Click for next page ( 34


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 33
SESSION 11: METHODS FOR CONTROLLING AMBIENT ILLUMINATION CHAIRMAN Dr. Henry A. Imus THE RATIONALE FOR USE OF CONTROLLED WHITE LIGHT IN RADAR AND SONAR SPACES Carroll T. White, U. S. Navy Electronics Laboratory SOME APPLICATIONS OF CONTROLLED WHITE LIGHTING Fred G. Henry, U. S. Navy Electronics Laboratory A BROAD-BAND-BLUE LIGHTING SYSTEM FOR RADAR APPROACH CONTROL CENTERS: EVALUATIONS AND REFINEMENTS BASED ON THREE YEARS OF OPERATIONAL USE Conrad L. Kraft, Laboratory of Aviation Psychology, The Ohio State University RELATIONS AMONG DARK ADAPTATION, THE SPECTRAL CHARACTER OF ILLUMINATION, AND THE VISUAL TASK John L. Brown, Acceleration Laboratory, U. S. Naval Air Development Center EFFECTS OF CERTAIN PRE-EXPOSURE VARIABLES ON DARK ADAPTATION John A. Hanson, Institute for Applied Experimental Psychology, Tufts University MAKING RADAR INDICATORS USEFUL IN HIGH AMBIENT ILLUMINATION J. R. Roeder, Air Arm Division, Westinghouse Electric Corporation 33

OCR for page 33
The Rationale for Use of Controfiec' White Light in Raciar and Sonar Spaces CARROLL T. WHITE, U.S. Navy Electronics Laboratory Summary Special lighting systems may be categorized in terms of general area i/Zumir~ation and localized illumination. Among the advantages of locally controlled lighting are its flexibility in complex systems and its elimination of reflections from displays. Principles for location of lamps and selection of type of illumina- tion are reviewed. Emphasis is placed on the use of white light. Polarized white light is specially advantageous for the illumina- tion of cathode-ray tubes. Work on specialized lighting systems at the Navy Electronics Laboratory has been directed at the general problem of critical lighting needs and not at any particular type of installation. This may help to explain some of the differ- ences between our concepts of light control and those which have evolved else- where. The first work along this line involved the selective spectrum lighting sys- tems, with which most of you are familiar. These systems represented an attempt to utilize the unique characteristics of certain illuminants and filters in order to achieve increased illumination with little interference with the displays. Exper- ience with these systems soon revealed distinct limitations and restricted appli- cability, so we began looking for other techniques that would better fit the needs of the many types of installations in which radar or sonar displays are involved. Out of this continuing study, our present concept of lighting has evolved; it is basically the idea of localized light control. This paper will attempt to describe the merits of such an approach. Regardless of the specific details, all the special lighting systems fall into one of two categories: general area illumination or localized illumination. One goal of the general area illumination approach is to achieve as nearly as possible the ideal interior lighting - fairly even over the entire area, with no high con- trasts between different sections. The localized lighting approach is based on an entirely different premise i.e., that each critical position in an area should be supplied with the proper quantity and quality of light for the job to be per- formed at that position. The overall illumination is considered to be of secondary importance. This does not mean that lighting is provided only at the scope con- soles. Properly controlled luminaires are installed for desk surfaces, equipment racks, and panels, and for safe passage throughout the area. One of the most important features of a locally controlled lighting system is its flexibility. It can be applied to every kind of installation from a single-scope sonar room to a complex installation where a number of different types of jobs are being done. In a complex installation, it is possible to supply each man with widely different amounts of light, according to his needs, without interfering - 34

OCR for page 33
with others who are working nearby. No single level of illumination can be chosen in the proper range for all the activities which go on in a complete situation. Locally controlled lights must be introduced into spaces for those tasks demanding much reading, writing, or repair of equipment. What has been said concerning the flexibility of a locally controlled lighting system has implied the use of various levels of white light, but it should be clear that any type of illuminant could be used at any particular station' or no illum- inant at all if it is found that it is operationally better to have a particular display in relative darkness. By means of a locally controlled lighting system, it is possi- ble to meet the special requirements of any display without penalizing those who must work in the vicinity of the display. Whenever light is introduced into an area, we face the problem of first- surface reflections from the glass or plastic masks covering the display scopes. If care is not taken, these reflections degrade the display beneath them. In a properly designed, localized lighting system, specular reflections are avoided by the careful placement and louvering of the luminaires. Reflections sometimes present a real problem for a system already installed, but can be handled without difficulty if the lighting is considered during the design of the system. The problem is different for a general area lighting system. If the illumina- tion level of an entire area is raised to an appreciable amount, specular reflections cannot be avoided. Major glare sources can be eliminated by the careful place- ment and louvering of the overhead luminaires, but reflections of objects (in- cluding the operator) will still be present to some degree. Since most plans for ares lighting systems have called for limited spectrum illumination it has been suggested that the operators wear special goggles which do not Pass wavelengths within the Illuminating spectrum, thereby e~m~nat~ng all first-surface reflec- tions. This would certainly do the job, but it is probably not very practicable. First-surface reflections can also degrade other visual displays, such as status boards and summary plots. This effect and the scope reflection problem make it quite evident that we cannot achieve successful illumination by arbitraril placing overhead luminaires throughout an area. The proper placement and control of light sources is a problem to be faced in both the general area lighting and the locally controlled lighting systems. Once it has been decided that a locally controlled lighting system should be developed, the next questions are: where should the lights be placed, and what type of lights should they be? Whenever a light source is introduced at a given work position, it must (a) supply an adequate level and distribution of light, (b) not "spill over" into adjacent work spaces unnecessarily, and (c) not be a source of direct or indirect glare either for the worker in that position or for other workers in the area. To meet these requirements, care must be taken in the choice of a luminaire, the placement of the luminaire, and in the design of the supplementary louver. 35-

