jeff43222
Senior Member
- Location
- Minneapolis, Minnesota
I just got back from a service call where the initial problem was that the homeowner found sparks jumping between a strap connected to a metal drain pipe and some EMT, and he also said that the lights were dimming. He separated the strap and the EMT and wrapped the EMT near the strap with electrical tape, then called me in.
My voltmeter showed about 30V between the drain pipe and the EMT (as well as 30V between the shower faucet and the drain), and I managed to duplicate the sparking when I shorted the two with a jumper wire. When I did that, the lights went from dim to full power. While I was scratching my head, a roommate came home and turned on a light upstairs, making the lights even dimmer, and when I checked the voltage between the drain and the EMT again, this time it was more than 50V.
I searched through the entire house and didn't find any visible connection between the plumbing and the wiring. Only one circuit is involved; the stray voltage disappears when the fuse is pulled.
What puzzles me here is how shorting the drain with the EMT results in full power back to the outlets. If there's a hot wire touching the drain somewhere causing a high-resistance short, wouldn't shorting the stray voltage result in less voltage to the outlets, rather than more?
My working theory at this point is that a neutral is touching the drain pipe somewhere and is preventing some of the current in the circuit from returning to the panel. Shorting the drain and the EMT provides a complete path back to the panel, so all the lights go back to full power. Alternately, I'm thinking that maybe an isolated neutral is inducing voltage on the drain pipe. It wouldn't surprise me if this is the case, as I saw plenty of hack wiring in the house, including at least one case of a lone current-carrying conductor entering a device.
Anyone want to weigh in on my theories and suggest their own?
In the meantime, I'd like to make the situation safer at least temporarily, so I was thinking that bonding the drain pipe to the grounded cold-water pipes or the grounded EMT would work. Long-term, my thought is that it's probably not worth trying to find the source of the problem, as that would likely involve tearing open a lot of walls (the circuit has plenty of outlets in the basement, first floor, and second floor). I told the homeowner it's probably best to cut the circuit at the source and start over with new wiring to replace the stuff that gets cut.
My voltmeter showed about 30V between the drain pipe and the EMT (as well as 30V between the shower faucet and the drain), and I managed to duplicate the sparking when I shorted the two with a jumper wire. When I did that, the lights went from dim to full power. While I was scratching my head, a roommate came home and turned on a light upstairs, making the lights even dimmer, and when I checked the voltage between the drain and the EMT again, this time it was more than 50V.
I searched through the entire house and didn't find any visible connection between the plumbing and the wiring. Only one circuit is involved; the stray voltage disappears when the fuse is pulled.
What puzzles me here is how shorting the drain with the EMT results in full power back to the outlets. If there's a hot wire touching the drain somewhere causing a high-resistance short, wouldn't shorting the stray voltage result in less voltage to the outlets, rather than more?
My working theory at this point is that a neutral is touching the drain pipe somewhere and is preventing some of the current in the circuit from returning to the panel. Shorting the drain and the EMT provides a complete path back to the panel, so all the lights go back to full power. Alternately, I'm thinking that maybe an isolated neutral is inducing voltage on the drain pipe. It wouldn't surprise me if this is the case, as I saw plenty of hack wiring in the house, including at least one case of a lone current-carrying conductor entering a device.
Anyone want to weigh in on my theories and suggest their own?
In the meantime, I'd like to make the situation safer at least temporarily, so I was thinking that bonding the drain pipe to the grounded cold-water pipes or the grounded EMT would work. Long-term, my thought is that it's probably not worth trying to find the source of the problem, as that would likely involve tearing open a lot of walls (the circuit has plenty of outlets in the basement, first floor, and second floor). I told the homeowner it's probably best to cut the circuit at the source and start over with new wiring to replace the stuff that gets cut.