Effective? Ground Fault Current Path

Status
Not open for further replies.

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
250.2 and 250.4(A)(5) define and require an Effective Ground Fault Current Path which must be permanent, low-impedance, electrically conductive, and which facilitates the operation of the overcurrent device (among other wording and requirements.

My question: what qualifies as 'facilitating the operation of the overcurrent device', and what qualifies as 'low-impedance'?

Do the characteristics of the overcurrent device change the requirements placed by 250.4(A)(5), and in particular, if a GFCI breaker is used, would an otherwise unsuitable conductive path (eg. the armor of old BX cable without the bonding strip) be rendered an acceptable equipment grounding conductor?

-Jon
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
Great topic!

The only hole I see in your train of thought is that facilitating the operation of an overcurrent device is not the same as aggravating GFCI detection circuitry, despite similar results. In a GFCI breaker, (I think) there is the normal thermal and magnetic "overcurrent" components, and then the GFCI secondarily operates those components to de-energize the circuit.

Very interested to hear other's thoughts on this, interesting way to think.
 

iwire

Moderator
Staff member
Location
Massachusetts
Ryan and I have 'fought' about 250.4(A)(5) before.

It is my opinion it is a meaningless section of code as it does not specify any perimeters.

How quickly must the OCPD open, 1 millisecond, 1 Minute or an hour?
 
Last edited:

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
iwire said:
It is my opinion it is a meaningless section of code as it does not specify any parameters.

How quickly must the OCPD open, 1 millisecond, 1 Minute or an hour?
Bob, given that aspect of it, I would tend to agree that this section is more handbook oriented than code oriented. But I think they're being honest and upfront about it.

250.4 General Requirements for Grounding and Bonding. The following general requirements identify what grounding and bonding of electrical systems are required to accomplish. The prescriptive methods contained in Article 250 shall be followed to comply with the performance requirements of this section.

Some sections of 110 are too generalistic for an inspector to feel good about citing when failing an installation. This section is so generalistic it could be in 110, but that would be more confusing, wouldn't it?

It does exactly what it promises to do, IMO. It identifies what the goal is. The rest of the Article is how to accomplish that goal.

It shouldn't be used in practical use of the code, IMO, because it'd be like failing an installation for a xxx.1 violation. If this Article doesn't apply, am I violating that article by not falling under the scope? Some narrative seems to be necessary, to make the rest have context.

So, I guess the answer to the original post's concept is that how sensitive the breaker is doesn't matter. 250.118 provides us with a prescription for an EGC, and doesn't offer lesser ones for more sensitive breakers. So the narrative of 250.4 doesn't really matter.

I still think it is a neat concept, though. I bought into the glitzy packaging and the free fanny pack. :D
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
winnie said:
250.2 and 250.4(A)(5) define and require an Effective Ground Fault Current Path which must be permanent, low-impedance, electrically conductive, and which facilitates the operation of the overcurrent device (among other wording and requirements.

My question: what qualifies as 'facilitating the operation of the overcurrent device', and what qualifies as 'low-impedance'?

Do the characteristics of the overcurrent device change the requirements placed by 250.4(A)(5), and in particular, if a GFCI breaker is used, would an otherwise unsuitable conductive path (eg. the armor of old BX cable without the bonding strip) be rendered an acceptable equipment grounding conductor?

-Jon

The presence or absence of a GFCI does not change the requirement for an adequate EGC. I think low impedence in this context means low enough to reliably trip the BC OCPD in the event of a ground fault, since that is pretty much what it actually says.
<added>
Note the use of the phrase overcurrent device. Clearly, this section is talking about an OCPD, not a GFCI. Just because they are part of the same device, does not mean they are not seperate functions.
 
Last edited:

iwire

Moderator
Staff member
Location
Massachusetts
georgestolz said:
It shouldn't be used in practical use of the code, IMO, because it'd be like failing an installation for a xxx.1 violation. If this Article doesn't apply, am I violating that article by not falling under the scope? Some narrative seems to be necessary, to make the rest have context.

So, I guess the answer to the original post's concept is that how sensitive the breaker is doesn't matter. 250.118 provides us with a prescription for an EGC, and doesn't offer lesser ones for more sensitive breakers. So the narrative of 250.4 doesn't really matter.

But 250.118 may not provide a large enough conductor if the circuit is long.

Ryan has mentioned he would have no problem with citing a violation of 250.4(A)(5) in a situation like that.

That is where he and I disagree.

BTW No one should take this as 'beating up' on Ryan, I have much respect for him and the job he is trying to do.:)
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
iwire said:
But 250.118 may not provide a large enough conductor if the circuit is long.
Wouldn't the lack of voltage at the point of use kinda self-prohibit that?

Ryan has mentioned he would have no problem with citing a violation of 250.4(A)(5) in a situation like that.

That is where he and I disagree.
I would tend to agree with your side in such a discussion (albeit without having heard the opposing view).

I think the frequent references to this section in educational materials diminishes the sections of code that more effectively (and directly) deal with a potential violation.

BTW No one should take this as 'beating up' on Ryan, I have much respect for him and the job he is trying to do.:)
Always feels a bit weird talking about an old argument when the other guy ain't around, doesn't it? I always kinda feel weird until they chime in to pick up the old argument anew. With your disclaimer out of the way, I'd say we can dispense with worrying about Ryan's toes, I'm sure he's got his boots on. :)

Got a link to the old discussion?
 

ryan_618

Senior Member
Bob, I know you aren't beating me up, so don't worry about me.

