Welcome! Here are the website rules, as well as some tips for using this forum.
Need to contact us? Visit https://heatinghelp.com/contact-us/.
Click here to Find a Contractor in your area.

GFCI keeps tripping

GCL Member Posts: 4
Hello all.
I’ve been wrestling with an electric radiant heat system that I installed roughly 6 years ago in my upstairs bathroom. I found one morning 2 weeks ago that the floor was cold and saw that the GFCI had tripped on my Honeywell TH115 AF-GA thermostat. For reference only, I have 2 other floors in my house that have this same type of system. This is the only problem that I’ve had with these 3 systems. The system in question has a 1’ x 15’ heating mat and it’s setup to run on 120 volts. It is on a dedicated 15 A line.
I would like to list all of the things that I’ve tried to find the reason for this situation.
1. Trouble shooting began with measuring resistance between the heating leads. Result was 64.2 ohms. This is suppose to be ok. Resistance between each lead to ground showed no change in value on my multimeter. Meter read “1”. (I measured another floor setup and got similar results from lead wires to ground).
2. I took thermostat from upstairs system and installed it on downstairs system. Thermostat worked ok.
3. I took downstairs thermostat and installed it into upstairs system and the GFCI tripped. At this point, one might think that the floor system might be failing and the thermostats are ok except step 1 showed mat was ok.
4. I borrowed an Insulation Tester and measured resistance between heater leads to ground again. I set tester to deliver 100 volts. Both leads read roughly the same. It would start at .2 mohms and climbed to .92 mohms within 15 seconds. I measured the system downstairs and both leads to ground read 110 mohms. These values are good so I’m told. The .92 mohms values are suspicious to me.
5. I then decided to call tech support and after listening to my attempts (the insulation tester results went over their heads which is ok - it went over mine too), I was told to hookup line voltage to heater leads and see if floor heats up. If floor gets warms, the thermostat is bad and floor mat is ok. Bathroom floor got warm. I let it warm up for about 30 minutes.
6. I ordered new thermostat and received it in 3 days.
7. Installed new thermostat into upstairs system and GFCI tripped. Installed new thermostat in downstairs bathroom. Thermostat worked correctly.

Ok???? Now what.

8. I inspected the heater and sensor wires from where it comes out of floor to thermostat looking for broken insulation etc and found one suspicious nick in heating wire. Cut that part off still leaving me barely enough to work with. Tested thermostat. Still trips.
9. Went to my breaker box and performed a few actions. Made sure all wiring was securely attached and also gently blew out dust and any other debris that was present. No change.
10. I ran a brand new length of 14-2 wire from upstairs bathroom to basement making it impossible to walk safely around my house. My wife loved this! Attached new wire to new 15 amp breaker. Thermostat still tripped.

I’m not sure how this idea got into my head but I pulled my generator out. I then installed a male plug on new wire and plugged it into generator. Tested thermostat and it didn’t trip! I left if for roughly 30 minutes and bathroom floor warmed up nicely. I am now completely baffled.

I can’t help but think that it’s something so simple and it’s staring me in the face. Can anyone shed some light on this problem? Could the .92 mohms reading in the heater leads be the fault shown by the insulation tester?



    JUGHNE Member Posts: 11,058
    Just to clarify, the GFCI tripping is part of the T-stat or in your panel?
  • pecmsg
    pecmsg Member Posts: 4,837
    Change the GFIC if it still trips probably a issue with the in floor.
  • GCL
    GCL Member Posts: 4
    The GFCI is part of the thermostat.
  • ratio
    ratio Member Posts: 3,623
    The megger showed less than 1 megaohm to ground on the faulting heater and 110+ megaohm on a good heater? The heater has failed for one reason or another. If you couldn't find the fault on the exposed part it's...

    Sorry buddy. I think you know what you're going to have to do to repair this.

    JUGHNE Member Posts: 11,058
    edited December 2018
    So after switching controls around the problem stays with the same floor mat?
    If you bypass the GFCI part (even the temp one), the breaker stays on and the floor heats?

    It seems tempting to bypass the GFCI permanently......however, you must have a leaker of current into the floor. I would NOT eliminate the GFCI protection.

    You could put in a standard T-stat and use a GFCI circuit breaker in the panel.....worth a try.

    The wet tile could become energized, you outside the shower with wet feet and wet hands are adjusting the well grounded water valve. That leaking current is looking for a path to ground and you could become it.

    IIRC, the trip current for GFCI is 4 milliamp. ….. .004 amps.
    When GFCI devices came on the market 40+ years ago it was determined that anything above 2 milliamp could be lethal to the human body under the ideal conditions of flow across/thru the heart. I believe they raised the standard to 4 ma because of constant nuisance tripping of the early models, which would make people bypass the devices.
    EBEBRATT-Ed Member Posts: 15,512
    GFCI measure current in hot wire to heating mat and also measures the current in the neutral coming back. The GFCI trips if their is an imbalance between the two measurements. The imbalance when it trips is very small like 5 milliamps.

    If the GFCI doesn't sense the same current on each wire than your leaking to ground somewhere.

    Seems like you have done about all you can to track this down.
  • Jamie Hall
    Jamie Hall Member Posts: 23,272
    You have an electrical fault, and the GFCI is telling you about it. It could be in the wiring from the thermostat to the mat. Unhappily, it could also be in the mat, particularly under a tile floor (doesn't take much for a shifting tile to damage the insulation).

    Like most safeties, pay attention to it -- don't try to bypass it or wish it away.

    The relatively low initial resistance reading is almost low enough to give a GFCI enough imbalance to trip, and you may not be reading the lowest initial resistance.

    Can you tap into the electric mat where the thermostat wiring reaches the mat? If so, you could try running new wire from the thermostat to that mat connection.
    Br. Jamie, osb
    Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England
  • GCL
    GCL Member Posts: 4
    Yes, I said I was suspicious of those .92 mohm readings but I needed someone smarter than me to understand the readings.
    I never liked that tile choice anyways.
    Thanks you very much and please have a pleasant and safe Christmas!
  • GCL
    GCL Member Posts: 4
    I guess I have one more question.
    Why did the system work ok when getting power from my generator and not my household power?
  • Jamie Hall
    Jamie Hall Member Posts: 23,272
    My first suspicion on that is... that your generator may not be properly grounded. A lot of them aren't.
    Br. Jamie, osb
    Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England