I do not think it is wise to deliberately put people at increased risk of a bad shock to prove a point that in someone else's opinion not RCD protecting a socket should have attracted more than a C3 whenever it was last inspected, That is what is being suggested.

Mike

I do not think it is wise to deliberately put people at increased risk of a bad shock to prove a point that in someone else's opinion not RCD protecting a socket should have attracted more than a C3 whenever it was last inspected, That is what is being suggested.

I don't quite follow. I'm not suggesting that 30mA RCD protection is omitted - just that where is it deemed necessary (according to local risk assessment - which may or may not align with codebreaker type guides), it could be seen as being better to provide it as part of the fixed installation rather than a portable device. (Less likely to be damaged, much easier to find & test every 3/6 months, can't be defeated by plugging in something else or using a different extension lead). This is all theoretical conjecture of course - we don't really know much about the OP's situation and whoever wrote the requirement might have had something quite different in mind.

Safe to say I've seen much more "curious" stipulations in some H&S policies....

   - Andy.

I think we are on the same page technically, but not approaching from the same end.

Under what conditions is it sensible to ban or remove RCDs inline with plugs or extension leads ?

1) when they cause a problem, false trips etc.

2) when, and only when, you can be sure, really really sure (!) , that the  device or lead will only ever be plugged into an already RCD protected socket.

As I said at the top, ideally they  should not be relied upon as the sole means of protection but that is not the same as demanding to cut them off willy nilly.

1 is clear cut

2) less so.
Firstly if you are dead certain there is already a working RCD in the line,  then you may not need them, but does it do any harm  to leave them in ?

Secondly, are we that sure there will always be an RCD upfront  if it is not a TT building ? The evolution of the regs and the time delay between changes and universal implementation is much longer than some authors on here imagine.

So a reg changes.

In the immediate year or two a few folk will carry on with projects already started and maybe the odd one who does not even know.  For a few more years the chancer types who do know really, but are happy to ignore it will offer cheaper jobs that use up old stock, especially at the shallow end of the cash job/ minor works market.

(some folk still have red and black T and E on part reels from nearly 20 years ago )

So I expect new socket circuits to be RCD protected, and most (but maybe even then not all) new additions to existing to have an RCD, but really a large fraction (maybe 75% plus of all wiring in use) will be pre-existing. These will only be updated for sure on an inspection fail that demands a change, so a C2 or C1 report, or in a private house when changing the CU.

I respectfully suggest that the average regs change to full implementation, the time delay should be measured in decades and perhaps quite a few of them.

As a non RCD but related real example, I live on a 1970s housing estate where none of the houses had lighting circuit CPCs when built, even though by then the regs requiring that had changed some years before. So, even now based on the neighbours I have visited, still less than half have been rewired to achieve this by 2022, though many who have new-ish plastic  CUs have an RCD for the lights and a 'no CPC' notice, so some nod to modern ways, safer than original,  but not compliant.  I do not live in some sink estate, or miles from civilization,  this is a well to do home counties market town, so I assume there are far worse out there.

Change is not measured in mere years.

And then there is the slight but non zero risk that the fully compliant RCD in the CU does not actually fire on demand, but for one reason or other sticks.

Still want to take the RCD plugs out of service ? I'd not want to justify that position.

Mike.

mapj1 said:

Still want to take the RCD plugs out of service ? I'd not want to justify that position.

Or similarly, when I change my CU (progress is being made!) should I replace the RCD sockets which go back to 15th Edn?

Chris, you will almost certainly have to remove any sockets with built in RCD protection, because if anyone presses a test button to test a socket RCD or just to turn the socket off I can almost guarantee the circuit RCD will trip in the consumer unit as well.

A bogof test button, nice. If I were Chris I wouldn't replace them until the tripping of the main RCD was verified. And even then I would consider whether it was a nuisance or not, Chris would probably go for all RCBO anyway, so the benefits of having two RCDs in series, thinking redundancy, might outweigh the nuisance.

It’s even more fun if the installation has an upfront 100 mA RCD and someone has fitted a SRCD socket with a built in RCD, then you need to swap the upfront 100 mA to a 30 mA.

ok, I read this with some interest. I'm not an electrician, but I do have a fairly good understanding of things. 

