AC 20-74 29 July 1971 1931 C.I.E. STANDARD OBSERVER
WAVELENGIH SPECTRAL LUMINOUS WAVELENGIE SPECTRAL LUMINOUS NANOMETERS EFFICIENCY, V (N NANOLETERS EFFICIENCY, V.(2) 380 0.0000 580 0.8700 390 0.0001 590 0 - 370 400 0.0004 600 0.6310 410 0.0012 610 0.5030 420 0.0040 620 0.3810 430 0.1116 630 0.2650 440 0.0230 640 0.1750 450 0.0380 650 0.1070 460 0.0600 660 0.0610 470 0.0910 670 0.0320 480 0.1390 680 0.0170 490 0.2080 690 0.0082 500 0.3230 700 0.0041 510 0.5030 710 0.0021 520 0.7100 720 0.0010 530 0.8620 730 0.0005 540 0.9540 740 0.0003 550 0.9950 750 0.0001 560 0.9950 760 0.0001 570 0.9520 770 0.0000
SPECTRAL LUMINOUS EFFICIENCY, VA Table 1.1 Chap 1 Page 4 Par 2
29 July 1971 AC 20- 74
- INTENSITY = I CANDELA, = 1 LUMEN/STERADIAN
• 1 METER 1 CM- W = 1 STERADIAN FLUX= ILL AREA: | CM?
ILLUMINANCE = | LUMEN/CM?
AREA = 1FT 3 ILLUMINANCE = I LUMEN/ FT?
AREA = I METER?
ILLUMINANCE = 1 LUMEN/ M?
INTENSITY AND ILLUMINATION Figure 1.3 Flashing lights are used oxtensively as signals and warnings because of their superiority in attracting attention. Because of this characteristic, flashing beacons were established as the required lighting for anticollision lights. These beacons are of several types: rotating, flashing incandescent, oscillating and gas discharge (strobe).
when a light signal consists of separate flashes, the maximum intensicy during the flash must be greater than the intensity of a steady light to have the same apparent intensity. It 18 convenient to evaluate flashing lights in terms of their EFFECTIVE INTENSITY, le or ETI, the intensity of a steady light which will appear equally bright when viewed at threshold, and is expressed in candelas.
Chap 1 Par 2 Page 5
AC 20- 74 29 July 1971 Current airworthiness requirements for aircraft anticollision lights specify the following equation (known as the Blondel Rey equation) for the computation of effective intensity:
where: le • effective intensity I(t) = instantaneous intensity as a function of time t2 - t1 - flash time interval (seconds) The maximum computed value of effective intensity is obtained when t2 and t1 are chosen so that the effective intensity is equal to the instantaneous intensity at t2 and t1. For short time flashes, t2 - t1 becomes insignificant compared to 02 seconds, and the total flash is integrated.
Short-duration flashtube lights have been used primarily as supplementary lights. Since flashtubes in general produce relatively small proportions of red light, about 90 percent of the light is lost in passing it through a red filter. Therefore, in order to use this source for an anti-collision light, it is necessary to operate them at higher energy levels than has been common in the small supplementary 11ghts.
3 . COLOR. The sensation of color is closely related to the wavelength of light and varies with the individual and the conditions of viewing.
Usually a color 1s said to have three psychological components: hue (red, blue, orange, etc.), brightness, and saturation (the amount a color differs from a grey of the same brightness). A measure of hue, sufficiently reliable for signal-color specification, is the wavelength of the part of the spectrun required to be mixed with the equal-energy (white) source to produce the color; this wavelength is called the dominant wavelength of the color.
ratio of the distance on the CIE chromaticity diagram (See Figure 1.5) from the equal-energy point to the color point, to the distance fron equal-energy point, in the same direction, to the spectrum locus. This ratio is called purity.
