Should T&E cable ends be twisted at socket terminals?

May 14, 2009 16 Replies

When using 2.5mm twin-and-earth cable, and where two wires go into one terminal on the back of a socket, should the two bare wire ends be twisted together?



I have read Google sources that think twisting weakens the wire, and say, e.g. "Insert wires fully into their terminals. Don't twist two or more wires together".



A different Google source says "Taking the conductors, twist each together ... now insert the pairs of conductors into the correspondingly labelled terminal".



Socket manufacturers such as MK have nothing to say on this matter in their installation instruction sheets.



What is good practice, should the wires be twisted, or not twisted?


Not with conductors of that size.

Not twisted. If you twist them it makes future testing and maintenance a pain.

I agree, I find that twisting them together often weakens the cores. As Andrew implies, with cores that size, they are both/all compressed by the screw and twisting doesn't improve the connection. Twisting makes them more difficult to wire up (cables combine together to fight against you)

Don't twist. It weakens the conductors. Of course make sure both straight cnductors are each gripped well.

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D\

=A0 =A0 =A0 =A0|

=A0 =A0 =A0 =A0|

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D/

What about the CPC then - if I've got two coming from the same access point in the backbox, I usually twist them before slipping the sleeving over.

Rob

same problem really - if you need to separate them to do low ohms tests etc then you need to unsleave, untwist etc to do it.

Copper 'work-hardens' which means that if you bend it about it gets harder and more brittle. I'd worry that if you twist 2.5mm wire it would work-harden to the point that it may fracture when clamped rather than deforming properly. Lighter gauge may well tolerate it better, the CPC is usually significantly lighter gauge.

I never twist them, and have never had a problem. You only need three hands if you didn't leave enough of a tail.

R.

Many thanks to everyone for your replies. I particularly take the point about copper work-hardening, and the point about ease of maintenance and fault testing. It is good to be told that no trouble results when the wires are not twisted.

Regards,

Anode.

What about the CPC then - if I've got two coming from the same access point in the backbox, I usually twist them before slipping the sleeving over.

Rob

Earth sleeving is cheap enough to cover both CPCs with seperate sleeving.

Adam

If you can't twist them, then why does this problem of work hardening come about when they are not twisted in a screwed terminatator?

Dave

The deformation by the screw can do it, but most likely it's mechanical or thermal movement of the wire. R.

That means cutting the CPC, which I try to avoid.

You don't need to twist the CPC's to slide into a piece of sleeving.

Some people try to install ring final circuits without breaking the conductors (i.e. cutting the cable) in the belief that this gives them a fundamentally more reliable and higher-integrity result than if they didn't bother. It takes a great deal of planning, lots of mucking about pulling the one-piece ring cable and ages to strip cables without cutting the conductors.

The OP may be doing this or a subset thereof.

I do try to avoid breaking the conductors at every connection and socket outlet. I have managed to install a few complete rings with no breaks in the conductors at all. It makes for an interesting challenge, but not something to get too anal about. If you can do it, each socket is only one connection from the fusebox, otherwise some may be accumulating 20 or more connections along the away.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required