Changing 240v, 20 amp plug can the conduit acts as the ground?

Users who are viewing this thread

Tim Fastle

Member
Messages
59
Reaction score
0
Points
6
Location
New Mexico
We recently purchase a lake house with a detached steel building/shop. I want to install a 20am 240v plug for my compressor. There is already a 240v 20amp circuit the prior owner had installed. I don't know what they ran on it but the plug is a 50A crow-foot, the wire is 8 gauge, there are two hot and a neutral and the breaker is 20amp. It's a steel building with all metal conduit and boxes and no ground wires. The wiring is not that old, I about 10 years, was done by an electrician and inspected. I gather that as long as the box is correctly grounded then the conduit is effectively the ground wire. There is a heavy ground wire that is attached to the neutral bus (actually attached to a single spot bus and that is bolted to the neutral bus).

I have installed 240v 20amp circuits before and always used 3 wires, 2 hot, 1 ground. Where I want to put this one (where the old plug was located) I have 2 hot, 1 common no ground wire but it is part of the "grounded" conduit in the building. It is heavier wire than I need but I assume that is fine. What is the most simple (and, of course, safe) way for me to wire this? Again, I am putting in a 240v 20amp plug.

Any input would be very much appreciated.

Thanks
 

wwhitney

In the Trades
Messages
6,532
Reaction score
1,822
Points
113
Location
Berkeley, CA
Yes, a properly installed metallic conduit system can be your EGC, and evidently that is how your system was installed, since you have no wire-type EGC. A "self-grounding" receptacle will have a contact on the yoke to one of the screws holding the receptacle to the metal box or mudring to provide the EGC connection. Or a pigtail can be run from the receptacle EGC terminal to a ground screw threaded into a metal box; that would be required if the receptacle is not self-grounding.

You may find that the #8 wires will not fit the terminals of your NEMA 6-20 receptacle (I looked up one brand, and it specifies #10 max). In that case you'll need to use some #10 or #12 pigtails to the receptacle, and figure out a way to splice those to the #8s (e.g. small Polaris-style connectors). If you use an EGC pigtail to a ground screw on the box, its size should match the ungrounded conductors going to the receptacle, since this is a 20 amp circuit.

Also, with the large conductors, you need to watch your box fill, but if this was an existing receptacle, that should be fine.

Cheers, Wayne
 

Tim Fastle

Member
Messages
59
Reaction score
0
Points
6
Location
New Mexico
Thanks very much for the replies. I had thought about wire size and had planned to do pigtails inside the box. I have wired a fair number of things myself over the years but never in a grounded metal conduit environment so, of course, always ran ground wires. In this case, I gather I would not use the white neutral wire and just connect the two hot wires and use the conduit as my EGC? Or, should I connect the neutral to the ground on the plug since the neutral bus and ground are tied together in the breaker box? Also, this breaker box has a heavy bare ground wire tied to the neutral bus that goes out of the building and, I assume, underground to the main breaker panel. Is that how a "properly installed metallic conduit system" should be grounded? The screws on this plug that connect it to the box go through a shinny metal piece of metal that basically makeup the rear body of the plug - the top and bottom screws that hold it to the box go through it and it then wraps around the back (running vertically) and has a tab that includes the ground screw so, I gather it would be a grounded plug (I guess what I am saying is the same piece of metal that the screws go through to mount it the ground screw also screws into to press the ground wire against it). Does that sound correct?

I really appreciate it.
 

wwhitney

In the Trades
Messages
6,532
Reaction score
1,822
Points
113
Location
Berkeley, CA
In this case, I gather I would not use the white neutral wire and just connect the two hot wires and use the conduit as my EGC? Or, should I connect the neutral to the ground on the plug since the neutral bus and ground are tied together in the breaker box?
Yes, and absolutely not. The only neutral ground connection is the main bonding jumper at the service entrance (which means that in the service panel the ground and neutrals may be intermixed).

