Grounding a 4 square metal box if not spliced in.

[JdoubleU]

I once heard that you don't have to ground a metal box with the equipment grounding conductor if it is only used as a raceway, Meaning there are no splices in it. Is this true? 250.80 I think says you can not do this unless it is a short run, but they don't specify what a short run is.

 

[infinity]

All metal boxes are required to be grounded. If the raceway is metallic and qualifies as an EGC then you can skip connecting a wire type EGC within the raceway to the box if the ungrounded and grounded conductors are not spliced. Are you referring to service raceways? Look at 250.148.

 

[don_resqcapt19]

The metal box needs to be connected to an EGC.

250.4(A)(3) Bonding of Electrical Equipment. Normally 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 groundfault current path.

 

[jap]

I once heard that you don't have to ground a metal box with the equipment grounding conductor if it is only used as a raceway, Meaning there are no splices in it. Is this true? 250.80 I think says you can not do this unless it is a short run, but they don't specify what a short run is.

I was leaning on a metal water pipe in a chicken house one day about to work on a 4 square junction box mounted on a wooden backboard.

The junction box was fed with romex and the ground wire was cut off.

The hot had rubbed into the side of the metal 4 square box.
As soon as I reached out touched this metal box, every aspect of poor bonding practices went through me.

Think of it that way.

The metal box needs a pathway back to the source either by metal conduit or a grounding conductor to quickly facilitate the overcurrent protection device should a ground fault occur.
If the grounded and ungrounded conductors are spliced in the box, so should the grounding conductor and it should be bonded to the box.

Not sure why this is, other than the fact that your removing insulation from a current carrying conductor in the Junction box whereas before it would have merely passed through the box undisturbed.

Maybe someone else can chime in on why the ground wire must be spliced and bonded when the grounded and ungrounded conductors are spliced. I dont fully understand this myself.

 

[Nitel]

250.118

 

[infinity]

Maybe someone else can chime in on why the ground wire must be spliced and bonded when the grounded and ungrounded conductors are spliced. I dont fully understand this myself.

The box is already grounded by it's attachment to a metal raceway that qualifies as an EGC. Since the box is merely used as a pulling point for the unspliced, pass-through conductors a bonding jumper from the EGC to the box is not required.

 

[JdoubleU]

Thank you

 

[jap]

The box is already grounded by it's attachment to a metal raceway that qualifies as an EGC. Since the box is merely used as a pulling point for the unspliced, pass-through conductors a bonding jumper from the EGC to the box is not required.

So why is the grounding conductor required spliced and bonded to the box only if the others are spliced?

 

[Smart $]

So why is the grounding conductor required spliced and bonded to the box only if the others are spliced?

When the others are spliced in the box, there's a greater chance of a splice failure and the box becoming energized.


There can be no splice failure if there are none...

 

[infinity]

Although I agree that Smart's answer is probably the best one you will find it's somewhat bogus. There could be splices and no wire EGC and it would still be compliant. I guess the NEC wants to give you an added margin of safety.

 

[jap]

Although I agree that Smart's answer is probably the best one you will find it's somewhat bogus. There could be splices and no wire EGC and it would still be compliant. I guess the NEC wants to give you an added margin of safety.

I agree its bogus.
Maybe they're looking out for us so in case we need to add something to the other splices in the future, the ground wire would also have slack in it because we spliced it instead of pulling it straight through the box without a loop.
Just another bogus thought.

I see no advantage of splicing the grounding conductor and bonding it to the enclosure only when the other conductors are spliced.

 

[Smart $]

...I see no advantage of splicing the grounding conductor and bonding it to the enclosure only when the other conductors are spliced.

Perhaps you should look at it from the perspective of not having to bond it when the other conductors are pulled through without splice. Otherwise, the alternative is splice and bond all the time...???

 

[jusme123]

I was leaning on a metal water pipe in a chicken house one day about to work on a 4 square junction box mounted on a wooden backboard.

Maybe someone else can chime in on why the ground wire must be spliced and bonded when the grounded and ungrounded conductors are spliced. I dont fully understand this myself.

If you allow the ground wire to be pulled strait through when other conductors are spliced, how are you going to connect any additional ground wires? That's why they require the ground wire to be spliced and bonded when the hot and neutrals are spliced within a box.

 

[jap]

If you allow the ground wire to be pulled strait through when other conductors are spliced, how are you going to connect any additional ground wires? That's why they require the ground wire to be spliced and bonded when the hot and neutrals are spliced within a box.

Thats not correct.
That'd be like the code telling me how to bend my pipe.
Thats an installation issue.

 

[jap]

If you allow the ground wire to be pulled strait through when other conductors are spliced, how are you going to connect any additional ground wires? That's why they require the ground wire to be spliced and bonded when the hot and neutrals are spliced within a box.

why would I need to?

 

[jap]

Perhaps you should look at it from the perspective of not having to bond it when the other conductors are pulled through without splice. Otherwise, the alternative is splice and bond all the time...???

Oddly enough actually I'd be ok with that.

 

[dicklaxt]

Something is bothering me in this post.If you have to bond the EGC to the box then in essence you are creating parallel fault current paths of unequal impedance.When more than one path exists then the clearing of that faulted circuit is slower, is NEC condoning that? This is an intentional path when a fault occurs,how come I stay confused in my old age.I didn't ask all these questions when I was 185 pounds of romping,stomping romance.

dick

 

[jap]

I look at it as reinforcing the return path that you do have.
A faults going to travel on everything regardless.

 

[cowboyjwc]

Something is bothering me in this post.If you have to bond the EGC to the box then in essence you are creating parallel fault current paths of unequal impedance.When more than one path exists then the clearing of that faulted circuit is slower, is NEC condoning that? This is an intentional path when a fault occurs,how come I stay confused in my old age.I didn't ask all these questions when I was 185 pounds of romping,stomping romance.

dick

Because when you were young and good looking you just did it, now that you've slowed down a little and look at it more you start to poner.:happyyes::happyno: (those two together means I have no idea if that's true or not.)

I guess it's no differnt than why do you bond all of the big metal beams and posts in a building, but not the steel wall studs? Which do you think is more likely to become energized? Now people will say that they are bonded, because the box is bonded and attached to the stud and that may be true.

 

[Smart $]

Something is bothering me in this post.If you have to bond the EGC to the box then in essence you are creating parallel fault current paths of unequal impedance.When more than one path exists then the clearing of that faulted circuit is slower, is NEC condoning that? This is an intentional path when a fault occurs,how come I stay confused in my old age.I didn't ask all these questions when I was 185 pounds of romping,stomping romance.

dick

How so? Parallel paths lower the total effective impedance. I could understand if you routed the grounding path(s) completely away from the circuit conductors.

As for the rest...