Parallel EGC

[david luchini]

The EGCs in the #12 and #14 home runs are in parallel, and, by 310.10(H)(5)and Table 250.122, the EGC for the #14 homerun has to be #12 . . .

The 15A and 20A circuits are not in parallel...their egc's are not parallel egc's. The rules in 310.10(H) provide the permission for paralleling circuit conductors, and establish the conditions for their installation.

 

[david]

It's been awhile since I've looked it up. I believe 250.148 is the appropriate citation.

that section requires a metallic box to be bonded to the circuit (s) equipment ground but does not require the # 14 and # 12 to be spliced together

I do not think it is the intent that every equipment ground be considered paralleled that are bonding a metallic piece of equipment

 

[al hildenbrand]

that section requires a metallic box to be bonded to the circuit (s) equipment ground but does not require the # 14 and # 12 to be spliced together

It says a LOT more than that. 250.148(C) is only a part of it.

 

[al hildenbrand]

The 15A and 20A circuits are not in parallel...their egc's are not parallel egc's. The rules in 310.10(H) provide the permission for paralleling circuit conductors, and establish the conditions for their installation.

:? If two conductors are electrically connected to each other at each end. . . that's parallel, no? Were they are used in parallel, they then have to meet the conditions of 310.10(H).

 

[david luchini]

:? If two conductors are electrically connected to each other at each end. . . that's parallel, no? Were they are used in parallel, they then have to meet the conditions of 310.10(H).

Where the circuit conductors are electrically connected at both ends, that's parallel. Where circuit conductors are paralleled in separate raceways, egc's are paralleled egc's.

The conditions in 310.10(H)(2) thru (6) apply when the General permission in 310.10(H)(1) is met...that is to say when there are parallel circuit conductors, their installation must comply with 310.10(H)(2) thru (6). Nothing in 310.10(H) applies to egc's from separate circuits.

(Though even if it did apply, the #14 and #12 egc would be fine.)

 

[LarryFine]

that section requires a metallic box to be bonded to the circuit (s) equipment ground but does not require the # 14 and # 12 to be spliced together

How could you accomplish the former without accomplishing the latter?

It's like bonding a gas supply without using it as a grounding electrode.

 

[david]

How could you accomplish the former without accomplishing the latter?

It's like bonding a gas supply without using it as a grounding electrode.

In a metallic gang -able box you would simply place a green ground screw in two different 10/32 tapped holes provided for bonding purposes.

I’m not saying it is wrong to pigtail your equipment grounds I’m saying it is not required to splice two separate branch circuit equipment grounds to accomplish the required bonding.

Consider metal bar joist in a building using metal J boxes if we are to consider every bonded equipment ground that has continuity established at the load end to be “paralleling” the circuit’s equipment grounds?

maybe i am not understanding your question

according to this thread you could never use two or three sets of "URD" (USE) as a feeder unless you special order the cable with a large equipment ground in each set.

i understand what Dennis is asking and i'm not sure yet if he is wrong , as long as the required size for the equipment grounds is established in each separate section of the circuit conductors being paralleled . The code states that a set of paralleled conductors are tied together at both ends to make a single conductor. I will follow this thread than make up my mind

 

[kwired]

Yes. And the first thing it says, tough not in so many words, is that if you pull an egc in conduit it has to be sized according to 250.122.

The second thing it says, not in so many words, is that if have some kind of fancy cable assembly, ie, mc or se that has some weird multi stranded egc inside the cable that's ok because it's part of a manufactured product. This is not unique, the code treats cable assemblies and conductors in raceways different in other sections.

Exactly what I tried to point out earlier. I think that is what is intended by the wording that is the focus of this thread.

Please consider: An occupancy is wired with a Nonmetallic Raceway Method. A 20 Amp OCPD #12 branch circuit with wire-type EGC is run in Nonmetallic Raceway to a bank of switches in a single multigang box. A 15 Amp OCPD #14 branch circuit with a wire-type EGC is run in a different Nonmetallic Raceway to the same bank of switches. The EGCs in the #12 and #14 home runs are in parallel, and, by 310.10(H)(5)and Table 250.122, the EGC for the #14 homerun has to be #12 . . .

Those EGC's are parallel to one another as are many others. Run a bunch of steel conduit attached to steel structure then pull green conductors through them and you have all sorts or parallel EGC paths whether you wanted to or not. But none of them are "in parallel" for the purpose of effectively creating a higher overall current carrying ability conductor.

