Tài liệu The SLL Lighting Handbook potx

In 1924, the CIE adopted the Standard Photopic Observer to characterise the spectral sensitivity of the human visual system by day.. The Standard Scotopic Observer is used by the lightin

The Society of Light and Lighting

is part of the Chartered Institution

of Building Services Engineers

The Society of

Light and Lighting

The SLL Lighting Handbook

The SLL Lighting Handbook

222 Balham High Road, London SW12 9BS

+44 (0)20 8675 5211www.cibse.org

This document is based on the best knowledge available at the time of publication However,

no responsibility of any kind for any injury, death, loss, damage or delay however causedresulting from the use of these recommendations can be accepted by the Chartered Institution

of Building Services Engineers, The Society of Light and Lighting, the authors or othersinvolved in its publication In adopting these recommendations for use each adopter by doing

so agrees to accept full responsibility for any personal injury, death, loss, damage or delay arisingout of or in connection with their use by or on behalf of such adopter irrespective of the cause

or reason therefore and agrees to defend, indemnify and hold harmless the Chartered

Institution of Building Services Engineers, The Society of Light and Lighting, the authors andothers involved in their publication from any and all liability arising out of or in connectionwith such use as aforesaid and irrespective of any negligence on the part of those indemnified.The rights of publication or translation are reserved

No part of this publication may be reproduced, stored in a retrieval system or transmitted inany form or by any means without the prior permission of the publisher

Note from the publisher

This publication is primarily intended to give guidance It is not intended to be exhaustive ordefinitive, and it will be necessary for users of the guidance given to exercise their own

professional judgement when deciding whether to abide by or depart from it

© February 2009 The Society of Light and Lighting

The Society is part of CIBSE, which is a registered charity, number 278104

ISBN 978-1-906846-02-2

Project and Print management by

entiveon Ltd www.entiveon.com

Design, linework and typsetting by

Squarefox Design Ltd www.squarefox.co.uk

Printed in England on FSC certified Mixed Sources paper by

Stones the Printers Ltd www.stonestheprinters.co.uk

ii

The Society of Light and Lighting

is part of the Chartered Institution

of Building Services Engineers The Society of

Light and Lighting

The SLL Lighting Guides, which provide detailed guidance on specific lighting applications

The SLL Lighting Handbook has been written to forge a link between them It is designed to be

complementary to the SLL Code for lighting but to go beyond it in terms of applications and

background information without getting into the fine detail of the Lighting Guides

The SLL Lighting Handbook is intended to be the first-stop for anyone seeking information on

lighting It is aimed not just at lighting practitioners but also at lighting specifiers and students oflighting For all three groups, we have tried to make it comprehensive, up-to-date and easily

understandable The contents summarise the fundamentals of light and vision, the technology oflighting and guidance on a wide range of applications, both interior and exterior

Authors

Peter Boyce PhD, FSLL, FIESNA

Peter Raynham BSc MSc CEng FSLL MCIBSE MILE

Acknowledgements

John Fitzpatrick

Lou Bedocs (Thorn Lighting)

Ted Glenny (Philips Lighting)

Jennifer Brons for Figure 20.2

Kit Cuttle for Figures 13.1 and 13.2

Lighting Research Center for Figures 9.1, 10.3, 18.8, 18.9 and 20.3

McGraw Hill Inc, for Figures 2.4 and 2.9

Mick Stevens for Figures 20.3 and 22.1

The Illuminating Engineering Society of North America for Figures 1.5, 1.6, 1.7, 1.8, 2.8 and 2.13Philips Lighting, iGuzzini Illuminazione, Havells Sylvania & Luxo

Charlotte Wood Photography for Figures 14.1, 14.2 and 14.3

Editors

Stuart Boreham (entiveon Ltd.)

Peter Hadley (Squarefox Design Ltd.)

SLL Secretary

Liz Peck

CIBSE Editorial Manager

Ken Butcher

CIBSE Director of Information

iv

CONTENTS

PART 1: FUNDAMENTALSChapter 1: Light

1.1 The nature of light 1.2 The CIE standard observers 1.3 The measurement of light — photometry

1.3.3 Illuminance 1.3.4 Luminance 1.3.5 Reflectance

1.4 The measurement of light — colourimetry

1.4.2 The CIE colour spaces 1.4.3 Correlated colour temperature 1.4.4 CIE colour rendering index

1.4.6 Scotopic/photopic ratio 1.4.7 Colour order systems

Chapter 2: Vision

2.1 The structure of the visual system

16 16 16 17 19 22 23 23 24 24 25 26 26 26 28 28 30 31 32 32 34 37 37 37 38 39 40 41

3.3.10 Electroluminescent 3.4 Electric light source characteristics

48 48 49 51 51 51 52 52 52 52 52 54 57 57 59 60 64 66 69 70 74 75 76 77 77 77 78 78 78 78 78 79 79 79 82 82 82 83

Chapter 6: Lighting design

6.1 Objectives and constraints 6.2 A holistic strategy for lighting

6.3 Basic design decisions

84 84 85 86 91 91 93 94 94 94 98 100 100 105

109 109 114 114 115 115 115 116

117 117 118 118 119 120 120 121 121 121 122 124 124 124 125 128

129 131 133 133

Chapter 8: Emergency lighting

8.1 Legislation and standards 8.2 Forms of emergency lighting

8.3 Design approaches 8.4 Emergency lighting equipment

Chapter 9: Office lighting

9.1 Functions of lighting in offices 9.2 Factors to be considered

140 141 141 142 142 144 144 144 145 145 146 147 148 148 149 149 149 150 153 153 153 153 154 154 155

156 156 156 157 157 158 158 159 159

9.4 Approaches to office lighting

Chapter 10: Industrial lighting

10.1 Functions of lighting in industrial premises 10.2 Factors to be considered

10.4 Approaches to industrial lighting

Chapter 11: Lighting for educational premises

11.1 Functions of lighting for educational premises 11.2 Factors to be considered

162 162 164 165 166 167 168 168 169 170 170 171 171 172 172

173 173 173 174 174 175 175 176 177 177 177 177 178 180 181 182 182 182 183 183 183 184

