250.122(b) sizing ground

Its only a issue with very long runs, say large parking lot lights or long path lighting If you play around with the numbers you could see situations where the available fault current is not enough to trip a breaker or it would take hours for either the breaker to trip or the weakest link in the equipment ground fault path burns out.
For example a large corporate campus with 480 service has a single phase step down transformer 8000A available fault current at the 120/240 secondary of the transformer,
Then a 500' run of #4 AL to a outdoor lighting panel out in a parking area,
then 1500' run of path lights on #8 AL and you pulled a #8 AL equipment ground.
Say the lights are really efficient LED so voltage drop is not an issue, they also are wired 240V to reduce voltage drop,
the #8 AL direct burial cable is on a 2-pole 40A breaker.
Now say we have a bolted fault to equipment ground at the last light pole.
The available fault current line to ground is something like 33 amps.
Arguably a #8AL EGC could carry that fault current all day long, or until the fault burns up and opens.
 
winnie said:
Thanks for the time pointer into the video.

The software Ryan links does _exactly_ the calculation necessary, evaluating wire resistance combined with the source available fault current to determine the maximum circuit length that will still permit the target ground fault current to flow.

The graph he shows is a common trip curve for a small breaker, and it is exactly what you need to understand to pick the target ground fault current.

Say you have a 20A circuit for a gate opener. You upsize the circuit conductors to 6 awg for voltage drop. Now you want to size the EGC per the exception.
1) Get the breaker trip curve from the manufacturer. They are all similar, but you need the one that matches your breaker.
2) Look at the trip curve graph, and select the 'rating multiplier' necessary to get the desired trip time.
3) Enter the wire information (circuit conductor AWG and guessed EGC AWG), the breaker rating, and the multiplier number into the software.
4) The software will calculate the maximum allowed length.
5) Adjust the EGC AWG until the maximum allowed circuit length matches the actual circuit length.

View attachment 2576224

So far so good, once you have a well selected 'current multiplier', you have a procedure that will give you the necessary EGC size.

Where I disagree with Ryan is selecting that current multiplier. Nothing in the code says how fast the breaker needs to trip for an 'effective' ground fault path, so Ryan has no basis other than his common sense for selecting that 5x multiplier. One key feature about trip curves: they all have a 'tolerance range'. If you look at the vertical line for 5x current, you will see that the breaker trips somewhere between 0.005 and 4 seconds.

If you want to _guarantee_ that the breaker trips faster than 0.02 seconds, you need to use a 10x multiplier. And if the breaker has a 'D' trip curve (they sneak into that diagram _two different_ breaker trip curves, type 'C' and type 'D'), and want to ensure a 0.02 second trip, you need a 20x multiplier. But since the code doesn't specify the required trip time, a qualified individual might say 'tripping in 10 seconds is fast enough' and get away with a 4x multiplier.

Or, as I previously mentioned, if the breaker has some sort of ground fault detection, perhaps you don't a high fault current at all. If you have a 20A breaker that trips in 1/20 second in the event of a 50mA ground fault, are you permitted to use a multiplier of 0.0025? (50mA = 0.0025 * 20)

If you have a strong basis for selecting the breaker multiplier (or breaker required trip time), then you will have everything necessary to use this procedure. But right now IMHO the only basis for selecting the breaker multiplier is Ryan's good sense.

-Jonathan
Click to expand...
Trip times of 5X are based on the insulated cable mfgs assn data, to limit overheating of ungrounded conductors. This is explained in the IEEE green book. The GEMI software program from the steel tube institute can determine the size of the EGC (free download).
Soares Book on Grounding has a section on testing steel conduit as an EGC
 
winnie said:
What if you have a GFCI, does that make a fault current of 20mA 'effective'?
Click to expand...
I say yes absolutely in these situations with very long runs of path or parking light poles in very public places I'd look at using a GFPE breaker (formerly a class B GFCI) If I remember correctly a GFPE breaker trips in 2 seconds at 45ma 150% of its 30ma rating.
 
tom baker said:
Trip times of 5X are based on the insulated cable mfgs assn data, to limit overheating of ungrounded conductors. This is explained in the IEEE green book. The GEMI software program from the steel tube institute can determine the size of the EGC (free download).
Soares Book on Grounding has a section on testing steel conduit as an EGC
Click to expand...
Ok, so the basis for requiring fault current of at least 5x breaker trip current is from the insulated cable mfgs association. Presumably when you have a breaker normally sized for the conductor, if the fault current is > 5x the handle rating then the slowest the breaker will trip is still fast enough to protect the cable.

But is this 5x number still valid with cables that are increased in size? If the ratio of <cable ampacity> : <breaker trip rating> increases, then the cable sustain less heating for the same breaker trip time; think about the extreme case of a system where the fault current is less than the ampacity of the conductor.

But I guess 'fault current of 5x trip rating' is considered good practice, and that supports using that number in the calculations
 
winnie said:
But I guess 'fault current of 5x trip rating' is considered good practice, and that supports using that number in the calculations
Click to expand...
Its easy to fall below that 5x rating with long runs of outdoor lighting.
 
tortuga said:
For example a large corporate campus with 480 service has a single phase step down transformer 8000A available fault current at the 120/240 secondary of the transformer,
Click to expand...
How does the available fault current relate to the ground faults of branch circuits on a system. I know it affects it I’m just not sure how/why.
 
infinity said:
If the circuit requires #8 ungrounded conductors for a 20 amp circuit then the EGC should also be #8.
Click to expand...
Now put a 40 or 50 amp breaker on same conductors and suddenly a 10 AWG can possibly carry sufficient current to trip a breaker with higher trip setting.