OCR for page 33
For work positions where no CRT displays are involved, there is no real problem. Since a great deal of reading, writing, troubleshooting' etc., must be done at these positions white light is the logical choice. The variety of color coding encountered in printed material and electronic circuitry is the major reason for this decision. The level of illumination is determined by the particular tasks to be performed. It is usually desirable to provide supplementary illumination for those por- tions of the area not adequately lighted by spill-over from the work positions. Additional illumination should include a series of lights placed near the floor to define and illuminate safe pathways through the area. In the floor design itself, pathways can be indicated by light tiles set against a dark background. A well- designed system should also include lights installed specifically for the conven- ience of maintenance personnel. The majority of these can be located within the consoles and racks themselves. They can easily be arranged so as to be of full benefit to the technician and only a minor distraction for the operating per- sonnel. The problem of providing illumination for those positions at which CRT displays are located is most critical, but it must not be divorced from the light- ing needs of the system as a whole. There are a number of ways in which we may introduce light around a scope without having light reach the display itself. Limited spectrum illumination of the console, with a filter over the scope which rejects that illumination, is the best known approach. The use of a polarized light source with a second polarizer over the scope is also a possibility. Another method is the use of focused white light, arranged so that panels and writing surfaces are illuminated and relatively little light reaches the scope itself. Still another approach, which has been suggested, is the use of a special wire mesh as In ~ ~'rr. ~ ~ 1 1 1 a covering for the scope. vvlth ordinary overneao 1uminaires, the mesh prevents light from reaching the display plane. And, if all else fails, there is always the hood. Each of the above techniques serves the purpose of keeping extraneous light from reaching the display screen, but it would be futile to try to apply any one technique in every situation. There are good arguments pro and con even for the scope hood. Admittedly, it is too restrictive for most purposes, but there are undoubtedly situations where display characteristics might demand extreme measures of this kind. It is a basic tenet of the locally controlled lighting concept that any of the special light control techniques may be used whenever necessary. The light control principle utilizing polarized white light) is probably the most widely applicable of all. Since the polarizing filter is not color selective, it may be used with any phosphor or mixed-phosphor display. At the same time, the operator can be supplied with white light so that he may be more effective in those phases of his work other than scope-watching itself. For a particular phosphors a colored filter may be available which will transmit a higher per- centa~e Of the light. emitted hv that r~hosohor than would. be transmitted by a - -rid Or- - 0 o~ ~^ ~ 0~ ~ ~ ~--~~ r--~r~ ~~ ~White, C. T. Polarized-light illumination of radar and sonar spaces and comparison with limited spectrum methods. U. S. Navy Electronics Laboratory, 1956, Report 669. 36

OCR for page 33
polarizer. The crucial question, however, is not how much light is or is not trans- mitted, but rather whether or not this has any practical effect on the performance of the operator. In this regard, we must realize that there are factors other than the filter transmittance which enter the picture; among these are, for instance, the dynamic characteristics of the display itself and the overall duties of the man who must work at that display position. The first of these would determine the latitude the operator has in increasing the light output of the display' and the second would indicate what compromises might be justified from the point of view of the entire system. In conclusion' I would like to summarize our point of view in regard to the problem of the illumination of critical areas: ( 1 ~ This problem is of much greater importance than is usually thought, and it should be an integral part of systems planning and design. (2) The wide variety of tasks to be performed in such an area leads us to favor a system which supplies each working position with the quantity and quality of light most appropriate to it rather than a system which represents a compromise with the needs of every position. ~ 3 ~ Whenever possible, white light should be used. Full spectrum illumination best meets the needs for most work positions, including those with CRT displays. There may be situations where the display characteristics may require some other type of light control. Even in these situations the total job of the man at that position must be consid- ered before a decision is made7 and any decision made for one position should not dictate what is done at any other position. (4) Finally, we do not believe that there is any one solution to this problem. Any attempt arbitrarily to specify what is to be done could be a real disservice. What is needed is for us to learn more about the characteristics of the various means of light control, and for us to deter- mine the situations in which each might be most suitable. In this way, we will be able to supply the system designer with those techniques he needs in order to make appropriate illumination an integral part of that system. --37

