Every fire technician learns the isolation lesson eventually. Usually it arrives as a pumper in the forecourt, evacuated tenants on the footpath and a building manager asking who pays the brigade bill. Sometimes it arrives the other way round, which is worse: a Monday morning call saying the panel has sat in isolation since Friday afternoon.
Both failures come from the same place. Isolation is treated as a button rather than a procedure with a beginning, a middle and an end. This guide covers what you are isolating, how to plan and record it, the ancillary shutdowns that catch people out, unwanted alarm charges around the country, restoring and proving the system before you leave, and pricing the work.
What you are actually isolating
A monitored fire alarm system has several points where a signal can be stopped, and each one stops something different. Choosing the right one is the difference between a controlled test and either a brigade call-out or a building with no protection at all.
| Isolation point | What it stops | Typical use | The risk |
|---|---|---|---|
| Device, circuit or zone at the fire panel | Alarms from that part of the system reaching the panel outputs and the monitoring provider | Routine testing of detectors and call points, contractor works in one area | The isolated area has no automatic protection until it is restored |
| Output or ancillary isolation at the panel | Specific control functions such as plant shutdown, lift recall, door release or warning system activation | Testing detection without dropping the building’s plant or releasing doors | Forgotten output isolations leave the cause and effect broken while the panel looks normal |
| Alarm signalling equipment (ASE) test mode | The alarm reaching the fire service; the monitoring provider still receives it | Proving the signal path from panel to monitoring centre during routine service | Using it as a general purpose way to avoid alarms during building work |
| ASE isolate or disable mode | The signal reaching the monitoring provider and the fire service | Exceptional maintenance that genuinely cannot be done with zones isolated at the panel | The whole building is unmonitored while it is in this state |
| Disconnection by the monitoring provider | All signals from the site | Equipment breakdown or major works, by arrangement with the provider | Usually needs formal approval, alternative arrangements and a reconnection date |
The principle that runs through every fire service’s guidance is minimum isolation. Isolate the smallest part of the system that lets you do the work, for the shortest time, and keep everything else live. Fire Rescue Victoria’s guideline on managing monitored alarm systems puts it plainly: most maintenance can be done with individual circuits, zones or detectors isolated at the panel, and isolating the ASE should be reserved for exceptional maintenance that cannot be done any other way.
Why isolation matters before testing
Routine service under AS 1851-2012 means deliberately putting devices into alarm. On a monitored system without isolation, that is an alarm of fire the brigade must attend, and the consequences stack up:
- A fire service attendance, which in most states can be charged to the building owner when it turns out to be an unwanted alarm.
- An evacuation, with the lost trading time, tenant complaints and warden fatigue that come with it.
- Ancillary functions operating for real: air handling plant shutting down, lifts recalling, smoke doors and fire curtains dropping, stair pressurisation starting, access control doors releasing.
- Complacency among occupants. Every unwanted alarm teaches people that the alarm probably means nothing.
Isolation is part of the test method, and whatever you isolate, you own until it is restored and proven.
What unwanted alarms cost
Every state fire service attends monitored alarms, and most recover a charge from the building owner when the alarm turns out to be unwanted or false. The schemes differ in how the charge is calculated, what counts as chargeable, and what leniency applies, and the amounts are reviewed regularly. The table below is a summary of how the main metropolitan services approach it at the time of writing. Check the current policy before quoting a figure to a client.
