Main Bonding Question

[Phil86]

I have the following situation and am not sure how to correctly provide main bonding:

1) The system is 480 V, 3-phase, 4-wire, grounded-wye.
2) The service incorporates a utility-owned CT cabinet for revenue metering.
3) A single set of service conductors leaves the CT cabinet and feeds an enclosed 400 A circuit breaker (ECB). The ECB is service entrance rated.
4) The 400 A ECB feeds an automatic transfer switch. The 3-pole automatic transfer switch feeds a 400 A main breaker panelboard.
5) A second set of service conductors from the CT cabinet feeds a second 400 A breaker in the same panelboard.
6) The two circuit breakers are kirk-key interlocked such that only one can be closed at any point in time.
7) The purpose of the direct feed from the CT cabinet to the panelboard is for temporary maintenance bypass of the 400 A ECB and the ATS. The 400 A ECB and the ATS will be energized during normal operation.

Question: Where should the main bonding occur in the above scenario? I'm struggling with the fact that normally the main bonding would be in the 400 A ECB; however, when operating in bypass, the main bonding would need to be in the 400 A panelboard. I can't figure out how to isolate the two and comply with the NEC. Any help is appreciated.
:?:?

 

[infinity]

So you have two service disconnects both in the same enclosure?

 

[Smart $]

A good question but there is no Code compliant answer. It is a flawed design.

BTW, the design does not bypass the ECB. Only the ATS can be bypassed.

 

[Smart $]

So you have two service disconnects both in the same enclosure?

If I am reading his description correctly, one service disconnecting means (SDM; the ECB here) is in its own enclosure, while the second is in the panelboard supplied by the first.

 

[Grouch1980]

A good question but there is no Code compliant answer. It is a flawed design.

BTW, the design does not bypass the ECB. Only the ATS can be bypassed.

Smart $, you mean that the second set of service conductors should tap on the load side of the ECB, instead of coming straight out of the CT cabinet? I imagine that would alleviate the main bonding issue, since you would only have to bond at the ECB now, and not at the panelboard?

 

[Smart $]

Smart $, you mean that the second set of service conductors should tap on the load side of the ECB, instead of coming straight out of the CT cabinet? I imagine that would alleviate the main bonding issue, since you would only have to bond at the ECB now, and not at the panelboard?

No. Switched to bypass mode, the line side of the ECB is still energized, so it is technically not bypassed. The design is flawed because a main bonding jumper is required in each service disconnecting means. With the second service disconnect in an enclosure also connected to the load side of the first disconnecting means, you can't have a main bonding jumper between the grounded conductor and the grounding bus and at the same time keep the grounded conductor from the first disconnecting means isolated from the ground bus. Amounts to two requirements that cannot be met in the same enclosure.

 

[Grouch1980]

No. Switched to bypass mode, the line side of the ECB is still energized, so it is technically not bypassed. The design is flawed because a main bonding jumper is required in each service disconnecting means. With the second service disconnect in an enclosure also connected to the load side of the first disconnecting means, you can't have a main bonding jumper between the grounded conductor and the grounding bus and at the same time keep the grounded conductor from the first disconnecting means isolated from the ground bus. Amounts to two requirements that cannot be met in the same enclosure.

Hmmm... there's something i'm not seeing. i attached a sketch... is this how you're interpreting the OP's description? (and Phil, please let me know if this is what you're describing). Wouldn't the ECB be bypassed? Or did i draw this wrong? :weeping:

 

[texie]

I think my first question is why the OP would even want to do this. Why not just use a bypass/isolation ATS that is made for this?

 

[Dennis Alwon]

I don't get the diagram. You have a loop from the first disconnect to the second????

 

[Grouch1980]

I don't get the diagram. You have a loop from the first disconnect to the second????

The 2nd service disconnect is located in the panelboard. it's interlocked with the other main breaker in the panelboard with a kirk-key interlock so only one main breaker is on at one time. The 1st and 2nd service disconnects are isolated from each other. Unless i drew something wrong. Let me know!

 

[Grouch1980]

I don't get the diagram. You have a loop from the first disconnect to the second????

The 'K' symbol is the kirk key interlock the OP mentioned in his post. so there's no loop. The 2 main breakers in the panel are isolated.

 

[Chamuit]

The 'K' symbol is the kirk key interlock the OP mentioned in his post. so there's no loop. The 2 main breakers in the panel are isolated.

A Maintenance Bypass Panel?

 

[Smart $]

The 'K' symbol is the kirk key interlock the OP mentioned in his post. so there's no loop. The 2 main breakers in the panel are isolated.

Aye... no loop on the ungrounded conductors. However, the grounded conductor is not switched, and each service entrance conductor set has an associated grounded conductor. Going by the first disconnect, the grounded conductor and EGC bus must be separated in the panel... but going by the second disconnect, they must be bonded by a main bonding jumper. There is no possible way to meet both requirements.

Bypass, by definition, is to completely de-energize equipment while still powering said equipment's load by alternate means.

As drawn, the ECB is always energized, so it is not bypassed.

 

[Grouch1980]

Aye... no loop on the ungrounded conductors. However, the grounded conductor is not switched, and each service entrance conductor set has an associated grounded conductor. Going by the first disconnect, the grounded conductor and EGC bus must be separated in the panel... but going by the second disconnect, they must be bonded by a main bonding jumper. There is no possible way to meet both requirements.

Bypass, by definition, is to completely de-energize equipment while still powering said equipment's load by alternate means.

As drawn, the ECB is always energized, so it is not bypassed.

oh woops. i thought he was talking about a loop on the ungrounded phase conductors. i see the loop you guys are talking about now... makes sense.