Submain configuration

Post Reply
JamieP
Posts: 478
Joined: Tue Apr 14, 2020 11:08 am
Answers: 0
Has thanked: 92 times
Been thanked: 18 times

Submain configuration

Post by JamieP »

I have an existing submain coming from a MSB to a DB

What I have been asked to do is to extend this submain to another board, so having the 2, supplied via the one submain, which I can't see any issue at all

But my issue is rather than continue the submain from the line side of the DB, they have asked me to take it from the load side of the DB isolating switch, which has confused me as now it seems like rather than a single submain I wonder if it should be considered a second one from this DB and although the upstream device affords over current protection I still wonder if I need some sort of device locally

Is this a compliant method? Am I best to just push for connection to the line side of the DB isolating switch?
AlecK
Posts: 912
Joined: Thu Apr 16, 2020 11:24 am
Answers: 5
Has thanked: 2 times
Been thanked: 351 times

Re: Submain configuration

Post by AlecK »

Nothing wrong with having a common submain supplying 2 (or more) DBs.
Also nothing wrong with having them cascaded; however in that case you must deal with discrimination.
Also note that, except for large submains to outbuildings as per 2.3.4.1 (a), a DB isn't required to have any isolating device on the incoming supply.

But this request appears to be neither fish nor fowl, it's neither truly common submain nor truly cascaded.
I can't think of any good reason for having the 2nd part of a common submain controlled by the first DB it supplies.
On other hand, a good reason not to do it is it will confuse people unless very carefully labelled.

Ok; that's an initial reaction.
Now for the application of Rules.

Definition of submain 1.4.89.
Pay particular attention to the words specifying the origin, and the termination, of a submain.
Bearing in mind that the way the N is treated is just as important as, if not more important than, the way any control / isolation switching or overcurrent protection is done.

A true common submain will have either tee-offs to each DB it supplies, else the submain must be looped in & out;
and all terminations done in such manner that continuity of A,N & E to DB2 cannot be broken by disconnecting any connections within DB1.

Active(s):
If the 1st length of a submain is terminated unto an isolating switch, that controls the 2nd part; then by definition that is the termination of a submain.
And therefore the run to DB 2 must be a different submain.
That 2nd submain then requires overcurrent devices at it's origin. Even if it's the same sized conductors; it's a new submain 'circuit", and every circuit must have overcurrent protection at its origin.

Neutral:
Unless the incoming & outgoing submain Ns are terminated in same terminal (or equivalent config); then again it's not 1 common submain but 2 cascaded.

Same for earthing.

For submains with PEC, I'd accept the incoming / outgoing conductors being in same terminal.
But for PEN submains, it must not be possible for the ongoing connection of the PEN to be broken at DB1. This then requires being eg crimped in common lug or otherwise permanently connected together, which is then connected to DB1 busbar [5.5.3.1(b)(v)].
JamieP
Posts: 478
Joined: Tue Apr 14, 2020 11:08 am
Answers: 0
Has thanked: 92 times
Been thanked: 18 times

Re: Submain configuration

Post by JamieP »

AlecK wrote:
Tue May 04, 2021 3:51 pm
else the submain must be looped in & out;
and all terminations done in such manner that continuity of A,N & E to DB2 cannot be broken by disconnecting any connections within DB1.
I thought this was only a case for PEN conductors when submains used the outbuilding earthing clause, I can't find anything in relation to them having to be terminated in such a way that continuity must remain if disconnections at DB1

Isn't it ok just to have them in the same terminal if supplying two boards from the same submain?
AlecK
Posts: 912
Joined: Thu Apr 16, 2020 11:24 am
Answers: 5
Has thanked: 2 times
Been thanked: 351 times

Re: Submain configuration

Post by AlecK »

Yes this requirement is specifically stated for PEN conductors passing through one DB on their way to another.

And true it's not stated directly for common submains that similarly pass through an intermediate DB.
But it's a result of the definition I mentioned - though maybe i wasn't clear enough WRT how the connections can be made.

What I was referring to was terminating the incoming & out-going conductors in different terminals; which would then make it easy to inadvertently disconnect one and not the other - the risk arising mostly around the outgoing N & E conductors.
Which of course should never be done without first isolating the associated A(s).

If you use tee-off branches to feed into DB1; no problem.
Or if you put incoming & outgoing conductors into same terminals, all OK.

But if the incoming conductors of first "section" of submain, eg MSB to DB 1, are terminated into different terminals than the outgoing conductors of the second section; then - by definition - it's two different submains (cascaded) and not a common submain at all.
Which is perfectly OK; but means you need overcurrent protection & at DB1 (origin of the outgoing submain).
And have to consider discrimination - which is usually why you'd want to use a "common submain" approach in the first place, so as to not have too many steps in the discrimination series.


The PEN scenario [5.5.3.1(b)] goes even further; and forbids simply 2 N wires in / on same terminal of the intermediate DB.
It requires the incoming & outgoing PENs to be joined together (eg into same lug , or equivalent , as long as it's not a "terminal of the switchboard"), so that it is absolutely impossible to separate them just by disconnecting at the N-Bar.
JamieP
Posts: 478
Joined: Tue Apr 14, 2020 11:08 am
Answers: 0
Has thanked: 92 times
Been thanked: 18 times

Re: Submain configuration

Post by JamieP »

That's exactly what I thought, just the way you worded the part I highlighted made me feel you were implying that they did infact need to be done in a similar way to the PEN example

Thank you
Post Reply