Also, this breaker box has a heavy bare ground wire tied to the neutral bus that goes out of the building and, I assume, underground to the main breaker panel.
That wire sounds like the Grounding Electrode Conductor, but that would only make sense in the service panel itself. What do you mean by "underground to the main breaker panel"? Is this building supplied by a service, or by a feeder? And if it's a feeder, does the feeder have an EGC, or is it an ungrounded feeder (a type of previously allowed installation that is still OK if the feeder is the only metallic path from the supplying building).

The screws on this plug that connect it to the box go through a shinny metal piece of metal that basically makeup the rear body of the plug - the top and bottom screws that hold it to the box go through it and it then wraps around the back (running vertically) and has a tab that includes the ground screw so, I gather it would be a grounded plug (I guess what I am saying is the same piece of metal that the screws go through to mount it the ground screw also screws into to press the ground wire against it). Does that sound correct?
If one of the two screws is held loosely to that yoke by a piece of paper, and the other is held tightly to the yoke by a little metal tab, then that little metal tab makes the receptacle self grounding. If both screws are loose in the yoke, and there's no metal tab, then you need a bonding jumper from the ground screw to the metal box.

Cheers, Wayne
 

Tim Fastle

Member
Messages
59
Reaction score
0
Points
6
Location
New Mexico
I should add, I was going to use wire nuts for my pigtails in the box. Would that be a bad approach or a code issue or just not as good as Polaris type connectors?
 

Tim Fastle

Member
Messages
59
Reaction score
0
Points
6
Location
New Mexico
I must say, that is an excellent reply.

I am not sure how to copy what you replied with like you did with mine but I will try to make sense.

1st point - ok, so that is what I thought and I will not use the white (neutral wire). What I would be incline to do is to cap it on both ends then.

2nd point - When I noticed that there were no ground wires in the building I thought that the conduit must be the ground. To confirm I looked in the panel (It's a sub-panel, the main is on a pole outside the house and it feeds the house and the out building (shop)) to make sure there was a ground from the panel to the main. This is part of why I asked the question. There are three heavy wires coming out of conduit in the slab and into the box f, 2 hot (black) and 1 neutral (white) - they look to be 6 gauge, maybe 4. Then there is a bare copper (ground) wire going from the ground bus (one space bus so not sure what that is called and it's bolted to the neutral bus) out of the bottom of the box and out the edge of the slab. I assumed it had to be the ground going back to the main because, an earth ground there doesn't make any sense and, from what I recall (mostly from physics 2 and electricity/electronics way back in college), shouldn't be there. I am not positive what you mean by "underground feeder," but it does come up through the slab so it is underground but I am fairly sure those 3 wires (2 hot, 1 neutral) are in conduit from the main panel. The building is fairly old, probably 60s or 70s but the prior owner said most of the electrical had been redone or added fairly recently. This doesn't look that old but I thought possibly it was before grounds were required and maybe they added the ground wire later and ran it underground to the main. Is that possible and if so should I be able to look in the main for it and if I found such a wire check for continuity to the shop and then I would know? It does seem to me that there should be a system ground to the shop.

3rd point - then this is not a self grounding receptacle. How would one best go about jumping to the box? There's no ground post in this box. Might be most simple to just get a self grounding receptacle.

Sorry for the slow reply, I had to run an errand.

Thanks for the input and information.
 

wwhitney

In the Trades
Messages
6,532
Reaction score
1,822
Points
113
Location
Berkeley, CA
0) You hit reply and note that the text is within tags that say
Code:
[quote=blah]text[/quote]
(where I just had to use some special tags to keep that from getting parsed.) Then you take the text chunks you want and similarly bracket them with the same tags. I usually highlight the first tag and copy it, then just type the second tag.

1) Capping at the source end is optional.

2) Does the main on the pole outside the house disconnect power to the outbuilding? Is the conduit from the main on the pole to the outbuilding's upstream most panel plausibly all metallic without interruption?

3) Normally a 4 square box has a tapped hole in the back, often on a raised boss. Then you get a 10-32 ground screw (often green, but not required to be), screw it into the hole, and land a conductor on it just like the side screws on a receptacle. If there is no tapped hole, then ideally you drill and tap one yourself.

