Today my mother switched her portable air conditioner from “fan†to “cool†mode and immediately lost power on the circuit it was plugged into. But it wasn’t a flipped circuit breaker.
<!—-><!—-> <!—->
She asked the apartment manager to look, and he said something about having had a similar problem in another unit, and he would probably need to replace the GFI outlet in the bathroom (which is on the same circuit) and the circuit breaker.
<!—-> <!—->
I went to her apartment to try to fix the problem. It’s a one-bedroom apartment that was probably built in the late 50’s or 60’s. It has modern copper romex-style wiring.
<!—-> <!—->
My inductive voltage tester showed that each outlet was live, but nothing plugged into the outlets worked. I pulled the extremely old bathroom GFI outlet. There was only one 14/2 romex cable in the box, so nothing was wired to its load side. I removed it so I would know it could not possibly be the source of the problem. But that didn’t solve anything (I didn’t think it would).
<!—-> <!—->
I also temporarily swapped the wire in the circuit breaker with that of another (working) breaker to see if the breaker was at fault. Everything on the other breaker continued to work fine, and everything on the circuit branch in question continued to malfunction exactly as before (I then put the wires back in their original breakers and re-installed the bathroom GFI outlet).
<!—-> <!—->
So I went home and got some more equipment. I plugged my receptacle tester into several outlets on the branch in question and it revealed that the hot/ground were reversed on each.
<!—-> <!—->
I pulled the outlet the portable A/C was plugged into, and it was fine, but the ground was live and I found no volts across the hot to the neutral, but 119v across the ground to the neutral.
<!—-> <!—->
I thought maybe a ground had come loose somewhere in the circuit and came in contact with a hot. But it seems to me that should trip the breaker instantly. I turned the breaker off and ran some extension cords to another circuit so my mother could have power.
<!—-> <!—->
I’d really appreciate any input you might have into what could be causing this.
<!—-> <!—->
Replies
What kind of voltmeter were you using? If it's an electronic meter then it will read voltage when there essentially is none, due to capacitative pickup. Plug a table lamp into the circuit when measuring to be sure you get good readings.
You said it was 14/2 wire, yet you speak of "ground". Is there a ground wire, or did you mean "neutral"?
No, it's modern romex-style wiring with a separate, bare ground.
I used an inductive voltmeter to determine whether the receptacles were live, and a digital meter to determine how much voltage was present. I measured the voltage on two different receptacles several times and it was always 119.
Edited 9/26/2008 11:51 pm ET by JDLee
You have an open neutral someplace.Most likely it in a backstab connection in a receptacle. But it could be in the neutral at the panel or anyplace else.The hot is hot so the inductive tester shows "power".And you have a load that is ON connected from the hot to the open neutral.Thus pulling the neutral up. Thus the indication on the receptacle tester and measurement of 119 from neutral to ground..
.
A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.
I'm confused.OP measured 119V across ground and neutral and outlet tester indicated ground/hot reversed.I can see how bridging hot to an open neutral through a load at one of the outlets could put voltage on the dead leg of neutral downstream, but how would that make the tester see hot/ground reversed?Why wouldn't he have found 120V across hot and ground? The hot should still be energized under the circumstances you propose, shouldn't it?BruceT
Those testers have limited "logic". For normal operation they generally good. But for problems they don't always give the exact details..
.
A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.
Hi Bill-
I used a "True RMS Auto-Ranging Multimeter" to measure the voltage across the neutral and ground. I've only ever used it to measure volts. I have to admit that I don't even know how to use it to measure amps.
http://www.sears.com/shc/s/p_10153_12605_03481079000P?vName=Tools&cName=Electrical+Shop&sName=Multi-Meters%2C+Testers+%26+Accessories
The problem with most digital meters is that they have extremely high input impedance, meaning that they'll pick up voltages (caused by inter-wire capacitance) on wires that are not connected at all. An electromechanical meter (the old style that doesn't need batteries and has a needle that moves across a scale) generally has lower input impedance and isn't confused by these "ghost" voltages.You can use a digital meter with more confidence if you do something like plug in an (incandescent) lamp (turned on) into the other outlet of the duplex recptacle -- the lamp provides a low-impedance "shunt" that will drain off "ghost" voltage.
There is no absurdity that human beings will not resort to in order to defend another absurdity. -- Cicero
Yeah, that's what I would've thought.
That was my guess. He could also do a continuity test if the problem is not found in one of the receptacles.
