GEC and SSBJ on transformer with multiple secondary

[Ccarlson32]

Looking for some clarification.
I have a 150kva transformer fed from 250cu on 225amp breaker for primary.
Secondary I have two sets of wires(not paralleled.)
One set is 250cu feeding 250amp main breaker panel
Second is 250cu feeding 200 amp main breaker panel.

Question 1) for sizing GEC I just take the largest ungrounded derived conductor (250cu)
Or do I treat them as a parallel and follow that for GEC sizing?

If I just follow largest ungrounded conductor for sizing, we get #2 cu for GEC.

The question arises if both secondaries fault, there is a potential for the ampacity of both 250cu and 250alum feeds.
Do I not think about the total ampacity of the transformer secondaries when sizing this and just look at the largest un grounded conductor?
I attached a picture to better describe my situation.
Thanks in advance for your input!

https://imgur.com/a/2KFcQho

 

[augie47]

Is your SBJ going to be at the transformer ??
If so your GEC should connect there and be sized based on the sum of the kcmils of the two secondaries.
The SSBJ to each panel can be sized per the individual feeders per 250.102

 

[Ccarlson32]

Is your SBJ going to be at the transformer ??
If so your GEC should connect there and be sized based on the sum of the kcmils of the two secondaries.
The SSBJ to each panel can be sized per the individual feeders per 250.102

Yes at the transformer.
And where do I find the code to size the GEC based on the sum of the kcmils of the two secondaries? (I thought that only applies to parallel installations?) 250.66- size of largest ungrounded service entrance conductor(secondary derived conductor) or equivalent area for paralell conductors.

Thanks, a very odd scenario that I have not ran into before and just want to ensure it I am doing things how they should! Common sense tells me to size based on the sum of the kcmils... If there is a secondary fault, the GEC could see the protentional from both of the secondary loads before primary protection clears.

 

[augie47]

Best I can come up with is the fact that 250.30 uses the word "conductors when referencing sizing and 250.26(D) does require the sum of conductors when sizing the SBJ,

 

[CoolWill]

I'd probably just size it as if there was a conductor carrying the full load of 150 kVA.

 

[Elect117]

I found 250.28(D)(3) for the system bonding jumper.

If you have the handbook or NFPA link, in the enhanced content, the last example is for more than one enclosure and the SBJ is at the transformer. It says to take the equivalent circular mils for each set of secondary. So in your case, 250kcmil+250kcmil and use table 250.102(C)(1) for the appropriate size, so (if all copper) 1/0 AWG.

 

[Ccarlson32]

I have found that sizing the SBJ specifically address this problem of taking the equivalent circular mils for each set of secondary conductors in sizing the SBJ. I cannot find this when sizing the GEC though? If they intended you to size it this way for your GEC wouldn't the wording be the same under there?
Regardless I am going to size the GEC 1/0 AWG the same size as the SBJ, but just want to find a hard answer for future purposes.
Right now, from what I find Sizing GEC tells me to use largest ungrounded conductor -- equivalent circular mils for PARALELLED installations. (This is technically not parallelled as they do not terminate in the same place at each end.)

From what I found it should be sized as follows:
SBJ --- 1/0
GEC--- (largest 250 cu = #2 cu GEC) -- #2? (or 1/0 awg)?
SSBJ---(#2 for 250cu, #4 for 250 AL)

Just going to have a 1/0 AWG SBJ and GEC, with #2 SSBJs for peace of mind.

 

[augie47]

I'm still of the opinion I noted in Post #4 that the wording in 250.30(A)(5) supports sizing it by the size of the conductors (plural)

Grounding Electrode Conductor, Single Separately Derived System. A grounding electrode conductor for a single separately derived system shall be sized in accordance with 250.66 for the derived ungrounded conductors.

 

[wwhitney]

I'm still of the opinion I noted in Post #4 that the wording in 250.30(A)(5) supports sizing it by the size of the conductors (plural)

I don't think I buy that, but Note 1 for Table 250.66 would by analogy mean the same thing. Seems like the language in Note 1 should be generalized not to be service specific.

Cheers, Wayne

 

[Elect117]

I agree with both of you.

If you size a common GEC for a service with multiple disconnects, it is based on the largest phase conductor of the sets.

If you size a common GEC for multiple separately derived systems to tap off of it, that common GEC can not be smaller than 3/0 (largest size).

The taps for individual GECs for the multiple separately derived systems are based on the largest phase conductor.

In this scenario, where it is a single separately derived system, with more than one panel being fed, I am inclined to say it is more similar to a service with multiple disconnects. I just can not find anywhere in the NEC that has language to allow that the GEC for a single separately derived system be sized on the largest of the set.

If I try to think about it from an engineering perspective, the goal is to provide a low resistance to earth as a reference point but also a path. If the overcurrent protection protects the wires, and the wire size can handle the fault for the overcurrent protection to trip, then I don't know why the GEC would need to be so large as to handle the impedance for each individual panel. To me, it is each individual system's low impedance to earth. Both systems would need to fault to the earth at the same time in order for the separate panel wires to be considered parallel. So maybe that is the point? Maybe that is justification to size the GEC based on their parallel mils?

