GEC and H20 Bond

[jar546]

Here is a good one to think about.

Overhead service 120/240 to a single gang meter base. From the meter base there are two main panels. One "A" is located in the basement and the other "B" is located on the other side of the house. Immediately inside the basement where the SE comes through the wall is Panel "A" and right along side it is the disconnect for Panel "B".

There is a #4 Cu GEC coming from a CEE that ties directly to the Disco for Panel "B". From there it jumps over to Panel "A" to provide the GEC. There are 2 separate lugs on the neutral bar of the disco.

Panel "A" provides the 1/0 Cu bond to the water pipe.

Is this a legal install under the 2005 NEC?

Here is the story in pictures:

Panel B disco on left, Panel A on right

Inside the disco to Panel B where the GEC feeds over to Panel A

Inside Panel A which is the only point that the H20 bond comes from

Subfed Panel B that has issues with the equipment grounding bars

 

[chris kennedy]

Re: GEC and H20 Bond

Subfed Panel B that has issues with the equipment grounding bars

OK, lets hear it.

 

[jar546]

Re: GEC and H20 Bond

Only 1 egc bar is connected directly to the grounding conductor. The other one (although not used yet) is only connected to the grounding conductor through the enclosure which relies on the screws that hold the bar in place only.

 

[chris kennedy]

Re: GEC and H20 Bond

Only 1 egc bar is connected directly to the grounding conductor. The other one (although not used yet) is only connected to the grounding conductor through the enclosure which relies on the screws that hold the bar in place only.

And this violates which section?

 

Re: GEC and H20 Bond

The GEC usually stops at the first point of service disconnection and is bonded to the neutral at that point. From there the additional panels are bonded to the main disconnect ground buss. At all points after the main disconnect, the ground is isolated from the neutral. The description given states two main panels which indicates two separate services.

 

[jar546]

Re: GEC and H20 Bond

Subfed Panel B that has issues with the equipment grounding bars

OK, lets hear it.

Maybe I am misapplying 250.24(B) My reasoning is as follows.

The terminal bar on the right is only connected to the enclosure via 2 screws. The terminal bar on the left is connected to the enclosure via 2 screws and also attached directly to the main equipment grounding conductor. If a 100A sub is run from that sub and the grounding conductor of the feeder to the new sub connects to that terminal bar (with a new lug of course) then it is relying on the threads of the 2 screws for connection back to the main disconnect because it is not directly connected to the main equipment ground wire that feeds that panel.

 

[chris kennedy]

Re: GEC and H20 Bond

Maybe I am misapplying 250.24(B)

Compliance with 250.24(B) happens at the other end of the house at the mains. At first I thought you where confusing this issue with the new in 08 200.2(B);

200.2(B) Continuity. The continuity of a grounded conductor shall not depend on a connection to a metallic enclosure, raceway, or cable armor.

I find nothing that says the continuity of the EGC can't be made through the enclosure, (250.118 maybe???).

Picture this, panel B is feed from its main with EMT and no wire EGC. This install is made in compliance with 250 an 358. Now I install my ground bar with the factory screws in the factory holes. The enclosure in this case is functioning in the same manner as your picture of panel B. Am I missing something here?

Respectfully,

Chris

 

[jar546]

Re: GEC and H20 Bond

No Chris, you are not missing anything. Your explanation of using EMT as the grounded conductor brings the situation into focus a little better. Since this is a sub-panel I can see where 250.24(B) would not apply no matter how much sense I think it makes. I just had another 2008 update class (we are still on 2005) and think I am in sensory overload.

The 2008 200.2(B) section would only apply to the neutral and that may be where I was confusing myself although I could swear someone went over this with me a few years ago because I alway bond the equipment ground bars together with wire.

Thanks for keeping me on the straight and narrow

 

[EPrice]

Re: GEC and H20 Bond

Jeff,

I wonder if the conversation you were remembering from a few years ago might have involved the grounding electrode conductor. At the service, if there is a neutral bar and an equipment grounding bar, IMO the GEC must land on the neutral bar rather than the equipment grounding bar unless the two are connected with a wire or busbar. My reference for that would be 250.24(A) with particular attention to 250.24(A)(4).

 

[peach]

Re: GEC and H20 Bond

I think this situation is a good reason to have an exterior disconnect at the service (for the fire boys).

The GEC is primarily for lightning.. not a fault within the building (that's that the EGC is for)... which is why the equipment grounding conductor needs to be sized.. the Grounding Electrode Conductor needs to be #6 copper max.. period.

 

Re: GEC and H20 Bond

The essential function of the grounding electrode is to dissipate over-voltages into the earth. These overvoltages can be caused by high-voltage conductors being accidentally connected to the lower-voltage system such as by a failure in a transformer or by an overhead conductor dropping on the lower-voltage conductor. Overvoltages can also be caused by lightning.

Grounding electrode conductors are sized per table 250.66 based on the size of the largest ungrounded service entrance conductor except as permitted in 250.66{A} thru {C}.

250.66{A} If the GEC is the sole connection to a rod, pipe or plate electrode then it is not required to be larger than #6 awg.

