Why is it that the Americans and Canadians can with reasonable success (I'm not saying there are NO 'nuisance' trips) use GFIs which trip on 5 mA.? (GFI = Ground Fault Interrupter = RCD)
Use of 120V. instead of 240V. might account for a 5mA. trip level instead of a 10mA. trip level, but OTOH 60Hz instead of 50Hz would increase capacitive leakage currents by 20%.
Shouldn't we be devoting more attention to reducing the various types of leakage which cause nuisance trips, and then increasing not reducing RCD sensitivity ?
You were getting terribly aerated because the question, which was "If we changed from 30mA rcds for general use to 100mA, would it solve more problems than create?" was not being answered. You now get aerated because it has been!!
If there is only a single RCD in the house (common 10-20 years ago) and it is 30mA it will probably suffer from nuisance tripping. When this occurs all the lights go out. It is common for 30mA RCD's to trip when bulbs fail - usually when they are switched on.
If it is dark there is a greatly increased risk of falls. If you are upstairs you have to find your way down and to an often inaccessible consumer unit in the dark.
The number of people killed by electric shock is about 10 every year,
8 of these are due to using extension cords and appliances with faults or similar. 2 are due to faults in fixed wiring.
The number of people killed in falls is about 2,000, nearly half of these are killed falling down stairs. Those killed in fires amount to another 350 or so. The risk of falling when investigating why all the lights have gone out is far greater than the risk of electrocution.
The fitting of whole house 30mA RCD's was a classic example of poor quality risk assessment - looking at a part of the problem rather than the whole of the problem. Someone decided that if you held a live wire and stepped in the bath you stood a better chance of surviving if a 30mA RCSD was fitted than if a 100mA RCD was fitted. No one could argue with that but they were not looking at overall risk.
In a _house_ over a period of time the risk caused by power being removed with no warning is far greater than the risk of electrocution.
It is difficult to say as the first people to arrive often don't make a note of it. The only reliable source of information I have seen was a study carried out some years ago which looked at detailed "first attender" reports only of deaths in fires. In a significant number of these the lights were out (RCD tripped) and the occupants had failed to find their way out. Most of the fires occurred after they had gone to bed and were usually caused by discarded or forgotten cigarettes. In a 12 month period it was possible to identify about 5 deaths (in one fire authority area) in which the lack of light was most probably a major contributor to death in that the fire was insufficiently developed at the time the occupants woke to kill them but they failed to find their way out and died in rooms where they had not been asleep and which were not on the escape route.
I am personally aware of a fire (not included in those figures and not in that area) in which two adults and two children died against a locked front door having dropped the key to allow them to get out and having failed to find it. The evidence was that they had more than sufficient time to escape and had been trapped behind the front door for some minutes. The lights in that instance had almost certainly failed because the RCD had tripped as witnesses reported the house in darkness some time before the deaths (they didn't even think there was anyone in the house).
As far as falls are concerned most of the evidence is in the statements of people who have survived rather than have died. I am not aware of any formal study which tracks this data but a small scale study in an elderly care unit at Salisbury (Odstock) done as part of a student thesis in about 1990 indicated that about 5 out of
20 people had fallen part of the way down stairs at night because all their lights had gone out when they turned the bedroom light on and they had gone down to fix the problem.
Why "all" the lights failed wasn't recorded and some were undoubtedly simply a bulb failing and the occupant not checking any other lights. However it was noted that a higher than expected number of falls at night took place in Local Authority housing - all of which in that area had recently been rewired and had whole house 30mA RCD's fitted.
As the (albeit imperfect) figures for two small areas of the UK show a high probability of more people being killed in those areas alone by failure of lighting at critical times than are killed in the whole of the UK by electric shock I suggest it is extremely likely that on balance the 30mA whole house RCD has killed far more than it has saved.
The fall mortality data is from the Office of National Statistics and the Fire Data from there and the ODPM (which collates fire information). I'm not sure the same years were used for both so there may be some variations but they are not huge. The cause of death in fires study is not in the public domain as far as I know. The Odstock one should be as it was a thesis but I have no idea where to find a copy (the student concerned came from Bristol University if I remember correctly).
a lot of smoke and the visibility is probably zero. Whether or not an RCD tripped; and/or there was no lighting is academic under this circumstance. With due respect and this is not a dig at you personally: - There are lies, damn lies and statistics. Jaymack .
