2-wire smoke detectors

[Mike]

An issue I have had out there for awhile and recently went back to try and resolve:

I had 4 System Sensor 2WTAB detectors installed and I added a 5th. When I added the fifth I had also added a reversing relay and a 2W-MOD2 (clean-me type signal addition to interface with the elk).

http://www.automatedoutlet.com/product.php...&cat=199&page=1

I started getting some false alarms, which I assumed was tied to the other pieces or the latest detector being added.

Some time passed and I removed the reversing relay and the 2W-MOD2 so that it is just a series of 2-wire detectors, on zone 16, with the jumper set for 2-wire smokes and the panel programmed as 10-Fire alarm, and type 6 - 2 wire smoke. When I check the status in ElkRP, the state is EOLR and the volts is 12.2.

What I was attempting to do was verify the cutoff timers worked so I could leave the detectors connected and run tests to see if a false occurred, and if so, then remove the latest detector and run it again.

I set the cutoff timers for 30 seconds and triggered the detectors using a spray tester.

The alarm sounded, but did not cut off. I went down to the keypad and tried to reset the detectors using option 2. When the alarm is sounded it would not let me hit the key to the right to 'activate' the command. When the alarm was not sounded, I could hit the option to the right and reset the alarm (I would hear a relay type reset sound at the panel and a brief message at the keypad).

The speaker connected to the panel (output 2) still makes the clicking sound that I thought might be related to the reversing relay when the alarm for the smokes is activated.

My assumptions have been dashed, and it seems I have missed something. If there are any recommendations on how to troubleshoot this or review further I would appreciate any ideas.

 

[Spanky]

Mike,
You will not be able to enter the user menu with the ELK key until you have acknowledged the alarm by entering your arm/disarm code.

The fire alarm cutoff time (installer menu 6, CT02) is defaulted to 65535 which is never cutoff. Make sure you changed the fire alarm cutoff time to 30 seconds.

You say that the speaker connected to output 2 is making a clicking sound. It should be making a siren sound. If not, check Global option 26 to make sure that output 2 is set for siren output not voltage output. DC voltage on output 2 connected to a speaker will damage the speaker. If everything is programmed correctly and you still do not get a siren sound, output 2's driver transistor could have been damaged.


Here is a tidbit of trivia on 2 wire smoke detectors. There are at least two connection ratings of S09 and S10. The M1 uses S09 rated 2 wire smoke detectors which basically means when the detector goes into alarm it pulls the 2 wire fire alarm loop lower in voltage than S10 detectors.

 

[Mike]

Thanks Spanky.

You will not be able to enter the user menu with the ELK key until you have acknowledged the alarm by entering your arm/disarm code.

I realized I was using the installer code, but when I retested (using the default user code) it still would not let me 'go right' on the keypad to trigger the smoke detector reset. It eventually did (10 seconds or so after trying, meaning i did not see the 'resetting smoke detectors' and it stayed on option 2 for that amount of time). The panel went silent, but the detector continued to go off. The behaviour when resetting the smoke detectors does not seem to be right (almost like it was locked up to a degree), and is different from normal operation using the keypad. I tried the same operations when the panel was not in alarm and it worked smoothly.

The fire alarm cutoff time (installer menu 6, CT02) is defaulted to 65535 which is never cutoff. Make sure you changed the fire alarm cutoff time to 30 seconds.

This one I checked again, it is set properly under cutoffs in ElkRP. It seems the panel resets, but the detector does not. Seems like a clue, but I'm not sure if perhaps anything can be related to my suspicion of the detector being bad. This last test was triggering the suspected bad detector.

You say that the speaker connected to output 2 is making a clicking sound. It should be making a siren sound. If not, check Global option 26 to make sure that output 2 is set for siren output not voltage output. DC voltage on output 2 connected to a speaker will damage the speaker. If everything is programmed correctly and you still do not get a siren sound, output 2's driver transistor could have been damaged.

That was it, my mistake. The global option was set improperly. It is working properly now.