Chap 1 Page 6 Par 3
29 July 1971 AC 20-74 It can be shown that any color can be matched by combining three properly chosen colors called primary colors. It a colored sample is placed in one half of a photometric field, a mixture of three primary colors such as red, green, and blue, in the other half of the field can be made to match the colored sample to the satisfaction of the eye. This is done by varying the relative brightness of the three primary colors. A problem with this method is that the matching judguent of one observer cannot be taken as representative of the average person. Only when a large number of observers are used in each experiment can consistent values be obtained for the relative brightness of primary colors required to match any given color. In order to avoid this difficulty, the tristimulus method was revised and standardized. Three primary colors were agreed upon; then, by.
experiments with a number of normal observers, standard values for the relative amounts of each primary color were established to match each wavelength in the visible spectrum. These numbers were associated with a "standard observer". With these values made available in 1931 by the Commission Internationale de L'Eclairage (CIE), a more objective and economical technique 18 available to specify color. Figure 1.4 is a graphical presentation of the relative amounts of each primary color required to match any wavelength. Standard notation for these functions 18: & for the red primary, y for the green primary, and 2 for the blue primary. It should be noted that the f function has been adjusted to correspond to the luminous-efficiency function (Figure 1.2). Table 1.2 shows the values of these functions in tabular form.
the of we time natured toù 1O a
SPECTRAL TRISTIMULUS VALUES ACCORDING TO THE 1931 CIE STANDARD OBSERVER Chap 1 Par 3 Page 7
The 1991 CIE : tandard observer Page 8 Spectral tristimului spectrum alues of equal-eDeres spectral tristimulus values of equal energ! spectzum: AC 20-74 leD/ Dm -
0.8700 .8103 .. . . ... ... ... ... ...
.3570
49.
300 $10 120 460 Par 3 Chap 1 1000 29 July 1971 0000 0000 0000 :0000 Totale ____ 21.9714 21.8715
SPECTRAL TRISTIMULUS VALUES Table 1.2
29 July 1971 AC 20-74 Capital letters X, Y, and 2 have been assigned as notation for the amounts of the three primaries required to visually match a color containing multiple wavelengths. These values can be plotted graphically if the following transformations are made:
× - (Z) y • (3) z • (4) X+Y+Z X+Y+Z X+Y+Z where X, I, and 2 are the amounts of the three primaries and x, y, and z are the 'chromaticity coordinates" in the CIE system. The horizontal of the graph becomes x and the vertical becomes y; ¿ may be obtained from the relationship x + y + z - 1. Such a graph, shown in Figure 1.5, 18 called a chromaticity diagran. Any color may be located on the diagram by specifying its chromaticity coordinates. The colors used for lighting of aircraft are shown with the limits as established by the airworthiness regulations.
On this diagram, there 1s a central white point marked I, where X = Y - Z corresponding to the color of a source having an equalenergy spectrum. As x + y + 2 = 1, this white point appears graphically at x • .333; y = .333 and represents a position of zero purity. The outer periphery of this diagram is a locus of points of 100% purity for visible wavelengths. A line drawn fron any point on this periphery to the point of equal energy (⅖) represents all purities between 100% and O for that dominant wavelength. For example, a line is shown for a light of dominant wavelength 0.9 micrometers with the 0%, 50%, and 100% purity points indicated. An aviation green light of this dominant wavelength would have to have a purity of only 32% to meet the regulatory requirements. Aviation red, however, would require a purity of nearly 100%.
The temperature scale along the bottom of the diagram, showing a range from 1000 k to infinity, contains calibrations for source color temperatures. Note that each temperature has a calibration mark on the curved line above. The curved line is called the "planckian locus," and represents the chromaticities of blackbodies at different temperatures.
In practical applications, it 1s accepted as representing the chromaticities of incandescent bodies much as lamp filaments. The numbers along this curve on the CIE diagram indicate the color temperature in Kelvins. CIE 1lluminant "A" can be represented on this curve at the 2854 K point, 11luminant "g" at about 5000 K and "C" at about 6800 k. Illuminant A represents the spectral distribution of typical tungsten-filament incandescent lamps, 1Iluminant B represents the spectral distribution of average noon sunlight, and illuminant C represents the spectral distribution of average daylight.
Chap 1 Par 3 445-499 0 - 71-3 Page 9
AC 20-74 29 July 1971 C.LE. CHROMATICITY DIAGRAM U.S. AVIATION COLOR LIMITS 55 .6 .57 'ELLOW .60 = 0.370 : 0.540 .61y=0.335 RED .3 raum
LIGHTING RED I URPLE
VIATION BLUE .8 AVIA,WHITE: C.I.E. CAROMATICITY DIAGRAM *EFFECTIVE AUGUST 197I
•A PRIOR TO HAUGUST IOTI Figure -.S chap i Page 10 Par !
