Understanding Illuminants

®
Author - Michael DiCosola

Understanding Illuminants

Many questions have been presented to the staff at X-Rite as to the meaning and application
of the various standard illuminants referred to in an X-Rite Spectrophotometer. This document
attempts to help our customers properly understand and apply illuminants when controlling
color.

Illuminants Defined
Relative Spectral Power Distribution
To understand an illuminant it is necessary to first understand how we describe light. Light is
described for our purposes by what is called a spectral power distribution curve. A specific
type of light will emit different amounts of energy at each part of the visual spectrum. The
graph of this power emittance over the visual spectrum is called the relative power distribution
curve for that specific illuminant or source.

Daylight
120

100

80

60

40

20

0
300 340 380 420 460 500 540 580 620 660 700 740 780 820

Above we see the relative spectral power distribution curve for natural daylight. The horizontal
axis represents the visible spectrum which is approximately between 380 and 750
nanometers1. The highest peak in the graph occurs around 460 nm which represents the blue
portion of the spectrum. This tells us that daylight, while appearing to be pure white, is largely
composed of blue light. Every type of light has a unique, relative spectral power distribution
curve that describes what it looks like, or more importantly, what other things look like when
illuminated by it.

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995
Illuminant vs. Source
Also important to understanding illuminants is the difference between an illuminant and a
source. As described by Billmeyer and Saltzman, a source is an actual physical light,
something that can actually be turned on to illuminate something. An illuminant however, is a
light that has been defined by a spectral power distribution, but may not actually exist. For
example, the light in your living room is a source. It can be turned on and off, and through
experimentation, its spectral power distribution can be determined. However, if we take a
blank spectral power distribution graph and draw a squiggly line across it, we have defined an
illuminant. This light we have defined may not exist, but we can use its curve to test
numerically how a certain color would look when viewed under it.

Color Rendering Index

A color rendering index (CRI) is a method by which sources are described. A CRI is a scale
from 1 to 100 which rates a source by comparing color appearance viewed under the source
to color appearance viewed under a standard illuminant; such as D65. A CRI value of 100
means the source is identical to the standard illuminant. Using the CRI as a guide, one can
choose a light source based on that light source's ability to accurately display color as
compared to a standard illuminant. For example, a common light source for office space may
have a CRI in the 60's, but in areas where color appearance is very critical, a light source with
a CRI in the 90's would be more appropriate.

So an illuminant is basically a definition used to determine how the appearance of a color will
be changed under a specific type of lighting. D65 illustrated above is an illuminant
representing daylight. There is no actual light source that exactly matches the theoretical
illuminant (though many sources come close). This illuminant is applied to color
measurements that attempt to represent what a color would look like when viewed under
daylight or some approximate light source.

How Illuminants Effect Color

So how does this concern someone measuring color? Many colors undergo a common color
phenomenon called metamerism when viewed under different light sources. Two colors may
appear to match under one source but appear completely different under another source.
This can be a potentially big problem if the color of your product changes when it hits the
showroom floor. Let's take an extreme example. If your product was to be used in a red light
dark room environment and the box was printed in blue, green and black, the box would
appear solid black in the dark room. The pigments of the ink used to print blue, green, and
black, all absorb the red portion of the spectrum. If your illuminant only emits light in the red
portion of the spectrum, all the light will be absorbed and the whole box will appear black.

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995
The following example combines the reflectance curves of two sample colors with the spectral
power distribution curves of a daylight and an incandescent illuminant. These two color
samples are metameric to one another.