185 185 185 186 186 186 186 186

11.4 Approaches to lighting educational premises

Chapter 12: Retail lighting

12.1 Functions of retail lighting 12.2 Factors to be considered

12.4 Approaches to retail lighting

Chapter 13: Lighting for museums and art galleries

13.1 Functions of lighting in museums and art galleries 13.2 Factors to be considered

187 187 187 187 188 188 189 189 189 189 189 190 190 190 190 190

191 191 191 192 192 192 192 192 193 193 193 194 194 194 195

198 198 198 198 199 199 199 200 200 200 200 201 201 201 201 201 201 202

Chapter 14: Lighting for hospitals

14.1 Functions of lighting in hospitals 14.2 Factors to be considered

14.3 Approaches for the lighting of different areas in hospitals

wash and shower rooms

14.3.10 Operating theatres

Chapter 15: Quasi-domestic lighting

15.1 Functions of quasi-domestic lighting 15.2 Factors to be considered

16.2.3 Lighting recommendations for conflict areas 16.2.4 Coordination

16.2.5 Traffic route lighting design

203 203 203 203 203 204 204 204 205 205 205 206 206 207 207 207 211 211 211 212 212

214 214 214 214 214 215 215 216 216 217 217 217 218 218 219 219 219

220 220 220 223 224 225 225

16.3 Lighting for subsidiary roads 16.3.1 Lighting recommendations for subsidiary roads 16.3.2 Lighting design for subsidiary roads

16.4 Lighting for urban centres and public amenity areas 16.5 Tunnel lighting

Chapter 17: Exterior workplace lighting

17.1 Functions of lighting in exterior workplaces 17.2 Factors to be considered

17.2.1 Scale 17.2.2 Nature of work 17.2.3 Need for good colour vision 17.2.4 Obstruction

17.2.5 Interference with complementary activities 17.2.6 Hours of operation

17.2.7 Impact on the surrounding area 17.2.8 Atmospheric conditions 17.3 Lighting recommendations 17.3.1 Illuminance and illuminance uniformity 17.3.2 Glare control

17.3.3 Light source colour properties 17.3.4 Loading areas

17.3.5 Chemical and fuel industries 17.3.6 Sidings, marshalling yards and goods yards 17.4 Approaches to exterior workplace lighting

17.4.1 High mast floodlighting 17.4.2 Integrated lighting 17.4.3 Localised lighting

Chapter 18: Security lighting

18.1 Functions of security lighting 18.2 Factors to be considered 18.2.1 Type of site 18.2.2 Site features 18.2.3 Ambient light levels 18.2.4 Crime risk

18.2.5 CCTV surveillance 18.2.6 Impact on the surrounding area 18.3 Lighting recommendations

18.3.1 Illuminance and illuminance uniformity 18.3.2 Glare control

18.3.3 Light source colour properties 18.4 Approaches to security lighting

18.4.1 Secure areas 18.4.2 Public spaces 18.4.3 Private areas 18.4.4 Multi-occupancy dwellings 18.5 Lighting Equipment

18.5.1 Light sources 18.5.2 Luminaires 18.5.3 Lighting columns 18.5.4 Lighting controls 18.5.5 Maintenance

230 230 232 232 233

236 236 236 236 237 237 237 237 238 238 238 238 239 239 239 240 241 243 243 243 244

245 245 245 246 247 247 247 247 247 247 249 249 249 249 252 253 254 254 254 255 255 256 256

Chapter 19: Sports lighting

19.1 Functions of lighting for sports 19.2 Factors to be considered 19.2.1 Standard of play and viewing distance 19.2.2 Playing area

19.2.3 Luminaires 19.2.4 Television 19.2.5 Coping with power failures 19.2.6 Obtrusive light

19.3 Lighting recommendations 19.3.1 Athletics 19.3.2 Bowls 19.3.3 Cricket 19.3.4 Five-a-side football (indoor) 19.3.5 Fitness training

19.3.6 Football (Association, Gaelic and American) 19.3.7 Lawn tennis

19.3.8 Rugby (Union and League) 19.3.9 Swimming

19.4 Lighting in large facilities 19.4.1 Multi-use sports halls 19.4.2 Small sports stadia 19.4.3 Indoor arenas 19.4.4 Swimming pools

Chapter 20: Lighting performance verification

20.1 The need for performance verification 20.2 Relevant operating conditions 20.3 Instrumentation

20.3.1 Illuminance meters 20.3.2 Luminance meters 20.4 Methods of measurement 20.4.1 Average illuminance 20.4.2 Interior lighting 20.4.3 Exterior lighting 20.5 Measurement of illuminance variation 20.5.1 Illuminance diversity 20.5.2 Illuminance uniformity 20.6 Luminance measurements 20.7 Measurement of reflectance

Chapter 21: Lighting maintenance

21.1 The need for lighting maintenance 21.2 Lamp replacement

21.3 Cleaning luminaires 21.4 Room surface cleaning 21.5 Maintained illuminance 21.6 Designing for lighting maintenance 21.7 Determination of maintenance factor for interior lighting 21.7.1 Lamp lumen maintenance factor (LLMF) 21.7.2 Lamp survival factor (LSF)

21.7.3 Luminaire maintenance factor (LMF) 21.7.4 Room surface maintenance factor (RSMF)

257 257 257 258 258 258 259 260 261 261 262 263 264 264 265 265 266 266 267 267 267 268 268

270 270 271 271 271 272 272 272 274 275 275 276 276 276

278 278 278 280 280 280 280 281 281 282 284

21.8 Determination of maintenance factor for exterior lighting 21.9 Disposal of lighting equipment

Chapter 22: On the horizon

22.1 Changes and challenges 22.2 The changes and challenges facing lighting practice 22.2.1 Costs

22.2.2 Technologies 22.2.3 New knowledge 22.2.4 External influences 22.3 The evolution of lighting practice

Chapter 23: Bibliography

23.1 Standards 23.2 Guidance 23.3 References

Index

285 286

287 287 287 287 287 290 290

293 296 298 303

1.1 The nature of light

Light is part of the electromagnetic spectrum that stretches from cosmic rays to radio waves

(Figure 1.1) What distinguishes the wavelength region between 380–780 nanometers from

the rest is the response of the human visual system Photoreceptors in the human eye

absorb energy in this wavelength range and thereby initiate the process of seeing

