System Bonding Jumper and Grounding for Multiple Gensets

[pyrodood]

Building with primary service being (5) 350KW gensets parallelled. Gensets feed into MSB which feed transfer switches. Back-up power is 800A utility service.
All gensets are outside the building.
Is there a system bonding jumper required in each genset enclosure? Or is only one system bonding jumper required and can it be made in MSB?
I've been trying to wrap my head around this for a while. 250.30(A)(2) Exception No. 2 Allows for a system bonding jumper to be be at both the source and first disconnecting means if there is not a parallel path for the grounded conductor. However, it's not clear in this case with multiple generators where the system bonding jumper should go.

Thanks

 

[ActionDave]

You said the backup power is the utility. That's not common but not impossible.

Where is the transfer switch? Where is the utility service disconnect?

Welcome to the forum

 

[pyrodood]

Utility Service disconnect is in MSB2
MSB, MSB2, and Transfer Switches are in the building in the Main Electrical room.
There is a 400A ATS and a few 400A Manual TS loadside of MSB they transfer to loadside of MSB2, they are all 4-pole Transfer Switches, so they switch the Neutral conductor. So I believe they separately derived systems according to the NEC. So 250.30 applies.

 

[ron]

You are using equipment names as if we know your oneline diagram.
Typically the paralleled generators would have a N-G bond at the paralleling gear, but you need to draw out a 5 line diagram to think through fault current paths especially if there is any ground fault protection / detection.

 

[pyrodood]

Sorry, ATS Is Automatic Transfer Switch. Manutal TS are Manually operated transfer switches.
MSB and MSB2 are two different Switchboards. All five generators feed to the one switchboard, the utility feeds the other switchboard. The transfer switches switch between the two sources.

 

[pyrodood]

You are using equipment names as if we know your oneline diagram.
Typically the paralleled generators would have a N-G bond at the paralleling gear, but you need to draw out a 5 line diagram to think through fault current paths especially if there is any ground fault protection / detection.

Received more information. The paralleling is done internally the gensets are connected with a low volt cable. The switchboard is just a typical switchboard.

I can see 250.30(A)(6) allows for a common grounding electrode conductor. This also requires the neutral to be bonded to the common grounding electrode conductor.

The "shall be permitted" language seems to indicate that a common grounding electrode conductor is not required but allowed. If one is not installed are each generator to be treated as a single separately derived system? Each get their own grounding electrode, and bond the neutral at each generator as well? If so all of these neutrals are going to be tied together on the same neutral bus of the switchboard, would it be allowed to have these multiple bonds of neutral to ground?

 

[pyrodood]

You said the backup power is the utility. That's not common but not impossible.

Where is the transfer switch? Where is the utility service disconnect?

Welcome to the forum

Here is the relevant portion of the single line.
My main question is, where is it required to bond the neutral to the ground? In each generator? At MSB1?

 

[ActionDave]

That's interesting. Since you said you have 4 pole transfer switches you need a bond in MSB1 if there is a common neutral bar there, and a separate EGC back to each generator case. I'm not sure about a common grounding electrode system/conductor, I think I'd just go for a ground rod at each generator. MSB 2 would have its own bond and grounding electrode system.

 

[ron]

That is an odd oneline for sure. MSB1 is rated at 4000A, yet there is only ~2100A worth of generators dangling on one end and a single point of failure where all of the power has to flow through common bus to get to all of the load ATS's. Then there is only 800A worth of utility to support that load. They should consider load flow.
In any case, as the drawing shows, each generator will get a GEC connection to the grounding electrode. The N-G bond I would suggest for simplicity at MSB1 (there is no code required GFP because it is less than 1000A, but they may add it above minimum code). Draw the 5 line diagram to see if GFP would lend better detection by moving the N-G bonds to each gen individually. The code allows either for that reason.

 

[pyrodood]

That's interesting. Since you said you have 4 pole transfer switches you need a bond in MSB1 if there is a common neutral bar there, and a separate EGC back to each generator case. I'm not sure about a common grounding electrode system/conductor, I think I'd just go for a ground rod at each generator. MSB 2 would have its own bond and grounding electrode system.