OCR for page 33
Some Applications of Controlled White Lighting FRED G. HENRY, U.S. Navy Electronics Laboratory Summary- This paper describes the use of controlled white lighting in two specialized installations where CRT displays are employed. The advantages of using controlled white light for in- installatzons of this type are: ( 1 ~ the reduction of distracting first- surface reflections from CRT displays; (2) good color discrimina- tion for various displays; (3) controlled variation of brightness values in the visual field, depending on the operators' require- ments; (4) adequate localized lighting which does not degrade scope displays, through the use of linear and circular polarizers; (5) nominal installation costs; and (6) flexibility of the con- trolled white lighting. The technique lends itself both to designing systems prior to construction and to altering operating areas which require lighting improvement. This report describes the application of controlled white light in two spec- ialized installations. The first is a Navy radar air traffic control center in which the lighting has been an important part of system planning. The second is a ground control approach (GCA) mobile radar van in which an experimental lighting scheme was implemented to improve the operator's visual environment. The objective for each installation was to design a lighting system which was operationally efficient, flexible, and economical to apply. Under a Bureau of Ships problem assignment, the Navy Electronics Lab- oratory (NEL) has provided technical assistance during design and implemen- tation of the NAS Miramar Radar Air Traffic Control Center (RATCC). Tech- nical assistance by NEL on this problem has, in the main, been confined to de- signing a lighting system. Figure 1 shows the general configuration of the RATCC, the major equip- ments involved, and the room dimensions. The system includes sixteen PPI displays of which thirteen are radar indicators, two are NTG projection displays, and one is an Iatron (not shown). Twenty air-controllers comprise the normal daytime watch. Among the general requirements of the lighting system for RATCC, the two rather specific lighting objectives were that the system be designed so that each member of the RATCC team has brightness relationships in his visual field which allow optimal, or near optimal, performance over prolonged watch periods; and that the system design be adaptable to other RATCCs with minimum difficulty and expense. On the basis of extensive experimentation with various lighting techniques, the laboratory concluded that both of these objectives could be met by using controlled white lighting In essence. this technique draws upon two basic prin- ciples of light control. the first principle is that light can be effectively controlled through the use of linear and circular polarizers. Radar consoles, illuminated by _ _ ~--~ 38

OCR for page 33
a polarized source, have a second circular Polaroid filter placed over the CRT display. The second filter is rotated by 90 with respect to the source polarization. Almost all incident light on the radar scope is thus cancelled. ^^'^ ~/~ ~ Fig. 1. General configuration of NAS Miramar RATCC, including equipments and room dimensions {ft). Dotted outline represents opaque wall map. Since only polarized light is cancelled by a correctly oriented Polaroid filter over the radar display, it is necessary to minimize the presence of stray, uncon- trolled light. This is accomplished through application of the second principle: light can be effectively controlled through manipulating the absorption and re- flection characteristics of surfaces in the visual field. This lighting principle is applied throughout RATCC but is especially effective at all operating stations, including radar indicators. The second principle is the basis of localized illumina- tion wherein the light flux is restricted to a well-defined region through the aid of black lattice louvers and spectral characteristics of surfaces in the visual field. Careful application of these two principles has met the lighting requirements of controllers at fixed positions and of supervisors who must move freely in RATCC. To aid in implementing these principles at the major operating positions, two mockup studies were undertaken. One was concerned with the application of localized light at the flight progress consoles; the others with the application of controlled white light at the radar consoles. From these studies it was possible to specify optimal lighting components including a fluorescent lamp, a special lattice louvered device, and a mounting assembly which positions the luminaire for optimal results. When properly grouped, these components apply to both flight progress consoles and radar consoles. Flight progress consoles Three requirements influenced the lighting installation at the flight progress console: ( 1 ) sufficient lighting for all aspects of the operator's seeing task, (2) - 39

OCR for page 33
localizing illumination to prevent spill-over on the adjacent radar operators, and (3) minimizing glare from the flight-strip display area so that adjacent radar operators are not affected by peripheral "hot-spots." In satisfying the first requirement, it was necessary to employ a luminaire which would provide good lighting for several specific seeing tasks. These tasks include the viewing of a fli~ht-strip display area, multichannel communication box, desk space, and clock. In order to obtain an optimal light source for tne operator's work space, several luminaire configurations were tested, both custom- built and commercially available. A desirable feature of the luminaire we decided to use is its adaptability to various types of lattice louvers. This was considered important in satisfying the second requirement of directional light control. Light control at the flight progress console is essential in view of its close proximity to adjacent radar indicators. Louvered assemblies were easily designed to fit the luminaires. Lattice louvers have been effective in minimizing stray light which might otherwise adversely affect scope visibility. Two longitudinal plates of the louvered assembly provide a sharp light cut-off in front of the operator, while the close spacing of the trans- verse plates provide an even sharper cut-off point at the console sides. Thus, 6 ft-c of illumination was measured on the extreme edge of the flight progress board in contrast with less than 0.10 ft-c of light spill-over at the adjacent radar . . posltlon. To satisfy the third requirement, action was taken to minimize glare from the flight progress display. Single flight strips currently used by the CAA are printed on light green or buff lusterless paper. The light reflectance from these strips ranges up to 63 per cent. The controller, who sits directly in front of this display, is in a comfortable visual environment since the greatest portion of his visual field is uniformly illuminated. On the other hand, the radar operator im- mediately adjacent to this position is faced with a somewhat different visual environment. The region to which he attends may absorb as much as 80 per cent of the incident light and therefore appear much darker than the flight progress strips for the same amount of incident light. Relatively bright surfaces in the radar operator's peripheral visual field are distracting and result in general visual discomfort. In an effort to minimize the peripheral glare from the flight progress board, the laboratory has experimented with black flight progress strips. A white blueprint pencil makes clearly legible mares on the black paper, which has a light reflectance value of 5.8 per cent. By using the black strips, a 91 per cent reduction in peripheral glare is achieved. Radar consoles There were two main considerations in optimizing lighting at the radar indicators. The first was the efficient application of Polaroid filters over both the illuminating source and the radar displays. The louvered assembly employed in the fluorescent luminaire was designed to accommodate a strip of linear Polaroid. The polarizer is seated on the top of the louvered assembly. A 1/16-in. _40

