[SML] Smoke, heat, fire sensors for HVAC systems

Dougherty, Jim jdougher at middlebury.edu
Tue Jan 26 14:43:16 UTC 2021


Anecdotal response, starting with an enormous caveat - your jurisdiction, codes and circumstances are your own; what I or others say may not apply!

We have a theater that had as its only alarm system the water-flow-bell that Dale mentioned, and even that wasn't tied to the FD, just a bell on the exterior of the building operated by the flowing water.  Grandfathered in, but the institution decided correctly that something a bit more might be warranted.  Another facility we use has those beam detectors in the (huge) lobby and not only have we had a balloon trip it, the fog from our theater made it through the air gap hallways and up to break the beam.  Not typical (it was a _lot_ of fog) but not ideal, so that was out.

For the new system, the designers were very responsive to our needs for atmospheric effects not tripping the alarms.  In addition to rate-of-rise detectors, they installed a most likely $$$ and complex system of cameras and computers to do visual analysis of the images to detect actual flames rather than smoke.  All was tied to main panels that transmitted the alarms to facilities and campus public safety.  I'm not an expert so forgive the lack of details, but the downside to these systems is that the computers rely on the images they see being stable over time in order to learn what would be a flame - the same hotel lobby, ballroom, etc.  They're not meant for constantly relocating soft goods, flashy light cues etc.  The tipping point was when the alarm tripped several times, each with a full evacuation, during one performance - a patron fanning themselves with their program and an actor putting on a kimono in the dressing room were among the causes.  We know this because we had facilities techs review the trip footage.  We ended the show with the alarm disabled and people at the panel (both facilities and public safety) to disable the alarms and check each activation (I believe the goal is to check that it's NOT fire, rather than that it IS).

Emergency planners dislike false activations almost as much as no alarms - they train occupants to ignore alarms.  A solution that relies on a person as an integral component is also not ideal - see the anecdote about the desk clerk pushing the silence button too long.  We were very fortunate to have both an AHJ and an institution working together well and footing the bills to do the rare thing of pulling a currently installed and operating system (the cameras) and ensuring the remaining system protected the facility and occupants sufficiently to meet code and need. No false trips since.

Jim Dougherty
Middlebury College Theatre Department

P.S.  That poor rigger on the bed of the hotel room must have been filthy - sprinkler water isn't exactly Evian!




More information about the Stagecraft mailing list