Lighting for turbulent spaces

Lighting controls are often claimed to cut 50% off the energy consumption of a conventional lighting installation. Such claims are highly dependent on the quality of the controls or their ease of use by building occupants.

People often prefer lower levels of illuminance than that assumed at the design stage. On that basis, provision of local choice may yield greater success than a blanket illuminance level suggested by current guidance.

An investigation into the value of user-controlled lighting has been carried out by the University of Liverpool and the Building Research Establishment (BRE). The work sought to discover whether the need for organisations to be adaptable and flexible in their working arrangements should be reflected in the provision of electric lighting. Organisations also have an increasing need to act responsibly, both towards the environment and to the well-being of their staff. These requirements fit neatly with the lower capital costs and energy consumption of well controlled lighting.

The research team looked at 12 buildings where occupants have control over their electric lighting. The case studies generally involved dimmable luminaires arranged in control groups of one to six people who were able to set light output to suit their preferences. The control devices were either hand held remote control units, wall-mounted potentiometers or rocker switches.

The primary light sources in the case studies were usually mirrored downlighters (sometimes supplemented by decorative uplighters). Nine buildings were in urban locations, and all buildings had blinds – automatic and manual – to control glare.

Research findings

Trade literature commonly cites flexibility as a major advantage of the systems under study, largely due to the possibility of unfettered switching arrangements and floor layouts.

Generally speaking, the most flexible systems are those using infrared control devices. They need few drop switches, and the whole lighting installation can be installed on one horizontal layer within the ceiling void.

However, observations revealed that occupants make little attempt to derive real benefits from the lighting controls. The reasons varied from unfriendly lighting control software to wiring layouts that fixed control groups in permanent positions.

There were also few signs of effort being made to co-ordinate workstations with luminaires. More often than not the structural grid dictated the lighting layout. This commonly led to situations where a group of luminaires was divided by two separate work teams, which were themselves divided by office furniture.

In this case, individually addressable luminaires offer the best opportunities to improve flexibility, particularly as these relieve facilities managers from much of the burden of managing and co-ordinating control groups.

In theory at least, devolving control to users will mean that electric lights will only be switched on or turned up according to need. However, experience shows that lighting intensity is related to the size of the workgroup. In large workgroups occupants tend to assess the entire space and switch the lighting in relation to the room's darkest zone. With smaller control groups, decisions become based on the local area below.

It is also the case that people will even prefer daylight to electric light, and will work by it even though established guidance says it is too dark to do so. With individually controlled lighting, those that want electric light can switch on a luminaire leaving others to work by daylight.

The changing pattern of office work also plays its part. Although modern offices are commonly occupied for over twelve hours a day, holidays, illness and peripatetic working means that desks are often only occupied for 60% of the available time. In the absence of lighting controls, such luminaires can be on wastefully for 40% of the time.

In the twelve case studies, the highest energy savings were found where the lighting default setting was at a low output, or where users had to consciously make decisions to switch on the lights. The poorer performing systems tended to be those with switches located remotely on walls, often without line of sight to desks, or where the systems defaulted to a timed on at a maximum setting.

Additionally, investigations have revealed that once controls have been used the lighting output tends to be constant, irrespective of changing daylight. This not only applies to the daily case, but also where the on setting has a user default.

These observations lead to two conclusions. First, conditions should be suitable for discouraging wasteful energy consumption before occupants enter a space. For example, there is no point in raising blinds halfway through the afternoon and expecting occupants to turn their lights down.

Second, people should have the opportunity at the start of the day to assess the daylight at their workstations before choosing the level of electric light they desire.

Relating comfort to control

Office lighting is often designed to give a horizontal desk illuminance of 300-500 lux – recommendations based as much on technology, economics and politics as on visual requirements.

Current vision theory suggests visual performance may be modelled by a plateau and escarpment approach. In other words increasing illuminance improves task performance to a given point, but beyond this task performance remains static.

Visual wants differ from visual needs: while a preferred illuminance can be set, individual preferences will vary widely. Figure 2 shows the variance in desktop (electric lighting) illuminances for the 12 case studies. Given the mean of 292 lux and a standard deviation of 199 lux, a set illuminance is an unlikely basis for a successful lighting installation.

Further research has attempted to map the nature of lighting control on illuminance levels. It shows that people tend to be more positive about their perceived comfort and productivity when controls are easy to use.

Many controls are regarded as user-unfriendly: infrared controls were often marked down when they gave no indication of change occurring. As people tended to be more positive depending on the familiarity of these controls, there is clearly a case for people being taught how to use them. Some people avoid using lighting controls through fear of upsetting others, a finding that once again highlights the value of individually addressable luminaries.

So what does all this mean for lighting design?

First, local control can yield significant energy savings at the same time as providing people with the desired level of electric lighting independent of task and daylight conditions.

Second, improvements need to be made on both these fronts. Luminaires should be made individually addressable and the controls should be improved for ease of use. More thought needs to be given as to what constitutes a minimum level of illuminance, particularly in relation to health and safety requirements. Presence detection is also deserving of closer attention.