Lighting Research & Technology

Glare which affects comfort or causes distraction is distinguished from glare which affects the ability to see. The results of an investigation to determine the factors which cause discomfort glare in lighted streets are summarised, and the influence of these factors on problems of street lighting design is studied.

Discomfort glare was found to be influenced by the general brightness of the scene, by the angle between the direction of a glare source with the line of sight, and, under street lighting conditions, by both the intensity and intrinsic brightness of the source. The glaring effects of a number of sources were found to be additive. No difference in the glaring effects of yellowish, bluish, and white light could be detected.

It has been found possible to express the results in a form easily applicable to problems of street lighting design.

Attention is drawn to the difficulty of assessing the visibility of an object when the difference in brightness between the object and its background exceeds the value necessary to produce the best performance under the particular experimental conditions. It is suggested that if the visibility scale is to be extended to brightness differences which exceed this value, then the visibility of the object must be considered in terms of the observer's performance under all conditions of sensitivity, distraction, attention, and unexpectedness which can occur in practice during the carrying out of the visual task.

Meters which have been designed to measure the visibility of an object depend upon making the visual conditions more difficult by producing optically a condition in which the difference in brightness between the object and its background is equal to a threshold value. It is shown that in the case of objects which are presented against backgrounds of different brightness, these objects will be arranged in an order of visibility which depends on the exact manner by which the threshold condition is produced. Inconsistencies which may arise from the use of a meter both in the presence and absence of glare sources are also indicated. The need is stressed for careful consideration of the results given by a meter both in relation to the nature of the visual task and the differences in lighting conditions under which comparisons of visibility are made.

Consideration is also given to visibility meters which have been designed to measure the efficiency of street lighting installations. As these meters take into account many of the factors affecting the visibility of objects in the street, they can usefully be employed to study the effect of these factors. On the other hand, since the meters do not take all the important factors into account, the results given by the meters in different installations should not be taken as measuring the efficiency of these installations in promoting object visibility.

Photometric data of some luminescent paints and lacquers are given which enable these materials to be treated as items of illuminating engineering equipment.

Factors which affect the brightness of a luminescent material are shown to include the type of material, its weight per unit area, the nature of the backing surface, the type and intensity of the exciting radiation and the temperature of the material. The effect of these factors is described and the general effect of deterioration in luminescence with life investigated.

The phenomenon of phosphorescence has also been investigated for certain materials.

A method is given by which luminescent effects may be calculated from a knowledge of visual data of a similar installation.

After a brief historical survey of attempts at the realisation of a physical eye of defined spectral sensitivity, the earlier methods and technique employed are appraised. Subsequently, the problem is stated, both from the theoretical and practical aspects.

The main treatment of the paper deals with the description of the multi-purpose apparatus which the authors have developed and which is suitable for use as an “average eye,” as defined by the Commission Internationale d'Eclairage. It can be adapted for use as a physical colorimeter indicating directly the C.I.E. trichromatic coefficients of coloured light and as an instrument for measuring spectral luminosity distribution in terms of the proportion of the total luminosity in each of 8 defined spectral bands.

Finally, the possible general application to commercial photometry, colorimetry and colour rendering measurements, is discussed.

This paper first outlines the problems encountered in the photometry of coloured light sources, in particular as they are associated with the colour-characteristics of the eye. The optimum conditions for such work are shortly discussed. In the light of these findings the various experimental determinations of the mean relative luminosity function are examined seriatim under two headings—first, the investigations forming the basis of the agreed Standard Curve, and, second, more recent measurements differing somewhat from the older ones. Possible reasons or the observed differences are enumerated and examined.

The author concludes that there are at present no certain grounds for disputing the acceptability of the Standard Curve, and supports this conclusion with evidence derived from general photometric experience. A final section of the paper deals with the peculiar difficulties encountered in heterochromatic photometry at low brightness-levels, where the colour sensitivity of the eye varies greatly with the level used.

Observations were made in the laboratory, on the effect of varying the flash frequency on the apparent intensity of flashing lights having durations of 1/40th second and 1/10th second. The majority of the readings were taken at an illumination level of 0.5 mile-candle on the eye. This is the desirable lower limit of illumination on the eye for adequate conspicuity of marine and aviation lights. Observations had been reported indicating that when distant lights were flashed at frequencies above that at which the flashes were distinct, the intensity appeared greater than Talbot's Law would suggest. It is shown that these observations, which were not made under controlled conditions, were evidently in error. Up to frequencies of about 5 flashes per second the light appears as flashing and its apparent intensity can be obtained from the fixed light intensity by the application of the Blondel-Rey correction. At higher frequencies than about 13 flashes per second (with the shorter flashes) the light appears continuous and the Blondel-Rey correction is no longer applicable; instead Talbot's Law applies. Between about 5 and 13 flashes per second the light appears to flicker and the apparent intensity is greater than either Talbot's Law or the Blondel-Rey correction would indicate. This flickering light is not thought to be of any practical value as its apparent intensity is less than a flashing light consuming the same amount of energy but giving a readable flashing indication.