Figure 1.1 A schematic diagram of the electromagnetic spectrum showing the location of the

visible spectrum The divisions between the different types of electromagnetic radiation are

indicative only

The sensitivity of the human visual system is not the same at all wavelengths in the range

380 nm to 780 nm This makes it impossible to adopt the radiometric quantities

conventionally used to measure the characteristics of the electromagnetic spectrum for

quantifying light Rather, a special set of quantities has to be derived from the radiometric

quantities by weighting them by the spectral sensitivity of the human visual system The

result is the photometry system (see Section 1.3)

The Commission Internationale de l’Eclairage (CIE) has established three standard

observers to represent the sensitivity of the human visual system to light at different

wavelengths, in different conditions In 1924, the CIE adopted the Standard Photopic

Observer to characterise the spectral sensitivity of the human visual system by day

Wavelength (m)

RADIO WAVES

MICRO WAVES

INFRA RED

ULTRA VIOLET

X RAYS

GAMMA RAYS

COSMIC RAYS

780 nm

380 nm VISIBLE

In 1990, in the interests of greater photometric accuracy, the CIE produced a Modified

Photopic Observer, having greater sensitivity than the CIE Standard Photopic Observer atwavelengths below 460 nm This CIE Modified Photopic Observer is considered to be asupplement to the CIE Standard Photopic Observer not a replacement for it As a result, theCIE Standard Photopic Observer has continued to be widely used by the lighting industry This

is acceptable because the modified sensitivity at wavelengths below 460 nm has been shown tomake little difference to the photometric properties of light sources that emit radiation over awide range of wavelengths It is only for light sources that emit significant amounts of radiationbelow 460 nm that changing from the CIE Standard Photopic Observer to the CIE ModifiedPhotopic Observer makes a significant difference to photometric properties Some narrow bandlight sources, such as blue light emitting diodes, fall into this category

In 1951, the CIE adopted the CIE Standard Scotopic Observer to characterise the spectralsensitivity of the human visual system by night The Standard Scotopic Observer is used by the lighting industry to quantify the efficiency of a light source at stimulating the rod

photoreceptors of the eye (see Section 2.1.4)

The CIE Standard and Modified Photopic Observers and the CIE Standard Scotopic

Observer are shown in Figure 1.2, the Standard and Modified Photopic Observers havingmaximum sensitivities at 555 nm and the Standard Scotopic Observer having a maximumsensitivity at 507 nm These relative spectral sensitivity curves are formally known as the

1924 CIE Spectral Luminous Efficiency Function for Photopic Vision, the CIE 1988 ModifiedTwo Degree Spectral Luminous Efficiency Function for Photopic Vision, and the 1951 CIESpectral Luminous Efficiency Function for Scotopic Vision, respectively More commonly,they are known as the CIE V (λ), CIE VM(λ), and the CIE V’ (λ) curves These curves arethe basis of the conversion from radiometric quantities to the photometric quantities used tocharacterise light

Figure 1.2 The relative luminous efficiency functions for the CIE Standard Photopic Observer,

the CIE Modified Photopic Observer, the CIE Standard Scotopic Observer, and the relativeluminous efficiency function for a 10 degree field of view in photopic conditions

Relative luminous efficiency

= Standard photopic observer

= Modified photopic observer

= Standard scotopic observer

The most fundamental measure of the electromagnetic radiation emitted by a source is its

radiant flux This is the rate of flow of energy emitted and is measured in watts The most

fundamental quantity used to measure light is luminous flux Luminous flux is radiant flux

multiplied, wavelength by wavelength, by the relative spectral sensitivity of the human visual

system, over the wavelength range 380 nm to 780 nm (Figure 1.3) This process can be

represented by the equation:

where: Φ = luminous flux (lumens)

= radiant flux in a small wavelength interval ∆λ (watts)

= the relative luminous efficiency function for the conditions

Km= constant (lumens/watt)

= wavelength interval

In System Internationale (SI) units, the radiant flux is measured in watts (W) and the luminous

flux in lumens (lm) The values of Kmare 683 lm/W for the CIE Standard and Modified

Photopic Observers and 1699 lm/W for the CIE Standard Scotopic Observer It is always

important to identify which of the CIE Standard Observers is being used in any particular

measurement or calculation The CIE recommends that whenever the Standard Scotopic

Observer is being used, the word scotopic should precede the measured quantity, i.e scotopic

luminous flux Luminous flux is used to quantify the total light output of a light source in

all directions

Figure 1.3 The process for converting from radiometric to photometric quantities The

lefthand figure shows the spectral power distribution of a light source in radiometric quantities

(watts/wavelength interval) The centre figure shows the CIE Standard Photopic Observer

Multiplying the spectral power at each wavelength by the luminous efficiency at the same

wavelength given by the CIE Standard Photopic Observer, the right hand figure is produced

The right hand figure is the spectral luminous flux distribution in photometric quantities

(lumens/wavelength interval)

1.3.2 Luminous intensity

Luminous intensity is the luminous flux emitted/unit solid angle, in a specified direction Solid

angle is given by area divided by the square of the distance and is measured in steradians An area

of 1 square metre at a distance of 1 metre from the origin subtends one steradian The unit of

measurement of luminous intensity is the candela, which is equivalent to one lumen/steradian

Luminous intensity is used to quantify the distribution of light from a luminaire

1.3.5 Reflectance

As might be expected, there is a relationship between the amount of light incident on asurface and the amount of light reflected from the same surface The simplest form of therelationship is quantified by the luminance coefficient The luminance coefficient is the ratio

of the luminance of the surface to the illuminance incident on the surface and has units ofcandela/lumen The luminance coefficient of a given surface is dependent on the nature ofthe surface and the geometry between the lighting, surface and observer

There are two other quantities commonly used to express the relationship between theluminance of a surface and the illuminance incident on it For a perfectly diffusely-reflectingsurface, the relationship is given by the equation:

(illuminance × reflectance)

π

For a diffusely-reflecting surface, reflectance is defined as the ratio of reflected luminous

flux to incident luminous flux For a non-diffusely-reflecting surface, i.e a surface with some

specularity, the same equation between luminance and illuminance applies but reflectance

is replaced with luminance factor Luminance factor is defined as the ratio of the luminance

of the surface viewed from a specific position and lit in a specified way to the luminance of a

diffusely-reflecting white surface viewed from the same direction and lit in the same way It

should be clear from this definition, that a non-diffusely-reflecting surface can have many

different values of the luminance factor Table 1.1 summarises these definitions

Table 1.1 The photometric quantities.

