Fire Alarm Batteries
Standby Power for the System
Every fire alarm panel is required to survive a utility outage and still run a full alarm. That survival depends on sealed lead-acid batteries sized to the 24-hour + 5-minute rule. Here is how the calculation works, what load testing actually proves, and why cold weather quietly kills capacity.
Why Batteries Are Not Optional
Fire alarm systems must continue to function during a utility power failure. Fires frequently follow electrical faults, and the first thing a structure fire does is take down the main service. NFPA 72 §10.6 requires every fire alarm control unit, NAC power supply, and off-premises transmitter to have a secondary power source capable of sustaining the system automatically for a defined period. In nearly every installation, that secondary source is a bank of sealed lead-acid (SLA / AGM) batteries.
The alternative — engine-driven generators — is rare for fire alarm standby because FA loads are small (watts, not kilowatts) and batteries are simpler, cheaper, and require no fuel management. Large campuses may back up the entire panel through the building generator, but the panel still needs 4–8 hours of battery capacity to bridge the generator startup and any cold-start delays.
The 24 + 5 Rule (and Its ECS Variant)
NFPA 72 §10.6.7.2.1 sets the minimum secondary-power capacity for a fire alarm system: the batteries must sustain the system in its supervisory (normal) state for 24 hours and then drive every notification appliance in full alarm for an additional 5 minutes. This is calculated as:
Standby Ah = (total standby current in A) × 24 hours Alarm Ah = (total alarm current in A) × (5/60) hours Battery Ah = (Standby Ah + Alarm Ah) × 1.20 [20% aging factor]
The 20% safety factor accounts for battery aging, temperature derating, and measurement error. It is not optional — NFPA 72 §10.6.10.1 explicitly requires a design margin. Under-sizing batteries by skipping the safety factor is one of the most frequent plan-review rejections.
For Emergency Communication Systems (ECS / mass notification), §10.6.7.2.2 extends the alarm portion from 5 minutes to 15 minutes because voice evac and mass-notification messaging continues well beyond the horn phase of a conventional alarm.
Sizing Example — Mid-Sized Outpatient Clinic
Assume a 60,000 sq ft outpatient facility with a Notifier NFS-640 panel, 120 SLC devices, 2 NAC circuits, a DACT communicator, and a graphic annunciator:
| Component | Standby (A) | Alarm (A) |
|---|---|---|
| Panel CPU | 0.35 | 0.45 |
| SLC loop (120 × 300 µA avg) | 0.036 | 0.60 |
| Annunciator | 0.06 | 0.09 |
| Communicator | 0.10 | 0.35 |
| NAC 1 (20 horn/strobes) | 0.00 | 1.80 |
| NAC 2 (18 horn/strobes) | 0.00 | 1.62 |
| Totals | 0.546 | 4.91 |
Applying the calculation: Standby = 0.546 × 24 = 13.1 Ah; Alarm = 4.91 × (5/60) = 0.41 Ah; Total = 13.5 Ah × 1.20 = 16.2 Ah minimum. The nearest standard battery size is 18 Ah, which is typical for this scale of system.
Battery Chemistry and Form Factor
Essentially every fire alarm battery shipping in 2026 is a sealed valve-regulated lead-acid (VRLA / AGM) type. They are low-maintenance (no water to add), safe to mount on their side, and tolerate the float-charge regime inside a FACP. Common sizes: 7 Ah, 12 Ah, 18 Ah, 26 Ah, 55 Ah, 100 Ah, typically paired in series for the panel's 24 VDC bus.
Lithium iron phosphate (LiFePO4) batteries are now approved in a handful of listed FACP models but have not displaced lead-acid because the float-charge behavior and the legacy charging circuitry in most panels is optimized for SLA. Check the panel's installation manual before substituting a chemistry the panel was not listed with.
Load Testing — What It Actually Proves
A battery load test verifies that the installed batteries can still deliver their rated capacity under real alarm current. NFPA 72 Table 14.4.5 requires a semi-annual load test. Two methods are in current use:
- Discharge test. A calibrated resistive load (or the panel's own alarm mode) is applied for 5 minutes; terminal voltage is measured at the start and end. A healthy 12 V SLA battery should finish the test above 10.5 V under load. Voltage dropping below 10.5 V during the test indicates capacity loss.
