Concord 4 CO gas sensor false alarm

I’m a happy Surety customer since 8/2017 when you helped me migrate my professionally installed hybrid Concord 4 system to Surety/ADC.

Since the very beginning, I’ve had one CO sensor (Zone 14 “Second Floor Gas”) generate a false alarm in certain situations, causing the local fire department to show up at my door. The central monitoring station told me that because this is a gas detector, they cannot undo the dispatch even when I tell them it’s a false alarm. I’ve been putting it off, as it doesn’t cause issues in normal situations, but I want to address and fix it now (I’ve had the same fireman a friend come out twice this summer!)

It happens in certain specific situations (I cannot tell exactly what triggers it), when (I think) I try to arm the system when some windows are open. Again, I’m unable to tell exactly what causes this. The most recent false alarm was last night at around 8:10pm ET (Tues 9/7). I was outside, wanted to arm the house (some windows were open) using the ADC app, and then the dreaded call came from the central monitoring station, followed by the fireman ringing my doorbell.

When I arm stay (overnight) or arm away (leaving for vacation), all my windows and doors are closed, hence no issue.

Anyway, a bit more background:

  1. My “detector” sensors

Sensor name Zone Reporting Type Physical Type
Basement Fire Alarm 8 Smoke/Heat Smoke/Heat
Basement Gas Detector 7 Carbon Monoxide CO
Garage Gas Detector 2 Carbon Monoxide CO
Second Floor Fire 16 Smoke/Heat Smoke/Heat
Second Floor Gas 14 Carbon Monoxide CO
Smoke First Floor 15 Smoke/Heat Smoke/Heat

The problem sensor is Zone 14 (Second Floor Gas).

  1. FWIW, in 8/2017 when we activated ADC with Surety for the first time, Zone 14 was not detected. I don’t remember exactly how we fixed it, but maybe after a scan it popped back up in Zone 14 on ADC list.

  2. There are 6 detector type sensors around the house (see table above), and some are reporting as “Smoke/heat” and some “Carbon Monoxide.” But when I look at the two 2nd floor sensors, they both look identical (see attached pictures).2F i (Gas 14 or Fire 16) 2F ii (Gas 14 or Fire 16)

Because they look alike, I don’t know which is Gas and which is Fire. How does ADC determine which is which? Is the same sensor able to function as either one?

  1. FWIW, both sensors on the 2nd floor (shown in pics) have flashing green light.

  2. What could be causing the “Second Floor Gas” detector to generate an alarm when clearly there’s no CO present?

  3. Can you help me identify which of the two 2nd floor sensors is the problematic sensor?

  4. Once the problematic sensor is identified, what can I do to eliminate the false alarm?

Any help would be greatly appreciated!

Because they look alike, I don’t know which is Gas and which is Fire. How does ADC determine which is which? Is the same sensor able to function as either one?

ADC does not determine what the sensors are or what they do. The panel does, and reports that to ADC.

Those are both the same CO detector, looks like the GE 260-CO.

Of course neither of those are smoke/fire detectors, so if you believe that they represent zones 14 and 16, zone 16 is improperly programmed.

Those I believe are wired sensors and would be wired into terminals on your panel or a zone extender near the panel.

It is more probable that both of those are wired in series as one zone, 14, and you have a smoke detector elsewhere.

If it has been a long time since you installed and you aren’t sure, I would definitely check again to make sure you are aware of all the compatible smoke and CO detectors on the system. It is important to know where these are located.

Can you help me identify which of the two 2nd floor sensors is the problematic sensor?

Is the detector in the second image right next to a window, an outside door? Interior door? It would be the most likely culprit if it is right next to a window or outside door, CO detectors should not be placed close to a window like that.

I would check the manufacturer replace by date on the sensors, replace any that are expired, then consider removing and replacing it with a wireless model if that is wired right next to a window.

Thank you Jason. Yes, all my sensors (smoke, CO, motion, perimeter) are wired sensors from the original install, except 2 glass break sensors we installed after moving in.

It is more probable that both of those are wired in series as one zone, 14, and you have a smoke detector elsewhere.

