Raceway as EGC

ron · Mar 21, 2004

Re: Raceway as EGC

How about this angle.
250.118 identifies the EMT as the "equipment grounding conductor".
250.112(A) tells us to comply with 250.4(A)(5).
250.112(B) says to increase the equipment ground conductor (EMT in this case) proportionately.

Keep in mind, I have never even considered this concept as a code requirement (just logical), so I'm glad we are discussing it. See, engineers learn stuff too every day.

pierre · Mar 21, 2004

Re: Raceway as EGC

Bob
I was wondering when that issue might pop up

I think that falls to the prescriptive part of the code as opposed to the performance part of the code. You are right maybe it could be addressed by the CMP, just remember the CMPs like and want our input and help.

Scott
I do not know the values in the provided tables and I am too lazy today to look them up. But think of the 500 foot run and all of the fittings that will also be installed, that has to be a lot of Zs

Pierre

bob · Mar 21, 2004

Re: Raceway as EGC

Iwire said
"A 20 amp 120 volt circuit 500' long on 10 AWG is almost a worthless circuit in the first place."

Bob
I agree with your comments regarding the practical
limitations of a 500 ft circuit but Ryans initial post was regarding the use of conduit as a good
EGC.
One possible solution to Ryans question would be to use AL conduit. The impedance of AL conduit is
lower that EMT but I seldom see it used. Why not?

It is my understanding when using tables 250.66 and 250.122 the distance is limited to 100 ft approx. This of course is not mentioned in the NEC. It becomes a design problem. Can someone comment on this issue.

iwire · Mar 21, 2004

Re: Raceway as EGC

Originally posted by bob:
Iwire said
"A 20 amp 120 volt circuit 500' long on 10 AWG is almost a worthless circuit in the first place."

Bob
I agree with your comments regarding the practical
limitations of a 500 ft circuit but Ryan's initial post was regarding the use of conduit as a good
EGC.

Right but the conduit is only half the circuit.

A line to line fault at the end of these 10 AWGs would only result in a fault current of about 90 amps at the breaker.

If the conduit had zero resistance (I know not possible) a fault to ground at the end of this circuit would provide 180 amps of fault current.

The 1/2" EMT is only half of clearing the fault.

My point is both the conductors and the conduit have to be taken into account, one with out the other means little.

ryan_618 · Mar 21, 2004

Re: Raceway as EGC

I'm glad to see that this thread is starting to take off a bit. When I try to teach people about grounding, I spend a lot of time discussing 250.2 and 250.4., since I think they do an admirable job of telling the intent of a very misunderstood section of the code (art. 250).

The only problem that I have with this section is, like Bob pointed out, it is very subjective. In my opinion, it gives the AHJ the authority to require a test on each circuit if he/she beleives it is neccassary. Now, I'm not saying that I would ask for a test to be performed, but I do think it would be just about the only way to ensure compliance.

roger · Mar 21, 2004

Re: Raceway as EGC

I want to comment on this issue in the practical installation sense.

Personally I like to see a wire conductor.

In many cases the engineer calls for "Steel Compression Fittings" over "Steel Set Screw Fittings" in interior installations. (WHAT IS THE REASON FOR THIS) I did work with an engineer on a Hospital project once who specified Compression fittings were not acceptable in Patient Care areas for the redundant grounding requirement.

Many installers will and want to do a good job, but will get frustrated at times in tight areas where two pairs of adjustable pliers just won't fit, so this fitting is left hand tight at best.

Now the same fitting if Set Screw will be tight.

In either situation, if installed perpendicular to metal joist or similar constructed buildings,
There would be a much better path due to the spans between straps simply being short jumpers.

Roger

iwire · Mar 21, 2004

Re: Raceway as EGC

Originally posted by pierre:
Bob
I was wondering when that issue might pop up
I think that falls to the prescriptive part of the code as opposed to the performance part of the code. You are right maybe it could be addressed by the CMP, just remember the CMPs like and want our input and help.

I think we look at this from different angles. What a news flash.

I think your perspective is this issue is something the CMPs overlooked.

My perspective is that during the creation and modifications of 250.4(A)(5) they discussed many possibility's on the wording and they intentionally chose the wording or lack there of, for reasons that I do not understand.

I am not trying to take short cuts, talk to my PMs

they would like it better if I was not so into following the derating rules, EGC sizing, circuit loading etc.

Ryan if you where to ask for a test what is the acceptable result?

What is the correct amount of fault current needed to operate a OCPD and where does the NEC tell me that?

ryan_618 · Mar 21, 2004

Re: Raceway as EGC

Originally posted by iwire:

Ryan if you where to ask for a test what is the acceptable result?

What is the correct amount of fault current needed to operate a OCPD and where does the NEC tell me that? [/QB]

Hi Bob. I think that 110.9/110.10 might be used.

