DB board - no isolation

[burntwire]

Hi, can anyone tell me in what edition of the regs or year it became necessary to have a DB with an isolation switch. Im guessing along time ago!?!

Reason is I tested a restaurant, extension built around 20 years ago and has an old crabtree board (c50, i think!) with no main switch and need further info on this,

Thanks

BW

 

[333rocky333]

does such a reg exist, in commercial work it was quite common for a board to have no switch and often fed by a seperate main switch fuse

 

[RF Lighting]

It tends to be installed as standard these days, but I'm not aware of any regulation which says DB's must actually have an isolator.

There is a requirement to have a isolator at the origin of the installation (which only needs to isolate the line conductor(s) in a non-domestic property).

On an EICR, it wouldn't even get a mention from me.

 

[streetlighter]

Seen this a few years back in a nursing home a 3-phase DB the old type with the redspot fuse carriers.
It was pulled up on the PIR for having no local isolation and a MEM excel isolator retrofitted onto the board.

 

[Adam_151]

On an EICR, it wouldn't even get a mention from me.

Even if it was a 3036 re-wireable board and was itself feed from a set of 63A ways in in a HRC fuseboard?

 

[RF Lighting]

Which regulation(s) does it contravene?

 

[securespark]

I think in the 1st Ed. it said thou shalt have a means of isolation at the origin but I can't remember the exact wording.

EDIT: As if by magic, my memory enlightens me!

If the pd within any house exceeds 200V, the house should be provided with a "switch", so arranged that the supply of electricity can be at once cut off.

There must be a modern equivalent???

 

[EFLImpudence]

537.1.4

 

[JohnW2]

537.1.4

It's interesting that this reg says that the switch/isolator should be "as near as practicable to the origin of the installation".

Does this mean that very many domestic (and some other) installations which don't have an isolator upstream of the CU/DB are technically non-compliant, since it would usually/often be 'practicable' to install a switch/isolator nearer to the origin than in the CU/DB? !

Kind Regards, John

 

[EFLImpudence]

537.1.4

It's interesting that this reg says that the switch/isolator should be "as near as practicable to the origin of the installation".

In most cases it would be, wouldn't it?
Either because the CU will be near the cut-out or if not then a switch-fuse will be there,

Does this mean that very many domestic (and some other) installations which don't have an isolator upstream of the CU/DB are technically non-compliant, since it would usually/often be 'practicable' to install a switch/isolator nearer to the origin than in the CU/DB? !

It depends on the definition of 'near' and, indeed, 'installation'.

However, you have digressed to discussing the position of a double-pole main switch as per 537.1.4.

The thread was merely asking should/must there be one.
Does not 537.1.4 state that there must?

 

[JohnW2]

It depends on the definition of 'near' and, indeed, 'installation'.

It doesn't just say 'near', it says, 'as near as practicable'. As I said, it would usually be 'practicable' to install it upstream of the CU/DB (indeed, there will not uncommonly be an isolator in that position). I agree that one could debate where the 'origin of the installation' is.

However, you have digressed to discussing the position of a double-pole main switch as per 537.1.4. The thread was merely asking should/must there be one. Does not 537.1.4 state that there must?

I'm not sure that I've really digressed that much. The OP was actually about whether there should/must be an isolator within a DB. As people have said, I'm not aware of any reg that requires that. As you say, 537.1.4 says that there must be a switch/isolator, but it certainly doesn't say that it has to be in the DB/CU. On the contrary, it requires that such a switch/isolator be positioned 'as close as practicable to the origin' - which, as I said, often theoretically might be upstream of the DB/CU (depending upon what point one regards as 'the origin').

Kind Regards, John

 

[plugwash]

Even if it was a 3036 re-wireable board and was itself feed from a set of 63A ways in in a HRC fuseboard?

Which regulation(s) does it contravene?

AIUI the regs require installations to be separated into circuits to avoid danger and minimise inconviniance in the event of a fault. Surely part of avoiding danger and minimising inconviniance in the event of a fault involves not just the automated response to the fault but the manual response to the automated response.

Ideally when replacing fuses one would remove the supply so there is no power to the fuse being replaced. This is possible if the other circuits fed by the DB can reasonablly be isolate without causing too much inconvenience.

If that is not possible then removing all significant load will reduce (but not eliminate) the risk of something bad happening when removing and reinserting the fuse. Afaict this is what the DNOs do when working on service fuses. They are also supposed to use PPE so they are sheilded in the event that they get it wrong or there is a fault and the fuse does blow up in their face.

However with the two level board scenario with fuses at both levels and only a single isolating switch covering everything things get nasty if the fuse feeding the sub-board blows. It's likely to be highly inconviniant to isolate upstream (given that the upstream board is likely a major board) and tedious and error-prone to isolate downstream (many individual isolators on bits of equipment to turn off individually)

On the other hand similar problems would apply if a socket circuit with lots of sockets is fed directly from a major fuse board.

Ultimately I guess this needs to be risk assesed in the individual case. What faults are likely to happen and is there a way of resolving them safely and with acceptable levels of disruption.

 

[securespark]

537.1.4

Thanks!

I thought it was in there.

Having said that, I can't find it in the 16th, or 15th, but it is in the 14th, right in the front of the book: A1 and A2!

 

[Lectrician]

There has always been a requirement to have an isolator for the "installation". If you used a DB that had no integral isolator connected directly to the origin of the installation, you would have to install a separate isolator before it.

If the DB was a submain from another board, or from a switch fuse at the origin, there is no requirement for a local isolator in or near the DB. These days, you would generally fit one as standard, although, TP DB's do come with various incomer configurations, one of these being a "direct lug kits".

 

[JohnW2]

Ideally when replacing fuses one would remove the supply so there is no power to the fuse being replaced. This is possible if the other circuits fed by the DB can reasonablly be isolate without causing too much inconvenience.

Getting back on-topic, what the OP didn't tell us is whether the DB in question is fed from a dedicated isolator, switch-fuse or whatever. If it is, then the only question is whether the regs are unhappy about the means of isolation being external to the DB, rather than within in - and I think everyone who has commented has agreed that the regs do not appear to specifically require a within-DB isolator.

As you say, if the only menas of isolating the DB would involve also isolating other DBs, or other parts of the installation, then other consdierations (such as you discuss) come into play - but we don't know whether that it is the situation with the installation described nby the OP.

Of course, if there is no means of isolation (deidcated or otherwise) upstream of the DB, then, as has been said, it would be non-compliant with 537.1.4.

Kind Regards, John