Interesting Case of Phantom or induced voltage

Status
Not open for further replies.

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
Here is the scenario
A non grounded 2 wire circuit feeds a 3 gang box box in the kitchen—The 3 gang has a separate feed for the receptacle and it is grounded. Then there are 2 switches that control undercabinet lights and a ceiling exhaust fan that are fed fom the ungrounded 2 wire circuit. There is also a feed to the dishwasher from this circuit

All the undercabinet lights, fan and the dishwasher have equipment grounding conductor’s in the cable which are tied together at the load as well as in the box.

Countertops and cabinets are metal and there is phantom voltage on the equipment grounding conductor because they are not tied back to the panel. When you touch the dishwasher and the countertop you get a shock?????? I know this is phantom voltage and I know the countertop is not energized. The countertop is grounded thru contact with copper water pipes not thru the dishwasher.

The voltage from countertop to dishwasher went from 5 voltage to 38 v or so depending on how many loads were turned on—ie, undercabinet light or ceiling exhaust. Why?

Also why would you get a shock? I did not think you can feel phantom voltage. I have convinced myself that there is leakage somewhere else but I cannot find it if it exists and the voltage and shock disappeared a soon as I tied the equipment grounding conductor from the receptacle circuit to those equipment grounding conductor’s on the ungrounded circuit.

Problem solved but why a shock???? And why did the voltage vary
 

big john

Senior Member
Location
Portland, ME
Were the under cabinet lights fluorescent by chance? I agree, I've never heard of normal phantom voltage severe enough to cause shocks, and I would've said you were feeling neutral current going through a parallel path except for the fact that 38V difference would obviously be way too high for that.
 

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
Were the under cabinet lights fluorescent by chance? I agree, I've never heard of normal phantom voltage severe enough to cause shocks, and I would've said you were feeling neutral current going through a parallel path except for the fact that 38V difference would obviously be way too high for that.


They are fluorescents and that is what I thought but it is not the case. I tried to shock myself at home by creating phantom voltage and I could not. Something else must be going on but the shock issue is gone.... In fact they said that when the used to open the metal cabinet doors it would cause interference on the radio.. Now it does not.
 

Gene B

Member
Location
USA
It would be interesting to measure current on the EGCs. I would say one of the loads is failing or has a design flaw.
 

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
It would be interesting to measure current on the EGCs. I would say one of the loads is failing or has a design flaw.

There was no true equipment grounding conductor. The equipment grounding conductor's that had induced voltage had anywhere from 5 volts to about 38V depending on if the fan and/or lights were on.

I thought perhaps the countertop was energized or maybe the plumbing but I tested that and that was not the case. I felt really stupid because it took me awhile to realize I had induced voltage because I didn't realize the feed didn't have an equipment grounding conductor.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
You may have eliminated the shock symptoms, but the installation is not code compliant with the effective EGC for the groundless (not strictly ungrounded!) circuit is following a different path from the rest of the circuit conductors.
There are some exceptions for supplying a ground for a previously groundless receptacle circuit, but they may not apply in your case.

I also strongly suspect that you have insulation leakage rather than simple capacitive phantom voltage. In which case you need to fix that too. I would measure the current in the EGC connection to get more info.
 

mivey

Senior Member
There was no true equipment grounding conductor. The equipment grounding conductor's that had induced voltage had anywhere from 5 volts to about 38V depending on if the fan and/or lights were on.

I thought perhaps the countertop was energized or maybe the plumbing but I tested that and that was not the case. I felt really stupid because it took me awhile to realize I had induced voltage because I didn't realize the feed didn't have an equipment grounding conductor.
Oh, if you would only have had some of the original BX in place instead of the ungrounded 2 wire. :D

I am surprised you got a non-leak shock except that an ungrounded fluorescent can nip you a bit. BTW, vent hoods are notorious for leaking because I guess the grease breaks things down. We have re-wired/replaced a many for that very reason. Remember the old metal counter edging?: It can carry the leak all over the kitchen.

Did you try to sustain the voltage through varying resistance values? It would be interesting to get an idea of the source's impedance (delta V / delta I).
 

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
You may have eliminated the shock symptoms, but the installation is not code compliant with the effective EGC for the groundless (not strictly ungrounded!) circuit is following a different path from the rest of the circuit conductors.
There are some exceptions for supplying a ground for a previously groundless receptacle circuit, but they may not apply in your case.
The entire install is not compliant. Disposal, dishwasher, undercabinet lights, kitchen ceiling light and kitchen fan are all on the same circuit. Yes, the equipment grounding conductor is not following the same path back but I don't see that as a real issue. The only way to fix this is to totally redo the circuit which I have talked to them about doing. Maybe someday?????


I also strongly suspect that you have insulation leakage rather than simple capacitive phantom voltage. In which case you need to fix that too. I would measure the current in the EGC connection to get more info.

