Occupancy and Daylight Sensors: Getting Lighting Controls Right in Practice

An LED installation saves energy by producing light more efficiently. Lighting controls save energy by not producing it when nobody needs it. In most non-domestic buildings the second saving is as large as the first, and it is what Part L of the Building Regulations now expects: automatic presence or absence control and daylight-linked dimming are the default assumption for offices, corridors, stores and most other spaces, not an optional extra. The technology is mature and inexpensive. What still goes wrong is the application.

PIR or microwave: choosing the detector

Passive infrared (PIR) sensors detect the movement of a warm body across a segmented lens. They need a clear line of sight, are most sensitive to movement across their field of view and least to movement directly towards them, and can miss small movements — a person typing — after a period of stillness. They are unaffected by anything behind walls, which makes them predictable. Microwave sensors emit a low-power radio signal and detect the Doppler shift from anything moving; they are more sensitive to small movements, work through thin partitions, glass and cubicle doors, and are unaffected by temperature. That same penetration is their weakness: a microwave sensor in a corridor can be triggered by someone walking past a door, or by a tree outside a window. Dual-technology sensors require both to trigger and are the answer for spaces where false switching is intolerable.

A working rule: PIR for open-plan offices, classrooms and stores where line of sight is good; microwave for toilets, corridors, plant rooms and high-bay warehouses where the sensor has to see round obstructions or down from height; dual-tech for meeting rooms and anywhere a light switching off mid-meeting will generate a complaint. Whatever the type, check the detection pattern at the actual mounting height — a sensor designed for 3 m does not simply see further from 10 m, it sees a coarser pattern and may miss a person entirely.

Presence or absence: the setting that decides how people feel about it

A presence detector switches lights on when someone enters and off after they leave. An absence detector requires a manual switch to turn the lights on, and only automates the turning off. The energy saving is similar; the experience is not. In offices and meeting rooms, absence detection is almost always better: people switch on when they actually want light, which on a bright morning may be never, and nobody is startled by lights coming on as they walk past a doorway. Presence detection belongs in corridors, toilets, stores and anywhere people are passing through with their hands full. CIBSE and SLL guidance leans the same way.

Daylight linking: dimming, not switching

Daylight-linked control uses a photocell to reduce electric light when daylight is providing some of the required illuminance. Done well, it is invisible: the lights near the windows dim smoothly as the sun comes out and recover as it goes in, and the illuminance on the desks stays where the design put it. Done badly — with switching rather than dimming, a single sensor controlling a whole floor, or set points left at factory defaults — it is the control system people ask to have disconnected. The essentials are a dimmable installation, sensors that control the row of fittings nearest the window separately from the rows further in, and commissioning that sets the target illuminance in the actual room on a real day.

Commissioning: where most schemes are won or lost

• Hold time. Ten to fifteen minutes is a sensible starting point for offices; two to five for toilets and corridors. Too short and lights cycle; too long and the saving evaporates.
• Sensitivity. Set it for the room, not the maximum. A PIR on full sensitivity facing a warm air outlet, or a microwave sensor facing a lift lobby, will keep the lights on all night.
• Placement. PIRs away from heaters, radiators and direct sunlight; microwave sensors away from windows, doors and thin partitions with traffic behind them; all sensors positioned to see the entrance to the space, so the lights are on before the person is in the dark.
• Corridor hold. On DALI and similar systems, keep corridor lighting at a low level while any adjacent room is occupied, and only let it drop fully when the whole zone is empty. It is one setting, and it prevents the common complaint of a dark corridor outside a lit office.
• Dim, do not switch. Dropping high bays or corridor fittings to 10–20% on vacancy rather than off avoids the black-hole effect, extends driver life, and still captures most of the saving.
• Go back. Walk the building a month after handover, at different times of day, and adjust. The settings that were right at commissioning, in an empty building, are rarely right once it is full.

A sensor that annoys people gets taped over, and a taped-over sensor saves nothing. The energy saving lives or dies on the commissioning.

Lighting controls have a reputation problem that they no longer deserve. The sensors are reliable, the protocols are open, and the savings are real — commonly 30–50% on top of the LED saving in an intermittently occupied building. The discipline that makes them work is the same one that makes any building service work: the right device for the space, put in the right place, set up for the people who use it, and revisited once they have moved in.