Daylight vs Incandescent
150

100

50

0
400 440 480 520 560 600 640 680

Incandescent Daylight

Sample 1 vs. Sample 2

150

100

50

0
400 440 480 520 560 600 640 680

Sample 2 Sample 1

Samples' Reflectance under Daylight

120
100
80
60
40
20
0
400 440 480 520 560 600 640 680

S.1 reflected S.2 reflected sample 2

sample 1 daylight

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995
 Samples' Reflectance under Incandescent
120
100
80
60
40
20
0
400 440 480 520 560 600 640 680

S.1 reflected S.2 reflected Incandescent

Sample 2 Sample 1

The two colors appear identical to the observer when viewed under the incandescent
illuminant, but Sample 1 appears very blue compared to Sample 2 under the daylight
illuminant. The two colors are spectrally identical from 520 to 700 nm, so they always reflect
the same amount of light in the green and red portion of the spectrum. However, the two
samples are very different in the 400 to 520 nm or blue range of the spectrum. While Sample
1 has the ability to reflect three times the amount of blue light as Sample 2, the incandescent
light source has very little blue light available to reflect. Both samples only reflect a small
amount of blue light and the colors appear the same. However, the daylight illumination has
an abundance of light available in the blue spectral range. Under the daylight illuminant,
Sample 1 reflects much more blue light than Sample 2 and therefore appears more blue than
Sample 2. It is easy to see why choosing the proper illuminant can be critical to achieving
quality color results in your product.

Choosing an Illuminant
So how does one use this information to choose a standard illuminant when measuring color
with a spectrophotometer? The answer is actually very simple. Choose the illuminant closest
to where you make your money. In other words, choose a standard illuminant that best
represents the source that will light your product where it is sold. If your product is displayed
outdoors, choose a few of the daylight illuminants to represent sunlight at different times of the
day. If you sell your product at a department store, choose a representative fluorescent
illuminant. The best way to determine what illuminant to use is to find out exactly what source
your product will be exposed to at the point of sale. If you can find out the exact brand and
type of bulb that will be used at the point of purchase, you can then obtain the spectral power
distribution curves of that source from the bulb manufacturer. Now simply pick an illuminant
whose spectral power curve most closely resembles the spectral power curve of the source.
For example; if the lighting used in one particular store that sells my product is GE® Cool
White fluorescent, the first step would be to obtain the spectral power distribution curves from
General Electric. Next we would look at the spectral power distributions of the available
standard illuminants.

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995
 GE Actual Cool White Source* CIE F2 Standard Illuminant
250 200
200
150
150
100
100
50 50

0 0
300 400 420 440 540 560 580 680 780 300 400 420 440 540 560 580 680 780

* data was approximated (GE Lighting)

The F2 standard illuminant was based on the cool white variety of fluorescent bulbs available
on the market. Comparing the two relative spectral power curves, we see that the GE® Cool
White source is not an exact match to the F2 standard illuminant, but measuring under the F2
illuminant will give a good approximation of the results under the GE bulb. This illuminant
would be an adequate choice for making your color measurements.

This solution however, is not always applicable to all situations. Not every bulb manufacturer
makes spectral power data easily obtainable, and certain products are displayed under an
infinite amount of light sources. In this situation we must look to the product first. The problem
of metamerism is most common when a product is constructed of different materials, such as
the plastic, cloth and metal in an office chair. If we insure that the different dyes and pigments
are not metameric to each other, we have won most of the battle. Because we know the
product will appear as one single color in all lighting conditions, we can now choose any
standard to take our color measurements. As long as we use this same illuminant when
measuring the color of our product, we will ensure color consistency.

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995
Appendix
This appendix displays the relative spectral power distributions for standard illuminants A, C,
D50, D65, F2, F7, and F11. All data was normalized to a relative power of 100 at 560
nanometers. Data was obtained and cross-referenced from multiple sources specifying CIE
standard illuminant data. Some data points were approximated or interpolated.

Illuminant A - Incandescent
300

250

200

150

100

50

0
300 340 380 420 460 500 540 580 620 660 700 740 780 820

The CIE standard illuminant A was first recommended in 1931 to represent incandescent light
sources with an approximate color temperature of 2856 degrees Kelvin.

Illuminant C - Daylight
140

120

100

80

60

40

20

0
300 340 380 420 460 500 540 580 620 660 700 740

CIE standard illuminant C was recommended in 1931 to represent average daylight with a
color temperature of 6774 degrees Kelvin.