As far as the range of overcurrent required to trip the OCPD, I have been led to beleive that it is about 4 times the rating of the breaker, but I agree, the NEC doesn't tell me that.

As far as the test goes, perhaps a meggar could be used to determine the exact impedance of the EGC (raceway) and Ohm's law could be used to determine whether or not the OCPD would be initiated.

Now, 250.4(A)(5) tells us that the effective ground fault current path must safely carry the fault. With that in mind, I think the only way to SAFELY carry the fault is to initiate the OCPD. If the OCPD doesn't open and you have an energized raceway, have you safely carried teh fault? I wouldn't think so.

bob · Mar 21, 2004

Re: Raceway as EGC

Now, 250.4(A)(5) tells us that the effective ground fault current path must safely carry the fault.

Ryan
What that statement means is that the EGC must be able to with stand the available fault current for a given amount of time. As an example #6 cu can with stand about 15000 amps for 1 cycle. If the available fault was 20000 amps, the #6 would not be adequate if you are using the 1 cycle time limit.
As far a measurements, a megger would not give you the results you want. Meggers are usually used to measure insulation resistance. There is equipment available ,with which I am not familar,
that will measure the impedance of the grounding system.
One additional point regarding the OC device opening. 70 amps will trip a 20 amp breaker. The question is how long will it take. Don't know without seeing the curves.

[ March 21, 2004, 03:14 PM: Message edited by: bob ]

pierre · Mar 21, 2004

Re: Raceway as EGC

There are a lot of parameters involved here. Short circuit current available and the impedance of the circuit conductors. Calculating all of this is not my strong suit and is better left up to the experts like Ron.
The circuit breaker and fuse manufacturers have a lot of this calculated, learning how to read and understand their charts/tables is part of the key to this - again it is not easy. Bussmann has great material for this issue.
On some jobs it takes the coordination of the POCO, engineer(s), manufacturers and electricians to help make it work.
A groundfault on a 20 ampere circuit can produce thousands of amperes of current flow depending on the above mentioned parameters. This is the goal as long as the devices used are selected for the proper AIC, and the wiring is correct.

This leads back to the original post - can a long run of conduit be depended on to ensure enough current flow to open the OCPD in a safe fashion?

Pierre

iwire · Mar 21, 2004

Re: Raceway as EGC

Pierre I am in the same boat as you I am certainly no expert here the fault current at the panel this circuit originates in may be thousands of amps.

But I am under the impression that once you run 120 volts through 1000' (500' x 2 of 10 AWG copper the highest fault current at the far will be about 93 amps.

10 AWG is 1.29 ohms per 1000'

120 volts / 1.29 = 93 amps for a line to line fault.

93 amps should certainly open the breaker but the EMT at the length will not be subjected to the full fault current available at the panel.

Please someone that knows this info please set me straight if I am wrong.

bob · Mar 21, 2004

Re: Raceway as EGC

IWIRE writes
But I am under the impression that once you run 120 volts through 1000' (500' x 2 of 10 AWG copper the highest fault current at the far will be about 93 amps.
10 AWG is 1.29 ohms per 1000'
120 volts / 1.29 = 93 amps for a line to line fault.
93 amps should certainly open the breaker but the EMT at the length will not be subjected to the full fault current available at the panel.

Bob
Your caculations are correct assuming a "bolted fault". However in a phase to ground fault you can have arcing. In the Soares study the arc had an impedance that resulted a voltage drop of 40 volts. Obviously this will reduce the level of the fault current. The remainder of the fault circuit is as you say: impedance of the #10 + the
impedance or the EMT.

bennie · Mar 21, 2004

Re: Raceway as EGC

Trying to get the loop impedance to a value that will trip an internal logic circuit breaker is a balancing act.

This is why a GFP is on a 277 volt to ground system. Creating current for a 1000 amp internal logic breaker is near impossible, external logic is required.

scott thompson · Mar 21, 2004

Re: Raceway as EGC

This thread is really getting into motion!

Bennie has brought up a very important key issue with the GFPE for >150VAC to Ground, upto 600VAC L-L, 1,000 Amps and larger services.

The Arc Fault on these Wye connected 4 wire systems will be frying away for days - and the OCPD just looks at it as an unstable Transiently Peaking Load - so no trip!
So the GFPE is used - very similar to the method behind GFCIs, to trip incase of an excessive Ground fault Current (more like excessively unbalanced load currents measured).
The CT used (Donut) has all 4 wires run through it.

So, why do only 4 wire Wye systems have this problem, and 3 Wire Deltas do not (above 150VAC to Ground, upto 600VAC L-L, 1000 amps and larger).
Hint: Look at the connection schemes of each Transformer's setup.