OCR for page 33
lip around the edge of the louver prevents the polarizer from shifting and fixes its plane of polarization with respect to the luminaire. Since the luminaire is supported by a rigid mounting arm, accidental bumping of the fixture will not change the axis of polarization. Light intensity at each operating position is also fixed only an on-off switch is available to control the light source. Efficient application of circular Polaroid filters was similar for each of thirteen radar displays. This was accomplished, for example at the medium- range indicators, by removing the scope bezel and placing the filter below the implosion shield. The Polaroid filter was then rotated so that its axis of polariza- tion was crossed to that of the light source. After reassembly, the polarizer was locked in place and its relatively soft surface was protected from scratching by the Lucite implosion shield. Since these two surfaces were not sealed, there was a slight tendency for dust to collect between the Polaroid filter and the bottom side of the implosion shield. The best solution to this problem is found in com- mercially available Polaroid glass laminates. These laminates are a suitable sub- stitute for the conventional transparent plastic implosion shields and are con- siderably more scratch resistant. The second consideration was optimal placement of the luminaire over the radar indicator. Figure 2 shows diagrammatically the approximate light pattern produced by the luminaire in relation to the operator. It should be noticed that the luminaire is mounted so that lamp reflections are not visible to the operator, either seated or standing. Furthermore, the lattice louver protects the operator from an otherwise annoying glare source when he looks up to throw a channel switch. Similarly, the louver eliminates stray light on the operator and distract- ing first-surface reflections of face and clothes. Environ merit Other factors affecting the general visual environment of the RATCC in- clude the finishing of floor, ceiling, and walls, and the lighting of large vertical displays. The floor covering for most of the RATCC room is a dark asphalt tile, which was installed before serious consideration was given to optimal interior finishing. Ideally, the entire floor could have been much lighter. As a com- promise solution, aisles were designated and the dark tile in these sections of the floor was replaced with off-white asphalt tile. Lighting of the aisles is provided by several low-output directional lamps that illuminate the walk space in front of the major operating positions. Low-intensity, localized floor lighting can be sufficiently well controlled so that it presents no problems to the radar operator. At the same time, floor lighting increases the overall brightness of the room, thus providing a more pleasing environment. The ceiling and the walls are covered with Celotex tile in a matte finish which reduces the effects of stray light. The ceiling, which tends to be visible as a first-surface reflection in the CRT displays, was painted black; any stray light striking this black matte surface is almost completely absorbed. The walls were painted with a nonreflecting forest green, since they are less likely to reflect light on the CRTs to the extent of affecting scope visibility. -41

OCR for page 33
1 I I 1 / l \\ \\ D \? ~ B \ ~ \ _ ~ ' ( \ fig. 2. Elements of light control at radar console: {A) Light ray strikes filter, specular reflections directed downward, {B) path of light from display, (C) extent of the penumbra, {D) representation of theoretical limit of the light beam, as de- termined by the depth/width ratio of the louver cell. \ \ Two vertical displays have also been considered in the overall RATCC lighting scheme. The first is a large opaque wall map showing the local operating area. This map is 18 ft in length by 8 ft in height. The achromatic background with shades ranging from dark grey through white provides a suitable field for displaying multicolored, magnetic, visual aids. Color coding is used to designate the federal airways, navigation aids, and various holding patterns and approach routes. The map is illuminated by two 96-in. fluorescent lamps mounted in tan- dem. The light from this source is controlled by louvers which confine light to the immediate vicinity of the map and do not affect the radar display. Proper positioning of the luminaires and louvers assures uniform distribution of light over the entire map surface and does not interfere with scope visibility. The second display is a clear-plastic' edge-lighted, status board designed for posting weather information. All external light sources are shielded from this surface in such a way that no direct reflections are visible to any personnel utilizing the data presented. The edge lighting of the plotting board also raises the brightness level of that portion of the room. -42-

OCR for page 33
While lighting of the areas described has been largely as a result of func- tional considerations, action has also been taken to raise the subjective light level of the entire center. In so doing, the under-lighted appearance sometimes associated with directional lighting systems has been greatly reduced. Our work in lighting the Miramar RATCC has included control over the selection, design, and location of luminaires and over the reflectances and textures of the entire environment and its contents. The plan has resulted in viewing conditions suited to the most stringent seeing tasks in a radar air traffic control center and higher, more psychologically satisfying, light levels for personnel in noncritical areas. GCA lighting The remainder of this paper is concerned with an experimental lighting installation designed to improve illumination in standard GCA trailers of the Gilfillan and Bendix types. This example demonstrates the flexibility of using controlled white lighting in areas where initial systems planning did not ade- quately consider the operator's visual environment. Traditionally, the darkened operating area has been considered necessary for maintenance of optimum CRT viewing conditions. There is good reason to believe, however, that the GCA controller is actually at a disadvantage when required to operate his radar display in a darkened area. There are four main objections to the dark operating conditions for GCA consoles. The first objection is that GCA controllers frequently come into the trailer from bright ambient conditions with the result that it requires at least three minutes for them to dark- adapt to the trailer interior. The second objection is glare or halation from the radar displays. To compensate for the high contrast between the bright video display and the usually much darker scope surrounding, operators frequently turn on supplementary light to reduce the contrast. Unfortunately, the only available auxiliary light sources are flashlights, which are difficult to control and sometimes shine directly into the scope face. The third objection is lack of sufficient illumination to carry out emergency maintenance. Under operating conditions, if a failure occurs the technician must either work in the dark or use a poorly controlled, auxiliary light source which may be inadvertently directed at the controller's scope. The fourth objection is lack of sufficient illumination on various knobs and switch controls. Operators frequently use an auxiliary illuminating source to establish positive identification of a particular control. The objections cited indicate the nature of the lighting deficiencies in both the Bendix AN/MPN-5 and the Gilfillan AN/CPN-4 mobile radar sets. The elimination of these difficulties required raising the level of illumination in both trailers, especially at the operating positions. The NEL cross-polarized lighting system appeared to satisfy the objections and to be otherwise functionally suit- able. Accordingly, two experimental lighting installations were set up and studied under operational conditions. Controlled white lighting at the operating positions in the GCA trailer solved satisfactorily nearly all the shortcomings of the visual environment. _43 ~ ~ret ~