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995
 Illuminant D50 - Daylight @ 5000K
120

100

80

60

40

20

0
300 340 380 420 460 500 540 580 620 660 700 740 780

CIE standard illuminant D50 is correlated color temperature D illuminant which is calculated
from the standard illuminant D65. This standard illuminant represents Daylight at 5000
degrees Kelvin and is the ANSI standard illuminant used in the graphic arts industry.

Illuminant D65
120

100

80

60

40

20

0
300 360 420 480 540 600 660 720 780

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995
The CIE standard illuminant D65 represents daylight at approximately 6500 degrees Kelvin.
This standard and the method for calculating different correlating color temperatures was
introduced in 1964.

Illuminant F2 - Cool White

180
160
140
120
100
80
60
40
20
0
300 400 420 440 540 560 580 680 780

The CIE standard illuminant F2 represents the typical cool white fluorescent source. The
spikes which protrude above the continuous graph in the fluorescent illuminants represent the
power measurements at the principle mercury emission lines.

Illuminant F7 - Broad Band Daylight

350

300

250

200

150

100

50

0
300 400 438 520 577.8 660 760

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995
The CIE standard Illuminant F7 represents a broad band daylight fluorescent source.
Examples of this type of source are the GE® and Philips® Daylight fluorescent sources.

Illuminant F11 - Narrow Band White

1000
900
800
700
600
500
400
300
200
100
0
300 360 404.7 428 460 520 556.1 587.8 640 700 760

The CIE standard illuminant F11 represents a narrow band white fluorescent source. Note the
spikes at approximately 430 nm and 550 nm peak at approximate power values of 1200 and
2500 respectively. Sources similar to this illuminant are the Philips® F40AX41 and TL841, as
well as the GE® SPX41.

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995
References
ANSI CGATS.5-1993, Graphic Technology - Spectral Measurement and Colorimetric
Computation for Graphic Arts Images, NPES the Association for Suppliers of Printing and
Publishing Technologies, 1899 Preston White Drive, Reston, Virginia 22091-4367, 1993.

ASTM E308-90, Standard Test Method for Computing the Colors of Objects by Using the CIE
System, American Society for Testing and Materials, 1916 Race Street, Philadelphia , PA
19103, 1990.

Billmeyer, F. W. Jr., & Saltzman, M., Principles of Color Technology, 2nd Edition, John Wiley
& Sons, New York, 1981.

Color Selection Guide for Fluorescent Lamps, Philips Lighting Company, 200 Franklin Square
Drive, P.O. Box 6800, Somerset, NJ 08875-6800, 1994.

Guide to Fluorescent Lamps, Philips Lighting Company, 200 Franklin Square Drive, P.O. Box
6800, Somerset, NJ 08875-6800, 1992.

Judd, D.B., & Wyszecki, Günter, Color in Business, Science, and Industry, 3rd Edition, John
Wiley & Sons, New York, 1975.

Lighting Application Bulletin: Specifying Light and Color, GE Lighting Resource Center, 1975
Nobel Road, Cleveland, Ohio 44112, 1994.

Metamerism: The Influences of Light Sources on Color, Colorcurve Systems, Inc., 123 North
Third Street, Minneapolis, MN 55401, 1989.

Product Information: Advantage X Fluorescent Lamps, Philips Lighting Company, 200 Franklin
Square Drive, P.O. Box 6800, Somerset, NJ 08875-6800, 1991.

Product Information: TL80 Series 2'-5' Fluorescent Lamps, Philips Lighting Company, 200
Franklin Square Drive, P.O. Box 6800, Somerset, NJ 08875-6800, 1993.

Stiles, W.S., & Wyszecki, Günter, Color Science: Concepts and Methods, Quantitative Data
and Formulae, 2nd Edition., John Wiley & Sons, New York, 1982.

X-Rite World Headquarters © 1995 X-Rite, Incorporated Doc# CA00002a.doc

(616) 534-7663 Fax (616) 534-8960 August 25, 1995