I don't think a ground-neutral bond should go at MSB1 if the generators each have a Ground-neutral bond unless you don't have a parallel path for the neutral conductors under Exception #2 250.30(A)(1). I was getting hung up on the fact that a neutral-ground bond at each generator would create 5 different ties of the neutral to ground, I wasn't sure that was correct under the code. However, under 250.30(A)(6) it says to bond the neutral at each source. Of course that is in reference to a common grounding electrode and the consensus seems to be to just go ahead and put a ground rod at each generator.

 

[pyrodood]

That is an odd oneline for sure. MSB1 is rated at 4000A, yet there is only ~2100A worth of generators dangling on one end and a single point of failure where all of the power has to flow through common bus to get to all of the load ATS's. Then there is only 800A worth of utility to support that load. They should consider load flow.
In any case, as the drawing shows, each generator will get a GEC connection to the grounding electrode. The N-G bond I would suggest for simplicity at MSB1 (there is no code required GFP because it is less than 1000A, but they may add it above minimum code). Draw the 5 line diagram to see if GFP would lend better detection by moving the N-G bonds to each gen individually. The code allows either for that reason.

I'm not sure what you mean by 5 line diagram. Do you mean instead of a single line you draw a diagram that has a line for each conductor; 3 ungrounded (hot), 1 grounded (neutral), 1 egc(ground)?

Yeah, the utility will run the ATS and one of the manual transfer switches at a time, switched by the customer.
I get that MSB1 becomes a single point of failure but is that a problem? I don't see the likelihood of that 4000A bus in the switchboard failing.

I like the idea for the system bond being at MSB1 but I wasn't sure if it was required to bond at the source. 250.30(A)(1) States that if the source is located outside the building or structure supplied, a system bonding jumper shall be installed at the grounding electrode connection in compliance with 250.30(C) which requires a grounding electrode connection at the source. It would seem the code requires the system bond (ground-neutral bond) at the same point which is at the source. Correct me if i'm wrong.

Thanks for the replies.

 

[ron]

I don't think a ground-neutral bond should go at MSB1 if the generators each have a Ground-neutral bond unless you don't have a parallel path for the neutral conductors under Exception #2 250.30(A)(1). I was getting hung up on the fact that a neutral-ground bond at each generator would create 5 different ties of the neutral to ground, I wasn't sure that was correct under the code. However, under 250.30(A)(6) it says to bond the neutral at each source. Of course that is in reference to a common grounding electrode and the consensus seems to be to just go ahead and put a ground rod at each generator.

I was not being clear. Never put the N-G bond at both places. Either at each gen OR at MSB1, or you will get parallel paths

 

[ron]

I'm not sure what you mean by 5 line diagram. Do you mean instead of a single line you draw a diagram that has a line for each conductor; 3 ungrounded (hot), 1 grounded (neutral), 1 egc(ground)?

Yes, a 5 line is like you describe, and you show the CT positions for any ground fault protection so it is easier to understand where current flows in the event of a fault.

Yeah, the utility will run the ATS and one of the manual transfer switches at a time, switched by the customer.
I get that MSB1 becomes a single point of failure but is that a problem? I don't see the likelihood of that 4000A bus in the switchboard failing.

MSB1 will be the single point of failure in this design. That is a given. The 4000A bus is oversized and is also the bottleneck for all power flow. Normally the generator connections to loads are intermingled with the load breakers so the "horizontal" bus or load bus can be sized much smaller when you consider the load flow of power coming in and out to the loads, and no one section of bus is carrying the bulk of the power from "in" to "out".

 

[ActionDave]

I don't think a ground-neutral bond should go at MSB1 if the generators each have a Ground-neutral bond unless you don't have a parallel path for the neutral conductors under Exception #2 250.30(A)(1). I was getting hung up on the fact that a neutral-ground bond at each generator would create 5 different ties of the neutral to ground, I wasn't sure that was correct under the code. However, under 250.30(A)(6) it says to bond the neutral at each source. Of course that is in reference to a common grounding electrode and the consensus seems to be to just go ahead and put a ground rod at each generator.

You're right. I kinda messed that up. One bond at MSB1 with an earth connection. I'm agreeing w Ron.

The individual ground rods would be something I would expect to see speced even if not required so that's what was rolling around in my brain when I mentioned them relating to the generator.