Offset Nipple between Meter and Disco

[shepelec]

Thanks, but...

My understanding (what I've read) is that you are required to bond the GEC to both ends of a metallic conduit it passes through because if you don't the the conduit acts as a choke on the GEC that restricts the flow of current.

Through the conductor yes, but not through the conduit.

For what it is worth, the whole point of bonding the conduit is not to bond the cabinets. That is already accomplished by bonding the neutral in the socket and main disconnect. The point of bonding the conduits is just that, to bond the conduits. If the conduit connections were not tight and a fault occured it would creat quite a mess. By bonding you ensure a low resistance path to hopefully open the primary ocp on the pole. That is why bonding jumper are only required on the line side of the main ocp.

 

[ike5547]

Through the conductor yes, but not through the conduit.

For what it is worth, the whole point of bonding the conduit is not to bond the cabinets. That is already accomplished by bonding the neutral in the socket and main disconnect. The point of bonding the conduits is just that, to bond the conduits. If the conduit connections were not tight and a fault occured it would creat quite a mess. By bonding you ensure a low resistance path to hopefully open the primary ocp on the pole. That is why bonding jumper are only required on the line side of the main ocp.

I think we are mixing two different topics here. My reply to post #27 (the reply you quoted) was regarding a separate issue outside the general topic of this thread.

The principles behind the bonding of conduits containing GECs vs. bonding in general have different purposes and objectives in mind.

 

[e57]

Thanks, but...

My understanding (what I've read) is that you are required to bond the GEC to both ends of a metallic conduit it passes through because if you don't the the conduit acts as a choke on the GEC that restricts the flow of current.

Yes - that too is my understanding as well...

Through the conductor yes, but not through the conduit.

That I disagree with both, on the topic of GEC's and conduits and enclosures with service conductors

For what it is worth, the whole point of bonding the conduit is not to bond the cabinets. That is already accomplished by bonding the neutral in the socket and main disconnect. The point of bonding the conduits is just that, to bond the conduits. If the conduit connections were not tight and a fault occured it would creat quite a mess. By bonding you ensure a low resistance path~

The code says "bonded together" By bonding only one end - this is not a low resistance path. It is both a high resistance parallel electrical path, and an inductive choke. Current passing on the conduit creates magnetic fields acting upon the neutral conductor - much like it would with a GEC in a conduit which is why both ends of it are bonded - so Faraday's Law comes in to play in a macro sense. While a PVC conduit will elliminate that - bonding both ends of metal conduit reduces the resistance of that path and reducing the effects of inductance.

~to hopefully open the primary ocp on the pole. That is why bonding jumper are only required on the line side of the main ocp.

Most POCO transformers in my area do not have any OCP at all... Even then if it did - the trip point would be much higher than the current required to damage or destroy any of the service equipment at the many buildings on it.

The main bonding jumper is excluded from being elsewhere at any point beyond the main (in most situations - except by specific exception) due to the parallel path of objectionable current it would create. Not long ago I was on a site where the plumber was getting shocked - why - because someone had not taped off neutral conductors in the rough-in and 1/2 of the neutral current was flowing on the EGC's through the building due to inadvertant connection in metal boxes on metal studs. Diverting current out on #12 neutral conductors, then back on the EGC. When he would touch the studs and his plumbing he would become another path for that current flow. From the metal studs - through the plumbing back to the main water - and back to the neutral in the main electrical panel. Sure - it was only a tingle, but it was the little sparks that would happen when he would touch copper to the tin studs that freaked him out more....

 

[shepelec]

Yes - that too is my understanding as well...

That I disagree with both, on the topic of GEC's and conduits and enclosures with service conductors The code says "bonded together" By bonding only one end - this is not a low resistance path. It is both a high resistance parallel electrical path, and an inductive choke. Current passing on the conduit creates magnetic fields acting upon the neutral conductor - much like it would with a GEC in a conduit which is why both ends of it are bonded - so Faraday's Law comes in to play in a macro sense. While a PVC conduit will elliminate that - bonding both ends of metal conduit reduces the resistance of that path and reducing the effects of inductance.Most POCO transformers in my area do not have any OCP at all... Even then if it did - the trip point would be much higher than the current required to damage or destroy any of the service equipment at the many buildings on it.

Let me elaborate on my previous statement.

When fault current flows through a GEC that is in metallic conduit that is not properly bonded the conduit will act as a choke on the GEC. This will inturn cause the fault current to flow on the conduit instead of the GEC.

The line side bonding jumpers have nothing to do with induction.
They are required to clear faults on the line side of the main OCP. Either by tripping the primary OCP or by allowing the fault to burn clear.
If induction were the issue why would they not be required throught the rest of the system?

I do find it hard to believe the POCO has no primary protection, either internal or external of the transformer, and would allow a large area to go out for a local fault. Then again I have not seen a large area of the country so it could be possible.[/B]

 

[elohr46]

Not buying it...

If there were no metal to metal contact on the un-bonded side - I would say yeah sure.... But there is... Potential difference depends on load and distance, and the current will flow on all available paths - not the one you figure is best for it....

