Can someone explain the logic behind the code?

Status
Not open for further replies.
charlie b said:
This article never made sense to me, and I think it should be dropped. I don?t believe that the physics of the situation supports the requirement.

Consider the following example (NOTE: This is a discussion of physics, not code, so please don?t cite 250.122 back at me):
? 20 amp breaker, #12 wire, ?fairly long run,? total voltage drop at 3.2%.
? Nothing requires me to upsize the conductors (i.e., VD is not a code requirement).
? I infer that there is a low enough impedance in the #12 conductors, including the #12 EGC, to result in a high enough fault current to trip the breaker, should there be a fault at the end of the run.
? Let me emphasize: This is a safe installation, because I have confidence that the OCPD would trip, on a fault. The fault current is high enough already!
? Now I choose to replace the phase conductors with #10. I have not yet biggie-sized the EGC.
? That just caused the overall circuit impedance, in the event of a fault, to be lower than it was before.
? Therefore, the fault current will be higher than it was before.
? Therefore, the OCPD will trip even faster.
? Now, tell me, from a physics point of view, why I need to increase the fault current even more, by using a #10 EGC?
Click to expand...

My understanding is different: As the impedence lowers the current goes up like you said and the trip curve should cause a faster tripping. However, EGC are sized smaller than the normal current carrying conductor. Thus there is a concern that the EGC might get damaged before the breaker trips with the higher current going through it. If the EGC gets damaged, then you might not get an OCPD tripped.

I think this has to do with much larger feeder/service conductors and not circuit conducts. Thus I would propose this rule only go into effect for circuits 100A and larger. Or a better one would be that you don't have to increase the size if the next size up conductor OCPD in table 250.122 is used(see the following example of why).

To me here is a better example of the flaw in this code:
You connect a 30A breaker to #10 current carrying conductor (CCC) and a #10 EGC as normally you would do. Then because of voltage drop, you go to a #6 CCC on the 30A breaker. 250.122(B) says that I now have to go to a #6 EGC too. However, if instead I put a 60A breaker in there with the same load calcs I can put a #6 CCC with a #10 ECG. This would be legal and have the same effect. The difference is that the 30A breaker would have actually strip sooner than the 60A breaker solution which would make it safer. So in this case as in all the smaller circuit issues, it seems silly to have to do this if the next CCC lets me use the same EGC from table 250.122
 
Ground fault. I know that the difference in time to trip on a large fault is minimal. The point is that if I can put a #6 with a # 8 EGC on a 60a OCPD surely I won't have any problems with it on a 30a OCPD. However the code thinks I could and wants me to increase the EGC to #6. I believe the instanatious trip time of a 30a breaker is slightly sooner than a 60, but I didn't look.
 
I have to admit that I first thought this increasing the EGC if the circuit conductors are increased made sense, but after more thoughts on it do not think it is necessary either. I think the idea was to make sure there was not a higher impedance in the circuit to ensure operation of the overcurrent device, when in reality increasing the circuit conductors does decrease the circuit impedance and will cause more fault current and faster operation of overcurrent device. Maybe long circuit runs should have some considerations for larger EGC but this is not a case of one rule covers all installations. The way 250.122(B) is worded it means any increase in circuit conductor size requires a equal increase in the EGC.

I have another example of where I don't think this rule should come into play.

Say you have a 10hp 240volt 1phase motor. Typically for a single circuit installed in a raceway and no long run to involve voltage drop one would install 6AWG 75deg copper conductors with 90 amp circuit breaker and 8AWG copper EGC, OR 60 amp time delay fuse and 10 AWG copper EGC.

Now install the same circuit in a raceway with 10-20 current carrying conductors and assume the circuit in question has the largest overcurrent device of the 10-20 conductors.

Derating will require us to increase the 6AWG circuit conductors to 3AWG. You still have the same length of conductors, and the same overcurrent device. You have decreased the circuit impedance because of the increase in conductor size, so if you have a ground fault more current should be allowed to flow resulting in even faster opening of the overcurrent device than in the single circuit in the single raceway.

According to 250.122(B):

If you have a 90 amp circuit breaker the 8AWG EGC will need to be increased to 4 AWG. But I can have a 200 amp feeder in a raceway right next to it with only a 6 AWG EGC:confused:

If you have a 60 amp time delay fuse the 10AWG EGC will need to be increased to 6 AWG. But I can have a 100 amp feeder in a raceway right next to it with only a 8 AWG EGC:confused:
 
hmmm,.. I thought it was because , lets stick to code reasons and voltage design issues ,. the same issue that would drive one to increase an ungrounded conductor would have a like effect on the equipment grounding conductor? Say high temp or wicked long run. those won't effect the impedance of the fault path if it were a wire?
 
M. D. said:
hmmm,.. I thought it was because , lets stick to code reasons and voltage design issues ,. the same issue that would drive one to increase an ungrounded conductor would have a like effect on the equipment grounding conductor? Say high temp or wicked long run. those won't effect the impedance of the fault path if it were a wire?
Click to expand...

and in those cases you still can find times where a smaller EGC and smaller circuit conductor can be used on a higher OCPD than on a circuit with lower OCPD and larger conductors. It may be a rule with good intentions but it does not need to apply in all situations, there needs to be some exceptions or more detail as to when the EGC needs to be larger IMO.

For ever we have sized EGC's to overcurrent devices, now they are leaning towards sizing them to the associated conductors.

An EGC does not need to be sized for continuous loading, it only needs to be large enough to carry ground fault current long enough to operate an overcurrent device without getting hot enough to compromise the EGC itself. A conductor without any insulation can carry many more times the T310.16 values for a few seconds without melting the conductor.
 
M. D. said:
hmmm,.. I thought it was because , lets stick to code reasons and voltage design issues ,. the same issue that would drive one to increase an ungrounded conductor would have a like effect on the equipment grounding conductor? Say high temp or wicked long run. those won't effect the impedance of the fault path if it were a wire?
Click to expand...
Yes, which is why only a proportionate increase is required, not actually matching the ungroundeds.

It's not this article's fault that #'s 14, 12, and 10 require the EGC to match the circuit conductor's size.
 
charlie b said:
This article never made sense to me, and I think it should be dropped. I don?t believe that the physics of the situation supports the requirement.

Consider the following example (NOTE: This is a discussion of physics, not code, so please don?t cite 250.122 back at me):
? 20 amp breaker, #12 wire, ?fairly long run,? total voltage drop at 3.2%.
? Nothing requires me to upsize the conductors (i.e., VD is not a code requirement).
? I infer that there is a low enough impedance in the #12 conductors, including the #12 EGC, to result in a high enough fault current to trip the breaker, should there be a fault at the end of the run.
? Let me emphasize: This is a safe installation, because I have confidence that the OCPD would trip, on a fault. The fault current is high enough already!
? Now I choose to replace the phase conductors with #10. I have not yet biggie-sized the EGC.
? That just caused the overall circuit impedance, in the event of a fault, to be lower than it was before.
? Therefore, the fault current will be higher than it was before.
? Therefore, the OCPD will trip even faster.
? Now, tell me, from a physics point of view, why I need to increase the fault current even more, by using a #10 EGC?
Click to expand...

Just a thought: Maybe the "NEC Gods" are figuring that whatever equipment is being fed can be upgraded someday to a larger piece of equipment and the electrician will say "no problem, the conductors going to the breaker are big enough!":-?
 
Status
Not open for further replies.
Top