The luminous flux/unit area at a point on

a surfaceThe luminous flux emitted in a given direction divided by the product of the projectedarea of the source element perpendicular to thedirection and the solid angle containing thatdirection, i.e luminous intensity/unit areaThe ratio of the luminance of a surface to theilluminance incident on it

The ratio of the luminous flux reflected from asurface to the luminous flux incident on it

The ratio of the luminance of a reflecting surfaceviewed from a given direction to that of a perfectwhite uniform diffusing surface identically illuminated

luminance = (illuminance × luminance factor) / πluminance = (illuminance × reflectance ) / π

Photometry has a long history that has generated a number of different units of

measurement for illuminance and luminance Table 1.2 lists some of these obsolete units,together with the multiplying factors necessary to convert from the alternative unit to the SIunits of lumens/m2for illuminance and candela/m2for luminance

Table 1.2 Some photometric units of measurement for illuminance and luminance and the

multiplying factors necessary to change them to System Internationale (SI) units

* Luminous exitance is the product of the illuminance on the surface and the reflectance of the surface

It is only meaningful for completely diffusely reflecting surfaces Luminous exitance has the dimensions

of lumens/unit area Luminous exitance is deprecated in the SI system.

1.3.7 Typical values

Table 1.3 shows some illuminances and luminances typical of commonly occurring

situations, all measured using the CIE Standard Photopic Observer

Quantity Illuminance

Luminance

Luminous exitance*

Unit lux metre candlephotfootcandlenit stilb

Multiplying factor1.001.0010,00010.761.0010,0001,55010.760.320.323,1833.43

SituationClear sky in summer

in temperate zones

Overcast sky in summer

in temperate zones

Textile inspectionOffice workHeavy engineering

Good road lighting

Moonlight

Illuminance (lm/m2)100,000

16,000

1,500500300200.5

Typical surfaceGrass

Grass

Light grey clothWhite paperSteelConcrete road surfaceAsphalt road surface

Luminance (cd/m2)1,910

300

140120202.00.01

Table 1.3 Typical illuminance and luminance values.

1.4 The measurement of light — colourimetry

Photometry does not take into account the wavelength combination of the light Thus it is

possible for two surfaces to have the same luminance but the reflected light to be made up

of totally different combinations of wavelengths In this situation, and provided there is

enough light for colour vision to operate, the two surfaces will look different in colour The

CIE colourimetry system provides a means to quantify colour

1.4.1 The CIE chromaticity diagrams

The basis of the CIE colourimetry system is colour matching The CIE Colour Matching

Functions are the relative spectral sensitivity curves of the human observer with normal

colour vision and can be considered as another form of standard observer The CIE colour

matching functions are mathematical constructs that reflect the relative spectral sensitivities

required to ensure that all the wavelength combinations that are seen as the same colour

have the same position in the CIE colourimetry system and that all wavelength

combinations that are seen as different in colour occupy different positions Figure 1.5 shows

two sets of colour matching functions The CIE 1931 Standard Observer is used for colours

occupying visual fields up to 4° of angular subtense The CIE 1964 Standard Observer is

used for colours covering visual fields greater than 4° in angular subtense The values of

the colour matching functions at different wavelengths are known as the spectral

tristimulus values

Figure 1.5 Two sets of colour matching functions: The CIE 1931standard observer (2 degrees)

(solid line) and the CIE 1964 standard observer (10 degrees) (dashed line)

The colour of a light source can be represented mathematically by multiplying the spectral

power distribution of the light source, wavelength by wavelength, by each of the three colour

matching functions x(λ), y(λ) and z(λ), the outcome being the amounts of three imaginary

primary colours X, Y, and Z required to match the light source colour In the form of

equations, X, Y and Z are given by:

x(λ), y(λ), z(λ) = spectral tristimulus values from the appropriate

colour matching function

λ = wavelength interval (nm)

h = arbitrary constant

If only relative values of the X, Y and Z are required, an appropriate value of h is one that

makes Y = 100 If absolute values of the X, Y, and Z are required it is convenient to take

h = 683 since then the value of Y is the luminous flux in lumens.

If the colour being calculated is for light reflected from a surface or transmitted through a

material, the spectral reflectance or spectral transmittance is included as a multiplier in theabove equations For a reflecting surface, an appropriate value of h is one that makes

Y =100 for a perfect white reflecting surface because then the actual value of Y is the

percentage reflectance of the surface

Having obtained the X, Y, and Z values, the next step is to express their individual values as

proportions of their sum, i.e

x = X / (X+Y+Z) y = Y / (X+Y+Z) z = Z / (X+Y+Z) The values x, y and z are known as the CIE chromaticity coordinates As x + y + z = 1, only two

of the coordinates are required to define the chromaticity of a colour By convention, the x and y

coordinates are used Given that a colour can be represented by two coordinates, then all colourscan be represented on a two dimensional surface Figure 1.6 shows the CIE 1931 chromaticitydiagram The outer curved boundary of the CIE 1931 chromaticity diagram is called the spectrumlocus All pure colours, i.e those that consist of a single wavelength, lie on this curve The straightline joining the ends of the spectrum locus is the purple boundary and is the locus of the mostsaturated purples obtainable At the centre of the diagram is a point called the equal energy point,where a colourless surface will be located Close to the equal energy point is a curve called thePlanckian locus This curve passes through the chromaticity coordinates of objects that operate

as a black body, i.e the spectral power distribution of the light source is determined solely by its temperature

The CIE 1931 chromaticity diagram can be considered as a map of the relative location of

colours The saturation of a colour increases as the chromaticity coordinates get closer to

the spectrum locus and further from the equal energy point The hue of the colour is

determined by the direction in which the chromaticity coordinates move The CIE 1931

chromaticity diagram is useful for indicating approximately how a colour will appear, a valuerecognised by the CIE in that it specifies chromaticity coordinate limits for signal lights andsurfaces so that they will be recognised as red, green, yellow, and blue (CIE Publication

107:1994)

Figure 1.6 The CIE 1931 Chromaticity Diagram showing the spectrum locus, the Planckian

locus and the equal energy point)