That would make sense! When I get home, I’ll go around the house to locate and identify all the 6 smoke and CO sensors. Smoke sensors are installed on the ceiling usually, correct?

Is the detector in the second image right next to a window, an outside door? Interior door? It would be the most likely culprit if it is right next to a window or outside door, CO detectors should not be placed close to a window like that.

Both of these 2 CO sensors are inside the 2nd floor hallway, away from a window or outside door. (The second image is right outside a bedroom door that opens into the hallway.) This seems to make sense given your explanation of CO sensor locations.

As I investigate further, can you help me with the following:

  1. If the CO sensors are located correctly, why would this zone throw off a false alarm when a certain Arm/Disarm sequence is triggered? Would I be able to isolate the problem to one of the two sensors in the same zone?
  2. Other than using logical guessing, is there a way to check which zone the sensor is programmed to? (e.g., you can open a window and the panel will let you know “Kitchen window open”)
  3. As I attempt to open the CO sensor and check the mfr replace by date, I’m afraid it will send out an alarm to dispatch firefighters again. Is it ok to tamper with detectors like that?
  4. Is there a way to take the system offline (test mode?) temporarily so whatever experiments I perform, the panel doesn’t notify the Central Station? For example, I want to test Arming the system with a couple of windows open without triggering the CO alarm.

Thanks again.

Adding to this post, I wanted to show you the Activity log from Tue 9/7 at 8:10pm ET when my mobile Arm Away command immediately triggered “Second Floor Gas” alarm. (See below.)

I also noticed in the Activity Log something strange: Basement Door and Garage Window opened. Basement Door and Garage Window were never opened. They were always closed. I’m wondering if these three sensors are related somehow (Second Floor CO, Basement Door, and Garage Window).

Any guidance here would be greatly appreciated.

8:10 PM - 8:11 PM

—————————

  • Second Floor Gas

Closed

  • Garage Window

Closed

  • Basement Door

Closed

  • Panel

Disarmed

  • Panel

[Mobile] Command: Disarm (onum)

  • Garage Window

Opened

  • Basement Door

Opened

  • 8:11 PM

Second Floor Gas

Carbon Monoxide

  • Second Floor Gas

Pending Alarm (Awaiting Panel’s Programmed Delay)

  • 8:10 PM

Panel

[Mobile] Command: Arm Away (onum)

I also noticed in the Activity Log something strange: Basement Door and Garage Window opened. Basement Door and Garage Window were never opened. They were always closed. I’m wondering if these three sensors are related somehow (Second Floor CO, Basement Door, and Garage Window

If those devices always open in tandem with the Alarm event from the CO detector, you are likely onto something. Are those three zones wired into the same Zone extender? It is highly likely that the extender is causing the issue. Perhaps loose wiring or just an internal malfunction.

Adding to this post, I wanted to show you the Activity log from Tue 9/7 at 8:10pm ET when my mobile Arm Away command immediately triggered “Second Floor Gas” alarm. (See below.)

A CO detector does not rely on arming status to trigger. This is coincidental.

As I attempt to open the CO sensor and check the mfr replace by date, I’m afraid it will send out an alarm to dispatch firefighters again. Is it ok to tamper with detectors like that?
Is there a way to take the system offline (test mode?) temporarily so whatever experiments I perform, the panel doesn’t notify the Central Station? For example, I want to test Arming the system with a couple of windows open without triggering the CO alarm

Yes, you can place your account on test mode with the monitoring station at any time in the system manager here. Visit the test my system tab to view test mode options.

Other than using logical guessing, is there a way to check which zone the sensor is programmed to? (e.g., you can open a window and the panel will let you know “Kitchen window open”)

You can tamper the sensor and manually open the circuit. The panel will then show an alarm or open condition for the programmed zone.

Thanks. I’ll investigate the Zone extender and see if there is loose wiring, wires touching, etc. My panel consists of the main Concord 4 panel (8 HW zones), an expander (8 HW zones) and another expander (8 HW zones; 4 used). I can visually confirm 20 zone terminals are being used (8+8+4) in the panel.

Would it make sense to physically move the wires (for the 3 problem zones) to the free terminals? In that case, do I need to delete and relearn the zones, or will the panel recognize the moved wires?