Now, what would happen if you had a 1,000' length of 1/2" emt for a 30 amp circuit, with the EMT being the equipment grounding conductor? If you were an inspector, would you buy it?
 

iwire

Moderator
Staff member
Location
Massachusetts
ryan_618 said:
if you had a 1,000' length of 1/2" emt for a 30 amp circuit, with the EMT being the equipment grounding conductor? If you were an inspector, would you buy it?

I might have to, can I prove that the breaker will never open?
 

ryan_618

Senior Member
Welll....according to the program that I use, it would pull 36 amps during a ground fault. Assuming a typical inverse time breaker, that would take about five minutes to trip. Does that satsify 250.4(A)(3) and 250.4(A)(5)?
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
Ryan, don't you think 250.118(4) should have the same wording as 250.118(8) which via 320.108 takes us to 250.4 if the NEC wanted/intended to make a cut off distance enforcible?

At what length could AC for this same circuit be tagged?

Roger
 
Ryan
I tend to agree with what you have to say.

1. 250.4(A)(5) - you need to read the paragraph before (A) - which states that the subsection (A) is PERFORMANCE guided, not PRESCRIPTIVE which is what most of the code is related to.

Going beyond 250.118, I think that we need to take a quick look at Table 250.122, the note after the table. Most people tend to not see them or think they are code, but they are. And the title of Table 250.122 state MINIMUM. Take these two items together, and I believe it beefs up Ryan's part of the discussion. We all know that long runs for feeders or branch circuits are a problem for facilitating overcurrent devices in the case of a groundfault, as the amount of fault current along the "Effective Ground Fault Current Path" may not be too effective ;)

I believe this correlates very well with Winnie's first question as to what facilitates the overcurrent device.

As has been mentioned already, we still need an effective ground fault current path regardless of the GFCI function.
 

dereckbc

Moderator
Staff member
Location
Plano, TX
I know of a very simple solution to the problem, that will never be adopted. Limit voltage drop to %5 and EGC must induce 10X the rated OCPD.
 

ryan_618

Senior Member
I tend to agree with Dereck on this. I also don't think it would ever be a part of the code....HOWEVER...I think the voltage drop limiations in 647 is for this reason, particularly when dealing with a 60/120V system, such as those in 647.

Roger: To be quite blunt, I don't know. :(
 

bob

Senior Member
Location
Alabama
iwire said:
But 250.118 may not provide a large enough conductor if the circuit is long.Ryan has mentioned he would have no problem with citing a violation of 250.4(A)(5) in a situation like that.
That is where he and I disagree.
Bob
Why would you disagree? Is it because the NEC does not provide clear
instructions to solve the problem?
I think this is a design issue. Many opinions use 1 cycle. That will require about 4 x OC rating.

georgestolz said:
Originally Posted by iwire
But 250.118 may not provide a large enough conductor if the circuit is long.
Wouldn't the lack of voltage at the point of use kinda self-prohibit that?
No it would not. Table 250.66 may not provide a large enough conductor to
supply fault current to trip the breaker at or near 1 cycle.

georgestolz said:
Originally Posted by ryan_618
if you had a 1,000' length of 1/2" emt for a 30 amp circuit, with the EMT being the equipment grounding conductor? If you were an inspector, would you buy it?
I might have to, can I prove that the breaker will never open?

The point is not whether it will ever open. Its how fast will it open.
roger said:
At what length could AC for this same circuit be tagged?


Length depends on the size of the conductor and breaker size. 250.4.A(5)
makes reference but does not give a method to determine the size and max length of the EGC.

dereckbc said:
I know of a very simple solution to the problem, that will never be adopted. Limit voltage drop to %5 and EGC must induce 10X the rated OCPD.

Dereck
That would do it for sure. However tests have shown 4 0r 5 times the OC rating will do it.

Someone tell me how to do a spell check.
 
Last edited:

dereckbc

Moderator
Staff member
Location
Plano, TX
Ryan, you are right about 647, but it has more to do with the low voltage nature of the circuit. In my industry we go by developed standards rather than code. Part of our standard is limiting voltage drop to %5 on general purpose/mechanical loads, and %3 on technical loads, all must have an EGC. By doing this, ground fault impedance is never an issue, it is a calculated risk.
 

ryan_618

Senior Member
I really think that Pierre's though are right on. People tend to struggle with this section, like Pierre said, it is a performance based section, with no prescriptive rules. It doesn't make it any less enforcable, however.

If you want to see soemthing quite odd, ICC has a building code that is 100% performance. It tells you what your building must do...how your do it is your business. For example, you must have a "safe" means of egress.
 

iwire

Moderator
Staff member
Location
Massachusetts
Ryan I don't think I am struggling at all. :)

I am going to change all the speed limit signs around my house to 'Don't go too fast' and each inspector..(opps).. I mean cop can decide what is to fast.

That should work well. :D
 
250.4 General Requirements for Grounding and Bonding.
The following general requirements identify what grounding and bonding of electrical systems are required to accomplish. The prescriptive methods contained in Art 250 shall be followed to comply with the performance requirements of THIS SECTION.

(A)(5) is part of the section, which is a performance issue. Using the precriptive requirements of Art 250 helps one to try and comply with 250.4(A)(5).
 

iwire

Moderator
Staff member
Location
Massachusetts
Pierre I am not going to convinced an open ended section can be enforced.

I say I have provided an effective ground fault path, the inspectors says I have not.

Who's right?

Who can tell?

The NEC has not told me the perimeters of the requirement.


JMO, Bob
 
Status
Not open for further replies.
Top