Just came from an interesting service call. It was a halfway house for veterans that I have worked at before. They have a number of duplex units built in the 50's, no equipment grounds at any receptacles.

Anyway a resident complained of shocks received off the refrig. When I got there they described the shocks as a snap accompianied by a blue spark. They get them whenever they touch anything metal and also the phone and the thermostat. Well I told them it was static electricity, but I would check the fridge anyway.

Well when the fridge is running, it will light my non contact tester when held close to the cabinet. So I got my analog meter and checked from a part of the fridge that was not painted to the grounded gas line nearby. I got about 13 volts. Remember this fridge has no equipment ground with the old 2 wire system.

So now I am thinking, in addition to the static shocks, they could've gotten a real current shock off the fridge. Is the 13 volts excessive? As soon as the fridge shuts off it goes away. I told them, short of running a new circuit, the only thing to do is get a new fridge. I know if I put in a GFCI it will not hold.
I know the code requires refrigerators to be grounded, but if they can't pay for a new circuit, what else is there to do?

You are seeing the real reason refrigerators trip GFCIs. You have a short inside the compressor. If it was grounded it would quietly arc and spark inside the can in a freon bath until it finally quits.

I've run into "hot fridges" many times. Once you ground the fridge, the casing will no longer be hot. Thanks for the explanation gfretwell....I've always wondered how a compressor can have a short and energize the casing, and yet when you ground it, it not only solves the problem but the breaker doesn't trip.