Voltage on equipment ground(s)

Status
Not open for further replies.

andyman

Member
Our company was called to trouble shoot a problem with what appears to be return current (neutral current) on the equipment grounds from the branch circuits. Upon inspection of the panel we discovered that if we removed the grounds from the ground bar with the branch circuit connected we have what appears to be return current present on the ground wires. If we then also lift the neutral wire that voltage increases substantially. The readings we get are as follows on the EGC for the given circuit:

Kitchen receptacles, 120V

Bathroom, 70V with neutral disconnected and 34.5V connected

Upstairs lights, 102V with neut disconnected and 46V connected

Lights downstairs, 80.6V neut. disconnected and 41.6V connected

upstairs receptacles, 112V neut. disconnected and 55.4V connected

downstairs receptacles, 77V neut. disconnected and 36.5V connected

Refer, 76V neut. disconnected and 38.9V connected

Of coarse this would lead you to believe that the equipment gounding conductor and the neutrals have been tied together somewhere, except the only circuit that shows continuity between the EGC and neutral is the kitchen receptacles, nothing else rings out. Also, this house had a direct hit with lightning about one year ago, we live in an area that gets very little lightning and so I have little experience with those consequences. Any ideas on what to do next to correct problem? Thanks!
 

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
You can't have any current flowing if you disconnect the wires and you are going to get weird readings with a digital meter checking them the way you described.

I would try checking voltages at the panel first and then running an extension cord from a properly wired receptacle to the kitchen and bath and use it to test hot to hot, hot to neutral, neutral to ground.
 

andyman

Member
You can't have any current flowing if you disconnect the wires and you are going to get weird readings with a digital meter checking them the way you described.

I would try checking voltages at the panel first and then running an extension cord from a properly wired receptacle to the kitchen and bath and use it to test hot to hot, hot to neutral, neutral to ground.

True, we cannot have return current with wires disconnected but when I touch the ground wire to the grounded panel I get spark like you would if had done the same thing with a neutral conductor and that should not be and this happens with or w/o the neutral of that particular branch circuit terminated to the neutral bus. There should be zero return current on the EGC but there is on 7 of the branch circuits in this panel with no obvious connection between ground and neutral.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Our company was called to trouble shoot a problem with what appears to be return current (neutral current) on the equipment grounds from the branch circuits. Upon inspection of the panel we discovered that if we removed the grounds from the ground bar with the branch circuit connected we have what appears to be return current present on the ground wires. If we then also lift the neutral wire that voltage increases substantially. The readings we get are as follows on the EGC for the given circuit:

Kitchen receptacles, 120V

Bathroom, 70V with neutral disconnected and 34.5V connected

Upstairs lights, 102V with neut disconnected and 46V connected

Lights downstairs, 80.6V neut. disconnected and 41.6V connected

upstairs receptacles, 112V neut. disconnected and 55.4V connected

downstairs receptacles, 77V neut. disconnected and 36.5V connected

Refer, 76V neut. disconnected and 38.9V connected

Of coarse this would lead you to believe that the equipment gounding conductor and the neutrals have been tied together somewhere, except the only circuit that shows continuity between the EGC and neutral is the kitchen receptacles, nothing else rings out. Also, this house had a direct hit with lightning about one year ago, we live in an area that gets very little lightning and so I have little experience with those consequences. Any ideas on what to do next to correct problem? Thanks!

How did you measure this voltage? (that is what meter or tester did you use)
Also in your first post you talk about voltage and in post 3 you talk about current. If you measured current, how did you do that?
 

andyman

Member
I did not measure current but you can see it as a spark when you touch the ground wire to the grounded panel, I don't think induced voltage would be visible as a spark, to me a visible spark indicates electricity trying to get back to the source.

Voltage measurements where measured between the grounded panel and the EGC's with a Fluke 335 digital meter.
 

andyman

Member
More specifically the voltage measurements were taken between the branch circuit equipment grounding conductors and the utility supplied grounded (neutral) conductor.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
More specifically the voltage measurements were taken between the branch circuit equipment grounding conductors and the utility supplied grounded (neutral) conductor.
The utility supplied grounded conductor at the service disconnect, or just a neutral wire you assume is connected to it?
If the former, then you have at a minimum a high resistance neutral in the branch or else a wiring error.
 

andyman

Member
The utility supplied grounded conductor at the service disconnect, or just a neutral wire you assume is connected to it?
If the former, then you have at a minimum a high resistance neutral in the branch or else a wiring error.

We were using the utility supplied neutral which is bonded to the service panel.
 

rlundsrud

Senior Member
Location
chicago, il, USA
Start with the simple things, check that the neutral is correctly bonded on the main panel. Check the terminations there to make sure there isn't anything loose, especially the neutrals. Then isolate the problem to a single circuit by turning the others off and track it from there. Once you eliminate the problem from that circuit, you will know why any others are having the same problem.

