Ground fault protection

[CarRamrod]

I have a question regarding ground fault protection.

We are installing emergency power distribution. We are transforming from 600V to 4160V, then back down, to power some equipment that is far away from the emergency generator. XFMs are delta to resistance grounded Y.

We are installing the 600V cable for 500A and the 4160V cable for 75A.

My question is:
Is ground fault protection necessary for this kind of setup? For the 4160V or the 600V equipment? I'm in Canada, but NEC references (and quotes!) would still be helpful for some guidance at least, as I've never done this kind of thing before.

Thanks in advance.
Ron

 

[ramsy]

I remember a similar thread that involved an airport ops several hundred feet from the svc.. People suggested running the HV service underground to one stepdown, rather than multiple step downs & backfeeds.

That thread seems lost to mikeholt.com & google, perhaps purged like others, but if you are referring to NEC 240.13

240.13 Ground-Fault Protection of Equipment.
Ground-fault protection of equipment shall be provided in accordance 
with the provisions of 230.95 for solidly grounded wye electrical systems 
of more than 150 volts to ground but not exceeding 600 volts phase-to-phase 
for each individual device used as a building or structure main disconnecting 
means rated 1000 amperes or more.

The provisions of this section shall not apply to the disconnecting means for the following:
(1) Continuous industrial processes where a nonorderly shutdown will introduce additional or increased hazards
(2) Installations where ground-fault protection is provided by other requirements for services or feeders
(3) Fire pumps installed in accordance with Article 695

The NEC may not required GFPE for Equipment, in cases like yours, but designers may still include them.

My "Practical Electrical Wireing 19th ed." claims, with an adj.trip, 1200A @ 1sec delay is the max setting permitted for a main breaker, but actual settings are typically more sensitive than NEC max limits, as users come to understand the awesome destructive power involved in these systems."

I am learning, svc. & feeder GFPE settings are higher than NEC branch GFI's of 6-30mA, but I don't know how common GFPE of 600vac 500A should be. Anyone else?

 

[CarRamrod]

The CEC has a similar rule. So am I right in interpreting that for anything above 600 V, ground fault protection is not necessary?
And for <600V equipment, if it's rated less than 1000A it's unnecessary too?

What would be the general reason for the rule? Why is 600V @ >1000A the cutoff?

I'm not looking for an actual explanation of why the code writers do this, but a general idea/guess of why this is necessary would be great. I'm just looking to understand what might happen if you have over 1000 amps, and a ground fault occurs. If you have over 1000A, does a ground loop scenario become hazardous?

 

[CarRamrod]

Also, what if you have a resistance grounded transformer? Since it's not solidly grounded, do ground faults become less of an issue?

I know I've asked a lot of questions here, but as I friend of mine recently said, "It's important to go to sleep less stupid than you were when you woke up." Any help is appreciated!

 

[ron]

From the NEC
250.186 Impedance Grounded Neutral Systems.
Impedance grounded neutral systems in which a grounding impedance, usually a resistor, limits the ground-fault current, shall be permitted where all of the following conditions are met:
(1) The conditions of maintenance and supervision ensure that only qualified persons will service the installation.
(2) Ground detectors are installed on the system.
(3) Line-to-neutral loads are not served.
Impedance grounded neutral systems shall comply with the provisions of 250.186(A) through (D).

(A) Location. The grounding impedance shall be inserted in the grounding conductor between the grounding electrode of the supply system and the neutral point of the supply transformer or generator.

(B) Identified and Insulated. The neutral conductor of an impedance grounded neutral system shall be identified, as well as fully insulated with the same insulation as the phase conductors.

(C) System Neutral Connection. The system neutral shall not be connected to ground, except through the neutral grounding impedance.

(D) Equipment Grounding Conductors. Equipment grounding conductors shall be permitted to be bare and shall be electrically connected to the ground bus and grounding electrode conductor.

 

[ron]

On a resistance grounded system, the ground faults are less destructive because the resistor limits ground fault current.

 

[winnie]

I remember a similar thread that involved an airport ops several hundred feet from the svc.. People suggested running the HV service underground to one stepdown, rather than multiple step downs & backfeeds.

