250.122 and Upsizing for Available Fault Current

[necnotevenclose]

I've been doing some reading on short circuit calculations and came accross a book that mentions upsizing the ground conductor (250.122) for the available fault current. I've ran lots of short circuit calcs and have yet to look at upsizing the grounding conductor, nor has the size of the GEC been brought up during plan review. I'm not saying this is right, but I have to wonder how many people are upsizing the GEC during design for the available fault current?

 

[Dennis Alwon]

I've been doing some reading on short circuit calculations and came accross a book that mentions upsizing the ground conductor (250.122) for the available fault current. I've ran lots of short circuit calcs and have yet to look at upsizing the grounding conductor, nor has the size of the GEC been brought up during plan review. I'm not saying this is right, but I have to wonder how many people are upsizing the GEC during design for the available fault current?

Not sure how many do but it is a requirement based on art. 250.122(B). I have done it.

 

[necnotevenclose]

Thanks Dennis. I have to wonder why this has not been brought up during plan review. Could it be that the plan reviewer is assuming that a steel raceway with a GEC sized for the OCPD would be able to support the AFC?

 

[kwired]

I've been doing some reading on short circuit calculations and came accross a book that mentions upsizing the ground conductor (250.122) for the available fault current. I've ran lots of short circuit calcs and have yet to look at upsizing the grounding conductor, nor has the size of the GEC been brought up during plan review. I'm not saying this is right, but I have to wonder how many people are upsizing the GEC during design for the available fault current?

You or the author of that book are looking at this the wrong way.

We do not upsize the EGC in these situations because of available fault current, we do so to decrease resistance of the EGC because of length of the conductor, which will ultimately increase the available fault current at the load end of the circuit. Higher current means faster response time from the overcurrent protection (usually) which is the main idea behind this requirement.

 

[Dennis Alwon]

What I don't get is if you have a metal conduit as the equipment grounding conductor then nothing needs to be done however if you choose to add an equipment grounding conductor then it must be sized accordingly in spite of the fact that the conduit is there. I do actually get it , I guess, but it would be a rare situation where the conduit would not be enough even with an undersized conduit. I guess it depends on where the fault occurred.

 

[kwired]

What I don't get is if you have a metal conduit as the equipment grounding conductor then nothing needs to be done however if you choose to add an equipment grounding conductor then it must be sized accordingly in spite of the fact that the conduit is there. I do actually get it , I guess, but it would be a rare situation where the conduit would not be enough even with an undersized conduit. I guess it depends on where the fault occurred.

Increase conductors enough in size and the raceway will need to increase in size also.

 

[don_resqcapt19]

necnotevenclose,
Are you asking about cases where the ungrounded conductors are larger than what would normally be installed with the OCPD, or are you asking about upsizing the EGC with standard sized ungrounded conductors in a circuit that has a very high available fault current?

I believe that one of the Bussmann publication has some information about upsizing the EGC for high available fault currents where the ungrounded conductors have not been increased in size.

 

[necnotevenclose]

Don,

I was asking about upsizing a standard EGC for a high available fault current. During design I typically size conductors based on the load / breaker rating. I've yet to upsize the standard GEC based on the AFC. Its been my understanding that the breaker / fuse should trip prior to the conductor melting. By providing a breaker that has an AIC for the AFC does that mean that the GEC is protected or would I need to upsize the GEC to handle the AFC?

 

[Little Bill]

Not sure how many do but it is a requirement based on art. 250.122(B). I have done it.

Dennis, I think the OP was talking about just increasing the EGC not the ungrounded conductors.
Of course if you upsize the ungrounded, you have to proportionally upsize the EGC.

But he is talking about just upsizing the EGC to account for a higher fault current.

I could be wrong, but that is the way I'm reading the OP.

 

[__dan]

I saw something but it was like 25+ years ago, in the IEEE colorbooks. They ran field tests of typical large plant fault currents comparing wiring methods and measuring the voltage elevation at the load end, during the fault carrying. RGS was best, barely raising an elevated voltage under fault, IMC, heavywall aluminum, and EMT were good, copper by itself was OK but not as good as EMT by itself, and worst by that test was cable tray. I do not recall if I saw them test conduit plus copper EGC, but the result was clear, lower impedance due the surrounding and net cancelling of the magnetic field was more important than the resistance of the conductor.

That is old information but should representative IIRC. Could use a second look.

Fault destruction due to heating can be expected to be localized, where P = I^2R, P, in heat, will develop where R is located, switching contacts, the arc location, while R or Z is low along the rest of the fault path.

 

[Smart $]

Don,

I was asking about upsizing a standard EGC for a high available fault current. During design I typically size conductors based on the load / breaker rating. I've yet to upsize the standard GEC based on the AFC. Its been my understanding that the breaker / fuse should trip prior to the conductor melting. By providing a breaker that has an AIC for the AFC does that mean that the GEC is protected or would I need to upsize the GEC to handle the AFC?

FWIW, I've not experienced or heard of anyone upsizing an EGC because of higher AFC. Also not aware of any requirement to do so either.

