I have a 240V dedicated circuit installed in my bathroom for in-floor heating. I have a GFT 2118 cable kit from warmtiles installed under tile. It was installed in 2009 and ran perfectly until 2015 when the thermostat had an end of life issue. The original thermostat was a GT-2 and I had no issue with it or GFCI trips prior to the end of life issue.

I installed a new FG thermostat and the thermostat GFCI has been intermittently tripping the since it was installed.

I checked the resistance in the heating cable:
Black to white - 38 ohms (exactly what it should be according to manufacturer)
Black to ground - infinite resistance
White to ground - infinite resistance

I have also checked the resistance in the sensor cable:
12,000 ohms (exactly what it should be according to manufacturer)

Initially I thought it must be a thermostat issue so I replaced the original FG thermostat with a new one and the same issue has continued.
The thermostat GFCI appears to trip every few days at the current temperature setting, but if I increase the temperature setting it will trip more often.

any ideas? I am not sure why it would work perfectly well for 6 years and suddenly I would have a cable issue at the exact same time I change my thermostat.

 

Did you possibly nick the heating cable when cutting out grout to install the new t-stat sensor? If it is intermittent, it may be related to the amount of moisture in the floor.

One other possiblity that comes to mind is if there is a skinned wire or ground wire brushing up against another wire in the junction box. Perhaps inspecting and repacking the box might help.

 

Thanks for your reply ibpooks. I used the existing thermostat sensor that is already under the tile. I made the assumption it would function with the new thermostat(same brand, just newer model thermostat). Do you think this could be a problem?

I thought it may be related to wiring within the box so I have taken it apart redone it about 3 times without success..

although moisture in the floor is on the list of possibilities I am doubtful this is contributing as it often trips in the middle of the night when the floor is bone dry.

I appreciate your thoughts and others are more than welcome.

 

When you checked the resistance to ground.... what meter did you use ?
Does it have multiple ohms settings ?

If you could post the make and model.... that would be helpful.

When you are checking for leakage..... you need to be using a high resistance setting on the meter.

 

Thanks for you reply PJmax.

I used at Mastercraft Digital Multimeter (Product #52-0060-2 from canadian tire in canada).

It's highest ohms setting is 2000K which is what I used when testing for leakage. It does have multiple ohms settings (2000K, 200K, 20K, 2000 and 200) and I started at the lowest and worked my way up. All settings gave a infinite resistance measurement.

I am quite confused by what may be happening and curious as to what my next logical step would be in trouble shooting the issue. Any thoughts would be appreciated.

 

This is just a thought, but what temperature was the floor when you checked the resistance of the cables? It's possible that the resistance values change as the floor temperature increases.

 

It's ok for the resistance of the cable to change. What is not ok is any leakage to ground.

Checking it using the 2000k setting was correct. Any leakage should have shown up there.
What are you using for a controller.... they list several ?

In reading more of the tech specs on your system they describe the cable as a braided ground cable. That would mean that the heating conductors are contained in a metal braid. What could be happening is that one of the conductors upon heating up is touching the braid. You would need a megger to check the cable for a problem like that.

The megger sends a high voltage thru the cable and measures or looks for an arc.

 

The resistance values I was thinking of would be black to ground and white to ground. The tested values listed by the OP were infinte, do they change as the floor gets warm?

 

I was typing as you were. Yes.... I agree that could be the problem.

 

All the trouble shooting I have done has been when the floor is cold. I will test out the resistance values when the floor is warm and see if there is a difference.

The thermostat is a FG thermostat. emerson industrial/wt_thermostats/FG-FGS/spec_sheet.pdf

I have wondered about a issue with insulative resistance and trying to get a hold of a megger tester, but it seems odd that it worked so well for 6 years and this issue has come up only since changing the thermostat.

thanks again for your thoughts.. i will post again after i check the resistance while the tile is warm.

 

In the spec link I put in your post.... the sensor required for the FG stat is 10k ohms.
I don't think that's your problem though.

Some further info on the FG stat. emerson industrial/FG_installation_instructions.pdf

 

Who was the manufacturer of the original thermostat? Most thermostats should last longer than 6 years, what happened to the original one? The problem coincides with the original thermostat being replaced, can you tell us more about how the original thermostat failed?