| Fire service | How charging works | Points to note |
|---|---|---|
| Fire and Rescue NSW | A fixed charge per chargeable false alarm from a monitored automatic fire alarm, applied under the alarm monitoring agreements | Leniency and waiver rules apply, including for some first alarms in a set period and causes outside the owner’s control. IPART reviewed the charging structure in 2022 |
| Fire Rescue Victoria | A time-based charge per appliance. At the time of writing FRV states $675 per truck per 15 minutes for 2026/2027 | Charged where the alarm was preventable or foreseeable. FRV bills the owner of the alarm system and looks for evidence of AS 1851 maintenance |
| Queensland Fire Department | A fixed charge per attendance at an unwanted alarm, set in the department’s annual schedule of charges under the Fire Services Act 1990 | Charged to the applicant for the alarm monitoring agreement, normally the building owner. QFD states it does not endorse on-charging attendance fees |
| Other states and territories | Most have a charging or cost recovery arrangement for unwanted alarms from monitored systems | Rates, leniency and dispute processes vary. Check the local fire service’s current policy |
Even a single fixed fee typically runs well into four figures per attendance, and Queensland notes that its charge recovers only a fraction of the real cost of attending.
Who actually pays
The fire service bills the owner of the alarm system or the party to the monitoring agreement. When a contractor caused the alarm, recovering the cost is a commercial matter between the owner and whoever caused it, and your contract terms, isolation records and notification emails are the evidence that decides whether it lands on you.
Planning the isolation before you arrive
Good isolations are planned in the office, not improvised at the panel. AS 1851 expects the responsible entity, or a person they nominate, to be told before routine service starts that it may affect the occupants or operation of the building, and to be told the extent and expected duration of any impairment so alternative safety arrangements can be made. In practice that means:
- Notify the building manager in writing with the date, the expected start and finish times, which parts of the system will be isolated, and which ancillary functions will be affected.
- Contact the monitoring provider in the way it requires for the test or isolation you are planning, and keep any reference number it gives you.
- Check the cause and effect matrix or programming record so you know exactly which outputs each zone drives. On an unfamiliar site, this is the step that prevents surprises.
- Agree alternative arrangements for longer isolations, such as warden patrols, fire watch, or signs at manual call points directing people to call triple zero.
- Plan the sequence so each zone is isolated for the shortest possible time, rather than isolating the whole building at the start of the day and restoring it at the end.
The isolation procedure on site
- Sign in, meet the site contact and confirm the notification was received and passed on to tenants and wardens.
- Check the panel is in a normal state before you touch it. Record any existing faults, isolations or alarms, so you are not later blamed for a condition you inherited.
- Contact the monitoring provider and confirm the site is in the agreed test or isolate condition before you put anything into alarm.
- Isolate the specific zones, devices or outputs needed for the next block of testing, and confirm on the panel that the isolations are indicated.
- Fit isolation tags at the panel stating what is isolated, by whom, from what time, and a contact number.
- Test, record results and reset alarms before restoring each block.
- Restore, confirm, and move on to the next block.
What to record in the isolation log
FRV’s guideline sets out a sensible standard for what each recorded isolation should include, and it works anywhere in the country. Every isolation entry should capture:
- The event type, such as a zone isolation, an output isolation or ASE test mode.
- Where it takes effect, described so a stranger could find it, for example zone 3A, first floor tenancy.
- The reason for the isolation.
- The name of the person responsible, with their employer’s contact details.
- Their qualification or role.
- The time the isolation started and the time it was restored.
That record belongs in the site logbook as well as your own system. When a charge is disputed or an incident is investigated, a timestamped log showing exactly what was isolated, when, and by whom is what protects you.
Time limits
In Victoria, FRV describes ASE test mode as deactivating automatically after 2 hours if the key is left in, and ASE isolate mode after 12 hours, while a monitoring provider may agree to a temporary disconnection of up to 24 hours on written request for emergency or maintenance purposes, with anything longer needing a building permit. Other states and equipment differ. Never rely on a timeout as your restoration procedure: panel zone isolations generally stay isolated until somebody restores them.