Family.....They're always there when they need you.
I had one that was similar over the summer. A waire had burned in half. I'm guessing you have something similar in a box somewhere. The voltage meters will show full voltage, but won't show amps. You can just barely have a connection and still get full voltage. But a bad connection won't transfer full amps.
A liberal is a socialist who won’t admit it. [Warren Robbins]
I'll have to read up on how to get my multimeter to measure amps. I'm sure it's not difficult; I just haven't done it before.
If a wire had burned in half somewhere, why would it likely be in a box and not just somewhere along the length of the cable in the walls or ceilings?
Miracle of miracles, the landlord showed up later yesterday. My mother said he fiddled with the box and the GFI outlet in the bathroom and somehow got the circuit working again. She said he might have replaced a breaker but wasn't sure (she's really no good for relaying information back and forth).
Since I had disconnected the GFI entirely and still had the same problem, I'm presuming it could not be at fault in any way. And since I had tried swapping breakers, I don't see how the breaker could be the problem.
Anyway, he somehow got the circuit branch in question working again. But, when he left, he said he should come back and look in all the boxes on that branch. My mother said he hadn't opened any of them.
So now I'm even more confused.
Edited 9/27/2008 10:20 am ET by JDLee
If a wire burns in half it is usually in a box and because someone nicked the conductor when he stripped it. Opens are most likely caused by loose back-stab connections or screws or wirenuts that are loose. I have only found one piece of romex that went open in a run. The installer (homeowner) had pounded in a staple and damaged the sheath and conductors. He kept on flipping the tripped breaker and the staple/wires charred pretty bad. Had to replace the wirerun and the breaker. He thought I was trying to generate work/revenue. There were other suspect staples and I didn't want to read about this guy's kids perishing in a fire.
Or a wire that did not get into the wire nut and just laying againts the other wires. Or backstab connections which are weak points.In addition to over driven staples you can have cables that bent too sharply, wedged in by framing or a long nail or screw penatrating it.And some one here once said that they had a spool of wire with a bad place where wires where spliced as the cable was made and the bad spot not cut out.Don't have any numbers but I would no be surpised that problems in boxes are not a 100 time, if not 1000 times, more common than a problem in a cable run..
.
A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.
> And some one here once said that they had a spool of wire with a bad place where wires where spliced as the cable was made and the bad spot not cut out.Yeah, that was me. I wish I'd saved the piece of wire as an "exhibit", since many have questioned whether it could happen or not. (But usually you don't think of that sort of thing when you're busy making a repair.)
There is no absurdity that human beings will not resort to in order to defend another absurdity. -- Cicero
This action by the landlord would have me checking the neutral conections in the circuit box. Just a few jiggling of the wires would remake the connection, at least for a time.
If a loose neutral was the problem, why would I find 119v across the neutral to the ground?
Because the loose neutral is the problem.
There is no absurdity that human beings will not resort to in order to defend another absurdity. -- Cicero
Because there was a "load" that was on between hot and neutral.And with a digital voltmeter it can be a very small load. For example a light fixture controlled by a lighted switch. When the switch is OFF the there a neon bulb and resister in series with the light. IIRC the resistor is aobut 47k ohms..
.
A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.
So, if I unplugged everything, you would expect to see that I'd measure 0v across the neutral and ground?
No, with the digital voltmeter I'd expect something in the neighborhood of 50V.
There is no absurdity that human beings will not resort to in order to defend another absurdity. -- Cicero
With the digital voltage meter you will measure a phantom voltage on an open wire.It be capacitively couple to the nearby ground and hot wires.I would expect to see it read between 20 and 100 volts.If it was "closed" and connected to the bonded neutral bus at the panel then it would read zero, or nevery near zero if there is current flowing in the neutral.If there is short between the neutral and hot and the neutral was open back to the panel then it would read whatever the supply voltage was. typicaly close to 120 volts.Note if the neutral was not also open back to the panel then you would have a dead short and the breaker would trip..
.
A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.
By "inductive voltmeter" you mean a non-contact detector? Or a electromechanical voltmeter of some sort?
There is no absurdity that human beings will not resort to in order to defend another absurdity. -- Cicero
Sorry, Dan. I found it described online somewhere as an "inductive voltmeter." But Greenlee describes it as a "Non-Contact Voltage Detector." It's one of those pen-shaped devices that beeps and lights up when its tip is placed near a live hot wire.
I don't even know what that other thing you described is.