Cheers, Wayne
 
Last edited:

Jadnashua

Retired Defense Industry Engineer xxx
Messages
32,771
Reaction score
1,191
Points
113
Location
New England
If your compressor only needs L1, L2 and a safety ground, if you reidentified the neutral wire, and moved it to the ground on both ends, that should work. Green tape or marker, at least functionally...not positive on code.
 

wwhitney

In the Trades
Messages
6,532
Reaction score
1,822
Points
113
Location
Berkeley, CA
If your compressor only needs L1, L2 and a safety ground, if you reidentified the neutral wire, and moved it to the ground on both ends, that should work. Green tape or marker, at least functionally...not positive on code.
For wires in conduit, relabeling is prohibited for #6 and smaller. You'd have to pull out the white wire and pull in a green wire. Which would be redundant as there's already an acceptable EGC, the metal conduit.

Cheers, Wayne
 

Tim Fastle

Member
Messages
59
Reaction score
0
Points
6
Location
New Mexico
jad, thanks for the suggestion and I will consider that but if I am confident I have a good EGC via the metal conduit I will be fine with that as well. I think at this point my main concern has become making sure that indeed the panel and conduit in the shop are correctly and safely grounded.

To answer you questions WW:

Yes, the main at the pole does disconnect power to the shop via a 60amp breaker and no, metal conduit does no go from the shop to the main box.

My thinking is that I need either the bare ground wire (I believe it's 6 gauge, solid) or metal conduit making a continuous run from the shop to the main. From visual inspection, clearly metal conduit is not making that run. So I decided I need to figure out if the bare ground wire is. There is the same type of wire in the main box attached to the ground bus so I hoped that was it. The next part is a bit on the "hack" side of the equation but sometimes, ... that's how you gotta roll.

To determine if the bare copper ground wire in the main is the same as in the shop I disconnected it from the ground bus in the shop. I then ran a wire from it (the bare ground in the shop panel) to the main panel and indeed there was continuity. In my thinking, with the ground wire disconnect from the shop panel, this means it's the same wire and thus the box and conduit in the shop are correctly grounded. Does that make sense or is there any other way that wire could have continuity from the end in the shop to the end in the main panel?

I think there is a threaded hole in the bottom of the box for the jump wire, if not, there will be one when I am done.

I would love to have used the "quotes box" but, even after reading your instructions, it's still a bit outside my wheel house.

If you see something wrong with my thinking (often the case) please let me know!

Thanks
 

wwhitney

In the Trades
Messages
6,532
Reaction score
1,822
Points
113
Location
Berkeley, CA
Yes, the main at the pole does disconnect power to the shop via a 60amp breaker and no, metal conduit does no go from the shop to the main box.
OK, so if the main at the pole disconnects power to the shop, but the feeder from the pole to the shop is just (2) ungrounded (hot) conductors and (1) grounded (neutral) conductor, and there is no EGC in that feeder (neither a wire nor continuous metallic conduit), then you have an ungrounded (without EGC) feeder. Those were allowed in the past when there is no other metallic connection between the source and the outbuilding. So no metal water lines, no phone lines, no cables lines, etc. In existing installations they are allowed as long as there are still no other metallic connections between the source and the outbuilding. [In this case, since the source is a pole with a feeder serving the main house, you need to consider whether there is a path from pole to house to outbuilding.]

For that situation, the feeder to the outbuilding is treated at the outbuilding like a service. So the outbuilding's main panel should have an EGC-neutral bond (in this case the neutral bar can just be bonded to the case of the main panel). And the outbuilding's grounding electrode system (always required when an outbuilding is supplied by a feeder) should connect to the neutral bar. So that bare ground wire is likely the grounding electrode conductor (GEC) from the grounding electrode system (e.g. there should be two ground rods, or a connection to the outbuilding's concrete foundation.

Now as to your test that you have continuity between the grounding electrode conductor and the EGC in the main house, in the situation above, you actually don't want continuity. The only path should be through the earth. Because if there is continuity (e.g. a common metal water pipe or a communications cable shield), then you want an intentional path, namely an EGC in the outbuilding feeder. Which would mean eliminating the neutral-ground bar in the outbuilding main panel, and landing the GEC on the ground bar, not the neutral bar.

