Grounding electrode conductor brought to service transformer

[noel westphal]

I have a electrical plan which is requiring a #3 grounding electrode conductor from a 400 amp 3 phase 277/480 volt service diconnect back to the utilitys transformer? Why would they require this? The service conductors are 4-500 kcmil. conductors and 1-#3 grounding conductor.

 

[fc]

You sure it said grounding electrode conductor ? or equipment grounding conductor ?
Who's plan is it? The utility co.?
For equipment grounding conductor use 250.122 --400 amp rating would be #3 copper.
For grounding electrode conductor would be table 250.66 and would be 1/0 copper

 

[coulter]

noel -

Nothing that comes to mind that sizes the grounding conductor back to the xfm by T250.122. There are a couple of reasons why one would run a conductor back, but I would expect the conductor to be sized by 250.66.

1. effectively grounded 480/277: xfm is treated as an SDS. There is a bonding jumper at the xmf or at the first disconnect - but not both. Either the xfm belongs to the customer or there is a special deal with the utility. I would have expected a 1/0cu (250.30, 250.102, 250.66)

2. impedance grounded 480/277: No bonding jumper at the first disconnect. Again, either the xfm belongs to the customer or there is a special deal with the utility. There won't be any 277 loads in the plant. Conductor sized by 250.36G1 - 250.66 (1/0) or 250.36G2 and 250.36B - as small as #8

3. Whoops - there is one other I have seen that did size the EBC by 250.122 (#3). The service point was at the xfm primary. The xfm OCPD watched the primary current, and also watched the secondary current (differential protection). There was also a function that would trip the primary CB for secondary over current. This was considered to be overcurrent protection at the xfm, protecting the secondary conductors. So the secondary conductors were considered feeders.


And there is alyways one other - I don't have a clue, and neither did the person that laid it out

carl

 

[noel westphal]

I am assuming it is a equipment grounding conductor base on the size (# 3) but i don't see the reason for ether one to be running to the utility's transformer from the service disconnect.

 

[noel westphal]

Carl, This project happens to be a open sided parking garage, so nothing fancy or unusual just 277 volt lighting, a possible elevator (as a alternate) and one transformer for a 225 amp 120/208 volt panel for receptacles. Noel

 

[fc]

The best thing to do is ask the person who design it. If you do lets us know I would be interested in the answer.

 

[don_resqcapt19]

Carl,

1. effectively grounded 480/277: xfm is treated as an SDS. There is a bonding jumper at the xmf or at the first disconnect - but not both.

If this is a customer owned SDS, then you are correct, but if it is an utility transformer, there will be a bonding jumper at both the transformer and the service disconnect.
Don

 

[don_resqcapt19]

I do not know of any reason or code compliant way to install this extra conductor, assuming that the transformer is owned by the utility.
Don

 

[coulter]

... but if it is an utility transformer, there will be a bonding jumper at both the transformer and the service disconnect. ...

Don -
I understand that would be the norm, especially if one were dealing with cookie cutter installations that use the NEC as a design guide.

Perhaps you could give me the code section supporting that there has to be a bond at the xfm.

carl

 

[noel westphal]

Just to clear up the question of who owns the transformer, it is the utility (Xcel Energy) in this case. I will be calling the engineer will let you all know what he says. thanks Noel

 

[coulter]

... I will be calling the engineer will let you all know what he says. ...

Here's my guess - especially if you ask for supporting code sections, "mumble, mumble, whoops"

Cause if it is bonded at both ends, there is a parallel neutral current path (pretty much following Don's reasoning).

carl

 

[coulter]

...nothing fancy or unusual just 277 volt lighting, a possible elevator (as a alternate) and one transformer for a 225 amp 120/208 ...

Just an aside: I wouldn't consider any of the ones I laid out as "fancy or unusual" (especially the one about me or the designer not having a clue) - rather, the norm. Of course, I don't do houses, very little commercial, almost no single phase, more self generation that connections to a utility.

Luckly the utility engineers I deal with are pretty good and quite willing to offer excellent advice.

