Relocating Consumer Unit to new location

[richardg]

A builder has asked for a quote for rewiring a domestic property all very straight forward until he mentioned relocating the consumer unit to another part of the house with a run of 10m to the new location. TNC-S at orgin.

Looking at options to make this as painless as possible. Thoughts on the different solutions?

Option 1 (no mechanical protection of cable run, in plaster and under floor boards)
Small metal clad unit with DP 100A main switch, 80A 100maRCD 2 pole with 16mm Double insulated meter tails to new consumer unit. Separate 16mm earth. Current rating 87A Consumer unit fitted with RCBO type A

Option 2
Small metal clad unit DP 100A main switch, 80A MCB with 16mm 3 core SWA -banjo on supply end, Current rating 94A. Consumer unit fitted with RCBO type A

So far these are the 2 options i'm looking at to provide necessary protection to the extended tails -
are there any options to use 25mm cable as I don't think 25mm fits into most MCBs and neutral bars unless you know of a brand that would.

Many thanks

 

[newfutile]

I would not use an rcd or rcbo , which means it will have to be armoured cable fed from a dedicated switch fuse with isolator.
if they are buried in a wall you need a 30 ma RCD for additional protection or earthed mettalic screen .

 

[buzzlightyear]

b type would do.option 2

Schneider Acti9 63A Double Pole 10kA B Curve MCB. Co in side with swa cable.​

 

[richardg]

b type would do.option 2

Schneider Acti9 63A Double Pole 10kA B Curve MCB. Co in side with swa cable.​

The Schneider would do the trick - have you installed 25mm into the output side before?

 

[telectrix]

if he's using a 63A MCB, it's pointless to use 25mm SWA. 16mm will handle more than 63A all day long.

 

[buzzlightyear]

it's pointless to use 25mm SWA

it will be like trying to put a python head in a small box and still struggling to put the tails in .lol.

 

[Dustydazzler]

Do what the last sparks did on a house I’m working on. recently relocated a board due to an extension.
wired in 2 x 2.5 twin a earth connected in parspell directly from the meter then under the suspended floor , along about 15-16 metres , then chased up the wall into a new 18th ed board.

lovely jubbly

 

[123]

Neither option. If no mechanical protection you'd need 30mA RCD. If mechanical protection an MCB up front isn't great either.

SWA into a metal clad fused switch for me. Like this one

 

[radiohead]

I am somewhat surprised and concerned at your option one, which is very obviously non compliant.

 

[richardg]

I am somewhat surprised and concerned at your option one, which is very obviously non compliant.

Thanks for pointing that out Radiohead _ I believe I missed including a OCPD in the brief for option 1.

(From my understanding 411.3.2.3 states 5s disconnection is acceptable for a distribution circuit, without mechanical protection the supply to the sub main would require RCD protection and a MCB.)

123 not sure why you mentioned MCB up front isn't great? Even with SWA you cannot use DNO fuse as your cutout and need to supply a overcurrent device?

Many thanks

 

[DPG]

Thanks for pointing that out Radiohead _ I believe I missed including a OCPD in the brief for option 1.

(From my understanding 411.3.2.3 states 5s disconnection is acceptable for a distribution circuit, without mechanical protection the supply to the sub main would require RCD protection and a MCB.)

123 not sure why you mentioned MCB up front isn't great? Even with SWA you cannot use DNO fuse as your cutout and need to supply a overcurrent device?

Many thanks

I think 123 means why you would use an upfront MCB rather than an upfront fuse.

 

[pc1966]

The problem with an up-front MCB is poor selectivity with any MCB/RCBO in the CU. Basically any fault that hits the magnetic trip point will take out the whole supply, and for a 63A C-curve MCB that can be as low as a 0.32kA fault.

Unless you can be sure the tails are not going to be < 50mm from any surface you end up needing RCD protection at 30mA, which again is a selectivity problem.

So the best option is generally a fused-switch feeding SWA, has better selectivity on over-current faults and no need for up-front RCD (unless a TT system).

If going for 63A or 80A fuse / overall limit then 16mm SWA ought to be fine.

 

[richardg]

The problem with an up-front MCB is poor selectivity with any MCB/RCBO in the CU. Basically any fault that hits the magnetic trip point will take out the whole supply, and for a 63A C-curve MCB that can be as low as a 0.32kA fault.

Unless you can be sure the tails are not going to be < 50mm from any surface you end up needing RCD protection at 30mA, which again is a selectivity problem.

So the best option is generally a fused-switch feeding SWA, has better selectivity on over-current faults and no need for up-front RCD (unless a TT system).

If going for 63A or 80A fuse / overall limit then 16mm SWA ought to be fine.