This is getting a bit complicated but I would suggest you carefully consider if there are any other metallic paths between the pole/house and the outbuilding, and then possibly repeat your test (with power off in the outbuilding) while taking a resistance measurement.

Cheers, Wayne
 

Tim Fastle

Member
Messages
59
Reaction score
0
Points
6
Location
New Mexico
Well Wayne, it is getting complicated. Seems most things do in life.

So, if you are still with me I looked more closely at things took some pictures. Not sure they will makes sense but I labeled things things with letters.

A = Feed to shop
B = Bare Ground Wire in Shop
C = Feed to Shop
D = Feed to Home #1 (the home has an addition, 2 different panels both fed from the main on the pole outside)
E = Feed to Home #2
F = Neut to Shop (I think or, my guess)
G = Nuet to Home #2

I am not sure if this shows what you described above as how is should be but I am hoping so.

I also rechecked the bare from the shop wire for continuity to the wire in the main panel. I disconnected it at both ends. It had continuity. But it also has continuity to the main panel even disconnected from it. So, I assume it is somehow bonded to the ground to one of the other two feeds and not a direct run from the main to the shop.

The original electrical was wired in the 70s or 80s. Then, in 2018, I think in coordination with the addition, the main panel was redone and the feed to the addition added (that feed has a ground wire - the other two do not except the one I thought came from the shop). The inspection sticker from 2018 is in the main panel. I assume, since it was inspected it is likely up to code but would like to be sure. I also am guessing, as per your post above, that is why some neutrals are on the neutral bus and some on the ground bus but, I am not sure about that.

What everything is may be a little unclear but I suspect it will make sense to you.

Thanks,

Tim

ShopPanel A.jpg
BareCopperFromShopPanel.jpg
MainPanel.jpg
 

wwhitney

In the Trades
Messages
6,532
Reaction score
1,822
Points
113
Location
Berkeley, CA
Okay, a little background to help you understand the issues here:

The purpose of the EGC (ground wire) is to bond together all the metal things that should not be energized by electricity, and if they do get energized, cause the breaker to trip. To do the latter, a hot connected to the EGC needs to complete a circuit. That is done by a single connection between the EGC and the grounded supply conductor (the neutral). However, any EGC system should have only one connection to the grounded conductor; if there is more than one connection, then that means some normal grounded conductor current will flow on the EGC (current will take all paths). And the EGC should not be carrying any current under normal conditions (if it does, it's called objectionable current).

A service from the power company has no EGC. So the EGC is established at the main service disconnect (the upstream most breaker) by having a single EGC-grounded conductor bond there. In your last picture, that green screw to the right of the right yellow plastic piece is your EGC grounded conductor bond. The grounded conductor bus is otherwise insulated from the panel cabinet, but that green screw goes through the bus, through the insulator, and screws into the back of the cabinet.

The modern way to run a feeder between structures is to include the EGC with the feeder. Then in the structure being supplied, the main panel there should not have a bond between the EGC and the grounded conductor. The EGC system in the building supplied is just an extension of the EGC in the supplying structure, and you don't want a second EGC grounded conductor bond.

However, an older way to run a feeder between structures was to omit the EGC in the feeder. Then the feeder at the supplied structure is treated like a service, where a new EGC system is established in the supplied building via an EGC grounded conductor bond in that building's main panel. The new EGC system and the supplying structure's EGC system should have no direct connection to each other. If they did, you'd again have multiple EGC grounded conductor bonds, causing objectionable current.

Hence the restriction for buildings supplied by a feeder without an EGC, that there be no other metallic paths between the supplying structure and the building. Because if there were, e.g. a water pipe, that water pipe would be bonded to the local EGC system at each end, causing the EGC systems to be interconnected. So that's something to watch out for, where you have an existing structure compliantly supplied by a feeder without EGC, and someone later comes along and adds a metallic water supply to the building. When they do that, they should also upgrade the feeder to include an EGC, and remove the grounded conductor EGC bond in the building's main panel, but they may not realize that.

