3000 A conductor sizing for service entrance

Status
Not open for further replies.
Surprising to hear that some utilities still direct bury service conductors. Makes it extremely time consuming and costly to repair/replace if it fails. We as a utility (both CA and Alaska) did away with any new direct bury and also in CA, replaced (at our cost) older existing direct burial with conduit. Agree on EGC. Utility will never (AFIK) install or require any grounds in a service drop.

makes me wonder if it is really a service if an EGC is required. If it is required than it has to be full sized.
 
Update

Update

Things have changed a lot on this project and another has already started. They have minor differences. Below is a summary of what was done and what will be done at the next project. Sorry if my original post was all over the place.

We have settled on 8 runs of conduit with 500MCM for phase wire. However, I disagree with some of the comments in here. Let me add some clarification and say I have 3000A service, 100% rated, and am using THHW in PVC conduit underground. Ambient of earth is 20 deg C. The parts not in earth are less than 10' or 10% of the run so I can use 20 ambient and have an adjustment factor of 1.11 and since we service L-N loads (and harmonics for that matter), the neutral is CCC. So I have to derate 80% for 4 CCC's in one raceway.
From what I understand, you can't use the 90 deg column unless the terminals are rated at 90 and they are not, they are rated for 75 deg C. 500MCM can carry 380A (75 deg C terminals, our wire is 90C). So I adjust this by 1.11 then multiply times 80%: 380*1.11*80%=337A. This times 8 (8 runs of A,B,C) is only 2700A. We do have 277VAC loads tied to this service. Do you guys agree with this calculation?

So 8 runs of 500MCM is not sufficient. If I could eliminate my neutral load, then I can eliminate the 80% derating, which would allow me to use 500MCM with 8 runs in 4". We do have other L-N loads but they are serviced through a 'separately derived system' (transformer) so I am not counting these as L-N loads for the sake of CCC's derating. Assuming loads on separately derived systems don't qualify as L-N loads for the number of CCC derating, I may still be able to eliminate my 277V loads (LED's and Heat Trace) in order to get away with 8 runs of 500MCM (no CCC derating).
Thanks!
 
No second guessing the details of your installation and acknowledge you have some L-N load but, as already been mentioned, having to count the neutral as a CCC on a 3000 service is extremely rare due the Art 310 wording:

(c) On a 4-wire, 3-phase wye circuit where the major portion of the load consists of nonlinear loads, harmonic currents are present in the neutral conductor; the neutral conductor shall therefore be considered a current-carrying conductor.
 
Things have changed a lot on this project and another has already started. They have minor differences. Below is a summary of what was done and what will be done at the next project. Sorry if my original post was all over the place.

We have settled on 8 runs of conduit with 500MCM for phase wire. However, I disagree with some of the comments in here. Let me add some clarification and say I have 3000A service, 100% rated, and am using THHW in PVC conduit underground. Ambient of earth is 20 deg C. The parts not in earth are less than 10' or 10% of the run so I can use 20 ambient and have an adjustment factor of 1.11 and since we service L-N loads (and harmonics for that matter), the neutral is CCC. So I have to derate 80% for 4 CCC's in one raceway.
From what I understand, you can't use the 90 deg column unless the terminals are rated at 90 and they are not, they are rated for 75 deg C. 500MCM can carry 380A (75 deg C terminals, our wire is 90C). So I adjust this by 1.11 then multiply times 80%: 380*1.11*80%=337A. This times 8 (8 runs of A,B,C) is only 2700A. We do have 277VAC loads tied to this service. Do you guys agree with this calculation?

So 8 runs of 500MCM is not sufficient. If I could eliminate my neutral load, then I can eliminate the 80% derating, which would allow me to use 500MCM with 8 runs in 4". We do have other L-N loads but they are serviced through a 'separately derived system' (transformer) so I am not counting these as L-N loads for the sake of CCC's derating. Assuming loads on separately derived systems don't qualify as L-N loads for the number of CCC derating, I may still be able to eliminate my 277V loads (LED's and Heat Trace) in order to get away with 8 runs of 500MCM (no CCC derating).
Thanks!

I think eight runs or 500 THWN is good for 3055 amps.
You need to use the 90° ampacity in your calculation, then size the wire to the 75° column for the termination.

430*1.11*80%=381.84*8=3054.72
500 THWN @75°c is good for 380
380*8=3040
Eights set of 500 is good.
 
I think eight runs or 500 THWN is good for 3055 amps.
You need to use the 90° ampacity in your calculation, then size the wire to the 75° column for the termination.

430*1.11*80%=381.84*8=3054.72
500 THWN @75°c is good for 380
380*8=3040
Eights set of 500 is good.

I agree he is good, although personally I would dump the 4CCC derate and dump the ambient temp adjustment.
 
AS Augie47 pointed out, unless the majority (50%+) of your loads are non-linear, you do not count the neutral as a CCC. If the neutral isnt counted as a CCC, then you have 3 CCC per conduit and 8 sets of 500MCM copper will suffice. What type of building/facility are you wiring? I could see a grow operation or large stadium using tons of high power HPS/LED lights as majority non-linear, in which case you'd need more conduits or larger wire. If it's an option, Al will be much less costly than Cu.

eta: it doesnt matter if 1 or 100% of your loads are L-N so long as the majority of the total load is non-linear. If you have 3000A L1-N, 3000A L2-N, and 3000A L3-N, say all 277V resistive loads, the load on the neutral is 0 amps.
 