I'm not sure but arent their gfcis usually used on one socket, whereas ours usually cover 2 floors of appliances, or in some cases a whole house. Immersion heaters, assorted elements, computers, all on one rcd.
Our 30mA RCDs trip at typ 15mA, and you've got a whole house full of stuff on just one, or maybe half a house full.
I'm not sure theres any significant safety mileage in it tbh. And if one did go that way, what method would you propose to improve the common heating element leakages, IT equipment filter leakage, and filament bulb current surge on capacitive wiring? I dont see any easy solutions to any of those, just spending big with no real benefit.
If that were true, then the logical conclusion would be that lit escape signs in buildings are of no worth. I doubt thats true though. I dont think visibility is always zero. Maybe Peter has more info.
Domestic fires do indeed generate a fair amount of smoke but not usually enough to render visibility zero in lit conditions. There is also the unpredictable reaction of people woken from sleep to find themselves in the middle of a fire to consider.
There is no doubt at all that escape from a building full of dense smoke is much easier if there are visual cues. Something like seeing a light bulb can be enough to regain some orientation and glimpsing a piece of wallpaper or any known object at close range can equally help. Conversely total darkness and smoke is extremely disorientating. I have seen experienced fire fighters wearing breathing apparatus and in training buildings (so no threat to life) become totally lost very quickly.
The case of the family who died looking for the key to open their door (it was found under one of the bodies) is perhaps exceptional but there is plenty of evidence of people who have gone into a spare bedroom or box room by mistake and died there.
The number killed in fires however is dwarfed by the number who die and are injured in falls.
The case for RCD protected sockets for equipment used outside and in garages is indubitable. However, using a whole house 30mA RCD does little for safety and introduces different risks which were not considered at the time. It is only since evidence of the danger of losing lighting has become better known that the regulations have been altered to remove whole house 30mA protection.
For those houses with the old fitting my contention is that replacing the 30mA RCD with a 100mA (or removing it altogether unless it is needed for supply earthing needs) is safer than leaving it in place. Sockets used for external equipment can be replaced with RCD protected ones. Ideally a new split Consumer unit would be fitted but if this isn't practical then no RCD is safer than a 30mA whole house fitting.
I have first hand experience of a situation a few years ago, in South Africa. Central heating is not usual over there, in the wintertime it can get very cold so we had a paraffin heater in the bedroom, with an open window for ventilation. My wife rose around 7am and everything was ok, as it was a Saturday; and I had had a few Glenmorangies the night before, My wife fortunately came back into the bedroom about 9am; and found the room full of thick sooty smoke. I was still asleep but when she wakened me, my mouth and nostrils were full of soot but I was completely unaware of the situation, (the heater had run out of fuel and the wick was burning away). We spent a long time later, washing the walls, ceiling and carpets. A smoke alarm would probably have avoided this incident; but the cheaper type were not available at that time; and I would probably not have fitted one in the bedroom anyway. I give this as an example here, since I think statistics show that more people die, because of smoke inhalation rather than burns, I presume this occurs when they are asleep, are dis-orientated or have restricted visibility due to smoke.
Or went into another room to avoid the smoke?
What are the statistics for persons who fell: -
1) Due to lack of lighting and: -
2) Due to smoke density.
With you on the last statement; but the whole house RCD, usually applies to TT installations where the distributor does not provide an earth; and the earth fault loop impedance is too high, to provide the necessary disconnection times for consumer safety. There is no way that this RCD should be removed. For the newer TT installations, there is still a requirement for a whole house RCD, as per the older regulations, normally
100mA with a time delay. However for TT installations and other TN-S and TN-C-S, as you seem to be aware, a lower rated 30mA RCD with no time delay, now requires to be installed for socket outlets (and other fixed electrical equipment, where recommended by the supplier, e.g. showers). IMO, this arrangement makes for a safer electrical installation.
I don't agree with your contention of substituting a whole house 30mA RCD; with a 100 mA unit; or removing it altogether, this would be compounding the felony and a criminal offence, if there was a subsequent court case. (Sorry, M'Lud I took it out because it was giving me problems). Any competent electrician, given time, can isolate the root cause of a tripping problem; with the help of a portable appliance tester; and other testing equipment. One of the reasons for the instigation of IEE Periodic Inspection Reports is to track the deterioration in the electrical installation over the years, (insulation etc). I would advocate, that if the statistics justify it; that a government grant be given towards uprating such older installations.
