Using metal conduit as EGC

[pv_n00b]

It's an accepted practice in commercial electrical work to use metal conduit as the EGC and not to run a separate conductor. Why is this not done in the PV industry? It's always puzzled me since it would save on conductor and labor.

 

[electrofelon]

It's an accepted practice in commercial electrical work to use metal conduit as the EGC and not to run a separate conductor. Why is this not done in the PV industry? It's always puzzled me since it would save on conductor and labor.

It seemail like in the last 20-30 years there has been a mass movement away from using a raceway as the egc. I dont know why. I am not aware of any evidence of a wire egc being better or safer. Personally I rarely use a wire egc if the raceway meets code as such.

 

[pv_n00b]

I've read a couple of white papers where EGCs were tested and it was found that even in systems that had an EGC conductor over 70% of the fault current was still carried in the parallel path of the raceway. The raceway has lower resistance and is grounded so when the fault current splits it has a preference for the raceway.

This assumes that the raceway is installed correctly and provides a good path for fault current. Can't even remember how many poorly connected EMT conduits I have seen, or how many broken connections in older conduit. In those, it's good to have the EGC conductor.

 

[jaggedben]

Well I think in the PV industry we have a history of a couple different things, and old habits die hard. One of those things is thinking of the green wire from the roof as a GEC, which it sometimes technically was and sometimes wasn't depending on equipment and code cycle. Another is bonding requirements over 250V, which doesn't technically require a wire EGC all the way but does call for bonding jumpers to bonding bushings and such. And there's been the awful 690.47 (D). I have to say also that I don't intuitively trust raintight EMT fittings all that much, and I feel safer with a redundant ground. Relying on the raceway does require a higher standard of workmanship, which... well ... is often an effort to find these days.

 

[electrofelon]

I would likely pull a wire EGC in EMT outdoors. Washington state requires this actually.

 

[SceneryDriver]

I would likely pull a wire EGC in EMT outdoors. Washington state requires this actually.

When I lived in southern OR, a local code amendment required that we pull a wire EGC, and not count on the conduit. Reasoning was that it is a seismically active area and the conduit could come apart during an earthquake, leaving potentially live sections of a run ungrounded. I thought that was good logic, actually.

I almost always pull a wire EGC anyway, as I've seen what age and forklifts can do to conduit.



SceneryDriver

 

[electrofelon]

When I lived in southern OR, a local code amendment required that we pull a wire EGC, and not count on the conduit. Reasoning was that it is a seismically active area and the conduit could come apart during an earthquake, leaving potentially live sections of a run ungrounded. I thought that was good logic, actually.

I almost always pull a wire EGC anyway, as I've seen what age and forklifts can do to conduit.



SceneryDriver

Seems like in a serious earthquake, conduits coming apart would be pretty low on my list of things to worry about.....

Probably the most prevalent grounding myth is the role of dirt, and of course that comes up on the forum here frequently. But I think the importance of bonding is over-emphasized as well. I just don't see equipment bonding for fault clearing to be so critically important as most make it out to be. There are many many things that can make an unsafe electrical system and I don't see non-redundant bonding as being anywhere at the top of the list. Think about it: For improper bonding to result in injury, death, or fire, first there has to be a fault - really not that common. How many times, really, have you seen a fault from a chaffed wire in an appliance, sharp corner in a pull box, etc......? Then we have to have the "inadequate" raceway not clear the fault - that is quite unlikely. I have seen some rather mongrel pipe runs clear a fault just fine. I have never seen or seen evidence of a fault not clearing through a boc connector, coupling, etc...THEN, if those two things happen, we have to have someone actually get shocked or a fire actually start. For the case of someone getting shocked, well the probably hurt a bit, but very unlikely to cause injury. That is just a lot of stars to align.

Why don't people find other NEC minimums to be completely inadequate? I don't see people putting gfi's on circuits that don't require them, using a 2nd redundant GFI on a circuit that requires GFCI, putting things on smaller breakers, avoid concentric KO's, using EMT of MC in Resi, etc. Its just ridiculous and a waste of time money and resources. If you want to play the "use all the resources up before China does" game then by all means, pull that magic wire.