From my personal experience, 70 volts usually indicates an open neutral with return current running through the ground. The 70 volts is an induced voltage, the spark on the ground is the return current.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
I am assuming that the meter you used is a high impedance meter and the EGC to neutral voltages are "ghost" voltages.
Those voltages, in my experience, are capable of producing a visible spark when the wire is touched to the grounded metal.

What was the issue that caused you to look at the voltage on the EGC?

With the loads all off and or disconnected and the neutrals and EGCs disconnected at the panel, what is the resistance between the neutrals and the EGC?
 

andyman

Member
I am assuming that the meter you used is a high impedance meter and the EGC to neutral voltages are "ghost" voltages.
Those voltages, in my experience, are capable of producing a visible spark when the wire is touched to the grounded metal.

What was the issue that caused you to look at the voltage on the EGC?

With the loads all off and or disconnected and the neutrals and EGCs disconnected at the panel, what is the resistance between the neutrals and the EGC?

Thank you for taking the time to consider my dilemma. Of all the years that I have done electrical work (about 10 years, so not long compared to most) I have never come across induced voltage (what you refer to as "ghost" voltage) in that high of a range (up to 80V on one of the circuits) and I don't know if I have ever seen it on branch circuits of a separate cable, I have noticed around 30V on multiwire branch circuits where you have one of the circuits on and you still read some voltage on the circuit that you have off. Also, in my experience it is totally unacceptable to have any voltage that high on the equipment grounding conductor, that is the conductor that bonds the metal on equipment and appliances, I don't think you would want enough voltage on appliances to make a spark to a bonded pipe.

The issue came up when another electrician alerted the homeowner to the problem and we were then hired to come in and trouble shoot. To me, and I am repeating myself but there is voltage on the EGC that wants to get back to the source and that is not an acceptable situation. If it makes a spark to a bonded panel it is return current because it is trying to get back to its source, otherwise why would it spark?

Am I way off base in my thinking on this one?
 

electricalist

Senior Member
Location
dallas tx
So when everything is wired correctly from the meter to the panel,grounds are bonded to neutral, there is 2 means of gec ,cold water/ground rod.
What is wrong?
I think if there is a parallel path on the ground you might be able to turn off one circuit at a time til you find the one thats creating the path .
 

kwired

Electron manager
Location
NE Nebraska
Start with the simple things, check that the neutral is correctly bonded on the main panel. Check the terminations there to make sure there isn't anything loose, especially the neutrals. Then isolate the problem to a single circuit by turning the others off and track it from there. Once you eliminate the problem from that circuit, you will know why any others are having the same problem.

From my personal experience, 70 volts usually indicates an open neutral with return current running through the ground. The 70 volts is an induced voltage, the spark on the ground is the return current.
70 volts is not induced, it is real voltage rise on the normally grounded conductor, it rises because there is a high resistance in the path back to the source, fix that high resistance or bypass it and voltage will drop.

I am assuming that the meter you used is a high impedance meter and the EGC to neutral voltages are "ghost" voltages.
Those voltages, in my experience, are capable of producing a visible spark when the wire is touched to the grounded metal.

What was the issue that caused you to look at the voltage on the EGC?

With the loads all off and or disconnected and the neutrals and EGCs disconnected at the panel, what is the resistance between the neutrals and the EGC?
I was leaning toward a bad neutral from the service/feeder or whatever is upstream, but so far the classic higher then 120 volts on one leg and less then 120 on the other hasn't yet been reported - but maybe is still there just not discovered yet, there is more readings under different conditions to be taken before this is solved.

Thank you for taking the time to consider my dilemma. Of all the years that I have done electrical work (about 10 years, so not long compared to most) I have never come across induced voltage (what you refer to as "ghost" voltage) in that high of a range (up to 80V on one of the circuits) and I don't know if I have ever seen it on branch circuits of a separate cable, I have noticed around 30V on multiwire branch circuits where you have one of the circuits on and you still read some voltage on the circuit that you have off. Also, in my experience it is totally unacceptable to have any voltage that high on the equipment grounding conductor, that is the conductor that bonds the metal on equipment and appliances, I don't think you would want enough voltage on appliances to make a spark to a bonded pipe.

The issue came up when another electrician alerted the homeowner to the problem and we were then hired to come in and trouble shoot. To me, and I am repeating myself but there is voltage on the EGC that wants to get back to the source and that is not an acceptable situation. If it makes a spark to a bonded panel it is return current because it is trying to get back to its source, otherwise why would it spark?

Am I way off base in my thinking on this one?

What you may be doing wrong is assuming there is voltage on the EGC, but instead you possibly have voltage that is above "ground" on the neutral conductor - some current is leaking via other grounded items that are in contact with your EGC system. By opening the path when you unhook EGC's you expose that voltage to where it is easier to see - but may be assuming it is coming from the wrong direction.

