smoke detectors in air ducts and water condensation

[Eyeap]

Hi, I'm thinking of installing smoke detectors into air conditioning ducts and I am wondering if I'm going to have problems with water condensation getting into the smoke detectors and killing them.

Anyone have any experience in this area and could tell me if there is going to be a lot of water condensation? If there is then any ideas on stopping it getting into the smoke detectors?

Thanks.

 

[quark]

Eyeap!

Generally smoke detectors are installed in return ducting and are used to control fire control dampers. We have been using them for years and I am not clear about your possible doubt of condensation. Why do you anticipate it?

Regards,

Repetition is the foundation of technology

 

[kitha]

In most of the regulations duct smoke detectors are necessary only if the ducts cross the boundaries of fire zones causing spread of fire to other zones. Otherwise there's no need of having so called duct smoke detectors. The construction of duct smoke detectors fully prohibit the condensation. They have probes going into the ducts and rubber gommets that prevent air leakage. I have installed them in several occasions with no problem at all.

 

[Eyeap]

Thanks for the input guys. It's just that some of my sales guys are telling me that the duct mounted smoke detectors would fill up with water and I was wondering if anyone out there had any experience with condensation problems in duct mounted smoke detectors.

 

[ChasBean1]

Negative. In U.S., duct-mounted smokes are required to serve air handling units with supply air >2,000 cfm, and in each branch of return air systems greater than 15,000 cfm (courtesy of NFPA-90A).

In most of the installations I see, the detector housing uses pitot tube technology in that total (velocity plus static) pressure enters ports in a tube protruding into the air stream, facing the moving air. Static pressure is sensed by a simple tube from the detector housing into the duct. Basically, the difference between the total pressure and static pressure (equal to velocity pressure - preferably 400 feet per minute or greater) drives flow through the smoke detector, which is mounted within a chamber that experiences flow-through by this differential sensed by the tubing protruding into the duct.

Condensation will only occur if the surface temperature within the tubing is below the dewpoint of the air in the duct. Since this is a continually flowing system, the smoke detector sensing tubing attains equilibrium temperature with the duct air temperature. This means that if air doesn't condense randomly within the duct, it will also not condense on the wall of the smoke detector sensor tubing.

Basically, no worries, as long as the sensing location is away from water sources such as humidifiers.

Personally, in my experience, I have never seen problems with condensate adversely affecting in-duct smokes.

 

[LCatey]

ChasBean1 did a very nice job of highlighting smoke detectors, in my opinion. I would only add that I have had a case where a humidifier caused a smoke duct detector to trip. The humidifier controls failed to shutdown the steam valves when the air handling unit shutdown, so the duct was essentially flooded with humidity because the air was no longer moving. The humidifier high limit also was unable to cut out the humidifier because it was bypassed for some testing.

The result was "raining" in the ductwork, which caused all kinds of raining below the ductwork, and the smoke detector was activated. This was an abnormal condition and the only occurence I have ever witnessed in my career.

Generally speaking, this should not be a problem if your smoke detector is located an appropriate distance downstream of your humidifier (if you have one in the duct) and if your high humidity limit is properly set.

 

[kitha]

In Canada, as per Canadian ULC requirements, smoke detectors are placed in the supply air duct. It is basically in contradiction with NFPA90A.

 

[ChasBean1]

LCatey - I've seen the same. During commissioning one project (no names mentioned), I ran a humifier high limit test immediately followed by a high static shutdown. The shutdown worked immediately, but the humidifier valve took a while to close, hence, a lot of steam in a dead duct --> supply smoke goes off --> building evacuates. Not cool. It wasn't related to condensation in or on the duct smoke or its tubing, but the fact that the steam from the humidifier looks like smoke, causing a device no-response signal from the detector.

For Eyeap's purposes, I would not worry about condensation in a detector that will be at equilibrium with the duct air temperature.

 

[Eyeap]

Thanks a lot for the input guys!