Voltage on Washing Machine frame; I am baffled.

[oldsparky52]

I appreciate all these ideas from everyone.

I am so glad you posted this in here. I sure hope you come across the problem and share it but I don't hold out hope based on what you have said. I've actually enjoyed reading the ideas, thinking, and doing something other than what I've been doing.

 

[tortuga]

Wayne and Syncro I think you nailed it:

.. it's unclear if the energized frame as on #10 or the unenergized frame as on #12 is the typical behavior of this washer on an ungrounded 2 wire circuit. It does show that there's a difference between circuits #10 and #12, the specifics of which have yet to be determined.

I suggest powering the washer with a known-good temporary 2-wire circuit and seeing if it develops a voltage from case to N. If it does, add the EGC and see if it goes away. If so and it's not defective then it simply can't be used on a 2-wire circuit (which the NEC and the instructions both tell you). And circuit #12 must somehow be bonding the washer case.

If it doesn't develop a voltage from case to N on a known-good 2-wire circuit, then there's some defect in the wiring on circuit #10, although I'm at a loss as to what sort of defect would exhibit the symptoms described.

Cheers, Wayne

A L-G filter capacitor in the washer could bring an open ground wire up in voltage considerably if it was loaded with only a high impedance meter and a low stray capacitance. And so wiring up just the washer or just the freezer leg of circuit #10 would reduce the capacitance and allow the G-N voltage at the receptacle (and therefore the washer housing to neutral voltage) to rise higher when the washer with its internal filter caps is plugged in.

I assume that the voltages you quoted above are at the receptacle without the washer plugged in? If so then you might pull the receptacle from the box and measure those voltages at the terminals when the washer is plugged in. The washer wouldn't have to be running to have its filter capacitor drawing current across L-G. I'm thinking such a current would have to be less than 4mA because otherwise the washer would not be compatible with GFCI's. But even a few mA can give you a noticeable shock.

Seems like standard protocol for a new washer in a old house with undrounded wiring should not be the GFCI replacement, rather adding a EGC as in 250.130(C)

 

[romex jockey]

Which calculation? The cost when the hack gets paid, then add to it cost to get it fixed right? Cost in dangerous conditions created by hack installation with potential for loss of life? Or overall industry cost with "savings" adjusted in because customer says "the lights come on, so why is it wrong"?

Good one Fred

anecdotally, 1/2 my biz is following up the trunk slammers

I just get tired of our trade getting blamed for their work, and the CMP's making codes over it all that they'll never read

~RJ~

 

[brantmacga]

A couple further comments:

If I have read correctly, the washer has only been tested on circuits 10 and 12. So it's unclear if the energized frame as on #10 or the unenergized frame as on #12 is the typical behavior of this washer on an ungrounded 2 wire circuit. It does show that there's a difference between circuits #10 and #12, the specifics of which have yet to be determined.

it was tested on two other receptacles, I don't know which number they are, and I didn't mention it because I didn't think it important. The owner first plugged in an extension cord in the kitchen for me because I couldn't see receptacle #12 as it was blocked by a clothes rack. He plugged into two different kitchen receps before I saw #12.

 

[brantmacga]

Could it be that the difference is they on different "phases"? There are actually 2 faults, one on the dryer and one on the washer? The dryer fault is on the same phase as circuit 10 so there is no voltage fault to fault if they are on the same phase.

Yea, reaching but ...

I swapped breaker #10 into the position of #12 to make sure I'd tried the same wiring on opposite legs.... I did mention that somewhere in the first post. I know we're getting long here and its getting easy to miss things.

 

[wwhitney]

it was tested on two other receptacles, I don't know which number they are, and I didn't mention it because I didn't think it important. The owner first plugged in an extension cord in the kitchen for me because I couldn't see receptacle #12 as it was blocked by a clothes rack. He plugged into two different kitchen receps before I saw #12.

OK, if the kitchen receptacles have EGCs (kitchen receptacles may be more recently installed), then that would have no bearing on determining the typical behavior of the washer on a ungrounded circuit. But if they don't and have no receptacle EGC to N connection anywhere upstream, then it would.

Cheers, Wayne

 

[oldsparky52]

I swapped breaker #10 into the position of #12 to make sure I'd tried the same wiring on opposite legs.... I did mention that somewhere in the first post. I know we're getting long here and its getting easy to miss things.

Oops, you are correct, I forgot.

 

[brantmacga]

OK, if the kitchen receptacles have EGCs (kitchen receptacles may be more recently installed), then that would have no bearing on determining the typical behavior of the washer on a ungrounded circuit. But if they don't and have no receptacle EGC to N connection anywhere upstream, then it would.

Cheers, Wayne

I only saw one newer 12/2 NM cable with a ground when I opened the panel cover. Can’t say for sure it was or wasn’t the kitchen.


Sent from my iPhone using Tapatalk

 

[Strathead]

I tested voltage on the receptacle with washer running, and there was 0 voltage drop; I also used an extension cord running from receptacle #12 to measure against the neutral of that circuit.

I appreciate all these ideas from everyone.