250.66{B} Where the GEC is the sole connection to a concrete encased electrode it shall not be required to be larger than a #4 awg.

250.66{C} Where the GEC is the sole connection to a ground ring it shall not be required to be larger than the conductor used for the ground ring. {the minimum size of a ground ring is #2 awg}

The essential function of equipment grounding conductors is to provide a low impedance path for a ground fault current to get back to the source. ie. the transformer. Without a low impedance path back to the source there is the possibility that the overcurrent device will not function. The EGC is sized according to the rating or setting of the circuit overcurrent device in the circuit ahead of the equipment, conduit,etc. {table 250.122}

The best way to figure it all out is a line drawing. In the pictures I see too many places where the ground and neutral are bonded together thereby providing parallel paths for a ground fault to follow back to the source. The idea that current will follow the path of least resistance is wrong because current will flow on every path. When the current has multiple paths there is the possibility that not enough current will flow on any path to trip an overcurrent device.

 

[chris kennedy]

Re: GEC and H20 Bond

In the pictures I see too many places where the ground and neutral are bonded together thereby providing parallel paths for a ground fault to follow back to the source.

I believe this would have been a compliant install if the tap conductor had been a 1/0 as per;

250.64(D) Service with Multiple Disconnecting Means Enclosures. Where a service consists of more than a single enclosure as permitted in 230.71(A), grounding electrode connections shall be made in accordance with (D)(1), (D)(2), or (D)(3).

(1) Grounding Electrode Conductor Taps. Where the service is installed as permitted by 230.40, Exception No. 2, a common grounding electrode conductor and grounding electrode conductor taps shall be installed. The common grounding electrode conductor shall be sized in accordance with 250.66, based on the sum of the circular mil area of the largest ungrounded service-entrance conductor(s). Where the service-entrance conductors connect directly to a service drop or service lateral, the common grounding electrode conductor shall be sized in accordance with Table 250.66, Note 1. A tap conductor shall extend to the inside of each service disconnecting means enclosure. The grounding electrode conductor taps shall be sized in accordance with 250.66 for the largest conductor serving the individual enclosure. The tap conductors shall be connected to the common grounding electrode conductor by exothermic welding or with connectors listed as grounding and bonding equipment in such a manner that the common grounding electrode conductor remains without a splice or joint.

The idea that current will follow the path of least resistance is wrong because current will flow on every path. When the current has multiple paths there is the possibility that not enough current will flow on any path to trip an overcurrent device.

:?: :?: :?:

 

[jar546]

Re: GEC and H20 Bond

The idea that current will follow the path of least resistance is wrong because current will flow on every path. When the current has multiple paths there is the possibility that not enough current will flow on any path to trip an overcurrent device.

True

 

Re: GEC and H20 Bond

Both 230.40 and 230.71 take me back to 230.2. It does not appear as though this installation would fall in any category allowing multiple services. I really don't have that good a grasp of the condition presented in this thread or electricity for that matter. In an effort to better acquaint myself with the "why" of the code we enforce, I have been studying Soares Book on Grounding and Bonding. It is from that book that I will now quote:

"Current, be it normal current or fault current will take all of the paths available to it to try to return to its source. If the grounded conductor [neutral] and equipment grounding conductors are connected at points downstream of the service or separately derived system, such as at sub-panels, multiple paths will be available on which the current will try to return to the source. This can lead to normal neutral current on water piping systems, conduit, equipment grounding conductors, and any other electrically conductive paths and can compromise electrical safety and even overcurrent device operation in ground fault conditions."

The book does not explain how multiple paths could compromise an overcurrent device in a ground fault condition, or at least I haven't found it yet. I may have taken a leap and miss-spoke when I stated that multiple paths may not provide enough current flow to trip a breaker but that is about the only thing I can think of that would compromise an overcurrent device.

Anyway, on Monday I will ask the County Chief Electrical Engineer and share his explanation.

 

Re: GEC and H20 Bond

Well here it is Tuesday and I haven't been able to get with the engineer. He wants to have lunch next week and go over it then.

 

[timnetzley]

Re: GEC and H20 Bond

Been wiring for more than a few years, now an inspector. For my money, tigerloose, you are right on.

 

[EPrice]

Re: GEC and H20 Bond

The idea that current will follow the path of least resistance is wrong because current will flow on every path. When the current has multiple paths there is the possibility that not enough current will flow on any path to trip an overcurrent device.

The first part of this statement is correct, but if I understand what you are saying in the last sentence, then it is not correct. The current flowing through the overcurrent device is what causes it to trip. The current flowing through the overcurrent device will be equal to the sum of the current flowing back over the multiple return paths. There is a formula for calculating the resistance of parallel paths, which I will not go into, but the total resistance of parallel paths will always be less than the resistance of any one of the individual paths. If at least one of the paths is of low enough resistance to allow enough fault current to trip the overcurrent device, then the sum of the current flowing over the multiple paths will be enough to trip the overcurrent device.

There are reasons why the code prohibits re-grounding of the neutral downstream from the service, but it is not to prevent multiple paths for fault current. It is to prevent multiple paths for neutral current with the resulting possibility for shock.