What's the correct way to solder twin and earth cable?

[clangers_snout]

I need to join some T&E cable that will be inaccessible and I don't want to use maintenance free junction boxes, so what is the approved way to solder and sleeve T&E cable?

[Roger Mills]

Probably better to crimp them - as explained in the DIY Wiki.

[tony sayer]

In article , clangers [e-mail] scribeth thus

If your having to ask that.. Don't do it .. crimp it, I would anyday over solder much less fuss and a better joint and self insulating:>...

But make sure to get a decent set of RATCHET crimp pliers not the cheap 'n useless ones..

[Harry Stottle]

I have a couple of junction boxes in a ring main circuit with spurs off, which apparently don't comply with the latest regs, is it OK to replace these with crimped connectors, and if so, what size connectors would be needed for the extra cores of the spur. If this is not OK, would it be OK to neatly solder the joints, and then lay them in the junction box as before, using the original screws to hold the now soldered joints in place?

[John Rumm]

You need some heatshrink in a couple of sizes (about 12mm does the overall sheath). Strip the wires - about 15mm ought to do it. Slide a smaller heatshrink onto the L & N wires, and the larger one onto one of the cables. Twist the ends together tightly inline - much of the mechanical strength of the joint will come from the twisting - you don't want to rely just on the solder. You then want an iron with a decent sized tip and heat capacity. Using a flux cored electronics solder, tin the tip, apply the iron to the joint, wait, then apply solder to the joint and allow it to flow into it. Remove the solder and then the heat. Allow to cool before moving. Repeat for the other wires.

Now using a multimeter on a low ohms range, test the joint quality from the next adjacent test points (i.e. sockets either side in the ring). Disconnect the repaired wires at both ends from the circuit, and at one and twist together all three wires, then measure the resistance between L&N, and L&E at the other position.

Compare your results with the expected values shown here:

(you will need to estimate the length of the cable you are testing)

If all is well, heatshrink it up. And wire the cable back into circuit.

Don't underestimate the importance of testing - one dry joint could cause significant localised heating of the cable.

[Andrew Gabriel]

Why? Is it because they are inaccessible for testing/inspection?

There isn't any general requirement to bring such items up to current regs, as long as they conform with the regs in place at the time they were installed. Having said that, I think you have to go back before

16th edition before that conformed, and this is something I would bring to current standards.

Secondly, unless you are already very experienced at electrical soldering, I would suggest using crimps. Crimping is likely to produce a more reliable result in less experienced hands, providing you are using a proper ratchet crimper and do a few practice goes first on offcuts, and make sure you can't pull them apart and the conductor is clamped firmly.

If you are _very_ experienced at electrical soldering, then a way to do this is to use a standard screw terminal junction box as normal, and afterwards solder each terminal and its conductors together. This is not trivial because the T&E insulation won't stand soldering temperatures with any stress on it, such as bends near the terminal, or pressing on anything (other wires, box edge). Also you'll need a powerful iron (50W absolute min) or you'll take too long heating the metal and cause too much damage to the insulation. Make sure the box itself is made of thermosetting plastic (non-melting) -- the circular ones are normally OK.

[Harry Bloomfield]

John Rumm laid this down on his screen :

Rather than try to twist them together, which with 2.5mm will be difficult, overlap the two ends then bind with some small diameter copper wire (30amp fuse wire?), then solder. I would also suggest offsetting the three joints so they don't all occur at the same place so the bulk is reduced.

[clangers_snout]

> (you will need to estimate the length of the cable you are testing) >

Thanks John, this is exactly the way I would have tackled it, being an electronics engineer. However what concerned me is that the heatshrink sleeving is normally very thin, certainly much thinner than the cable sheath. I'm concerned that this would fail an electrical inspection. Any thoughts?

[Harry Stottle]

Thanks Andrew, I am pretty good at soldering so it should not be a problem. I had thought of binding the cores with 5A fuse wire before soldering, the cores should then fit in the terminal box OK, and could be also screwed, a bit like a belt and braces approach, but when I got the soldering iron out to do it, I realised that I had turned the power off and hadn't got any power to do the soldering, so it was placed on the to do list :-) I will have to run an extension from the garage to get it soldered, before nailing down the floorboards.

[John Rumm]

With a couple of pairs of pliers, twisting 2.5mm^2 inline is actually quite doable - just make sure you start with enough length of wire.

Yup, staggering the joints helps.

[John Rumm]

>>> (you will need to estimate the length of the cable you are testing) >>

Heatshrink IME is a little thinner - but also a bit tougher. It certainly withstands 500V testing with an insulation resistance tester. If you are particularly concerned there is nothing to stop you using two layers. The main thing is to ensure the joint itself does not have any sharp protrusions like wisps of solder or wire ends poking out.

Presumably if you are soldering, then it is in a place where it can't be inspected in future anyway!

[John Rumm]

Another reason crimps are popular! ;-)

Nothing to stop you placing crimps in the JB and just using it as a housing to save titting about with heatshrink.

[Tim S]

John Rumm coughed up some electrons that declared:

As it happens:

need to use such a big box (I had a couple of old ones kicking about and this is a temporary circuit).

I'd never bothered with crimps on mains before, but I knew I'd need to and with the correct rachet tool, they are very solid and convincing. Rapid sell the correct tool at a reasonable price.

Cheers

Tim

[Mike Clarke]

Crimps will join 2 cables end to end without any overlap, they aren't suited to adding a branch.

[Harry Stottle]

But I've got a soldering iron, and not got a crimper :-) For very occasional use, would this be OK?

[Tim S]

Harry Stottle coughed up some electrons that declared:

The trick is to give the wires a bloody good pull after crimping. If they don;t come out you're winning.

Good insulated crimpers will give you a double crimp on each go - that's something to check for too.

[Clive]

Soldering iron and solder. Was it a trick question or are you genuinely stupid?

[John Rumm]

[Graham.]

> (you will need to estimate the length of the cable you are testing) >

What kind of solder? There seems to be more choice in this post lead era.

[John Rumm]

I always use 60/40 tin/lead solder. Something with a flux core, and a eutectic melting profile. I usually get a fairly fine solder for electronics work (say 22 swg) and hence would use that - however if I were buying it for the purpose, then would go for a heavier gauge.