GFI Circuit Breaker Trip

[DeltaT]

Interesting service call this week. About 6 to 7 year old Dectron pool dehumidification/A/C system that trips the main GFI circuit breaker immediately when compressor contactor kicks in.

Told owner to get electrician out and change circuit breaker cause everything else checks out cept the swollen up Mexican made compressor run capacitor, which I changed out before I tried starting the compressor.

Owner calls back and says new GFI in place. Return to job, start compressor again, runs for about 20 to 30 seconds and GFI trips again. Does this every time.

Here is the layout. Main panel about 50 feet away. Wiring OK in panel, wiring OK at sub-panel at unit. Unit has separate 30 or 40 amp circuit breaker disconnect, which has never tripped.

When compressor runs it pulls about 19 to 20 amps, which is normal. Including blower motor entire unit pulls about 25 amps or so.

Called owner & told him to get electrician out and check GFI cause somewhere the circuit breaker is picking up a short, leak or something causing it to trip.

Any input? Or do you guys have any fancy way of testing the GFI for GFI?

[hvac1000]

A GFCI needs to have a bleed or short to trip.Check all grounds and check the unit for grounding at the compressor. It does not take much to trip the GFCI. Been there on a 220 volt GFCI circuit on a in ground spa years ago. The problem was with the wiring methods on the sub panel ( four Wires). You have to keep the grounds and neutral separate and also check to see if this circuit provleds 110 Volts to anything. A 110 Volt circuit can and will cause problems on the 220 volt GFCI panel.

[keepitsimplestupid]

Look here: http://www.ce-mag.com/archive/06/ARG/webb.htm
Might give you a bit of understanding. GFCI's trip when the difference in currents flowing in L1 and L2 differ by about 5 milliamps. IF and that's a big if, ALL 5 mA were leaking from one side of the line, that's 44K ohms to ground.
If you had 2.5 mA leakage to ground on each leg, them there would be 88 K ohms from L1 to ground and 88 K ohms from L2 to ground. This ASSUMES (ASS U ME) that it's always there and not thermally or vibration induced. It's an exercise to look at 4 mA and 1 mA. Kinda cool, heh?

[jim baughman]

get rid of the GFIC and put a regular breaker in there...go on to the next job and forget about it.

[elkhvac]

Originally posted by jim baughman
get rid of the GFIC and put a regular breaker in there...go on to the next job and forget about it.

Before you remove a safety switch...check with your insurance company see if they will cover YOU in the event some one is killed.

After that if you decide not to have the liability hanging over your head; start isolating some circuits in the system. As equipment gets older they can start to loose the insulation on electrical components. Windings on motors, defrost timers and such are common causes. The condenser fan motor is a good place to start.

The hardest part is to explain to your customer why you have to replace a part that works fine (refer back to paragraph one if needed).

[johnl45]

Does it trip when you push the test button?

[DeltaT]

Originally posted by johnl45
Does it trip when you push the test button?

Yes, it trips every time so it appears to look good. This system has run for the last 3 or 4 years with no problems.

And the NEC requires anything in and around a pool or outdoors requires GFI protection so I am not going to eliminate that protection.

My thought is this. The run capacitor swelled and failed, like most of the Mexican made run capacitors, so I am wondering if the compressor, when running, is now showing some kind of bleed to ground although no resistance is showing on any of the terminals in an off position. The only connection on the load side of the compressor is the compressor contactor and crankcase heater, both prove out OK with no grounds.

So I am asking this question to see what is available to test the GFI under operating conditions to catch why this GFI breaker in the main panel is tripping.

The old breaker tripped immediately. The new GFI breaker at least allows a 20 to 30 second compressor run before it trips out.

My gut reaction tells me the compressor may have an operating short to the oil/ground while running. I hate to condemn a perfectly good compressor on a guess, especially if a new compressor would do the same.

So how in the world do I pick up a 6 or so ma leak while the system is running which, apparently, is tripping this GFI breaker?

This is a standard off the shelve Copeland, I think, about 3 ton A/C compressor.

[wink]

Originally posted by DeltaT
Interesting service call this week. About 6 to 7 year old Dectron pool dehumidification/A/C system that trips the main GFI circuit breaker immediately when compressor contactor kicks in.

Told owner to get electrician out and change circuit breaker cause everything else checks out cept the swollen up Mexican made compressor run capacitor, which I changed out before I tried starting the compressor.

