The metal label is still riveted to the curtain: a rating, a laboratory stamp, a promise made on a test rig somewhere far from this building. Nothing on the label says whether the door will actually fall when the alarm sounds tonight. That is the gap the fire door drop test exists to close. Once a year, at the opening itself, the curtain is commanded to fall, and the promise is renewed in front of witnesses. Skip the renewal and the building keeps the paperwork while quietly losing the protection it describes.
We build fire-rated roller shutter doors in our specialty doors family, and every one of them ships with a factory inspection report and its certification documents. What no manufacturer can ship is next year’s proof. That proof is produced at the opening, by the procedure described below.
What a Drop Test Actually Verifies
Start with what the event is. A drop test is the periodic test of automatic closing. It applies to automatic-closing fire doors: rolling service and counter doors, smoke and fire curtains, and swinging doors with powered closers. One manufacturer’s compliance guide puts it plainly: the test exists to verify that the assembly will close by itself when it is needed.
The fire door drop test is therefore not a formality attached to the label. It is the annual demonstration that the label still describes this door. EI ratings classify a fire-rated roller shutter the same way, before it is ever hung.
The demonstration has three results, and the DASMA technical data sheet for rolling steel fire doors names them: proper operation, full closure, and proper reset. Its form records each as pass or fail under three headings: Visual, Operational, and Drop Test. The operational check comes before any falling. The door is cycled through its normal opening and closing travel, so imbalance or rough motion shows up while nothing is being asked to drop.
The same document sets the frequency floor. The door must be inspected and drop-tested not less than annually by a trained technician, and the requirement applies even to doors that are never used and remain closed.
Audit day is where this becomes concrete. An inspector rarely asks what the label says; the folder with last year’s form is the first thing opened. Frequency also runs on two tracks, which audits often conflate. Acceptance testing happens immediately after installation, and one manufacturer’s guidance says its record is kept for the life of the assembly. Annual periodic testing follows for the door’s service life, with records kept for at least three years.
Our reading of the compliance chain is simple. The label is the laboratory’s verdict on the model. The annual record is the verdict on this particular door, and the authority having jurisdiction only ever sees the second one. The ratings read the same in a warehouse as in a hotel corridor. What 30-, 60- and 90-minute labels describe is laboratory performance.
The field record must keep re-proving it.
So the test verifies, at minimum:
- That the release chain triggers the curtain without manual help.
- That the curtain descends under control and fully closes the opening.
- That the door can be returned to service and will repeat the performance.
- That the assembly is free of the damage and obstructions that quietly invalidate all of the above.
Prepare the Opening Before You Release It
Who is allowed to run the procedure is part of the standard, not a preference. NFPA 80 wording, as quoted by the same manufacturer’s guide, requires a trained rolling steel fire door systems technician, someone with documented training and experience on this door type. The DASMA drop test form also requires a witness, and the association’s technical director notes that person can be an authorized representative of the customer or the authority having jurisdiction. One person may lawfully test the door; the form still wants a second signature.
Preparation then follows a fixed order, because a drop test is the one moment the curtain is commanded to fall on schedule:
- Run the visual inspection first and correct every deficiency before anything moves.
- Clear the doorway of anything that could impede the curtain or get crushed beneath it.
- Block access to the area around both sides of the door with caution tape.
- Check the release chain’s cleanliness: fusible links and cables must carry no paint, dust or grease, and no drop arm may be tied, blocked or wedged.
- Remove anything attached to the assembly that is not part of it, from conduit clips to signage.
The order is the point. Painting crews and renovation dust have a way of reaching fusible links, and a coated link is a trigger that will not melt on cue. A drop arm lashed back out of convenience behaves the same way: the mechanism looks intact, photographs well, and will not release.
These are the failures that make preparation more than tidiness. None of them announce themselves while the door sits open and quiet. Every one of them is caught in minutes by the visual pass the standard insists comes first. A deficiency corrected on the ground costs nothing. A deficiency discovered mid-descent costs the opening.
Deficiencies are corrected before the operational test for the same reason: you want the descent judged on a healthy door, not a sabotaged one.
Safety line: a commanded curtain is a falling machine. Nobody works or stands under it, and the tape stays up until the door has been reset and secured.
Judge the Descent: Speed, Closure and the Sill
The release starts from the fully open position, and it is not a single event. Each automatic closing option on the door is exercised in turn. Every detector and release device in the chain must prove itself: the alarm interface, the smoke detectors, the fusible links, the local test switch. A door that drops for one trigger and not for another has failed the chain even though it passed one test.
Descent is measured, not admired. The standard sets the average closing speed in a band. The curtain must fall at no more than 24 inches per second and no less than 6. The same figures give 152 to 610 millimetres per second. The readings in between are where judgment lives, and the four observations below are what a fire door drop test record should capture:
| Observation | Pass reads as | Fault reads as |
|---|---|---|
| Closing speed | Between 6 and 24 in/s on average | Faster than 24 is free fall; slower than 6 risks a stall |
| Landing | Curtain rests on the sill or floor | Stops short of the floor; the opening is not sealed |
| Track | Stays inside both guides | Jumps or bows out of the guides |
| Travel character | Smooth, no abnormal noise | Juddering, scraping, banging on the way down |
Free fall is its own verdict. A UK shutter guide calls sudden free-falling movement a serious fault even when the door lands perfectly, because the brake or governor that should meter the descent has already quit.
The reading is an average on purpose, and that detail disciplines the whole test. One clean run can hide a curtain that hesitates halfway; averaging across the commanded descents turns a lucky drop into a measured behaviour. Our field rule is to log the reading for each release path separately. A door that falls fast on the alarm interface and slow on the local switch cannot hide behind a generous mean.