Ancillaries: the shutdowns people forget
Isolating the brigade signal only solves half the problem. A detector in alarm still drives every output it is programmed to drive, and those outputs are where most of the expensive surprises come from.
| Ancillary function | If you forget to isolate it | What to do |
|---|---|---|
| Air handling plant shutdown | The building loses air conditioning mid-afternoon; some plant needs a manual restart and may fault on the way back | Isolate the plant shutdown output, and test it deliberately in a planned window with the mechanical contractor aware |
| Lift recall | Lifts return to the recall floor and may stay out of service until reset | Isolate the recall output, then prove it as a separate test with the lift contractor or building manager present |
| Door hold-open release and access control | Fire doors close across busy corridors; secured doors unlock | Isolate, and tell security before you prove the release |
| Occupant warning and EWIS | An evacuation tone through an occupied building | Agree a time for audibility tests and tell occupants beforehand |
| Gas or special hazard suppression interfaces | A release signal to an extinguishing system | Follow the suppression system’s own isolation procedure before going near the detection that triggers it |
Proving the signal path, not just avoiding it
Isolation can become a habit that means the alarm actually reaching the fire service never gets tested. Routine service is meant to prove the system end to end, and that includes the brigade signal.
That is what ASE test mode, or the equivalent arrangement with the monitoring provider in your state, is for. The alarm is transmitted to the monitoring centre but not passed to the fire service. Arrange the test with the provider, put a device into alarm, and confirm with the monitoring centre that the alarm was received at the time and from the point you expected. Do the same for fault signals and, where fitted, sprinkler valve monitoring. Record the time and the person you spoke to. FRV notes that monitoring companies automatically record test mode use, so your log and theirs should agree.
Restoring and confirming de-isolation
Restoration deserves the same discipline as isolation. Work through it as a checklist:
- Reset every alarm before you de-isolate. Restoring a zone that still has a device in alarm will transmit that alarm, which is a common way to generate a call-out at the very end of an otherwise clean visit.
- Make sure test equipment is removed: smoke detector covers, heat test heads, magnets and anything taped over a sampling point.
- Restore isolations in a controlled order, and confirm on the panel that each one has cleared.
- Confirm the panel shows normal, with no isolations, alarms or faults you did not record on arrival.
- Take the ASE out of test or isolate mode and call the monitoring provider to confirm the site is back to normal and that they are receiving it as normal.
- Confirm ancillary outputs are restored, plant has restarted, lifts are back in service and doors are back on hold-open.
- Remove isolation tags, update the site logbook with restoration times, and tell the site contact in person or in writing that the system is fully restored.
If something cannot be restored, for example a zone with a faulty device waiting on parts, that is a defect, not an isolation. Raise it, classify it, tell the responsible entity before you leave, and agree the alternative arrangements. A long-term isolation that nobody owns is one of the most dangerous conditions you can find on a site.
Reading the panel event log
The panel’s event history is the most useful record on site and the most often ignored. Read it on arrival and before you leave.
- On arrival, review events since the last service. Repeated alarms from the same device, recurring faults and isolations nobody mentioned all tell you what is going on in the building between visits.
- Check for unwanted alarms and their causes. A cluster from one detector near a kitchen or loading dock is a quotable fix, not something to keep resetting.
- Before leaving, confirm your own isolations and restorations appear in the log with times that match your records.
The mistakes that cause call-outs and isolated buildings
| Mistake | What happens | How to prevent it |
|---|---|---|
| Isolating the wrong zone | The device you test is live and transmits; the zone you meant to protect stays isolated | Confirm zone numbers against the block plan and the panel text, then test one device and check the panel before working through the zone |
| De-isolating with a device still in alarm | A transmission to the brigade as you pack up | Reset first, check the panel is clear, then de-isolate |
| Leaving isolated over the weekend | A building unprotected for days, with nobody aware | End of day restoration checklist, written handover for anything left isolated, and a follow-up call |
| Isolating the ASE for building works | The whole site is unmonitored, contrary to fire service guidance | Isolate only the affected zones, for the minimum time, with a documented plan |
| Resetting the panel before the brigade arrives | Lost information, longer search time, and in Victoria a possible offence | Leave the panel for the fire service and meet them with the information |
Hot works and contractor-caused alarms
Many preventable unwanted alarms come from other trades: welding, grinding, dust, steam and spray painting. Queensland Fire Department’s guidance to building owners specifically calls for an on-site works management plan to prevent activations by workers using tools such as welders and blowers, and for detectors to be protected and isolated during work that produces dust, smoke or steam.
