Correct selection of RCD with regards to EICRs

[GBDamo]

I'd go C3 with an FI to examine the RCD for DC leakage with an appropriate clamp meter. If the investigation doesn't show DC leakage then C3 (with a note on how DC leakage can have a negative effect on Type AC RCDs and how the client should consider RCD type when making future purchases (EV charger, DC heavy white goods), but if it does show a sizable amount of DC leakage (potentially enough to blind the RCD and prevent proper operation) then I'd C2 then I'd look to replace the RCD for a more suited type.

Think of it as half a dozen sockets on an unused spur off a 32A ring.

You wouldn't, on inspection, say there was no overload at point of inspection, or even evidence of previous overloading, therefore C3.

You would say there is potential to overload an insufficiently protected circuit, therefore C2.

Its about what can reasonably be expected.

 

[Lister1987]

Think of it as half a dozen sockets on an unused spur off a 32A ring.

You wouldn't, on inspection, say there was no overload at point of inspection, or even evidence of previous overloading, therefore C3.

You would say there is potential to overload an insufficiently protected circuit, therefore C2.

Its about what can reasonably be expected.

As you say, an EICR is a snapshot in time and we can only code by what we can see.

If we get an idea of present DC saturation (based on what is installed) then we are in a position to C2 (if saturation is present) or C3 (if none present). We can only code what we can see. Obviously if MIs state an RCD type and they aren't being followed then you could code as a C2 citing 134.1 and 510.3 - By not following instructions them there is the potential for danger, and the C2 then forces that MIs to be followed, whereas a C3 can be 'ignored'

If we are coding on what IS plugged in then does it then not follow that we should code on what MAY be plugged in?

No, because we don't know what the client is likely to buy, all we can do is educate or future proof by changing the RCD (which may need changing as a result of the C2 for 134.1 and 510.3

 

[Pretty Mouth]

Think of it as half a dozen sockets on an unused spur off a 32A ring.

You wouldn't, on inspection, say there was no overload at point of inspection, or even evidence of previous overloading, therefore C3.

You would say there is potential to overload an insufficiently protected circuit, therefore C2.

Its about what can reasonably be expected.

Subtly different to your example IMO.

A cable unprotected from overload could be dangerous in normal use - there are no additional failures required, just equipment drawing a heavy load.

An incorrect RCD has 2 points of failure before it becomes a dangerous situation - DC leakage from faulty equipment, and an AC earth fault. Hence why I'd code C3.

 

[GBDamo]

Subtly different to your example IMO.

A cable unprotected from overload could be dangerous in normal use - there are no additional failures required, just equipment drawing a heavy load.

An incorrect RCD has 2 points of failure before it becomes a dangerous situation - DC leakage from faulty equipment, and an AC earth fault. Hence why I'd code C3.

Fair point, but some equipment produces DC leakage current during normal operation it is not necessarily a fault that caused the RCD saturation. C2 now?

I'm not saying I have the answer and will be having a word with NAPIT before my next EICR, when ever that is.

 

[GBDamo]

As you say, an EICR is a snapshot in time and we can only code by what we can see.

If we get an idea of present DC saturation (based on what is installed) then we are in a position to C2 (if saturation is present) or C3 (if none present). We can only code what we can see. Obviously if MIs state an RCD type and they aren't being followed then you could code as a C2 citing 134.1 and 510.3 - By not following instructions them there is the potential for danger, and the C2 then forces that MIs to be followed, whereas a C3 can be 'ignored'




No, because we don't know what the client is likely to buy, all we can do is educate or future proof by changing the RCD (which may need changing as a result of the C2 for 134.1 and 510.3

I don't disagree with any of that.

However 513.3.3 states "Where DC leakage current is confirmed to be present or may be produced by equipment....."

Now, on fixed equipment, we have a known unknown a case of finding out if DC current leakage is likely under normal operation or a likely fault product.

With socket circuits we have an unknown unknown. That is, we know there is potential for appliances to be used but we don't know what they are and how they could affect DC current levels.

