Why do I measure voltage on a breaker that's off?

My breaker #1 (115VAC) is connected to my dishwasher. My breaker #2 (115VAC) is connected to my garbage disposal.

Why do I get 38.5VAC on the dishwasher line, even when breaker #1 is off? When breaker #2 (Garbage Disposal) is also off, I get 0 VAC on dishwasher. Why?

• Does this answer your question? Unexpected voltage with switch turned off Feb 25 '20 at 16:41
• Please do not use all caps on public forums. It is hard for most people to read even if it might be easier for you. Feb 25 '20 at 16:42
• @jphi1618, there are a lot of phantom voltage answers with accepted answers I would choose one of those but phantom voltages are common in most wiring methods where parallel runs of cable in close proximity, multi wire branch circuits have required handle ties for decades so if one was off they both would be off, even a short section of cable in close contact is enough to induce a voltage on a non energized conductor for the answers below.+ Feb 25 '20 at 17:29
• agreed Ed. A while back I was working (helping) in a commercial building with long runs in conduit. While the circuit I was working on was clearly turned off, there was still enough induced current to give enough of a shock that I couldn't work on the circuit without rubber gloves. I put an incandescent bulb on it just to make sure it was just induced voltage (it was bc voltage when to zero). Just saying this bc induced voltage can be a bigger issue in some instances. Feb 25 '20 at 17:44

Almost certainly an induced voltage. Is this part of a Multi Wire Branch Circuit (MWBC)? If you have black, red & white wires in the cable it's probably a MWBC. Common to have induced voltages in them.

You have a multi-wire branch circuit feeding these two appliances. This means that it's a single /3 cable, with two hots and a shared neutral. Since the two hots run together in the same cable, you are getting capacitive "crosstalk" between the two wires. This is very weak, and you can only see it because you're using a DVM. DVMs are very sensitive.

The bigger problem is the breakers

And the biggest risk is in phasing the multi-wire branch circuit incorrectly, which will overload the neutral wire. The surest way to get that right is to use a factory provided handle-tie, either built in to a 2-pole breaker, or made by the factory to tie two individual breakers. The handle-ties are keyed so they won't fit on the breaker in a mis-phase situation. 2-pole breakers are much easier to obtain. Note that you will need a 2-pole breaker if you ever want to install GFCI protection.

Second, it should not be possible to turn only half the circuit off. That is precisely to protect you from getting nailed like you almost did. The 37 volts wouldn't have harmed you, but if you interrupted the neutral, that would've! Bigtime. Interrupting a neutral on a live circuit can kill you, that's why neutral wires have insulation.

So, when you use a 2-pole breaker or factory handle-ties to tie the two breakers together, that takes care of that also.

Again, the reason not to tie them with a nail is because a nail will let you tie, say, the two halves of a duplex breaker, which is something you should not do. That would cause precisely the overloaded-neutral scenario I mentioned.

• Harper, what do you mean by "interrupting a neutral..."? Feb 25 '20 at 20:26
• Hey Harper: Great, thorough resp. as usual. But I have one issue with this resp. and others that refer to phantom voltages as "capacitive". I believe the correct term is "inductive". Capacitors can store power, transformers can't. Wires running very close to each other act like transformers inducing (hence the term inductive) power from one wire to another via electromagnetism. Feb 25 '20 at 22:34
• @GeorgeAnderson yup, I used to think that too, until I was corrected by many people here. They may all be wrong, but I have not been willing to sit down and do the research to prove it. Feb 25 '20 at 22:58
• @clwhoops44 I mean say, there's a neutral in the box coming and going on a wire nut, so you need to add a pigtail wire so you separate the wire nut to add the wire. You just interrupted neutral. Worse if the supply and onward neutrals are on the two receptacle screws, which they better not be in a MWBC, and you go to change the recep. Obviously what's in your head is "They're neutrals, they're harmless" not at all. Feb 26 '20 at 2:23
• @clwhoops44 correct, the pigtailing is only required where 2 or more hots are present. Feb 26 '20 at 16:48

In a lot of homes in the US, the circuits for your dishwasher and garbage disposal are fed to a single duplex receptacle under the sink, which has the two hot wires coming from separate breakers to the separated receptacles, but they share the same neutral wire (called a Multi Wire Branch Circuit, or MWBC). What you are reading, because you most likely have an inexpensive digital meter, is just the capacitive coupling though the motor windings and starting capacitors of those motors because they are sharing the neutral. If you had used an older meter with an inductive coil on it or a more expensive digital one that has proper filtering, the meter burden would have taken that voltage reading to zero. Bottom line, don't worry about it.

• I must disagree with part of this answer. There's no way that "coupling though the motor windings and starting capacitors of those motors " would create a phantom voltage. The phantom voltage is generated simply because wires running in very close proximity on AC current are acting like a transformer. Simple electromagnetic principles. I do agree that highly sensitive meters can mislead a lot of people. Even a tiny "load" across the measured wires can help interpret the readings correctly. Feb 25 '20 at 23:45