The CIE 1931 chromaticity diagram is perceptually non-uniform Green colours cover a

large area while red colours are compressed in the bottom right corner This perceptual

non-uniformity makes any attempt to quantify large colour differences using the CIE 1931

chromaticity diagram futile In an attempt to improve this situation, the CIE first introduced

the CIE 1960 Uniform Chromaticity Scale (UCS) diagram and then, in 1976, recommended

the use of the CIE 1976 UCS diagram Both diagrams are simply linear transformations of the

CIE 1931 chromaticity diagram The axes for the CIE 1976 UCS diagram are

6500 4800 3500

590

600

510 620 630 640 780

2360 1900

1500

1.4.2 The CIE colour spaces

All chromaticity diagrams are of limited value for quantifying colour differences becausesuch diagrams are two-dimensional, considering only the hue and saturation of the colour

To completely describe a colour a third dimension is needed, that of brightness for a

self-luminous object and lightness for a reflecting object In 1964, the CIE introduced the U*, V*, W* colour space for use with surface colours, where

U* = 13 W* (u – un)

V* = 13 W* (v – vn)

W* = 25 Y0.33– 17 (where Y has a range from 1 to 100) W* is called a lightness index and approximates the Munsell value of a surface colour (see Section 1.4.7) The coordinates u, v, refer to the chromaticity coordinates of the surface colour in the CIE 1960 UCS diagram while the chromaticity coordinates un, vnrefer to a

spectrally neutral colour lit by the source, that is placed at the origin of the U*, V* system This U*, V*, W* system is little used now, about the only purpose for which it is routinely

used is the calculation of the CIE colour rendering indices (see Section 1.4.4) For other

purposes, the U*, V*, W* colour space has been superseded by two other colour

spaces known by the initialisms CIELUV and CIELAB

770 nm

440

400 nm 420

V'

U'

The three coordinates of the CIELUV colour space are given by the expressions:

L* = (116 (Y/Yn) 0.33 – 16) for Y/Yn>0.008856

L* = 903.29 (Y/Yn) for Y/Yn≤0.008856

u* = 13 L* (u' – u'n)

v* = 13 L* (v' – v'n)

where u' and v' are the chromaticity coordinates from the CIE 1976 UCS diagram and u'n,

v'n, Ynare values for a nominally achromatic colour, usually the surface with 100%

reflectance (Y = 100) lit by the light source.

The three coordinates of the CIELAB colour space are given by the expressions:

Again, Xn, Yn, Znare the values of the X, Y and Z for a nominally achromatic surface,

usually that of the light source with Yn= 100

Each of these colour spaces have a colour difference formula associated with them For the

CIELUV colour space, the colour difference is given by

1.4.3 Correlated colour temperature

While the CIE colourimetry system is the most exact means of quantifying colour, it is

complex Therefore, the lighting industry has used the CIE colourimetry system to derive

two single-number metrics to characterise the colour properties of light sources The metric

used to characterise the colour appearance of the light emitted by a light source is the

correlated colour temperature The basis of this measure is the fact that the spectral power

distribution of a black body is defined by Planck's Radiation Law and hence is a function of

its temperature only (see Section 3.1.1)

0.33

appearance is quantified as the correlated colour temperature, i.e the temperature of the temperature line that is closest to the actual chromaticity coordinates of the light source Thetemperatures are usually given in kelvins (K).

iso-As a rough guide, nominally-white light sources have correlated colour temperatures

ranging from 2,700 K to 7,500 K A 2,700 K light source, such as an incandescent lamp, willhave a yellowish colour appearance and be described as ‘warm’, while a 7,500 K lamp,

such as some types of fluorescent lamp, will have a bluish appearance and be described as

‘cold’ It is important to appreciate that light sources that have chromaticity coordinates thatlie beyond the range of the iso-temperature lines shown in Figure 1.8 should not be given acorrelated colour temperature The light from such light sources will appear greenish whenthe chromaticity coordinates lie above the Planckian locus or purplish if they lie below it

Figure 1.8 The Planckian locus and lines of constant correlated colour temperature plotted on

the CIE 1931 (x,y) chromaticity diagram Also shown are the chromaticity coordinates of CIE

Standard Illuminants, A, C, and D65 (from the IESNA Lighting Handbook)

1.4.4 CIE colour rendering index

The CIE colour rendering index measures how well a given light source renders a set ofstandard test colours relative to their rendering under a reference light source of the samecorrelated colour temperature as the light source of interest

colour temperature above 5000 K The actual calculation involves obtaining the positions of a

surface colour in the CIE 1964, U*, V*, W*, colour space under the reference light source and

under the light source of interest, correcting for any difference in white point under the two

light sources and expressing the difference between the two positions on a scale that gives perfectagreement between the two positions a value of 100 The CIE has fourteen standard test colours.The first eight form a set of pastel colours arranged around the hue circle Test colours nine to

fourteen represent colours of special significance, such as skin tones and vegetation The result

of the calculation for any single colour is called the CIE special colour rendering index, for

that colour The average of the special colour rendering indices for the first eight test colours

is called the CIE general colour rendering index (Ra) It is the CIE general colour rendering

index that is usually presented in light source manufacturers’ catalogues The CIE general colourrendering index varies widely across light sources (see Section 3.4.10)

1.4.5 Colour gamut

The colour gamut of a light source is obtained by calculating the position of the first eight CIE

standard test colours under the light source of interest and plotting them on the CIE 1976 UCS

diagram When the plotted positions are joined together, the colour gamut is formed The colourgamut can be reduced to a single number by calculating the gamut area Figure 1.9 shows the

colour gamuts for a number of different light sources A great deal can be learnt from the colourgamut From a consideration of its shape and the spacing between the positions of the individualtest colours, the extent to which the different parts of the hue circle can be discriminated is

apparent From its location on the CIE 1976 UCS diagram, the appearance of colours can be

appreciated to some degree By plotting different light sources on the same diagram it is easy to

make comparisons between light sources Further, by including the colour gamut of an ideal lightsource, such as daylight, it is possible to evaluate how close to the ideal light source is the light

source of interest, as far as colour rendering is concerned

+

*

+ +

+ + +

Figure 1.9 The colour gamuts for high pressure sodium, incandescent, fluorescent and metal

halide light sources, and for the CIE Standard Illuminant D65, simulating daylight, all plotted onthe CIE 1976 uniform chromaticity scale diagram The dotted curve is the Planckian locus

by the CIE Standard Scotopic and Photopic Observers and expressing the resulting

scotopic lumens and photopic lumens as a ratio The value of scotopic/photopic ratios isthat they express the relative effectiveness of different light sources in stimulating the rodand cone photoreceptors in the human visual system A light source with a higher

scotopic/photopic ratio will stimulate the rods more than a light source with a lower

scotopic/photopic ratio when both produce the same photopic luminous flux This

information is useful when considering light sources for applications where the operation ofboth rod and cone photoreceptors is likely Table 1.4 gives scotopic/photopic ratios for a

number of commonly used light sources

Table 1.4 Scotopic/photopic ratios for a number of widely used electric light sources (from