 

The original thermostat was also an Easy heat warm tiles thermostat. It was a GT-2 model which was the model made prior to the FG one I have in there now.

That original thermostat just randomly failed one day this past summer. The in-floor heating was off and had been for months. One day I walked to find the thermostat's GFCI light on and unable to be reset. I called the company and after a brief discussion/bit of trouble shooting they felt it was an end of life indication.

Interestingly, I have another bathroom in the house that also has in floor heating and the thermostat down there also had an end of life issue around the same time as the other one. It is a 120V system and had a GT-1 thermostat in it...

I changed both systems to the new FG thermostats as they are supposed to be good for either 120 or 240 V.. The 120V system works perfectly while the 240V system continues to have this intermittent GFCI issue.

I will post the resistance of the heating cable while the floor is warm tomorrow when I have some free time.

Thanks again for all your thoughts. I appreciate the help.

 

I could never find a definitive description of what "end of life" actually meant. I was trying to figure out what actually defines the end of it's life. Is it based on "on" time, actual use time or when it fails it just gets blamed on end of life.

CO/smoke detectors now have an EOL timer built in but that is understandable.

It may be interesting to switch the two stats and see if the problem remains.

 

I agree about the end-of-life vagueness. I asked the company about it and was not told much beyond a generic statement.

I have switched the 2 thermostats and the 240V system continues to trip the GFCI intermittently and the 120V system runs perfectly regardless of which thermostat is in which system.

The company has suggested that maybe I should try a thermostat with a 15mA GFCI trigger as opposed to the 5mA trigger on the FG.

 

A bit of an update..

I heated my floor up to 82 F (28 C) today and checked the resistances, usually I run it at 77 F (25 C) but figured a higher temp may help expose a problem if there is one. I used the same digital multimeter set to 2000K during the test.

Heating Cable resistances
a) black to white - 41 ohms
b) black to ground - infinite
c) white to ground - infinite

sensor wire resistance - 9400 ohms

Seems like the heating cable checks out, only thing left to do is get a megger tester i guess??

I checked all the connections/wires in the box and nothing seems to be cracked or an obvious issue. any other thoughts/comments would be appreciated.

 

You could run it until it trips..... and then check it. That could take days.
That leaves a megger.

 

Another update..

I used a 500 VDC megger and checked the insulation resistance:

1) between black and ground - Greater than 550 MOhms
2) between white and ground - Greater than 550 MOhms

It now seems that the heating cable checks out in every way and yet the GFCI continues to trip intermittently. Nuisance tripping seems to be the only possibility. Maybe a change to a thermostat with a 15mA GFCI trip threshold will be the next step.

I will be speaking again with the manufacturer and will let the forum know for completeness sake what they say. Maybe it will help someone else in the future.

 

I would think the purpose of the GFCI was personal safety. 15Ma would decrease safety. I know the GFCI is new but it could still be defective. Maybe try another one?

 

I agree everything about the cable looks to be ok. That strongly suggests GFCI failure even though you have tried 2 units. It would be a long shot but we have heard on this forum of high frequency RF devices (like HDTV tuners) causing GFCI devices to trip. Perhaps you have something like that running at the same time? Again, long shot.

Do they even make a t-stat with 15mA trip threshold? That does not meet the requirement for a "life safety" classification so I can't imagine why they would sell it in a residential format. You could see if the mfr would be willing to RMA your current unit and exchange for a new one on the chance you happened to get 2 faulty GFCI units (e.g. maybe they moved their production to China since you bought the last one...)

 

After reviewing all the information I sent them, the company agreed that the heating cable appears to be functioning fine and it is likely a "nuisance tripping" issue.

They sent me a TH115-AF-240GA to try instead.

Aube Technologies

It appears to be very similar to the FG thermostat I had installed and has the same 5mA GFCI trip threshold.

I am happy to say the floor and thermostat have worked perfectly together since I installed it. Ultimately, I am not sure what the issue was but it is a huge relief to know the cable is not the issue and simply changing the thermostat can be a solution in certain scenarios.

Thanks again for all the help and advice on this forum. Hopefully this can help someone else in the future.

 

Thanks for stopping back and letting us know how you made out.