grounding a sub panel

[bennie]

Re: grounding a sub panel

Roger: I know you are aware that a continuous metal path is OK if not bonded at both panels. 250.32.2.

This was the issue when I got trashed for stating "there is no code section that requires a neutral/ground separation at a panel". The code section is to eliminate parallel neutral paths. The method for doing this is not spelled out.

A three wire cable assembly, in a dwelling, does not create parallel paths.

The neutral is not permitted to be grounded (earthed)on the load side of a service, the connection at a panel for ground fault purposes was not considered grounding (earthing).

Three wire, cable supplied panels, were supplied by SE cable, W/bare neutral/ground conductor until 1996.

The change to 4 wire, with a reduced size safety ground conductor for fault clearing was started in 1996, by a select committee. No one could contest, or debate, the personal decisions, and misunderstandings that created the changes.

This change caused a higher impedance to be present in the fault path, and elevated the neutral/ground voltages. This is not forward engineering, it is two steps back.

 

[roger]

Re: grounding a sub panel

Hello Bennie,

I know you are aware that a continuous metal path is OK if not bonded at both panels.

I have actually been involved in similar discussion this past week with an engineer (never have worked with him before) and his design of a remote standby generator.

The Generator is supplied with a 600 amp output breaker which feeds a second 600 amp breaker inside the building feeding a 4 pole transfer switch.

He shows rods at both locations along with metallic conduit and a #1 EGC.

I called him and humbly pointed out the error which didn't go over very well, kind of like I dare you question my design.

He tries some explanation of the inside switch being the service disconnect and the #1 EGC being necessary to clear a fault back to the Generator and must be grounded and bonded at both locations.

I said, WHAT, this is a triple path which current will divide and travel on all the time, not just during a fault.

The more he tried to rationalize his design the more ridiculous it got.

I'm used to working with engineers as a team, not them against me.

He finally says to remove the internal neutral case connection inside the Gen and run PVC.
(bottom line is he wants me to have buy the #1)

I tell him he needs to write a letter instructing me to alter the Generator factory wiring, and we can do the same thing by isolating the EGC and grounded conductors in the inside switch, this also didn't go over well. (again he wants me to buy the #1

Roger

[ October 05, 2003, 02:46 PM: Message edited by: roger ]

 

[bennie]

Re: grounding a sub panel

Roger: Very good article. My next myth expose is the requirement for removing the neutral/ground bond in a generator.

All fault current should return on the neutral conductor, not a smaller equipment ground conductor. Parallel paths can be avoided by not providing the parallel path.

An internal fault, in the generator winding to the core, will appear as zero voltage on the faulted phase sensor. The exciter will immediately be dumped. This takes place much faster than an inverse time circuit breaker that may never trip, due to the absence of sufficient current. A faulted power production device is a lot different than a power consuming device.

 

[iwire]

Re: grounding a sub panel

Originally posted by bennie:
"there is no code section that requires a neutral/ground separation at a panel". The code section is to eliminate parallel neutral paths. The method for doing this is not spelled out.

Bennie I know you are aware of the following code section, what part of this allows me to use the Neutral for the ground at a panel that is not the main panel (which most of call a sub panel but you will not)

250.142 Use of Grounded Circuit Conductor for Grounding Equipment.
(B) Load-Side Equipment. Except as permitted in 250.30(A)(1) and 250.32(B), a grounded circuit conductor shall not be used for grounding non?current-carrying metal parts of equipment on the load side of the service disconnecting means or on the load side of a separately derived system disconnecting means or the overcurrent devices for a separately derived system not having a main disconnecting means.

Originally posted by bennie:
A three wire cable assembly, in a dwelling, does not create parallel paths.

That is true in a wood frame dwelling, but the code is to fit all dwellings and there are areas that will frame an entire house with metal studs.

Originally posted by bennie:
Three wire, cable supplied panels, were supplied by SE cable, W/bare neutral/ground conductor until 1996.

Bennie can you tell me what changed in 1996?

I have been in since 1982 or so and that has not been permitted during that time, can anyone from any area say that they where permitted to use a 3 wire feeder to a second panel in a dwelling unit in any year?

Originally posted by bennie:
The change to 4 wire, with a reduced size safety ground conductor for fault clearing was started in 1996, by a select committee. No one could contest, or debate, the personal decisions, and misunderstandings that created the changes.

Can you explain what this "select committee" changed?

Bennie I really think you should make clear that a lot of what you post is your personal opinion.

There are people here that are new to the trade and it is possible that you are making a tough part of the code (250) even more confusing.

 

[bennie]

Re: grounding a sub panel

Sorry, I assumed everyone would determine the statements I make are my own, and anyone accepting them as fact is at their own discretion.
This is why I use my real name.

I also research most of the subjects before I submit an opinion or perspective.