He et al., 1997)

Light sourceIncandescentFluorescentMercury vapour

Metal halideHigh pressure sodium

Low pressure sodium

Photopic efficacy (lm/W)14.784.952.3107.4126.9180.0

Scotopic efficacy (lm/W)20.3115.966.8181.780.540.8

Scotopic/photopic ratio1.381.361.281.690.630.23

1.4.7 Colour order systems

A colour ordering system is a physical, three-dimensional representation of colour space.There are several different colour ordering systems but one of the most widely used is theMunsell system Figure 1.10 shows the organisation of the Munsell system The azimuthalhue dimension consists of 100 steps arranged around a circle, with five principal hues (red,yellow, green, blue and purple) and five intermediate hues (yellow-red, green-yellow, blue-green, purple-blue and red-purple) The vertical value scale contains ten steps from black towhite The horizontal chroma scale contains up to 20 steps from gray to highly saturated Theposition of any colour in the Munsell system is identified by an alphanumeric reference made

up of three terms, hue, value and chroma, e.g a strong red is given the alphanumeric 7.5R/4/12.Achromatic surfaces, i.e colours that lie along the vertical value axis and hence have no hue orchroma, are coded as Neutral 1, Neutral 2 etc depending on their reflectance To a first

approximation, the percentage reflectance of a surface is given by the product of V and (V–1) of

the surface, where V is the Munsell value of the surface.

Figure 1.10 The organisation of the Munsell colour order system The hue letters are B =

blue, PB = purple/blue, P = purple, RP = red/purple, R = red, YR = yellow/red, Y = yellow,

GY = green/yellow, G = green, BG = blue/green

The existence of several other colour ordering systems, such as the Natural Colour System,

the DIN system and BS 5252 system, would seem to be a recipe for confusion This is

avoided by the fact that conversions are available between many of the colour ordering

systems For more detail on the Munsell system, other colour ordering systems and the

relationships between them, see the SLL Lighting Guide 11: Surface reflectance and colour.

5P 10P

5RP

10RP

10R 5R

5YR

1 2 4 6

10YR

5Y 10Y

10GY

5GY 5G

2.1 The structure of the visual system

The visual system consists of the eye and brain working together Functionally, the visualsystem is an image-processing system that extracts specific aspects of the retinal image forinterpretation by the brain

2.1.1 The visual field

Humans have two eyes, mounted frontally Figure 2.1 shows the approximate extent of thevisual field of the two eyes in humans, measured in degrees from the point of fixation Theenclosed white area can be seen with both eyes The shaded area to the left is visible to theleft eye only The shaded area to the right is visible to the right eye only

Figure 2.1 The binocular visual field expressed in degrees deviation from the point offixation The shaded areas are visible to only one eye (after Boff and Lincoln, 1988).Given this limited field of view for a fixed position, it is necessary for the two eyes to be able to move There are two ways this can be done; by moving the head and by moving theeyes in the head Humans have a limited range of head movements but a wide range of eye movements

100

80

60

20 40

Figure 2.2 The pattern of fixations made by two inspectors examining men’s briefs held on

a frame S = start of scan path, C = end of scan of front and one side, rotation of frame and

continuation of scan across back and sides, E = end of scan Inspector M examines only the

seams while Inspector D examines the fabric as well (after Megaw and Richardson, 1979)

Movement between the fixation points is made by saccades Saccades are very fast,

velocities ranging up to 1000 degrees/second depending upon the distance moved

Saccadic eye movements have a latency of about 200 ms, which limits how frequently the

line of sight can be moved to about five movements per second Visual functions are

substantially limited during saccadic movements Fixations and saccades both occur in a

single eye, but movements in the two eyes are not independent Rather, they are

coordinated so that the lines of sight of the two eyes are both pointed at the same target at

the same time Movements of the two eyes that keep the primary lines of sight converged

on a target, or which may be used to switch fixation from a target at one distance to a new

target in the same direction but at a different distance, are called vergence movements

These movements are very slow, up to 10 degrees/second, and can occur as a jump

movement or can smoothly follow a target moving in a fore-and-aft direction Both types of

movement involve a change in the angle between the two eyes

2.1.3 Optics of the eye

Figure 2.3 shows a section through the eye, the upper and lower halves being adjusted for

focus at near and far distances, respectively The eye is basically spherical with a diameter

of about 24 mm The sphere is formed from three concentric layers The outermost layer,

called the sclera, protects the contents of the eye and maintains its shape under pressure

Over most of the eye’s surface, the sclera looks white but at the front of the eye the sclera

bulges up and becomes transparent It is through this area, called the cornea, that light

enters the eye The next layer is the vascular tunic, or choroid This layer contains a dense

network of small blood vessels that provide oxygen and nutrients to the next layer, the

retina As the choroid approaches the front of the eye it separates from the sclera and forms

the ciliary body This element produces the watery fluid that lies between the cornea and the

lens, called the aqueous humor The aqueous humor provides oxygen and nutrients to the

cornea and the lens, and takes away their waste products Elsewhere in the eye this is done

by blood but on the optical pathway through the eye, a transparent medium is necessary

Figure 2.3 A section through the eye adjusted for near and distant vision

After passing through the pupil, light reaches the lens The lens is fixed in position, butvaries its focal length by changing its shape The change in shape is achieved by

contracting or relaxing the ciliary muscles For objects close to the eye, the lens is fattened.For objects far away, the lens is flattened