OCR for page 33
sensitivity? Although the main purpose of the experiment was to determine the pre-exposure tolerance of the peripheral retina, a limited number of pre-exposure luminances and durations were used to retest the fovea so that the present ex- periment would have some common conditions for comparison with the previous studies. A modified Crozier-Holway discriminometer was used. The instrument consists of two main paths which can be individually controlled with respect to size, brightness, position, and duration of exposure. A third auxiliary path is used to provide fixation points. The pre-exposures are presented by means of one main path, the test flashes by the other. The eye tube enters a light-tight booth where the observer sits. A head holder fixes the observer's eye in position relative to the eyepiece of the instrument. The optical arrangement provides a Maxwel- lian view of both the pre-exposure field and the test patch. For calibration of luminance, three observers made binocular matches in which one eye viewed a field presented by the discriminometer and the other eye viewed a field of similar size and known luminance through an artificial pupil. The luminances presented to the eye by the discriminometer in this ex- periment are equivalent to luminances of a diffusing surface viewed with a pupil size of 3. 77-mm diameter. In the main part of the experiment, the testing of peripheral locations, fixa- tion was provided by a red spot approximately 20 min of arc in diameter. For foveal threshold determinations, fixation was provided by four blue dots posi- tioned at the extremities of an imaginary, vertical cross. When the observer fixated the intersection of the imaginary cross lines, the four dots were positioned 3 above and below, and 3 to the right and left, of the central fovea. The test patch was a square, 1 on a side, which was presented for a duration of 0.033 sec. The pre-exposure field was circular, measuring approximately 55 in diameter. Under all conditions, fixation of the field was central; fixation was provided by ~ ~ ~ ~ . . ~ ~ 1 ~ . _ _ cllagona1 cross-hairs. l Free peripheral locations were tester. 1 ne locations, measured from the central fovea, were 2, 6, and 18 on the horizontal temporal . ~ . meridian. There were two observers, both females in their mid-twenties. Before pre- exposure, the observers adapted to darkness. When foval thresholds were to be determined, adaptation lasted 10 mini when peripheral thresholds were to be determined, adaptation lasted 30 min. Pre-exposure luminances investigated there 0.01, 0.1, 1., and 10 ft-l for durations of 1 and 10 sec. Therefore, the range of luminance-duration products was 0.01 to 100 ft-l-sec. Peripheral locations were tested with all combinations of pre-exposure luminance and duration. The fovea, however, was tested only following pre-exposures of 10 ft-l for 1 and 10 see and of 1 ft-l for 10 sec. Absolute thresholds were determined using a modified method of limits in- volving the ascending series only. The first transtional judgment was accepted as the threshold. One dark-adaptation curve was obtained following each pre- exposure condition for each observer. 66-

OCR for page 33
The results of the retest of the fovea were in slight disagreement with the results of the previous study. There appeared to be a measurable loss of sensitivity following the pre-exposure of 100 ft-l-sec. Whether adaptation is demonstrated following the two pre-exposure combinations yielding 10 ft-l-sec is impossible to state positively, in view of inherent variability in threshold determinations. However, slight adaptation appears to have taken place in at least one of the conditions ( 1 ft-l for 10 see). In general, the following results were similar at all three peripheral loca- tions tested: ( 1 ~ The three pre-exposure combinations of greatest magnitude re- sulted in a substantially greater sensitivity loss than the remaining combina- tions. (2) Reciprocity of luminance and duration appeared to hold for all three reciprocal relations tested. At 2, all conditions resulted in a measurable sensitivity loss. However, the effects of pre-exposure combinations of 1, 0.1, and 0.01 ft-l-sec were not appre- ciably different. At both 6 and 18, pre-exposure combinations of 0.1 and 0.01 ft-l-sec produced little or no loss of dark adaptation that could be measured by the technique used. Effect of pre-exposure size on foveal dark adaptation The same apparatus was used as in the previous experiment, with the follow- ing modifications: (1) A 10x wide-field eyepiece replaced the previously used 20x wide-field eyepiece in order to increase the maximum obtainable brightness. This reduced the maximum subtense of the pre-exposure field from 55 to 37 5, however. (2) The conventional slit mechanisms in both pre-exposure and test- patch paths were replaced by ones which provided circular apertures. (3) The previously used blue fixation points were replaced by red ones (in the same orientation) in order to make the fixation points visible after the higher pre- exposures. A check comparison indicated no differences between dark-adaptation curves as a function of the color of the fixation points. The pre-exposure sizes investigated were centrally-fixated circular fields with diameters subtending 1, 2.5, 5, 10, and 37.5. Each size was presented at the to~ow~ng ~um~nance-oura~on combinations: 1 ft-l for 10 see; 10 ft-l for 10 see; and 1000 ft-l for 100 sec. Thresholds were obtained using the same tech- niques as in the previous studies except that the test patch was a 1 circle rather than a 1 square. .~ r ~ ~ ~ ~ ~ Two observers, a female (EA) and a male (RW), both in their late twenties, served in the experiment. Three curves were obtained, and average curves were computed for each observer under each condition. The average curves were com- puted as follows: The obtained threshold points of each of the three curves were drawn perpendicular to the abscissa at regular time intervals. The points of intersection of the three curves on each perpendicular were read off the ordinate and averaged. In Fig. 1, the average points are enveloped by line segments connecting the highest and lowest intercepts on each perpendicular. The effects of increasing 67