Yes there is a lot of confusion in 250. You will notice that panels, other than service panels are not included in the NEC.

All panels during one era were regarded as service panels. There was and still no such thing as a sub panel. The old knob and tube method did not present a parallel path for the load except through the earth. The earth is permitted.

The service panel is grounded by the neutral/ground on the load side. The neutral/ground service entrance conductor is grounded on the supply side one time, and that is at the transformer. There was a time when the meter base was the ground electrode conductor point.

Panels in buildings with no parallel path available were allowed to be fed by 3 wires until 1996. The basis for enforcement was the article concerning using the neutral/ground for equipment grounding on the load side.

When the 4 wire circuits for dryers and ranges was introduced in 1996, the 3 wire to a panel also went away.

I'm sure there are many reading this who have fed panels from the service with 3 conductor SEC.

Only panels with metallic grounding path required separation of the neutral ground at a loadcenter.

Anyone out there who have witnessed 3 wire feeders to panels please jump in. I need all the help I can get.

 

[iwire]

Re: grounding a sub panel

1990 NEC

250-61(b) Load-Side Equipment. A grounded circuit conductor shall not be used for grounding noncurrent-carrying metal parts of equipment on the load side of the service disconnecting means or on the load side of a separately derived system disconnecting means or the overcurrent devices for a separately derived system not having a main disconnecting means.

Exceptions to this where for ranges and dryers, separate buildings, meter enclosures, 710-72(e)(1), 710-74 and DC systems.

What part of this article (pre 1996) would have allowed 3 wire feeders for any panels?

Bennie I have great respect for your knowledge and I am not saying you are wrong in your description of how a separate ground and neutral has disadvantages like a difference of potential between them and a reduced conductor size for fault current.

But I do dispute your opinion that the NEC allowed 3 wire panel feeders to be installed as recently as 1996.

[ October 05, 2003, 06:23 PM: Message edited by: iwire ]

 

[bennie]

Re: grounding a sub panel

Iwire: I respect your opinion and dedication to the industry also.

The error in interpretation of 250.24.5 is in the definition of a circuit conductor, versus feeder cable.

A circuit conductor can not be used for equipment grounding, there is nothing about a feeder cable grounding a panel, when there will be no undesirable current flow. Like I say "sub-panels do not exist, this is merely a panel.

How long have you noticed ground buses in panels?

The only reason for separating the ground from the neutral conductor is to prevent undesirable current flow. When there is no parallel current flow, there is no reason for the separation.

 

[karl riley]

Re: grounding a sub panel

Bennie, your last sentence I would agree with. The only problem is that even in wood frame buildings using NM cable there is no way to be sure that parallel paths will not exist for neutral current; hence there is no way that the Code could condone 3-wire to subpanels.

As an example, all the circuits going out of a subpanel contain EGCs. Some of the EGCs may be connected to metal cases such as a washer or garbage disposal or any number of appliances which may in turn be connected to or touching a water or gas pipe or vents or ducts, etc. which then become parallel paths for the neutral back to the service point.

In this case some neutral from the subpanel neutral bus backfeeds through the EGC to the connected water pipe, which gives a low impedance path through the building back to the service point.

I am not theorizing. I know this because I trace these net current paths with clamp-on ammeters when I am called into buildings with magnetic field EMI problems.

I would like to add for those who are getting their bearings that in no way are you allowed to run 3-wire to a subpanel. Neither now or pre 1996. I think 1937 was the earliest we could trace this prohibition back to.

The exception for dryers and electric ovens was the 1996 change. Separate buildings could still be fed with 3-wire even when there was a parallel path for neutral in 1996. That was corrected in the next NEC. Now 3-wire to separate building is only allowed if there is no metallic parallel path.

Karl

 

[iwire]

Re: grounding a sub panel

I agree with you from a electrical point of view that if there are no parallel paths that there is no need for separation, other than the fact I must follow the NEC

I still feel that the NEC requires us to provide separation whether there is a parallel path or not.

I can not agree with your interpretation of 250.24(5)

250.24(5) Load-Side Grounding Connections. A grounding connection shall not be made to any grounded circuit conductor on the load side of the service disconnecting means except as otherwise permitted in this article.

The term circuit conductor IMO is all encompassing, it could be service conductors, feeder conductors or branch circuit conductors.

IMO This article would use the term "Branch Circuit Conductor" if they did not want the article to apply to feeders.

Anyone else have some opinions on this, for or against my position, all views welcome.

Also, Thanks for the kind words Bennie

Bob

[ October 05, 2003, 08:08 PM: Message edited by: iwire ]

 

[karl riley]

Re: grounding a sub panel

Correction to my post: It may have been in 1996 that 3-wire could not be used for a separate building when there was a parallel metallic pathway between buildings.

Karl

 

[karl riley]

Re: grounding a sub panel

Wow! I shouldn't post after 8PM! In my post I meant to say that the exception that allowed dryers to be fed by 3-wire was revoked in 1996.

