Grounding separately derived systems

[bdktoys]

I have an existing condition where I have a 480 volt feed from the service switchboard to a transformer. From the transformer the wiring goes into a wire way where it is tapped off to 3 disconnect switches and a panelboard. The disconnect switches then feed other panelboards through out the building.
The issue:
The disconnect switches and panelboard have the neutral and ground bonded and, the transformer also has the neutral and ground bonded. There is a ground from the transformer to building steel as well.
I understand that the neutral/ground bond should not be in both places.
Where should the correction be made. Should the neutral/ground bond be separated at the transformer or at the disconnect switches and panelboard? And why?

 

[pete m.]

The system bonding jumper should be located at the same place that the grounding electrode conductor is terminated. You should not have a system bonding jumper at both locations unless you meet the requirements of 250.30(A)(1) exception 2.

Pete

 

[augie47]

250.30 allows the bond to be anywhere between the source (transformer) to the first disconnecting means but not at both (it may be at both if it meets the somewhat rare exception).
The grounding electrode conductor should be connected at the same point (with some exceptions).
I have found the most common point is at the transformer.
If you elect to make the connection at the transformer you need to make sure you have a system bonding jumper to the 1st disconnect.
A Neutral-ground at the panel is a violation of 250.24(A)(5),
(unless a separate building is involved), but again you need to assure you have a EGC meeting 250.118 to the panel.

 

[infinity]

I have found the most common point is at the transformer.
If you elect to make the connection at the transformer you need to make sure you have a system bonding jumper to the 1st disconnect.

I believe that the 2011 NEC has changed the terminology and now refers to this connection as the Supply Side Bonding Jumper (SSBJ).

 

[augie47]

I believe that the 2011 NEC has changed the terminology and now refers to this connection as the Supply Side Bonding Jumper (SSBJ).

figures.. I just learned to say the other

 

[RUWired]

From the transformer the wiring goes into a wire way where it is tapped off to 3 disconnect switches and a panelboard.

This would be a violation of article 240.21(C). You would need to run three seperate feeders from the transformer.

Rick

 

[bdktoys]

The system bonding jumper should be located at the same place that the grounding electrode conductor is terminated. You should not have a system bonding jumper at both locations unless you meet the requirements of 250.30(A)(1) exception 2.

Pete

I don't totally understand what exception 2 tells me. How would you know if a second parallel path is established or not.

 

[pete m.]

I don't totally understand what exception 2 tells me. How would you know if a second parallel path is established or not.

Any conducting path regardless if it is intentional or not that will conduct neutral current would be a violation to the exception.

Pete

 

[RUWired]

I don't totally understand what exception 2 tells me. How would you know if a second parallel path is established or not.

IE to what Pete said would be, an equipment ground conductor, a metal conduit, a conduit rack, between the transformer and panel would create a parallel path.

Rick

 

[PaulWDent]

[QUOTE A Neutral-ground at the panel is a violation of 250.24(A)(5),
(unless a separate building is involved).[/QUOTE]

It is always a very bad idea to bond the neutral to ground at two places, EVeN when one end is a separate building . The reason is that it creates two paths for the return current in parallel - one through the neutral wire and one through the ground. The ground wire may not in fact be sized to take the share of the current it gets.

The solution for a separate building should be to use an isolating transformer at source, with neither hot nor neutral grounded at the transformer, and to bond the neutral to the local ground at the destination only.

 

[infinity]

It is always a very bad idea to bond the neutral to ground at two places, EVeN when one end is a separate building . The reason is that it creates two paths for the return current in parallel - one through the neutral wire and one through the ground. The ground wire may not in fact be sized to take the share of the current it gets.

Do you mean ground as in a conductor or the earth?

 

[Rockyd]

Read 250.30(A)(1)Ex. No.2 (a) & (b)

Use one or the other, but not both - in the transfomer case is typical, but not mandatory.

Also see 250.32(B) Ex and the conditions required.

Check out 250.142(A) & (B)

That should help clear the issue up.

 

[ActionDave]

Read 250.30(A)(1)Ex. No.2 (a) & (b)

Use one or the other, but not both - in the transformer case is typical, but not mandatory.

If you check out most of the transformers in my area you will find the bond in the first disconnect.

The head inspector liked it that way. He never required it, but if anybody asked how to bond the secondary or failed an inspection because of improper bonding he would tell them, "Do it just like you would at the service." The result was a lot of electricians learned to do it "Steve's Way" and passed it on.

Since none of his kids became electrical inspectors this is his legacy.