It is questionable whether the consideration of luminance ratios should be in a separate section of this report. Discomfort from excessive contrast (ie: one surface which has a very high luminance in comparison to a neighbouring surface – a large luminance ratio) is often referred to as glare. Many of the glare indices discussed above were simply measuring the luminance ratio.

However, the existing research in the field of visual comfort is divided. Some researchers simply consider luminance ratios to be glare and thus do not distinguish between the two. In contrast, other researchers specifically investigate luminance ratios as a critical subset of glare. It is because of the clear impact of luminance ratios on the level of perceived glare, that a separate section has been designated to it in this report.

A study by Newsham and Veitch indicated that occupant visual satisfaction was driven by the amount of screen glare (veiling reflection) and the desktop luminance ratio. A mock-up office was prepared in which the users each worked on a computer for a day under various lighting conditions. At the end of each day the occupant was asked to adjust the dimmable lights to achieve the most satisfactory environment. The desktop luminance ratio was clearly a significant factor in the level of satisfaction. Although this experiment was targeted at artificial lighting applications, it still shows the importance of luminance ratios.

Recommended Practice

An extensive study conducted by van Ooyen et al investigated the preferred luminances of the walls and work surfaces in three identical offices. Every aspect of the offices was the same except for the luminance of the walls and working plane, which were variable. In total, the experiment involved 180 participants who performed reading, writing, VDU and conferencing tasks in the test rooms. The participants each completed a questionnaire consisting of 38 semantic scale-based questions. Unfortunately, the investigation did not consider daylit conditions. The results of the study are interesting nonetheless.

• The preferred working plane luminance is inversely proportional to the wall luminance.
• The preferred luminance ratio between the task, the working plane and the wall is dependent on the specific task – the ratio is lower for VDU tasks.
• The wall luminance contributes the most to the spatial experience.
• Work tasks can be satisfactorily combined under one luminance condition.

• Preferred Luminances:
o Non-VDU: Wall 30 – 60 cd/m2
Work Plane 45 – 105 cd/m2
o VDU (white text on black): Wall 20 – 45 cd/m2
Work Plane 40 – 65 cd/m2

• Luminance Ratio (Task : Work Plane : Walls)
o Preferred 10:4:3
o Relaxing 10:5:2
o Stimulating 10:3:4

Whilst these results are for artificial lighting, some of the results are still of use. The actual figures (eg the specific luminance values preferred) may be unreliable for daylit situations. Howeve,r it is evident that the occupants want the task to be brighter than its surrounds. It also appears that VDU tasks should have a lower luminance than non-VDU tasks.

The study by van Ooyen et al above is just one of the many investigations completed to identify the preferred luminance ratios for artificial lighting. The historical development of these recommendations, from 1928 to the present day, is detailed by Osterhaus. It was found that most of the studies into preferred and acceptable luminances produced similar results. There was a noticeable trend that the recommended luminance ratio for task : immediate surround : distant background was 10:3:1. However, most of the research reviewed in the paper was based on artificially lit environments. Whilst this ratio is reliable for artificial lighting, it may not apply in daylit offices. Osterhaus suggests that this ratio is unrealistic when looking at a window. Screen luminances of computers are typically 50 – 100 cd/m2, whilst sky luminances can reach 5, 000 – 10, 000 cd/m2 even under overcast conditions. This will lead to a luminance ratio of 100:1 – far greater than the suggested preference and yet people still position their workstations so they face the window, indicating the luminance contrast cannot be that bad. This ratio is also questioned in the fact that the human eye can adapt to contrasts of 100:1 and possibly 1,000:1.

Berrutto and Fontoynont have also questioned the merit of such rules of thumb for luminance ratios. They raise very relevant points regarding poorly explained details. What is the precise distinction between task, surroundings and background? Should average or extreme luminance measurements be used? Where exactly should the measurements be taken, particularly for the background?

The limited research regarding acceptable luminance levels in daylit offices means there is no definitive goal. However, there are several sources for recommendations on acceptable luminance ratios, some of which are listed below:

• Egan
o Task to adjacent darker surroundings 3:1
o Task to remote darker surfaces 10:1
o Lighting fixtures / windows 20:1
to sizable adjacent surfaces
o Field of view max 40:1
• IEA Sourcebook
o Variation in luminance across task max 2.5:1 to 3:1
o Task to background max 3:1
o Task to remote surface 10:1
• IESNA
o Task to surrounding max 5:1
o Normal field of view max 4:1
o Windows to adjacent surfaces max 2:1
• Newsham and Veitch
o Field of view max 20:1

NB: Definition task = 50 solid angle [field of view from eye]
Near surround = 50 < 600 solid angle
Far surround [background] > 600 solid angle

Assessment Methods

Luminance meters are the obvious tool to use for measuring the luminance values of surfaces / objects in an office. But typical luminance meters only provide a spot measurement. In the time it takes to perform the number of measurements required in the office (which in itself is laborious) the sky conditions could have changed considerably.

The development of CCD technology for luminance mapping could provide a solution. “Basically, CCD photometers are based on picture digitization [taking a digital photo] and the conversion of pixel values into luminance values. Accuracy is lower than for standard spot luminance meters but acceptable for lighting quality evaluation (i.e. within  10%). Measuring time is tremendously reduced, as well as the time spent for analysis and reporting. Generally speaking, one luminance distribution requires 2 or 3 picture shots, i.e. less than 1 minute.” However, Berrutto and Fontoynont go on to say that the calibration required for the equipment is difficult. This complication may make the technology unsuitable for a simple and easy assessment procedure targeted in this report. Furthermore, the cost of the necessary equipment is also a hindrance, although the price is slowly decreasing.

A final alternative is to use illuminance meters to measure luminance as suggested in Aizlewood’s research above. This process is even more complicated and time- consuming than simply taking a luminance measurement. The only advantage is that it does not require a luminance meter or CCD equipment.

Possible Measurements Needed in Assessment Procedure

None of the possible assessment methods above are perfect. In the future the CCD technology is likely to be the most suitable technique of assessing the luminance ratios for visual comfort. At present, simply using a luminance meter appears to be the best method available. There is an opportunity for the required measurements to be taken when performing the glare calculations discussed previously.

Even if there was a perfect method available for recording the luminance values in the office, the assessment of those values would still pose a problem. There is no agreed set of recommendations for luminance ratios in a daylit office. The 10:3:1 rule of thumb could be used (with caution) for parts of the office with no window in the field of view. However, when looking at the window there is no accepted standard for comfortable luminance ratios and very little research in the area.

.1 Future Research Needed

Further development of the CCD technology is needed to make it cheaper and remove the difficulty of calibration. Once CCD photometry is readily available the data it is able to gather will need to be satisfactorily analysed. This will require a lot more research into what the acceptable luminance values and ratios for daylit office are.