Karl again

 

[bennie]

Re: grounding a sub panel

If a feeder cable is considered to be a circuit conductor, there is a new slant on the feed to a separate building. We can't have both. It has to be a feeder or a circuit conductor.

 

[don_resqcapt19]

Re: grounding a sub panel

Bennie,
What is a "circuit" conductor? I know what a "feeder circuit conductor" is and I know what a "branch circuit conductor" is. I have no idea what a "circuit conductor" is.
Don

 

[hurk27]

Re: grounding a sub panel

Bennie, In the definitions it uses the term circuit conductor in the discription of the main bonding conductor now this main bonding conductor is between the grounding and the service neutral?
now I think there is a need to look at the artical inwhich the text is taken to determine which conductor they are talking about.
In this case they are talking about a curcuit conductor that is part of the service, it is still a curcuit conductor only it carry's the service neutral current. A circuit conductor can be any conductor that is part of a circuit, not just a branch circuit. in this case it is part of the feeder's feeding a subpanel it is still a circuit conductor. Why would they be talking about a branch circuit in a artical that is about
service wires?

250.24 Grounding Service-Supplied Alternating-Current Systems.

100.
Bonding Jumper, Main. The connection between the grounded circuit conductor and the equipment grounding conductor at the service.

 

[pierre]

Re: grounding a sub panel

Don
Without qualifying if it is a branch, or feeder circuit conductor my thoughts are they mean any circuit conductor. That is the way it is worded in section 250.24(A)(5).
I think they are trying to say that the practice of grounding a grounded conductor on the load side of the service is objectional for any circuit. I know that is not news to you, but it might help clarify what a 'circuit conductor' is.

An EGC sized according to 250.122 and properly installed will function as designed and I do not see it necessary to be the same size as the ungrounded conductors, although it is not prohibited from being the same size. Economics have a play in this.

Grounding the Grounded Conductor on the load side of the service to me is potentially dangerous and a practice that is not understood by as many as should be. The evidence is in how many panels we see in the field wired incorrectly. Karl said it well when he said that even in wood structures there are many ways this problem can rear its ugly head.
The electrons do not care how they get home, even if it is through a persons body, they will take any and all paths. The death of many poor souls is proof of that.

 

[Guest]

Re: grounding a sub panel

Half my town was built around 1957 and all the "sub-panels" I've come across (including my own) were originally wired 3-wire from the meter base to the "sub-panel(s)". That's twenty years after 1937 and nearly 40 years before 1996.

[ October 06, 2003, 01:12 AM: Message edited by: awwt ]

 

[hurk27]

Re: grounding a sub panel

Wow! You mean there was no protection of the entrance wire going to these panels or they close to the outside of the building? these would not follow what I think is a sub panel. I think these would be an extention of the service as they connect to the meter base. and would and could cause paralleling of the neutral of there is any common water pipes that also connect's between the two service panel. or duct work, gas pipe ETC...

 

[Guest]

Re: grounding a sub panel

The meter base is on the outside corner of the garage. Back-to-back inside the garage is a FPE panel with 8-handles, but a tie-bar to bring it down to six or less handles. Of course since this is the "main" panel the N-G are bonded. There is a water pipe ground wire from this "main" panel to where the water service enters the dwelling.

One of the tie-bar breakers is a sub-feed to another FPE panel in the garage on the common wall (fire wall) to the dwelling. This "sub-panel" has the majority of the dwelling circuits including the stove (3-wire); oven (3-wire); dryer (3-wire); lights; kitchen; bathroom; etc. This subpanel had N-G bonding too.

../Wayne C.

[ October 06, 2003, 01:37 AM: Message edited by: awwt ]

 

[iwire]

Re: grounding a sub panel

Originally posted by awwt:
Half my town was built around 1957 and all the "sub-panels" I've come across (including my own) were originally wired 3-wire from the meter base to the "sub-panel(s)". That's twenty years after 1937 and nearly 40 years before 1996.

I am not sure what you are saying here Wayne, my own house is 3 wire from meter base to my main panel, there is no breaker at the meter so this is code compliant.

If I add another panel it would need to be a 4 wire feed.

 

[Guest]

Re: grounding a sub panel

Originally posted by iwire:
<snip>I am not sure what you are saying here Wayne, my own house is 3 wire from meter base to my main panel, there is no breaker at the meter so this is code compliant.

If I add another panel it would need to be a 4 wire feed.

I was trying to answer Hurk27's (Wayne#1) "Wow!" question. The second panel which is a "sub-panel" only had a 3-wire feed and a bonded N-G. The second panel is about 35 feet away from the main panel.

Subpanel is in quotes for those that call it something else.

../Wayne C.

[ October 06, 2003, 01:20 PM: Message edited by: awwt ]