Now here's something totally different to ponder:
An OCPD activates from an Overcurrent / Overload situation, right?
Say the Loop Impedance was great enough on a Circuit (Circuit Conductors + EG paths), to allow only 35 amps flow on a 20 amp circuit.
Is this still a Ground Fault?
Since this flows to Ground, will the 35 Amps ever trip an OCPD with a 20 Amp rating?

Of course it will! It won't trip fast, it might trip after 15 minutes, but it will eventually trip.
Besides the Shock Hazards to Personnel, will this pose any other dangers to the system or Building?
This should get some different views indeed!

Super Bonus Question!!!

What will result from the following:

120/240V 1? 3W 37.5 KVA Transformer - Z = 2%,
Transformer is 20 feet from Panelboard,
Panelboard is 200 Amp bus, fed with 3 # 3/0 CU + 1 #1 cu for the EGC,
Branch Circuit has 40 Amp OCPD - 10KAIC,
Branch Circuit is 2 #6 cu, 1 #8 cu EGC - 1" Aluminum IMC,
Six 5 HP 1? Cap. start Induction Motors running on system, loaded to 90% capacity,
Ground Fault on the above Branch Circuit occurs 10 Feet from Panelboard.

What happens if we swap the #6s with #10s?
What if we increase the KVA to 75 KVA, or 112.5 KVA with Z = 1.4%?
What if we move the Transformer farther away, use smaller feeders and higher %Z?
What if the Ground Fault occurs 100 feet away from Panelboard?
What if all the Motors were starting at the time of the Fault?
What if all the Motors shut off at the time of the Fault?
What if both Ungrounded Conductors Fault to Ground at the same time?

To most, these will be easy - even laughably simple! To others, it might be a bit difficult at first. Some may not know there's any difference, or know of these factors - which is why I wanted to toss this in and get some feedback.

I picked the numbers at random, so feel free to do the calcs and post them.

Scott35

pierre · Mar 21, 2004

Re: Raceway as EGC

Hello Scott
I bet the temps in California are a little warmer than the 30 degree spring like weather we are having here.

that the short circuit fault current available to these homes is higher than most homes that are pole fed, especially the ones with the pole located further from the home.
How many of us are contacting the POCO to find out what the short circuit fault current available is for this type of installlation?

Pierre

ryan_618 · Mar 21, 2004

Re: Raceway as EGC

Originally posted by pierre:
How many of us are contacting the POCO to find out what the short circuit fault current available is for this type of installlation?

Pierre

The POCO here will not give the available fault current of their transfomers. They want you to you a Z rating. I believe this is true with Charlie E's power guys as well.

don_resqcapt19 · Mar 22, 2004

Re: Raceway as EGC

Ryan,
While I agree that the utilities don't like to give the available fault currents, that lack of information can in some cases create a safety hazard. If you use the impedance, the KVA rating of the transformer and assume an infinite primary current you can come up with a short circuit value. This value will be artificially high and using it to pick the electrical system components will provide a safe system, but using this high value for arc/blast/flash calculations may in some cases result in a selection of PPE that is not adequate to provide the required protection. This is because the PPE selection is partly based on the time it will take to clear the fault. Where the fault current used for these calculations is higher than what is actually available, you will show a shorter fault duration time(faster clearing time) and this may lead to the selection of PPE that is not suitable for the actual fault duration time.
Don

bob · Mar 22, 2004

Re: Raceway as EGC

Don
Thats a great point you make. Had not thought about that. I did a quick caculation using
2500 ava 480 volt 5%. Assumed a utiltiy system
of 12.47 kv and available fault of 6000 amps. This
is a large fault for many utility systems. Most
will have smaller available fault currents.

Fault current using only 5% = about 60 ka
Fault current including the utiltiy system was
43 ka.

[ March 22, 2004, 01:20 PM: Message edited by: bob ]

charlie · Mar 22, 2004

Re: Raceway as EGC

Don, you are correct and that is the reason that EEI is opposing the requirement to mark the actual values of incident energy. Assume that we look on the nameplate to give you the actual impedance of the transformers and then we replace the bank during a storm. Who gets notified and when? It may take us a couple of months to wade through all the paper generated by a major storm and who knows what information the line crew will write on the trouble sheets (especially if it is from out of town).

scott thompson · Mar 29, 2004

Re: Raceway as EGC

Darn,
No one came up with any numbers or replies to the example scenarios I posted!

OTOH, the post did seem to make a crashing conclusion to the discussion

Anyone want to post any numbers?
(per my reply message in this page, with the "Super Bonus Question")

Scott35

Raceway as EGC

Senior Member

Senior Member

Senior Member

Moderator

Senior Member

Moderator

Moderator

Senior Member

Senior Member

Senior Member

Moderator

Senior Member

Esteemed Member

Senior Member

Senior Member

Senior Member

Moderator

Senior Member

Senior Member

Senior Member