Well, actually I do one house.

carl

 

[noel westphal]

Carl, Well, you know engineers. Carl I really do appriciate all your help, and all the good direction i get from this forum. Thanks Noel

 

[don_resqcapt19]

Carl,

Perhaps you could give me the code section supporting that there has to be a bond at the xfm.

250.24(A)(2) requires a bond either at the transformer or elsewhere outside the building.
Don

 

[LarryFine]

Carl,

250.24(A)(2) requires a bond either at the transformer or elsewhere outside the building.
Don

But, wouldn't that bond normally not be to both ends of a parallel conductor run from the transformer's GEC to the service GEC?

I'm thinking that, like a second building service, you either bond the neutral at both ends, or run a GEC, but not both.

 

[Pierre C Belarge]

But, wouldn't that bond normally not be to both ends of a parallel conductor run from the transformer's GEC to the service GEC?

I'm thinking that, like a second building service, you either bond the neutral at both ends, or run a GEC, but not both.

As Don has mentioned, the bonding of the system depends on the type of system. If it is a Utility system, then bond both the utility transformer side and the building side. If it is a SDS, then your choice of which side, but not both.

 

[coulter]

...If it is a Utility system, then bond both the utility transformer side and the building side. ...

Humm, even if it makes a parallel path for the neutral current? Some might consider this unsafe and poor design. But it would be code compliant.

Pierre - I'm not poking fun at you, cause you are right. It's just some of the code legal requirements that are a bit ridiculous.

...If it is a SDS, then your choice of which side, but not both.

Along the same line, it is truly amazing that the physics of design for safety depends on who owns the transformer.

carl

 

[coulter]

...Along the same line, it is truly amazing that the physics of design for safety depends on who owns the transformer. ...

Now, just for a thought experiment:

Susposing the utility bought the transformer, and the install, on the customer's property, included primary metering. Probably have to include a primary CB with provision for customer as well as Utility control.

Now, is the secondary a service, or is the transformer an SDS?

carl

 

[coulter]

...250.24(A)(2) requires a bond either at the transformer or elsewhere outside the building. ...

Good call Don. Plenty close enough.

Consider the following installation:
1 Industrial site, ground mat through the entire area.

2 Utility installs the transformer on the customer property, effectively grounded 277/480.

I worked on one just like that about ten years ago. It had been installed well before my watch.

Following your interpretation, (and I believe you are code correct) code compliance requires the service is bonded and grounded at the transformer and at the first disconnect.

Amazingly the AHJ and the utility managed to figure it out.

Hummm. Can the utility supply an ungrounded 5W, 480Y and let the customer ground it? 250.20B2 would seem to say, "No".

carl

 

[don_resqcapt19]

Carl,

Hummm. Can the utility supply an ungrounded 5W, 480Y and let the customer ground it? 250.20B2 would seem to say, "No".

I agree that you can't do that.

Humm, even if it makes a parallel path for the neutral current? Some might consider this unsafe and poor design. But it would be code compliant.

Not only code compliant...code required. There are a number of places where the code requires parallel paths on the line side of the service.
1) if the there is a common metal undergound water piping system it will be a parallel path and it would not be uncommon to find 20% or more of the grounded conductor current flowing on the water pipe.
2) if the building is served by a cable TV system, the coax shield will be bonded at each occupancy making another code required parallel path. I watched as this path started a fire one night. There had been a storm with broken wires and on this one house, the only the grounded conductor in the service drop was broken. The #14 cable bonding wire was acting as the grounded conductor for the house and it got hot enough to ignite the wood siding.
3) underground phone cables also often have a metallic shield that becomes a parallel path.
4) and always the earth itself as the utility always makes a bonding connection at the transformer if there a grounded conductor and if there is also a primary grounded conductor, the two will be bonded to each other. In many cases the same conductor is used as both the primary and secondary grounded conductor.
The CMP requires that these parallel paths be created, but also have gone on to say that the same types of paths on the load side of the service disconnect are dangerous and should be prohibited. I don't understand how the electrons know what side of the service disconnet that they are on, because the CMP seems to be under the impression that the electron behavior changes from one side of the service disconnet to the other.
Don