I recently encountered a strange situation in which a double-pole breaker had failed in a spectacularly unusual way. I can say this confidently because replacing the breaker solved the problem. What I want to know is what could have possibly caused this mode of failure.

In short: Both legs of the double-pole breaker appeared to be on the same phase. I have no idea how this could have happened because a straight breaker swap rectified (figuratively, not literally) the issue.

In long: Referring to the letters in the picture below, I measured the following voltages using my multimeter:

  • 0V A-H
  • 120V A-B
  • 120V A-C
  • 240V B-C
  • 120V A-D
  • 120V A-E
  • 120V A-F
  • 240V E-F
  • 120V A-G
  • 5V F-G

Subpanel, labeled with letters to identify test points

Of course,

  • A is the Neutral
  • B is the Hot (Leg 1)
  • C is the Hot (Leg 2)
  • D is a blade on Leg 1
  • E is a blade on leg 2
  • F is a screw-down terminal on the double-pole breaker (nominally Leg 1)
  • G is a screw-down terminal on the double-pole breaker (nominally Leg 2)
  • H is the Earth Ground

I tested points D and E as extra validation to make sure the contact springs on the double-pole breaker were in fact receiving power from separate legs. In retrospect, I should have also checked D-F, D-G, E-F, E-G, but I didn't do that at the time.

What could possibly cause the 5V F-G reading?

  • A couple things: a floating wire might take on almost any potential; an AC voltmeter might read a DC value in a funny way. Apr 10 '17 at 17:58
  • @CarlWitthoft I'm not sure what you're implying with that statement... I'm especially unclear about the DC part, unless you are talking about a DC offset, which should be the same for both wires anyway
    – Hari Ganti
    Apr 10 '17 at 18:35

Your double-pole breaker is not switching both legs simultaneously as it should be.
The top D-F pole is connected, while the bottom E-G pole is not connected.

Whatever 240V appliance you have connected to that circuit is feeding the 'B' phase from D-F back into your G terminal.
Even with no appliance connected, capacitive coupling between the wires in that cable will fool your mutimeter.

If you had a load to neutral from both poles you would see 120V on one and 0V on the other.

  • I'm not quite sure how the supplied device is wired internally, but it would seem strange to me that it could backfeed one of the wires. That said, an easy way to test is with a continuity test on the breaker itself. I can check if it switches both legs on or not.
    – Hari Ganti
    Apr 10 '17 at 18:37
  • Well something as simple as a water heater would very easily backfeed into the dead phase as the heater elements are just big resistors. YMMV with other devices, but if it has no neutral connected (as your setup seems to be) you're almost certain to measure something on that dead phase when the other one is powered. Your multimeter has a very high resistance and even the teeniest tiniest trickle of current can cause it to show you a significant voltage present.
    – brhans
    Apr 10 '17 at 19:36
  • Good point, I forgot that the impedance of the multimeter prevents truly loading the circuit to see it's "actual," under-load voltage. The device is an EVSE, which shouldn't even operate without the correct voltage applied, unlike a simple resistor.
    – Hari Ganti
    Apr 10 '17 at 19:38
  • We see back feed voltages all the time everything from transformers, to clock motors causing the voltage on the dead leg, add inductive coupling and it is quite common.
    – Ed Beal
    Apr 10 '17 at 20:11
  • Thanks @EdBeal. I also ran a continuity test on the breaker itself and found that this answer is correct. One of the breakers on the double-pole was not switching correctly.
    – Hari Ganti
    Apr 11 '17 at 0:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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