 

[ActionDave]

Seems like in a serious earthquake, conduits coming apart would be pretty low on my list of things to worry about.....

Probably the most prevalent grounding myth is the role of dirt, and of course that comes up on the forum here frequently. But I think the importance of bonding is over-emphasized as well. I just don't see equipment bonding for fault clearing to be so critically important as most make it out to be. There are many many things that can make an unsafe electrical system and I don't see non-redundant bonding as being anywhere at the top of the list. Think about it: For improper bonding to result in injury, death, or fire, first there has to be a fault - really not that common. How many times, really, have you seen a fault from a chaffed wire in an appliance, sharp corner in a pull box, etc......? Then we have to have the "inadequate" raceway not clear the fault - that is quite unlikely. I have seen some rather mongrel pipe runs clear a fault just fine. I have never seen or seen evidence of a fault not clearing through a boc connector, coupling, etc...THEN, if those two things happen, we have to have someone actually get shocked or a fire actually start. For the case of someone getting shocked, well the probably hurt a bit, but very unlikely to cause injury. That is just a lot of stars to align.

Why don't people find other NEC minimums to be completely inadequate? I don't see people putting gfi's on circuits that don't require them, using a 2nd redundant GFI on a circuit that requires GFCI, putting things on smaller breakers, avoid concentric KO's, using EMT of MC in Resi, etc. Its just ridiculous and a waste of time money and resources. If you want to play the "use all the resources up before China does" game then by all means, pull that magic wire.

No doubt. We may not know how this whole electricity thing works but one thing is for sure, we can run a green wire and tag to all kinds of metal things and go home satisfied that the world has been made a safer place by those efforts.

 

[Coffee Cup]

Belt and suspenders

Belt and suspenders

I did some MSHA certifications at a gravel plant a few years ago. MSHA requires all catwalks, motors and conduits be within 1 ohm of ground. We were just supposed to report, but I would have filled volumes of required fixes for those over an ohm or open altogether. I tightened a lot of fittings and locknuts. The code now requires an EGC in a high vibration areas, but this made a believer out of me, I don't rely on the raceway for ground, you loosen a fitting to adjust and forget to tighten it will still probably work, but if it slips out at some point. The real reason I do it is because I love using bonded bushings, I mean how fun is that

 

[electrofelon]

I dont buy the loose coupling or box connector or forklift argument for as a reason to install a wire EGC at all. Think about it: Even with a wire EGC, the conduit will be the only fault path for a substantial portion of that conduit system. We could have a several hundred foot run with some pull boxes along the way. We all know and use the exception that doesnt require us to bond the box with the EGC if there are no splices. Any fault on those boxes or raceway will have to travel through them. You need to have good tight connections regardless of if there is a wire EGC or not.

 

[ggunn]

I dont buy the loose coupling or box connector or forklift argument for as a reason to install a wire EGC at all. Think about it: Even with a wire EGC, the conduit will be the only fault path for a substantial portion of that conduit system. We could have a several hundred foot run with some pull boxes along the way. We all know and use the exception that doesnt require us to bond the box with the EGC if there are no splices. Any fault on those boxes or raceway will have to travel through them. You need to have good tight connections regardless of if there is a wire EGC or not.

I guess I'm a belt and suspenders guy; I put a ground wire in every conduit run irrespective of the conduit type.

 

[ActionDave]

I guess I'm a belt and suspenders guy; I put a ground wire in every conduit run irrespective of the conduit type.

That's you perrogative. Almost all the time I'm running conduit that is not underground I'm in a steel structure with steel conduit tied to steel framing members with steel fasteners and the steel is bonded to the electrical system so I'm already set up with belt, suspenders, and a spare underwear and really don't need a silly green wire to make me feel better.

 

[petersonra]

an unnecessary green wire is usually cheaper than the time spent arguing over it.

