We opened a panel in our commercial building this morning to trace a few circuits. While in there, I found that one of the neutrals on a branch circuit had been spliced. It is a soldered connection and insulated with what looks like heat shrink tubing. It looks as though a GFCI breaker was removed and a standard 20A breaker installed in it’s place, and they spliced the neutral to reach the bus. Is this legal? Do I have to re-pull this incredibly long branch circuit to correct it or could this be considered a repair to a “damaged neutral”?

It is perfectly legal to splice a wire in a panel, that being said, solder is not an approved method. Also find out where it goes and why it had GFCI protection to start with. The GFCI breaker was there for a reason, and probably needs to be put back.

Soldering is an approved method of splicing.



As a matter of fact, soldering is one of the only approved methods for some applications.

The exceptions to this are for service conductors (230.81), grounding and bonding conductors where the connection relies solely on solder [250.8, 250.148(E)] and connections to grounding and bonding electrodes [250.70].

I think some clarification is necessary.

Solder is a perfectly acceptable method - but not just solder alone. Where solder is used, the wires must have some other means physically holding the wires together - such as twisting them together. Solder will make the 'electrical' connection; the 'mechanical' must come from elsewhere.

Back in the olden days you made a western union splice and soldered it. You can see why they invented wirenuts.
Assuming this splice was twisted and soldered the thing I might question would be the shrink tubing. Is it "equivalent to that of the conductors". Shrink tube comes in different thicknesses.

Reno, 110.14(B) needs no clarification.

Heat shrink does not meet 110.14(B) regardless of thickness, as it is applied after the splice is soldered, vice before. Seems a pretty nitpicky point, though. Is twisting the wires sufficient, though? If so, the heat shrink may only be required to function as electrical insulation. Not sure how you'd check up on that.

Steve, I'm not sure I follow you. Tape is a proper means of insulation, and it's certainly applied 'after' the splice is made. I don't have any problem with heat shrink tube as covering for a splice.

I would have a problem is the tube was left with an open end. Otherwise, use it.

For the mechanical connection, twisting the wires is generally enough. Sure, there are exceptions ... where you need to use a hydraulically crimped or welded connection ... but for most purposes, twisting is plenty good.

Naturally, for larger wire (like you might see used in a service), the practice is not to simply try to twist the wires, but rather to braid the strands together. There's an entire art to these splices, that's pretty much been lost.

I thought tape was not UL approved to "replace" the original insulation. That was why you have rubber tape and then a outer layer of electrical tape. But why wouldn't the person(s) just have used a wirenut. Seems to me that they must have been "nervous" about electricity, probably a handyman special.

All electrical tape, of whatever form, pretty much has to pass the same tests. The only real variable is the voltage rating - and that's based upon only a two-layer thickness.

"Ordinary" electrical tape is listed for 600v. So are the usual splicing tapes and putties. The 'self fusing' nature of the splicing tapes does make the final set waterproof, and the thickness is a real help to building up thickness.

There are reasons for building up some thickness, apart from electrical insulation. For example, to cover sharp edges.

NEC is quite clear about soldered connections being electrically and mechanically secure *before* applying solder. This is impossible with tape or anything else applied *after* the solder connection is made.

If a simple twist suffices as the primary electrical and mechanical connection (is this in NEC anywhere? If so, then tape for insulation should be fine. But a twist is not sufficient for ground wires without a crimp ring or other listed device, so I don't see how it would be acceptable here, either.

The twist gives mechanical connection, the solder gives electrical connection (like the crimp ring). Where is the confusion.
The insulation has nothing to do with the electrical connection so it is not part of that equation. Insulation is another step, just like taping up a split bolt.

But where is twisting allowed as an approved connection?

"Twisting" is the same thing a wire nut does IMO

Twisting is implicitly allowed by 110.14 because what constitutes a "mechanically and electrically secure" connection isn't explicitly defined and twisting is not specifically prohibited.

So any soldered splice that is mechanically and electrically secure before soldering is approved and twisting is just one means to that end.

But it is sufficient

250.8 Connection of Grounding and Bonding Equipment.
Grounding conductors and bonding jumpers shall be
connected by exothermic welding, listed pressure connectors,
listed clamps, or other listed means. Connection devices
or fittings that depend .........SOLELY ON SOLDER........ shall not be used.

Twisted together and soldered is different than relying solely on solder.

There are a lot of things not specifically prohibited by NEC that aren't legal. NEC is always very clear about requiring LISTED connectors, clamps and other devices. Twisting may be common, but unless we expand 110.14 as carte blanche to use any method we feel like including unlisted knock-off products and home brew bolts, etc, it's not authorized.

I have yet to see anything in NEC authorizing twisting + solder beyond 90.2.(C). Not that 90.2.(C) isn't plenty for most jurisdictions, but it's not the same as actual code allowance.

Not to mention there are other factors involved- the difficulty of reversing such a splice, for instance. Might not be an issue if you've got 3' of slack in a panel as you can just cut it off, but what if every connection in a 2x4 outlet box was made up in this fashion?