OCR for page 33
BRIGHTNEss SIZE- lo 2.S DURATION I FT-L 10 SEC. 21 ~2 5 10 37 5 2L .~2: lO FT-L 8 IO5EC. I I I 24___ 2: _2~ 2~__2_ a' -- 2 jet 22 24 air: 0 100 200 400 600 0 100 200 600 0 100 200 400 600 o 100 200 400 600 0 100 200 400 600 TIME -SECONDS Fig. 1. Average foveal dark adaptation curves for two observers (filled circles - RW; open circles - EA) following various sizes of pre-exposure field presented at three luminance-duration combinations. pre-exposure size for a given luminance and duration of pre-exposure are shown by the five curves that appear side by side in a given row. In general, there appear to be no differences in the curves as a function of pre-exposure size for any of the luminance-duration combinations employed except after 1000 ft-l for 100 see with a 1 pre-exposure field. This curve shows a lower initial threshold and reaches a final threshold earlier than the other curves of the row. This may be a real effect, or it may be an artifact resulting from the inherent difficulty in reliably testing a small retinal area with a test patch the same size as the pre- exposure field. Imperfect fixation could result in both incomplete pre-exposure of the retinal area to be tested and inadvertent testing of unexposed retina. To test the hypothesis that imperfect fixation could be a factor, a 1/3 test patch was used to compare the effects of 1 and 37.5 pre-exposure fields at 1000 ft-l for 100 sec. The results were inconclusive since both observers showed some size effect, although the magnitude of the effect was markedly reduced. Effect of pre-exposure size on peripheral dark adaptation The apparatus used in this experiment was the same as in the previous experiment except that fixation was provided by a single red fixation point. Three peripheral locations were tested: 2, GO, and 15 on the horizontal tem- poral meridian of the right eye. A preliminary experiment, using a wide range of centrally fixated pre-exposure sizes, had indicated no differential effect of pre-exposure size when the pre-exposed area did not include the area being tested. Therefore, in order to concentrate the data, three pre-exposure sizes were chosen to test each peripheral location. These sizes were such that the small size would not pre-expose the area to be tested, the intermediate size just included the area to be tested, and the large size was as large as the apparatus would per- mit. The sizes used to pre-e~`pose each location were: 1, 5, and 37.5 at the 2 68

OCR for page 33
location; 2.5 ' 15' and 37.5 at the 6 location; and 5, 32, and 37.5 at the 15 location. Two pre-exposure luminance-duration combinations were used: 0.1 ft-l for 10 see, and 10 ft-l for 10 sec. Three dark-adaptation curves were ob- tained for each condition by the same observers as in the previous experiment. The results are shown in Figs. 2 and 3. Average points and ranges are shown as before. All pre-exposure conditions resulted in some loss of subsequent sen- sitivity. For all sizes and locations' the greater luminance during pre-exposure resulted in more loss of sensitivity. For those sizes which did not stimulate the tested area, this effect may have been due to the increased amount of stray light in the instrument and eye. RE~NAL LOCATION TESTED 2 6 tn 3 he or m 15 SIZE 2 t . .~ ...: ~ ~ ~ SMALL INTERMEDIATE LARGE 2t 3 E. . . ~ 3- . con- 3 4' , ,, ,, , 1 ~ , . . ,, , I , i.t,l.' . ~ ~ l 1 l 1 ~ 111 I'DI: 4~' 11.11 , ~ I ~ I ~ I I'1'1~ I ~ I I ~ ~ islet- ~ ~ ~ ~ I ~ ~ In 1 . ~ 34= 4 o loo 200- 400 600o 100 200 400 600 TIME-SECONDS 2- . . ~ . 3 .' . O. j Q..,'t. ~ .~,,,..~ ' - ` ~ 2 ~ ~ Fig. 2. Thresholds at three peripheral locations for two observers {filled circles - RW; open circles - EA) following presentation of three sizes of pre-exposure field at a luminance of 0.1 ft-l for 10 sec. For all locations, the pre-exposure sizes that stimulated the area to be tested had a much greater effect on dark adaptation than those which did not stimulate this area. The sensitivity loss following the intermediate pre-exposure size was as Great as the sensitivity loss following the large pre-exposure size. The sensitivity loss at the 15 location, following the intermediate- and large-size pre-exposures, was greater than the loss at 2 and 6 locations. How- ever~ the overall effects of size were similar at all retinal locations tested. The results of these three experiments, together with the results of the previous studies, indicate that maintaining foveal sensitivity with "white" illumi- nation is probably feasible. Adaptation appears to be only slightly affected 69

