CU move. How do I connect this lot up?

[Colin Stamp]

Hi all,

We finally moved into our doer-upper house last Monday and, working my way down a very long list of jobs, I've come fit a new CU in order to get RCDs and a cooker circuit. The job is getting pretty urgent because we're all getting cheesed off with takeaway food. It's only a temporary fix I'm after - all of the final circuits are going to go when I do a complete re-wire some way further down the job list.

The original plan was to site the new CU where the old fusebox was in the bottom of a cupboard in the kitchen. In-case some of the wiring needed extending inside the new CU, I bought some butt crimps and one of these:-

course, now I've got into the house, I've decided not to put the new CU where the old one was, but to move it to the opposite side of the wall, and at the top instead of the bottom. The new tails are going to end-up 2.5M long. I've stuck with the previous 16mm^2 for the tails and upgraded the CU earth to 16mm^2. It's a TN-C-S system.

The new CU is now in and I'm currently wiring up the extension wires to it. Back at the meter, I'm left with this :-

I'll take the plunge, rip out the fusebox and connect the extensions to the old soon-to-be-disused final circuits. My question is how to make the connections. The easiest way would seem to be to use the crimps I've already got, gut the old fusebox and put it back to enclose the mess. Any ideas on a better way to do it?

Cheers,

Colin.

[TMC]

I know that this doesn't answer your question in any way but I could not resist asking was there a specific reason why you clipped all of the cables individually to the wall rather than using a length of trunking?

[Colin Stamp]

I had clips but I didn't have any trunking. I actually prefer the look of the cables too - it appeals to the engineer in me. Also, I don't need to worry about de-rating the cables.

It's a temporary job, by the way. After the re-wire, only the tails and earth will remain.

Cheers,

Colin.

[Andrew Gabriel]

I would install the new CU and connect the tails to it. I would fit a suitably high current spare MCB or MCBO in the new CU, and use that to daisy-chain to the old CU with suitably thick cable, until such time as the old CU and wiring it all ripped out. Don't run any high current load such as a hob or shower via the old CU though.

BTW, have you really left the installation with exposed live conductors (which is not safe even temporarily), or is the main fuse removed?

[Colin Stamp]

I must admit, I hadn't thought of that, but I'd rather continue with my current scheme and get a proper two RCD setup until the big re-wire.

I'm just putting in the temporary cooker circuit now, direct from the new CU and all surface wired across tiles. It's going to look really pretty. ;o)

When I took the photo, The lid was off so I could count how many cables I'd need to extend. It's back on now until I work out how to make the connections.

Cheers,

Colin.

[John Rumm]

The short answer is yes, if you really want the circuits fed via dedicated MCBs on the new CU, then you could make crimped connections and use the old CU just as a "box" to enclose them.

However personally, I would not bother linking all the existing circuits through to the new CU, I would either:

1) Install a henley block near the old CU. Take the existing tails into the Henley, and then new short tails from there to the old CU, and also connect your long new tails to the new CU. So you have both CUs connected and working. Then as you migrate circuits, connect them to the new CU. When all complete, simply remove the old CU and its connection to the Henley.

or

2) As above with henley, but only connect the existing tails to the new long ones. Then run a single high power submain from the new CU (say a 45A circuit (RCBO if you want)), to the incomer of the old one).

(it depends a bit on how long this temporary arrangement is going to last)

[Colin Stamp]

The original plan was to get to the re-wire quite early since the whole house currently needs heavily re-decorating but, knowing how these things go, it could take a while before I get round to it. Loads of "urgent" jobs have materialised that I wasn't expecting...

The submain option is tempting, making one of the RCDs in the new CU a whole house one for a while, but I would like to give it some more

17th-ish leanings until the re-wire. The clincher is that the extension wiring is all done now. Perhaps I should have asked the question before starting the job ;o)

Anyway, it's good to hear that crimps in the shell of the old fusebox is a reasonable way to go. Hopefully, a morning of crimpage tomorrow will see us all up and running on the new CU. It'll have to - the in-laws are coming in the afternoon. We've dug out the camping stove just in case...

Cheers,

Colin.

[John Rumm]

Often the way ;-)

Well possibly. What you propose is ok, but perhaps more work than was required...

[Andy Wade]

Grouping factors. The rating factor for nine cables, bunched and enclosed is 0.5 [Table 4C1]. Given that the rating for 2.5 mm^2 enclosed in conduit or trunking is 23 A [Table 4D2A], this gives an as-installed rating of only 11.5 A - quite some way short of the 20 A required for a ring circuit.

Bundling all the cables leaving a CU into a piece of 2x2 trunking is commonly seen, but isn't necessarily compliant.

[jgharston]

Can I ask why you don't want the CU next to the meter?

JGH

[Colin Stamp]

It's not that I don't want it near the meter. If it had stayed where it was, it would have ended up really inaccessible when we re-do the kitchen. Ideally, the meter would move too but I bet that costs...

Cheers,

Colin.

[Colin Stamp]

Yep. I sort-of led myself into the current scheme, since I was initially going to put the new CU where the old one was, so the old one had to go. More haste, less speed...

I do still like the idea of having the new CU fully up and running from the outset though.

Right then. Let the crimping begin.

Cheers,

Colin.

[jgharston]

Looking at the photo, you've got one of those dinky modern ones and the supply is PVC sheathed, not brittle lead'n'paper, so it should be easily movable at some future point, especially as the backboard is screwed to the wall with accessible screws. Try to arrange whatever encloses it in such as way that somebody can moderatly easily get the backboard off the wall and shift the lot en mass to a new position.

JGH

[Tim Watts]

But those derating factors assume all the cables are simulatenously fully loaded IIRC, which is unlikely in a domestic setting - given that the sum of the beaker capacities could easily exceed the main fuse capacity.

I must admit, I am unsure how to formally allow for diversity WRT grouping factors - other than if the sum of the circuits is say 150A and the supply is 100A then in the case above, factor for a grouping of 6 cables.

Is there a formal approach?

[Colin Stamp]

All done, Just before the in-laws arrived :o)

Here it is, in all it's glory:-

the new CU:-

're all up and running again and we have a cooker now :o)

Right then, a couple of dozen little jobs and then TRVs next, I think...

Cheers,

Colin.

[Colin Stamp]

Yep. I'll try and keep it vaguely accessible. It'd be nice to get it moved to the outside of the wall eventually. I'll have a think when I do the kitchen...

Cheers,

Colin.

[John Rumm]

[Andy Wade]

I certainly agree with your drift. Domestic load factors are generally low and the probability of having several circuits simultaneously heavily loaded is low - unless there is an extensive space heating installation. Even so, I don't think that it's a good idea to throw all the outgoing cables in to one piece of trunking, without further thought.

Not that I'm aware of, other than the full sizing procedure in Appendix

4 (of BS 7671). Section 7.2.1. in the OSG sets out some rules which, if followed, mean that "derating for grouping is not necessary." Heating and water heating circuits are excluded and the cable installation conditions are constrained to avoid any bundling.

By the way the assessed FLC for an installation is the diversified sum of the individual design currents (Ib), rather than of the OPD ratings (In). There's no difference for the standard socket circuits, but for most others there is.

[John Rumm]

Our summary of those references here:

By the way the assessed FLC for an installation is the diversified sum