Would it have complied to use 2 only RCD's in my house?

[Risteard]

Doesn't that mean that the regs, as well as 'us', probably do "get unnecessarily excited/concerned about RCD trips"?
Very true, but from the point-of-view of safety (rather than 'convenience'), loss of power (and hence lighting, assuming that it happens during the hours of darkness) for just 15 seconds or so could be enough to result in the postulated catastrophes involve pans of boiling oil, ladders, stairs etc - regardless of the cause of that loss of power.
... but not necessarily 'totally easily' in the case of N-E faults.

I have to say that, for reasons I've often discussed, I would personally be pretty comfortable with no RCDs at all. That's what I (and most other people of my vintage) lived with for the first 3+ decades of my life - and during the subsequent 3+ decades, during which I have 'enjoyed' RCD protection, I can't really think of any RCD trip I've experienced which resulted from a fault which would have represented a significant risk to life, health or property in the absence of RCDs. However, as BAS often used to point out, I could say the same about the car seat belts I've been 'religiously' wearing (since long before I was 'required' so to do) for around 50 years, which have so far not saved me from any significant harm.

Kind Regards, John

I've survived car crashes both with and without seat belts, so again it doesn't prove anything.

 

[ericmark]

I only have BS7671:2008 but can't see where it says at least 2 RCD's. It says

(iv) reduce the possibility of unwanted tripping of RCDs due to excessive protective conductor currents produced by equipment in normal operation

the supply should be split into circuits, but the only reference of two is where in series not parallel.

It does not actually differentiate between use of RCD and MCB to form circuits.

In times gone by it was common to have one 100 mA delayed RCD covering all the house. And it would rarely trip, only when we moved to 30 mA did we start to have problem with tripping for no apparent reason.

If the RCD trips for no apparent reason then I would say that was "normal operation" so the installation does not comply. And when it trips it is often inconvenient and it does state

(i) avoid hazards and minimize inconvenience in the event of a fault

So where a fault causes the RCD to trip, if that fault is not on the lighting circuit it should not cause the lights to fail as that would be inconvenient.

 

[JohnW2]

Does that make getting into the bathroom in the morning a pain?

No - at least, not in the bathroom 'zones', because RCDs aren't allowed there!

Kind Regards, John

 

[JohnW2]

... It says the supply should be split into circuits, but the only reference of two is where in series not parallel.

What on earth (in an electrical installation) are "circuits in series" - and where does BS7671 refer tothem?
It says

(iv) reduce the possibility of unwanted tripping of RCDs due to excessive protective conductor currents produced by equipment in normal operation

That presumably is saying that one should not have loads whose total expected N-E 'leaks' ("in normal operation") would be enough to trip an RCD protected by the same RCD (e.g. rooms full of PCs)?

It does not actually differentiate between use of RCD and MCB to form circuits.

We've discussed this before. Although you argue that an RCD trips because of "excessive (residual) current", I really don't believe that when BS7671 talks about "protection against overcurrent" it intends that to relate to a residual current - so, in turn, I don't believe that, per BS7671 definition, an RCD 'creates a circuit' (although, of course, it will usually be protecting one or more 'circuits' which already exist, per BS7671 definition, because of an OPD).

... where a fault causes the RCD to trip, if that fault is not on the lighting circuit it should not cause the lights to fail as that would be inconvenient.

Both 'minimising' and 'inconvenience' leave a lot of scope for judgement and discretion. For myself, I would say that the possibility that I might lose a lighting circuit, during the hours of darkness, once every few years is already so little of an 'inconvenience' that it needs no further 'minimisation'!

Kind Regards, John

 

[John D v2.0]

Very true, but from the point-of-view of safety (rather than 'convenience'), loss of power (and hence lighting, assuming that it happens during the hours of darkness) for just 15 seconds or so could be enough to result in the postulated catastrophes involve pans of boiling oil, ladders, stairs etc - regardless of the cause of that loss of power.

Yes I guess there's the avoid hazards and the minimise inconvenience part, and I'm thinking more of the inconvenience part there.
Since bs7671 covers installations with potentially reliable/reduandant power supplies, the hazard thing may be more relevant.

 

[JohnW2]

Yes I guess there's the avoid hazards and the minimise inconvenience part, and I'm thinking more of the inconvenience part there.

Fair enough, but I presume we would agree that matters of safety (i.e. 'hazards') are more important than matters of 'convenience' (without safety implications).

Particularly given that views about desired and expected levels of 'convenience' will vary a lot between individuals, I have to wonder whether it is necessarily appropriate for BS7671 to concern itself with issues of convenience which don't have safety implications.

Kind Regards, John

 

[John D v2.0]

well true but I suppose inconvenience could turn into a safety issue if someone does something inadvisable.
e.g. Now we don't use rewirable fuses in domestic because of misuse (bit of earth cable in the carrier), also need for lots of multi way adaptors is an inconvenience, but it would still be flagged on an EICR.
I think the boundary is blurred intentionally, so that people have at least some sockets available so they don't go doing anything to get the faulty ones back on.

 

[JohnW2]

well true but I suppose inconvenience could turn into a safety issue if someone does something inadvisable.