The space between the lens and the retina is filled with another transparent material, thejelly-like vitreous humor After passing though the vitreous humor, light reaches the retina,the location where light is absorbed and converted to electrical signals The retina is acomplex structure, as can be seen from Figure 2.4 It can be considered as having threelayers: a layer of photoreceptors, which can be divided into four types; a layer of collectorcells which provide links between multiple photoreceptors, and a layer of ganglion cells The axons of the ganglion cells form the optic nerve which produces the blind spot where itpasses through the retina out of the eye Light reaching the retina, passes through the

ganglion and collector cell layers before reaching the photoreceptors, where it is absorbed.Any light that gets through the photoreceptor layer is absorbed by the pigment epitheliummounted on Bruch’s membrane

Iris contracted

Distant vision

Pupil Cornea

Iris opened

Lens flattened

Sclera

Blind spot

Optic nerve

Cilary muscle

Near vision

Fovea Retina

Lens rounded

Figure 2.4 A section through the retina (after Sekular and Blake, 1994)

2.1.4 The structure of the retina

The retina is an extension of the brain The visual system has four photoreceptor types in

the retina, each containing a different photopigment These four types are conventionally

grouped into two classes, rods and cones All the rod photoreceptors are the same,

containing the same photopigment and hence having the same spectral sensitivity The

relative spectral sensitivity of the rod photoreceptors is shown in Figure 2.5 The other three

photoreceptor types are all cones, each with a different photopigment Figure 2.6 shows the

relative spectral sensitivity functions of the three cone photoreceptor types, called short (S),

medium (M) and long (L) wavelength cones

Figure 2.5 Log relative luminous efficiency of the rod photoreceptor

Log relative luminous efficiency

Long wavelength cones

Medium wavelength cones

Short wavelength cones

Figure 2.7 The distribution of rod and cone photoreceptors across the retina The 0 degree

indicates the position of the fovea

The three cone types are also not distributed equally across the retina The L- and M-cones

are concentrated in the fovea, their density declining gradually with increasing eccentricity

The S-cones are largely absent from the fovea; reach a maximum concentration just outside

the fovea and then decline gradually in density with increasing eccentricity

Over the whole retina there are approximately 120 million rods and 8 million cones The fact

that there are many more rod than cone photoreceptors should not be taken to indicate that

human vision is dominated by the rods It is the fovea that allows resolution of detail and other

fine discriminations and the fovea is entirely inhabited by cones There are three other

anatomical features that emphasise the importance of the fovea The first is the absence of bloodvessels The second is that the collector and ganglion layers of the retina are pulled away over thefovea The third is the fact that the outer limb of the cone photoreceptor can act as a waveguide,making cones most sensitive to light rays passing through the centre of the lens This last

characteristic, known as the Stiles-Crawford effect, compensates to some extent for the poor

quality of the eye’s optics by making the fovea less sensitive to light passing through the edge of

the lens or scattered in the optic media The fovea is populated only with cones Rod

photoreceptors, which dominate the population of the rest of the retina, do not show a

2.1.5 The functioning of the retina

The retina is where the processing of the retinal image begins Recordings of electrical

output from single ganglion cells have shown a number of important characteristics Thefirst is that the electrical discharge is a series of voltage spikes of equal amplitude

Variations in the amount of light falling on the photoreceptors supplying signals to the ganglioncell through the network of collector cells, produce changes in the frequency with which thesevoltage spikes occur but not in their amplitude The second is that there is a level of electricaldischarge present even when there is no light falling on the photoreceptors, called the

spontaneous discharge The third is that illuminating photoreceptors with a spot of light, canproduce either an increase or a decrease in the frequency of electrical discharges, relative to thelevel of frequency of discharges present when light is absent

Further studies of the pattern of electrical discharges from a single ganglion cell have revealedtwo other important aspects of the operation of the retina The first is the existence of receptivefields A receptive field is the area of the retina that determines the output from a single

ganglion cell A receptive field always represents the activity of a number of photoreceptors, andoften reflects input from different cone types as well as from rods The sizes of receptive fieldsvary systematically with retinal location Receptive fields around the fovea are very small Aseccentricity from the fovea increases, so does receptive field size

Within each receptive field there is a specific structure Receptive fields consist of a centralcircular area and a surrounding annular area These two areas have opposing effects on theganglion cell’s electrical discharge Either the central area increases and the annular surrounddecreases the rate of electrical discharge, or, in other receptive fields, the reverse occurs Thesetypes of receptive fields are known as on-centre/off-surround and off-centre/on-surround fields,respectively If either of these two types of retinal receptive fields is illuminated uniformly, thetwo types of effect on electrical discharge cancel each other, a process called lateral inhibition.However, if the illumination is not uniform across the two parts of the receptive field, a neteffect on the ganglion cell discharge is evident This pattern of response makes the retinal fieldswell suited to detect boundaries in the retinal image

While every retinal ganglion cell has a receptive field, not every ganglion cell is the same Infact, there are two types of ganglion cell, called magnocellular (M) cells and parvocellular (P)cells There are a number of important differences between the M-cells and P-cells First, theaxons of the M-cells are thicker than the axons of the P-cells, indicating that signals are

transmitted more rapidly from the M-cells than from the P-cells Second there are many moreP-cells than M-cells and they are distributed differently across the retina The P-cells dominate

in the fovea and parafovea and the M-cells dominate in the periphery Third, for a given

eccentricity, the P-cells have smaller receptive fields than the M-cells Fourth, the M-cells andP-cells are sensitive to different aspects of the retinal image The M-cells are more sensitive torapidly varying stimuli and to small differences in illumination but are insensitive to differences

in colour The P-cells are more sensitive to small areas of light and to colour

This brief description shows that the retina extracts information on boundaries in the retinalimage and then extracts specific aspects of the stimulus within the boundaries, such as colour.These aspects are then transmitted up the optic nerve, formed from the axons of the retinalganglion cells, along different channels

Figure 2.8 A schematic diagram of the pathways from the eyes to the visual cortex

(from the IESNA Lighting Handbook)

