Peak And Off Peak Submain Isolation

[chris_k]

Evening.

Ive got a job to price and im after a soloution.

I have a flat to wire that is going to have all electric heating / water. Its an upstairs flat with the service head / meter downtstairs.

The head has been installed already but we are waiting for the meters. The distance from the head to the proposed consumer unit location in the flat is approx 18-20m. with a torturous route up walls,  through floors and ceilings.

The spec is to have 2 storage heaters and water heater on econ 7 and the rest panel heaters.

Whenever ive done work with systems with storage heaters they have their own board running off the off peak supply. These have always been older properties with older timeclocks. I have seen a couple of newish properties that have only 1 cu but has econ 7 and the whole house turns to cheap rate after 1am or whatever it is. Whats the norm nowadays with newbuild and econ 7?

If i have to have 2 cu`s then i will need at least 4c from the meter to the cu`s. Due to the route to be taken it will have to be SWA and because of length will have to have isolation and fusing down for discrimination against the 100a fuse.

What isolation / switching would you go for?

Ta Muchly

 

[ProDave]

The norm is a dual rate meter with a built in time switch.  Gone are the days of separate meter and time clock.

So you want to run TWO submains from the meter point to the two cu't in the flat.  Each will need over current protection and isolation so two switch fuses would do the job.  Make sure the cable run / type does not need RCD protection.

The whole house gets metered at the cheap rate at night, but only the off peak CU is actually energised.

 

[D Skelton]

I normally use hi-integrity DBs with a four pole main switch up front for dual tarrif supplies. Dual RCD for normal circuits and a third RCD protecting the off peak circuits. The same principle would apply if I was protecting this as a submain, a four core SWA (armour as the cpc) and a TP switch fuse utilising two of it's fuses. The thing is, both supplies need a single point of isolation really, a TP sw/f kills two birds with one stone as opposed to having a TP isolator and then two SP sw/f.

 

[ProDave]

A "normal" instalation with the two CU's right next to the dual rate meter / time switch has a separate isolator for each CU and never normally has a "single point of isolation"

 

[D Skelton]

A "normal" instalation with the two CU's right next to the dual rate meter / time switch has a separate isolator for each CU and never normally has a "single point of isolation"

And such 'normal' installations don't comply with the requirements of 537.1.4.... Technically

 

[SPECIAL LOCATION]

And such 'normal' installations don't comply with the requirements of 537.1.4.... Technically

Another myth.... 

I think you will find it only mentions isolating installations...  NOT buildings...

An electrical installation is just an assembly of associated electrical equipment having co-ordinated characteristic to fulfill a specific purpose... 

(you can find that in definitions page 27)

One building can have more than one installtion..

One supply can energise more than one installation...

People often assume that just because everything is in the same building it must all be the same installation......

Bit like that confusion (common with 5WW short course idjuts) where they assume an 80A 30ma RCD gives 80A overload protection...

Or that a 100A main switch gives 100A overload protection as well!!!!!!

opcorn

 

[D Skelton]

Another myth.... 

I think you will find it only mentions isolating installations...  NOT buildings...

An electrical installation is just an assembly of associated electrical equipment having co-ordinated characteristic to fulfill a specific purpose... 

(you can find that in definitions page 27)

Not quite. You are quite correct that it only mentions installations, a point I have discussed many times with others as to the abiguity of the word with regards to electrical work. The common definition of an electrical installation, one which many of us accept but is rightly open to interpretation, is the definition found in section 2 of the BGB and the one which you have mentioned. This is widely interpreted however to mean from the point of supply onwards. This is certainly taken as the interpretation from more than one supply company anyway.

One building can have more than one installtion..

Yes it can. In my mind, this is true if they are seperately supplied, seperately metered and seperately distributed.

One supply can energise more than one installation...

This is where we disagree. From one supply comes one installation. Although one installation can have multiple sources of supply.

People often assume that just because everything is in the same building it must all be the same installation......

Nine times out of ten it is however. Take the case of main protective bonding in a building of multiple occupancy (a block of flats for example). There is a good argument that each flat is it's own installation where bonding is concerned, on the flip side however, if each flat is centrally distributed and metered from one supply, there is a good argument that the whole block is one installation. The fact is, the regulations leave this matter widely open to interpretation. GN8 goes some way to make sense of it, but this is guidance only. The point is, we could argue until the cows come home on the finer points of the definition of the phrase 'electrical installation' but as far as the regs are concerned, with regards to the OP's situation where we have one supply, which supplies one installation that is seperately metered, they are clear. A main linked switch or circuit breaker shall be provided as near as practicable to the origin of the installation as a means of switching the supply. That is 'the' supply, not supply A but not supply B, THE supply. Which I interpret to mean no matter how many sources of supply an installation has, a main linked switch shall be able to make dead, in one throw, the whole lot.

If you were to take your interpretation literally, you could divide your average domestic installation an infinite amount of times. Also, I've never seen anyone omit a main switch fuse before a busbar chamber citing your interpretation as a get out clause. That said, even GN2 would seem to support your interpretation, but in my opinion, that support is out of nothing more than an interpretation of a word.

This is definitely an interesting argument though, and one that we will all continue to debate for all eternity until someone comes along and actually clarifies it in the regs! lol

 

[batty]

I normally use hi-integrity DBs with a four pole main switch up front for dual tarrif supplies. Dual RCD for normal circuits and a third RCD protecting the off peak circuits. The same principle would apply if I was protecting this as a submain, a four core SWA (armour as the cpc) and a TP switch fuse utilising two of it's fuses. The thing is, both supplies need a single point of isolation really, a TP sw/f kills two birds with one stone as opposed to having a TP isolator and then two SP sw/f.

Would armoring be suitable to take fault currents on a TNCS supply?

 

[D Skelton]

You'd need to use a couple of adiabatic equations to confirm that, but on a standard domestic TN-C-S supply, 999 times out of 1000, yes, it will.

 

[D Skelton]

There are also charts that are easier  :innocent

Or that yeah