Maybe place a probe in the earth away from any grounding electrodes connected to the system and run a lead from there to where you are measuring these voltages - you can then tell what is truly at ground potential and what is not. Doesn't hurt to play with both high and low impedance meters just to see the differences in readings, but use a low impedance meter and you will eliminate low capacitance, capacitive coupling (AKA ghost voltages).
 

Fulthrotl

~Autocorrect is My Worst Enema.~
Thank you for taking the time to consider my dilemma. Of all the years that I have done electrical work (about 10 years, so not long compared to most) I have never come across induced voltage (what you refer to as "ghost" voltage) in that high of a range (up to 80V on one of the circuits) and I don't know if I have ever seen it on branch circuits of a separate cable, I have noticed around 30V on multiwire branch circuits where you have one of the circuits on and you still read some voltage on the circuit that you have off. Also, in my experience it is totally unacceptable to have any voltage that high on the equipment grounding conductor, that is the conductor that bonds the metal on equipment and appliances, I don't think you would want enough voltage on appliances to make a spark to a bonded pipe.

Am I way off base in my thinking on this one?

not so much off base, as just following your meter where it leads you....

the problem with high quality DVM's is that they have a phenomenally high ohm/volt, which is nice 'cause the
meter doesn't have much influence on the circuit under test... when it's a sensitive electronic circuit.

the meter that i use for all my field work is a discontinued fluke 12. when you flip it on, the default meter
has a 2,000 ohm shunt internal in the meter, so it puts a bit of load on what you are measuring, and gives
you real world voltages that you can trust.

and you can get a spark off transient voltages. i'd bet you a cheeseburger that if you use an analog meter,
like an old simpson, or something like the fluke 12, all those weird voltages will disappear.

day before yesterday, i went to megger some long feeders, with a fluke digital megohmmeter.
the feeders are good. i KNOW they are good, for a number of reasons. however, the conduits
are now full of water, and the capacitance of the insulation is all wonky 'cause of the water....
short version is, the megger was batshit crazy. got a second meter, and the readings were good.

if i'd a not gotten a warm fuzzy feeling off the second megger, i'd gone and got a old hand cranked 500V,
and that would have told me if it was safe to energize... sometimes, low tech analog is better.
 
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
Power off at the Main. Shut off all CBs.
Gas or electric range and dryer?
Are they 3 or 4 wire if electric? If 4, check for proper connection at the appliance. If 3, unplug them.

Check continuity of EGs that you have isolated from the Neutral bar. Most likely at least some of them have continuity.
Remove all the neutrals from the bar. Check between each of them. There should be no continuity. Chase down the one(s) that do.
Check for continuity between each isolated neutral and the group of EGs that appear to be common. There should be none.
Chase down the neutral that shows continuity. Check against the other EGs. Repeat as needed.


With all power off, what is the current on the GEC(s)? What is the current on the Neutral?
Do you share a common transformer with neighbors?
Common metallic water line? Some current flow would be normal.


EGs will not have current on them unless there is a fault. Hot to EG or Neutral to EG. No current, no voltage.

I have in the past chased what I thought was V, until I adjusted my glasses and found Mv.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
...
The issue came up when another electrician alerted the homeowner to the problem and we were then hired to come in and trouble shoot.
Alerted them to what problem?
To me, and I am repeating myself but there is voltage on the EGC that wants to get back to the source and that is not an acceptable situation. If it makes a spark to a bonded panel it is return current because it is trying to get back to its source, otherwise why would it spark?

Am I way off base in my thinking on this one?
In my opinon what you are seeing is not a problem, just the normal result of using a high impedance meter.
The capacitive coupled voltage in the isolated EGC has some energy...it is not unusual to see a small spark.

If you do your testing using a meter with a much lower impedance and you still see the voltage, or if you measure the current and see current, then I would say you have some type of problem. From what you have posted here, I am not seeing any problems.
 

andyman

Member
Alerted them to what problem?

Problem is unacceptably high voltage on equipment grounding conductor.

In my opinon what you are seeing is not a problem, just the normal result of using a high impedance meter.
The capacitive coupled voltage in the isolated EGC has some energy...it is not unusual to see a small spark.

I would bet if you were barefoot on a concrete floor with sweaty feet and hands and you touched something like a metal toaster that you would be feeling that 80V on the equipment ground.

If you do your testing using a meter with a much lower impedance and you still see the voltage, or if you measure the current and see current, then I would say you have some type of problem. From what you have posted here, I am not seeing any problems.

Based on everyone's reply's it sounds like I need an analog tester to help determine problem.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Based on everyone's reply's it sounds like I need an analog tester to help determine problem.

One of the very revealing numbers will be the current in the EGC. That is most easily and safely determined with a clamp-on ammeter. Whether digital or analog does not matter much for that function.
 
Status
Not open for further replies.
Top