The optic nerves leaving the two eyes are brought together at the optic chiasm where the nervesfrom each eye are split and parts from the same side of the two eyes are combined This

arrangement ensures that the signals from the same side of the two eyes are received together

on the same side of the visual cortex The pathways then proceed to the lateral geniculate

nuclei Somewhere between leaving the eyes and arriving at the lateral geniculate nuclei, some

optic nerve fibers are diverted to the superior colliculus, responsible for controlling eye

movements, and to the suprachiasmatic nucleus which is concerned with entraining circadian

rhythms After the lateral geniculate nuclei, the two optic nerves spread out to supply

information to various parts of the visual cortex, the part of the brain where vision occurs

The visual cortex is located at the back of cerebral hemispheres About 80% of the cortical

cells are devoted to the central ten degrees of the visual field, the centre of which is the fovea,

a phenomenon that again emphasises the importance of the fovea

2.1.7 Colour vision

Human colour vision is trichromatic It is based on the L, M and S cone photoreceptors Figure

2.9 shows how the outputs from the three cone photoreceptor types are believed to be arranged.The achromatic channel combines inputs from the M- and L-cones only Its output is related to

luminance The other two channels are opponent channels in that they produce a difference

signal These opponent channels are responsible for the perception of colour The red-green

opponent channel produces the difference between the output of the M-cones and the sum of

the outputs of the L- and S-cones The blue-yellow opponent channel produces the difference

between the S-cones and the sum of the M- and L-cones

2.1.6 The central visual pathways

Signals from the retina are transmitted to the visual cortex of the brain over the central

visual pathways (Figure 2.8)

Lateral geniculate nucleus

Visual cortex Superior

colliculus

Optic chiasm

Retina

Optic tract Optic

nerve

Cortical cells

The ability to discriminate the wavelength content of incident light makes a dramatic difference

to the information that can be extracted from a scene Creatures with only one type of

photopigment, i.e creatures without colour vision, can only discriminate shades of grey, fromblack to white Approximately 100 such discriminations can be made Having three types ofphotopigment increases the number of discriminations to approximately 1,000,000 Thus, colourvision is a valuable part of the visual system, and not a luxury that adds little to utility

2.2 Continuous adjustments of the visual system

2.2.1 Adaptation

To cope with the wide range of luminances to which it might be exposed, from a very dark night(10–6cd/m2) to a sunlit beach (106cd/m2), the visual system changes its sensitivity through aprocess called adaptation Adaptation is a continuous process involving three distinct changes

Change in pupil size: the iris constricts and dilates in response to increased and decreased levels of

retinal illumination The maximum change in retinal illumination that can occur through pupilchanges is 16 to 1 As the visual system can operate over a range of about 1,000,000,000,000 to 1,this indicates that the pupil plays only a minor role in the adaptation of the visual system

Neural adaptation: this is a fast (less than 200 ms) change in sensitivity produced in the retina.

Neural processes account for virtually all the transitory changes in sensitivity of the eye atluminance values commonly encountered in electrically lighted environments, i.e belowluminances of about 600 cd/m2 The facts that neural adaptation is fast, is operative at moderatelight levels, and is effective over a luminance range with a maximum to minimum ratio of 1000:1 explain why it is possible to look around most lit interiors without being conscious ofbeing misadapted

S cones

M cones

L cones

Achromatic channel M+L

Blue/yellow channel [(M+L) vs S]

Red/green channel [(L+S) vs M]

Photochemical adaptation: the sensitivity of the eye to light is largely a function of the percentage of

unbleached pigment in each photoreceptor Under conditions of steady retinal illumination, the

concentration of photopigment produced by the competing processes of bleaching and

regeneration is in equilibrium When the retinal irradiance is changed, pigment is bleached and

regenerated so as to re-establish equilibrium Because the time required to accomplish the

photochemical reactions is of the order of minutes, changes in the sensitivity can lag behind the

irradiance changes The cone photoreceptors adapt much more rapidly than do the rod

photoreceptors Exactly how long it takes to adapt to a change in retinal illumination depends onthe magnitude of the change, the extent to which it involves different photoreceptors and the

direction of the change For changes in retinal illumination of about 2–3 log units, neural

adaptation is sufficient so adaptation should be complete in less than a second For larger changesphotochemical adaptation is necessary If the change in retinal illumination lies completely

within the range of operation of the cone photoreceptors, a few minutes will be sufficient for

adaptation to occur If the change in retinal illumination covers from cone photoreceptor

operation to rod photoreceptor operation, tens of minutes may be necessary for adaptation to be

completed As for the direction of change, once the photochemical processes are involved,

changes to a higher retinal illuminance can be achieved much more rapidly than changes to a

lower retinal illuminance

When the visual system is not completely adapted to the prevailing retinal illumination, its

capabilities are limited This state of changing adaptation is called transient adaptation Transient

adaptation is unlikely to be noticeable in interiors in normal conditions but can be significant

where sudden changes from high to low retinal illumination occur, such as on entering a long

road tunnel on a sunny day or in the event of a power failure in a windowless building

2.2.2 Photopic, scotopic and mesopic vision

This process of adaptation can change the spectral sensitivity of the visual system because at

different retinal illuminances, different combinations of retinal photoreceptors are operating

The three states of sensitivity are conventionally identified as follows

Photopic vision: this occurs at luminances higher than approximately 3 cd/m2 For these

luminances, the retinal response is dominated by the cone photoreceptors so both colour vision

and fine resolution of detail are available

Scotopic vision: this occurs at luminances less than approximately 0.001 cd/m2 For these

luminances only the rod photoreceptors respond to stimulation so colour is not perceived and

the fovea of the retina is blind

Mesopic vision: this is intermediate between the photopic and scotopic states, i.e between about

0.001 cd/m2and 3 cd/m2 In the mesopic state both cones and rod photoreceptors are active As

luminance declines through the mesopic region, the fovea, which contains only cone

photoreceptors, slowly declines in absolute sensitivity without significant change in spectral

sensitivity, until vision fails altogether as the scotopic state is reached In the periphery, the rod

photoreceptors gradually come to dominate the cone photoreceptors, resulting in gradual

deterioration in colour vision and resolution and a shift in spectral sensitivity to shorter

wavelengths The relevance of the different types of vision for lighting practice varies Scotopic

vision is largely irrelevant Any lighting installation worthy of the name provides enough light to

at least move the visual system into the mesopic state Most interior lighting ensures the visual

system is operating in the photopic state Current practice in exterior lighting ensures the visual

system is often operating in the mesopic state