1. Lighting circuit safe area, does it need RCD? 2. RCD maximum current smaller than combined rings

bernardgreen said:

To re-start a heart that has started to fibrillate ( rapid random contractions that do not pump blood ) the casualty is given an electric shock to stun the heart's nerves and stop it. After the shock the heart can self re-start into an effective pumping rhythm. Hence a short duration shock ( such as that from a defibrillator ) is very un-likely to kill. Death happens when the duration of the shock is enough to cause tissue damage to vital organs.

Click to expand...

The problem is that when (accidental) electric shock (no matter how short the duration) results in immediate death, it is usually due to ventricular fibrillation, and once that has happened, nothing other than a (controlled) further shock will start the heat beating/pumping normally again.

bernardgreen said:

Is the 40 milli-second duration before the RCD trips short enough to prevent tissue damage ?

Click to expand...

As with all these things, it is essentially "I²t" that matters, so the magnitude of the current (and, obviously, its path), as well as its duration which matters. However, although you often talk about it, I don't think that significant "tissue damage to vital organs" is often much of an issue in those who survive the immediate effects of electric shock (unless the head/brain is involved, but that's generally only seen in relation to judicial electrocution) - so I very much doubt that any credible current for 40 ms would result in much significant/serious organ tissue damage.

Kind Regards, John

Taylortwocities said:

And 30mA can be enough to kill you. IIRC a current of only 10mA MAY be enough to start AF. That could be fatal in some people. It’s not a precise science.

Click to expand...

[I assume that you mean 'VF' (ventricular fibrillation), rather than 'AF' (atrial fibrillation', which is far less of a threat to life]

In 'some people', it's far worse than that. There are plenty of people walking about who, because of heart disease, have a heart that is so 'electrically unstable' that they are at constant risk of spontaneously developing ventricular fibrillation (the classic 'dropping dead'), even in the absence of any electrical stimulus from outside their bodies. In such people, any current due to an electric shock, even 1 mA or less, might be adequate as the final straw which tips them into fatal VF.

The point, of course, is that the concept and recommendations of protection against electric shock relate to 'most people', not that very small group of 'some people', and it is generally believed/accepted that a shock of less than 40 msec duration will not kill 'most people'.

The 30 mA (or whatever) trip threshold is, of course, potentially misleading since there is no way that an RCD can limit the current through a person. The current which flows is simply down to Mr Ohm - with wet skin that can be appreciably more than 30mA even at 230V, but obviously a lot more if higher voltages are involved.

The 30 mA threshold has been chosen because it is felt that currents lower than that will not kill 'most people', no matter what the duration of the shock.

Kind Regards, John

JohnW2 said:

However, in terms of the OP's concerns, it's no different from Main Switches. They are usually rated at 100A, sometimes only 80A, but if one adds up the ratings of all the MCBs in the CU, it would not be unusual to get a figure approaching 200A.

Click to expand...

RCD overload protection cannot be by diversity alone, so if the MCB total is more than the RCD rating and the main fuse is also more, it does not comply.

flameport said:

RCD overload protection cannot be by diversity alone, so if the MCB total is more than the RCD rating and the main fuse is also more, it does not comply.[/ATTACH]

Click to expand...

Well, as I just wrote, if one temporarily forgets about "RCD overload protection" and instead considers "Main Switch overload protection", then I would think that, per what you have cited (highlighted), a high proportion of CUs "would not comply". Is that not the case?

Having said that, I'm not sure how one is meant to interpret the totality of the sentence in the reg you have highlighted together with what is said in the three options which immediately precedes that sentence/paragraph.

Kind Regards, John

You know I'm not good at cryptic clues but what does the last sentence mean?



Which is THE OCPD? Is it suggested there should be a single separate one to protect the RCCB or switch?

Which manufacturer? That of the RCCB or switch or the OCPD?

EFLImpudence said:

You know I'm not good at cryptic clues but what does the last sentence mean?

Click to expand...

I'm glad to hear that it's not just me! As I wrote ...

JohnW2 said:

... I'm not sure how one is meant to interpret the totality of the sentence in the reg you have highlighted together with what is said in the three options which immediately precedes that sentence/paragraph.

Click to expand...

