Smoke detector testing proves that smoke can reach a detector's sensing chamber and trigger an alarm. Pressing the test button does not prove that, because it bypasses the path smoke must travel. NFPA 72 therefore sets specific test methods for smoke, heat and specialist detectors. Those methods decide whether a fire alarm AMC visit is genuine evidence or a formality.
This guide explains the methods NFPA 72 accepts, using the code's own testing methods table. It covers aerosol and smoke testing, beam and air-sampling detectors, sensitivity checks and heat detector tests. It also explains what the test record should show for each detector tested. Owners and facility managers can use it to judge whether their contractor is testing detectors properly.
Why isn't the test button enough?
The test button is not enough because it checks electronics and signalling, not smoke entry. NFPA 72 requires smoke detectors to be tested in place to confirm smoke entry and an alarm response. This functional test is annual in buildings other than one- and two-family dwellings.
Dust, insects or paint over the vents can block smoke while the button still works. That is why the code focuses on the path from the room into the chamber. A report saying only that detectors were tested hides which method was actually used.
Quick Answer
The test button checks electronics and signalling, not whether smoke can reach the sensing chamber. NFPA 72 requires an annual in-place functional test proving smoke entry and an alarm response.
Which smoke test methods does NFPA 72 accept?
NFPA 72 accepts smoke or a listed and labelled product acceptable to the detector manufacturer, used in line with its instructions. A note in the code's test table says manufacturers customarily test specific aerosol products for acceptability. Magnets are not acceptable for smoke entry tests under the code.
Beam detectors are tested by introducing smoke, other aerosol or an optical filter into the beam path. Air-sampling systems are tested from the end sampling port where airflow lacks automatic supervision. Each detector type therefore needs its own method in the AMC.
Quick Answer
NFPA 72 accepts real smoke or a listed product acceptable to the detector maker, but not magnets, for smoke entry tests. Beam and air-sampling detectors have their own test methods.
What is smoke detector sensitivity testing?
Sensitivity testing confirms that each smoke detector is within its listed and marked sensitivity range. It is separate from the functional test, which only proves smoke entry and an alarm response. NFPA 72 sets its own schedule for sensitivity testing in section 14.4.4.3.
A detector drifted too sensitive can cause false alarms, while one drifted the other way may respond late. Some detectors use a magnet as the manufacturer's calibrated sensitivity test instrument, which NFPA 72 recognises. Where the panel reports detector sensitivity, those readings belong in the AMC record.
Quick Answer
Sensitivity testing confirms each smoke detector stays within its listed and marked range, separate from the functional smoke entry test. NFPA 72 sets its own testing schedule under section 14.4.4.3.
How are heat detectors tested?
Restorable rate-of-rise and rate-compensated spot heat detectors get annual heat tests, while fixed-temperature types rotate over five years. The code calls for a listed and labelled heat source or the manufacturer's published method. The test must not damage the non-restorable element in combination detectors.
Non-restorable line-type heat detectors must not be heat tested; they are checked mechanically and electrically, with loop resistance recorded. Non-restorable spot types are replaced after 15 years, or two per hundred are sent for laboratory testing. Owners should know which heat detector types their building uses.
Quick Answer
Restorable heat detectors get listed heat-source tests, annually or on a five-year rotation, while non-restorable line types are checked electrically. Non-restorable spot types are replaced or sampled after 15 years.
What should the detector test record show?
The record should list each detector by address or location, its type, the method and the result. Where an aerosol is used, the record should name a product acceptable to the detector manufacturer. Detectors with control outputs, such as door releases, also need those functions verified.
Our fire AMC service report guide sets out the wider format for each visit report. For placement and spacing rules, see our smoke and heat detector placement guide. Detectors that fail or show blocked vents should be listed as defects with a repair date.
Quick Answer
The record should show each detector's address, type, test method and result, the aerosol product used and any control functions verified. Failed or blocked detectors become dated defects for repair.
Frequently Asked Questions
Can I test a smoke detector with the test button?
Is canned smoke acceptable for detector testing?
Can magnets be used to test smoke detectors?
How often should smoke detectors be functionally tested?
How are beam smoke detectors tested?
How are air-sampling smoke detectors tested?
Why do some heat detectors need replacing after 15 years?
What happens if a sampled heat detector fails?
Who should test detectors in a Dubai building?
Does detector testing cause false alarms to the monitoring centre?
Main Takeaways
Smoke detector testing must prove smoke entry into the sensing chamber, so the test button alone is not enough. NFPA 72 accepts smoke or a manufacturer-accepted listed product, never magnets, for that annual functional test. Heat detectors follow separate rules for restorable and non-restorable types. Each record should show the detector, method, result, aerosol product and any control functions verified.
Your next step for detector testing
Detector testing is only meaningful when the test method proves what a real fire would actually require. For smoke detectors, that means smoke entry into the chamber, not a button press. Heat detectors, beam detectors and air-sampling systems each need their own method under NFPA 72. Sensitivity checks then confirm detectors are neither too sensitive nor too sluggish to respond.
Ask your contractor which method is used for each detector type and which aerosol product is approved. Check that reports list detectors individually, with results, methods and any control output functions verified. Our fire alarm maintenance service page explains how QSERV scopes alarm testing. QSERV can review your latest detector test records and flag any gaps before the next visit.
Sources & References
- NFPA 72: National Fire Alarm and Signaling Code, 2025 edition (Table 14.4.3.2, section 14.4.4.3)
- NFPA 72 technical committee public input responses, 2025 edition cycle, test methods table (PDF), accessed 23 October 2026