upsizing ground wires

[don_resqcapt19]

I've been rethinking this and you're correct. I was thinking of the word capacity instead of ampacity. In the example of a motor and a large voltage drop you could still have a code complaint conductor size from an ampacity perspective but it may not have the capacity to allow the motor to start.

So from this list which ones would require an increase in size of the EGC according to 250.122(B)?

1-Voltage drop compensation
2-Derating compensation
3-Just had larger conductors one the truck

In my opinion only 1 and 3 trigger 250.122(B). Item 2 changes the actual ampacity of the conductor.

Again, this is a very difficult section to work with and it remains my opinion that Table 250.122 should be based on the size of the ungrounded circuit conductor and not the size of the OCPD. That would get rid of all of these questions and doing any calculations. It works in 250.102 for bonding and in 250.66 for GECs, I see no reason why that won't work for Table 250.122.

 

[infinity]

Again, this is a very difficult section to work with and it remains my opinion that Table 250.122 should be based on the size of the ungrounded circuit conductor and not the size of the OCPD. That would get rid of all of these questions and doing any calculations. It works in 250.102 for bonding and in 250.66 for GECs, I see no reason why that won't work for Table 250.122.

I agree and have always liked this approach since it helps to simplify the process of properly sizing the EGC. Now if only you can convince the CMP. :slaphead:

 

[petersonra]

Again, this is a very difficult section to work with and it remains my opinion that Table 250.122 should be based on the size of the ungrounded circuit conductor and not the size of the OCPD. That would get rid of all of these questions and doing any calculations. It works in 250.102 for bonding and in 250.66 for GECs, I see no reason why that won't work for Table 250.122.

what is magic about the size of the ungrounded conductor?

the real problem is that in most cases there is no real reason to upsize the EGC but the code forces you to do so arbitrarily.

 

[iwire]

what is magic about the size of the ungrounded conductor?

the real problem is that in most cases there is no real reason to upsize the EGC but the code forces you to do so arbitrarily.

So you want to eliminate the upsizing? That may well make sense but I doubt that is going to be an easy case to make with the CMP.

On the other hand it seems Don's approach is If we are stuck with it lets at least make it easy to apply.

You would think the CMP would jump to simplify it with an already successful method demonstrated by 250.66.

 

[Smart $]

Wow... thanks everyone for the responses. I saw the first 5 responses and meant to fix my question, just didn't have time to sit down... yes, if you have a #8 wire (at 75 deg C), you have to use the standard size of a 50 amp breaker. A 60 amp breaker is not allowed. Sorry about that! Let me use different numbers, one's that work: (ahem, allow me to try again :angel

I can use a #6 wire (copper, at 75 deg C), which is rated for 65 amps, on a 70 amp breaker... my ground wire would be a #8.
Most engineers i know, including me, would use a #4 wire (at 75 deg C), which is rated for 85 amps, on this 70 amp breaker. Now.. can my ground wire still be a #8, or does it have to be upsized to a #6? (since the 85 amp rating of the wire is much greater than the 70 amp breaker)

I saw infinity's comment that mentioned "Where ungrounded conductors are increased in size from the minimum size that has sufficient
ampacity for the intended installation, wire-typeequipment grounding conductors, where installed, shall be increased in
size proportionately according to the circular mil area of the ungrounded conductors."

So for my example above, is a #6 wire the minimum size that has sufficient ampacity? (since the code allows this for anything less than 800 amps), or does the #4 have sufficient ampacity?

Thanks all again.

All I can say is there are two views on this issue: 1) the minimum required size includes any reduction in size or ampacity by way of permission or exception, and 2) the converse, where the minimum required size does not include any reduction in size or ampacity by way of permission or exception.

I go with the latter, which for your example would mean #4 is not an increased-in-size conductor because using #6 takes advantage of the permission given in 240.4(B).

 

[ActionDave]

You would think the CMP would jump to simplify it with an already successful method demonstrated by 250.66.

We would all like to hope so. What has been the hold up in getting 250.122(B) changed?

 

[don_resqcapt19]

what is magic about the size of the ungrounded conductor?

the real problem is that in most cases there is no real reason to upsize the EGC but the code forces you to do so arbitrarily.

Because then there is no 250.122(B). There is no discussion about "up-sizing" the EGC. If you have "X" size ungrounded conductors you use an "X" size EGC.

The 2017 NEC will go back to up-sizing the EGC only where the ungrounded conductor has been up-sized to account for voltage drop. It will still require calculations and gets us back to the installer telling the inspector, "that was just what I had on the truck, I didn't up-size the ungrounded conductors for voltage drop".

A table requirement based on the size of the ungrounded conductor gets rid of those problems.

 

[ActionDave]

Because then there is no 250.122(B). There is no discussion about "up-sizing" the EGC. If you have "X" size ungrounded conductors you use an "X" size EGC.