Last edited:
AS Augie47 pointed out, unless the majority (50%+) of your loads are non-linear, you do not count the neutral as a CCC. If the neutral isnt counted as a CCC, then you have 3 CCC per conduit and 8 sets of 500MCM copper will suffice. What type of building/facility are you wiring? I could see a grow operation or large stadium using tons of high power HPS/LED lights as majority non-linear, in which case you'd need more conduits or larger wire. If it's an option, Al will be much less costly than Cu.

eta: it doesnt matter if 1 or 100% of your loads are L-N so long as the majority of the total load is non-linear. If you have 3000A L1-N, 3000A L2-N, and 3000A L3-N, say all 277V resistive loads, the load on the neutral is 0 amps.

There are three provisions listed under 310.15(B)(5) when determining if the neutral is CCC. (b) says 4 wire 3 phase has common neutral so any load is considered CCC. If you are using 277VAC, you are using one phase and the neutral to supply power to the device. You will have neutral current regardless. This current will return to the source via the neutral (electricians taught current flow in the opposite direction as engineers, usually). So there is neutral current. It is not just a matter of non-linear loads or unbalanced loads to determine if neutral is CCC - at least in my understanding.
This is feed mill and we are using all LED lighting via a lighting transformer (K factor) so the harmonics won't be an issue getting back to the service (at least from the LED's). My biggest concern is the VFD's. We have many of them and the largest is 500HP. I have been told this 500HP has no line filtering/reactor.
 
I think eight runs or 500 THWN is good for 3055 amps.
You need to use the 90° ampacity in your calculation, then size the wire to the 75° column for the termination.

430*1.11*80%=381.84*8=3054.72
500 THWN @75°c is good for 380
380*8=3040
Eights set of 500 is good.

Thanks for this. I get it now. We don't need to derate anything for terminal temperature limitations since there is no heat generated from other conductors, as with wire. I have 19 years of experience in product design at the board level, power generating facility O&M, etc. but don't have the much experience with NEC, as you may be able to tell.
 
There are three provisions listed under 310.15(B)(5) when determining if the neutral is CCC. (b) says 4 wire 3 phase has common neutral so any load is considered CCC.

That is not what 310.15(B)(5)(b) says. 310.15(B)(5)(b) applies to a 3-wire circuit (two phase conductors and the neutral) of a 4 wire, 3 phase, wye connected system.
 
There are three provisions listed under 310.15(B)(5) when determining if the neutral is CCC. (b) says 4 wire 3 phase has common neutral so any load is considered CCC. If you are using 277VAC, you are using one phase and the neutral to supply power to the device. You will have neutral current regardless. This current will return to the source via the neutral (electricians taught current flow in the opposite direction as engineers, usually). So there is neutral current. It is not just a matter of non-linear loads or unbalanced loads to determine if neutral is CCC - at least in my understanding.
This is feed mill and we are using all LED lighting via a lighting transformer (K factor) so the harmonics won't be an issue getting back to the service (at least from the LED's). My biggest concern is the VFD's. We have many of them and the largest is 500HP. I have been told this 500HP has no line filtering/reactor.

As David mentioned, (b) doesn't apply. You should be looking at (a) which says you don't have to count the neutral for a 3 phase, 4 wire system unless the major portion has harmonics.

Imagine a perfectly balanced 3 phase, 4 wire system. All the current flows on the 3 phase conductors, and none flows on the neutral.

Now imagine you somehow unbalance the load without going over 3000 amps total. If the current on the neutral goes up to 600 amps for example, the current on the other conductors has to go down by a total of 600 amps. So the neutral only carries the current unbalance, and it can't create any more heat than the full load balanced load.

Regarding the VFD, that is pretty big, but still only in the 600 amp range for a 3000 amp service. And if the VFD is that big, it might us a 12 or 18 pulse input circuit which helps reduce the harmonics.

NEC service calculations seem to be pretty conservative. So if someone did the load calc. right, the odds of the service ever being loaded to 3000 amps is pretty slim. That's why a lot of them are just 8 sets of 500 KCM.
 
As David mentioned, (b) doesn't apply. You should be looking at (a) which says you don't have to count the neutral for a 3 phase, 4 wire system unless the major portion has harmonics.

Imagine a perfectly balanced 3 phase, 4 wire system. All the current flows on the 3 phase conductors, and none flows on the neutral.

Now imagine you somehow unbalance the load without going over 3000 amps total. If the current on the neutral goes up to 600 amps for example, the current on the other conductors has to go down by a total of 600 amps. So the neutral only carries the current unbalance, and it can't create any more heat than the full load balanced load.

Regarding the VFD, that is pretty big, but still only in the 600 amp range for a 3000 amp service. And if the VFD is that big, it might us a 12 or 18 pulse input circuit which helps reduce the harmonics.

NEC service calculations seem to be pretty conservative. So if someone did the load calc. right, the odds of the service ever being loaded to 3000 amps is pretty slim. That's why a lot of them are just 8 sets of 500 KCM.

I don't know if I agree with the statement that just because the neutral current is 600A doesn't necessarily mean the current in the other two phases increases my 600A. Ill have to do some thinking on that one but my initial thought is no that each phase may increase 200A not 600A. However, I think only the phase with the load to neutral would increase in current not all of them. Have to draw that out. After rereading (a) it is clear they don't care about unbalanced loads since the neutral is sized for this already. I just don't get why they don't count it as current carrying. I believe the 500HP drive is a PWM type and not PAM but I will find out at a site visit next week. I agree that the service will never have to provide 3000A.
 
Status
Not open for further replies.
Top