One more thing to be aware of is that each building (except outbuildings fed by only one branch circuit) is required to have a grounding electrode system. If the building is supplied by a feeder with EGC, that grounding electrode system is connected to the EGC; if the building is supplied in a way that requires a grounded conductor EGC bond in the main panel, it is often instead connected to the grounded conductor in the main panel (which may not be a real distinction, often there's just one terminal bar serving both).

In your first picture, the solid bare conductor B is your grounding electrode conductor (GEC), it's not part of the feeder to the building (99% sure). In which case that panel should have a grounded conductor EGC bond in it. I can't tell if it does, is one of the screws through the neutral bar just extra long so that it goes into the case? And what is that blob to the left of where B terminates?

Anyway, assuming we are correct that the feeder to the outbuilding has no EGC, if you lift that GEC, the only connection between it and the EGC at the pole/house should be through the earth. So if you took a resistance measurement, I wouldn't expect anything lower than 25 ohms (which would be a very earthing connection). I'm not sure what the threshold is for your continuity tester, it might show that as continuity. But if you take a resistance measurement and get a low number like 1 ohm, that's a problem. Because most likely that means there's some other metallic path between the buildings, and the GEC is bonded to that path. [I assume the incidental contact between the case and the GEC as it exits the panel is not involved, but you could try wiggling it there to see if changes your measurement.] In which case the proper thing to do is to replace your feeder with a new feeder with EGC.

So that's the theory involved, and if I have a chance I'll look more closely at your pictures.

Cheers, Wayne
 

Tim Fastle

Member
Messages
59
Reaction score
0
Points
6
Location
New Mexico
The "blob" is my cold fusion generator, it's experimental and top secret, generates all my power so I can be off the grid. Shhhhhh, don't tell anyone.
Actually, it's an old mud-daubers (type of wasp) nest. Not sure how it got in there but, there it is. I'm going to clean it out this trip I hope.

Do you teach this stuff or just actually paid really good attention when you learned it. You ought to write a book.

I read your reply a few times and am still trying to digest it all but I think I get most of it. I have to run a quick errand and will look at it again when I return and see if I can grasp it and apply it to my situation.

And yes, the bottom screw on the neutral bar does thread through the case and sticks out the back an 1/8" or so.

The well is fairly new and is pvc casing and I am fairly certain the water supply lines are all PVC. The well equipment is in the shop (power, pressure switch, pressure tank) and on one of the breakers in the shop panel. The old well had steel casing (it's abandoned now) but I don't know if the water lines on it were steel pipe or not. I'll see if I can figure out if there is anything that could be a metallic path between the buildings. Obviously any metal touching the shop would be connected but on the home wouldn't that metallic path have to be connect to the wiring somehow?
 

Tim Fastle

Member
Messages
59
Reaction score
0
Points
6
Location
New Mexico
I redid my test but rather than just listen for continuity I actually looked at the reading. 2.5 ohms.

I looked around and there is metal pipe that comes up from the old well and looks like it may have gone into the equipment in the shop. This pipe comes out of the ground just around the corner from where the 99% GEC goes through the slab and probably about 7 feet away in a straight line. I guess the norm back in the day would have been to tie it something like a steel pipe that goes to a metal well case that goes 250' into the ground?

There are no water pipes touching the steel building, just going under the slab and then up to the pressure switch and tank. But I think what you are saying may have happened is that when this sub-panel went in it would have been normal to not pound a ground rod into the ground and just bond it to the steel pipes that were coming out from under the building and going to the house. And that is why there is continuity (or should I say very good continuity) between the 99% GEC and the wire in the Main Panel - that it's possible that wire is tied to a steel water pipe that runs to the house the GEC for the house is also tied to that pipe. And then from the home breaker panel out to the main breaker panel

In the main panel then, I wonder were that ground wire (top right of my 3rd picture), the one I thought went to the shop, actually goes. The only place that makes sense would be the house's original breaker panel. I think I will take a look at that. If you are still around, I'll let you know what I find.

Sorry if this was confusing.
 

Tim Fastle

Member
Messages
59
Reaction score
0
Points
6
Location
New Mexico
I pulled the cover off the original home panel and there is a bare copper wire like the one in question in the main. If I had to guess, I'd guess that's it. The feed wires for both the original home panel and the shop all go into the same conduit in the main panel (the center conduit). I have no ideal where they split. I'd have to guess the answer is under the house, must be a J-box or something. I don't have good access to the crawl space right now but plan to in the coming months.

