Natural Lighting

In early November CIBSE unveiled its new Environmental performance toolkit for glazed facades (TM35). Its production was funded by the DTI and industry, and it is intended to provide designers with a means of establishing the types of facade that can be used when a given set of environmental variables has been specified.

The authors of this toolkit did not imagine that it would eliminate the need for the individual design of facades, but that it would provide the starting point for the consideration of how the choice of facade might be limited by various environmental loads. The toolkit is an effective means of highlighting the importance of the facade as a principal moderator of the external environment and also provides a relatively easy way of gaining some sense of the importance of various environmental factors.

It is inevitable that in developing such a tool various assumptions have to be made about the external environment, the design of the facade and the way that the internal environment is controlled. These clearly impose restrictions on the applicability of the results. This does not invalidate the usefulness of the toolkit, but it makes it important that its users do not apply its results blindly.

This is particularly true when considering how the facade determines the daylighting of a building, and in designing for natural lighting.

Designers and researchers need sometimes to be a little more honest with themselves and their clients. Good environmental conditions should be provided for humane reasons and to help productivity. However, it should be recognised that proving a link between good environmental conditions and productivity is very difficult.

Those studies that purport to show improvements are rarely designed to separate the effects of motivation and changes in the physical environment. We are all aware of how crucial morale is to our own performance. My own productivity can fall to almost zero under stressful circumstances. It would be very difficult to demonstrate that small improvements in individual productivity were significantly related to changes in the physical environment when such enormous changes in productivity result from changes in a person’s mood.

A series of experiments undertaken at the factories of the Western Electric Company in the US during the last century led to the identification of a phenomenon known as the Hawthorne effect, named after the location of the experiments. It describes the overwhelming nature of motivational effects when workers were presented with changing environmental conditions. This does not mean that there is nothing to be gained by providing a good physical environment, but that it is difficult to prove any gains conclusively in most practical circumstances.

Natural light is constantly changing and indeed this is one of the perceived attractions of natural light. However, this natural variability is a cause of some difficulty to designers. One way of coping with this variability is to ignore it and consider only the extremes of natural lighting. Thus in the past, the design of natural lighting was primarily directed to ensuring that a design worked under two extreme conditions. These two conditions or limiting cases were, the completely overcast sky and a clear sky with sunlight. It is possible to distinguish between the two by describing the design of natural lighting under an overcast sky as daylighting, and that under a clear sky as sunlighting.

There are good reasons for choosing these two limiting conditions. Both can be simply defined and therefore easily used within a design procedure. The overcast sky may be taken as representative of a minimum worst case, and if a design is acceptable under such conditions, then it may be assumed that it will provide sufficient daylight for 85% of the working year. A clear sky with sunlight may be taken as representative of a maximum worst case for solar glare and heat gain, and if the shading or control can withstand such conditions then it is accepted that overheating and glare are tolerable.

The criteria used to assess daylighting would usually be the minimum or average daylight factor within a room. Additionally, a designer might assess the angle and intensity of the daylight vector and the maximum discomfort glare under the brightest overcast sky likely to be experienced.

Sunlight design would usually depend upon the maximum solar gain and a qualitative assessment of where sun patches would occur. Solar shading might be designed to eliminate or reduce solar gain and it might also be that a check would be made on the balance of brightness between patches of sunlight and the principal areas of view.

However, although these criteria help in the design of the daylight under an overcast sky and sunlight when there is a clear sky, they do not necessarily result in a design that will ensure a minimum energy expenditure throughout the year. Indeed the very nature of the criteria being based upon limiting conditions can result in problems.Two examples demonstrating this are: the overheating of classrooms designed for a minimum daylight factor of 2%; and the over-specification of cooling plant designed to cope with shading coefficients that relate to sunlight at normal angles of incidence.

The increased importance of energy expenditure and the need to seriously consider the sustainability of any development has meant that it is no longer appropriate to consider only these limiting conditions. It is now necessary to consider how a building will respond to natural lighting in all its variability.

Design is an iterative process that depends upon the assessment of an existing solution and a subsequent modification that hopefully produces an improved solution.

Designers are now called upon to look at the balance between daylight provision and the use of electric lighting; the net balance between solar gain and heat loss; and the utilisation of daylight when shading is used to reduce solar gains.

It is not at all easy to draw all these strands together to produce an efficient design. Software that allows designers to simulate the use of a building throughout the year has provided the illusion that a design process is available. There is a temptation to use a simulation program as if it were a black box. Various arbitrary designs might be loaded into the package and the best solution of those submitted chosen, but it is seldom, if ever, the case that this id the true optimum. Also, without understanding the interactions of the various factors, it would be very easy to end up going round and round in circles rather than homing in upon a viable design.

Designing for natural lighting can only be successful if the role of the facade is clearly understood. Interactions between the use of electric lighting and the deployment of shading are critical to the successful exploitation of natural lighting.

Often though, one can still see offices where all the electric lights are on and venetian blinds haphazardly lowered towards a cloudy sky. The savings in electric lighting that are predicted if natural lighting is properly exploited just do not seem to be being delivered. Designers cannot simply shrug their shoulders in despair at the way users respond to their designs. Clearly, designers of natural lighting have not yet got it right, and we need to apply additional considerations.

The control of the facade and the way in which users respond to the control of both the facade and the control of the internal environmental services is very important. Top up electric lighting controls that switch the lighting suddenly on and off are known to produce an adverse reaction from users, and yet this is the control strategy adopted in many simulation programs. It is most unlikely that we will optimise a design until we only adopt control strategies that users find acceptable. Users must understand why an automatic control activates a particular device and they do not want to be alerted constantly to changes. This might mean extra effort being put into educating the user, but just as likely is the need for a reassessment of the control strategy itself and the need to apply some principles of ergonomics to interaction between the user and the building.

Users appreciate the view through a window and the degree to which fenestration relieves the sense of architectural enclosure. Because for so many buildings there is a need to provide some shading, the manner in which the shading affects the response of users to their sense of view and enclosure needs to be properly considered. Views through screens and shades depend upon many factors and among them are the area of opening, its spatial frequency, the orientation and the relative luminance of the screen and the view. The sense of enclosure has received less study, but it is an important aspect of the window in a room.

Incorporating some form of shading or combination of specialist glasses not only affects the view out, but it also has a considerable effect on the external appearance of a building. How often is it that designers do not achieve the image that they desire? In descriptions of their architecture, designers often use the property of transparency as one of the driving objectives of their designs. It would appear that often the reality falls far short of the intention. The reasons for this may be complex, but it seems to be time that designers fully appreciated the visual implications of using specialist glasses, external shading, the variety of natural lighting and the direction from which the building is viewed.

The final point to make is one about feedback on buildings. It is often difficult to gain from building owners and occupants, but it is crucial for designers.

It is quite likely that the need to save energy is going to fall disproportionately on building users, and without proper feedback on new buildings, designers are going to find it difficult to achieve the savings demanded while achieving acceptable internal comfort.

Design is an iterative process that depends upon the assessment of an existing solution and a subsequent modification that hopefully produces an improved solution. Design education is much the same. Without examples of old solutions and a sensible critique of their designs then how are designs and design abilities to be improved ?

It may be that the industry now needs some type of forum where designs can be critically assessed without the threat of legal action. Without access to fair feedback, it would not be surprising if the quality of design did not improve.

• Martin Wilkinson is a lecturer in environmental design at the University of Bath department of architecture and civil engineering. See www.bath.ac.uk