The 2017 NEC will go back to up-sizing the EGC only where the ungrounded conductor has been up-sized to account for voltage drop. It will still require calculations and gets us back to the installer telling the inspector, "that was just what I had on the truck, I didn't up-size the ungrounded conductors for voltage drop".

A table requirement based on the size of the ungrounded conductor gets rid of those problems.

So why are we stuck with what we have now? This has to be one of the top two worst code sections.

 

[Smart $]

So why are we stuck with what we have now? This has to be one of the top two worst code sections.

Some people like controversy?


:happyyes::angel::slaphead::blink::happyno:

 

[Carultch]

A table requirement based on the size of the ungrounded conductor gets rid of those problems.

What about a rule such as given X rating of OCPD, the EGC must at least be as large as the value from table 250.122, and the total resistance of the EGC cannot exceed Y ohms?
Y would be a new column in table 250.122, which would then be used with Chapter 9 Table 8, to calculate when EGC upsizing is required, above the default size.

 

[don_resqcapt19]

What about a rule such as given X rating of OCPD, the EGC must at least be as large as the value from table 250.122, and the total resistance of the EGC cannot exceed Y ohms?
Y would be a new column in table 250.122, which would then be used with Chapter 9 Table 8, to calculate when EGC upsizing is required, above the default size.

I am too lazy to want to do all of that. I just want a simple single direct reading table. X size ungrounded = X size EGC.

 

[Carultch]

I am too lazy to want to do all of that. I just want a simple single direct reading table. X size ungrounded = X size EGC.

I'm aware that it may not be the most practical calculation to do, but perhaps a better question to ask is:
Do you think that calculation would better match the theoretical reason behind the requirement to upsize EGC when curtailing voltage drop?

I've heard the argument that the reason behind this is that larger conductors cause more fault currents, and I disagree. I think that is a misleading way to explain the effect of conductor size on fault current. Larger conductors simply don't diminish the fault current as much as their smaller counterparts for the same length. They don't cause more fault current than what would be present on a negligible length circuit.

Limiting the ohms of the EGC, makes the most sense to me, as the reason behind this rule.

 

[ggunn]

I'm aware that it may not be the most practical calculation to do, but perhaps a better question to ask is:
Do you think that calculation would better match the theoretical reason behind the requirement to upsize EGC when curtailing voltage drop?

I've heard the argument that the reason behind this is that larger conductors cause more fault currents, and I disagree. I think that is a misleading way to explain the effect of conductor size on fault current. Larger conductors simply don't diminish the fault current as much as their smaller counterparts for the same length. They don't cause more fault current than what would be present on a negligible length circuit.

Limiting the ohms of the EGC, makes the most sense to me, as the reason behind this rule.

At some point this gets to be tilting at windmills. Compliance with the NEC as written today with respect to this issue isn't that hard, and it isn't the only point in the code that one could argue against on logical grounds. I just want to pass my inspection and move on.

 

[don_resqcapt19]

I'm aware that it may not be the most practical calculation to do, but perhaps a better question to ask is:
Do you think that calculation would better match the theoretical reason behind the requirement to upsize EGC when curtailing voltage drop?

I've heard the argument that the reason behind this is that larger conductors cause more fault currents, and I disagree. I think that is a misleading way to explain the effect of conductor size on fault current. Larger conductors simply don't diminish the fault current as much as their smaller counterparts for the same length. They don't cause more fault current than what would be present on a negligible length circuit.

Limiting the ohms of the EGC, makes the most sense to me, as the reason behind this rule.

I think that a table method eliminates any need for calculations. If a #10 EGC is good for clearing faults on a 60 amp circuit that uses #6 as the ungrounded conductors, than it is good for clearing faults where the #6 is supplied by a 20 amp OCPD. Under the current code, the EGC for a 20 amp circuit with #6s as the ungrounded conductors requires a #6 EGC.

I agree that there is little increase in the fault current remote from the OCPD where the conductor has been up-sized from the code minimum conductor for that OCPD. I just don't think that CMP 5 will ever accept something that completely gets rid of increasing the size of the EGC where the size of the ungrounded conductor has been increased from the code minimum. A table method still provides a limited increase in the size of the EGC where the ungrounded conductors have been up-sized.

 

[kwired]

I have witnessed on field irrigation applications where circuit distances are often quite long that even a properly 250.122 sized conductor doesn't allow enough current to flow to open overcurrent protection. Total circuit impedance is current limiting enough that often a fuse never blows but a connection burns open when there is a fault in these applications. One time I was there when trying to find a problem. Turned on a 60 amp fusible disconnect that I later found out had a line to line fault on the other end and it never blew any fuses. I could hear that it must have had significant current flowing and did not leave it on for very long so don't know how long it may have held, but customer had done same thing sometime before calling me there. This was 480/277 volts with 200 amp supply conductors anywhere from 800 to 1000 feet on the supply side, and 90 amp conductors on load side with maybe 1400 feet to the fault location. The ungrounded conductor has resistance as well as the EGC on such long runs and has an impact on how fast an overcurrent device will operate.