Fridge trips GFI


  #1  
Old 03-04-07, 02:13 PM
K
Member
Thread Starter
Join Date: Dec 2004
Location: Indianapolis IN
Posts: 33
Received 0 Votes on 0 Posts
Fridge trips GFI

I have a Frigidaire freezer in the garage. Recently I opened it and noticed no light - well, I was busy getting ready for a business trip, so I figured I would replace the light when I got back.

My wife calls me as I am on the plane waiting to take off for my business trip the next day. She had found the whole circuit was out, and our chest freezer on the same circuit was starting to thaw. Ain't that the way it always works So I walked her through unplugging things in the garage until the GFI stopped tripping. It was the fridge.

To make the rest of the long story shorter: when I got home, I discovered that when I unplug the defroster element, my GFI doesn't trip. So I removed it from the coil, and discovered that when the defroster element touches the evaporator coils, the GFI trips, but otherwise everything seems fine.

The thing I can't figure out is why this would ever *not* trip the GFI. The evaporator should be connected to ground somewhere down at the compressor, so how could this element be touching the evaporator coils and not basically be shorting to ground?

The element is bare metal, something like what you would expect from an oven heating element. It says it is 475 watts, which jibes with my measurement of 28 ohms. Is there perhaps supposed to be something covering this element that keeps it from shorting to the evaporator?

Thanks for any help.
ken
 
  #2  
Old 03-04-07, 02:56 PM
DaVeBoy's Avatar
Banned. Rule And/Or Policy Violation
Join Date: Mar 2006
Posts: 1,662
Received 0 Votes on 0 Posts
Don't forget that compressors float on rubber mounts.

Refrigerators plugged into GFCI's can be touch and go the way it is. Not the best of idea due to motor kick on extra draw. And if the GFCI is 15 amp rather than 20 amp, that could be even more potential grief. It is done, and I have units that are plugged in GFCI's, but I've also got calls to rentals where the GFCI tripped. Luckily it gets caught by someone before the food spoils.
 
  #3  
Old 03-04-07, 03:10 PM
K
Member
Thread Starter
Join Date: Dec 2004
Location: Indianapolis IN
Posts: 33
Received 0 Votes on 0 Posts
Hmm, I see what you mean. However, this couldn't be motor overcurrent. I have connected it currently so that only the defroster has current - the compressor is disconnected.

So the only thing that could be causing this is shorting of the defroster element to the evaporator. I moved the defroster element so it wasn't touching the evaporator, and no GFI pop. As soon as I touched it to the evaporator coil, pop goes the GFI.

Are you saying that the evaporator coils should be electrically floating? I wonder what could cause them to suddenly have low enough impedance to ground to cause this?

ken
 
  #4  
Old 03-04-07, 03:19 PM
DaVeBoy's Avatar
Banned. Rule And/Or Policy Violation
Join Date: Mar 2006
Posts: 1,662
Received 0 Votes on 0 Posts
Plug the thing into a regular outlet to take a weak GFCI out of the equation. If the circuit breaker blows, you know you have a problem for sure.

Floating, response = no. You said earlier that the evaporator coil and compressor are connected and could be grounded through the compressor and I said that the mounts on the compressor may cause the compressor to float some and not make real good ground contact. This is just a theory at this stage on my part.

Why are you getting the defrost and evaporator to touch?

But what this sounds like the problem could be is the outer lining of the defrost coil is now carrying current, which it shoud not. Just like how a Calrod burner of a stove should not carry current in it's outer metal skin. It means the element is shot and needs to be replaced.
 
  #5  
Old 03-04-07, 03:35 PM
GregH's Avatar
Super Moderator
Join Date: Jan 2001
Location: Manitoba
Posts: 10,191
Received 51 Votes on 48 Posts
Ken,

It is likely that the defrost element has shorted internally to ground.
It will not show up as a dead short because when the element is internally touching the metal outer wrap of the heater there will resistance from the element.
If you were to disconnect the wires from the heater and measure the resistance of one of the terminals to the metal on the heater you will find some resistance. Not zero ohms as if it were a dead short.
 
  #6  
Old 03-04-07, 08:58 PM
K
Member
Thread Starter
Join Date: Dec 2004
Location: Indianapolis IN
Posts: 33
Received 0 Votes on 0 Posts
OK, I will get the extra long extension cord out. It's worth checking. But...

My current working theory is exactly what you are saying - the defrost coil should have a very high impedance coating of some sort, and something has worn through. I don't think it's a direct short, but leakage - the digital ohmmeter was flashing some values back and forth that I can't quite read. My analog VTVM is packed away and probably has some leaking caps after being so many years in storage anyway, so I can't really tell what the resistance is here.

If this is all true, the GFI is a *good* thing here. I could waste a couple of amps all day as far as a circuit breaker cares, but hot side leaking a few milliamps to ground will pop that GFI. Not to mention the safety hazards.

I'll let you know if this works.

ken
 
  #7  
Old 03-05-07, 02:53 PM
DaVeBoy's Avatar
Banned. Rule And/Or Policy Violation
Join Date: Mar 2006
Posts: 1,662
Received 0 Votes on 0 Posts
Impedence in resistors is normal, but from your OP, impedence is not your concern; a short is. The way elements are made is that the current flows through a wire that is inside what you visually see as the defrost element. That inner electrical wire and the outer metal of it should be separated, physically and electrically. But it is apparent to me, from what you say, that your outer metal is being energized... and when it contacts some other metal in the refrierator it trips out the GFCI.

If you were to run an ohms check between the inner wire inside the defrost element and the outer metal of it...if you got any reading at all on your meter (on the ohms test), there is a short between the inner electrical wire inside the defrost element and the outer metal of the defrost element.

Elements on stoves/ovens can be tested the same way. Those should not have current leaking to the outer shell of them either.
 
  #8  
Old 03-05-07, 05:31 PM
K
Member
Thread Starter
Join Date: Dec 2004
Location: Indianapolis IN
Posts: 33
Received 0 Votes on 0 Posts
I agree. I didn't realize the wire was actually *inside* of the element - I thought the element was the wire. This makes perfect sense, and is exactly what I am seeing. It's not exactly a short, but it's closer to that than an open.

I will be ordering a new element and let you know how it goes.

ken
 
 

Thread Tools
Search this Thread
 
Ask a Question
Question Title:
Description:
Your question will be posted in: