SDS generator under 2002

[ceb58]

At work I have discovered 2 gen. sets that were installed in 04 under the 2002 code at communication sites. The gen.sets were installed as SDS using a three pole ATS. The installers picked up on the underground single point grounding ring. They cad welded to the ring and bonded to gen. frame and then bonded the neutral on a bonding bar. From the gen. set, which has a breaker, they ran 2 ungrounded cond. and the grounded conductor to the gen. disco on the building. They once again picked up on the underground ring and ran the grounding conductor,through separate conduit, to the disco where they again bonded the neutral.

I know that bonding to neutral in two places was wrong but what about not running a EGC with the phase conductors. Boss said the sites were thoroughly inspected at the time and were compliant. My thought is the gen. has the first OCP an EGC should have been ran to the disconnect on the building and not the way they did it.

 

[Smart $]

An EGC was not required to be run with the circuit conductors under 2002, IIRC.

I'm fairly certain bonding the disco with a GEC of the structure's GES was required. What I'm not certain is whether bonding to neutral (again) was permitted or required. (OK, tell me I'm no help )

 

[ceb58]

An EGC was not required to be run with the circuit conductors under 2002, IIRC.

I'm fairly certain bonding the disco with a GEC of the structure's GES was required. What I'm not certain is whether bonding to neutral (again) was permitted or required. (OK, tell me I'm no help )

No, you were a help. When I look at the system I step back and pretend it is a normal service. Now with that I know of no code that would allow bonding the neutral in two places.

Generator with OCP neutral and ground bonded. Same as main service bond neutral and ground.

Feeders leave gen. to a disconnect with OCP. Same as leaving main panel to a sub panel.

I see it as a violation bonding the N-G in the disconnect the same as if you bonded the N-G in a sub panel.

Or am I off base?

 

[hurk27]

Wow I thought I was having deijavu until I thought of this
Thread

 

[Smart $]

No, you were a help. When I look at the system I step back and pretend it is a normal service. Now with that I know of no code that would allow bonding the neutral in two places.

Generator with OCP neutral and ground bonded. Same as main service bond neutral and ground.

Feeders leave gen. to a disconnect with OCP. Same as leaving main panel to a sub panel.

I see it as a violation bonding the N-G in the disconnect the same as if you bonded the N-G in a sub panel.

Or am I off base?

No, you are not off base. But as long as the grounding conductor (GEC) is not exposed, I don't see it as creating a personnel safety concern. With all the finer points there are to grounding, I don't want to go back to the 2002 Edition to determine whether or not it is a violation thereunder, because I will likely never install under 2002 ever again :roll: It does create a parallel path for neutral current, and IIRC 'objectionable current' was the same then, or darn close to what it is now.

On that note, it wouldn't bother me in the least if you removed the N-G bond at the disco'.

 

[RUWired]

On that note, it wouldn't bother me in the least if you removed the N-G bond at the disco'.

I'm not sure if the alternate source is the other gen set or a utility source that is connecting to the transfer switch. CEB said he had two gen sets but that could be two locations with a gen set at each location.

If the alternate source is the gen set, i agree that the n-g bond at the disconnect should be removed to keep compliance with 250.30(A), A1 and exception # 2 of A1. This method would require an EGC be installed unless feeders are in a metal raceway.

If the alternate source is the utility source, the n-g bond shoud be removed at the gen set as well as at the disconnect as you said, because of the solid interconnection at the ATS. This would then require an EGC be installed unless the raceways were metallic.

Rick

 

[ceb58]

I'm not sure if the alternate source is the other gen set or a utility source that is connecting to the transfer switch. CEB said he had two gen sets but that could be two locations with a gen set at each location.

If the alternate source is the gen set, i agree that the n-g bond at the disconnect should be removed to keep compliance with 250.30(A), A1 and exception # 2 of A1. This method would require an EGC be installed unless feeders are in a metal raceway.

If the alternate source is the utility source, the n-g bond should be removed at the gen set as well as at the disconnect as you said, because of the solid interconnection at the ATS. This would then require an EGC be installed unless the raceways were metallic.

Rick

Two different sites, each has utility and gen. The generators are installed as separately derived systems. The neutrals are switched in the ATS. The N-G bond should be in the gen. I can see it would be OK to use the underground halo as they did but not re-bonding the N-G in the disconnect as they did because the gen. sets have OCP on them.

 

[RUWired]

OK the picture is better. Then i would agree with you that the bonding jumper should have been removed at the disconnect and an equipment ground conductor should have been ran with the feeder.

If the GEC had not been common, the n-g bond would have been ok then in the 2002 and now using the exception# 2. But since the GEC is common the n-g bond should not have been inplace at the disconnect then or now.

Equal potential grounding is ok between equipment, but the placement of bonding jumpers have to be thought out.

Rick