I have a practical problem in E&M which has baffled me and and my electrician. The copper water pipes in my house are carrying a current of small potential (~1V) AC current. Here is my hypothetical circuit diagram:

         R1     R2
|           |        |
| R3      source   ground
|                    |

Where o is my shower head and ground is actual ground from which the supply pipe enters the house. R1 and R2 represent the assumed (low) resistances of the copper piping. The way I detected the current was while changing a light fixture above the shower head and detecting a flow from my showerhead to the grounded junction box through my forearm. This current produced a mere tingle in my arm and has been confirmed with two different multimeters: my "toy" and the electrician's Fluke. R3 represents my arm, an intermittent connection to ground.

There is also a 2m copper ground spike embedded in the outside earth connected to nothing which can be used as a reference. The common/neutral line of the breaker box is currently tied at R2 near where the municipal supply line enters the house's basement. The source is unidentified but assumed since there is a measured, felt potential on the copper pipe.

If the circuit is how I have drawn it, I believe the following would be true following Maxwell, Ohm, and/or Volta:

  1. in the typical case with R3 disconnected all current goes from source to ground via R2
  2. while my arm is touching the pipe and the grounded junction box, most of the current goes from source to ground via R2 while a small current goes through R1 and R3 where R3 is presumed to be of much higher impedance than R2.

Is this a correct interpretation of how the hypothetical circuit would behave?

For the curious, the house is 80-ish years old with some olde-fashioned knob-and-tube (poorly insulated) wiring inside the walls that we guess could be in contact with the plumbing and thus be presumed current source.


2 Answers 2


Yes, this would be a fair assessment of how the circuit is probably working.

As a bit of an advisory, do not use the shower until you have traced the source of this!. It takes about 1mA for you to feel a tingling. Assuming it was dry when you were changing the light, then your resistance would be around 1-20kOhm in a damp environment (like a shower). This means you were experiencing a voltage around 1-20V. You claim you measured it at 1V. That means with dry skin, you have a resistance around 1000 Ohm. Wet skin significantly reduces this and you could find yourself with a resistance around 100 Ohm while taking a shower. That would put the current going through you at 10mA at least. This level of current can be extremely painful and potentially life-threatening if it crosses your heart.

One common thing that isn't always taken into consideration is that with old houses, sometimes part of the plumbing is replaced with PVC or other non-conducting materials. This breaks the ground connection. You should check you plumbing. If you find something like this, create a current bridge and that should solve the problem. Also, watch for dim/flickering lights, appliances not acting properly, and outlets that don't work or don't work well. Any of these could indicate the source of the stray electricity.

In the end, if you follow the pipes from the water main to your shower, you should find the problem. As a temporary fix, you could try connecting the shower head to a (insulated) ground line. Note the word "temporary" as this could drain a lot of power over time.

  1. Rewire the house properly with modern cable. Knob and tube is obsolete and well beyond it's design life.
  2. Either pay your baffled electrician the going rate to actually trace the problem OR get a better electrician.
  3. Don't attempt to use this forum to solve circuits involving your arm and water/showers. Concentrate on getting a safe electrical system in your house. Theories and diagrams here should be WAAAAAYYYYY down your list.
  • $\begingroup$ I'd have to agree that the advice to rewire with modern wiring is the best in the long run. When you start having to "decode" the wiring that's potentially lethal, it's time to recognize the danger. And I'd note @Jim 's remarks about potentially lethal current in this context. $\endgroup$ Apr 12, 2017 at 0:15

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.