Emergency Light Inspection Checklist
Two tests at two frequencies, logged fitting by fitting. The monthly test proves the lamp lights. The annual test proves the battery lasts, and that is the one that finds failures.
- Monthly function test and annual duration test, tracked separately
- Twelve checks on every fitting, not one tick per building
- The duration each fitting actually held, not the duration required
- Coverage recorded alongside pass rate, so an untested stair shows up
Emergency Light Inspection Checklist
Two regimes, every luminaire logged
| # | Check on every fitting | Pass | Fail | N/A |
|---|---|---|---|---|
| 1 | Charge or healthy indicator lit under normal supply | |||
| 2 | Lamp illuminates within five seconds of supply loss | |||
| 3 | Correct light output, not dim, flickering or partly lit |
The document you will get. Download for the full, editable file.
Who this emergency light inspection checklist is for
Four people rely on this record, and the one it protects has usually never seen it.
The technician or testing contractor
You are the one on the steps with the key switch. The per-fitting checks and the reference label are what make the job repeatable, because a fitting you cannot name again is a fitting nobody can chase.
The responsible person or duty holder
Your name is against the logbook. What you need from this is evidence that the test happened, that failures were dated, and that the fittings tested are the fittings the building actually has.
Facilities or building manager
You have to schedule a three-hour discharge without leaving the building uncovered. The pre-test section and the recharge window are the two parts that decide when the test can safely run.
Fire safety or compliance lead
You are reading this a year later, or handing it to an inspector. Coverage, outstanding failures and the logbook entry are the three lines that tell you whether the system is genuinely proved.
Which section matters the most in your sector
The fittings and the tests are the same everywhere. What changes is when you are allowed to discharge the system and how much of it you can reach, and that is usually decided by who is in the building. If you run one of these, the sector page goes further than the template does.
- FM service providers
The luminaire test log. When testing is subcontracted, a log with a reference and a held duration per fitting is the only evidence that the visit covered the building rather than the ground floor.
FM service provider software - Healthcare
The recharge window. A ward cannot be left without compliant emergency lighting, so the duration test has to be planned around occupancy rather than fitted into a maintenance slot.
Healthcare maintenance software - Retail and malls
Test scheduling against trading hours. A three-hour discharge that starts too late in the day leaves the recharge running into opening, which is the window nobody accounts for.
Retail maintenance software - Commercial real estate
Coverage across the portfolio. The risk is not the failed fitting, it is the building where the same accessible fittings are tested every month and the risers never are.
Portfolio maintenance software - Education
Escape route walking. Campus buildings change use between terms, and a fitting that was correct for a corridor is not automatically correct once the direction of travel changes.
Campus maintenance software - Data centres
The pre-test checks. Discharging life safety lighting in a data hall is a planned event with its own notification list, not something to run on a routine visit.
Data centre maintenance software
What an emergency light inspection checklist should contain
An emergency light inspection checklist is a record of two tests carried out on every emergency luminaire and exit sign in a building. It captures the system and its rated duration, the checks made on each fitting, the duration each one held under discharge, the failures found, and the logbook entry that closes the test.
