Plugging an RCD into a cct already protected by an RCD/RCBO

[ericmark]

As already said we tend to jump up or down by a factor of 3. Mainly as the RCD has to trip between half rated value and whole rated value.
So we see 10mA, 30mA, 100mA and 300mA in common use.

After 500mA most are designed for use by a skilled or instructed person and are not able to be accessed by a normal person.

There are three values listed in the regulations 30mA, 300mA and 500mA the latter is a special case but basically 30mA protects against a shock and 300mA against fire.

The other use is where the earth system is not good enough and the value is determined by calculation. In the main rule of thumb sorry should say historically we used 100mA where the earth was not good enough as being robust enough not to trip too easy and sensitive enough to ensure it did trip under fault conditions.

The 2008 regulations have changed that some what and now most items are protected by 30mA.

As already said where a cabin is protected by 30mA non time delayed, the distribution unit with 100mA time delayed type S, and the main distribution unit with a 1A at 1 minute unit. A worker who knocked a nail through the wall to hang his coat on and hit the cable tripped out all the devices even though in theory only the 30mA one should have tripped.

However where the fault is not so dramatic for example dust and damp build up then using a 10mA plug in unit could stop the rest of the system tripping.

I have a 10mA socket supplying my shed to avoid any faults likely tripping my whole house. The house has two RCD's both 30mA and it is not unknown for both to trip together. Likely spike caused when first one trips then trips the second although so quick it seems both trip together.

Also resetting one often trips the second one. But my trips are old fitted in 1990 and today the trips are better made and don't trip as easy with spikes.

I have also used RCD's for a completely different job. At work I used active RCD's on all our grind stones. Not to protect from fire or shock but so that in event of a power cut on the power returning the grind stones would not auto re-start. This with old lawn mowers without deadman's controls this may still be a good idea.

The higher quality RCD's give warning between 50% and 90% and only trip between 90% and 100% of rated value. There are even auto resetting RCD's.

As well as active and passive there are RCD's for sine wave only, some will work with square wave and very special types for DC.

All now work with current but back in the third quartile of 20th Century we had voltage types. Now these could not be run in tandem and so you may find old books telling you how only one ELCB can be used. That would now be called an ELCB-v and a RCD can also be called an ELCB-c.

As to name I do agree it could cause people to think you are a spark but you did to be fair declare you were not so not really a problem. One one forum the admin started putting pro against qualified people but this back fired when tradesmen answered questions away from their expertise. And on another site all posts were monitored before being published and again this back fired as the admin then took responsibility for being a correct answer. We all make mistakes and for anyone to take any of the information given on any forum as true without checking they are taking rather a high risk.

Remember I have not talked about the special RCD's used by skilled persons and there are so many slight variations that I am sure there are types I have never seen.

But using two standard 30mA RCD's on same circuit in tandem will not reduce the effectiveness of either.

 

[JohnW2]

As already said we tend to jump up or down by a factor of 3. Mainly as the RCD has to trip between half rated value and whole rated value. So we see 10mA, 30mA, 100mA and 300mA in common use.

... and I suspect therein derives the common misconception that a 3:1 ratio of RCD ratings will always provide discrimination - although, as we know, that is very often not the case.

As The other use is where the earth system is not good enough and the value is determined by calculation. In the main rule of thumb sorry should say historically we used 100mA where the earth was not good enough as being robust enough not to trip too easy and sensitive enough to ensure it did trip under fault conditions.

That's the situation we've been discussing - in particular, the reason why, historically, 100mA (rather than, say, 30mA), time-delayed, was chosen for this purpose. You may be right that it was due to avoid unwanted tripping, but no-one seems to know for sure.

I have a 10mA socket supplying my shed to avoid any faults likely tripping my whole house.

As above, and particularly if the L-N imbalance is fairly large, I don't think there is anything approaching a guarantee you'll get that discrimination.

As I have also used RCD's for a completely different job. At work I used active RCD's on all our grind stones. Not to protect from fire or shock but so that in event of a power cut on the power returning the grind stones would not auto re-start.

Fair enough - although, particularly in the industrial setting, NVR switches are surely well-established for that job?

Kind Regards,John.

 

[Sparky54]

Well, as stated, I am not a mains electrical person and I can assure anyone concerned with the username that I have absolutely no intention of answering any technical questions in this forum !

Coming from the radio transmission engineering business I do understand most of the interesting technical discussion though...

My concern stemmed from an extension cable I saw being coiled up after use which had the coloured inner wires hanging from the plug, the live wire had damaged insulation and bare copper was visible... The lead had been plugged into a socket protected by a 30mA RCBO. My new extension lead has an RCD on the end and I was planning to loan the user that next time he needed one, (and to destroy his extension lead, which also had a taped joint a few cm from the unshielded and unclamped plug...). Just had a doubt over the performance of the two devices in series, now answered.

That lead was almost as bad as the electric fire I once saw being used in the bathroom of an old lady's 1950's-wired house, where both ends of the fuse holder were protruding through the back of the rubber plug

The one which made me scratch my head was a radio site where I accidently tripped the site RCD, only to have the big diesel genny start and put power back on - AFTER the RCD...

Thanks again.

 

[JohnW2]

My concern stemmed from an extension cable I saw ....Just had a doubt over the performance of the two devices in series, now answered.

Thanks for the background- a frightening, but only too familiar story! As for the two devices in series, several of us have answered that for you. RCDs/RCBOs work by simply sensing a small difference between currents flowing in the L and N conductors - and those currents will obviously be the same through as many devices as you care to connect in series.

