I'm measuring the leads on a light switch that's wired behind the load side of GFCI.

Somehow I'm getting 45 Volts between Hot and a metal faucet, 120 volts between hot and ground, and zero volts between ground and the faucet.

What could this be caused by?

  • 1
    I suspect your metal pipe isn't grounded. Whether it should be depends on location amd other factors. What sort of meter are you using (a digital multimeter I reckon)? If it has a resistance mode, what do you get (how many ohms) if you connect it between ground and the faucet? Don't try to measure resistance from hot to anything.
    – Chris H
    Commented Jul 2, 2023 at 8:11
  • I don't think this pipe is grounded, no, but the house is piped using copper piping which in theory contacts the ground at some point? Given that they touch the same ground, why would the voltage measure differently?
    – Dong Shi
    Commented Jul 2, 2023 at 9:19
  • 2
    My house is also all copper. The water comes into the house with a plastic pipe.
    – crip659
    Commented Jul 2, 2023 at 10:48
  • @crip659 same here. In some places they're cross-bonded, others not
    – Chris H
    Commented Jul 2, 2023 at 12:26
  • That just sounds like the pipe is floating, like not connected to anything. Get a drone and tell it to hover 3 feet off the ground. What's the voltage between hot and the drone? Who knows, they're not connected, they have no relationship. A cheap DVM will give wacky readings in that case. It's phantom voltage, it means nothing (except that they have no relationship). Commented Jul 2, 2023 at 19:42

3 Answers 3


In my neck of the woods (Washington state), all metal piping, such as the hot water line from the water heater, the cold water line, any gas lines need to connected with a run to the grounding bus bar in the main panel, which also of course has to connected to a couple of ground rods at least 6' apart or a Ufer ground.

OK, on to your question: As some others have suggested it could be phantom voltage. to test put an amp clamp style meter over the pipe and see if it's actually drawing any real current, if so, that's a problem and dangerous.

As to the cause, somewhere in your plumbing system you lost ground. Perhaps by a repair of a broken pipe using PEX. If so, the plumber should have known that a ground wire would need to be connected of either side of the PEX repair.

  • Accepting this answer as it includes a way to test for safety!
    – Dong Shi
    Commented Jul 6, 2023 at 0:36

Those readings are what I would expect from water pipes that have a poor or no ground connection. In the old days, connecting to a water pipe was the thing to do but now there are driven ground rods. Even though you might have copper piping, the main line might not be or there might be a non metal connection at your water meter. Your pipes then would not be deep enough for a good ground. In my sub panel that feeds my house, I have a ground wire from my main breaker panel and ground rod. Then the electrician ran a #6 copper wire from my sub panel ground bus to my copper water pipe. It's questionable if it's to code but they did it.

  • Just wondering if you have a water system that goes into dirt/ground and an electrical system with ground rods, but both systems are not bonded/connected to each other except by the dirt, will the voltage be different.
    – crip659
    Commented Jul 2, 2023 at 12:35
  • @crip659 My house is on a slab on top of coral rock fill dug up to form a lake. The pipes are in that fill. The ground wire went from my sub panel to the inlet copper pipe on my water heater. The voltages were the same. I did disconnect the ground at the pipe when I replaced the heater and the voltage from hot to water pipe dropped significantly. The fill probably wasn't a good enough ground. I did reconnect the ground to the water pipe.
    – JACK
    Commented Jul 2, 2023 at 13:30

As mentioned, this is normal, if mains cables are running close to the pipes. The voltage may sound scary, but it is not dangerous because the current capacity is very very low. Anything that will have a water connection will either be grounded or be double-insulated, so the pipes are unlikely to come into contact with mains.

In my case, I had 100V between steel beams and neutral in my previous house built in 1995. We have 230V mains, neutral, and ground running in TPS cables. I was concerned about interference with electronics and got the Electrician to strap them. It was unusual then to strap beams in residential houses. It is more commonplace now.

In your case, the voltage on the pipe is unlikely to affect any electronics. I have not grounded my copper pipes in any house, The running water does take away some of the buildup. And of course, entry to the house is via PVC pipes.

  • No. The OP is seeing capacitance at work, not induction and voltage is not being induced on their pipes. This has nothing to do with the pipes running near wires. Large conductive objects (including long wires) have enough capacitance to let a tiny amount of current flow - enough for a very high-impedance digital voltmeter to register voltage relative to a hot wire, but not enough to do anything useful (like even light a 4W incandescent nightlight bulb). That’s why they’re not seeing any voltage difference between the pipes and neutral or ground.
    – nobody
    Commented Jul 3, 2023 at 2:05
  • Capacitance is the same reason those old-style neon bulb voltage testers work: one side of the bulb is connected to the test probe, the other side to the user’s body. The user’s body has enough capacitance to let enough current (in the realm of 1 mA) flow to light the bulb, even if the user is not directly connected to ground or neutral.
    – nobody
    Commented Jul 3, 2023 at 2:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.