A. Fields specific to an emergency light inspection
| Field | What goes in it | Why it earns its place |
|---|---|---|
| System type | Self-contained luminaires, central battery or hybrid | A central battery system fails differently from self-contained fittings, and it adds a daily visual check that a self-contained system does not have. |
| Required duration | One hour or three hours, taken from the design | This is the number the annual test is judged against. Taking it from habit rather than the commissioning documents is how a three-hour system gets signed off on a one-hour test. |
| Design or commissioning standard | NFPA 101 and NFPA 111, BS 5266, or AS 2293 | The standard sets the duration, the test frequency and what the logbook has to show. Naming it makes the record readable by somebody who did not design the system. |
| Testing method | Manual key switch, test facility, automatic self-test or addressable | An addressable system reports per fitting and a key switch does not, which changes how much of the log you have to fill in by hand and how much you can trust. |
| Test regime and next due dates | Both regimes, with the last done and next due for each | Two tests on two clocks is the thing that goes wrong. Holding them in one table with a countdown is what stops the annual test quietly becoming a fourteen-month test. |
| Pre-test checks | Eight lines, including the recharge window and who has been told | The system is not compliant while it recharges. These checks exist so the test is not scheduled the day before the one event the lighting would have mattered for. |
| Per-fitting checks | Twelve checks per luminaire, each pass, fail or not applicable | A building-level tick proves nothing. The five-second start, the light output and the exit sign legend are separate failures with separate fixes, so they are separate lines. |
| Luminaire test log | One row per fitting, with the duration it held | This is the heart of the record. Held for is the column that turns a pass rate into an engineering finding, because it shows a battery declining before it fails outright. |
| Compliance block | Pass rate, plus coverage of the system and outstanding failures | Pass rate on its own flatters. Coverage says what share of the building was actually tested, and outstanding failures says whether last time was ever closed out. |
| Failures and remedial actions | Fault found, action, owner and a date | A failed fitting on an escape route is a fire safety defect rather than a maintenance backlog item, so the date against it belongs in days rather than in a quarter. |
| Duration test record | Times off and on, duration achieved, and recharge complete | The recharge completion time is the field people leave blank. Between the end of the discharge and the end of the recharge the building is not covered, and somebody should know that window. |
| Sign-off and logbook | Tester, competence, logbook entry and next due dates | A test with no logbook entry is a test that did not happen, as far as an inspection is concerned. This is the line that turns work done into evidence held. |
If you cut this down, keep the Held for column and the coverage line. Everything else on the form records whether a fitting worked on the day. Those two record whether the system will work next time. Held for shows a battery losing capacity while it can still pass a monthly test, which is the only early warning the system gives you. Coverage stops the record flattering itself, because a hundred percent pass rate across the fittings somebody could reach on a stepladder says nothing about the riser nobody opened.
B. What it looks like filled in
Six rows from one annual duration test on a three-hour system, taken from the luminaire log. The building passed, and the log still found two things worth acting on.
| Ref | Location | Lights | Held for | Result |
|---|---|---|---|---|
| EL-014 | Level 2 stair core B | Yes | 3h 00m | Pass |
| EL-019 | Level 2 stair core B | Yes | 3h 00m | Pass |
| EL-022 | Basement corridor | Yes | 1h 40m | Fail |
| EL-023 | Basement corridor | Yes | 2h 05m | Pass |
| EX-07 | Rear exit, sign above door | No | n/a | Fail |
| EL-031 | Plant room stair | Not tested | n/a | Not tested |
Three of these rows matter more than the pass rate. EL-022 lit correctly on every monthly test for a year and then held one hour forty of a required three hours, which is exactly the failure a function test cannot see. EL-023 passed, and it is the more useful row: two hours five on a three-hour system is a battery on the way out, and it will be next year's failure unless the held duration was written down. EL-031 was not tested at all, because the plant room stair needed access nobody had booked. Had that row been left off the log rather than recorded as not tested, this record would read as a clean sweep of the building instead of a test of most of it.
Word to adapt the per-fitting checks to your system and standard, Excel for the luminaire log and the rolling test record a logbook inspection asks for, PDF for the clipboard copy that goes round the escape routes. Free, and yours to rebrand.
How do you carry out an emergency light inspection?
Fill in the system details once, then work the form in order on every visit. The two regimes are recorded in the same document rather than in two places. Six steps.
Record the system, its rated duration and the standard
System type, number of luminaires and exit signs, the required duration taken from the commissioning documents rather than from habit, the design standard, the testing method and where the logbook is kept. This section is filled in once and then checked rather than rewritten.
Set both test regimes and their next due dates
The monthly function test and the annual full-duration test, each with what it proves, the date last done and the date next due. A central battery system adds a daily visual check. Two clocks in one table is what stops the annual test slipping.
Clear the pre-test checks, especially the recharge window
Key switch or test facility located, occupants told the lighting will be interrupted, spares and access equipment available, and the previous test reviewed so you know what was already outstanding. Then confirm no event, drill or high-occupancy period falls inside the recharge window.
Walk the escape routes and check every fitting
Twelve checks per luminaire: the charge indicator, a start within five seconds, light output, the exit sign legend and whether the sign type matches the direction of travel, diffuser condition, mounting, obstruction, overheating, battery condition, the test facility and the reference label.