The title of the thread almost sounds like something I would put in place, but not as a fixed rule. I don't use the words 'banned' or 'forbidden' I tend to say 'the company standard is'.

Having consistency is a big benefit, and making rules that reduce the amount of ongoing maintenance or inspection is always good, as is reducing the amount of confusion.  

Making it a company standard, rather than banning things also means you can have non-standard things where economics or circumstances get in the way, such as an area served with a BS3036 board, an SRCD is an appropriate solution, sometimes. 

I made our standard preclude the use of SRCDs, as they are expensive, unreliable, easily damaged and don't protect the cable supplying the socket if someone was to damage it. The cost of replacing two SRCDs is higher than fitting a single RCBO to a modern board. 

One of our sites has bought plug in RCDs and has put them on every appliance, I shudder to think about the cost of that, or the practicalities of it, plus the PAT test that will be added and then, of course, the RCD part wont be getting tested.

I was a bit concerned about the type A and AC RCDs, but having just checked my stores, all the Schneider RCBOs are type A, but I have a Dorman Smith RCBO, and that is type AC, slightly bizarrely, the Dorman Smith RCBOs fitted to the board some 15 years ago are type A. 

I think I just need to check that they are type A when I purchase them in the future. 

I just thought it might provide a bit of insight as to where these rules come from

Thank you, Sparkingchip, I hadn't thought of that.

Pressing the test button every 6 months wouldn't be a problem, but we also have an RCD protected FCU which supplies the pump of a small fountain. In effect, the test button is the "off" switch so that would be operated on a daily basis.

And yes, the CU will be all RCBO (and will probably incorporate an SPD).

Under what conditions is it sensible to ban or remove RCDs inline with plugs or extension leads ?

Perhaps add:

3) Where (a single layer of) additional protection is needed and you want it done in the best (least unreliable) way available.

From a H&S perspective the can be a danger of having too obvious "backup" protection - i.e. people tend to start to rely on it. I remember a H&S course years ago where the presenter said he had 5 lights in one of his rooms - one day there was a pop and the room was left in darkness. What happened he asked? One lamp blew and the MCB tripped suggested most of us with any electrical knowledge. No he said - MCB was still on. A few puzzled expressions. He explained that a few ears earlier all five lamp where working, one blew, but he didn't bother replacing it as there was still plenty of light from the remaining 4, Later another blew, but three was still more than enough light and it was a lot of effort to get the step ladder out, when it got to two left he started thinking that he should do something about it but didn't quite get around to it, when there was only one left it moved up the metal priority list as it were, but as there was still enough light for what was needed he still hadn't quite got around to replacing any of the lamps before the 5th eventually blew.

I can see an (avoidable) risk that if someone had a plug-in RCD to use they won't bother checking that the wall socket is RCD protected.

Secondly, are we that sure there will always be an RCD upfront  if it is not a TT building ? The evolution of the regs and the time delay between changes and universal implementation is much longer than some authors on here imagine.

This is a workplace policy - so presumably the authors are aware of the situation on their particular site, and the procedures they expect their staff (both electrical and 'ordinary') to follow. They may have all sockets RCD protected or have some means of identifying which are.

If I were in that situation - say writing a procedure for use of electrical equipment outdoors, I think I'd be happier saying something like "only plug into one of the sockets marked 'for equipment outdoors'" knowing that the RCD behind that was reasonably well protected and subject to regular checks & tests, rather than allowing the option to plug into any socket you like but use a plug-in RCD that's might have been collecting dusk in the back of a filing cabinet for 10 years or been bouncing around in the back of someone's van or was left out in the rain yesterday.

Certainly, personally, I'd be more than happy to have a plug-in RCD in addition to RCD protection in the installation, but I'm the kind that had enough shocks during my miss-spent youth to be pretty respectful of electrical dangers. But most people will just to the absolute minimum the procedures demand of them - often not even that if they're seen as complicated or time consuming - so simple and reliable perhaps trumps the ideal.

    - Andy.