Reconciling the differences

During the early 1960s a number of new criteria of good lighting were introduced. First glare index, then vector/scalar ratio as an index of modelling strength and the direction of the vector as direction of modelling, in addition to the illuminance of surfaces of all orientations, measured by scalar illuminance alone.

There were also authors who recommended preferred ratios of luminance between task, surround and background. These were all in addition to the previously accepted criteria of working plane illuminance, economy and emphasis. Although there was no accepted measure of this last, it was generally agreed to be important.

I began to ask if it was possible to design an installation which complied with all criteria simultaneously, and if it were not, which should take precedence?. For instance, glare reduction involves either ensuring a high illunuinance on the ceiling (how does this square with emphasis?) and/or a strict limitation on the illuminance of the walls (how does this square with apparent brightness and/or preferred luininance ratios?).

Satisfying all criteria
It turns out that in rooms with reasonably light decorations (as recommended in the then IES Code) where the working plane is horizontal, and lighted by a regular array of not less than four luminaires on the ceiling, (that is to say, most ordinary offices and classrooms etc) there is a range within which all criteria of good lighting, with one exception, can be satisfied simultaneously.

This range could be defined in many different ways, but the most convenient is a set of limiting ratios between the illuminances of walls and working plane, and the ceiling and working plane. The figures are:

• walls/working plane 0·5 – 0·8;
  
• ceiling /working plane 0·3 – 0·9.

Recommended reflectances are:

• walls 0·3 – 0·7;
  
• ceiling cavity 0·6 or more;
  
• floor cavity 0·2 – 0·3.

It must be stressed that illuminance ratios are a convenient way of defining the limits of the optimum range, and are not solely criteria in themselves, although they are consistent with good practice for installations of their type. The one criterion of good lighting which is not satisfied within these limits is direction of modelling which under a regular array of luminaires will always be vertical, whereas the preferred direction is oblique. However, in rooms with side windows the direction of modelling will be acceptable over a large part of the working year, since the amount of daylight required to correct this angle is far less than is required for adequate horizontal illuminance.

My 1968 paper1 appeared in the references to successive Codes, and, for instance, appears to form the basis of figure 2.1 on page 57 of the 1984 Code. However, no further explanation is given except that compliance will ensure 'visual comfort', without any reference to modelling or illuminance of three dimensional objects inside the room.

A modified Figure 2.1 appears in the 1994 Code, but in this case the relative wall illuminance is limited to 0· 5 - 0·6, which does seem very restrictive, while recommended floor cavity reflectance is between 0·3 and 0·7. No reason is given to justify the figures except to say that they are based on research findings, 'modified by design application and experience'.

I am not aware of any application or experience which could possibly justify such a straightjacket for wall illuminance ratio, nor have I seen any real room with a floor cavity reflectance as high as 0·7, which requires that the surfaces lining the cavity have reflectances not less than 0·8. Perhaps the Code panel was thinking of an integrating cube. Further, the 1994 Code implies that the luminance of the walls is the real criterion, and that the recommendations are expressed as illuminances solely because this is what lighting designers use to control the appearance of a wall of given reflectance.

If LG3 now merely quotes minimum figures for illuminance ratios and suggests that these are desirable purely to prevent the walls and ceiling appearing unduly dark they have, once again, totally missed the point.

Further, with a floor cavity reflectance of 0·7 it is very difficult to keep the ratio of wall illuminance to working plane illuminance down as low as 0·6.

I understand that the new Code says much the same as the 1994 Code, except that in one place it gives limiting wall ratios of 0·5 - 0·6 and in another 0·5 - 0·8. Further, the recommended floor cavity reflectance is now 0·3 - 0·5 which is a small sign that reality is beginning to break in, but one cannot help asking where these people have been living – in igloos perhaps? The question of what to do when the reflectances are not in accordance with standard recommendations, or where there is some special reason to deviate from normal practice, is important. Kit Cuttle took the basic idea and from it developed the 'multiple criterion design method' published as IES Technical Report No 15. This includes IR charts which enable one to read off the detailed properties of any given installation, whether it complies with the recommended limits for reflectances and illuminance ratios or not. This method is described in all the Codes. What a pity that so few people seem to have read about it.

Engineers often make the mistake of assuming that what matters are relative luminance ratios within the visual field. They mention simultaneous contrast in colour, but appear to be unaware of the importance of brightness or lightness constancy. That is to say, in normal interiors we do not perceive in terms of luminance, but of objects and surfaces of defined colour and texture which appear to be more or less strongly illuminated. But in a room whose walls are panelled in dark timber, judging by relative luminance ratios would mean that the illuminance of those walls should be increased to several times that of the working plane solely in order to ensure that the ratio of their luminance to that of light coloured desks is not less than 0·5. I do not understand this obsession with the luminance of walls – as if totally empty rooms are especially important? Many years ago J M Waldram pointed out that it is no use discussing the details of lighting where there is nothing to light.