I think that may be stretching things a bit. Someone 'doing something inadvisable' could well be a safety issue if there was nothing 'inconvenient' going on - and there's a limit to how far we can (and, some would say, should) try to protect people from their own stupidity.

e.g. Now we don't use rewirable fuses in domestic because of misuse (bit of earth cable in the carrier)....

... that sounds like a reasonable suggestion, but I don't know whether it is really the reason 'why' we don't use re-wireable fuses in domestic installations. I would have guessed it was more a matter of 'convenience' and, probably more to the point, "because we have new-fangled gizmos which can be used instead of any type of fuses"!

... also need for lots of multi way adaptors is an inconvenience, but it would still be flagged on an EICR.

It might be mentioned by an inspector, but since it's really beyond the scope of BS7671, I'm not sure that it could actually be 'reported' ('or coded') as part of an EICR.

Kind Regards, John

 

[ericmark]

As to RCD in series, maybe wrong word, but looking at 100 mA feeding 30 mA and also with caravans and boats 30 mA feeding 30 mA. However point is it does not say you need two RCD's in a home, that is what people have interrupted it as saying.

As to earth leakage, 3.5 mA is the limit for normal equipment, and a RCD should trip between 15 and 30 mA. So the limit could be exceed with just 5 sockets, clearly not every item will be to the limit, but also having a 100 sockets on one RCD is clearly over stepping the mark. As to what can be considered reasonable hard to say.

This house has 16 areas/rooms 3 will not have sockets as contain bath or shower, so 13 rooms which includes the hall and landing. Not got the guide to hand, but seem to remember 4 sockets per room, and more in the kitchen, so reasonable to expect 60 to 70 socket outlets, many will have items plugged in 24/7, kettle, washing machine etc. My house splits the sockets over 6 RCBO's if you include the socket on the cooker supply, all together there are 14 RCBO's (13 in use) only time they tripped it was correct, the flat roof leaked so socket got wet. However old house with just two RCD's it seemed to go in batches, could last a year with out a trip, then would get 5 trips in a month, then another year, no faults found, as it if spike on the line, or what not a clue, and resetting one could cause the other to trip. Maybe it was their age, fitted in around 1993 when my lad started to study to be a radio ham.

As to danger yes clearly a danger, I have noted RCD tripped and reset it, and have not told my wife, as far as I knew it had just tripped, then find some thing in freezer which looks as if it has defrosted. Today the freezers show the highest temperature they reached during a power cut, if you notice it before opening a door which cancels the display, and with the RCD in garage or granny flat it can be an occupant of the house can't reset the RCD as no assess. So possibly no heating.

In the main the house was wired before we started using RCD's so no consideration has been made that for example it feeds a shed or outside light which may cause it to fail. Often there is simply no consideration, we fit a consumer unit with two RCD's and it really does not matter where they feed.

I had mothers house rewired, the kitchen had a mini consumer unit fed with SWA, the guy wiring the main consumer unit was going to take it from one of the pair of RCD in main unit until I said something. Even then it seems he did not use the normal split even for the bits he had rewired, it was more down to what was easy. So when the overhead supply to garage went down, lost all but kitchen sockets.

I returned to UK around 1992 and all I heard was 16th edition says this, and it was also the first time I had come across the RCD, it clearly could have saved lives specially when some one knocks a nail through a cable to hang up his coat, but that took out the 30 mA, 100 mA, 1 amp and 5 amp RCD the last three all with time delay, it took out a huge chunk of the site, but was why I fitted them at home. Two RCD's fed two old Wilex fuse boxes. And with them been fitted for around 27 years, I have got use to them. And likely did safe my son who would play with electrics, now an electrical engineer.

But when I fitted them no rules said I should, and in hind sight two was not enough, but fitting after the fuse box was not easy. Even today with CU being split the electrician often has no option but to fit just two, as the box is not designed to fit 3 or 4 unless using RCBO. But with the reduction in price of the RCBO it is not so cut and dried, clearly in a caravan we fit one RCD, so to say a home must have x number RCD's seems daft, it clearly must be down to size of the home, but I would still say for an electrician to fit just 2 in my house would be wrong, and the same with neighbours houses. It needs at least 3, maybe 4, so lights in any room are not on same RCD as sockets in same room.

 

[JohnW2]

As to RCD in series, maybe wrong word, but looking at 100 mA feeding 30 mA and also with caravans and boats 30 mA feeding 30 mA.

You appeared to be talking about circuits in series, not RCDs. I have no problem with the meaning of having two (or more) RCDs in series - but that then changes my question, since although I know (in a literal sense) what it would mean to have two RCDs 'in parallel', it would surely be a crazy thing for anyone to do, which would reduce safety and thereby create a potential risk/hazard (in fact, partially undermining the whole reason for having RCDs).

However point is it does not say you need two RCD's in a home, that is what people have interrupted it as saying.

Exactly ... and if one has a good argument as to why a single RCD (protecting the entire installation) would not represent a safety hazard (e.g. because emergency lighting was installed), then I don't thing it would be correct to interpret the regs as requiriung two or more RCDs.

