250.122(B)

[hhsting]

If one increases phase conductors due to voltage drop then 250.122(B) also requires wire type EGC to be increased in size.

Anyone have have idea if one does NOT increase the wire type EGC and increase phase conductors only due to voltage drop then what happens?

 

[electrofelon]

Anyone have have idea if one does NOT increase the wire type EGC and increase phase conductors only due to voltage drop then what happens?

What usually happens, is a bus full of nuns has a head on collision with a tractor trailer full of kittens fresh from the kitten factory, and everyone dies.

 

[LarryFine]

Anyone have have idea if one does NOT increase the wire type EGC and increase phase conductors only due to voltage drop then what happens?

In case of a fault, the reduced (not increased) EGC might have too much impedance to operate OCP, or slowly.

 

[Carultch]

If one increases phase conductors due to voltage drop then 250.122(B) also requires wire type EGC to be increased in size.

Anyone have have idea if one does NOT increase the wire type EGC and increase phase conductors only due to voltage drop then what happens?

If you have too many ohms in the ground-fault current path, you may not effectively trip the breaker in the event of a fault. 250.122(B) is included as a way to mitigate this issue for particularly long circuits, and reduce to the ohms of the ground-fault current path so it can be effective.

The rule may not really match theory, but it is set up this way so that the inspector doesn't have to measure any lengths or resistances to determine if it is compliant or not. You simply determine the minimum local size for ampacity reasons, determine the number of gauge size increments from that, and see if the EGC is also increased by the same number of gauge sizes. The way the rule is currently written, you could have a 10 ft run where voltage drop is insignificant, but install an oversized wire because "that's what we have on the truck", and still be required to increase the EGC in proportion to it. You could also have a situation where the manufacturer's terminals specifically require you to connect a larger-than-code minimum size, and you'd still also have to increase the size of the EGC, even when length is insignificant.

A way the rule could be written to match the theory behind the rule, is if there were a maximum number of ohms permitted for the EGC. The table in 250.122 would specify the default ground size based on OCPD, and a maximum number of EGC ohms. If your circuit length exceeds that number of ohms, you then increase the EGC size until it is below the maximum number of ohms.

 

[electrofelon]

If you have too many ohms in the ground-fault current path, you may not effectively trip the breaker in the event of a fault. 250.122(B) is included as a way to mitigate this issue for particularly long circuits, and reduce to the ohms of the ground-fault current path so it can be effective.

The rule may not really match theory, but it is set up this way so that the inspector doesn't have to measure any lengths or resistances to determine if it is compliant or not. You simply determine what the minimum local size is, determine the number of gauge size increments from that, and see if the EGC is also increased by the same number of gauge sizes.

A way the rule could be written to match the theory behind the rule, is if there were a maximum number of ohms permitted for the EGC. The table in 250.122 would specify the default ground size based on OCPD, and a maximum number of EGC ohms. If your circuit length exceeds that number of ohms, you then increase the EGC size until it is below the maximum number of ohms.

IT seems like the big flaw is the rule counts on the phase conductors being increased in size due to a longer than "normal" length. Wouldn't it actually be MORE important to increase the size of the EGC if the phase conductors were NOT increased?

 

[don_resqcapt19]

Most of the math shows that the increased size ungrounded conductors will flow more current into the fault and that even with the additional length of an EGC that is not up-sized that approximately the same amount of fault current will flow as compared to a circuit without up-sized ungrounded conductors.

 

[Dsg319]

If you have too many ohms in the ground-fault current path, you may not effectively trip the breaker in the event of a fault. 250.122(B) is included as a way to mitigate this issue for particularly long circuits, and reduce to the ohms of the ground-fault current path so it can be effective.

The rule may not really match theory, but it is set up this way so that the inspector doesn't have to measure any lengths or resistances to determine if it is compliant or not. You simply determine the minimum local size for ampacity reasons, determine the number of gauge size increments from that, and see if the EGC is also increased by the same number of gauge sizes. The way the rule is currently written, you could have a 10 ft run where voltage drop is insignificant, but install an oversized wire because "that's what we have on the truck", and still be required to increase the EGC in proportion to it. You could also have a situation where the manufacturer's terminals specifically require you to connect a larger-than-code minimum size, and you'd still also have to increase the size of the EGC, even when length is insignificant.

Good explanation

 

[Carultch]

Most of the math shows that the increased size ungrounded conductors will flow more current into the fault and that even with the additional length of an EGC that is not up-sized that approximately the same amount of fault current will flow as compared to a circuit without up-sized ungrounded conductors.

By that logic, shortening the length of the circuit, would also "cause more current to flow" in to the fault. Cutting the circuit length in half, would be approximately equivalent to doubling the kcmil, in terms of impact on the available fault current at the end of the circuit opposite the source.

Besides, regardless of what size or length the conductors of a circuit are, the fault current will never be larger than what it is at the source of the circuit. Increasing the size of the circuit conductors, just decreases the fault current by a smaller amount. Given 20kA as the fault current at the panelboard where the circuit is supplied, you will not make the fault current 25 kA by installing larger branch circuit conductors. You might increase the fault current at the end of the circuit from 15kA to 18kA by using larger conductors to mitigate voltage drop, but you aren't going to increase it above 20kA. You'd have to install another source, or a motor load, that adds to the fault current, to actually increase it above 20kA in this example.

 

[don_resqcapt19]

By that logic, shortening the length of the circuit, would also "cause more current to flow" in to the fault. Cutting the circuit length in half, would be approximately equivalent to doubling the kcmil, in terms of impact on the available fault current at the end of the circuit opposite the source.

Besides, regardless of what size or length the conductors of a circuit are, the fault current will never be larger than what it is at the source of the circuit. Increasing the size of the circuit conductors, just decreases the fault current by a smaller amount. Given 20kA as the fault current at the panelboard where the circuit is supplied, you will not make the fault current 25 kA by installing larger branch circuit conductors. You might increase the fault current at the end of the circuit from 15kA to 18kA by using larger conductors to mitigate voltage drop, but you aren't going to increase it above 20kA. You'd have to install another source, or a motor load, that adds to the fault current, to actually increase it above 20kA in this example.

I am not saying you end up with more at the end of the circuit, than at the beginning of the circuit if the size of the circuit conductors is increased, but I am saying a larger portion of the original system fault current will be available at the point of the fault if larger circuit conductors are used.

I worked the following example using Bussmann's FC² app. (note that this app is not intended to do what I did with it, but it is probably good enough for this comparison)

Starting with 10kA of fault current using 3/0 conductors at the end of a 100' run there would be 6065 amps of current available into the fault. I put that as the available fault current for 100' of 6AWG and the fault current at the end of the 6AWG would 1966 amps.

Doing the same thing but up-sizing the 3/0s to 250 kcmil, the available fault current at the load end of the 250s is 6642. Again using the 6642 as the starting point current for the 6 AWG EGC, there would be 2023 amps of available fault current at the other end of the circuit.

In this example there would be almost no difference in the trip time of these too circuits.

If the EGC would be up-sized in proportion, it would need to be increased to 4 AWG. That would result in more available fault current at the end of the EGC than in the original example with 3/0 circuit conductors. The current at the end of the EGC would be 2706 amps. This would result in a quicker tripping time than either of the other two cases.