A sub panel where the N was not isolated from EGC.

[templdl]

I just corrected 2 subpanels where the the N and EGC were not isolated from one another. The bonding scew was used in one panel and missing from the other. Needless to say I added a ground bar and moved all of the EGC over to the ground bar. Then I removed all of the neutrals removed the bonding screw in the one and checked both for insulation and isolation from ground.
That was obvious. But, on 2 previous occasions the home experienced extensive damage to electronics and appliences amounting to thousands of dollars due to a lightning strike as told be the home owner. In retrospect I strongly suspect that the incorrect wiring of the 2 subpanels contributed to the damage by allowing a surge to appear on both the N and EGC down stream from the SE.
What are your thoughts?

 

[Dennis Alwon]

I doubt the mis-wiring had anything to do with it

 

[jumper]

I doubt the mis-wiring had anything to do with it

Ditto. :thumbsup:

 

[mjf]

I doubt the mis-wiring had anything to do with it

Agree.

I find (and correct) miss wired sub-panels constantly, especially in older installations.

Anyone know what code cycle required separation in downstream panels?
Or were ground and neutral always supposed to be separate?

 

[ActionDave]

Agree.

I agree also.

I find (and correct) miss wired sub-panels constantly, especially in older installations.

Anyone know what code cycle required separation in downstream panels?
Or were ground and neutral always supposed to be separate?

I don't know that it was ever allowed once the wide open frontier of electrical was tamed. As far as how it was enforced..... Do you know any retired electricians or inspectors in your area?

 

[templdl]

Thanks guys,
The home I refered to was built in1978 and I doubt if bonding the neutral to the ground in a subpanel was allowed at that time. But I thought that I would bounce the lightning issue off you as the home have substantial damage done to electronics and appliances on two separate occasions and questioned whether the subpanels were a contributing factor.
Dave

 

[broadgage]

AFAIK, ground and neutral had to be seperrated in sub-panels since it least the 1930s, reference is made to this requirement in an REA* leaflet from then.

From a practical point of view, as distinct from being a code violation, I dont believe that it much matters. Failing to seperate the ground and neutral in a sub panel must be a very common violation, and very seldom indeed has any serious consequences.

* Rural Electrification Authority, set up as part of the national deppression recovery programe to supply low cost electrical service to areas lacking service.

 

[greenjourneyman]

bonding sub panel to water main ?

bonding sub panel to water main ?

We put a new service on a garage and overhead feed the house to main panel and refed a subpanel downstairs. I was told to bond the water main with #6 to sub panel. Does this seem right? As for a grounding electrode condutor to water main I thouhjt you sized according to 250.66 ( largest nongrounded v onductor). So would it be largest ungrounded to the sub panel? I'm wearing out 250 looking

 

[greenjourneyman]

Sorry guys that was supposed to be new thread

 

[templdl]

From a practical point of view, as distinct from being a code violation, I dont believe that it much matters. Failing to seperate the ground and neutral in a sub panel must be a very common violation, and very seldom indeed has any serious consequences.

My concern has always been that when the neutral and EGC are connected together down stream from the SE panel that there will be some portion of the neutral current returning to the SE panel via the EGC which would be greater should the neutral be compromized.
This in itself would be a code violation as the EGC is not considered to be a current carrying conductor.
When any device the has a plug that grounds in plugged into a grounded receptacle you will have the potential of exposing the device that is plugged in and grounded to any neutral current that may be flowing on the EGC which can possibly present a which hazard.
Your thoughts?

 

[ggunn]

I just corrected 2 subpanels where the the N and EGC were not isolated from one another. The bonding scew was used in one panel and missing from the other. Needless to say I added a ground bar and moved all of the EGC over to the ground bar. Then I removed all of the neutrals removed the bonding screw in the one and checked both for insulation and isolation from ground.
That was obvious. But, on 2 previous occasions the home experienced extensive damage to electronics and appliences amounting to thousands of dollars due to a lightning strike as told be the home owner. In retrospect I strongly suspect that the incorrect wiring of the 2 subpanels contributed to the damage by allowing a surge to appear on both the N and EGC down stream from the SE.
What are your thoughts?

The neutral and ground are still connected at the main panel and the resistance between the main and the subs on the neutral and ground wires is very low. I don't think that having the neutral and ground touching in the subs made any difference to the damage due to the strike.

 

[GoldDigger]

AFAIK, ground and neutral had to be seperrated in sub-panels since it least the 1930s, reference is made to this requirement in an REA* leaflet from then.

From a practical point of view, as distinct from being a code violation, I dont believe that it much matters. Failing to seperate the ground and neutral in a sub panel must be a very common violation, and very seldom indeed has any serious consequences.

* Rural Electrification Authority, set up as part of the national deppression recovery programe to supply low cost electrical service to areas lacking service.

IMHO the biggest practical problem, in the absence of a massive ground fault, is that the multiple EGC to neutral bonds can cause problems with GFCI circuit breakers (or RCDs across the pond). But for a subpanel bond to be an issue there, the GF device would have to be at the feeder, main, or generator supply breaker.

 

[electrofelon]

AFAIK, ground and neutral had to be seperrated in sub-panels since it least the 1930s, reference is made to this requirement in an REA* leaflet from then.