In a metallic gang -able box you would simply place a green ground screw in two different 10/32 tapped holes provided for bonding purposes.

I’m not saying it is wrong to pigtail your equipment grounds I’m saying it is not required to splice two separate branch circuit equipment grounds to accomplish the required bonding.

Consider metal bar joist in a building using metal J boxes if we are to consider every bonded equipment ground that has continuity established at the load end to be “paralleling” the circuit’s equipment grounds?

maybe i am not understanding your question

according to this thread you could never use two or three sets of "URD" (USE) as a feeder unless you special order the cable with a large equipment ground in each set.

i understand what Dennis is asking and i'm not sure yet if he is wrong , as long as the required size for the equipment grounds is established in each separate section of the circuit conductors being paralleled . The code states that a set of paralleled conductors are tied together at both ends to make a single conductor. I will follow this thread than make up my mind

Using two green screws in a steel box is allowed, they are still considered bonded together when you do this just like the main bonding jumper to the service enclosure and a separately mounted EGC terminal bus are still considered bonded together.

Multiple URD cables for feeders - that often does create problems most common conductor combinations you will find in such a cable were primarily intended for use as service conductors. If you do have one with a green conductor in it - it was intended for feeder or branch circuit for a single run and not parallel runs. You very well can have same issue with SE or MC cables if you try to parallel them, NM not so much mostly because you won't find them in sizes large enough to be paralleled anyway.

 

[al hildenbrand]

In a metallic gang -able box you would simply place a green ground screw in two different 10/32 tapped holes provided for bonding purposes.

I’m not saying it is wrong to pigtail your equipment grounds I’m saying it is not required to splice two separate branch circuit equipment grounds to accomplish the required bonding.

David, look at the opening of 250.148. . . so far you are only thinking about 250.148(C).

Consider metal bar joist in a building using metal J boxes if we are to consider every bonded equipment ground that has continuity established at the load end to be “paralleling” the circuit’s equipment grounds?

You raise an interesting parallel conductive path scenario. My example above is trying NOT to look at everything at once, rather to consider a nonmetallic raceway method with wire-type EGCs. For your thinking, consider 250.148 as applied to smurf tube and smurf boxes where the box is nonmetallic.

according to this thread you could never use two or three sets of "URD" (USE) as a feeder unless you special order the cable with a large equipment ground in each set.

This, David, is starting to get to the point, . . . however, there is an exception, 310,10(H)(5) that allows CABLES to have smaller EGCs as long as the sum of the individual cable EGC circular mils equals or exceeds the EGC size given to us by Table 250.122 for the OCPD size.

David Luchini raised an excellent point about the reach of 310.10(H)(1), with respect to my #12 & #14 non metallic raceway hypothetical, that I am mulling over. He's basically saying, because of the Style Manual / Code Arrangement, by considering 310.10(H)(5) first I have my cart in front of the horse.

 

[david]

David, look at the opening of 250.148. . . so far you are only thinking about 250.148(C).

….Any equipment grounding conductors associated with those circuit conductors…..

To me in a non-metallic raceway system including non- metallic device boxes only the equipment grounds associated with their respective circuit conductors

I do not take that to include different branch circuits sharing a common device box requiring all equipment grounds to be tied together

Rather the requirement would be to tie together the equipment grounds respectively associated with their branch circuit

 

[kwired]

This, David, is starting to get to the point, . . . however, there is an exception, 310,10(H)(5) that allows CABLES to have smaller EGCs as long as the sum of the individual cable EGC circular mils equals or exceeds the EGC size given to us by Table 250.122 for the OCPD size.

.

That is talking about parallel sections of EGC within the same cable assembly, not multiple EGC's in multiple cables that are used in parallel to one another. Typical URD won't have a multi sectioned EGC. I mentioned earlier some VFD cables might be an example of what this may apply to, say you had individual conductors in the cable individually shielded within the cable and those shield conductors also compose the EGC of the entire cable assembly. I'm not all that familiar with VFD cables, but seems like that may be a design option for them.

If you had 4/0 aluminum URD containing a #4 EGC, do you think it is acceptable to parallel two of them and use for a 400 amp feeder with the two #4's being the EGC or do you think a #1AL EGC (per 250.122) is required in each cable?