Well perhaps but when I simulated this induced voltage at home I got the same voltage reading. What I still cannot understand is why the voltage changes when the load on the circuit changes---signicantly, imo
 

Gene B

Member
Location
USA
There was no true equipment grounding conductor. The equipment grounding conductor's that had induced voltage had anywhere from 5 volts to about 38V depending on if the fan and/or lights were on. I thought perhaps the countertop was energized or maybe the plumbing but I tested that and that was not the case. I felt really stupid because it took me awhile to realize I had induced voltage because I didn't realize the feed didn't have an equipment grounding conductor.
When you got shocked your body was completing the EGC path. The fact that the EGCs were connected together added to the risk, since a fault in any would energize all. What you could do is use an ampmeter on the (completed) EGC and turn on each load one at a time. When/if you find significant (>1ma) current, that's the faulty device. Or maybe more efficient, install a GFCI and see what, if anything, trips it.
 
Last edited:

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
...
...
What I still cannot understand is why the voltage changes when the load on the circuit changes---signicantly, imo
Simplest thing in the world to come up with one of many plausible explanation, but hard to pin it down to any one.
Example:
Consider a Calrod (TM) heating element consisting of an insulated resistive element inside a grounded metal shell.
If the only leakage is capacitive, then the leakage current will be identical to the current through the total capacitance but at half the applied voltage when operated line to neutral. The same element operated line to line on 120/240 single phase will produce zero net leakage current unless there is some asymmetry.
The only time the ground fault current (and therefore phantom voltage) will not be load dependent will be when the coupling is entirely in the supply wiring upstream of the local switch.
It is also possible for two loads with opposite phase leakage to produce a lower phantom voltage than either one alone.
Voltage dividers and vectors. :)
 

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
I wish I had done a lot deeper checking when I was there. I should have used the ampmeter and it wasn't until after I left that I said to myself- "Wait a minute, phantom voltage won't give you a shock."

I get that completing the equipment grounding conductor circuit should have told me that there is leakage somewhere. Cabinet guy had done some work there and we had him take the cabinets down because we were sure he got a screw in the wire. We could find no evidence of it. I still think that there must be something like that going on.

I am so mad at myself because I am generally very good at troubleshooting and this time I just wasn't thinking clearly. I have never not found the problem on a job when something like that happened. This was the first time.
 

mivey

Senior Member
Well perhaps but when I simulated this induced voltage at home I got the same voltage reading. What I still cannot understand is why the voltage changes when the load on the circuit changes---signicantly, imo
A phantom voltage probably would not change that much and would disappear with a small resistance. Also would not be shocking you.

The best I can figure, there must be a leak and you are getting a voltage rise across the leak impedance that is pulling the normally floating ground closer to line voltage as the leakage current increases.
 

mivey

Senior Member
Cabinet guy had done some work there and we had him take the cabinets down because we were sure he got a screw in the wire. We could find no evidence of it. I still think that there must be something like that going on.
Was that before or after the radio noise? Could be an original issue with previous cabinets.

I have also had siding guys on the outside hit the cable so you might consider any remodeling/repairs done outside as well.
 

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
Was that before or after the radio noise? Could be an original issue with previous cabinets.

I have also had siding guys on the outside hit the cable so you might consider any remodeling/repairs done outside as well.
I am not sure but I think the radio in that area was a new addition after the work.

The cabinets are the same and they are metal cabinets. They took them down and rehung them for some reason. Problem is the cabinet guy did some of this electrical work and I do know that the shock was an issue after he finished. Whether it is in the wiring he did or driving a screw into a wire I don't know. If there is a fault to the equipment grounding conductor it isn't enough to trip a breaker because it didn't trip when we connected it up.

I really want to go back and solve this issue because I spend half the day thinking about it. We did so many things it is hard to remember everything we did.... All I know is the voltage that I was reading, phantom or not, is gone.
 

mivey

Senior Member
Problem is the cabinet guy did some of this electrical work and I do know that the shock was an issue after he finished.
Then my guess would be he probably pinched the cable with a staple or getting it caught between something.

Cabinet guy doing wiring? Oh, the stories we could share. Some of the worst electrical hack jobs I have ever seen.
 

kwired

Electron manager
Location
NE Nebraska
Did you measure your presumed "phantom voltage" with a low impedance meter or a high impedance meter?

Low impedance meter and still reading a voltage even really low levels - probably not "phantom voltage".
 

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
Then I vote no phantom voltage. Get a low impedance adapter for your meter, or a light bulb in a rubber socket, or something similar.

Well are you saying that the meter won't read phantom voltage? Because if that is what you are saying then I disagree because I energized the black wire on a piece of nm cable and left everything else disconnected and I read 38 or so V from the bare wire to a grounded box. I repeated this at home with a 3' piece of nm and it read the voltage
 
Status
Not open for further replies.
Top