250.122(F)

[Dale001289]

Article 250.122(F), 2014 NEC: Let’s say you have 4 parallel circuits, consisting of single conductors in triangular configuration installed in cable tray. NEC says only one EGC is needed suffice for all four sets. I say this is inadequate and defer to the same scenario using 4 parallel circuits consisting of multiconductor cables in which case an individual EGC would be included. My rationale uses the same laws of physics governing magnetic flux, impedance and fault path, i.e. EGC must be in very close proximity to phase conductors in order to be effective. I believe this article needs to be revised to include a separate EGC for each set of parallel single conductor bundles. Only problem is, I don't have the capability to produce the math in order to prove my point.

 

[iwire]

Only problem is, I don't have the capability to produce the math in order to prove my point.

Than I respectfully suggest you do not request a code change.

 

[Dale001289]

Than I respectfully suggest you do not request a code change.

My hope is someone who is damned smart will take up my cause and produce the math.

 

[al hildenbrand]

My hope is someone who is damned smart will take up my cause and produce the math.

Not such a simple problem, eh? Multiple parallel EGC paths, of exotic shape and characteristic, physically spread out in three dimensional space relative to the parallel feeder conductors . . .

I'm not disputing 'safety'; I'm disputing functionality, effectiveness and efficiency of fault path. Once I've figured out a viable way to prove it (either through math/hand calc's or Etap) I'll let you know the result. But...I wont publish the numbers; someone on this thread would publish them and get all the credit

 

[jumper]

My hope is someone who is damned smart will take up my cause and produce the math.

You are asking if someone else here will do your homework for you? Prolly not gonna happen.

 

[charlie b]

This is not a situation in which math will provide any answers. Let's start here:

. . . EGC must be in very close proximity to phase conductors in order to be effective.

That is not true, and it leads me to believe that you have a misconception as to the purpose and function of an EGC. Please allow me to explain what I mean, and forgive me if I am not giving you enough credit for your knowledge and experience.

As far as the physics is concerned, I can run three ungrounded conductors from a panel located on the north wall of a room to a motor located on the south wall of the same room. I can route these conductors in a conduit along the east wall. I can then run the EGC in a separate conduit along the west wall.

EVERYONE PLEASE NOTE: I know this is a code violation, but I am talking about physics here.

Now suppose one of the ungrounded conductors breaks free of its termination point inside the motor and comes into contact with the motor's outer case. This will cause a high current to flow from the ungrounded conductor, to the case, to the EGC, and via the EGC back to the panel. This will cause the breaker to trip and thereby terminate the event. That is, indeed, the EGC's job. The fact that the EGC was nowhere near the phase conductors will not stop it from doing that job.

As far as multiple circuits in the same tray sharing the same EGC, the description I just gave applies here as well. The single EGC will allow the fault current to flow back to the panel, resulting in a trip of whichever circuit breaker serves the load that experienced the fault. I do not see a need for a code change.

 

[iwire]

My hope is someone who is damned smart will take up my cause and produce the math.

Do you have information about real life damage this issue has caused?

In my personal opinion without real world incidents there is no reason to add a code restriction.

To me you are putting the cart before the horse. Trying to produce a mathematical result to support a gut feeling.

 

[jumper]

Do you have information about real life damage this issue has caused?

In my personal opinion without real world incidents there is no reason to add a code restriction.

To me you are putting the cart before the horse. Trying to produce a mathematical result to support a gut feeling.

:thumbsup:

 

[Dale001289]

This is not a situation in which math will provide any answers. Let's start here:That is not true, and it leads me to believe that you have a misconception as to the purpose and function of an EGC. Please allow me to explain what I mean, and forgive me if I am not giving you enough credit for your knowledge and experience.

As far as the physics is concerned, I can run three ungrounded conductors from a panel located on the north wall of a room to a motor located on the south wall of the same room. I can route these conductors in a conduit along the east wall. I can then run the EGC in a separate conduit along the west wall.

EVERYONE PLEASE NOTE: I know this is a code violation, but I am talking about physics here.

Now suppose one of the ungrounded conductors breaks free of its termination point inside the motor and comes into contact with the motor's outer case. This will cause a high current to flow from the ungrounded conductor, to the case, to the EGC, and via the EGC back to the panel. This will cause the breaker to trip and thereby terminate the event. That is, indeed, the EGC's job. The fact that the EGC was nowhere near the phase conductors will not stop it from doing that job.