Also many appliances work in cycles so you could have the situation where it is only when appliances are at the correct point in their cycle that they leak DC current. How do you measure this? How do you account for the cumulative effects of various appliances?

To me this doesn't pass the reasonableness test, which is, could it reasonably be expected that a normal person ,acting normally, is highly unlikely to introduce a saturating DC current to a circuit?

When any person can go out and buy an appliance with a plug on it that can satuate an AC type RCD then the answer has to be, no.

 

[Dartlec]

If as it seems, Type A RCDs will become the standard in the next BS7671 which won't be fully in force till 2023, then that would seem the time to introduce this level of consideration for RCD types.

If we can still get a satisfactory EICR on properties without RCD protection at all, then I don't see how we can give an unsatisfactory to one that does - unless there is a clear case like a car charger, PV, etc. where a Type A is mandated.

I'm also a little wary of manufacturers of 'end user' white goods specifying things when it is often to avoid them having to make improvements to their products. As far as the end user is concerned, if it comes with a 1363 Plug, then it fits into a 1363 socket - and is either rated at 3A or 13A.

What would happen if a washing machine manufacturer makes a deal with Wylex and states that only a specific Wylex model RCD must be fitted?

Have there been any real world incidents of injuries caused by this situation as yet? It's hard to grasp how much of a practical risk it is, even though the theoretical risk is clear enough. If there is actually a significant risk out there now, then more should be done than expecting EICRs to always pick up the issue.

 

[GBDamo]

If as it seems, Type A RCDs will become the standard in the next BS7671 which won't be fully in force till 2023, then that would seem the time to introduce this level of consideration for RCD types.

If we can still get a satisfactory EICR on properties without RCD protection at all, then I don't see how we can give an unsatisfactory to one that does - unless there is a clear case like a car charger, PV, etc. where a Type A is mandated.

I'm also a little wary of manufacturers of 'end user' white goods specifying things when it is often to avoid them having to make improvements to their products. As far as the end user is concerned, if it comes with a 1363 Plug, then it fits into a 1363 socket - and is either rated at 3A or 13A.

What would happen if a washing machine manufacturer makes a deal with Wylex and states that only a specific Wylex model RCD must be fitted?

Have there been any real world incidents of injuries caused by this situation as yet? It's hard to grasp how much of a practical risk it is, even though the theoretical risk is clear enough. If there is actually a significant risk out there now, then more should be done than expecting EICRs to always pick up the issue.

It seems to be the latest fad to have an inverted motor on washing machines, certainly the high end ones, with EV chargers, home batteries and PV.

It certainly looks like we are heading this way.

 

[Dartlec]

It seems to be the latest fad to have an inverted motor on washing machines, certainly the high end ones, with EV chargers, home batteries and PV.

It certainly looks like we are heading this way.

Sure - and Type A don't have any disadvantages that I know of, so makes sense to make the change.

But with EV chargers, home batteries and PV there is usually (hopefully) a competent installer who will be certifying that work and can either upgrade the whole installation, or install new consumer unit for that bit with the correct RCD.

Washing machines are the tricky one, since they are almost never supplied or fitted by an electrician, but sold to the customer directly who will just plug it in.

I wonder if the "installation services" that Currys, AO etc offer, check whether the installation is correct for the appliance and notify the customer if there are issues. (hint: I doubt it).

I had to attend a flat earlier in the year where the people "installing" the washing machine wouldn't do so because there was a flex outlet plate under the worktop, not a socket.

 

[Pretty Mouth]

Fair point, but some equipment produces DC leakage current during normal operation it is not necessarily a fault that caused the RCD saturation. C2 now?

I'm not saying I have the answer and will be having a word with NAPIT before my next EICR, when ever that is.

I didn't know that. Yeah, could be a C2 then. What sorts of equipment does produce DC leakage in normal operation?

 

[Pretty Mouth]

What sorts of equipment does produce DC leakage in normal operation?

terrible misuse of the English language