Emergency lighting is tested on two cycles: a short function test at monthly intervals and a full duration test annually. The reason for two is that they prove different things. The short test proves the lamp and the changeover work. The long test proves the battery still holds enough charge to last the rated duration, which is the thing that degrades with age and which nothing else reveals.
The monthly function test
Simulate a supply failure to each luminaire, usually with a key operated test switch or, on an addressable system, from the panel, for long enough to confirm the lamp lights and the changeover works. Then restore the supply and confirm the charging indicator comes on. It is a short test deliberately, because every test discharges the battery and a battery that has been discharged needs time to recover.
The annual duration test
Discharge each luminaire for its full rated duration, commonly three hours, and confirm it is still providing light at the end. This is the test that finds aged batteries, and it is the reason an annual regime cannot be replaced by more frequent short tests. A luminaire that fails the duration test needs its battery or the whole unit replaced.
The recharge problem
After a full duration test every battery in the building is flat and takes many hours to recover. During that window the building has reduced emergency lighting, so the test is scheduled when the building is unoccupied or is done in sections so that a complete escape route is never unprotected at once. Doing it on a Friday afternoon at the end of a working week is the classic mistake.
The record
Per luminaire, not per building: the date, the type of test, the result, and what was done about any failure. Automatic test systems do this themselves and log it, which is why they are increasingly specified in buildings with large numbers of luminaires, where manual testing is a substantial recurring labour cost.
Questions people ask about emergency lighting testing requirements UK
What does BS 5266 say about emergency lighting testing?
BS 5266-1 and BS 5266-8 are the British Standard codes of practice that HM Government's fire safety guides point to for emergency escape lighting. The routine the guides set out from them is a monthly function test, an annual full discharge test and a record of both, with care after the full discharge because batteries typically take 24 hours to recharge and the premises should not be reoccupied until the system works again unless other arrangements are in place.
How to test emergency lights in a small premises?
Operate the test switch or key for each luminaire, confirm it lights, restore the supply and confirm it is charging. Record each one.
Can the whole building be duration tested at once?
Only if it can be left with degraded emergency lighting for the recharge period, which usually means out of hours. Otherwise test in sections.
Who can carry out the tests?
The monthly function test is normally done in house by trained staff. The annual duration test is commonly done by a contractor, particularly where remedial work follows.
What do the UK emergency lighting regulations actually say?
There is no regulation with that title either. The requirement comes from the Fire Safety Order, which obliges emergency routes and exits to be indicated by signs and, where they need illumination, provided with emergency lighting of adequate intensity in case of failure of the normal lighting. The detail is in the standards: the design standard for what to light and to what level, and the maintenance standard for the monthly function test and the annual full duration test. So the law says adequate and the standard says what adequate means, which is why a system designed to the standard and logged is the practical answer to an inspector.
How often should emergency lighting be tested?
Monthly for the short function test and once a year for the full duration test, with a daily look at any central control panel where the system has one.