EGCs

[DBoone]

Since all EGCs are bonded together at the main panel then under normal operation all EGCs and everything they are attached to will be at the same potential. Meaning, you wouldn't get a difference in potential between two appliance frames or between two switch yokes or between two light fixtures, etc. Is this correct?

 

[iwire]

Since all EGCs are bonded together at the main panel then under normal operation all EGCs and everything they are attached to will be at the same potential. Meaning, you wouldn't get a difference in potential between two appliance frames or between two switch yokes or between two light fixtures, etc. Is this correct?

That is true during normal operation, not true during a ground-fault condition.

During a ground fault there will be differences of potential due to voltage drop on the EGC.

 

[kwired]

An EGC that inadvertently is bonded to a "neutral" conductor somehow beyond the service equipment or first disconnecting means of separately derived system also will have elevated voltage due to voltage drop when current is flowing. The whole concept of how EGC is supposed to be installed is to not have any current during normal operation, and current during a fault condition should be limited in duration

 

[DBoone]

Thanks for the replies guys. Here's the reason I asked:

When making up multi-gang switch boxes (plastic), with multiple circuits spliced within the box, I have been keeping the EGCs of each circuit separate and attached to their respective switch yokes.

I wanted to make sure there couldn't be a difference of potential between the yokes (face plate screws). I didnt think there would be any potential during normal operation because there would be no current on the EGCs but I just wanted to check with the pros.

Im aware of 250.148 and I guess that makes me a code violator but my thought has always been the circuits don't know they are sharing a box and if those switches were in single gang boxes side by side, the EGCs couldn't be spliced together and it would be acceptable.

Another thing is I Find it slightly harder to splice so many EGCs. I prefer not to make two splice groups and then bond them with a jumper. What are some of your methods for splicing large groups of EGCs? I like to leave one long pigtail to bond all my switch yokes with.

Thanks for the time. Help me learn. I want to learn new and better techniques!

 

[iwire]

Thanks for the time. Help me learn. I want to learn new and better techniques!

I think a better technique would be connecting all EGCs together every chance you get. That reduces impedance during a fault causing the breaker to trip quicker. It is also my belief that 250.148 requires you too.

Yes, in multi-gang boxes the number of EGCs can be a headache to join but it's not impossible.

Make the groups you mentioned, or get connectors that are listed for more conductors, try leaving one long bare EGC to 'loop' through all the device ground terminals etc.

 

[roger]

Unlike neutral current, the more paths available for fault current to get back to the source the better off you are.

Roger

 

[DBoone]

I think a better technique would be connecting all EGCs together every chance you get. That reduces impedance during a fault causing the breaker to trip quicker. It is also my belief that 250.148 requires you too.

Yes, in multi-gang boxes the number of EGCs can be a headache to join but it's not impossible.

Make the groups you mentioned, or get connectors that are listed for more conductors, try leaving one long bare EGC to 'loop' through all the device ground terminals etc.

Im glad you mentioned leaving a long EGC and looping through all devices because that's what I do and so I'm glad someone like yourself considers it an acceptable practice.

So to sum it up, the way I have been making up my switches boxes thus far, keeping EGCs of each circuit separate, poses no hazard as far as switch yokes having different potentials, but by splicing EGCs of multiple circuits together whenever possible (junction boxes, switch boxes, etc) you create multiple paths back to the panel thus lowering resistance, tripping breaker faster, and if one circuit lost its EGC between the panel and the splice, there would still be a path for the fault current and trip the breaker.

Also most folks believe 250.148 requires all EGCs to be spliced together I have seen.

All around it's just a good practice to splice them all correct?

Thanks for the time on something so basic :thumbsup:

 

[roger]

All around it's just a good practice to splice them all correct?

Correct

Roger

 

[Dennis Alwon]

Another method is to use stakons . Leave one wire extra extra long on stakon the grounds at different points along the wire starting at the bottom first , of course , and working up to the top then you can either loop to each or use a large wirenut and go individually to each device.

Here is what I am talking about

 

[DBoone]

Another method is to use stakons . Leave one wire extra extra long on stakon the grounds at different points along the wire starting at the bottom first , of course , and working up to the top then you can either loop to each or use a large wirenut and go individually to each device.

Here is what I am talking about

Dennis forgive my ignorance but I couldn't quite follow. Would you mind explaining again?