Owner calls back and says new GFI in place. Return to job, start compressor again, runs for about 20 to 30 seconds and GFI trips again. Does this every time.

Here is the layout. Main panel about 50 feet away. Wiring OK in panel, wiring OK at sub-panel at unit. Unit has separate 30 or 40 amp circuit breaker disconnect, which has never tripped.

When compressor runs it pulls about 19 to 20 amps, which is normal. Including blower motor entire unit pulls about 25 amps or so.

Called owner & told him to get electrician out and check GFI cause somewhere the circuit breaker is picking up a short, leak or something causing it to trip.

Any input? Or do you guys have any fancy way of testing the GFI for GFI?

good place to start here.http://www.hvacoracle.com/cgi-bin/publish.cgi?p=306&s=cooling

[wink]

http://www.hvacoracle.com/cgi-bin/pu...=306&s=cooling

[hvac1000]

Disconnect the compressor wiring at the terminals. Run the unit and see what happens to the GFCI. If it still trips then reconnect and check other components.

[elkhvac]

Just last month we had two different accounts call about reach-ins that were tripping GFIC outlets. Both had no other problems, just tripping GFIC. In both cases it was the condenser fan motor. The tech support guy called it "dirty electrical" He said it is fairly common as equipment gets older.

[DeltaT]

Originally posted by elkhvac
Just last month we had two different accounts call about reach-ins that were tripping GFIC outlets. Both had no other problems, just tripping GFIC. In both cases it was the condenser fan motor. The tech support guy called it "dirty electrical" He said it is fairly common as equipment gets older.

But the question still remains is how did you guys find that the condenser fan motors were the tripping cause?

The unit runs correctly until the compressor kicks in. The fan runs, the valves all switch correctly and it is running now and has been for days on the ventilation cycle (uses 100% fresh air in place of compressor cooling).

Doesn't trip GFI when contactor is not pulled in or if contactor is pulled in and compressor terminals are not connected. So it's got to be the compressor but there is absolutely no indicaiton of short to ground, unusual resistance between R/C/S or anything else that I can find. It does run now for about 20 to 30 seconds before the GFI trips and amp draw is normal.

If the start relay or capacitor were defective the amp draw would be off.

The only thing I can think of is the winding in the compressor are leaking to ground through the oil only when power is applied. But I can find no way to prove this and condemming this compressor is something I don't want to do until I have proof.

Somebody somewhere ought to have a testing instrument for a 220 volt GFI circuit to test for problems, I would think. The only test left to me is a megger on the compressor but I have never trusted that method of checking out a compressor.

Got to figure this one out this week. The iron is this customer IS an electrical contractor and he's left this trouble shooting up to me. Funny, huh?

[hvac1000]

GFCI's are very sensitive. You might try to meg the windings to ground and see what the readings are. I no of no other tester available for this purpose.

[DeltaT]

Originally posted by hvac1000
GFCI's are very sensitive. You might try to meg the windings to ground and see what the readings are. I no of no other tester available for this purpose.

Thanks and thanks for all the replys. I think I am going to meg out the compressor - hope I can find a friend who has one - and become a parts changer including changing out the start relay and start cap just to be on the safe side.

Still amazes me that in all these years this is the first system I have had that acts like this. And that the home owner is an electrical contractor. You would think one of his guys could help.

Back to the trenches.

[ambarton]

have you concidered the wireing inside the cabinet. i once saw a transformer that had the high side connections just barely touching water (stopped up condensate line), tripped the breaker every time. it did not have a gfi breaker like your unit but mite be something to think about.

[elkhvac]

Delta-T

The hardest part is explaining to the customer why THEY HAVE TO PAY to replace parts until the problem is found. We know of no better way of testing it. We just disconnect components until we are reasonably sure we have found the culprit. We have not had any compressors cause that problem yet, but I'm sure one could. Try to explain it the best you can to the HO. Don’t guarantee changing the compressor will fix the problem.

Delfield tech support is who first explained it to us. He said “There is no way to test it. Isolate to narrow it down and start changing parts from the cheapest to the more expensive.
You may want to try Dectron’s tech support and see if they have had other compressors do the same.

[rclaoac]

Disconnect wiring to condenser fan and see if GFI trips. If it holds, connnect the fan and disconnect the compressor. Don't change anything until you can prove it's faulty.