I wouldn't expect a voltage drop, that would indicate a problem with the hot. I do a silly thing when I troubleshoot. I think of myself as a single little electron that can clone himself for parallel paths. Then I run along wires from the source and find my way back. In your case, you have a neutral and a ground bonded together. One is at a different voltage potential than the other out in the field. So starting at the transformer you run along the hot until you get through the washing machines load. After that, you have a resistance between there and the neutral ground bonding point, period. At this point it can either be on the neutral, or on the ground, but since you have a voltage potential between them, the ground is unlikely and easy to eliminate. For it to be the ground you would need a double problem, one being a leak to ground from your hot in the washing machine and also a resistance on your ground. Easy to check by running a ground wire from the panel and checking the washing machine frame to the ground for no voltage. Once eliminated, then it can only be on the neutral. From that point you need to find places you can read voltage from neutral to ground while the washing machine is running, without disconnecting the washing machine all the way back to the service connection. It is really as simple as that in my mind.

 

[tortuga]

Brant I believe as Syncro explained, there is an internal filtering capacitor circuit in the washer.
Here is an example of a appliance filtering circuit, caps C2 and C3 need to be connected to a 'frame ground'.
It is probably taking hi frequency powersupply 'noise' created by the washer to the equipment grounding terminal. When a equipment ground is not present the frame of the washer will have this hi frequency noise on it.
I am not saying the washer is operating normally as that seems like a lot of voltage.

250.114(3)(b) requires that washer to be connected to an EGC, so that precludes the use of 406.4(D) to use a GFCI on old two wire circuits. I think either way the fix that you can do as an EC is adding a EGC to the washer and freezer receptacles.

 

[wwhitney]

Agree that the above is most likely the problem, but I'm still curious why it only exhibits on circuit #10. Simplest explanation is that the other circuits tried have an EGC or a bootleg ground, even if not installed directly at the receptacle (which has been ruled out for #12).

Cheers, Wayne

 

[kwired]

Brant I believe as Syncro explained, there is an internal filtering capacitor circuit in the washer.
Here is an example of a appliance filtering circuit, caps C2 and C3 need to be connected to a 'frame ground'.
It is probably taking hi frequency powersupply 'noise' created by the washer to the equipment grounding terminal. When a equipment ground is not present the frame of the washer will have this hi frequency noise on it.
I am not saying the washer is operating normally as that seems like a lot of voltage.

250.114(3)(b) requires that washer to be connected to an EGC, so that precludes the use of 406.4(D) to use a GFCI on old two wire circuits. I think either way the fix that you can do as an EC is adding a EGC to the washer and freezer receptacles.

Next thing though is will such filtering trip GFCI's? And/or AFCI's? GFCI has been required in some situations in laundry areas in more recent codes and latest code requires in all laundry areas regardless of other situations. No EGC can maybe mask the problem until you find a way to make a return path.

 

[ramsy]

Next thing though is will such filtering trip GFCI's? .. No EGC can maybe mask the problem until you find a way to make a return path.

Roger that, EGC must be in place to make that determination.

..new washer in a old house with ungrounded wiring should not be the GFCI replacement, rather adding a EGC as in 250.130(C)

Yes, if new appliance can't be moved to newer outlet with grounding, then it must be supplemental EGC or new circuit.

Unless at some point the existing 2-prong outlet was replaced with 3-prong, then GFCI replacement would also be required for a laundry room, etc..

250.114(3)(b) requires that washer to be connected to an EGC, so that precludes the use of 406.4(D) to use a GFCI on old two wire circuits.

406.4(E) Note: calls 250.114 to clarify where an EGC is still required, regardless of GFCI's, that may also be required.

Without that 2-prong cord marked with double-insulation, there's no exception for the EGC at those appliances.

I think either way the fix that you can do as an EC is adding a EGC to the washer and freezer receptacles.

Yep!

 

[sparky234]

As soon as you described the wiring type "2 wire- no ground", I suspect the problem is precisely that. i.e. no ground. It is probably the asbestos sheathed cable. Most of that wiring went out shortly after WWII. I am sure that you double-checked to be sure there is no EGC. Back in the day, electricians would wrap the bare copper grounding conductor on the outside of the metal outlet box. But regardless, the solution here would be to ground the washing machine frame to ground by installing NMC cable with ground to the appliance receptacle. As it is now, this is a dangerous condition.

 

[sparky234]

As soon as you described the wiring type "2 wire- no ground", I suspect the problem is precisely that. i.e. no ground. It is probably the asbestos sheathed cable. Most of that wiring went out shortly after WWII. I am sure that you double-checked to be sure there is no EGC. Back in the day, electricians would wrap the bare copper grounding conductor on the outside of the metal outlet box. But regardless, the solution here would be to ground the washing machine frame to ground by installing NMC cable with ground to the appliance receptacle. As it is now, this is a dangerous condition.

As well as install a GFCI receptacle for the appliance receptacle.

 

[drcampbell]

Next thing though is will such filtering trip GFCI's? ...

Seems pretty unlikely. 2.2 nF will present an impedance of 1.21 MΩ at 60 Hz, and a current of less than 0.1 mA at 120 volts.

 

[synchro]

UL allows up to 0.75 mA leakage current for cord connected fixed or stationary appliances.

 

[romex jockey]

so how'd this sugar off Brant?

~RJ~

 

[brantmacga]

so how'd this sugar off Brant?

~RJ~

I haven’t made it back yet. The house is about an hour away from me, which is tough to squeeze into my schedule. Probably another week or two before I can go back.


Sent from my iPhone using Tapatalk