I'm thinking my troubleshooting may be backwards, perhaps I need to remove the suspected defective detector and retest that the system is working properly. Then once that is confirmed, link that one that may be bad back in to confirm it.

I also found a post you made about verified fire alarm settings. I might want to look at that once I confirm things are working properly from an initial perspective.

 

[Digger]

I may not fully understand your situation here but it sounds like you are using Alarm verifcation and also a detector with a built in siren.

1. maybe this is happening:

If the detector goes into alarm the alarm verification feature will automatially do a 5 second reset (powers down zone 16 for 5 seconds) and then the zone becomes energized again. For the next 20 or 25 seconds the panel is in a "retard" mode (its far from retarded its just a term that means it will ignore alarms on that zone allowing the detector to reset without false alarming). During that retard mode the detector is resetting. In your case the reset time for a System Sensor Detector is maximum 15 seconds (warm reset time for fire alarm verification). If the detector is in alarm and the panel is not yet in alarm it is because of the fire alarm verification feature until that 20 to 25 second retard time expires. At that point the panel should go into alarm and you should be able to acknowledge the alarm and reset the detector.


2. or you have a bad detector that will not reset. (honestly not to likely if its a new detector)

3. or the relay is not installed correctly (I never used a relay like that on a 2 wire zone so I am not familiar with it.

4. or none of teh above

 

[Spanky]

Digger knows the M1 fire system intimately.

 

[Digger]

The rumours of me sleeping with the M1 are greatly exagerrated. I swear I never slept a wink.

And to everyone reading this it is an inside joke...............

 

[Mike]

Let me clarify.

I removed the reversing relay and the 2W-MOD2. Those just added complexity and other items that could be introducting a problem. I went back to basics.

What I have left:

5 System Sensor 2 wire detectors, attached to zone 16, 2-wire jumper enabled.

Zone 16 is defined as 10 - Fire Alarm (not verified).

Fire Alarm cut off timer is set to 30

I was trying to troubleshoot to determine if the false alarms were related to the last detector that was installed, but needed to ensure that the detectors would turn off if they falsed (I would see it in the log, but I had an episode where they were on for hours). That led me to this circumstance and the detector does not seem to reset properly (preventing me from isolating the new detector as a culprit).

Does this make sense? This has gone through a few iterations.

 

[Digger]

the cutoff timer is for the M1 siren and not the detector sirens. To reset the detector and silence the sirens in the detector you need to acknowledge the alarm and then do a smoke detector reset. Until you reset the detector the siren in the detector will go on forever.

The M1 is unique where you might be able to write a rule that after "x" amount of time of Zone 16 being in alarm you could do a smoke reset. I am not to sure how happy the NFPA would be with that since I do not beleive that the detectors should be automatically reset (how would you know which head tripped if you reset it).

The real issue is why did you have a false alarm. That is a major taboo..... nothing worse than that except for a failure to alarm obviously.

If you can isolate the detector that falses try swapping it with another one in the zone (if head 5 falses swap 2 and 5 etc) and see if its the detector that is bad or a location problem. Photoelectrics are less likely to false so I would be surprised if you have a bad head. Sounds like you will have to do a lot of process of elimination unfortunately.

 

[Mike]

I was trying to go through a process of elimination, but my problem was the detectors did not reset (after 7 hours), which brought me to approach the reset issue so I could leave them on to test the falsing.

It all started when I added the fifth detector so I am assuming that the detector (or the location as you put it) must be the culprit. I originally thought the reversing relay and 2W-MOD2 might be complicating it so I took them out.

I'm trying to make sure the reset works so that if they false and do not reset that the alarms are not going off for hours until someone gets home. As long as I know the reset will work (which is why I went back to see if I had anything installed wrong) I can verify false alarms by looking at the log, but I can't have the alarm going off for hours.

You mention location being a possible issue, what might be an example of location based issues?

That and, am I approaching the issue properly?

 

[Digger]

Locations near humidity (bathrooms kistchens etc) and near the furnace room etc might be a problem. Also if you have forced hot air heating keep them aeway from vents.

hard to say what is going on..... you need to check the red LED's on the detectors to see which went into alarm

Slight chnace its a wiring issue?

