Nuisance tripping of ringmain RCCB

[JohnW2]

I cannot see how 'my' ramp test cannot work. ... Are we saying - not that it wouldn't/doesn't work but that it is unlikely to be useful because a circuit leakage of between the relevant values is unlikely?

Yes, that's what I'm saying.

It could, of course, be that a small leak (say due to water ingress), not itself enough to trip an RCD, was facilitating 'nuisance trips' when other things which also had leaks were connected to the circuit, therefore resulting in a 'totting' up of leaks to enough to cause the RCD to operate. In that situation, you might be called to the site, and your ramp tests ought to 'work' as you expect. However, if there are not enough such 'other leaky loads', then the primary sub-30mA leak is not going to cause any RCD trips, so you won't get called - and that is why I suspect that you won't often find yourself investigating these sub-30mA leaks.

I agree with the rest of what you say about how ramp tests ought to (and, I presume, do) behave in these circumstances - I just feel that, as above, you probably won't all that often find yourself investigating such situations.

There is, of course, one potential complication, although not one you're likely to see in relation to water ingress. If the primary fault is a N-E fault/'leak' that will presumably work 'in the opposite direction' to the test current applied by your meter, and therefore should increase the required trip current you measure with your ramp test.

Kind Regards, John

 

[EFLImpudence]

There is, of course, one potential complication, although not one you're likely to see in relation to water ingress. If the primary fault is a N-E fault/'leak' that will presumably work 'in the opposite direction' to the test current applied by your meter, and therefore should increase the required trip current you measure with your ramp test.

Oh, sneaky.

If that's how it works - is it? - it would still show up with -
30mA when no circuits connected and higher when circuits connected.

Haven't we fallen into this trap before when not realising that ANY leakage, even on the line, downstream of the RCD is seen as the same because it is not returning through the neutral?

 

[JohnW2]

Oh, sneaky. ... If that's how it works - is it? - it would still show up with - 30mA when no circuits connected and higher when circuits connected.

Yes, that's what I was suggesting, but ...

Haven't we fallen into this trap before when not realising that ANY leakage, even on the line, downstream of the RCD is seen as the same because it is not returning through the neutral?

Now you have me a little worried, since I vaguely recall such a discussion. I need to go away and think - watch this space!

Kind Regards, John

 

[JohnW2]

Haven't we fallen into this trap before when not realising that ANY leakage, even on the line, downstream of the RCD is seen as the same because it is not returning through the neutral?

Now you have me a little worried, since I vaguely recall such a discussion. I need to go away and think - watch this space!

I've 'thought', and I think you're right . Whether the 'leak to earth' is from L or N, the result will be the same - the current through the L side of the RCD will be greater than that through the N side. The phenomenon I was postulating (and which we've discussed before) would only arise if the current through the N side of the RCD was greater than that through the L side - and that just can't happen unless there are cross-circuit (cross-RCD) connections, since all the current going through the N side must come from somewhere (some L). The only other 'contrived' situation I can think of would be if (as well as there being an N-E fault) there were another fault such that 'earth' potential was appreciably higher than N potential - but that's really scraping the barrel!

So, assuming that I am now 'thinking straight', I think that 'afterthought' of mine was probably a mistake - but that does not alter the rest of what I said - i.e. that I suspect you don't often find yourself investigating sub-30mA leaks.

Kind Regards, John

 

[EFLImpudence]

but that does not alter the rest of what I said - i.e. that I suspect you don't often find yourself investigating sub-30mA leaks.

Indeed, not on circuits but with appliances and accumulated leakage the results are the same.

 

[JohnW2]

but that does not alter the rest of what I said - i.e. that I suspect you don't often find yourself investigating sub-30mA leaks.

Indeed, not on circuits but with appliances and accumulated leakage the results are the same.

True, but one has to consider the realities of small leakages. Given the inevitable variability of both trip times and trip currents of RCDs, you would probably be hard-pressed to confidently quantify a small leakage using the ramp testing method.

Kind Regards, John

 

[ericmark]

To me in the main it's the carbon track formed when you start to get leakage which is the problem.

