Bathroom RCD protection

[Jm1980]

Looking at a bathroom modification and wouldn't mind some input on the need for RCD protection.

Bathroom has existing light and wall fan. Customer wants downlights fitting and ive also said the fan needs an isolator as it currently doesn't have one. No other electrical circuits in the room. No supplementary bonding is present either and as the room is nearly finished, adding any is difficult.

Firstly, all work is outside the zones so my understanding is that part P doesn't apply?

The circuits obviously need RCD protection as per 18th and also as i have no supp bonding present. CU is a split load 16th edition style with upstairs and downstairs lights on the non protected side. The option to run a new bathroom dedicated circuit isn't feasible due to difficulty of run and customer not wanting this. As i see it then that leaves a couple of options.

1, switch the upstairs lights onto the RCD side at the CU thus protecting the entire circuit and thus the bathroom. This seems a good option although I'm risking any faults already present giving me RCD issues (borrowed neutrals and the like).

2, An easier option seems to be to run the lighting circuit into an RCD fused spur and then from this create my bathroom lighting and fan wiring. I see conflicting opinions on the RCD fused spur option though especially since upgrade to the 18th.

Any opinions appreciated.

 

[Jm1980]

my thoughts exactly, the BS7288 reference you linked Ian doesn't really make much sense to be honest, basically saying you can use them but its pretty pointless because it wont be doing anything that the devices we demand upstream wont be doing.

 

[DPG]

It really does need clarifying in plain and simple terms. I can't believe we are still in this situation.

 

[Jm1980]

Its annoying as it looks like i will have to RCD at the CU end to include the entire lighting circuit which means I'm risking running into pre existing faults that have to be explained to customer.

Who would be happy to use an RCD spur noted as a departure on the cert given that I'm using the RCD as a means to ommit the need for supplementary bonding?

 

[loz2754]

Its annoying as it looks like i will have to RCD at the CU end to include the entire lighting circuit which means I'm risking running into pre existing faults that have to be explained to customer.

Who would be happy to use an RCD spur noted as a departure on the cert given that I'm using the RCD as a means to ommit the need for supplementary bonding?

You could fit an enclosure with an RCD instead of an RCD FCU.

 

[brianmoooore]

If the existing lighting circuits have such a low IR that they would trip an RCD, then that is a C2 fault.
If there is a shared neutral, that is a C2 fault.
I would explain to your customer that fitting the required RCD protection to the bathroom circuits could potentially show up pre existing potentially dangerous faults, that they will need to pay to have sorted.

 

[Jm1980]

Fitting an enclosure outside the bathroom with an RCD in it did cross my mind but aesthetically this wouldn't look very good and it would be a strange setup to have an RCD enclosure half way along a lighting circuit.

In an ideal world Brianmoooore, the entire circuit i would RCD at source and rectify any faults but the loft above has been converted and a chipboard floor layed so i could really be opening a can of worms messing with the entire lighting. You know what customers are like when your there to fit a bathroom light yet trying to convince them you have good reason to have every bedroom rose down and switches off looking for a fault whilst charging them extra.......

 

[loz2754]

Just do an IR test on the lighting circuit and test for borrowed neutrals (more often than not it's a borrowed line at the 2 gang switch for the stairs).
Then you'll know whether an RCD would trip or not.
That's even if there are 2 or more lighting circuits.

 

[Dodd 1]

Basically the IET made a MASSIVE mistake MASSIVE, by not including bs7233 in reg 531.3.6, this is why when the new amendment that comes out in march 2022 will put it right. So the departure they want us to include on our certs is due to there MASSIVE mistake. Carry on using just include there mistake (we should get a refund for this one)

 

[Aaron b]

Basically the IET made a MASSIVE mistake MASSIVE, by not including bs7233 in reg 531.3.6, this is why when the new amendment that comes out in march 2022 will put it right. So the departure they want us to include on our certs is due to there MASSIVE mistake. Carry on using just include there mistake (we should get a refund for this one)

Sparky ninja posted a blog about this issue. It doesn't sound like a mistake or something that will be changed. If it doesn't provide the correct isolation we shouldn't use them as it doesn't give the same level of safety. Please correct me if I am wrong and they are now proposing this for amendment 2.

 

[Dartlec]

Sparky ninja posted a blog about this issue. It doesn't sound like a mistake or something that will be changed. If it doesn't provide the correct isolation we shouldn't use them as it doesn't give the same level of safety. Please correct me if I am wrong and they are now proposing this for amendment 2.

The biggest problem is that there is no one in a suitable position of authority willing to make a clear unambiguous statement on when and when not they should be used. Instead we get carefully worded statements that are more concerned with covering themselves, while not upsetting the manufacturers who make these products, so that they can't be used against them in a court of law.

Either they can be installed in situations where there is no current RCD protection to provide that protection (in which case they should be listed somewhere in the Regs with a clear definition of when it is safe to do so), or they cannot - in which case there is literally no point whatsoever in a 30ma version existing. (perhaps a 10mA one would have some usage where 30mA is already present for additional protection)

The NICEIC's view (for what its worth) does seem to be that they are perfectly fine to install to cover a new small installation such as the one OP describes, as long as they are noted as a departure from 7671, because in their view they meet a suitable level of protection.

There are clearly situations where they are a very handy method, so it's ridiculous that the standards haven't been designed to ensure that an entirely 'official' solution exists.

 

[Aaron b]

Protecting a whole circuit has got to be a better solution most of the time and the srcd's aren't cheap either.

The op should be testing the circuit regardless and issuing the customer with a certificate of tests confirming it is safe for the alteration.

 

[Aaron b]

While doing some CPD, GN8 amd. 3 has a section in it suggesting their use:

Section 8.7 "Low voltage socket-outlets (for example, 230 V) may be installed not closer than 3 m horizontally from the boundary of zone 1, provided they are protected by an RCD having a rated residual operating current, IΔn, not exceeding 30 mA, in accordance with Regulation 415.1.1. In cases of retrofit, this requirement could be met by the provision of a socket-outlet with an integral residual current device, as shown in Figure 8.6, or, alternatively, the circuit could be fitted with RCBO protection."

It is a poor situation, you would of course need an attractive big yellow label too ?