Increased EGC Sizing 250.122(B)

[Wyeman94]

Hey all, I'm using 2011 NEC and copper conductors.
In theory, let's say we have 208Y120V serving a 35-40A generic load at 300' away (subpanel).

Solution A
50/3 breaker. Initial feeder size would be 4#8, #10G, 3/4"C but due to voltage drop, we'd want to increase the size of wire to say 4#3, #4G, 1-1/4"C (#4G using 250.122(B) rule for circular mil upsizing from the initial).

Solution B
100/3 breaker, keeping the wiring at 4#3, #8G, 1-1/4"C (EGC would be smaller and cheaper than above).


Which solution is best, and am I thinking about this wrong?? The #4G doesn't really make sense considering solution B. I would appreciate any thoughts about this, thanks.

 

[iwire]

Solution B
100/3 breaker, keeping the wiring at 4#3, #8G, 1-1/4"C (EGC would be smaller and cheaper than above).

Unlikely the wire will cost so much less that it will pay for the added labor and material for overcurrent protection at the far end of the circuit for the 'generic load'.

In most cases you cannot supply equipment that needs 35-40 amps with 100 amps.

 

[JoeStillman]

In Solution B, you still need to up-size the equipment grounding conductor, even though #8 is "right" for a 100A breaker. By up-sizing the EGC you make sure there is a low enough resistance that you get sufficient current to trip the breaker during a ground fault. By using a 100A breaker, you compound the problem. You need a #4 there too.

 

[iwire]

In Solution B, you still need to up-size the equipment grounding conductor, even though #8 is "right" for a 100A breaker.

I do not understand what you mean? :?

 

[lielec11]

I do not understand what you mean? :?

I believe what he means is that although based on Table 250.122 a #8 EGC is correct for a 100A overcurrent device, you still need to upsize the EGC based on the current carrying conductors.


My two cents.... although I cannot seem to find it in the code at the moment, I believe there is a section that disallows protection of a 34-40A load by a 100A overcurrent device.

 

[iwire]

I believe what he means is that although based on Table 250.122 a #8 EGC is correct for a 100A overcurrent device, you still need to upsize the EGC based on the current carrying conductors.

Its a 100 amp circuit with 3 AWG 'hots' there is no NEC requirement to up size the EGC from 8 AWG.

I believe there is a section that disallows protection of a 34-40A load by a 100A overcurrent device.

For sure there are, you can find some of them in 422 for appliances.

 

[charlie b]

In Solution B, you still need to up-size the equipment grounding conductor, even though #8 is "right" for a 100A breaker.

I think not (this is one of those rare occasions in which I agree with Bob :lol: ).

By using a 100A breaker, you compound the problem.

Not necessarily. What it takes to trip a 100 amp breaker during a fault is not much different than what it takes to trip a 50 amp breaker under the same fault. In their instantaneous trip regions, the two breakers will function about the same.

You need a #4 there too.

I presume you meant for the EGC. But I disagree. You have not upsized the ungrounded conductors from the minimum size that has sufficient ampacity to handle the selected breaker size of 100 amps. Therefore, you don't have to upsize the EGC.

 

[charlie b]

In most cases you cannot supply equipment that needs 35-40 amps with 100 amps.

My two cents.... although I cannot seem to find it in the code at the moment, I believe there is a section that disallows protection of a 34-40A load by a 100A overcurrent device.

I think that you would both be right, if we were talking about running from a 100 amp breaker directly to a hard wired outlet for an appliance or other load that is expected to draw only 40 amps. But the OP mentioned a sub panel. So giving the benefit of the doubt here, I infer that the intended installation for solution B is a 100 amp feeder breaker, with appropriate sized wires, serving a sub panel that is rated for 100 amps, and for which the calculated load is only 40 amps. In that case, I would say that the proposed solution B is acceptable as described.

 

[iwire]

this is one of those rare occasions in which I agree with Bob :lol:

You can't always be wrong.

(Joking. I have much respect for Mr Beck and his knowledge)

 

[iwire]

But the OP mentioned a sub panel.

Where? :huh:

I mentioned overcurrent protection at the far end.


(Edit, I see it now in the preamble, not the solutions)

 

[Wyeman94]

Wow thanks for the replies!
I meant that the demand load is within a 100A MLO (SquareD NQ) subpanel at that distance. In solution B, the panel would be simply be underloaded. 100A protects #3 which holds voltage at that distance under the 40A load. Unless a voltage drop calc is done considering the 100A subpanel be maxed out (in which case I'd want to see 4#1, #4G with #3 being the "initial size"). But I don't believe it's required for solution B since the load is set, although it would be good design practice because things change over time. In my case, almost no load would be added over time to this particular subpanel.