As to earth leakage, 3.5 mA is the limit for normal equipment, and a RCD should trip between 15 and 30 mA. So the limit could be exceed with just 5 sockets, clearly not every item will be to the limit, but also having a 100 sockets on one RCD is clearly over stepping the mark.

Exactly - and that's where the designer's judgement comes into play. Since many domestic items have very low earth leakage, it's unlikley that that an RCD's trip threshold would be reached with quite a lot of sockets on a 'general domestic sockets circuit'. On the other hand, just half a dozen sockets in a school/whatever 'computer room' could be too much to have on a single RCD. That's why we need competent and sensible designers.

... My house splits the sockets over 6 RCBO's if you include the socket on the cooker supply, all together there are 14 RCBO's (13 in use) only time they tripped it was correct, the flat roof leaked so socket got wet. However old house with just two RCD's it seemed to go in batches, could last a year with out a trip, then would get 5 trips in a month, then another year, ....

My current experience (with a dozen or so residual current devices) is similar to yours - virtually no 'incorrect' ('nuisance') trips in 20-30 years, other than ones we sometimes got when there was lightning in the area but, for some inexplicable reason, that all stopped 10-15 years ago (maybe a lot of the overhead distribution was moved underground, as happened in our village?). Why some people (like you in your old house) apparently get/got all these 'nuisance trips', I really don't know - as you suggest, maybe it was a phenomenon of earlier devices?

Kind Regards, John

 

[John D v2.0]

It might be mentioned by an inspector, but since it's really beyond the scope of BS7671, I'm not sure that it could actually be 'reported' ('or coded') as part of an EICR.

I believe reg 133.3 says that electrical equipment must be suitable for its location. So you could say one socket per room isn't suitable for a domestic house nowadays. As evidenced by the need for ad hoc extension leads. However you could only code it if it gives rise to a danger.
See what you mean about the rewireable fuses though. It's more of a marketing thing (have some flashy trip switches) rather than a regs thing.

 

[JohnW2]

I believe reg 133.3 says that electrical equipment must be suitable for its location. So you could say one socket per room isn't suitable for a domestic house nowadays.

I agree.

I think I misunderstood what you were suggesting. I thought you were talking about the use of "lots of multi way adaptors" - which, as I said, I think that, per se, would be beyond the scope of BS7671, and hence of an EICR. However, have re-read your post, I think you were probably talking about the need (or potential need) for "lots of multi way adaptors" that might result from an inadequate number of sockets - and, as you say, such an inadequacy could legitimately be 'flagged' in an EICR (regardless of the presence/absence {or even actual 'need for} of adaptors).

Kind Regards, John

 

[ericmark]

There must I am sure be a figure which is taken as medium average, hope I have name right it is a long time ago when I did mean mode and medium, maths is not my strong point. So sure there is a number of sockets which is considered as acceptable from one RCD.

I was rather surprised at the electrical safety council as to what should be included in an EICR, as stated I would consider anything plugged in was beyond my remit. But they show an over loaded weight wise socket as reason for failure.

I do consider any device plugged into a socket not to BS1363 should be removed, as to if it should then fail as possibly strained not so sure? But if I find a blank in a scoket not marked BS1363 then it is removed.

 

[JohnW2]

There must I am sure be a figure which is taken as medium average, hope I have name right it is a long time ago when I did mean mode and medium, maths is not my strong point. So sure there is a number of sockets which is considered as acceptable from one RCD.

You mean median (not medium), and that's the value which is 'in the middle' when figures are sorted into ascending order (i.e. equal, or roughly equal, numbers above and below it. Mode is a very different animal, and not used a lot in statistical circles, being the 'most common' value. Mode is really just 'descriptive' and is almost impossible to deal with mathematically, but it is of importance to some - particularly manufacturers etc. - the 'average' (mean or median) size of shoes sold might be, say, 9, but if the greatest number of sales (i.e. 'mode') were of, say, size 11, then more size 11s should be made than any other size, despite the mean or median size sold being 9.

However, the point is that whilst there obviously will be an 'average' (of whatever sort) of number of sockets, the number appropriate for a given building or room (used for particular purposes) will vary widely, and it is for the designer to decide what is an appropriate/reasonable number for the building/room/whatever in question (and its intended usage). A bedroom or storeroom generally needs a very different number of sockets from the number required for, say, a kitchen or workshop.

I was rather surprised at the electrical safety council as to what should be included in an EICR, as stated I would consider anything plugged in was beyond my remit. But they show an over loaded weight wise socket as reason for failure. I do consider any device plugged into a socket not to BS1363 should be removed, as to if it should then fail as possibly strained not so sure? But if I find a blank in a scoket not marked BS1363 then it is removed.

The ESC is/was a rather odd organisation, which said some rather odd things (and I'm not sure still exists). However, what you talk of is perhaps a grey area, since if things are plugged into a BS1363 socket which 'threaten to damage' the socket mechanically, then that could be regarded as within the scope of BS7671, hence EICRs.

Kind Regards, John

 

[ericmark]

This is the picture and link to me nothing to do with an EICR as it may be temp, and 5 minutes latter removed.