- Conductance test. A handheld device (Midtronics MDX-715P, SBS-5000) injects a high-frequency AC signal, measures internal resistance, and estimates remaining capacity against a factory baseline in 10–15 seconds. NFPA 72 allows conductance testing as an approved alternative to discharge testing when the tester is listed and the manufacturer provides a baseline reference.
Conductance testing has largely replaced discharge testing in day-to-day ITM work because it is faster, non-invasive, and generates a printable report with pass/fail thresholds. Discharge testing is still used for acceptance testing and when a conductance result is borderline.
Temperature Derating
Lead-acid battery capacity is specified at 25°C (77°F). Actual capacity drops roughly 1% per °C below 25°C. A battery rated at 18 Ah at room temperature delivers only about 14 Ah at 0°C (32°F). This matters in three real-world scenarios:
- Outdoor NEMA-rated enclosures housing remote NAC power supplies — can see −20°C in northern climates; sizing must account for worst-case winter ambient.
- Unconditioned attics and mechanical penthouses — Summer temperatures accelerate aging (loss of 50% calendar life for every 10°C above 25°C), while winter temperatures reduce available capacity.
- Cold-storage warehouses — FA panels serving freezer sections sometimes share the cold environment; batteries must be installed in a conditioned enclosure or derated dramatically.
The practical rule: design to the minimum expected battery ambient. If that is 5°C, multiply your calculated Ah by 1.20 instead of 1.10 to cover both the NFPA safety factor and the cold derating.
Replacement Schedule
NFPA 72 does not set a hard replacement interval — batteries are replaced when they fail the load test or when visible condition indicates end of life. Practical guidance:
- 4–5 year target. Plan to replace SLA batteries on a 48-month or 60-month cycle regardless of test results. Calendar aging catches up with even well-maintained batteries, and the cost of scheduled replacement is trivial compared to the cost of a silent failure during a real alarm.
- Date-code every battery. Write the install date on the case in permanent marker. Photograph it with the load-test report. This creates an auditable lifecycle record.
- Swelling, leaking, terminal corrosion = immediate replacement. These are not load-test issues, they are end-of-life signs visible to anyone who opens the panel.
- Match batteries in a bank. When replacing, replace both (or all) batteries in the series string even if only one fails a test. Mixing new and old batteries stresses the new cells and shortens overall life.
ITM Summary
- Monthly visual. Confirm panel shows no battery trouble, no visible swelling or leakage, terminals tight and clean.
- Semi-annual load test. Per NFPA 72 Table 14.4.5, measured discharge or conductance test. Document results in the inspection record.
- Annual capacity calc review. When the system has been expanded or modified, recalculate the Ah requirement and verify the installed battery still meets it.
- Upon replacement. Functional test of the panel on battery alone for the full 24-hour supervisory + 5-minute alarm (or 15-minute for ECS) period as part of acceptance testing.
Frequently Asked Questions
What is the 24-hour + 5-minute rule?
How often do batteries need to be replaced?
What is a battery load test?
Do the batteries need to be in a specific orientation?
Why does cold weather matter?
References
1. NFPA 72 (2022), §10.6 — Power Supplies.
2. NFPA 72 (2022), §10.6.7.2.1 — 24 hour standby + 5 min alarm.
3. NFPA 72 (2022), §14.4.5 — Secondary power supply ITM.
4. IEEE 485 — Sizing Lead-Acid Batteries for Stationary Applications.
5. UL 924 — Emergency Lighting and Power Equipment.
Was this article helpful?
Rate this article to help us improve
Discussion (3)
The trick with battery calcs is remembering that EVERY device on the SLC draws standby current, not just the ones in alarm. I have seen designers forget to count the RMS modules, the relay modules, the graphic annunciator current draw — by the time you add it all up, the initial 18 Ah battery is undersized by 40 percent.
We document every battery with its manufactured-on date written directly on the case in Sharpie when it goes in service. When the annual load test happens, the technician photographs the date and attaches it to the report. If the date is older than 4 years and we are approaching the capacity threshold, it gets replaced that visit instead of being on a failure watch list.
Biggest finding I write: swollen batteries. A swollen AGM battery means an internal cell has failed and the battery is off-gassing. The case bulges visibly. Do not wait for the load test to catch this — it is an immediate-replacement condition and anyone opening the panel can see it at a glance.