Once it has formed then as it gets damp the leakage shoots up so be it 10 mA, 30 mA, or 100 mA it still takes out the trip.

But where the carbon track has not formed then you can get discrimination. With IT control systems one can watch the residual current monitor (RCM) bar graph raise in fall with humidity but then something somewhere starts to track and all of a sudden one has a major leakage problem. When found it does not seem to matter how much you try to clean it once the track is there it will return to that track again and again. Some times drilling a hole through the track will work I have done that many times with distributor caps with old cars.

So yes a 10 mA RCD may help but there is no guarantee. So there are two methods.

One fit a RCBO in the consumer unit which supplies the pond.
Two use an isolation transformer on the pond supply.

The latter would of course need some protection but much would depend on what it supplied. With a single item one could have no RCD protection as with a bathroom shaver socket.

The RCBO is easy but the isolation transformer method is open to debate it would clearly remove the RCD tripping problem but could introduce dangers so rather than debating which RCD will trip first with a ramp test let us debate the pros and cons of an isolation transformer.

So first is back to poster what power does the pond equipment require? At 200 mA then simple a tooth brush power supply but unlikely to be that low. For a small wattage it may be worth going down the isolation transformer route but as the power increases then so does the price of a transformer so as we reach the 3 kW mark then a RCBO with the extra wiring would be a better option.

I am sure this will cause a debate in fact I hope it does. So although posting power requirements is a good idea I would wait a little before taking action.

 

[JohnW2]

To me in the main it's the carbon track formed when you start to get leakage which is the problem. ... Once it has formed then as it gets damp the leakage shoots up so be it 10 mA, 30 mA, or 100 mA it still takes out the trip.

An interesting theory but, at the start of that progression, how exactly is a, say, 10 mA leak going to result in formation of a 'carbon track'?

Kind Regards, John

 

[ericmark]

To me in the main it's the carbon track formed when you start to get leakage which is the problem. ... Once it has formed then as it gets damp the leakage shoots up so be it 10 mA, 30 mA, or 100 mA it still takes out the trip.

An interesting theory but, at the start of that progression, how exactly is a, say, 10 mA leak going to result in formation of a 'carbon track'?

Kind Regards, John

I would agree the distance makes a huge difference over 10 mm carbon may form over 10 meter unlikely.

But what I am more interested with is not if a 10 mA RCD will trip before a 30 mA RCD but how does the poster progress. As I have stated simple RCBO or more complex isolation transformer? What is more interest to me is what does the team think about the isolation transformer option?

 

[JohnW2]

An interesting theory but, at the start of that progression, how exactly is a, say, 10 mA leak going to result in formation of a 'carbon track'?

I would agree the distance makes a huge difference over 10 mm carbon may form over 10 meter unlikely.

I was really thinking about distances but, rather, about the fact that very high temperatures would have to be generated in order for any of the plastic materials around to degrade to the extent of forming free carbon - and I would not have thought that 10mA (over any distance) would produce adequate temperatures for that to happen.

But what I am more interested with is not if a 10 mA RCD will trip before a 30 mA RCD ...

As discussed, that would only happen (reliably) f the leak increased gradually which some people (seemingly like yourself) don't seem to think is usually the case.

... but how does the poster progress. As I have stated simple RCBO or more complex isolation transformer? What is more interest to me is what does the team think about the isolation transformer option?

An isolation transformer would obviously prevent an earth fault downstream of the transformer affecting any upstream RCD, which might reduce 'nuisances'. However, one could not then ignore the fact that there were potentially water-ridden accessories downstream of the transformer, so it would only be part of the solution. The proper solution is, surely, to make the entirety of the circuit properly 'waterproof' (or, at least a lot closer to being 'waterproof'), then there would be no issues of tripping RCDs.

Kind Regards, John

 

[bernardgreen]

My vote is for an isolation transformer with a split bobbin where the winding for the incoming supply is double insulated from the winding for the outgoing supply.

( this is photo of a 440 to 230 volt used as a example ).

I would also consider a centre tap of the output to a ground rod and a RCD on the output.