The landing closes the argument. The door must close completely and rest on the sill or floor, without a helping hand from anyone. A curtain that hovers an inch above the concrete has not passed; gravity got it most of the way and the opening stayed open.
The Second Drop Is a Verdict on the Reset
The first descent tests gravity, curtain and guides. The second tests whether the door can do it again, which is the only question a fire actually asks. One manufacturer’s guide cites the requirement directly. The door must be dropped and reset entirely at least twice, per NFPA 80 section 5.2.3.7.3.1. The repetition exists to prove the automatic closing device resets correctly, not that the curtain fell nicely once.
DASMA’s own procedure says the same in plain steps. Drop once. Reset the door per the manufacturer’s instructions and open it by normal operation. Then drop it a second time.
Reset is also where the release chain gets its audit. Each trigger in turn, the alarm interface, the detectors, the fusible links and the test switch, must re-arm and command the door again. The weekly and monthly routine in a fire shutter inspection plan keeps those triggers exercised between annual tests. That is exactly why a seized release tends to surface in a drill rather than on test day.
A door that drops but does not reset has protected the opening exactly once. In fire protection, once is zero.
The second drop is also a useful trap for things the test itself displaced. Run it and you may hear the rattle a spanner left on the bottom rail makes on the way down. After the final drop, the door is reset per the manufacturer’s instructions and given one last visual pass, because a procedure this physical can leave its own evidence behind.
When a Drop Test Fails: Faults and Fixes
A fail is a verdict, not a conversation. The technician who ran the test specifies whether repair is enough or whether the door must be replaced. Either way, the door must pass a retest before it returns to service. Most failures trace to a short list of causes. Mapping the symptom to the cause turns a failed fire door drop test into a work order instead of a mystery:
| Symptom | Likely cause | Disposition |
|---|---|---|
| Stalls above the floor | Debris under the curtain; obstructions; friction in damaged guides | Clear and repair, then retest |
| Free-falling descent | Worn brake; failed governor or closing mechanism | Component replacement, then retest |
| No release on trigger | Broken fusible link; disconnected alarm interface; tied or wedged drop arm | Restore the release chain, then retest every trigger |
| Curtain leaves the guides | Bent guides; damaged slats or endlocks; struck curtain | Guide and curtain repair, then retest |
The handling protocol around that table is as codified as the test itself. Tell the responsible person, record the failure, and apply an immediate control measure: restrict access to the opening or arrange temporary fire separation. Arrange repair with parts from the original manufacturer, which the DASMA data sheet makes explicit, then repeat the test until the door passes. Components can fail during testing through no fault of the tester, which is one more reason the record and the retest belong to the same event.
When the verdict is replacement rather than repair, start where the door started. Our specialty doors line builds fire-rated curtains and shutters to the rating the file already names.
Our rule is stricter than habit: until the retest passes, the opening is managed as unprotected. Work practices change, traffic is rerouted, and nobody treats the label on a failed curtain as permission. The interim measure goes in the record too, because the auditor who reads about a failure will next ask what the building did about it that afternoon.
Records That Keep the Certification Alive
The paper trail is not an afterthought to the fire door drop test. It is the part the outside world can actually inspect. The DASMA form asks for the door’s identification and location, the date, the technician, and the three pass or fail results for Visual, Operational and Drop Test. It also carries the witness signature the association requires. A sound record adds the defect found, the corrective action, the retest result and the next due date.
Filled in at the opening, on the day, in ink or in the maintenance system, the form is short. Reconstructed later for an auditor, it is worthless:
- Door identification, location and rated label details.
- Test date and the name of the trained technician.
- Visual, Operational and Drop Test results, each pass or fail.
- Defects found and corrective actions taken.
- Retest result and the date of the next test.
- Witness signature and, where applicable, the AHJ reference.
Retention runs on the same two tracks as the testing. The acceptance record stays with the assembly for its life. Annual records are kept for at least three years on one manufacturer’s guidance. The authority having jurisdiction gets the final word on local requirements.
The tidier habit is a file per door: model, location, rating, last inspection date. An auditor’s question is then answered by opening a folder instead of a search. If the opening needs a replacement curtain or a new door altogether, the same file is where the fire-rated roller shutter door specification and its documents should land.
Keep the manufacturer’s documents and the field records together. The factory papers describe what the door was built to do; the drop test records describe what it did this year.
That pairing is how the chain closes. We supply the factory inspection report, the certification documents and the drawings with every door; the facility supplies the annual record that proves the assembly still performs. Certification is not a document you frame. It is a procedure you keep running, one commanded descent at a time.
Frequently Asked Questions
How often does a fire door need a drop test?
At least annually, by a trained technician, even if the door is never operated and stays closed. Acceptance testing after installation is a separate track with its own record.
Who can perform a fire door drop test?
A trained rolling steel fire door systems technician with documented training on this door type. General maintenance staff qualify only with that documented training, and the form needs a witness.
How fast should the curtain close?
The average closing speed must stay between 6 and 24 inches per second, or 152 to 610 millimetres per second. Faster reads as free fall; slower risks stalling above the floor.
Why does the test drop the door twice?
The first drop proves the curtain can close. The second proves the automatic closing device reset correctly, so the door can protect the opening again after the first use.
What happens if the door fails the drop test?
Report it to the responsible person and record the failure. Restrict the opening or add temporary separation. Repair with parts from the original manufacturer, then pass a retest before returning to service.
Which records does an audit ask for?
The acceptance test record, kept for the assembly’s life. Annual drop test records, kept at least three years on one manufacturer’s guidance. A per-door file with model, location, rating and last test date.