 

[Strathead]

I dont buy the loose coupling or box connector or forklift argument for as a reason to install a wire EGC at all. Think about it: Even with a wire EGC, the conduit will be the only fault path for a substantial portion of that conduit system. We could have a several hundred foot run with some pull boxes along the way. We all know and use the exception that doesnt require us to bond the box with the EGC if there are no splices. Any fault on those boxes or raceway will have to travel through them. You need to have good tight connections regardless of if there is a wire EGC or not.

Your statement doesn't make sense. If we install ground wire in all of our conduits then there is a wire path back to the panel. If a coupling gets loose, or a forklift or earthquake jars something as soon as it is jarred, it is a code violation as all fittings are no longer made up tight. There is no doubt that a properly installed conduit system doesn't need a ground wire, and the code doesn't require one, however you inspector is also fully with in his realm to tell you to get out a screw driver and a lift and get up there and show that every single coupling is made up tight even if it takes two days. And if he finds one loose, then he can stop the inspection and require corrective action prior to passing. I am definitely in the ggunn school of a ground in every conduit. .

 

[petersonra]

Your statement doesn't make sense. If we install ground wire in all of our conduits then there is a wire path back to the panel. If a coupling gets loose, or a forklift or earthquake jars something as soon as it is jarred, it is a code violation as all fittings are no longer made up tight. There is no doubt that a properly installed conduit system doesn't need a ground wire, and the code doesn't require one, however you inspector is also fully with in his realm to tell you to get out a screw driver and a lift and get up there and show that every single coupling is made up tight even if it takes two days. And if he finds one loose, then he can stop the inspection and require corrective action prior to passing. I am definitely in the ggunn school of a ground in every conduit. .

whether there is a green wire or not the conduit fittings still have to be made up correctly.

 

[ActionDave]

Your statement doesn't make sense. If we install ground wire in all of our conduits then there is a wire path back to the panel. If a coupling gets loose, or a forklift or earthquake jars something as soon as it is jarred, it is a code violation as all fittings are no longer made up tight. There is no doubt that a properly installed conduit system doesn't need a ground wire, and the code doesn't require one, however you inspector is also fully with in his realm to tell you to get out a screw driver and a lift and get up there and show that every single coupling is made up tight even if it takes two days. And if he finds one loose, then he can stop the inspection and require corrective action prior to passing. I am definitely in the ggunn school of a ground in every conduit. .

The green to everything crowd are some of the worst IMO. They have this attitude that conduit doesn't have to be made up right because Hey, it's got the green wire, don't worry about it.

 

[electrofelon]

whether there is a green wire or not the conduit fittings still have to be made up correctly.

Right, that is what I was trying to say. Pulling a wire egc because you are scared of loose or seperate fittings, well the egc isn't going to do anything to make up for it in many cases.

 

[roger]

In the majority of commercial buildings the conduit would perform as an EGC regardless of whether it has come apart in one or numerous places, this is due to all the metal supports, framing, steel, other piping systems, etc... it is in contact with.

NFPA 99 recognizes this and considers the metallic raceway the primary EGC and the wire a backup, hence the reason for 517.13(A)

Roger

 

[jaggedben]

In the majority of commercial buildings the conduit would perform as an EGC regardless of whether it has come apart in one or numerous places, this is due to all the metal supports, framing, steel, other piping systems, etc... it is in contact with.

...

With respect to the OP question, that doesn't apply particularly well to PV systems with conduit on the roof.

Although I don't have any data to back it up, I think a reasonable assertion can be made that with outdoor conduit runs there is a justification for the additional insurance of a green wire going to the solar array. Exposure to weather and critters, etc.

 

[petersonra]

With respect to the OP question, that doesn't apply particularly well to PV systems with conduit on the roof.

Although I don't have any data to back it up, I think a reasonable assertion can be made that with outdoor conduit runs there is a justification for the additional insurance of a green wire going to the solar array. Exposure to weather and critters, etc.

you really think the green wire is less likely to be damaged by critters and weather than EMT?