With electrics in a pond with a plastic liner there is a risk that the water is insulated from the surrounding earth and in fault conditions the water could become Live and with no path to ground no protective device will operate.

 

[JohnW2]

My vote is for an isolation transformer with a split bobbin where the winding for the incoming supply is double insulated from the winding for the outgoing supply.

I can't help feeling that all this talk of isolating transformers is somewhat sidestepping the issue that it's possible/probable that some of the OPs outdoor installation has been penetrated by water. As I've just written to eric, an isolating transformer would obvioulsy prevent RCDs in the house operating as a consequence, but they would do nothing to address the (possible) primary problem. That 'primary problem' is surely what should be addressed first?

I would also consider a centre tap of the output to a ground rod and a RCD on the output.

That invokes the usual discussion - why, particular outdoors, earth-reference the output of an isolating transformer (even if the CT) when it would probably be safer left 'floating'?

With electrics in a pond with a plastic liner there is a risk that the water is insulated from the surrounding earth and in fault conditions the water could become Live and with no path to ground no protective device will operate.

True - but, as above, if the supply to the pond were totally floating, 'Live water' ('live' relative to what?) would not really represent a hazard, would it? If you earthed any part (one end, or CT) of the secondary of the transformer, then there would be a potential risk.

Kind Regards, John

 

[ericmark]

I can't help feeling that all this talk of isolating transformers is somewhat sidestepping the issue that it's possible/probable that some of the OPs outdoor installation has been penetrated by water.

I would say the reverse the issue is not is there a fault in the outdoor installation but if there is a fault how to stop it tripping the supply to the fridge/freezer.

What I am saying the isolation transformer will stop it tripping the supply to the fridge/freezer.

As to the outdoor installation clearly one has to consider the protection required which could be achieved in many ways but that is a separate issue.

The major one is how to stop losing supply to fridge/freezer.

 

[JohnW2]

I would say the reverse the issue is not is there a fault in the outdoor installation but if there is a fault how to stop it tripping the supply to the fridge/freezer. What I am saying the isolation transformer will stop it tripping the supply to the fridge/freezer. ... The major one is how to stop losing supply to fridge/freezer.

I understand what you're saying - but unless you're proposing it as a temporary measure (which I doubt, given the cost of a transformer) I personally think that it's the totally wrong approach. ... if there is a fault, IMO what one has to do it to locate and rectify the fault - not leave the fault there and use a transformer to 'hide' it from the house's (and fridge/freezer's) RCD!

Kind Regards, John

 

[ericmark]

I totally agree one needs to locate the fault and correct. Maybe I have read the question wrong but what I read the question as was how do I stop a fault in the future from tripping my fridge/freezer supply.

However well one tries to water proof an out doors supply there is always the chance of water ingress and the resulting earth leakage.

I would agree that a 10 mA RCD may not trip before the 30 mA trip so I was looking for a cure.

To my mind there are two methods the independent supply from a RCBO or an isolation transformer. I would also agree with the latter likely an earth rod and RCD are also required and once these are included then the out side supply may still trip. But my understanding was it's not the tripping of out side supply that is the problem but that when it trips it takes the fridge/freezer supply with it.

I looked at isolation transformers for a narrow boat supply and found the price was rather silly. In fact cheaper to transform to 110 then back to 230 than 230 to 230 which of course raises the question why not use 110 volt outside electrics in fact go the whole hog and use 55 - 0 - 55 volt.

However at this point no idea of power required and at 500 VA the cost is far lower.

The point is where does the cost of isolation transformer cross with the price of a RCBO and clearly we don't have the information required to work that out.

So I am not saying the isolation transformer is the best way forward I am just saying it needs considering as an option.

Personally I had a pond with pump no longer running and while in use it never caused my RCD's to trip. Many other faults caused them to trip but not the pond. I have not had a RCD trip while on holiday causing the freezer to fail. I have had a freezer fail while on holiday but it was freezer motor not the RCD which caused it. I was lucky I caught it before everything was spoilt but the question is for one holiday a year does one install isolation transformers or RCBO's or simply have insurance?