 

[don_resqcapt19]

The Bussmann document referenced 250.4(A)(5) for the code rule that could require an increase in size of the EGC where there is high available fault current.

 

[don_resqcapt19]

FWIW, I've not experienced or heard of anyone upsizing an EGC because of higher AFC. Also not aware of any requirement to do so either.

I have never seen it done in the field, but there was information included on this subject in the NJATC grounding class. I have not taught that class for many years and not sure if they still include this information in the course.

 

[kwired]

Available fault current is going to change depending on size of conductors. If you run larger conductors you will have more available fault current at the load end, supply end remains constant if no changes are made ahead of that point.

Most of the time we are usually concerned about opening an overcurrent device rapidly should there be a ground fault. Increasing the size of the EGC will generally help increase that time because it will have less impedance. If available fault current is already high at the source end, most general purpose cases you wouldn't need increased size of an EGC, you will get quick response.

I suppose you could have circumstances where you have fuses designed to be faster responding than most general purpose applications and maybe you want larger EGC to assist the functioning of those fuses, but anything beyond what is mentioned in 250.122 is design issues and not code issues.

 

[don_resqcapt19]

I did not go back and look up the information from the old NJATC grounding class, but as I recall the point was that with very high available fault currents, that the temperature of the EGC could exceed the damage point for it and the circuit conductors before the OCPD cleared the fault.

 

[kwired]

I did not go back and look up the information from the old NJATC grounding class, but as I recall the point was that with very high available fault currents, that the temperature of the EGC could exceed the damage point for it and the circuit conductors before the OCPD cleared the fault.

But those same conductors are also going to limit how much fault current flows.

Connect a 100 amp circuit with 100 amp conductors and a certain amount of fault current is available at the load end.

Connect a 15 amp circuit with 15 amp conductors of same length to the same source, and the amount of available current at the load end will not be the same, even if you doubled the size of the EGC.

Available fault current at the supply end is the same in both cases.

 

[david luchini]

But those same conductors are also going to limit how much fault current flows.

Connect a 100 amp circuit with 100 amp conductors and a certain amount of fault current is available at the load end.

Connect a 15 amp circuit with 15 amp conductors of same length to the same source, and the amount of available current at the load end will not be the same, even if you doubled the size of the EGC.

Available fault current at the supply end is the same in both cases.

Consider a high available fault current installation, where you have a 400A panel feeder (4#500mcm + #3G) where the available ground fault current is 50,000A at the panel and the OCPD should clear the fault in 1/2sec. IEEE 141 says that the for a 50,000A fault and 1/2 sec clearing time, the minimum conductor size should be #1/0 AWG. The #3 ground cannot safely carry the fault (per 250.4(A)(5)) and should be increased to #1/0.

 

[kwired]

Consider a high available fault current installation, where you have a 400A panel feeder (4#500mcm + #3G) where the available ground fault current is 50,000A at the panel and the OCPD should clear the fault in 1/2sec. IEEE 141 says that the for a 50,000A fault and 1/2 sec clearing time, the minimum conductor size should be #1/0 AWG. The #3 ground cannot safely carry the fault (per 250.4(A)(5)) and should be increased to #1/0.

So if supplying the circuits in my example from the panel in your example - I guess you are talking about the supply end of the feeder, but if feeder is pretty short, most of that 50kA will still be available at the panel, then the 15 amp circuit is pretty much impossible to supply unless you can come up with a 15 amp device with lugs that will accept 1/0, and the 100 amp circuit will require 1/0 conductors minimum.

 

[iceworm]

slow poster: As dave, don, and others have discussed:

...I've ran lots of short circuit calcs and have yet to look at upsizing the grounding conductor, ...

I have only run into this a few times - always high available SCC and short feeders, 500A or more.

Part of coordination is verifying the feeder OCP trips inside of the cable damage curve. Consider a 2000kva, 480V, 5.7% xfm - 42KA at the secondary. Using a 600A feeder for example, parallel 350s, and a #1 EGC (yes, hard to arrange, but this is just an example). Another example is a 500A feeder with 750s, and #2 EGC. Either of these cases could have the cable damage curve outside of the OCP trip.

...how many people are upsizing the GEC during design for the available fault current?

Me personally? Never - I do as little design as possible. But I always look for the cable damage curve plotted on the coordination sheet. If it is not there and its a short, fat feeder with a high available SCC, I'll ask for it.

Generally the design firms have already accounted for any OCP/cable damage issues. A few times (maybe twice?) I've seen the CB curve crawl outside a cable damage curve.

ice

 

[Smart $]

Consider a high available fault current installation, where you have a 400A panel feeder (4#500mcm + #3G) where the available ground fault current is 50,000A at the panel and the OCPD should clear the fault in 1/2sec. IEEE 141 says that the for a 50,000A fault and 1/2 sec clearing time, the minimum conductor size should be #1/0 AWG. The #3 ground cannot safely carry the fault (per 250.4(A)(5)) and should be increased to #1/0.

Is there any ground fault detection schema which would mitigate the need for a larger EGC?