Take a look at the images below - schematically they are no different than a bonded meter, and bonded main enclosure with a conductive path between them.

E57, the two graphics you posted shows what happens when the neutral is improperly connected which results in objectionable current flow. I agree with those graphics. The metal nipple I was refering to was properly bonded to the neutral at the service on one side and therefore not a violation. Current flow on metal parts of the service is not objectionable current.

 

[ike5547]

Through the conductor yes, but not through the conduit.

For what it is worth, the whole point of bonding the conduit is not to bond the cabinets. That is already accomplished by bonding the neutral in the socket and main disconnect. The point of bonding the conduits is just that, to bond the conduits. If the conduit connections were not tight and a fault occured it would creat quite a mess. By bonding you ensure a low resistance path to hopefully open the primary ocp on the pole. That is why bonding jumper are only required on the line side of the main ocp.

250.64(E) should never have entered into this discussion in the first place.

 

[e57]

E57, the two graphics you posted shows what happens when the neutral is improperly connected which results in objectionable current flow. I agree with those graphics. The metal nipple I was refering to was properly bonded to the neutral at the service on one side and therefore not a violation.

That - in lies the debate we are having... Just because some members of Mike Holts forum believe in the "one side only interpetation" even Mike Holt himself does not make it so. The code clearly does not say one side only, and it clearly does not say both sides either - but in my own interpetation, and a few others it does imply that both are required. So clearly the code needs clarification.

'shall be bonded together' (in 250/92(A) ) means to me, all connecting parts are bonded... Not at one end or the other - but both.

And 'electrical continuity' - 'ensured by one of the following methods' means just that - continuity by those methods - nowhere there in 250.92(B) does it say to one side or the other... ('One of the following methods' does not mean to one side only to me...)

Nor does the use of the word "SOLE". As the base of this debate of 250.92(B) also stands on this wording - "Standard locknuts or bushings shall not be the sole means for the bonding required by this section" You can clearly have a bonding bushing or locknut inside of the enclosure, and a standard locknut out side the enclosure, but IMO this does not get you out of bonding the other end of the conduit as a whole... (To provide continuity by the provided means - bonding the enclosures together.)

For that matter the code itself makes as if the grounded conductor is connected in one place only because the meter does not exist....

Current flow on metal parts of the service is not objectionable current.

The current is still there, and it is still objectionable... (250.6) But in many localities alterations to stop this current are not available. i.e I can not have SEC's in other than IMC/RMC, and no acceptable fitting comination can be devised to isolate one end. Which is why I find the contradiction near and dear... I also can not disconnect either bonding jumper. (in most common cases)

 

[e57]

250.64(E) should never have entered into this discussion in the first place.

I think as an example it is relevant.

 

[ike5547]

I think as an example it is relevant.

The requirement to bond both sides of a GEC conduit to the GEC itself is different because of an effect that occurs when you have only one single conductor in a conduit and impose a high current through it. The conduit creates a high impedance or choke.

The bonding we've been discussing here is through conduits and nipples that contain all the circuit conductors.

 

[ike5547]

Through the conductor yes, but not through the conduit.

Even if the conduit is properly bonded as per 250.64(E) over 90% of the current flows 'on' the conduit and not the conductor. The purpose of 250.64(E) is not to prevent 'objectionable current' on the GEC conduit -- because it can't.

 

[shepelec]

250.64(E) should never have entered into this discussion in the first place.

Probably not, but since you brought it up........

 

[elohr46]

That - in lies the debate we are having... Just because some members of Mike Holts forum believe in the "one side only interpetation" even Mike Holt himself does not make it so. The code clearly does not say one side only, and it clearly does not say both sides either - but in my own interpetation, and a few others it does imply that both are required. So clearly the code needs clarification.

'shall be bonded together' (in 250/92(A) ) means to me, all connecting parts are bonded... Not at one end or the other - but both.

And 'electrical continuity' - 'ensured by one of the following methods' means just that - continuity by those methods - nowhere there in 250.92(B) does it say to one side or the other... ('One of the following methods' does not mean to one side only to me...)

Nor does the use of the word "SOLE". As the base of this debate of 250.92(B) also stands on this wording - "Standard locknuts or bushings shall not be the sole means for the bonding required by this section" You can clearly have a bonding bushing or locknut inside of the enclosure, and a standard locknut out side the enclosure, but IMO this does not get you out of bonding the other end of the conduit as a whole... (To provide continuity by the provided means - bonding the enclosures together.)

For that matter the code itself makes as if the grounded conductor is connected in one place only because the meter does not exist....

The current is still there, and it is still objectionable... (250.6) But in many localities alterations to stop this current are not available. i.e I can not have SEC's in other than IMC/RMC, and no acceptable fitting comination can be devised to isolate one end. Which is why I find the contradiction near and dear... I also can not disconnect either bonding jumper. (in most common cases)

I am not basing my opinions because "some members of Mike Holt's forum" agree. I attend IAEI (Ben Franklin Chapter, Phila) monthly meetings and this subject comes up from time to time. Just recently the local POCO Rep. was a guest speaker going over services along with the senior Electrical Inspector from the Middle Department and the City of Phila Electrical Inspector and they all concurred when asked about bonding metal conduits at the service.