Could even have a motor load, same 8 AWG conductors and possibly up to a 60 amp breaker and only 10 AWG EGC.

Kind of comes down to how high do we want the fault current to be to matter as to whether the breaker trips on magnetic vs thermal trip. One way resistance is already reduced by oversizing your conductors for voltage drop reasons, so that helps the cause some.

There is a difference between say a 500 foot run and a "normal sized EGG" being sufficient vs say a 2000 foot run where same sized EGC as ungrounded conductors still won't trip an overcurrent device before some other component in the circuit fails first. I see that happen frequently with long runs and irrigation equipment.
 
Pinnie said:
How does the available fault current relate to the ground faults of branch circuits on a system. I know it affects it I’m just not sure how/why.
Click to expand...
The more current that flows during a fault the faster it will reach a trip point in the trip curve of a particular OCPD.

Lower level of fault current may still trip the device but will have fault current for longer duration before it trips.

If magnetic trip setting is say 150 amps but your fault current is limited by circuit impedance to only 135 amps, then you won't get nearly instantaneous trip out of it, instead it will have to force it to trip on thermal overload function which may take a few seconds instead of being nearly instantaneous. The lower the fault current the longer it will take to trip on thermal overload.
 
Pinnie said:
How does the available fault current relate to the ground faults of branch circuits on a system. I know it affects it I’m just not sure how/why.
Click to expand...

In this discussion of long runs, we've been focusing on the resistance of the branch circuit wire limiting the current through the fault and leading to slow breaker tripping. But the reality is that _all_ of the impedance in the circuit reduces the current through a fault and slows the operation of the breaker.

The available fault current at the transformer is usually very high, which is the same as saying the source impedance is very low. Usually the source impedance is tiny compared to the branch circuit impedance, so we ignore it for this discussion. But if the source impedance is high it will slow down the breaker tripping in the same way that a long branch circuit will slow down the breaker tripping.
 
infinity said:
If the circuit requires #8 ungrounded conductors for a 20 amp circuit then the EGC should also be #8.
Click to expand...
See this is actually incorrect, it's not what the rule says.

If the circuit *requires* #8 ungrounded conductors according to NEC rules (e.g. derating for CCCs or temp), then a #12 egc on a 20A breaker is fine.

If #8 ungrounded conductors are used *optionally* (e.g. to avoid voltage drop, or because they were repurposed from a higher amp load), then the rule says to upsize the EGC. (Thank god for the exception.)

Never mind that optionally oversizing the ungrounded conductors (all else being equal) is *never* going to increase the impedance detrimentally to clearing a fault, but rather will do the opposite.

Dumbest rule in the book.
 
jaggedben said:
See this is actually incorrect, it's not what the rule says.
Click to expand...
The example in the OP is increasing the ungrounded conductors for voltage drop for a 500' run which is what I was referring to. Unless I've misread the quote below you've contradicted your response in the same post.


jaggedben said:
If #8 ungrounded conductors are used *optionally* (e.g. to avoid voltage drop, or because they were repurposed from a higher amp load), then the rule says to upsize the EGC.
Click to expand...
 
Pinnie said:
Yeah pretty stupid. the exception is crude but does what it should. How would you size the ground in an upsizing for voltage drop scenario?
Click to expand...
If I was really upsizing for voltage drop because of the *length* of the circuit, I would upsize the EGC per the rule. (Or maybe consult an engineer.) If I'm upsizing for any other reason, in a typical residential context I would consider myself qualified to size the EGC per table 250.122.

We upsize almost all our solar microinverter output circuits for voltage drop to increase energy harvest and avoid voltage-out-of-range dropouts, and we upsize the EGCs as well just because it's always been done that way. But I try to avoid it in other cases.
 
jaggedben said:
I would upsize the EGC per the rule. (Or maybe consult an engineer.) If I'm upsizing for any other reason, in a typical residential context I would consider myself qualified to size the EGC per table 250.122.
Click to expand...
This is definitely the safe approach and what infinity was getting at. The purpose of my posting was if someone wanted to do the steps to undersize the egc per the exception how would one do that.

The steel institute program I’m sure is great but id prefer a formula.

If the transformers available fault current is known then its possible no?
 
winnie said:
In this discussion of long runs, we've been focusing on the resistance of the branch circuit wire limiting the current through the fault and leading to slow breaker tripping. But the reality is that _all_ of the impedance in the circuit reduces the current through a fault and slows the operation of the breaker.

The available fault current at the transformer is usually very high, which is the same as saying the source impedance is very low. Usually the source impedance is tiny compared to the branch circuit impedance, so we ignore it for this discussion. But if the source impedance is high it will slow down the breaker tripping in the same way that a long branch circuit will slow down the breaker tripping.
Click to expand...
In an industrial plant the source impedance is usually rather low.

For services supplying limited load applications often the source is not much larger than absolutely needed, so the available fault current is fairly low at the source terminals compared to a large industrial facility with same supply voltage.

The irrigation equipment I mentioned earlier - some cases if the center pivot is all that is being powered (the well probably powered by diesel engine in this situation) and the POCO doesn't have three phase primary in the area it may only have a 10 kVA @ 240/480 volt single phase transformer and a rotary phase converter to derive three phase for the center pivot. The fault current from that is nothing compared to a plant supplied by a 1+MVA transformer. And when you have a fault clear out at the end of the system it often doesn't even blow any fuses, it burns some component until the circuit is open and the farmer calls someone because they don't know why the thing isn't working right even though nothing is blown/tripped.
 
If its 15 through 50 amp lighting circuit I'd just use a GFPE breaker and a regular sized EGC, if more than 30ma needs to flow on the equipment ground continuously I'd want to know why.
 
You must log in or register to reply here.
Top