r i n g c o l o u r s , 2 n d éd., Ellis Horwood, London, 1995 [1991] M U N S E L L , A., A c o l o r n o t a t i o n , Munsell color company, Baltimore 1981 [1946]. K O V A L S K Y , P., V i s i o n e t m e s u r e d e l a c o u l e u r , Masson, Paris, 1990. S E V E , R., P h y s i q u e d e l a c o u l e u r . D e l ' a p p a r e n c e colorée à l a t e c h n i q u e colorimétrique, Masson, Paris, 1996

A U T O R S

EZRATI jean-jacques, Lighting adviser, Research and Conservation Center of French Muséums, 6 rue des Pyramides, 75001 Paris, France. iean-iacques.ezrati(a).culture.fr N E ' E M A N Eliyahu, BSc , Dipl Ing, M S c DSc, Consulting Engineer, Faculty of architecture and Town Planning, Technion, Israël Institute of Technology, Haifa, 32000, Israël, arrentOKaitx.technion.ac.il

The Farnworth-Munsell 100 hue test is an effective method for determining colour vision abnormalities and testing colour discrimination. The set consists of 4 trays containing a total of 85 colored référence (12 mm diameter) caps. They are chosen from the Munsell atlas, of a lightness (Value) and saturation (Chroma). According to the chromatic order, they should be arranged in a continuous succession of hue.

u We base our considération on the formula of colour différence CIEL*a*b* 1976. It is an old formula, adapted to measure colour différence between two objects, but we apply it for finding the déviation between thèse différences.

U l According to Hunt (see Bibliography, p. 141), a déviation of 2 between two adjacent coloured caps of Chroma, is considered as a just perceptible différence.

i v Film SBI 25 R S 500 S R Xylar filter (Vecteurs Applications, 37 boulevard Jean Allemane 95100 Argenteuil, France).

v Light box GAMAIN M 6584 HF (31, rue des Annelets 75019 Paris, France).