I'm currently doing some renovation in my bathroom, and while the walls are open, running updated cable where I have access. This also includes the addition of a GFCI outlet. It's a 1940s home, and the bathroom and its wiring are original. I've lived here for 17 years, and just purchased it in December. Now that I own it, I am starting to correct all of the issues I know of.

With the intention of having a dedicated circuit for the new outlet, I went all the way back to the breaker panel with 12/2+G Romex. Currently, this circuit has many branches throughout the house, and the intention is to ultimately isolate the bathroom when I free up panel space. The breaker panel is full, so a simple "new circuit" is not immediately possible. The new 12/2 was installed at the first "half" of the circuit so far - forming the "trunk" so to speak - and all but 2 of the branches remain unchanged.

But the issue at hand is that one of the outlets on the circuit failed the day after I ran the new cable. EDIT: The dryer and heater would not work, and there were no audible or visible signs of a problem, nor a burning smell. Due to the fact that the current wiring is an absolute mess, I have drawn a schematic with more specifics to illustrate the current configuration rather than trying to fully explain. I am aware of all of the code (and plain stupidity) related issues that are present, and intend to correct them and reallocate circuits incrementally once I finish the bathroom renovation, and then map out the rest of the wiring.

At the end of one of the branches in the basement, is a receptacle that has my clothes dryer plugged in. I work from home, and my office (with its own circuit) is also in the basement. The basement is unfinished, so we have areas separated by shelving or furniture. 2-3 months ago, I added a space heater (set low), just to take the edge off on colder days, figuring while the dryer is not running, the heater set low would be fine there. Note #1: I have another space heater on my office circuit. Note #2: I was not aware that the laundry area was not on its own circuit until I started this. Regardless of these facts, this setup has been working without issues since I started using it. So, the question is that how, after 2+ months, did the old receptacle fail, and when replaced with a brand new one, fail again?

Also... The breaker did not trip. All other outlets are working fine, and the new bad receptacle is reading ~36v to neutral, and ~40v to ground. I get ~126v at every other known branch on the circuit. My only assumption is that the lower resistance of the 12/2 changed the dynamics of the whole circuit somehow?

  • What do you mean when you say the outlets "failed"? Did the breaker trip? The GFCI trip? Did it melt?
    – mmathis
    Mar 23 '17 at 21:15
  • @mmathis Thanks for your reply. By "failed," I mean stopped working properly. From Paragraph 4 above: "At the end of one of the branches in the basement, is a receptacle that has my clothes dryer plugged in." From the linked drawing (thought it was also above, I will update in paragraph 3): "The day after I updated wiring, the (+15 yr old) receptacle failed quietly. Metering yielded ~36v." From the last paragraph, above: "The breaker did not trip. All other outlets are working fine, and the new bad receptacle is reading ~36v to neutral, and ~40v to ground."
    – Jay
    Mar 23 '17 at 21:35
  • Simply replacing #14 wire with #12 wire will not cause any problems. Something in your installation went awry. Can you post pictures of your breaker panel and each of the boxes that contain new wire?
    – mmathis
    Mar 23 '17 at 21:39
  • @mmathis I didn't think the swap should have any effect. Yes. I can open up the boxes and get pictures this evening. Also, in case it was not clear based on my text alone, this receptacle is part of an old branch without the new wiring. The new wiring is only up to the 3rd of 5 junction boxes along the path to it. Thanks.
    – Jay
    Mar 23 '17 at 21:46
  • @mmathis OK. No need to continue. I found the issue, and it was a fluke. Turns out the issue was in a box that I hadn't opened until after the first failure. Checking the junction must have re-closed the circuit inside, leading me to believe the outlet change solved the problem. Outlet was fine. Clothes are now drying. Should I delete this thread altogether, or answer it myself with the troubleshooting process?
    – Jay
    Mar 24 '17 at 2:09

The short version of the solution is: when you see a very old junction that is taped together, unwrap it completely (while the circuit is off), and inspect the junction directly. Then, fix as needed, and replace the tape with wire nuts. The new upstream wiring had nothing to do with the actual problem, and the wiring inside the old downstream junction box must have shaken loose during demolition. Troubleshooting it the first time resulted in the accidental rejoining of the wires within the box, restoring continuity, and creating the illusion that the outlet was the problem. The first repair worked out on a fluke.

How I missed It: The issue was not with the outlet, but inside a junction box that was ruled out since it was not accessed until the problem with the receptacle arose. While the circuit was off, I removed the old receptacle and opened up the junction boxes between it and the panel. After restoring power to the circuit, testing the bare wire immediately yielded ~126v. So I shut it back off, installed a new outlet, closed the boxes and restored power. Done deal.

Three days later, no power again. Actually, very low power. After the second time the outlet stopped delivering proper voltage, I decided to diagnose the original receptacle before removing the new one. It was fine.

Checking for continuity and shorts with a meter: Disconnect and remove the outlet. Set your meter to resistance/ohms. This is a little more reliable than continuity (if you have that), because you can have a resistive short that may not register as full continuity.

Test the receptacle in groups:

  • Neutral Group = long slots & silver screws
  • Hot Group = short slots & brass screws
  • Ground Group = round holes & (usually) green screw on the mounting frame

Touching the probes to any two points within a group should read zero or nearly zero ohms. High readings or no response indicate a dirty, loose, or broken internal connection. Any points from two different groups should yield no response from your meter. If you get any reading, there is a short.

Finding nothing wrong with the original receptacle brought me to the realization that something else happened with the wiring. I removed the new one to confirm that the bare wires still read low. Working backward, the junction nearest to the the receptacle has an old porcelain light/power socket combo. Testing that power socket also yielded full power, so the problem was absolutely with the one cable between that box and the actual receptacle. Opening that box up immediately gave me a reading at the trouble spot. It became obvious that the issue was a loose hot connection in an old junction box hidden by tape.

One final tip... Testing a receptacle/wiring on a confirmed working circuit that is on. If you get a reading on your hot (black) to ground (bare), but nothing on your hot to neutral (white), then your neutral leg is open. If you get no reading from hot to either one, you most likely have a break in your hot leg. If you get a reading on hot to neutral, but not hot to ground, you have an open ground. In my case, I got about 36v to neutral, and 40v to ground. Ruling out the outlet, and getting nearly the same result on both made it rather obvious that a hot wire was loose somewhere.

  • 1
    Anytime you see voltages other than the expected (34V, 19V, 105V) that is almost always capacitive coupling, which is basically the wire rattling, i.e. picking up "60 cycle hum" via radio from the wires it's parallel to. That is not capable of driving any load, and putting even a tiny load on it like a night-light will make the phantom voltage disappear. The only exception I've ever seen is when a guy found 208V on 120/240V circuit, but he was in a shop with three-phase power (wild-leg). Mar 24 '17 at 17:00
  • @Harper Thanks. Having worked with small electronics, that makes total sense. Due to the age of the receptacle, I just thought the contacts must have been corroded (the first time), and I was getting mild arcing or some type of capacitance through the corrosion.
    – Jay
    Mar 24 '17 at 18:32

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.