As a service company representative, I want to share a couple of things from this side of the issue.
A heat detector is not a life safety device and they are not intended for the purpose of life safety. They are for the detection of a fire for property protection. Every one of them has clearly labelled on it “This is not a life safety device”. There are only a handful of situations where a heat detector would be deemed to be performing a life safety function. I would point out that Joseph’s quote supports this position. Additionally, it is not every detector that may not operate.
One would be as permitted by the Ontario Building Code in Sentence 184.108.40.206.(7) – Fire detectors are permitted to be installed in lieu of the smoke detectors required by Sentence (5) in Group F occupancies where the smoke detectors may be subjected to false alarms due to the activities in the building.
Another might be in an older building that still has heat detectors installed at the top of the stairwells or elevator shaft as was permitted by the building code of the day.
It also needs to be noted that the service companies cannot write the existence of these devices up as deficiencies. When we are performing an annual inspection under CAN/ULC-S536-04, there is no test or inspection criteria that the device would fail. Bear in mind that the only way to prove the fixed temp portion of a heat detector works is to perform a destructive test. Some of the detectors under the Health Canada alert also have a rate-of-rise detection capability and this is tested annually. The fixed temperature is never tested, but the electrical circuit that the activation shorts is tested annually.
I sit on the CAN/ULC-S500 working groups for S536, S537 and S524 and have submitted a proposal to ammend the standard to reflect recalled devices already. We’ll see where that goes.
15 year lot testing of fixed temp heat detectors is addressed in CAN/ULC-S536, but only in the informative index:
“G2.2 Non-restorable heat detectors may be replaced or tested on an annual lot sampling basis with the initial test following 15 years of service. The results of the initial tests and examination for deterioration will determine the frequency of subsequent tests. Sample sizes of one unit for lots of 20 or less, two units for lots of 21 to 99, and 2% for lots exceeding 99, are recommended as a minimum. Selected samples should be subjected to the Operating Temperature Test detailed in the Standard for Heat Actuated Fire Detectors for Fire Alarm Systems, CAN/ULC-S530. When failures are encountered, the lot sample size should be determined in accordance with normal quality plans.”
Because the statement exists outside the body of the standard (which it has to under Canadian Standards Council requirements and the nature of the statement), it is not required for a service company to perform or a building owner to abide by.
Rather than simply “ordering the detectors out”, I might suggest using the following from an bulletin ULC issued in 2006 (email me if you want a copy):
“The AHJ may enforce some or all portions of the Preface or any informative appendix”
Issue an order that the building owner have lot testing of the devices done (and this would include any and all fixed temps that exist and are not ever tested). For the number that exist in the building, it would likely be cheaper to replace them all then the cost of having ULC or another agent perform the testing.
Suddenly putting nearly every building in Ontario which has these devices on a fire watch for a potential failure of devices that have existed for decades seems a bit extreme for a “maybe it won’t work”. Issue the order to have them tested. It gives them 30 days to decide what to do. That seems like a more balanced approach to me. My opinion only of course.