When to require bonding both ends of a home-run conduit to the EGC?

[roger]

I wonder what that listing would mean. Would it mean that the box is water tight or just that it won't deteriorate if it gets wet?

Well maybe the OP can show us what he is using.

 

[Zee]

In nonresidential, would a metal bell box be listed for bonding over 250V (250.97)?

Or is this just for the DC circuit wiring on the roof / only works in residential because they don't make them large enough for commercial?

Frankly, I did not know about that requirement.
Over 100 or so inspections of both >250V DC installs and <250V Enphase installs I have never had an inspector complain about our bell box as J box. Take that for what it is worth.

I've been simply running on the requirement to use bonding bushings if over 250V AND also concentric/eccentric knockout in the box, or non conductive box.
A bell box is a solid, one piece molded box. Aluminum I believe. With threaded outlets/hubs.
Seems solid to me.
Though.... that is no Code reference!

Bonding bushings are the tiniest - YET most laborious - product to install PROPERLY. I avoid them like the plague. So i use metal boxes and cut my own hole or smallest KO that just fits with no welded rings remaining.

 

[jaggedben]

Any box used in a wet location needs to be listed per 314.15, so that's a given. The question is what the 2020 NEC 250.92(B)(2) language "Connections using threaded couplings or listed threaded hubs on enclosures if made up wrench tight" means. Do you have to use a listed hub? It's silent on integral threaded entries.

The 2023 NEC 250.92(B)(2) language is clearer: "Connections made up wrenchtight using threaded couplings, threaded entries, or listed threaded hubs on enclosures." So any integral threaded entry suffices, and the part about "listed threaded hubs" only applies to add on hubs.

Cheers, Wayne

You are right of course, but it's ridiculous, and I'm willing to go on record here to say that I literally don't care.

 

[roger]

If the plastic boxes are suited for wet installations I think they probably have some hub type entry and that would make attaching a grounding bushing impossible. If they are using a residential type plastic box I would think the whole installation is questionable, hence my post # 19

 

[photonboy]

My question to the OP is, if the plastic box is listed for wet locations what type of KO'S or hubs does it have?

They don't have any pre-stamped knockouts, each one is drilled on site. Here are a couple of examples, one of which I think is pretty cool:

 

[photonboy]

FFS if the bell box says it's UL listed that's good enough for me.

When bell boxes are used as roof j-boxes we do require the enclosure itself to be bonded with a lug or terminal, that bonding transfers to the metal conduit since there is sufficient thread engagement.

 

[jaggedben]

When bell boxes are used as roof j-boxes we do require the enclosure itself to be bonded with a lug or terminal, that bonding transfers to the metal conduit since there is sufficient thread engagement.

Right. I don't need some further proof that the threaded hubs are listed for grounding.

 

[roger]

So what is the point of using a plastic box if they are not installed to be weatherproof?

 

[pv_n00b]

Why does pv grounding matter anymore than grounding anything else?

Because PV is a source.

 

[ggunn]

Because PV is a source.

Why does that make a difference? Unless a PV system is supply side interconnected, doesn't the grounding look the same as for any other feeder?

 

[electrofelon]

I would say grounding for PV is less important if anything. All bonding will do is make it more likely the inverter will shut off hopefully alerting someone of a problem. Say you have a roof system that is isolated and has a fault. Isn't it just now a grounded system? Probably won't have any way to touch ground on a roof so even if it floats at non ground potential, a shock seems unlikely.

 

[petersonra]

Because PV is a source.

why would it matter for any source?

 

[photonboy]

So what is the point of using a plastic box if they are not installed to be weatherproof?

They are installed to be weatherproof, the snow pic was someone forgetting to close the box during install. Once buttoned up and covered with a module the boxes are weather tight, if installed correctly.

 

[roger]

They are installed to be weatherproof, the snow pic was someone forgetting to close the box during install. Once buttoned up and covered with a module the boxes are weather tight, if installed correctly.

I'm not talking about the snow picture. The fittings used for the single conductors are not weatherproof unless they are filled with some type of compound.

 

[electrofelon]

Bonding bushings are the tiniest - YET most laborious - product to install PROPERLY. I avoid them like the plague.

As a little side rant, yeah I hate BB and doubt they even accomplish much. They are so horribly designed. They need more lug and setscrew options around the perimeter. Also most fittings are not deep enough for the setscrew to grab, especially when you use a plastic enclosure . Finally there is no anti turn function on the lug so pretty hard to get them to stay tight in practice.

 

[photonboy]

I'm not talking about the snow picture. The fittings used for the single conductors are not weatherproof unless they are filled with some type of compound.

Fair point, although the strain relief fittings do have a foam/rubber gasket that snugs around the conductors when tightened and the boxes are typically under a module, best practice is to put the fittings on the downward side of the box but even when they side enter it's good enough, they aren't meant to be submerged, if they are then there are bigger problems..

 

[shortcircuit2]

The NEC only requires that one end be bonded, wire EGC or not. Anything more is design choice.

In general, nonmetallic boxes can only be used with nonmetallic cables and raceways.
If nonmetallic boxes are used with metallic raceway bonding at each end would be required per 314.3 Exception 1 & 2
So if EMT is used from the nonmetallic enclosure of the inverter to the nonmetallic deck-box, bonding (bushings) at both ends is required.

 

[LarryFine]

If nonmetallic boxes are used with metallic raceway bonding at each end would be required per 314.3 Exception 1 & 2

Even if the far box has nothing requiring grounding? What would you connect the bonding wire to?

I don't quite see the connection (no pun intended) you do in 314.

Exception 1 mentions integral means, and exception 2 mentions threaded entries.

So if EMT is used from the nonmetallic enclosure of the inverter to the nonmetallic deck-box, bonding (bushings) at both ends is required.

Does the presence or absence of an EGC in the EMT affect your comments?

If an EGC is present, I would think bonding one end would suffice.

If an EGC is not present, the EMT is the EGC, then I agree on needing the bushings.

 

[Zee]

They don't have any pre-stamped knockouts, each one is drilled on site. Here are a couple of examples, one of which I think is pretty cool:

Well those plastic boxes are better than the boxes I expected when you said plastic. I agree: definitely place any plastic UNDER the panels.

Last plastic box I saw was this on a 10-year old install:
(I am skeptical metal boxes will look like this in 10 years)

 

[Zee]

The PVC box reminds me! Pls tell your subs that there are long threaded EMT raintight connectors. That makes a bonding bushing possible. A PVC box has THICK walls and it eats up all your thread on the EMT connector. Not enough thread left for the bushing.
You cannot get a bonding bushing on there and set the setscrew into metal thread to save your life half the time. Even if you were willing.

Sometimes just skipping the locknut (which no longer serves a purpose if adding a bondng bushing) gets you enough thread.

To be fair your 3 pics show thinner walled "plastic" boxes. Whereas mine, Cantex, I think is actually PVC and thicker.