So my question boils down to:
Do you initially assign feeders to subpanels based on the demand load or simply the rating of the subpanel supporting a full capacity (where Table 250.122 is simply used as being "right" for EGC)? If the latter, and a voltage drop calc on the demand load could see nothing wrong with the Table derived EGC.

 

[lielec11]

So my question boils down to:
Do you initially assign feeders to subpanels based on the demand load or simply the rating of the subpanel supporting a full capacity (where Table 250.122 is simply used as being "right" for EGC)? If the latter, and a voltage drop calc on the demand load could see nothing wrong with the Table derived EGC.

Typically I design for the full rating as you never know what a client will do down the line in the future.

 

[charlie b]

Typically I design for the full rating as you never know what a client will do down the line in the future.

So do I. Otherwise there would be wasted capacity on the panel being served.

 

[JoeStillman]

I do not understand what you mean? :?

Increased in Size. Where ungrounded conductors are
increased in size, equipment grounding conductors, where
installed, shall be increased in size proportionately according
to the circular mil area of the ungrounded conductors.

Just because the breaker matches the wire size doesn't mean the wires aren't still oversized for voltage drop. Lets do the math; going from #8 (16,510 cm, table 8) to #3 (52,620 cm) is an increase of 52,620/16,510 or 319%. The EGC for a 50A CB is #10 (10,380 cm). 10,380 x 319% = 33,083 cm, somewhere between #6 and #4, so the correct EGC is #4.

Unless you can scrounge up some #5 AWG somewhere. :happyno:

 

[JoeStillman]

Not necessarily. What it takes to trip a 100 amp breaker during a fault is not much different than what it takes to trip a 50 amp breaker under the same fault. In their instantaneous trip regions, the two breakers will function about the same.
I presume you meant for the EGC. But I disagree. You have not upsized the ungrounded conductors from the minimum size that has sufficient ampacity to handle the selected breaker size of 100 amps. Therefore, you don't have to upsize the EGC.

I looked up the Instantaneous pickup for 50A and 100A Q0's. The 50A picks up about 350A and the 100A about 700A. With a 300' run of #8 and a 10,000A L-L source (3,333 L-G) you get around 280A of ground fault at the remote end - not enough to trip either breaker on instantaneous. Increase the wire size to #3 and your available GF current goes up to around 900, so you would get a trip, providing the source is really that stiff.

This makes me wonder what SKM uses for the ground-return impedance? There's no place to enter the EGC size.:?

 

[iwire]

Just because the breaker matches the wire size doesn't mean the wires aren't still oversized for voltage drop. Lets do the math; going from #8 (16,510 cm, table 8) to #3 (52,620 cm) is an increase of 52,620/16,510 or 319%. The EGC for a 50A CB is #10 (10,380 cm). 10,380 x 319% = 33,083 cm, somewhere between #6 and #4, so the correct EGC is #4.

Unless you can scrounge up some #5 AWG somewhere. :happyno:

As long as the wires match the breaker size. (3 AWG - 100 AMPs ) you can forget about 250.122(B). If he replaced the 100 amp breaker with a 50 than 250.122(b) would apply.

I know. .... It sounds nuts but that is the deal.

 

[charlie b]

Just because the breaker matches the wire size doesn't mean the wires aren't still oversized for voltage drop.

Oversizing for voltage drop is not the issue. Rather, that is not the starting point of the process. First you calculate the load. Then you pick a wire size that has sufficient ampacity for that load. Then you pick a breaker rating that will protect that wire. If after having completed this process you pick a larger wire size than is necessary to handle the load, then and only then you are into 250.122(B).

If you consider your load to be 40 amps (e.g., if you were to install a 40 amp fused disconnect directly upstream of the sub panel), and you pick a wire that has an ampacity of 100, then you must increase the EGC size. But if you consider the load to be the full 100 amp rating of the sub panel (as I believe should be done), then picking a wire that has an ampacity of 100 does not bring about the need to increase the EGC size.

 

[iwire]

But if you consider the load to be the full 100 amp rating of the sub panel (as I believe should be done), then picking a wire that has an ampacity of 100 does not bring about the need to increase the EGC size.

Day 2: Detente continues.


I was planing on asking Joe about panel feeder sizes. It seems his position would require an upsized EGC for every panel feeder that was not fully loaded.

 

[JoeStillman]

Day 2: Detente continues.


I was planing on asking Joe about panel feeder sizes. It seems his position would require an upsized EGC for every panel feeder that was not fully loaded.

Now wait a minute - don't put words in my mouth, even if they are logical. We're talking about what the code says, after all. If you push that #3 feeder out to 385', you don't have enough GF current to trip the breaker on instantaneous.