Ground wires from separate circuits in a junction box.

[Martinelli]

We just had a rough inspection for a small kitchen remodel. The inspector failed us because all the grounds were not tied together in a 2 gang switch/ receptacle box. The box is plastic and contains a lighting circuit and a small appliance branch circuit. In this instance I kept the equipment grounds from the 2 circuits separate. I usually tie them together but this time I thought it was cleaner to separate them. both circuits are properly grounded at the sub panel. The inspector said that for the past 30 years the code has required that all equipment grounds be tied together regardless of what circuit they originate from. He could not give me the code reference. I could not find anything in the 2008 NEC, can someone please point me in the right direction.

 

[iwire]

It is disturbing the inspector cannot provide a code reference for his request.

The following code section may or may not require what the inspector is requesting, it depends on who you ask.

250.148 Continuity and Attachment of Equipment Grounding Conductors to Boxes. Where circuit conductors are spliced within a box, or terminated on equipment within or supported by a box, any equipment grounding conductor(s) associated with those circuit conductors shall be connected within the box or to the box with devices suitable for the use in accordance with 250.148(A) through (E).

Exception: The equipment grounding conductor permitted in 250.146(D) shall not be required to be connected to the other equipment grounding conductors or to the box.

 

[infinity]

The following code section may or may not require what the inspector is requesting, it depends on who you ask.

If you ask me it's not required based on what 250.148 says. If that's the true intention of that section then it should say so directly, not in mumbo-jumbo form.

 

[mcclary's electrical]

If you ask me it's not required based on what 250.148 says. If that's the true intention of that section then it should say so directly, not in mumbo-jumbo form.

I agree,,,,,,,the words"associated with those conductors" IMO allow his two circuits to remain seperate.

The 15 amp circuit is not "associated with the conductors" of the 20 amp circuit. I always ire in the same fashion as the op when two circuits are in the same box.

 

[iwire]

I always ire in the same fashion as the op when two circuits are in the same box.

Why? ///////////////

 

[Dennis Alwon]

Interesting, I always tpook 250.148 to mean they all had to be tied together. I believe it is interpreted, as I said, around here.

 

[CFL]

I have left them separated before to save space. For example, above a kitchen counter, in a 4x4 box that is going to house a gfci rec. and a disposal switch. That is probably the only place I've done it and really wasn't necessary.

 

[mcclary's electrical]

Why? ///////////////

If I have a 20 amp disposal ciruit, and a 15 amp lighting circuit, with several legs, I keep the 14 gauge seperate from the 12. The are effectively bonded together on the "feed end" (service panel), there is no reason to group them together again. If I have a fault in the 20 amp circuit, there's no need in energizing the EGC's of the 15 amp circuit along with it. It's an unneeded ground loop. The main reason I seperate is so there is no need have this stiff ball of EGS's when each one can land on it's prospective switch without the other one being attached to it.

If it's all the same circuit, of course they all get joined.

I think it's actually safer not to mingle the two circuits together, IMO, the 20 amp breaker will react quicker taking a single path back to the panel rather than multiple paths. But I do it because of stiffness

 

[brian john]

. It's an unneeded ground loop.

I would not consider this a ground loop a parallel path yeah.

I think it's actually safer not to mingle the two circuits together, IMO, the 20 amp breaker will react quicker taking a single path back to the panel rather than multiple paths. But I do it because of stiffness

A lower impedance path to the NGB wouold be a problem?

 

[mcclary's electrical]

I would not consider this a ground loop a parallel path yeah.




A lower impedance path to the NGB would be a problem?

No problem,I'm really just making up reasons:grin: My main reason is the stiffness of all the grounds coming together in unnecessary

 

[jsmalone1]

proves my point about the code--tough to reach a consensus, an interpretation, etc. Those grounds should not be tied together--for crying out loud!!!!

 

[iwire]

Those grounds should not be tied together--for crying out loud!!!!

Why?

I am guessing you never work on steel framed buildings where all grounds are always electrically connected.

 

[iwire]

I think it's actually safer not to mingle the two circuits together, IMO, the 20 amp breaker will react quicker taking a single path back to the panel rather than multiple paths.

What happens when you are making these same connections in a metal j box?

 

[RICK NAPIER]

I think this article is clear that the grounds must be bonded together either directly or through bonding with the metal box. This wil not affect the overcurrent protecting the circuit since it is the current carrying conductor and not the ground that is protected from overcurrent so distributing the ampacity over the grounds wil not change the amperage flow through the current carrying conductor. Also there is a risk of arcing between grounding conductors in a fault condition that is eliminated if the grounds are bonded together. By the way I failed two installations yesterday in part for this issue and in NJ it has been a long time requirement.

 

[jsmalone1]

Why?

I am guessing you never work on steel framed buildings where all grounds are always electrically connected.

Better question--why would you? It's grounded! why mix? Isn't it about where they are all connected?

 

[roger]

The more parallel paths we can create for a low immpedance path back to the source, the better off we are.

In NFPA 99 there is commentary that supports this.

The reason for the metallic sheath being required to qualify as an EGC per 517.13(A) is that, it's contact with the metal framing and structural building members create a lower fault clearing path back to source than a single insulated wire conductor, this is the reason the insulated conductor is secondary and the metallic raceway in contact with all the other metal is the primary EGC.


In short, back to my first sentence, parallel paths are better.

Roger

 

[mcclary's electrical]

What happens when you are making these same connections in a metal j box?

If it becoomes more trouble than it's worth, then I just put them all together. But I will say that I've put more than one green screws in plenty of metal boxes.

 

[cmreschke]

you are paralleling your grounds by spliceing them together. although an effective ground path and i am guilty of doing so as well, i do however feel that you are in violation of article 310.4 of the 2005 code. they are smaller than 1/0 and fit none of the exceptions for conductors in parallel. that is just my opinion

 

[iwire]

If it becoomes more trouble than it's worth, then I just put them all together. But I will say that I've put more than one green screws in plenty of metal boxes.

Either way they are electrically connected together at that box.

 

[iwire]

you are paralleling your grounds by spliceing them together. although an effective ground path and i am guilty of doing so as well, i do however feel that you are in violation of article 310.4 of the 2005 code. they are smaller than 1/0 and fit none of the exceptions for conductors in parallel. that is just my opinion

That section does not apply to EGCs.

(A) General. Aluminum, copper-clad aluminum, or copper
conductors of size 1/0 AWG and larger, comprising each
phase, polarity, neutral, or grounded circuit conductor shall
be permitted to be connected in parallel (electrically joined
at both ends).