The 2008 Code seminar I attended was given by an IBEW Codes Instructor who also is an Electrical Engineer went into great detail on this very subject so that everyone there understood what was necessary for a proper install. "You may bond both ends of metal conduit if you like but only one side is necessary". I respect Mike Holt and all that he does but I don't see where he says to bond both sides of metal conduit between service enclosures.

 

[e57]

Even if the conduit is properly bonded as per 250.64(E) over 90% of the current flows 'on' the conduit and not the conductor. The purpose of 250.64(E) is not to prevent 'objectionable current' on the GEC conduit -- because it can't.

Thats why I added it to this discussion in the first place - it is a great example...

 

[e57]

I am not basing my opinions because "some members of Mike Holt's forum" agree. I attend IAEI (Ben Franklin Chapter, Phila) monthly meetings and this subject comes up from time to time. Just recently the local POCO Rep. was a guest speaker going over services along with the senior Electrical Inspector from the Middle Department and the City of Phila Electrical Inspector and they all concurred when asked about bonding metal conduits at the service.

The 2008 Code seminar I attended was given by an IBEW Codes Instructor who also is an Electrical Engineer went into great detail on this very subject so that everyone there understood what was necessary for a proper install. "You may bond both ends of metal conduit if you like but only one side is necessary". I respect Mike Holt and all that he does but I don't see where he says to bond both sides of metal conduit between service enclosures.

I'm not trying to impune your varacity... And I believe Mike Holt is a "One End'er" too...

If you look at it schematically - it makes little sense at all...

If you look at the very wording of the code - it says neither but seems to imply both ways depending on how you look at it. Thats the issue...

If you look at it politicaly - POCO's HATE GROUNDS OF ALL KINDS! They don't want an additional bonding jumper - so 'one end' fits their desire, and a compromise to those seeking meaning of the wording of the NEC and comfortable jumping on the "One End Bandwagon". But as mentioned one could use a bonding locknut... (Or in some areas - PVC) - then both the "Both End'ers" that may be involved and the POCO are happy...

But is still does not say one end only.... Does it?

 

[ike5547]

Probably not, but since you brought it up........

Whatever. ................................

 

[elohr46]

I'm not trying to impune your varacity... And I believe Mike Holt is a "One End'er" too...

If you look at it schematically - it makes little sense at all...

If you look at the very wording of the code - it says neither but seems to imply both ways depending on how you look at it. Thats the issue...

If you look at it politicaly - POCO's HATE GROUNDS OF ALL KINDS! They don't want an additional bonding jumper - so 'one end' fits their desire, and a compromise to those seeking meaning of the wording of the NEC and comfortable jumping on the "One End Bandwagon". But as mentioned one could use a bonding locknut... (Or in some areas - PVC) - then both the "Both End'ers" that may be involved and the POCO are happy...

But it still does not say one end only....Does it?/QUOTE]

No, it doesn't, because the NEC is not a training manual. They assume users of that book have had a formal education on electrical theory. There are training manuals available like Soares book on Grounding and Bonding that will provide education in that area. If you have access to the 10th edition of that book look on pages 86, 87 and 91, it shows the proper way to bond metal conduit at the service.

 

[iwire]

But is still does not say one end only.... Does it?

No, because it does not have to. It only requires us to bond the raceway it does not say I have to do it more than once.

 

[don_resqcapt19]

What is the UL listing category for an offset nipple? If the category is not one of the wiring methods listed in 230.43, then you can't even use the offset nipple on the line side of the service disconnect.

 

[George Stolz]

Thats why I added it to this discussion in the first place - it is a great example...

It amazes me that you point to an example that clearly demonstrates that when the Code wants a raceway bonded on both ends that it's clear in doing so as a proof to your belief - it actually proves the exact opposite point.

Lightning is a high frequency event and the GEC is designed to give that event a path to earth. Therefore, ferrous raceways containing a GEC have to be designed to limit that choking effect.

It is entirely different than a ground fault, making those provisions at an offset nipple in a service is not necessary, so it's not required.

250.64(E) is explicit that you must bond both sides of that raceway. If they wanted us to bond both sides of a raceway or nipple in service equipment, they would not make any bones about demanding it.

What ignorant power company employees, Frank the Barber down the street, or untrained/belligerent inspectors want to see is not relevant in the discussion. What is relevant is what the NEC requires, and does not require.

It clearly does not require bonding on both sides of this nipple between the meter socket and service disconnect, because if it were required, it would be specifically required the way that 250.64(E) does.

 

[iwire]

It amazes me that you point to an example that clearly demonstrates that when the Code wants a raceway bonded on both ends that it's clear in doing so as a proof to your belief - it actually proves the exact opposite point.

Glad I was not the only one that thought that.