Run the duration test and log what each fitting held
Disconnect the supply, record the time, and log the duration each fitting actually held rather than the duration required. Then record the recharge start, the time the system is available again, and whether it was retested after recharging.
Close out failures, then enter the test in the logbook
Each failure gets a fault, an action, an owner and a date measured in days, because a dead fitting on an escape route is a fire safety defect. Then the compliance block, the coverage figure, the sign-off, the logbook entry and both next due dates.
The monthly function test versus the annual duration test
These are treated as the same job done more or less often, and they are not. They prove different things, and only one of them finds a failing battery.
| Aspect | Monthly function test | Annual duration test |
|---|---|---|
| What it proves | Every lamp illuminates when the supply is lost | The battery still holds its full rated duration |
| How long it runs | A short discharge, a few minutes at most | The full rated time, one hour or three hours |
| What it finds | Dead lamps, failed chargers, missing or blocked fittings | Batteries that have lost capacity but still light |
| Recharge afterwards | Minutes, with no practical loss of cover | Typically 24 hours, and the system is not compliant until it finishes |
| Running it more often | Sensible, and the point of a monthly regime | Wears the batteries out and leaves the building repeatedly uncovered |
| If it is the only one you do | A battery can decline to nothing while still passing every month | A lamp that failed last week goes unnoticed for a year |
The monthly test is the one you do; the annual test is the one that tells you something. A battery near the end of its life will pass a function test comfortably, because lighting for a few minutes asks almost nothing of it. It will then fail at ninety minutes on a three-hour system, and the only way to know that before the night it matters is to discharge it fully once a year and write down what it held. The trap runs the other way too. Because a duration test is more thorough, it is tempting to run it more often, and a full discharge every month both wears the batteries out early and leaves the building without compliant lighting during each recharge. Run the short test monthly, the full test annually, and never put the full test in the day before something happens in the building.
When the template starts to feel limiting
A form handles one visit well. It handles a live building, and a portfolio of them, badly, and always in the same four ways.
Two clocks, and neither one reminds you
The monthly test and the annual test fall due independently, per building. On paper both are dates in a document that is filed after the visit, so the test that slips is the annual one, and it slips by months before anybody notices.
You cannot see what is still outstanding
Which fittings failed last time, which were never reached, and which remedial actions are still open. That is the question an inspection asks first, and answering it means reading every previous record rather than looking at one.
The fitting is not an asset anywhere
EL-022 has a reference on a form and no existence in any system. Its battery history, the two failures it has had and the replacement fitted last year live in three separate documents that only a person can join up.
Coverage cannot be proved from a stack of forms
Proving you tested the whole system means comparing the fittings on the log against the fittings the building has. Without a list of fittings to compare against, the pass rate is measured against whatever was tested that day.
What running this in Facilio looks like
The template is the paper version of this inspection. The checks are the same ones; the difference is that a fitting becomes an asset with a history, and a due date becomes something the system raises rather than something a document remembers.
Work Completion Validator
A test cannot close with fittings unaccounted for
On paper the visit is finished when the form is signed. Work Completion Validator holds the test against the fittings on the system, so a luminaire that was never reached is flagged as not tested rather than quietly missing from the log.
Ops Performance Intelligence
Both regimes are scheduled rather than remembered
The monthly function test and the annual duration test run on their own clocks per building, so the annual test surfaces before it is due instead of after somebody notices the last one was fourteen months ago.
Audit Report Intelligence
Held duration becomes a trend, not a tick
Because the duration each fitting held is recorded against the fitting, a battery falling from three hours to two to ninety minutes reads as a decline, which is a replacement you can plan rather than a failure you discover.
Contractor Work Tracker
The testing contractor's coverage is visible
Which fittings the visit actually covered, with references, sits against the contractor record, so a building tested at the ground floor every month shows up as partial coverage rather than as a full pass.
Hallucination-free by design. Atom AI answers from the records in your tenant rather than generating plausible text, so an empty field reads as empty rather than filled in for you.
Frequently asked questions
What is an emergency light inspection checklist?
An emergency light inspection checklist is the record of testing every emergency luminaire and illuminated exit sign in a building. It captures the system type and its rated duration, the checks made on each individual fitting, the duration each one held when the supply was disconnected, any failures found with an owner and a date, and the entry made in the system logbook.