I think that this was the airport thread:
http://www.mikeholt.com/code_forum/showthread.php?t=77681

-Jon

 

[CarRamrod]

So, if you have a location rated less than 1000A, the code doesn't require ground fault protection. Is it safe to say that if there is a ground fault on a system rated <1000A, other than the actual equipment or cable where the fault is, the other equipment/cable in the system shouldn't be harmed?

Obviously it'd be a judgement call depending on each specific location, and you may or may not want to have protection. But that's the message I'm getting here.

And if your system is rated at a really high current, you would want to have a neutral grounded resistor, to limit the ground fault current.

Neat.

Thanks.

 

[CarRamrod]

I think that this was the airport thread:
http://www.mikeholt.com/code_forum/showthread.php?t=77681

-Jon

Thanks for the link, but this is an emergency system. I.E. if the utility goes down, this system kicks in.

Cheers

 

[winnie]

Sorry, I should have sent that link as a PM. It was the discussion that ramsy was thinking of (I think), but is not relevant to the current discussion.

On your statement:

So, if you have a location rated less than 1000A, the code doesn't require ground fault protection. Is it safe to say that if there is a ground fault on a system rated <1000A, other than the actual equipment or cable where the fault is, the other equipment/cable in the system shouldn't be harmed?

This is not necessarily true. Code making is always attempting to balance cost versus benefit on a 'lowest common denominator' basis. The amount of fault energy available, the time that it could take OCPD to clear the fault, the cost of GPF, the manufacturers representatives for the GFP companies, etc. etc. convinced the code making panel that it was worth _requiring_ GFP on these high current systems, but not enough of a cost/benefit improvement to _require_ GFP on smaller systems. A good _design_ might conclude that you _want_ GFP on smaller systems.

But there could certainly be damaging faults on lower current or lower voltage systems which would not be quickly cleared by the OCPD, thus causing extensive damage. A high impedance ground fault on any system could permit sufficient current to flow to cause lots of damage, but not enough current to flow to trip the OCPD.

If you use an impedance grounded system, GFP may be required, not because of code directly, but instead by the manufacturer of the grounding impedance or other equipment in the system. The grounding impedance may have a time rating and require that the fault be cleared before the impedance overheats. I don't have much experience in this field, and don't know if anyone installs systems with time rated grounding impedances, but when I was looking at impedance grounded systems, I noted that the resistors often had both continuous and time limited ratings.

-Jon

 

[ramsy]

I think that this was the airport thread:
http://www.mikeholt.com/code_forum/showthread.php?t=77681

Yes, that was it. Searching for "airport" got me nothing. How'd you do it, prey tell?

Winnie and tx2step nicely explained the waste of using multiple xfrmrs when a distro. line is available. I see here we are stuck with an emergency generator's 600vac output.

I agree with Jon & Ron's suspicion, that an engineering consultant should check if multi-xfmr impendances will choke or overload that generator, if not match your orderly shut-down procedure with generator protections, which may exclude GFPE's altogether, or limit them to specific equipment.

Benaround also points out using GFPE's can depend on the design of your ATS. "The four pole ATS is normally used on services that have GFPE protection."

 

[CarRamrod]

Jon,

Thanks. I think we've decided to put in an NGR on our last transformer, because it's relatively cheap (it's a big project). We'll also use a monitor on the NGR to detect ground faults.

Regarding GCP, and just for this thread's edification, we are confident now that the OCP we are providing will be sufficient, now that I've investigated a few things in depth.

I appreciate the help. I frequently find that when I ask questions here or read threads looking for a certain piece of information, that I see a lot of helpful comments on other topics too.

Cheers,

The other Ron

 

[haskindm]

Ron,
Notice that the NEC requires that "qualified people" service the equipment if an impedence grounded system is to be installed. You may want to check how that is being interpreted in your area.
Also, the simple explanation for GFI on a high amperage circuit or feeder is to limit the damage done by an arcing fault. It would take a great deal of energy to trip a 1000 amp breaker with an arcing fault and much damage could be done or a fire start before the fault is cleared. The 1000 amp is an arbitrary number that was determined by a committee to be the best cost/benefit compromise. There is probably nothing magical about the number. There is no tremendous difference in danger between 900-amps and 1000-amps, it was just a number that the committee could agree on.