Steve,

110.14(B)specifically permits a soldered connection.

"Twisted and taped" does seem pretty secure, electrically, if done well. That is not how I would do things but I removed dozens of splices made that way when I ripped out all the previous owner's work in this house and it was at least 20 years old. (dating from a 1978 remodel as best I can tell) I didn't find any traces of heating, even on the 1440w bathroom heaters.

Steve, I think you've got it backwards.

Splicing, by twisting and solder, is the standard by which all else is judged. That is, other connectors are listed as a substitute for this, the original accepted method.

All else - wire nuts, split bolts, PGC's, NSI's, crimps, whatever ... came later.

Could be, but it's odd that NEC would be anything less than absolutely specific about this it were indeed the gold standard.

How many twists are acceptable? What length? What strength to separate? How to properly insulate? Etc. Back to my example before, if soldered twists are permitted, then I would also be allowed to use a bolt from the junk bin and a couple washers to make a mechanical connection, so long as I soldered it.

If you used a brass bolt, flux and dipped it in the solder pot I am not sure how you would ever get it apart.

Back in the olden days this is what they did.

Steve,
How many twists are acceptable in a wire nut? What length do you strip the wires in a wire nut? How much strength to separate a wire nut?

As stated earlier insulation is a different issue altogether. You're flogging a dead horse.

I imagine that Fig 21 "knotted tap joint" and Fig 13, "rat tailed joint" would frequently be merged to create a rat tail joint with an insulated wire extending beyond the joint (the difference in the way the electrical tape was applied). And that the rat tailed joint looks like the ancestor of the wire nut.

Must have been fun to use a solder pot in tight quarters at a junction box in some hard to get at spot... And not fumble the solder pot...

I'm amazed that this forum is looking for loopholes as opposed to the exact letter of the law. If you went into a box and did a bunch of Fig.21s and Fig.22s, would your AHJ approve it?

The difference is that wire nuts are UL tested and approved, and the instructions printed on the packaging- easy to do, easy to comply with, and easy to check. While the solder joint has has no official approval beyond what the AHJ is willing to allow. One author's instruction on how to make a western union joint isn't the same thing as NFPA endorsing it. As listings cost money, nobody is going to get a DIY soldered splice method UL-approval, so the status quo remains. Only exception I can think of would be official US Government mil-specs and such, which are widely recognized as an acceptable listing authority, and do go into detail on methods like this.

At the end of the day, though... Why? Why even bother? There's no reason to do a solder splice in this day and age. It takes more time, takes a higher level of skill, requires insulation to be applied separately, and is difficult to reverse. There's more reason for NFPA to explicitely ban its use than to allow it. (Thanks for the graphics, btw, it's great to see stuff like that!!)

I doubt there is going to be a resurgence of soldered splices anytime soon. This is just a relic in the code like K&T. I don't think anyone here seriously thinks there is a hazard with a joint that was twisted and soldered coming apart. If anything we are just saying the NFPA did not word 110.14 very well. It must be the same guys who wrote 682.

I learned how to make soldered connections when I was a kid helping my older brother. We used fire heated soldering irons for small joints not solder pots.
A few years ago an electrician did a new house and soldered all the joints, with crimps on the grounds. He said the way he did it was to prepare, i.e. twist and flux, all the joints first and then go through with a small torch and do the whole house.
He had good solder flow and no cold connections. So I approved the job. When I asked why he did it he explained that since nobody else did solder work he could tell if someone had messed with his job, changing or adding to it. Kind of like the plumbers when they used copper pipe and soldered everthing.

Well, I guess from averaging all the input I can rule out having to re-pull the circuit. I removed the insulation and the joint looked like it was well executed...probably by an 'old timer'...it bore a striking resemblance to the western union joint that was posted. Thanks for all the input. Oh, BTW, the circuit used to go to a ceiling hung cord reel. It appears they needed an additional dedicated circuit for a drop cord for a piece of cord connected equipment. The receptacle was replaced with a drop cord with strain relief. the box was supported in the grid with a t-bar hanger.

I'm sure there are lots of old timers who think the world is a worse place when wire nuts replaced solder. Sort of like the people in Chicago and their opinion of Romex.

Greg, I'm from Chicago, and I'd like you to answer me one thing:

What's Romex?

Renosteinke; thanks for starting my day off with a hearty laugh.

A few years ago we had a manufacturer try to sell us something in the heart of the downtown Chicago market and the question was asked of him about wiring it up...he explained exactly how to install their product (that would only work with NM cable) in great detail.
After he got all finished and saw a sea of stunned faces, someone asked him exactly what you asked.

"What's Romex???"

I guess he figured that Chicago would happily change their Codes so that he could collect his commissions. He was REALLY depressed when he found out that most of the Chicago suburbs had adopted the Chicago requirements and that he had to go 2 counties away before there was any hope of selling his wares.

Thanks again...I really needed that smile today.