That is the frustrating thing. The CO sensor never goes off randomly unless some specific arming status change occurs. It has happened at least 6 or 7 times in the last 4 years, only during arming / disarming. There’s never been a random false alarm from the CO. Could there be something faulty with the panel programming or wiring that triggers this strange behavior?

I’ll try that, and thank you for letting me know about the system manager. I assume tampering the CO sensor to manually open the circuit should only be done in test mode, correct? (I don’t want the fire department to show up again.)

Would it make sense to physically move the wires (for the 3 problem zones) to the free terminals? In that case, do I need to delete and relearn the zones, or will the panel recognize the moved wires?

I’m trying hard to remember and I don’t think in a decade of working with these systems that I have actually seen a terminal itself be the culprit. Rewiring has other benefits though, such as exposing fresh cable, tightening wires, reducing resistance, discovering breaks, etc., so it is a good thing to try, I would just generally do so with the same terminal.

You could try new terminals if you prefer, but you would need to reprogram. The terminals can be programmed as any zone number, but they have to be programmed in. It’s more likely an entire extender would have an issue rather than 3 specific terminals. And if it is 3 specific terminals, the wiring/resistance is likely the problem.

That is the frustrating thing. The CO sensor never goes off randomly unless some specific arming status change occurs. It has happened at least 6 or 7 times in the last 4 years, only during arming / disarming. There’s never been a random false alarm from the CO. Could there be something faulty with the panel programming or wiring that triggers this strange behavior?

There wouldn’t be anything specific to the act of arming the system that would have an impact on zone terminals like that.

If I had to guess, power draw may be at play here. The 2 8 zone expanders use about half of the max current draw for the panel aux outputs.

How many keypads do you have? Sirens? Do you have an interrogator two way voice module?

What else is powered off of the panel outputs?

I assume tampering the CO sensor to manually open the circuit should only be done in test mode, correct? (I don’t want the fire department to show up again.)

Correct, please be sure to place the account in test mode first.

Got it. I’ll try rewiring the same terminals and test. (Basic question: When rewiring/tightening terminals, should I first disconnect the panel from the backup battery power?)

I have 3 keypads around the house, and, I think maybe 1 Siren. No two way voice module. Other than the 20 wired sensors, the only other device drawing power is the LTE cellular module that also acts as my Z wave hub. The LTE module is connected using a Cat5 cable (twisted pair terminations using the 8 wires in the Cat5).

Other than the CO sensor going off at certain arm/disarm commands, I have not had any issues with the system. One other potential (?) culprit I can think of is: The few times the CO sensor went off at arm/disarm, I think a perimeter sensor may have been open. Could the false alarm be related to sensor bypass errors? (When we Arm Stay at night, all windows and doors are closed. When we Arm Away for vacation, again all windows and doors are closed.)

When rewiring/tightening terminals, should I first disconnect the panel from the backup battery power?

It is a good habit to avoid accidental shorts if you interact with power wires.

I have 3 keypads around the house, and, I think maybe 1 Siren. No two way voice module. Other than the 20 wired sensors, the only other device drawing power is the LTE cellular module that also acts as my Z wave hub. The LTE module is connected using a Cat5 cable (twisted pair terminations using the 8 wires in the Cat5

Is that a speaker siren or a piezo? What is the model number?

Just with the keypads and the extenders and idle current draw of the LTE module you are sitting at around 90% of the amp capacity of the Concord 4 output.

Spikes when certain actions occur may indeed be underpowering the extenders or the sensors themselves. It would be a strange circumstance but you have enough devices that underpowering does become a concern, and it is a hard thing to test. You might try recreating the issue with two of the keypads disconnected.

Are you using Force Bypass when arming your system with zones open?

If you set your account on test mode, and you open a couple windows, then remotely arm the system, do you get the alarm? The same sensors open?

Can you repeat that reliably a few times?

I will run some tests over the next couple of days and report back.

I never consciously arm the system with zones open. (Indirect or Force Bypass.) We prefer to have all windows and doors closed when arming.

I’ll definitely run a bunch of different tests and report back.

Thank you.