29 July 1971 AC 20-74 The color of an object depends on the spectral characteristics of the 1lluminant as well as the nature of the object. For example, fabrics change color when moved from the light of ordinary tungstenfilament bulbs to fluorescent light. Any accurate system of color specification must account for this fact by relating the color to a specified light source. In specifying a color that is going to be used under known conditions of illumination, the tristimulus values for the standard illuminant conforming most closely to those conditions should be used. For example, in coloring an aviation red light cover, the color temperature of the light-source must be considered.
Some red glass filters change color as their temperatures change.
As the temperature increases, the light passing through some red filters becomes more red (longer wavelength) and less intense. For this reason, red filters are generally designed to be near the yellow limit (shortest wavelength) at room temperature. Then, if the operating glass temperature is consistently higher than room temperature, aviation red light will be produced under all operating conditions. By designing o this yellow limit, the maximum intensity will also be obtained. Glass of other colors changes slightly with temperature, but not enough to be a problem.
Page 11(and 12)
29 July 1971 A2 20-74.
• CHAPTER 2
MEASUREMENT CONSIDERATIONS GENERAL. The measurement of visible radiated energy is called photometry. Instruments for measurement of the amount of a light, regardless of their calibration, measure only flux, and care must be taken when converting to various photometric units. An additional precaution is necessary because photosensitive devices, unless properly filtered, do not have the same characteristics as the human eye.
There are four general types of light measurements:
Luminous Intensity - The luminous flux emitted per unit solid angle in a given direction (candelas).
b. Illuminance - The luminous flux incident per unit area on a surface at some distance from the source (footcandles).
Luminance - Luminous intensity per nit projected area of surface (candelas per square foot).
d. Chromaticity - The color quality of light determined by its chromaticity coordinates.
In addition to the above light measurements, the efficiency of light covers 1s sometimes measured. There are two terms associated with such efficiency measurements; spectral transmittance (T) which refers to the ratio of transmitted to incident power at one wavelength or a very narrow band of wavelengths, and luminous transmittance (I) which is the ratio of transmitted to incident total light power. The transmittance of a light cover indicates the decrease in light due to material and color, and can be used to adjust luminous intenaity values measured without the cover. The filtered light may be measured directly provided the photometer is equipped with filters which accurately match the luminous-efficiency function.
INTENSITY. The instrument used to measure the luminous intensity of lights is called a photometer. Photometers can be placed in two general classes; "visual" and "direct-reading photoelectric".
Visual Photometers. Before the invention or photoelectric cells, most instruments for measuring intensity employed the principle of balancing two adjacent fields visually. With these photometers, a test light and a standard light of know intensity were viewed Chap 2 Page 13 Par 4
AC 20-74 29 July 1971 simultaneously. By adjusting the relative distances from the viewing point, a balance was obtained, and by applying the inverse square law, the intensity of the test light could be determined.
Although visual photometers have generally been superseded by direct-reading photoelectric type instruments, many are still being Direct-Reading Photoelectric Photometers. Photoelectric devices are now being used to detect and convert light energy to electrical energy for measurement. A photometric system for precisely measuring the magnitude and coverage of aircraft lights usually consists of the following:
(1) Goniometer. This device includes an attachment position for and or one post on m the light unit which can be moved around two axes of freedon.
The vertical and horizontal positions are calibrated, and in some cases the information is remoted to recorders so that a plot of intensity vs. direction can be made directly. A goniometer of this type is installed in the photometric laboratory of the National Bureau of Standards in Washington, Tunnel and Track. A dark tunnel, including baffles to eliminate stray light, is located in line with the goniometer.
A photodetector is placed on a carrier which can be moved on a track to vary the distance between the goniometer and the photodetector. Position information is usually remoted to a recorder.
(3) Photodetector. Photodetectors are of several types, and read-out in units such as milliwatts, microamperes, etc.