It covers two separate tests: a short monthly function test that proves each lamp lights, and an annual full-duration test that proves the battery still holds its rated time. Both are recorded in the same document so the two due dates cannot drift apart.
How often should emergency lighting be tested?
A short function test every month, and a full-duration test once a year. Central battery systems add a daily visual check of the panel indicators. Take the exact frequencies and the required duration from the standard your system was designed and commissioned to, because that is what an inspection will measure you against.
The monthly test is a brief discharge of a few minutes. The annual test runs for the full rated duration, one hour or three hours depending on the design, and the system needs to recharge fully afterwards before it is compliant again.
What is the difference between a monthly function test and an annual duration test?
The monthly function test proves that every lamp illuminates when the normal supply is lost. It asks almost nothing of the battery, so a battery close to the end of its life will pass it comfortably. The annual duration test discharges the system for its full rated time and proves the battery can still hold it.
That is why the annual test is the one that finds real failures, and why running the duration test monthly is a mistake rather than extra diligence. A full discharge every month wears the batteries out early and leaves the building without compliant emergency lighting during each recharge period.
How long do emergency lights have to stay on?
One hour or three hours, depending on how the system was designed. The figure is set by the design and the commissioning documents rather than by the type of building alone, so it needs to be read off the paperwork rather than assumed.
This matters because it is the number the annual test is judged against. A three-hour system signed off on a one-hour test has not been tested, and the fittings that would have failed between one hour and three hours are exactly the ones that would matter during a real evacuation.
What should I do if an emergency light fails the test?
Record what actually happened rather than only that it failed. A fitting that held one hour forty of a required three hours is a different fault from one that never illuminated, and the held duration is what tells you whether the battery, the lamp or the charger is the problem.
Then treat it as a fire safety defect rather than a maintenance backlog item. Give it a fault description, an action, a named owner and a date measured in days, and carry it forward as an outstanding failure until it is verified fixed. A failed luminaire on an escape route is a gap in the building's means of escape.
Do I have to test every emergency light?
Yes, and proving that you did is the part most records fail. Every luminaire and every illuminated exit sign needs to be checked individually and logged with a reference, which is why each fitting carries a label so it can be found again next time.
This is what the coverage figure is for. Testing the same forty accessible fittings every month and never opening the plant room stair produces a perfect pass rate on a system that has not been tested, so the record shows the share of the system covered alongside the share that passed.
When should I not run a duration test?
Not in the twenty-four hours before an event, an evacuation drill, or any period when the building will be unusually busy. After a full discharge the batteries need to recharge, typically for around a day, and the system does not provide compliant emergency lighting until that has finished.
The pre-test section of this checklist exists for exactly that decision. It asks whether the recharge period is known and whether anything is happening inside it, so the test is planned around the building rather than slotted into whichever visit was next.
Can I edit and rebrand this template?
Yes. It is free to use, edit, rename and put your own logo on, internally or for clients. No attribution required.
The Word version is the one to edit if you want to change the per-fitting checks or match the wording of your local standard. The Excel version is the one to use if you want the luminaire log and the rolling Test Log tab, which keeps one row per test event with fittings tested, passed, failed, duration achieved and the certificate reference.
In one paragraph
An emergency light inspection is two tests on two clocks, and confusing them is the usual failure. Run a short function test every month to prove each lamp lights, and one full-duration test a year to prove the batteries still hold their rated time, because a battery near the end of its life passes the monthly test easily and fails at ninety minutes on a three-hour system. Check every fitting individually against all twelve lines, log it against a reference so it can be found again, and write down the duration it actually held rather than the duration it needed. Watch coverage as well as pass rate, since a clean sweep of the fittings somebody could reach says nothing about the riser nobody opened. Give failures a date in days rather than a quarter, and make the logbook entry, because an untested fitting and an unrecorded test look identical a year later.
The template is the floor, not the ceiling
Take the checklist; it will run one inspection properly and give a logbook something to hold. When a folder of forms stops being enough, a connected CMMS runs the regime: every fitting an asset with its own test history, both test frequencies scheduled per building, held duration trended so a declining battery is a planned replacement, and coverage measured against the fittings that actually exist.
