GRI Water Sensor Problems


Active Member
I wired a new house build with a lot of GRI water sensors.
Some were 2600s, where a single sensor/single input was required (like under a washer).
Many were 2600Ts with one sensor, like under hose bibs in the wall, so we could remote mount the sensor in the wall and the switch in an accessible area for service.
Three were 2600Ts which allow up to ten sensors to be attached to one input. These were used in three places, for kitchen (sink, dishwasher and refrigerator), garage (utility sink, water filter, water heater), and shop (toilet, water heater and utility sink).
Only one is giving us trouble. The shop. The three remote sensors are glued to the concrete behind the toilet and under the sink and water heater contacts facing up (not touching the concrete). 99% of the time it is fine, but twice it has triggered a water alarm, and both times in the dead of night. Each time the area was bone dry. We have checked voltage, connections and resistance and it all seems to be fine. All three sensors legitimately trigger an alarm when using a wet finger across the contacts, but these false alarms are baffling.
Has anyone used a GRI 2600T and had it fail in this manner?
Are you sure it is the 2600T and not the zone input on the panel that's the problem?  First thing I would do is swap it to a different zone and see if the problem moves with the sensors.

How high is the humidity in the shop? High humidity can cause false triggering in some cases. GRI makes a GRI 2600T-LS, which is less sensitive, for those situations.
It's not impossible that the sensor is bad.  But I would also suspect an intermittent fault in the wiring as a possibility.  Some mice nibbling in the night?  
The shop has a mini-split and is temperature controlled with humidity around 45%. I haven't seen any evidence of mice but will do a good look around.
Leakage was not tested. Boy those meters are expensive.
Thanks for the zone idea RAL, I'll swap zones as a first step. If it happens again I'll swap out the switch. After that I guess it's time to bite the bullet on a leakage tester.
Thanks a lot guys!
For a quick test, just place your multimeter in the highest ohms scale (if not auto-scaling) and see if there is any leakage between the conductors.  Make sure one end of the cable is open (disconnected) with no conductors touching each other or the ground/frame work, then measure from conductor one to conductor two.  You should see infinite resistance unless there is some sort of crush or other factor causing resistance between these conductors.