By conversion factors, footcandles (1llumínance) can be computed from the read-out. Then, by means of the distance information, a calculation of intensity can be made. For steady lights, measurements are concerned with rate of flow of light entering the eye, analogous to gallons per minute into a container. For flashing lights, however, the measurements are concerned with the quantity of light per flash entering the eye, analogous to gallons. The quantities are proportional to footcandles and footcandle-seconds respectively. Since the instantaneous illuminance varies during exposure to flashing lights, integration is necessary to determine the footcandle-seconds. As discussed in paragraph 2 of Chapter 1, the computation of effective intensity requires a candela-second measurement. Candela-seconds Chap 2 Page 14 Par 5
29 July 1971 AC 20-74 can be computed from footcandle seconds by multiplying by the distance (in feet) squared. Care must be taken not to overload a photodetector by too much 11ght as it can saturate and its response become non linear.
Integrating photometers are available which electronically perform the integration and read out the total exposure in microcouloubs.
These photometers are used only on short duration strobes where the entire flash is integrated. A microcoulomb is a microampere-second, and a calibration factor is used to convert a microcoulomb reading to footcandle-seconds. Then, multiplying by the distance squared gives candela-seconds corresponding to the numerator of the equation for effective intensity.
Strip recorders or recording oscilloscopes are generally used for longer duration flashes, such as rotating beacons, and mechanical integration 1s then perforned.
800 INTENSITY 500 400 200 I INCH 12-11 -.05 SEC.
STRIP RECORDING OF FLASH figure 2.l Figure 2.1 shows an example of a recording made by a strip recorder.
The paper is moved horizontally at a linear rate so that each diviaion represents approxdmately ,05 seconda. The recorder is calibrated so that each vertical division represents 100 candlas. After the flash 1s recorded, a mechanical integrator (planineter) 1s used to trace the recording and obtain the area, in candela-seconds, between the limits of t2 and t1. This ares divided by 0.2 + t2 - t1 gives the effective intensity. The values of t2 and t1 are determined by experinentation, and are selected to maximize the computed effective intensity. In the Chap 2 Par 2 Page 15
29 July 1971 AC 20- 74 example of figure 2.1 the flash has an area between t2 and t1 of approximately Si candela-seconds. When divided byQ3 seconds (0.2 +
0.1) the effective intensity is 170 candelas. The peak intensity of 850 candelas, therefore, has an effective intensity of only 170 candelas. A method for selecting t2 and t1 is given in Chapter 3.
when measuring colored flashing lights, the photodetector cannot always be relied upon to give accurate results, even with luminous-efficiency-correction filters. The alternative is to measure the effective intensity with a clear cover, and then with a spectrophotometer, determine the relative luminous transmittances of the two covers The effective intensity value measured through the clear cover is then reduced by this luminous transmittance factor.
Spectrophotometers are discussed in paragraph 3., Color.
3. COLOR. Color can be evaluated with a visual photometer by comparing • to le efor temperature a sample against a filter of known color. In such comparisons, the same, correct color temperature should be used for both sources. However, there has been increasing interest in the use of photoelectric instruments for the measurement, of chromaticity of colored lights.
The following is a discussion of some of these instruments:
Spectrophotometer. Spectrophotometers break up a self-contained light source into a spectrum by prisms or gratings, so that narrow bands wíthin the visible spectrum can be individually applied to a test ware (such as a light cover). Although these are not single wavelengths, they are very narrow bands approximately five nanometers wide. Determining either luminous transmittance (I), or chromaticity coordinates (x, y, and z) with a spectrophotometer involves the measurement of 40 transmittances (t), spaced 10 nanometers apart, throughout the visible spectrum. Mathematically, these transmittances are used as follows:
X - Е360 КЕТАЛ (5) Y • 1380 FETA) (6) Z = [388 ZЕТAЛ (7) The values for x, y, and z are found in Table 1.2, and represent the relative amounts of primaries required at each wavelength. The E in the equations represents the spectral distribution of the light source, and is published for certain Illuminants such as C.I.E. Source Tables are available which list the products II, yE, and zi for certain illuminants (NBS Monograph 104). These numbers are sometimes found on a computation form as shown in table 2.1. If the 11ght 18 not a standard source, a table must be developed from measurements of the energy distribution. When each of the 40 lines have measured values for t logged, the remainder of the sheet can be filled out with data computed by hand, calculating machine, or computer. For example, assume a measured value for 1 at 380 nanometers of 0.078.
Then IET would be 0.078, JET would be 0.000, and IET would be 0.468.
Chap 2 Page 16 Par 5