That’s great thank you for clearing that up - in all honesty I never thought there would be selectivity issue between a 63A MCB on the supply end and the rcbos in the consumer unit - it’s a good day when you learn something new - thank you

 

[pc1966]

For overload there is no real selectivity problem between MCB if you have something like a 1:1.6 ratio (say 40A & 63A etc) as the down-stream thermal trip will always go first.

The problem with the magnetic trip is often both MCB's coils can be underway to tripping before either disconnects, so they both fire on the fault. Sometimes you can get acceptable behaviour depending on the chosen MCBs and circuit Zs values, and companies often have tables of selectivity limits for MCB/MCB or MCB/RCBO, etc to allow you check this.

You get fancy MCCB with electronic adjustable trip curves that have a short delay option to give you better selectivity, but they cost far more than a fuse-switch and really are not justifiable for domestic work.

 

[Rockingit]

You get fancy MCCB with electronic adjustable trip curves that have a short delay option to give you better selectivity, but they cost far more than a fuse-switch and really are not justifiable for domestic work.

I had to do just this on a job recently - had to put an upfront adjustable ELCB/MCCB inbetween the incomer and a Ryefield..... £800 to Mr Schneider later....

 

[pc1966]

Before heading off for the night, was going to add a last comment that if you ever find no reasonable option other than a MCB feeding a sub-main, say it is coming out of an existing DB or similar, then you generally want the supply MCB to have as high a trip point as you can get away with.

Typically that means going to a D-curve breaker and as high a current as you can meet either both:

• sub-main overload protection (if load not otherwise limited)
• end of sub-main Zs for 5s disconnection

Appendix B of the on-site guide is your friend here, as you can see 5s Zs for MCBs on page 131, or for a fused switch on pages 128 (industrial BS88-2 style fuses) or 129 (domestic style fuses BS88-3 similar to BS1361 on following page). Fuses are usually better as well for this, for example, some Zs values:

• 63A BS88-2 fuse = 0.62 ohm
• 63A BS88-3 fuse = 0.55 ohm
• 63A D-curve MCB = 0.28 ohm

You don't see 5s values for C or B curve MCBs as they reach the "instant" trip before the thermal curve gets down to 5s. While selectivity varies a lot from make to make and model to model, taking a couple of examples from the Hager commercial catalogue Jan 2020 that I happen to have on hand:

• 63A BS88 fuse feeding 32A B-curve RCBO fault limit = 1.5kA (page 133)
• 63A D-curve MCB feeding 32A B-curve RCBO = 0.80kA (page 136)
• 63A C-curve MCB feeding 32A B-curve RCBO = 0.51kA
• 63A B-curve MCB feeding 32A B-curve RCBO = 0.27kA

So you see the fuse has both twice the selectivity of the D-curve MCB and usable with double the Zs!

Also you see for the B and to some extent C-curve MCB feeds the fault level needed to trip it is easily managed by a typical RFC or modest length radial circuit.

 

[richardg]

Before heading off for the night, was going to add a last comment that if you ever find no reasonable option other than a MCB feeding a sub-main, say it is coming out of an existing DB or similar, then you generally want the supply MCB to have as high a trip point as you can get away with.

Typically that means going to a D-curve breaker and as high a current as you can meet either both:

• sub-main overload protection (if load not otherwise limited)
• end of sub-main Zs for 5s disconnection

Appendix B of the on-site guide is your friend here, as you can see 5s Zs for MCBs on page 131, or for a fused switch on pages 128 (industrial BS88-2 style fuses) or 129 (domestic style fuses BS88-3 similar to BS1361 on following page). Fuses are usually better as well for this, for example, some Zs values:

• 63A BS88-2 fuse = 0.62 ohm
• 63A BS88-3 fuse = 0.55 ohm
• 63A D-curve MCB = 0.28 ohm

You don't see 5s values for C or B curve MCBs as they reach the "instant" trip before the thermal curve gets down to 5s. While selectivity varies a lot from make to make and model to model, taking a couple of examples from the Hager commercial catalogue Jan 2020 that I happen to have on hand:

• 63A BS88 fuse feeding 32A B-curve RCBO fault limit = 1.5kA (page 133)
• 63A D-curve MCB feeding 32A B-curve RCBO = 0.80kA (page 136)
• 63A C-curve MCB feeding 32A B-curve RCBO = 0.51kA
• 63A B-curve MCB feeding 32A B-curve RCBO = 0.27kA

So you see the fuse has both twice the selectivity of the D-curve MCB and usable with double the Zs!

Also you see for the B and to some extent C-curve MCB feeds the fault level needed to trip it is easily managed by a typical RFC or modest length radial circuit.

Thank you for the very in depth answer appreciated ?