And to everyone else, share your techniques for splicing large groups of EGCs.

I bought a bag of big blue Ideals for large splices. Anybody know off the top of your head how many 12s a big blue is rated for?

I will check back in later today. We are really busy bringing a basement up out of the ground... In Mississippi heat and humidity. :slaphead:

 

[iwire]

I bought a bag of big blue Ideals for large splices. Anybody know off the top of your head how many 12s a big blue is rated for?

Sadly not nearly as many as they will take. Six 14s or 12s.

 

[DBoone]

P.S. - Dennis, do you splice all grounds when multiple circuits are in a box?

 

[Dennis Alwon]

What I am saying is to leave on wire very long then take say 3 other conductors and twist them on near where the wire exits the box then put a stakon on it. Do the same just above that splice and again till they are all splice and the tail either long enough to do all the devices or use a red wirenut and splice 4 pigtails together for the device.

Not sure how many the big blue wirenuts can use but the reds can handle 5 #12 wires so if you are using #14 you can get a good number of them on it-- check the box.

BTW, you can also use the big blues and jumper the wires with another piece and have 2 equipment grounding conductor splices with 2 blue wirenuts. Personal I find the blues take up a lot of space especially with dimmers and decora style switches.

 

[Dennis Alwon]

P.S. - Dennis, do you splice all grounds when multiple circuits are in a box?

Yes, I tie all the equipment grounding conductor's together if there is more than one circuit in the box

 

[iwire]

Yes, I tie all the equipment grounding conductor's together if there is more than one circuit in the box

As do I.

A poll to go along with this topic.

http://forums.mikeholt.com/showthread.php?t=162941

 

[suemarkp]

If using metal boxes, or you have more than say 8 EGCs to connect, use a small 4 or 5 hole grounding bar. The CH ones use a 10-32 screw which matches the threads tapped in metal boxes. Each hole is rated for 3 EGCs if the same gauge and material.

I also have some 8 hole push-in splice blocks. I know they are listed for regular conductors, but don't know about grounding conductors.

 

[don_resqcapt19]

...
I also have some 8 hole push-in splice blocks. I know they are listed for regular conductors, but don't know about grounding conductors.

250.8(A)(1) says those connectors are suitable for the EGC connection.

 

[DBoone]

Thanks for all the feedback guys. I was taught by an old schooler to just twist the grounds together in a switch box and push them to the back and not even ground the switches. Later on as I began to study electrical And seek more knowledge I started grounding my switches, just like I was taught to twist my grounds in a receptacle box and cut one off, and land the longer one on the receptacle, no wire nut or connector. Now I leave all EGCs normal length and splice a pigtail to the receptacle and wire nut my splice. Catch grief from my boss on both accounts because he thinks it's unnecessary.

I haven't looked at the poll that was posted yet but I know from Internet research that 250.148 requiring all grounds to be spliced isn't a unanimous conclusion. I also don't think keeping each circuit separate creates anymore risk than the many other circuits that only have one path back to the panel. Also to be a hazard it would require two malfunctions (have a ground fault and lose your EGC path back to the panel).

Something about having two bundles of grounds with a jumper between them just worries me, bothers me, and it shouldn't because they are electrically connected just as if they were in one bundle. And I don't get to buy the materials so I would have to jumper between wire nuts.

Hopefully after this discussion I will be settled on a method and not lose sleep over it.

 

[iwire]

I haven't looked at the poll that was posted yet but I know from Internet research that 250.148 requiring all grounds to be spliced isn't a unanimous conclusion. I also don't think keeping each circuit separate creates anymore risk than the many other circuits that only have one path back to the panel. Also to be a hazard it would require two malfunctions (have a ground fault and lose your EGC path back to the panel).

Even with just one malfunction more EGC connections back to the source decreases impedance which cause the breaker to trip quicker.

But that aside, you really have no choice in it at all when you have metal boxes. All EGCs will end up connected either directly or through the box.

 

[Little Bill]

Another method is to use stakons . Leave one wire extra extra long on stakon the grounds at different points along the wire starting at the bottom first , of course , and working up to the top then you can either loop to each or use a large wirenut and go individually to each device.

Here is what I am talking about

I've never heard those referred to as "sta-kons". I call those crimp sleeves or "Buchanan" crimps. I call the forked, ring, or spade connectors Sta-kons. Just different terminology for different areas I guess.