From a practical point of view, as distinct from being a code violation, I dont believe that it much matters. Failing to seperate the ground and neutral in a sub panel must be a very common violation, and very seldom indeed has any serious consequences.

* Rural Electrification Authority, set up as part of the national deppression recovery programe to supply low cost electrical service to areas lacking service.

I think how serious this is depends on the wiring method. Worked with a guy once where the job was all surface emt, no wire EGC to subs. He said to put the bonding screws in because "neutral and grounds are the same" In this case it is pretty bad because no you have neutral current flowing on all your piping. Could be a potential difference between different pieces of pipe, could arc between two pipes and start a fire. This case was particularly bad becasue most of the pipes were independently supported, i.e. no trapezes to bond it all together.

 

[GoldDigger]

I think how serious this is depends on the wiring method. Worked with a guy once where the job was all surface emt, no wire EGC to subs. He said to put the bonding screws in because "neutral and grounds are the same" In this case it is pretty bad because no you have neutral current flowing on all your piping. Could be a potential difference between different pieces of pipe, could arc between two pipes and start a fire. This case was particularly bad becasue most of the pipes were independently supported, i.e. no trapezes to bond it all together.

Since, in the absence of a neutral failure, the neutral would still present an alternate (shared current) path with the EGC, there might be very low voltage arcing contacts when interrupting the EGC path of the EMT, but I do not see it generating much heat or being a shock hazard.
If the neutral fails, then all bets are off and you have to depend on the EMT installation method to insure a solid connection.
But if you bond at the subpanel, there will be no way to detect a failure in the neutral, and that condition might last a long time, exposing workers on the EMT to injury.

 

[Galt]

If the neutral in some way gets unhooked in a correctly wired sub panel the lights go out.If the same happens in a sub panel with grounded neutrals the 120 volt loads run on the ground wire. Now ask what happens if both are unhooked? Any metal objects hooked to this panel will become effectively energized although in series with the load.

 

[templdl]

If the neutral in some way gets unhooked in a correctly wired sub panel the lights go out.If the same happens in a sub panel with grounded neutrals the 120 volt loads run on the ground wire. Now ask what happens if both are unhooked? Any metal objects hooked to this panel will become effectively energized although in series with the load.

Any way you look at it the the results are not good. Bottom line is to keep that neutral isolated from the EGC except at the SE.

 

[kwired]

My concern has always been that when the neutral and EGC are connected together down stream from the SE panel that there will be some portion of the neutral current returning to the SE panel via the EGC which would be greater should the neutral be compromized.
This in itself would be a code violation as the EGC is not considered to be a current carrying conductor.
When any device the has a plug that grounds in plugged into a grounded receptacle you will have the potential of exposing the device that is plugged in and grounded to any neutral current that may be flowing on the EGC which can possibly present a which hazard.
Your thoughts?

People have mentioned some of the worst case possibilities such as having raceways carrying 100% of neutral current should the intended neutral become compromised, and that is good information. Something nobody has mentioned yet that is there from the beginning - voltage drop. If a conductor (which a raceway carrying current is still a conductor from a basic electricity point of view) is carrying current there is a voltage drop across whatever resistance is in that conductor. Anything connected to this "conductor" will have a raised voltage to earth, or anything else bonded to earth. Now most of the time is will be such a small voltage it is negligible, but when things go wrong it can be pretty high. Now bring the right conditions to the mix and only a few volts can be deadly. This is exactly why we install equipotential grids in certain applications, sometimes the voltage to ground is still a problem even when we ran the branch circuits and feeders correctly as that equipment grounding conductor in a proper install still has any voltage drop on the service or from POCO's primary neutral imposed on it.

I think how serious this is depends on the wiring method. Worked with a guy once where the job was all surface emt, no wire EGC to subs. He said to put the bonding screws in because "neutral and grounds are the same" In this case it is pretty bad because no you have neutral current flowing on all your piping. Could be a potential difference between different pieces of pipe, could arc between two pipes and start a fire. This case was particularly bad becasue most of the pipes were independently supported, i.e. no trapezes to bond it all together.

Even with equipment grounds pulled in the pipes, you have three parallel paths, the intended neutral, the equipment grounding conductor, and the raceway. The intent of the NEC is for current to only flow on the actual intended neutral conductor itself.

 

[templdl]

People have mentioned some of the worst case possibilities such as having raceways carrying 100% of neutral current should the intended neutral become compromised, and that is good information. Something nobody has mentioned yet that is there from the beginning - voltage drop. If a conductor (which a raceway carrying current is still a conductor from a basic electricity point of view) is carrying current there is a voltage drop across whatever resistance is in that conductor. Anything connected to this "conductor" will have a raised voltage to earth, or anything else bonded to earth. Now most of the time is will be such a small voltage it is negligible, but when things go wrong it can be pretty high. Now bring the right conditions to the mix and only a few volts can be deadly. This is exactly why we install equipotential grids in certain applications, sometimes the voltage to ground is still a problem even when we ran the branch circuits and feeders correctly as that equipment grounding conductor in a proper install still has any voltage drop on the service or from POCO's primary neutral imposed on it.

Even with equipment grounds pulled in the pipes, you have three parallel paths, the intended neutral, the equipment grounding conductor, and the raceway. The intent of the NEC is for current to only flow on the actual intended neutral conductor itself.

This is so extremely important to recognize as well as appreciate and should never be taken lightly.
Well stated.