... and, yes, that question applies to the next sentence (which flameport didn't highlight), too!

Kind Regards, John

JohnW2 said:

The 30 mA (or whatever) trip threshold is, of course, potentially misleading since there is no way that an RCD can limit the current through a person. The current which flows is simply down to Mr Ohm - with wet skin that can be appreciably more than 30mA even at 230V, but obviously a lot more if higher voltages are involved.

The 30 mA threshold has been chosen because it is felt that currents lower than that will not kill 'most people', no matter what the duration of the shock.

Click to expand...

In
from 12:30 to 14:50 there is some discussion on an "experiment" to determine the maximum "Let Go" and "Fibrillation" current levels.

I have seen a few times where 63 amp and 80 amp RCD's have been used in a consumer unit where the MCB's would allow in excess of that rating, I don't think any overload would be long enough to cause a problem, however technically the RCD should be selected so over load is not possible, as to whom is responsible that is not so clear, a minor works certificate is not a legal requirement and it says "Comments on (including any defects observed in) the existing installation (Regulation 644.1.2):" the whole form is here and it is questionable as to what should be done with an old installation. However I would have thought a comment should be made as clearly could not enter the tripping time or current.

In Wales there is a requirement for a compliance or completion certificate but not in England.

Note:- If you were the person ordering the work, but not the owner of the installation, you should pass this Certificate, or a copy of it, to the owner.

Note not the occupant, so with a rented property the occupant may not get a copy, and the electrician may do everything correctly and it could be the builder who has not passed it on. The new EICR law will correct this to some extent.

The question is how far does some one working on a circuit go to ensuring it complies? For example some boilers state you should use a type A RCD, but some one working on the circuit is unlikely to find out installation instructions for all items fitted and check what is required. My house has a large sticker on the DNO fuse which says 60A but since sealed I have never pulled it to see if it really is 60A. If the fuse has ever blown it is likely it would be upgraded to 80A. It could even be 100A. As to if the DNO worker would in an emergency which if the fuse blows it would be an emergency would actually check to see if the RCD is big enough not so sure. Not even sure if he would change the label?

Yes today it would be nice to think every circuit is RCD protected, but when my mothers kitchen was done, the council who was doing the work was told the house wiring would not allow the use of RCD protection it had too much leakage, so a new consumer unit just for kitchen was fitted, then it was found the ceiling needed redoing and the electrician had not run in cables for the lights, so all but lights were RCD protected, but the lights did not even have an earth, I know what happened, but there was nothing on the paperwork to say lights were not protected. So is there any point demanding the minor works?

ericmark said:

I have seen a few times where 63 amp and 80 amp RCD's have been used in a consumer unit where the MCB's would allow in excess of that rating ....

Click to expand...

I would have thought that 'a few times' is probably an appreciable understatement.

63A RCDs are far from uncommon, yet it only takes a couple of 32A circuits ('just'!), or a shower circuit plus almost anything else, to theoretically overload them.

I think one probably has to be sensible, particularly with sockets circuits - the chances of two or more 32A circuits being 'fully loaded' simultaneously in a domestic environment must be incredibly small.

... and, as I said, what about Main Switches, where I suspect the opposite may well be the case - i.e. that you have only seen a few CUs where the sum of the ratings of the MCBs did NOT exceed the 'rating' of the Main Switch.

Kind Regards, John

FrodoOne said:

In ... from 12:30 to 14:50 there is some discussion on an "experiment" to determine the maximum "Let Go" and "Fibrillation" current levels.

Click to expand...

Thanks, I'll have a look.

Of course, per what I wrote, fibrillation thresholds in terms of just current are of little value in relation to RCDs, since an RCD does not in any way limit the shock current.

Needless to say, there's quite a lot of data available in relation to fibrillation thresholds in animals (primarily pigs, whose torsos can be quite similar in shape/size to humans) but, for obvious reasons, proper experiments cannot be done in humans. However, we do know a lot about human fibrillation thresholds etc. in terms of the heart itself (i.e. when an electrical stimulus is applied directly to the heart), since deliberate electrically-induced ventricular fibrillation is quite often used during open-heart surgery.