OCR for page 33
RETINAL LOCATION TESTED 2 2 3 l 4L 6 ~_ o 1 an an At, he - following ~easonably adequate luminances and durations of large pre-exposure fields. The maintenance of peripheral sensitivity, however does appear not to be feasible. An investigation is now being made to determine the length of dark period necessary between pre-exposures to avoid the accumulation of effects during long series of pre-exposures. This investigation will be followed by a wavelength study on the effects of various narrow-band pre-exposures on thresholds for colored test patches. Assuming that these studies enable us adequately to specify an illumination system capable of preserving foveal sensitivity' it seems de- sirable next to determine the visual tasks to which the system applies. Sl Z E SMALL W...:~-.-. 2-1 1 ~ I l 1 ~ 1lil1 , ~ I, 1 ~ - 4 2 3 '`.,,,,0 on' 4-: 1 l 1 ~ 16111 :~ 4 . INTERMEDIATE LARGE i_ ~ ~ l i ' I ' I 'I'm I :,~Do ., ~ o ~ . ~ ~ '1 ' 1 l 1 ~ 111 . . l ~ ~ 1 ~l 1 ~ 1'1'1 1,,,, 1 1 1 1 l ill _ _ . 2- 2 5 3 3 4'~- 4 o 100 200 400 600 ~_` 3 . o -- 4 t: [A 4~ ~- ..... . . . . . .. . . . . . , , . . . . . . o 100 200 400 600 o 100 200 40~0 600 TIME-SECONDS Fig. 3. Thresholds at three peripheral locations for two observers (filled circles - RW; open circles - EA) following presentation of three sizes of pre-exposure field at a luminance of 10 ft-l for 10 sec. -70

OCR for page 33
Making Raciar Inclicators Useful in High Ambient ItIumination J. R. ROEDER, Westinghouse Electric Corporation Summary The design of the electrical portion of the indicator must take into account pulse width, pulse repetition frequency, sweep speed of the indicator, antenna scanning pattern, and video amplifier bandwidth. The characteristics of the converter must be considered in selecting the picture tube and the phosphor in the picture tube. The design of the optical portion of the indicator must take into consideration the geometry of the specific installa- tion. Hood design cannot be divorced from geometry, and the magnitude of the reflection problem is determined by the geom- etry. The best present-day indicating system for fighter aircraft consists of a bright tube, a circular-polarizer filter with a non- reflecting first surface, and a sun shield. The radars of fighter aircraft are excellent today, with the indicating system probably the weakest link in the chain for the single-place aircraft. In a one-man aircraft when the pilot must look at the sun, clouds, and sunlight on his instrument panel, it is very difficult for him to look also at the picture tube of his radar indicator and see small, weak targets. The smaller and weaker a target the pilot can see, the greater is the detection range of the radar. Every mile that can be added to the range of a fighter aircraft radar is of great im- portance because of the supersonic speeds at which aircraft can fly. Obviously, the higher the speed of approaching aircraft, the less is the time available for maneuvering and for preparing to fire. Every second gained is important. High- speed aircraft, approaching each other at several times the speed of sound, do not have enough seconds to get ready to fire missiles. The pilot must see the target before the weapon system is of any value. In the 20-year development of radar, a great many improvements have been made. The signal-to-noise ratio has been improved, the noise of the first amplifier tubes has been reduced, improvements have been made in crystal noise, etc. A great deal can be found in the literature about these subjects, but prac- tically nothing can be found on the use of indicators under high ambient light conditions. It has been only recently that the operation of radar under high ambient light has been extensively studied. Therefore, it is the author's opinion that there is more to be gained at this time towards improving detection range by expending engineering effort on the indicating system than on any other portion of a radar. Detection range must be increased to make fighter radars more useful. Electrical inputs to the indicating system are shown in Fig. 1. The video signal input has noise plus pulses with the intelligence. The characteristics of the pulses are: (1) pulse width (parsec), (2) pulse repetition frequency (prf), and (3) sequence repetition, determined by antenna pattern. The output is optical. The hood is used to keep direct sunlight off the face of the picture tube. The optical output must be bright enough so that the pilot can see it in the presence of -71

OCR for page 33
bright sunlight. At high altitude, most of the light appears to come up into the pilot's eyes, being reflected off clouds. SUNLIGHT FIRE CONTROL RADAR I NDICATOR E LECTRI CAL OPTI CAL I N PUT O UTPUT 7 HOOD in / ~ / PILOT'S / EYE SUNLIGHT REFLECTED OFF CLOUDS Vl DEO SIGNAL Fig. 1. Fighter-aircraft radar indicator blocic-diagram. To obtain the best detection range with the electrical inputs, we must get as much light output as possible. One thing that must be done is to make the range- sweep speed compatible with the minimum scot-size of the picture tube. The slowest range-sweep in Fig. 2 takes about four times as long as the pulse width to cross the minimum spot-size on the picture-tube phosphor. This means that the electrical signal causing the light output of this spot is the average of the target pulse and the noise. The apparent target intensity is as shown at the bottom of Fig. 2. As the range-sweep is made faster, the time required to generate the minimum spot-size is about twice the target pulse time. Hence, the average of the target pulse and noise gives a larger apparent target intensity than for the slower sweep. The next sweep shown crosses the minimum picture-tube spot-size in a time equal to the target pulse width. At this sweep speeds the optimum ONE MILLIMETER DEFLECTION I N t 6 MILLIMETERS | 4 2 APPARENT ~ , TARGET I NTENSITY . _ _ M I N I MU M SPOT SIZE ~TIME(microseconds) ,L] ~11-RADAR SIGNAL Fig. 2. Detection is improved when sweep speed is optimized for pulse width and minimum spot size. 72

