Feeder Taps & EGC Sizing

[Elect117]

Lets say I wanted to tap a 1600A bus with a 500A fused disconnect within 25'.

If I wanted to use 4 runs of 1/0 (3 CCCs in each). Would the EGC need to be 4/0 per conduit or 1/0 per conduit?

Basically, does the rule where the EGC does not need to be larger than the feeder tap conductor reduce the EGC size when taking into account parallel runs or should I consider the feeder tap as the circular mils of 4x1/0 and that brings me back to 4/0 being the smallest size EGC?


What if I chose to have all of the CCCs in one conduit, like 4 x 2/0 (derated at 70% is still okay at 546A). Would my EGC be a single 2/0 or 4/0?

 

[infinity]

Your tap conductors are equal to (4) #1/0's.

 

[Elect117]

If I am understanding you correctly., an EGC sized 4/0,

 

[infinity]

Lets say I wanted to tap a 1600A bus with a 500A fused disconnect within 25'.

T250.122 EGC size #4/0 in each parallel raceway.

If I am understanding you correctly., an EGC sized 4/0,

Yes because although you have parallel #1/0 tap conductors there are 4 sets so your #4/0 EGC would not be larger than the ungrounded conductors consisting of 4 sets of #1/0.

 

[don_resqcapt19]

CMP 5 is previously on record as saying that the ungrounded conductor size for parallel sets is the sum of the sets in parallel, and that results in an ungrounded conductor size of 422,400 circular mils. 4/0 is 211,600 circular mils. Under the current code, a 4/0 is required in each raceway, even though the ungrounded conductors in the raceways are 1/0.
If this is a real installation, I strongly suggest you talk to the local inspection authority before designing the system.

The First draft report shows this being fixed in the 2026 code with new language in 250.122(F)(b)(2).

The equipment grounding conductor in each raceway shall not be required to be larger than the largest ungrounded conductor in the raceway.

 

[wwhitney]

So prior to the 2026 First Draft, we have two statements in 250.122 that appear to be in contradiction. Using the 2020 NEC for reference, the second sentence of 250.122(A) says "The equipment grounding conductor shall not be required to be larger than the circuit conductors supplying the equipment." If we are to interpret that for the case of parallel circuit conductors as referring to all the paralleled conductors in aggregate, the same should be apply to the equipment grounding conductor.

Yet 250.122(F)(1)(b) on conductors in parallel installed in multiple raceways says in the second sentence "The equipment grounding conductor installed in each raceway shall be sized in accordance with 250.122 based on the rating of the overcurrent protective device for the feeder or branch circuit." As the OCPD rating will be related to the circuit conductor size in aggregate across all the raceways, that could lead to the EGCs in aggregate exceeding the circuit conductor size in aggregate.

However, in resolving this conflict, it seems to me that 250.122(A) takes precedence. It has no reference to 250.122(F)(1)(b), while the second sentence of 250.122(F)(1)(b) has the language "in accordance with 250.122," which includes the second sentence of 250.122(A). So the requirement of the second sentence of 250.122(F)(1)(b) is subject to the second sentence of 250.122(A).

The First draft report shows this being fixed in the 2026 code with new language in 250.122(F)(b)(2). "The equipment grounding conductor in each raceway shall not be required to be larger than the largest ungrounded conductor in the raceway."

Seems to me this wording is not quite right--the raceway might have parallel conductors within it, and so the wording should be clear that those parallel conductors with the raceway should be taken in aggregate. A somewhat contrived example: a 2 wire single phase feeder, with two parallel raceways, each containing for each leg 14 1/0 Cu ungrounded conductors of in parallel. With 90C insulation, each conductor has an ampacity of 170A; with 28 CCCs, the ampacity adjustment is 0.45, so each conduit provides an ampacity of 1071A. With an OCPD of 2000A, 250.122 specifies a minimum EGC size of 250 kcmil. But if we interpret the first draft text to mean each raceway only needs a 1/0 Cu EGC, then 2 x 1/0 is only 211 kcmil.

Cheers, Wayne

 

[don_resqcapt19]

...


Seems to me this wording is not quite right--the raceway might have parallel conductors within it, and so the wording should be clear that those parallel conductors with the raceway should be taken in aggregate. A somewhat contrived example: a 2 wire single phase feeder, with two parallel raceways, each containing for each leg 14 1/0 Cu ungrounded conductors of in parallel. With 90C insulation, each conductor has an ampacity of 170A; with 28 CCCs, the ampacity adjustment is 0.45, so each conduit provides an ampacity of 1071A. With an OCPD of 2000A, 250.122 specifies a minimum EGC size of 250 kcmil. But if we interpret the first draft text to mean each raceway only needs a 1/0 Cu EGC, then 2 x 1/0 is only 211 kcmil.

Cheers, Wayne

Yes, that could be a problem. There are no Public Comments to address that issue, but there are 3 to delete the new language.