What Is a Fire Watch? A Plain-English Guide for Property Owners

A fire watch stands in for a system that is not working

Every occupied building is permitted on the assumption that its fire protection works: the sprinklers will knock down a fire before it spreads, the alarm will tell people to leave, the standpipe will give firefighters water on the tenth floor. A fire watch exists for the hours or days when one of those pieces is missing, either because a required system has failed or been shut off, or because the work happening in the building is itself capable of starting a fire the way welding, cutting and torch-applied roofing are. In both cases the fire code’s answer is the same: put a trained, dedicated person in the space with no job but to notice fire early, raise the alarm and call for help.

That is the whole idea. It is not a form of security patrol wearing a different name, though a Price Protection officer trained for both can run a security post and a fire watch on the same shift where the post orders allow it. A fire watch has a narrower purpose and a schedule set from outside the building, by a code section, a fire marshal or a hot-work permit, rather than by the property manager’s own judgment about risk.

The two situations that actually require one

Almost every fire watch a Texas property owner will ever book falls into one of two categories, and they come from different rulebooks.

A required system is out of service. Section 901.7 of the International Fire Code, which Houston and most Texas jurisdictions adopt with local amendments, requires that when sprinklers, a fire alarm, a standpipe or a fire pump go out of service, the fire department and the fire code official are notified immediately, an impairment coordinator is named, and, where the official requires it, the building is either evacuated or kept occupied under an approved fire watch until the system is back and tested. The IFC 901.7 article works through that section clause by clause.

Hot work is happening. Welding, cutting, brazing and torch-applied roofing throw sparks and slag that can travel and smolder in insulation or debris long after the torch is off. OSHA 29 CFR 1910.252(a)(2)(iii) requires a fire watch wherever combustibles sit within 35 feet of the work, sit farther away but could still catch a spark, or are exposed through an opening in a wall or floor, and the watch continues at least 30 minutes after the work stops. NFPA 51B, which most hot-work permits reference, asks for at least an hour. The OSHA hot-work article covers that standard in full.

A third, less common trigger is a contract or insurance condition: an owner’s representative, a general contractor or an insurer’s impairment permit can require a watch even when the code does not strictly demand one for a particular duration, and that requirement is treated the same way once it exists.

Trigger	Governing rule	What starts the clock	What ends it
A required system out of service	IFC 901.7, as adopted locally	The system goes down and is reported	Restored and tested
Hot work	OSHA 29 CFR 1910.252(a)(2)(iii), NFPA 51B	The torch is lit	Work stops, plus the post-work period
Contract or insurance condition	The permit, policy or site rule	Whatever the document sets	Whatever the document sets

What actually happens during the watch

Once a watch starts, the job is narrower than it sounds. For an impairment, the officer walks the areas the system would have protected, on an interval set by the fire code official or, absent a specific number, by the coordinator from the building’s fire plan, looking and smelling for smoke and heat, checking that exits are clear and extinguishers are where they belong, and logging each round with the time. For hot work, the officer does not patrol at all; the job is to stay where the sparks land, watch the floor below an opening or the far side of a wall being cut, keep an extinguisher within reach, and remain through the post-work period once the tools are down.

If a fire is found, the sequence is the same in either case: raise the alarm by whatever method the post orders set for a building whose system may be down, call 911, warn occupants, and use an extinguisher only on a fire small enough to put out safely. Everything gets written down, with times, because to an inspector or an insurer a watch that was not logged did not happen.

Who can perform the watch

The fire code official decides who counts as an approved fire watch for a given building, and the answer is not fixed. Some buildings cover a short daytime valve closure with their own maintenance staff, who know the layout and can call 911 as well as anyone. Longer watches, overnight coverage, emergencies that start at midnight and hot-work watches on construction and renovation sites more often go to a contracted security company whose officers are trained specifically for this and have no other duties competing for their attention during the watch. Price Protection staffs permanent, temporary and emergency fire watches across Greater Houston, and the who can perform a fire watch article goes further into where that line usually falls.

Getting a watch started

Dispatch needs a short list of facts to open a fire watch: the address and the type of building, which system is down or what hot work is planned, the areas affected, whether the fire marshal has already given a direction, the round interval if one has been set, and the name of the coordinator or permit holder. With that, officers can be assigned and post orders drafted while they are en route, and the log opens the moment they arrive on site. For a planned shutdown there is time to schedule the watch around the contractor’s window; for an emergency, the same call opens it the same day.

This is a summary, not legal advice. A property’s own fire code official has the final say on what a specific building must do, and the articles linked throughout this guide go deeper into each piece: the impairment side, the hot-work side, the coordinator’s role and the log that ties it all together.