That lead was almost as bad as the electric fire I once saw being used in the bathroom of an old lady's 1950's-wired house, where both ends of the fuse holder were protruding through the back of the rubber plug

At least it was a plug with a fuse, and therefore presumably not plugged into the lamp holder! When I think back to some of the things which,as a child, I saw going on 'electrically' in my grandparents house in the 50s, I still shiver!!

Kind Regards, John.

 

[Electrifying]

When I think back to some of the things which,as a child, I saw going on 'electrically' in my grandparents house in the 50s, I still shiver!!

Kind Regards, John.

Me too - did anyone elses dad used to plug the bare live & neutral of an appliance in, using the pins of another plug?

 

[ericmark]

Yep and my grand mother plugged iron into light socket. Seems only two 15A sockets in house and originally lighting was on a lower tariff than power. Hence people were in the habit of using a non earthed lighting supply.

 

[JohnW2]

Me too - did anyone elses dad used to plug the bare live & neutral of an appliance in, using the pins of another plug?

Yep,seen that one - or sometimes just matchsticks. ... and, of course,some amazing 'distribution systems' involving multiple 'adapters' of one sort or another! I think we originally only had one 15A socket in our house, plus a couple of 5A ones - so,as the number of appliances increased, so did the complexity of the tree of adapters (and trailing cables)!

Kind Regards, John.

 

[Paul_C]

Continuation of the practice certainly wasn't long ago - indeed, I'm sure that it continues to today. However, it was probably a fair while ago that the fitting of these RCDs started, and hence the 100mA habit got established, wasn't it?

I would say that the 100mA RCD probably became common place in the 1980's, but that's not to say that it wasn't employed before then, if earthing conditions warranted it (although it wasn't called an RCD then, but an RCCB or simply a current-operated ELCB).

Yes, that might well be the case. The 'voltage-operated' (somewhat of a misnomer!) devices which preceded them were also often/generally 100mA, weren't they?

In terms of the current needed through the coil to trip, they were typically rather more sensitive than that. I seem to recall figures for one of the common Crabtree types of it having a coil with a 500-ohm impedance and a tripping current of no more than 35mA (these coil characteristics being the limiting factor for the earth electrode resistance, since the latter had to be low enough for the trip current to be reached before the potential reached the 50V limit).

But remember that the current passing through the coil would be equal to the fault current only if there wasn't a parallel path to earth somewhere on the installation side (as I think we discussed on this forum some weeks ago). An earth path added on the "F" side of the coil would not raise the potential needed to trip the ELCB, but it would increase the fault current needed, since part of that current would then be bypassing the coil.

 

[JohnW2]

Yes, that might well be the case. The 'voltage-operated' (somewhat of a misnomer!) devices which preceded them were also often/generally 100mA, weren't they?

In terms of the current needed through the coil to trip, they were typically rather more sensitive than that. I seem to recall figures for one of the common Crabtree types of it having a coil with a 500-ohm impedance and a tripping current of no more than 35mA (these coil characteristics being the limiting factor for the earth electrode resistance, since the latter had to be low enough for the trip current to be reached before the potential reached the 50V limit).

Yes, a coil around 500Ω is what I remembered, which would correspond to a current through the coil of 100mA for 50V across it - hence my comment. If it tripped with a current flow of 35mA, that would correspond to only 17.5V across the coil. Are you saying that they were that much more sensitive than 'necessary' to limit the voltage to 50V?

Kind Regards, John.

 

[JohnW2]

Yes, a coil around 500Ω is what I remembered, which would correspond to a current through the coil of 100mA for 50V across it - hence my comment. If it tripped with a current flow of 35mA, that would correspond to only 17.5V across the coil. Are you saying that they were that much more sensitive than 'necessary' to limit the voltage to 50V?

Just thinking a bit further. If the required 50V limit for the CPCs was relative to true earth, then in what would probably be a worst case scenario, of a TT electrode with Ze=200Ω, that would resumably trip at around 71mA through the device's coil.

Kind Regards, John

 

[Paul_C]

If it tripped with a current flow of 35mA, that would correspond to only 17.5V across the coil. Are you saying that they were that much more sensitive than 'necessary' to limit the voltage to 50V?

As you've just noted, you need to allow for the resistance of the earth electrode in the calculation - The higher the electrode resistance, the higher the voltage on the "F" terminal (relative to true earth) needs to be to reach the trip current. I seem to recall that same Crabtree ELCB mentioned earlier actually specified that it was for use with an electrode resistance of up to 500 ohms, which is actually quite conservative in providing a wide margin if you calculate the values involved.

Although 50V was the limit, in practice most installations had sufficiently low electrode resistance values to provide for tripping at much lower voltages.

 

[JohnW2]

As you've just noted, you need to allow for the resistance of the earth electrode in the calculation - The higher the electrode resistance, the higher the voltage on the "F" terminal (relative to true earth) needs to be to reach the trip current. I seem to recall that same Crabtree ELCB mentioned earlier actually specified that it was for use with an electrode resistance of up to 500 ohms, which is actually quite conservative in providing a wide margin if you calculate the values involved.

Fair enough. If the electrode resistance was as high as 500, then the tripping current for a volatge (relative to true earth) would presumably be 50mA. If, as you said, the device actually tripped at 35mA through the coil, that would obviously then be at a voltage of 35V above true earth - i.e. an additional 'safety margin'.

I wonder what the conceptual thinking was behind this. Was the idea primarily about 'touch voltages', or was it seen more widely as a means of disconnection in the case of L-E faults? If the former then, if the concepts of bonding and equipotential zones were established at the time, the actual potential of CPCs (and hence exposed-conductive-parts) relative tob true earth would presumably not have mattered all that much, would it?

Kind Regards, John.