Kind Regards, John

JohnW2 said:

... and, as I said, what about Main Switches, where I suspect the opposite may well be the case - i.e. that you have only seen a few CUs where the sum of the ratings of the MCBs did NOT exceed the 'rating' of the Main Switch.

Click to expand...

Surely if there is a 100A main switch and a 100A main fuse then that doesn't matter.

As for RCCBs, doesn't the subject raise the question of why do they bother to make ones with a rating of less than 100A.

They are, quite surprisingly I think, a little bit cheaper but not enough to worry about.

EFLImpudence said:

Surely if there is a 100A main switch and a 100A main fuse then that doesn't matter.

Click to expand...

You're getting dangerously close to talking about common sense, whereas flameport was quoiting the chapter and verse of a regulation, which is not necessarily the same thing

I suppose that, strictly speaking, there is a theoretical issue (even if we feel we can over-ride it by common sense). A 100A fuse will allow appreciably more than 100A to flow 'indefinitely' and considerably more than 100A to flow for moderate periods (e.g. around 200A for 1 hour). If one held the view that any current appreciably >100A was unacceptable for a Main Switch 'rated' at 100A, then that could be an issue.

Kind Regards, John

EFLImpudence said:

As for RCCBs, doesn't the subject raise the question of why do they bother to make ones with a rating of less than 100A.

Click to expand...

It does. I've never really understood, but I imagine that the reason/explanation is probably 'historical' (and unthinkingly perpetuated).

Kind Regards, John

JohnW2 said:

I would have thought that 'a few times' is probably an appreciable understatement.

Click to expand...

Yes as you say common to see 80A RCD supplying MCB's which add up to over 80A, the RCD has two current ratings, one momentary often 4.7 kA and one continuous the 80A rating and it would be unlikely you would get over 80A continuous. A B32A MCB should trip on the magnetic part at between 3 and 5 times the rating, so up to 160A before it trips in 0.01 seconds, there is a sliding scale below that, but seen many a 40 amp shower on a 32A MCB.

There are very few items in the house which can draw high current for an extended time, likely the immersion heater is the longest, followed by maybe tumble drier, it is recommended that items fixed at over 2 kW have a dedicated supply, and normally the immersion heater does have a dedicated supply, but the tumble drier rare, more likely to find the oven with a dedicated supply and the oven with use mark/space control to maintain temperature so in real terms draws high current for less time than the tumble drier or washer/drier.

We see many things which don't really comply with BS7671, and some times there is a slight re-writing to high light miss reading of the regulations.
314-01-01 Every installation shall be divided into circuits as necessary to:
(i) avoid danger and minimize inconvenience in the event of a fault, and
(ii) facilitate safe operation, inspection, testing and maintenance.
314.1 Every installation shall be divided into circuits, as necessary, to:
(i) avoid hazards and minimize inconvenience in the event of a fault
(ii) facilitate safe inspection, testing and maintenance (see also Section 537)
(iii) take account of danger that may arise from the failure of a single circuit such as a lighting circuit
(iv) reduce the possibility of unwanted tripping of RCDs due to excessive protective conductor currents produced by equipment in normal operation
(v) mitigate the effects of electromagnetic interferences (EMI)
(vi) prevent the indirect energizing of a circuit intended to be isolated.
The (iii) (iv) and (vi) are really covered by (i) and (ii) but it became common practice to use a single RCD for many circuits, but even after the change people interpreted it as the RCD not forming a circuit even when it actually mentioned the RCD in the regulations. As to how anyone designing a house electrics can work out if a RCD is likely to trip I don't know? 543.7.1.1 seems to say any plug should not have over a 3.5 mA drain, the point when a RCD trips is between 15 and 30 mA so clearly we want less than 15 mA so just 5 sockets could exceed 15 mA unlikely, but to have 25 sockets plus other items on one RCD is pushing it a bit. But to be fair we can get area power cuts, so "such as a lighting circuit" is not really a valid argument as does not matter how many circuits it can still fail, personally I have a rechargeable torch at top of stairs which auto switches on with loss of power or movement.

So we do a risk assessment, we don't follow BS7671 to the letter, but use some common sense. It may say "A cable concealed in a wall or partition" but we read it as "A low voltage cable concealed in a wall or partition" and the 8 volt door bell does not have a RCD.