GROUND SCREWS IN METAL JUNCTION BOXES

[ryan_618]

Re: GROUND SCREWS IN METAL JUNCTION BOXES

I beleive we are over-thinking this.

If we decide to pull an equipment ground, even though it was not required, and we splice or terminate in a metal box, we must bond the metal box with a screw or listed device, to the equipment ground.

The only exception would be for an IG circuit.

 

[russ]

Re: GROUND SCREWS IN METAL JUNCTION BOXES

Ryan:

I agree with that, but I've never really done it or required it to be done. Guess I'll have to adjust.

 

[garrosan]

Re: GROUND SCREWS IN METAL JUNCTION BOXES

Well guys, I think we've about beaten the dead horse enough. And for Ryan and Russ; if you ever have the pleasure of building a WallGreens Drug store, you are not provided the option of whether or not to pull a circuit ground. WallGreens has their own Electrical Inspectors (we here in Greenville, SC refer to them as Nazi's). They require every piece of conduit to have a system ground pulled with it, including those IG circuits. Yep, for the IG's we have to pull a green for the system ground and a green/yellow for the IG ground, though of course the IG ground is not bonded to the juction box.

Onc again to everyone, thanks for the help. I can see this is a place to get straight answers, well mostly straight anyway.

 

[iwire]

Re: GROUND SCREWS IN METAL JUNCTION BOXES

Originally posted by garrosan:
They require every piece of conduit to have a system ground pulled with it, including those IG circuits. Yep, for the IG's we have to pull a green for the system ground and a green/yellow for the IG ground, though of course the IG ground is not bonded to the juction box.

That's shocking can you believe the nerve of them?

 

[russ]

Re: GROUND SCREWS IN METAL JUNCTION BOXES

Onc again to everyone, thanks for the help. I can see this is a place to get straight answers, well mostly straight anyway.

LMAO

 

[ryan_618]

Re: GROUND SCREWS IN METAL JUNCTION BOXES

Originally posted by iwire:
Sorry you just described 75% or more of all the work we do.

Job specs almost always go well beyond code requirements.

When I review plans that don't show an EGC, I require the engineer to show me impedance charecteristics of the raceways being used to ensure 250.4(A)(5) is being complied with. Surprise, surprise, they come back with an EGC in them

 

[shocker3218]

Re: GROUND SCREWS IN METAL JUNCTION BOXES

I pull an equipment grouond in everything I do (2 if I have an IG circuit). The extra peace of mind is worth the small additional cost.

The boss tells me not to, but somehow my jobs always come in under budget.

 

[don_resqcapt19]

Re: GROUND SCREWS IN METAL JUNCTION BOXES

Ryan,

When I review plans that don't show an EGC, I require the engineer to show me impedance charecteristics of the raceways being used to ensure 250.4(A)(5) is being complied with. Surprise, surprise, they come back with an EGC in them

Do you require any calculations for a nonmetallic cable with an EGC? I just run some numbers on the GEMI software from the steel tube institute to calculate the maximum length of a 100 amp, 120 volt circuit with a 700 amp fault. The software said that maximum length of a cable with a #8 EGC is 124', EMT without an EGC is 187' and EMT with a #8 EGC is 231'.
Don

 

[ryan_618]

Re: GROUND SCREWS IN METAL JUNCTION BOXES

Thats interesting Don. I have the same software that you speak of, and I have been crunching numbers on it this morning as well. I have my results to be pretty similiar to your own, which I find very intriguing. So, to answer your question, no I've never asked for the calc's on any non-metallic wiring method.

I alosfind it interesting that there is a note to table 250.122 saying that the EGC may have to increased in sized to satiisfy 250.4(A)(5), and yet doing the numbers, in seems as though they seldom would meet 250.4(A)(5), that is, assuming an effective ground fault current path to create about 4-5 times the amperage of the OCPD rating, which is what I have always learned to be the norm on fault clearing.

Thanks for pointing that out...that certainly surprised me.

 

[George Stolz]

I'm pulling an old thread from the crypt, blowing the dust off it, and breathing new life into it. This came up at work today, and I overheard a discussion and was confused.

Some large feeder pull boxes will not get bonded, if we pull right through.

This was the position of one guy, and the other guy didn't get it.

I overheard the discussion and avoided chiming in (first day and all ).

Isn't not bonding the box a violation of 250.4(A)(3)?

250.4(A)(3) Bonding of Electrical Equipment. Non?current-carrying conductive materials enclosing electrical conductors or equipment, or forming part of such equipment, shall be connected together and to the electrical supply source in a manner that establishes an effective ground-fault current path.

Just because 250.148 doesn't require a non-junctioned EGC to terminate in the box, doesn't seem to relieve the overall requirement to bond the box.

 

[Minuteman]

George - Do you bond LB's or other conduit bodies?

 

[George Stolz]

No.

Does my indiscretion legalize it?

 

[LarryFine]

I'm pulling an old thread from the crypt, blowing the dust off it, and breathing new life into it.

*Cough-cough!*

Isn't not bonding the box a violation of 250.4(A)(3)?

I would think that EMT connectors (or double-locknuts on rigid) and non-concentric/eccentric KO's would satisfy the requirement. You could always bond the conduit ends together inside the box with bonding bushings, etc, if you want.

 

[Minuteman]

I beleive we are over-thinking this.

If we decide to pull an equipment ground, even though it was not required, and we splice or terminate in a metal box, we must bond the metal box with a screw or listed device, to the equipment ground.

George, I believe Ryan put it about as clear as anyone could when he posted this back on 05/09/04

 

[George Stolz]

To clarify, a couple of points:

It could be the conduit system in question yesterday was metallic, and an EGC according to 250.118. I assumed it wasn't, eavesdropping.

So, moving forward with that assumption...

Supposing the raceway was non-metallic, with a metal j-box in the middle, with no splices, would the box need to be bonded?

I would say yes, it does, although not necessarily by the EGC running unspliced through it. But then again, the EGC would have to be sized according to 250.122 by the OCPD of the largest conductors in the box, right?

 

[don_resqcapt19]

George,
In that case the circuit EGC must be used to bond the box.
Don

 

[George Stolz]

Thanks for the sanity check, Don.

 

[Minuteman]

George

Obviously, I'm a little weak on code references, but I am a thinker. Here is how I understand the intent of the code.

The purpose for bonding to metal is for there to be a path for the fault current to go that would trip the OCP device. A nonmetallic conduit EGC must bond to the metallic box for that fault current path that could occur in that box.

Someone will be along shortly to put it in better words.

 

[infinity]

Here's a graphic from Mike Holt:

 

[mostra]

"Infinity" has a great picture for the situation! Imagine the Romex cable being Conduit in a solid run thru metal boxes Etc., The DEVICE still has to be BONDED to the box in any case, even if the code says that EMT is a suitable grounding means. Where else would you get your bonding if the device were pulled away from the box (Metal). Residential is differant from Commercial.