As far as multiple circuits in the same tray sharing the same EGC, the description I just gave applies here as well. The single EGC will allow the fault current to flow back to the panel, resulting in a trip of whichever circuit breaker serves the load that experienced the fault. I do not see a need for a code change.

"The fact that the EGC was nowhere near the phase conductors will not stop it from doing that job"

I have to agree/disagree with you at the same time. How does a 50/51 interact with a GE Multilin relay on ground fault? Or how does a neutral conductor interact on a perfectly balanced three phase circuit? See what I mean? Its all about time and effectiveness.
Consider this: Are you familiar with the Terra-Terra 'TT' system (IEC)? Where a protective earth connection is provided locally, usually a ground rod, along with another installed at the generator or transformer. There is no 'earth wire' between the two. This would be a NEC violation - 250.4(A)(5) reason: it is horribly ineffective.
Grounding is always highly subjective but if you stick to the fundamental physics of fault path, you'll see through the mist and fog.

 

[don_resqcapt19]

... The fact that the EGC was nowhere near the phase conductors will not stop it from doing that job. ....

But the impedance of that remote fault clearing path will be substantially higher than if the EGC was in the same conduit. Maybe high enough that the fault current will be limited enough that the breaker will operate in its short time trip range and not in the instantaneous trip range.
As I recall from the IEEE green book, a 3' separation between the ungrounded conductors and the EGC can double the impedance of the fault return path.

 

[Dale001289]

But the impedance of that remote fault clearing path will be substantially higher than if the EGC was in the same conduit. Maybe high enough that the fault current will be limited enough that the breaker will operate in its short time trip range and not in the instantaneous trip range.
As I recall from the IEEE green book, a 3' separation between the ungrounded conductors and the EGC can double the impedance of the fault return path.

Correct - when the EGC is routed with phase conductors, it creates a low impedance path due to magnetic flux of the surrounding phases - much like a neutral on a balanced system.

 

[roger]

Just install the tray continuous from point to point and use the tray as the EGC, you're done.

Roger

 

[Dale001289]

Just install the tray continuous from point to point and use the tray as the EGC, you're done.

Roger

Not so fast Roger - we're not discussing what the code says but rather what it should say...at least according to me

 

[jumper]

Not so fast Roger - we're not discussing what the code says but rather what it should say...at least according to me

Data, test studies with peer review, and published articles from respected sources are your back up or is this just a silly fishing expedition?

Yeah, I am bit harsh.

 

[ActionDave]

Didn't we have this argument earlier this year?

 

[jumper]

Didn't we have this argument earlier this year?

Yes, AFAICT the consensus was that Mcdonald's makes the best fries and that Carl's Jr. has a better burger.

I maybe confusing forums though.

 

[mwm1752]

It would seem the EGC is as protective as the grounded conductor as all EGC eventually tie to the grounded conductor -- when installing multi circuits in a raceway we only are required to size the EGC according to the largest OCPD. Adding EGC for each branch circuit would not create a better path.. GEC are a different subject matter when discussing purpose.

 

[Chamuit]

Yes, AFAICT the consensus was that Mcdonald's makes the best fries and that Carl's Jr. has a better burger.

I maybe confusing forums though.

Now wait a minute, are we talking french, wedge or waffle fries?

 

[al hildenbrand]

when the EGC is routed with phase conductors, it creates a low impedance path due to magnetic flux of the surrounding phases - much like a neutral on a balanced system.

You seem very hypnotized by the repetition of this meme. The way you state it implies that a low impedance path IS NOT present in any other configuration . . . which is just plain silly.

I support Bob's suggestion to actually work from real world situations that fail. Do a forensic analysis of the assembly, model it mathematically and demonstrate your case.

 

[GoldDigger]

You seem very hypnotized by the repetition of this meme. The way you state it implies that a low impedance path IS NOT present in any other configuration . . . which is just plain silly.

I support Bob's suggestion to actually work from real world situations that fail. Do a forensic analysis of the assembly, model it mathematically and demonstrate your case.

:thumbsup:

An EGC run separately from the line may, for example, decrease the current from a bolted fault from 25kA to 10kA, but is not likely to affect the trip time of OCPD.