Is My Masterplug RCD Safe? Please Help!

[securespark]

Well, can I balance that out a bit?

I'm sorry to report that I have found a great many failed RCD's.

Including two in the same dual RCD board fitted new 6 months earlier.

And another brand new out of the box!

However, you have to test them in isolation, as they can appear faulty if tested when connected to a circuit full of appliances.

I've had several that have failed the cycle of tests from a socket, but when carried out from the RCD in isolation, have sailed through!

 

[ebee]

Although I think we all believe that the in-service failure rate of RCDs is nowhere near as high as 7%, I don't think anyone has got a clue as to what the true figure actually is. It may,or may not, be 'acceptable' as the reliability of the only device which offers any hope of protection once an electric shock has arisen.

As I said the ESC (2006) data puts it at 2.8% and zero caused by environmental conditions.

As I recently wrote, this is a situation which electricians could remedy to a significant extent. There must be hundreds of RCDs tested every day, so it wouldn't take very long to generate some reasonable data (an awful lot more reasonable than we currently have) about in-service failure rates - and once we had some reasonable data to work from, we could have a sensible discussion. In the meantime, we can but speculate and guess.

I'll start the ball rolling on that one - over a five year period up until Sept this year I have installed 112 RCD/RCBO's and none of them have ever failed the standard tests.
In the same period I have conducted 12 PIR/EICR and only once has an RCD failed the test - in fact it was the mechanical test button - it was only when I looked at the device did I discover it had been shorted out by a previous ????
So someone found the fault but because it was/is difficult to find a replacement they took the short cut.

If it was our friend and he was using his own plug in RCD then that might have ended up saving him (he might have been unaware his "electrician" had done such a silly trick )

 

[securespark]

Indeed.

You (and anyone else) may think there are two RCD's in the circuit, but in reality, there's only one!

 

[JohnW2]

As I said the ESC (2006) data puts it at 2.8% and zero caused by environmental conditions.

Yes, we are familiar with that data, but it derives from a very small and quite probably unsatisfactory study. I strongly suspect (and hope) that the 2.8% is appreciably higher than the truth. If I believed that the risk of an RCD not 'doing what it said on the tin' when I needed it to (in order to limit the duration of a shock I was receiving) was as high as 2.8% (about 1 in 36), I would probably be muttering a lot more loudly about the need to consider doubling up on RCDs!

I'll start the ball rolling on that one - over a five year period up until Sept this year I have installed 112 RCD/RCBO's and none of them have ever failed the standard tests.

If you're talking about 'standard tests' undertaken at the time of installation, that's really of no interest to me. I would assume/hope that every one is tested at the time of installation and that any 'congenital failures' are not put into service. It is subsequent failure of an RCD which was functioning satisfactorily when put into service that concerns me.

In the same period I have conducted 12 PIR/EICR and only once has an RCD failed the test ...

Well, from what you go on to say, that one failure might not totally 'count' (did you test it after removing the short?) - but, taken at face value, that would obviously be an 8.3% failure rate. Hopefully that will drop dramatically if/when others can add a few hundred other tests to the melting pot!

Kind Regards, John

 

[riveralt]

we are familiar with that data, but it derives from a very small and quite probably unsatisfactory study.

Must be talking about different studies - 607 RCDs tested well exceeds the Central Limit Theorem - who is 'we'?

I'll start the ball rolling on that one - over a five year period up until Sept this year I have installed 112 RCD/RCBO's and none of them have ever failed the standard tests.

If you're talking about 'standard tests' undertaken at the time of installation, that's really of no interest to me. I would assume/hope that every one is tested at the time of installation and that any 'congenital failures' are not put into service. It is subsequent failure of an RCD which was functioning satisfactorily when put into service that concerns me.

Why? you raised the issue of out of the box problematic RCDs - all of the ones I used came from the same supplier.
The majority of those discovered faulty in the ESC survey were down to misaligned contacts which may be a factory fault and therefore should be included in any study to get the complete picture.

In the same period I have conducted 12 PIR/EICR and only once has an RCD failed the test ...