 

[al hildenbrand]

That is talking about parallel sections of EGC within the same cable assembly, not multiple EGC's in multiple cables that are used in parallel to one another. Typical URD won't have a multi sectioned EGC.

Ah. Thank you for that. My lack of experience with this material is showing. I see how I am misreading 310.10(H)(5). I haven't even seen anything in print showing a "multisectioned EGC" as part of a single multiconductor cable assembly.

 

[ActionDave]

This, David, is starting to get to the point, . . . however, there is an exception, 310,10(H)(5) that allows CABLES to have smaller EGCs as long as the sum of the individual cable EGC circular mils equals or exceeds the EGC size given to us by Table 250.122 for the OCPD size.

David Luchini raised an excellent point about the reach of 310.10(H)(1), with respect to my #12 & #14 non metallic raceway hypothetical, that I am mulling over. He's basically saying, because of the Style Manual / Code Arrangement, by considering 310.10(H)(5) first I have my cart in front of the horse.

You can't pull a bunch of little EGC's in a pipe and make them one bigger EGC. That's it. There is nothing left to mull over.

 

[Gary11734]

You can't pull a bunch of little EGC's in a pipe and make them one bigger EGC. That's it. There is nothing left to mull over.

You can at 1/0 and over, maybe!

 

[kwired]

You can at 1/0 and over, maybe!

To make a larger equivalent conductor - no. Only thing the 310.10(H)xx allowance is about is a multiple sections of EGC within an assembly being allowed to be considered one EGC for the entire assembly. Even bare concentric conductor of a type SE cable fits in here, all those strands can comprise one EGC for the entire cable assembly.

 

[al hildenbrand]

According to this thread you could never use two or three sets of "URD" (USE) as a feeder unless you special order the cable with a large equipment ground in each set.

David, based on the new language in the 2017 NEC at 250.122(F)(2), it now reads, to me, that your statement is correct, i.e., the direct burial URD (USE) EGC has to be sized for the OCPD rating of the paralleled set of cables.

2017 NEC
250.122 Size of Equipment Grounding Conductors.
250.122(F) Conductors in Parallel.
250.122(F)(2) Multiconductor Cables.
250.122(F)(2)(d) Except as provided in 250.122(F)(2)(b) for raceway or cable tray installations, the equipment grounding conductor in each multiconductor cable shall be sized in accordance with 250.122 based on the overcurrent protective device for the feeder or branch circuit.

Following on Augie47's post #43:

According to the Soares book on grounding, "one reason for a full size conductor in each raceway is that if a line to ground fault occurs in the raceway, current will be fed to the fault from both directions. The equipment grounding conductor will thus be called upon to carry the entire amount of the fault current".

It appears to me that the CMPs are going after Cables in the same way they have Raceways, with a concern for the fault that is midway along a multiconductor cable, a fault that is between the phase conductor(s) and the EGC where there are no other effective fault current conductive paths.

 

[kwired]

David, based on the new language in the 2017 NEC at 250.122(F)(2), it now reads, to me, that your statement is correct, i.e., the direct burial URD (USE) EGC has to be sized for the OCPD rating of the paralleled set of cables.



Following on Augie47's post #43:

It appears to me that the CMPs are going after Cables in the same way they have Raceways, with a concern for the fault that is midway along a multiconductor cable, a fault that is between the phase conductor(s) and the EGC where there are no other effective fault current conductive paths.

IMO this has always been the rule, someone must have thought it needed some additional clarification why we have a change in wording.

 

[al hildenbrand]

You can't pull a bunch of little EGC's in a pipe and make them one bigger EGC. That's it. There is nothing left to mull over.

I understand a little better your Zen koan of a statement, after adding David Luchini's insight, . . . :

310.10(H) does not say anything about equipment grounding conductors, only phase and grounded/neutral conductors.

 

[LarryFine]

I do not take that to include different branch circuits sharing a common device box requiring all equipment grounds to be tied together

Let's explore just that: a mix of 12's and 14's in a multi-gang plastic switch box. Join? Separate? Optional?

We already join all of one size cable in a box, even if they're separate circuits, without hesitation, don't we?

 

[al hildenbrand]

I do not take that to include different branch circuits sharing a common device box requiring all equipment grounds to be tied together

Let's explore just that: a mix of 12's and 14's in a multi-gang plastic switch box. Join? Separate? Optional?

We already join all of one size cable in a box, even if they're separate circuits, without hesitation, don't we?

Yes, let's explore that.

IMO, Join.