Fire services are equally clear about what isolation is not for. FRV’s guideline states that ASE isolations are not to be made to prevent false alarms from normal day to day activities or building work. In Victoria, resetting a panel that has detected a fire or simulated condition without the consent of FRV or a reasonable excuse is an offence under section 75B of the Fire Rescue Victoria Act 1958, and FRV notes that resetting does not avoid the charge.
Where you are the fire contractor on a site with other trades working, a workable approach looks like this:
- The building manager or principal contractor runs a hot works permit system, and the permit identifies which detectors or zones are affected.
- Only the affected zone is isolated, for the duration of the permit, and the isolation is logged with the permit reference.
- A fire watch covers the isolated area during the work and for a period afterwards, as the site’s hot works procedure requires.
- Restoration at the end of the day is your responsibility or the building manager’s, and that is agreed in writing, not assumed.
Isolating and restoring zones for other trades during fitouts is legitimate work, but price it as an attendance at each end of the day, and make the contract clear that the isolated area is the site’s responsibility in between.
Pricing and planning the work
Isolation, notification and restoration take time. Price only the device testing and you will either rush those steps or absorb the cost.
- Price the administration: notification letters, monitoring provider bookings and coordination with mechanical, lift and security contractors.
- Allow time on site for the panel check on arrival, the event log review, the isolation sequence and the restoration checklist, not just the tests.
- Price after hours or weekend attendance where the client cannot tolerate evacuations or plant shutdowns during trading hours, and agree that up front.
- Write into the contract who is responsible for unwanted alarm charges caused by your testing, by other contractors and by building occupants, and what records you will keep.
- Build in a restoration buffer at the end of every visit. A technician racing to the next job is the one who leaves a zone isolated.
Frequently asked questions
Should we isolate the ASE or just the zones we are testing?
In most cases, isolate the zones at the panel and use the ASE test arrangement with the monitoring provider to prove the brigade signal. Fire service guidance generally treats isolating the whole ASE as a last resort for maintenance that cannot be done any other way, because it leaves the entire building unmonitored.
If our technician causes a false alarm, who pays the charge?
The fire service bills the building owner or the party to the monitoring agreement. Whether the owner can recover it from you depends on your contract and what happened, so clear terms, written notifications and a timestamped isolation log are your protection.
Can we leave a zone isolated overnight if a device is faulty?
Only as a deliberate, documented decision. Treat it as a defect, tell the responsible entity before you leave, hand the isolation over in writing to a named person, agree alternative arrangements such as patrols, and set a time to return. In Victoria, long-term or regular isolations should be discussed with a registered building surveyor because they may need a building permit.
Can we reset the panel after an unwanted alarm before the brigade arrives?
Generally no. Fire services ask that the panel be left for the attending crew so they can identify the cause, and in Victoria resetting without consent or a reasonable excuse is an offence. Resetting also does not avoid the charge. Meet the crew and give them what you know.
How do we prove the brigade signal works if we always isolate it?
Use the test arrangement with the monitoring provider. In test mode the alarm reaches the monitoring centre but is not passed to the fire service, so you can confirm with the centre that the alarm and fault signals were received correctly. Record the time and the operator’s confirmation in your service record.
Sources and further reading
- Guideline No. 46: Managing monitored automatic alarm systems in accordance with fire services’ legislation, Fire Rescue Victoria (version 5)
- False alarms FAQ, Fire Rescue Victoria
- Managing your unwanted fire alarm activations, Queensland Fire Department
- Recommended charges for Fire and Rescue NSW attendance at false fire alarms, IPART information paper (March 2022)
- AS 1851-2012 Routine service of fire protection systems and equipment, Standards Australia