A customer from a row home calls me and tells me that he received a good shock, while in the shower, as he was adjusting the shower head. He also tells me he smells something burning in the basement. I will just state the facts, to keep this as short as possible. Facts: 1st day 60-volts from soil pipe to cold water pipe. No-hub fittings on soil pipe are melting and stainless clamps are glowing. Turn off all breakers, one at a time, still getting voltage. Out back of property, test down spout, that is emptying into soil pipe, to cold water spigot get 120v with all breakers off. Service is 100a cable, sleeved in non-grounded emt, 100-a cable is twisted and smashed, in a lb, going into the house. Most house's on the block are brick, stucco, with a 2nd floor bay, that is AL siding, siding is in contact with the stucco wire and stucco, all windows and barge boards are capped in AL. Conclusion: Service cable is in contact with stucco wire, traveling to siding? Going to replace service cable on 2nd day. Facts: 2nd day First thing,test down spout, no voltage. Neighbor next door asks to check a breaker that has tripped, reset breaker. Back to next door, disconnect one leg at taps,get 120v, disconnect other leg, still get 120v, disconnect neutral and no-hub fittings catch fire, re-connect neutral. Fire dept. and PECO on there way! Peco tests neutrals in both houses, everything is fine. What could it be?
Peco said that the neutrals were ok on their side, still getting 120v on down spout. I do not know of any new cable tv installs on the street. All services are overhead, one feed along the back of the houses. I have the answer, I was trying to see if anyone could guess what it was.
I think what you have is a source that has a bad neutral connection and is using any available ground to provide a neutral reference, hence a large amount of unbalanced current coming thru the grounds to get to a neutal. The cast iron sewer pipe can provide a ground.
I think that the one breaker seems to control the problem is a major hint. The problem almost has to be on that circuit.
Enough current to melt the no-hub connectors? I'm impressed. I suspect that neighbor has a few problems. Here are some ideas: First, I'd find out what's on the circuit. Then, I'd see if the problem persisted with every appliance unplugged. This removes the appliances as the source of the problem. (I don't think they are, but let's stop guessing, and be systematic). The next step is finding the place in the circuit where the hot contacts a pipe, the conduit, or whatever. A cheap megger is a good tool to use to check the wires.
You notice I am not thinking "neutral." Not, at least, if the problem persists without any appliances plugged in. Since you claimed to measure 120v at one point, I don't think that current is flowing through any load.
Once you find the short to ground, you've found half the problem. The other half? Grounding. I suspect that the neighbor has a good water bond ... it might even be to the sewer line. I also suspect that your customer has a poor one. I doubt either house has a ground rod.
Keep this in mind: The only way the piping can become energized is if there is no path for the electricity to get back to the PoCo transformer.
Walrus, from the OP's description, I was making the guess that there was arcing between the two pipes. Heck, since those connectors typically have a foil-like wrapper, the clamps and foil may be the primary current path.
Too bad most amp clamps aren't big enough to go around a sewer line!