Well, from what you go on to say, that one failure might not totally 'count' (did you test it after removing the short?) - but, taken at face value, that would obviously be an 8.3% failure rate. Hopefully that will drop dramatically if/when others can add a few hundred other tests to the melting pot!

Yes it does count because of course I tested it after removing the short. I have just checked the actual number of RCD/RCBO's is 29.

 

[JohnW2]

we are familiar with that data, but it derives from a very small and quite probably unsatisfactory study.

Must be talking about different studies - 607 RCDs tested well exceeds the Central Limit Theorem - who is 'we'?

[the 'we' is those of us here who have discussed it in the past]
OK, in the interests of continuity of discussion, I'll deal with that later. If you're happy that the 2.8% figure represents a satisfactorily estimate of the in-service failure rate, then fair enough - let's assume that for the time being. With a sample of 607 RCDs, the 95% confidence intervals of that 2.8% failure rate is approximately 1.5% to 4.4%.

If you're talking about 'standard tests' undertaken at the time of installation, that's really of no interest to me. I would assume/hope that every one is tested at the time of installation and that any 'congenital failures' are not put into service. It is subsequent failure of an RCD which was functioning satisfactorily when put into service that concerns me.

Why? you raised the issue of out of the box problematic RCDs - all of the ones I used came from the same supplier.

I've never raised that issue - certainly not in this thread. Given the manufacturer's QA and QC, the 'out of the box' failure rate will almost certainly be a lot lower than the in-service failure rate, so if one included 'out-of-the-box' tests (as well as in-service tests) in one's figures, one would end up seriously underestimating the in-service failure rate. Furthermore, if one assumes (OK, OK!) that all RCDs are tested at installation, and that those failing those initial tests are not put into service, it makes no sense to include initial tests in an exercise to estimate in-service failure rates (which is what obviously matters to users/consumers).

Yes it does count because of course I tested it after removing the short. I have just checked the actual number of RCD/RCBO's is 29.

Fair enough. So, if I understand correctly, you found one faulty in-service device out of 29? If so, at 3.4%, the failure rate you've observed is pretty close to the ESC's 2.8% figure (and certainly well within the 95% confidence interval of the ESC's figure) - so maybe that is roughly correct, even if (IMO) worryingly high.

Kind Regards, John

 

[JohnW2]

Must be talking about different studies - 607 RCDs tested well exceeds the Central Limit Theorem

Since you seem to be happy that the ESC's estimate of an in-service RCD failure rate of 2.8% is fairly accurate, I'd be interested in using this figure to gain some insight into your approach to 'mortal risk' - by presenting you with what I think is a reasonably close medical analogy ...

Imagine that it was known (e.g. because of family history or genetic tests) that you were at a small risk of developing a rare disease (c.f. a small risk of suffering a electric shock potentially serious enough to kill) which, if you developed it, in the absence of prior treatment would probably kill you.

Having taken whatever measures are possible to prevent you developing that disease (c.f. ensuring the basic electrical protection and ADS were in place), I then offer you some further options.

Firstly, I can give you a drug which, in the event of your developing the disease will have a 97.2% chance of preventing the disease killing you (i.e. a 2.8% chance of it failing to prevent your death).

Alternatively, I can give you another drug as well as the first, and this will increase the chance of your not dying if you develop the disease to, say, 99.6% (i.e. a 0.4% chance of it failing to prevent your death).

Neither drug has any particular downside (e.g. side effects) other than a small up-front monetary cost. If you chose the two-drug oprion, there would be a slight inconvenience in having to remember to take two drugs rather than one (or none) (c.f. the slight inconvenience of maybe having to reset two RCDs if both trip)

What option do you choose:

1...No drug treatment, because the risk of the disease even developing is already small (c.f. no RCDs at all, because the risk of your suffering a serious electric shock is already small)
2...Just the first drug – a risk of dying of 2.8% if you develop the disease (c.f. one RCD)
3...Both drugs – a risk of dying of 0.4% if you develop the disease (c.f. two RCDs)

?? Views will obviously vary - so I’d also be interested to know what readers other than you would choose.

