A contractor has provided the single-phase electrical installation in 12 new, very small, individual commercial units. For whatever reason he provided two 6way distribution boards one appears to be for lighting and the like and the other for power, both with 30mA overall RCD protections both boards have a main switch. The tails for both boards are brought to a set of ISCOs from which he left a short tail connection for the meter. Now 4 of the units have been connected to the supply but apparently connection is being refused to the remaining units as no main switch has been provided. I guess different DNOs, different rules and indeed attitudes but I can find no reference to the need for a main switch in the DNO connection guide other than that the installation has to comply with 7671.
Main switch meaning singular, is in the model form checklist. UKPN also stipulate a main switch for overall disconnection in their manuals. No idea what area your in, but i guess other DNO's might follow suit.
Crazy having 2 small boards, must have got them for free.
Hi lyle, were the 4 connected up different to the others or were all units the same? It would be unusual to have more than one DNO involved on a relatively small site!
Ah, but not so unusual to have a different DNO approved sub contractor for the first set , with a less rigorous checking of the rules than the one making the most recent visits. I suspect the DNO and the book of rules have not changed any time recently, but rather the interpretation of what sort of installation it should be treated as, and so if it is sensible to connect or not.
After all BS7671 section 462 and thereabout could be read to indicate that there should be a single clear point of isolation at the origin, not multiple ones, but henley blocks and multiple consumer units are very common.
Ah, but not so unusual to have a different DNO approved sub contractor for the first set , with a less rigorous checking of the rules than the one making the most recent visits. I suspect the DNO and the book of rules have not changed any time recently, but rather the interpretation of what sort of installation it should be treated as, and so if it is sensible to connect or not.
After all BS7671 section 462 and thereabout could be read to indicate that there should be a single clear point of isolation at the origin, not multiple ones, but henley blocks and multiple consumer units are very common.
Thanks Mike, although it would draw into question the already connected units, either the person connecting the first batch had a bad day or the person who didn't want to connect them up had a bad day, who is wrong? Equally, presuming the installation certificate was in order, if the DNO refuse to accept it then it would suggest they are suggesting a fraudulent document has been issued?
Have new isolators supplied by the supplier with accessible terminals for the BS7671 compliant installation gone out of fashion?
Those isolators always confuse me a little. Is this a new demarcation point, beyond the consumer side of the meter? If so, who owns it. If the customer, then BS7671 applies and it should be non combustible.
If a contractor fits one... does it then become the DNO's property de-facto, or is the contractor in breach of the regs for a combustible switchgear enclosure?
This does need working out.
I understand the reason for insulated enclosures (TT systems, pre RCD....) but surely there could be a non combustible but non conductive enclosure (phenolic anyone?) that bridges the gap?
Personally I wouldn't accept that a stand alone isolator is a similar switchgear assembly to a DBO, and therefore I don't believe that any Regulation is breached by the installation of such.
FWIW, 537.1.4 has migrated to 462.1.201, but the words remain the same.
It is interesting Chris that 462.1.201 refers to a main linked switch or circuit breaker, both singular terms not plural. This device to be positioned as near as practicable to the origin of every installation. It must be double pole switching for ordinary folk for a single phase installation and be capable of switching the supply ON LOAD.
This "isolator" must be able to disconnect when carrying a full load current. So it is more than an "isolator" as defined on page 31 which can also be called a "disconnector." A "disconnector" is defined on page 27 in note 2 as being able only to be capable of opening or closing a circuit when either a negligible current is broken or made, or when no significant change in Voltage across the terminals of each pole of the disconnector occurs."
So the device is a "SWITCH, LINKED" as defined on page 37, even though it affords "isolation". So it is best not to call it an "isolator."
Those isolators always confuse me a little. Is this a new demarcation point, beyond the consumer side of the meter? If so, who owns it. If the customer, then BS7671 applies and it should be non combustible.
If a contractor fits one... does it then become the DNO's property de-facto, or is the contractor in breach of the regs for a combustible switchgear enclosure?
This does need working out.
I understand the reason for insulated enclosures (TT systems, pre RCD....) but surely there could be a non combustible but non conductive enclosure (phenolic anyone?) that bridges the gap?
The NICEIC told me:
In a domestic setting, an RCD in its own enclosure is considered 'similar switchgear' re BS EN 61439-3, so it should be a non-combustible enclosure.
'If' an RCD was installed upstream of a non-combustible CU on a TT installation, the enclosure would have to be non-combustible (even though I argued the RCD offers fault protection to the CU, so would just be shifting the problem of no fault protection to the RCD enclosure, so you'd need an RCD upfront of that, and so on).
An RCD upstream of the metal CU on TT is not needed anyway, so no need to worry about using a metal enclosure for the RCD because it wouldn't be there, and there's no need for the RCD because an insulated gland should be used for the 'double insulated' tails so the tails don't contact the metal enclosure at the point of entry, so won't touch the metal anyway (even though I argued that the RCD would offer fault protection for the metal CU 'if' in a split load board the tails or the single insulated linking conductors to the RCCB's inside the board ever became damaged and contacted the metal CU).