As far as not reseting put a meter across zone 16 and do teh reset from the keypad. Zone 16 should drop to zero volts for 5 seconds. That should be more than enough for the system detectors to reset since they need less than 3 V for 1 second or something close to that.

if you get the zero volts then it looks like the panel reset is ok.

 

[Mike]

Thanks for all the feedback Digger. I don't think it is a location issue from your description. The location with the detector in question is a bedroom, the only difference between that and the other locations is the room is an addon and has three steps up.

As far as wiring, I did add on the leg of wiring when I added the last detector, and it was a little bit more tricky than the other wiring because of the extension and where it ran in the attic. When I had the 2W-MOD2 installed, I did the walk-test and it passed (although since the symptom is occasional falsing, perhaps that is not a clear positive test). However, when the panel was in alarm, the lights on the 2W-MOD2 were strange. I inquired a few times with System Sensor but did not get a response, perhaps I should track that down. If the 'wierd' light combinations (may have been fine, just wasn't documented under alarm condition) perhaps were an indication of a wiring problem, then that might give me another lead.

I had tried cleaning the detector (even though it was new). The wiring is the other 'change' prior to the falsing. Maybe I need to run another length to rule this out. Since (to me) the issue appears to be intermittent, is there a test I can do to help positively confirm that leg?

I will test zone 16 voltage this weekend before I do anything else, but from what I was seeing, the issue seemed closer to the detector (except for not being able to 'activate' the 'reset smokes' function for a bit) which I did not consider the wiring since the walk-test was ok. If a slight issue with the wiring could cause this, it makes more sense than the detector since it is brand new.

 

[Digger]

anything in the bedroom that could cause a problem??? drop down attic stairs, fireplace, heating vent blowing on detector, fresh paint?

Usually photoelectrics are not prone to false alarms.

You do have all of the same detector on the zone correct. That zone should handle up to 20 detectors of the same type. usually if a detector does not reset its because of a capacitance problem (so the detector stays latched in alarm when you do a reset).

For the false alarm if the detector LED was lit you know it was the head itself and usually not the wiring now that I think about it. The LED would not light if the wiring caused the panel to trip if you think about it.

Can you easily exchange the head?

 

[Mike]

anything in the bedroom that could cause a problem??? drop down attic stairs, fireplace, heating vent blowing on detector, fresh paint?

No vent, no attic, no fireplace, in fact we don't even go in that room at all.

You do have all of the same detector on the zone correct. That zone should handle up to 20 detectors of the same type. usually if a detector does not reset its because of a capacitance problem (so the detector stays latched in alarm when you do a reset).

They are all the same type.

For the false alarm if the detector LED was lit you know it was the head itself and usually not the wiring now that I think about it. The LED would not light if the wiring caused the panel to trip if you think about it.

Hmm. I don't remember the light being on but it has been quite awhile so I wouldn't bet on that (I don't remember checking it specifically as it was quite loud at the time). During the case where it was on for hours, my wife shut off the panel so I'm not sure. Perhaps I can try to reproduce the issue this weekend and check (if I can get it to false). However, if it were the wiring, would the detector go off for hours? Then again, at that time the reversing relay was in place, so with noone to reset it, it should not go off. Now however, with no such relay, shouldn't the detector stop sounding once the alarm condition clears, right?

I'll run the tests on voltage this weekend to see about the reset in general.

Can you easily exchange the head?

If you mean swap it with the head of another detector, absolutely, there are four others to choose from at the moment.

 

[Digger]

the detector will latch into alarm until reset. so it would go forever until reset.

I meant can you return it and get a new one where you purchased it.

 

[Mike]

I was trying to get to the point of proving that unit was the culprit in order to do so. Granted this has taken me longer than expected, but I will try to exchange it. I don't expect a problem there.

I'll run those tests though over the weekend without that unit, which should prove this out as well.

It also sounds like I should switch it to fire verified to help prevent false alarms as well.