Kind Regards, John

 

[echoes]

John, for the purposes of your little hypothesis, in the population as a whole, what is the probability of developing the disease?

And what is the probability that the genetics/family history reliably predict the onset of the disease?

 

[JohnW2]

John, for the purposes of your little hypothesis, in the population as a whole, what is the probability of developing the disease (being electrocuted)?

That's somewhat of a $64,000 question. IIRC, the number of UK 'domestic/leisure' deaths recorded as being due to electrocution (despite widespread use of RCDs) is currently around 20 per year - i.e. about 1 in 3million. What on earth the figure would be in the absence of RCDs (which is what we need to know), I haven't a clue ... but we are (surprisingly, IMO) certainly talking about incredibly small risks. I confess to being a bit of a sceptic, who isn't convinced that RCDs have actually 'saved many lives' at all (as I've discussed before, most people seem unaware, in their experience or 'circles', of anyone who has ever suffered a shock which has caused an RCD to operate)

And what is the probability that the genetics/family history reliably predict the onset of the disease?

For the analogy to be be valid, it would obvioulsy have to be similar to 'the figure above'. What I had hoped was that people would have some personal 'notion'/belief of what they believed to be the risk of a potentially lethal electric shock (without RCD protection), and would thus apply the same figure to the 'disease' analogy.

Given my scepticism, it's a bit ironic that I've got involved in this discussion about 'multiple RCDs', at least as far as deaths are concerned, since electrocution is (surprisingly) so incredibly rare. Even though RCDs are very far from ubiquitous (IIRC, about half of UK domestic installations don't have RCD protection), there are 'hardly any' electrocution deaths to be prevented by extending RCD protection to all installations. Someone will probably come up with the 'one death is one death too many' argument, but I would counter that with the suggestion that if all the moneywhich has been spent on buying, installing and testing RCDs (other than those in TT installations) had, instead, been spent on something else (e.g. the right aspects of road safety), then far more lives might well have been saved.

In terms of death, the risk is so incredibly small that it is tempting to go with my option 1 (no RCDs at all). My question is therefore really directed at those who actually believe that RCDs are a valuable 'life saving' measure, and is asking them whether, having decided that we need RCDs to save lives, they are happy with there being a 1 in 36 chance of a (single) RCD not actually saving the life that RCD(s) could have saved.

Of course, much more common than electrocution are serious injuries due to electric shock, and there might be major benefits of RCDs there - but there is AFAIAA very little in the way of useful statistics about that.

Kind Regards, John

 

[ban-all-sheds]

electrocution is (surprisingly) so incredibly rare.

Not as rare as it was, according to those clowns at the OED.

http://oxforddictionaries.com/definition/english/electrocute

 

[JohnW2]

electrocution is (surprisingly) so incredibly rare.

Not as rare as it was, according to those clowns at the OED. http://oxforddictionaries.com/definition/english/electrocute[/QUOTE]
We've been through all this before. Dictionaries do vary, and if you do some searches, you'll find plenty of references to 'non-fatal electrocution', including in learned journals. However, you (and others) have convinced me in the past that the word is most commonly used to refer to death due to electric shock.

Kind Regards, John

 

[ban-all-sheds]

That's what the word meant when first coined, and it remained with that meaning for many years.

The first person who used it to mean a non-fatal electric shock was wrong. It really is that simple, no matter how "learned" he was, he was wrong.

 

[JohnW2]

That's what the word meant when first coined, and it remained with that meaning for many years. The first person who used it to mean a non-fatal electric shock was wrong. It really is that simple, no matter how "learned" he was, he was wrong.

I believe that's all correct. However, language evolves and, if you think about it, evolution of use/meanings of words can only ever start by someone using the word in a manner which is 'wrong' in terms of what is regarded as 'right' at the time that first step in evolution occurs.

Kind Regards, John

 

[ban-all-sheds]

Absolutely.

But the OED gave in to the ignorami far too quickly.

 

[EFLImpudence]

It is literally awful but it seems that the folk who decide these things are staunch democrats insofar as when the majority of people are wrong it therefore becomes right.

You may consider this process wicked whether you agree with it or not.