OCR for page 33
apparent target intensity is achieved. Making the sweep faster, as shown, will obviously make the target larger but will not make the target more intense per unit area. In order not to deteriorate detection range, the indicator must have a sweep speed greater than that determined by the minimum spot-size of the picture tube. In actual equipment for the case of a 50-mi range sweep, a 10 per cent increase in detection range was calculated. Actual flight test data indicated better detection range for a very small target on a 12-mi range than on a 24-mi range, thus verifying the calculations. The effect of picture-tube phosphor characteristics on light output is shown in Fig. 3. The pulse width and the time between pulses affect the light output. These parameters are often determined by other requirements of the radar, but information about the relationship of these parameters to the light output and what the pilot can see can be used to improve detection range. Conventional picture tubes do not store information long enough to make the antenna pattern of any importance. However, with the storage tube there is a buildup of light output by additional passes of the antenna across the target. The insert in Fig. 3 shows the relative light buildup for a conventional picture tube with P2 and PI phosphors under pulsed operation which is typical of radar indicators. This shows how important the selection of the phosphor is to obtain the most light output. The time between target pulses is typically 1,000 times as long as the target pulse. The greater the duty cycle or ratio of pulse time to off time, the better is the detection range. mono 900 800 ~ 700 ~n At 600 by 500 400 - / 300 1 \1 / / / / ,, ~ Ah\ ~ ~ ~ P2 TIME- ~ Vl DEW Fig. 3. Pulse width and time between pulses determines brightness of picture tube. The transfer function of voltage input to light output for a conventional picture tube is shown in Fig. 4. Target signals of three different amplitudes are shown with idealized noise signals. To improve detection range, the small target signals are the ones to study. The amount of light visible to the pilot's eye will change with the ambient illumination. Hence it follows that the electrical signal required to produce this light will vary with the ambient light. It is also inter 73

OCR for page 33
esting to note that the gain of the conversion of the electrical signal to light is not constant because of the changing slope of the curve. There is better gain at high- intensity light output. LOGA R ~ TH M I 1 1 VOLTAGE FOR / I ZERO LIGHT OUTP;/ I ~- - ~ ELECTRICAL SIGNAL (volts ) == , - o - z w I J Fig. 4. Transfer function (volts to light). Thus far we have considered electrical-output signal characteristics, as well as how variation of the component parts of the signal affect detection range. The transfer characteristics of the picture tube (the electrical to optical converter) have also been shown. This leaves the strictly optical portion to be considered. Reflections are a big problem. When enough target light is available, a circular-polarizer filter has been found to improve greatly the usefulness of the indicator. Test pilots report very favorably on circular-polarizer filters when used with bright tubes. The light from the pilot's bright-orange life vest, his white, shiny helmet, and other bright spots pass through a linear polarizer. It is then rotated 45. If it should be reflected from any one of a number of surfaces, it makes no differ- ence, since it is rotated another 45 on its return trip. It is now rotated 90 with respect to the linear polarizer and' hence, is greatly attenuated as it passes through on its return trip. It should be noted that, with conventional picture tubes, the 60 per cent one-way loss of light cannot be tolerated, but with the brighter picture tube, the circular polarizer appears to give a very great improvement. The 74

OCR for page 33
circular polarizer reduces all rear-surface reflections. If a nonreflective coating is added to the first surface, we have improved our equipment greatly as far as reflections are concerned. If light could be eliminated from the face of the indicator, we would not have a reflection problem. Many types of hoods have been built, but as each hood must be designed for a particular installation, it becomes a problem in geometry. The installation of the radar indicator in fighter-aircraft cockpits is very important. From all the work Westinghouse and others have done, both in development and on specific problems, it is felt that the best state-of-the-art fighter-aircraft indicating system consists of a bright tube, a circular Polaroid filter with a non- reflective coating on the first surface, and a simple sun shield, as shown in Fig. 5. The bright tube is used to overcome the high-ambient-light problem. The cir- cularly polarizing filter is used to eliminate all multiple rear-surface reflection. The nonreflective coating on the first surface is used to eliminate first-surface reflections, and the indicator has a simple sun shield to keep off direct sunlight. Over a dozen flight tests with the bright-tube indicator have yielded excellent results in comparison with existing equipment. With direct sunlight on the in- dicator at 30,000-ft altitude, targets are visible, but further improvement ap- pears possible. BRIGHT TUBE I NDICaTOR - l SIMPLE SUN SHIELD 'en\ C IRCULAR NON-REFLECTIVE POLAROID SURFACE FILTER Fig. 5. Best indicating system flight tested. 75

OCR for page 33