You have been grinding this particular axe a long time Peter, and no, don't stop grinding it either.
In this case I think you are entirely right.
In my house a trip downstairs is via a corridor, landing and passageway which has NO windows..and after the umpteenth time nearly killing myself (why do people always tack things on stairs?) and fumbling for a torch, I DID uprate to 100mA.
Now I MAY have a neutral earth problem...and one day I WILL investigate that.
BUT in my reasonably long and varied life I have not ever come across a fault that was potentially lethal that a 100mA RCD wouldnt have protected against.
I have not had a shock in YEARS..since the 60's really..off of mains wiring. Had far more off electronic equipment running of 100v, or off car ignition electrics. Or magnetos.
It takes a really gross amount of damage to todays kit to get a shock off it.
The cause of death in almost all fires is recorded as smoke inhalation even if the victim truly did burn to death (quite rare). It is certainly true that more people die in their bed from the effects of combustion fumes without ever waking up than die while trying to escape.
You usually find those victims in wardrobes or under beds in their own bedroom. Those in other rooms are usually trying to escape (there is often ample evidence of this).
In a modern house, or any with plasterboard ceilings, there is a lot to be said for fitting at least one upstairs room (preferably facing the road) with a 30min fire/smoke stop door and a wired telephone.
Rather difficult to collect as most of the victims don't say. However any fireman will tell you there is a vast difference between moving in a smoke filled room in darkness and one with even some light.
I think you will find most 30mA RCD installations were fitted in the great enthusiasm for them 10-20 years ago. Local authorities in particular fitted them to all house being upgraded whether they were needed or not. This followed a campaign by RoSPA in about 1990. Most are fitted in houses which do not use TT.
If you removed it for rational reasons and could justify those reasons then there is no offence.
No, "Sorry, M'Lud I took it out because it was more dangerous than not having it fitted". I would have no problems defending that statement.
They can't, because occasional tripping is more or less a characteristic of whole house 30mA RCD installations, the more so now that more electronic devices with noise filters are in use. The problem will become much worse as energy efficiency eurorules force the replacement of simple Class 2 "Wall Wart" plug in power supplies with switched mode units which usually have an earth and noise filters to achieve EMC compliance. Moreover many appliances with electronic controls such as washing machines cannot be fully tested by simple devices like a PAT as they cannot be forced into an operational cycle while being tested.
There are many 30mA RCD installations which have nothing "wrong" with them but will trip whenever a bulb fails or whenever they feel like it.
That was just to make money for electricians and allow them to make customers pay about GBP200 for the privilege of having to listen to a sales spiel pushing unnecessary "upgrades".
There is no evidence worth talking about that PVC wired electrical installations deteriorate significantly over time (and as the test records are rarely kept or referred back to no way at all of "tracking deterioration"). Even on a perfectly serviceable installation the average report will state "Not compliant with BS
7671, action required" (often in red) simply because it won't be up to this weeks standard and it allows a chance to sell something - usually a new Consumer Unit.
The money to be wasted on contracts to nulabors donors to "develop" ID cards would more than pay for such a scheme, you should suggest it to the ODPM. However, use small words and include lots of simple sketches and large denomination notes if you expect them to do anything about it.
Let me put forward the suggestion that for TT installs a whole house ELCB would be safer than either 30mA RCD, 100mA RCD or no earth impedance protection . And that the ideal solution would be whole house ELCB followed by RCBOs for sockets, or a split CU with the usual RCD on the sockets.
ELCBs do fully protect against the problem introduced by high earth impedance, by tripping when the earth rod voltage begins to rise. They dont sense L and N current at all, so have none of the safety problems that RCDs introduce. They can nuisance trip, but the incidence of these trips is a couple of orders of magnitude lower than with RCDs.
ELCBs are unfashionable, so I expect to hear some knee jerk resistance to this, but logically they seem to be the best choice for high R_earth TT installs. I've said for a while there is a case in favour of ELCBs being reintroduced as best practice in some situations. Another example would be to earth impedance protect a TT feed to several CUs in a block of flats.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.