As you said, it's getting a little complicated. I think I am going to call the electrician that did it and talk to him, maybe have him come out and take a look and tell me what he knows. From there, if something needs to be rectified (I might have to point it out to him), then get it rectified. I am not going to be able to do that for a while. I assume, since it's worked like this for many years, that it's not a major risk to continue as is for a few months until I can get it addressed?

We left the scope of this forum a while back and I REALLY APPRECIATE all the help and input but I don't think it's realistic that I will get to the bottom of it today. I do think it's good though that I now know it should be looked at. On the plus side, I did get my box wired just as you suggested and my compressor is working just fine.

What do you think?
 

wwhitney

In the Trades
Messages
6,532
Reaction score
1,822
Points
113
Location
Berkeley, CA
A few comments:

A simple test you can do to see if you have a parallel metallic path is to turn on an unbalanced set of loads in your shop (i.e. 120V loads all on one leg; 240V loads don't matter, and adding 120V loads on the other leg lessens the effect) and then take a clamp on ammeter to measure the current on the shop GEC. It should be 0, but if it's in parallel to the feeder grounded conductor, it will be carrying some of that current, and will read non-zero.

That brings up an important point that there's a possible big hazard here if you go around disconnecting GECs and bonding wires. Which is that when you have a parallel metallic path like that, it can happen that the grounded conductor in the feeder gets damaged, and the alternate path becomes the only current path. Then if you disconnect it, and you end up touching both sides of the previous connection, you now because part of the path for the unbalanced current on the grounded conductor. Not good at all, people (often plumbers) have died that way. The upshot is that you shouldn't disconnect any bonding wires without checking them for current with an ammeter first. If there is current, shutting off the main breaker of the outbuilding should make it go away, so you can do your testing.

Since your service panel is on a post and that makes it a separate structure, it should have its own grounding electrode system (GES) and the bare copper wire would be its GEC. Now if it's very close to the house, then it could share the GES with the house. That is a bad idea if there is an ungrounded (without EGC) feeder from the post to the house, as this path: post panel -- post GEC -- common GES -- house GEC -- house panel is a parallel path to the feeder grounded conductor, given that you will have GEC - neutral bonds in each panel. In which case I would recommend you replace any ungrounded feeder from the post panel to the house panel with a feeder with EGC.

Any metal piping system in a building should be bonded to the EGC, even if it's abandoned. That's so that if it accidentally got energized, it would trip a breaker, rather than sit around energized waiting for you to come along and complete the circuit through the earth. So if the abandoned metal well casing has metallic water piping to both buildings, that will be your other metallic path. Also, the abandoned metal well casing is a great grounding electrode, and so one of the buildings should use it as a grounding electrode. But as long as you have an ungrounded feeder between the buildings, you can't connect it to both buildings GESs, without creating the problems described in the previous paragraph.

The upshot is that you need to either (a) replace the shop feeder with a new grounded (with EGC) feeder, and eliminate the neutral-EGC bond there or (b) find the other metallic path between the buildings and sever it. For example, if it's abandoned metal water piping from the well to both buildings, you could find a place in between the buildings, dig up a few feet of the pipe, remove it and cap the ends. [Of course you'd want to be sure it's (one of) the culprit(s), so you could locate the piping at both ends, temporarily remove any electrical bonding wires from the pipes (after checking for current), and then check the resistance between both ends.] The advantage of not severing the pipe at the building is that if the metallic piping is buried at least 10 feet, it becomes a grounding electrode to connect to the local GES.

Cheers, Wayne
 
Top
Hey, wait a minute.

This is awkward, but...

It looks like you're using an ad blocker. We get it, but (1) terrylove.com can't live without ads, and (2) ad blockers can cause issues with videos and comments. If you'd like to support the site, please allow ads.

If any particular ad is your REASON for blocking ads, please let us know. We might be able to do something about it. Thanks.
I've Disabled AdBlock    No Thanks