 

[wwhitney]

If you size a common GEC for a service with multiple disconnects, it is based on the largest phase conductor of the sets.

If you mean largest looking at the sets one at a time, then no. Note 1 to Table 250.66 says that you add up the corresponding conductors from all the sets and use that equivalent size.

Cheers, Wayne

 

[david]

Table 250.102 (C) (1)
ex #3
The equivalent size of the largest of the ungrounded supply conductor

determined by the largest sum of the of the areas of the corresponding conductors of each set.

it does not say,
determined by the sum of the of the areas of the corresponding conductors of each set.

you could have( parallel 1/0) as one set and just 1/0 in another set.

the largest conductor of the two sets would be the equivalent size of the one set with the parallel 1/0

if in your equation you are adding the area of 3 1/0's your over sizzing your bonding jumper and that's ok because the code min is still being met

 

[Elect117]

If you mean largest looking at the sets one at a time, then no. Note 1 to Table 250.66 says that you add up the corresponding conductors from all the sets and use that equivalent size.

Cheers, Wayne

My apologies, I misread a reference book I use. You would take the sum of the mils for each set of conductor for the common GEC.

I am still on the fence as to whether the GEC would need to be #2 or 1/0.

I guess if you bonded the neutral to the electrode in the panels, the individual GECs would be #2 and the common GEC would be 1/0? or would it be multiple separately derived systems and be required to be 3/0?

 

[wwhitney]

Table 250.102 (C) (1)
ex #3
The equivalent size of the largest of the ungrounded supply conductor

determined by the largest sum of the of the areas of the corresponding conductors of each set.

it does not say,
determined by the sum of the of the areas of the corresponding conductors of each set.

The word "largest" there refers to doing the sum for each of the ungrounded legs, and then taking the largest of those sums. Usually each of the sums would be same, but there's no reason that you couldn't have different size conductors on different ungrounded legs.

you could have( parallel 1/0) as one set and just 1/0 in another set.

the largest conductor of the two sets would be the equivalent size of the one set with the parallel 1/0

True, but per Note 3, you would need to use 2 * 1/0 + 1/0 = 3 * 1/0.

Cheers, Wayne

 

[david]

Table 250.102 (C) (1)
ex #3
The equivalent size of the largest of the ungrounded supply conductor

I'm going to base t as stated, on the largest of ungrounded supply conductor(s). if you have more than one set your going to have a set with the largest and you will also have a set with the smallest or all sets will have the same size. if there are parallel conductors in one or more sets all base the largest set on the equivalent size as if it was a single conductor by the sum of areas of the conductors in parallel.

i understand your reading of this section i disagree with it, the same as you disagree with my reading of this section

 

[wwhitney]

I'm going to base t as stated, on the largest of ungrounded supply conductor(s).

No, your reading does not match the text. Your reading matches: "the sum of the areas of the conductors of the largest set." That is obviously different from "the largest sum of the of the areas of the corresponding conductors of each set."

The phrase "corresponding conductors of each set" means that you have to look at all the sets at once, and correspond the conductors (A leg to A leg, etc), and that phrase is acted on by "sum", so you have to add those conductor sizes across all sets.

Cheers, Wayne

 

[david]

. Usually each of the sums would be same, but there's no reason that you couldn't have different size conductors on different ungrounded legs.


Cheers, Wayne

just curious do you have an example where one ungrounded conductor in a supply set would be a different size than another ungrounded conductor in the same set of secondary supply conductors

 

[wwhitney]

just curious do you have an example where one ungrounded conductor in a supply set would be a different size than another ungrounded conductor in the same set of secondary supply conductors

You could make up such an example, but a simpler case for what is intended by the wording would be a 3P4W 240V High Leg Delta where you have two sets of SECs or SDS secondary conductors, one set being 3P3W 240V Delta for 3-phase loads, the other set being 1P3W 120/240V for single phase loads. Then ignoring parallel sets, there's only one high leg conductor, while the other two ungrounded legs have two conductors each, one for each set. So obviously the sums will be different for the high leg vs the other two legs.

Cheers, Wayne

 

[david]

No, your reading does not match the text. Your reading matches: "the sum of the areas of the conductors of the largest set." That is obviously different from "the largest sum of the of the areas of the corresponding conductors of each set."

The phrase "corresponding conductors of each set" means that you have to look at all the sets at once, and correspond the conductors (A leg to A leg, etc), and that phrase is acted on by "sum", so you have to add those conductor sizes across all sets.

Cheers, Wayne

you would not have the largest sum; you would only have one sum. if you had the largest sum than you would also have the possibility of smaller sums

we are not going to agree on this one

 

[wwhitney]

you would not have the largest sum; you would only have one sum.

You have one sum for each ungrounded leg of the supply. See the preceding example.

Cheers, Wayne