Plug in balcony solar UK , just a Lidl question

[JohnW2]

https://electrical.theiet.org/wirin...-march-2024/bidirectional-protective-devices/

“The concern highlighted by BEAMA is that a voltage present on the outgoing terminals of the protective device, either due to the device operating in the event of an earth fault or by use of the functional test button, could cause irreparable damage.”
That’s where i read about 2 issues with unidirectional rcds ,whether true or not

I'm not at all sure how an RCD could suffer as the result of it operating because of an earth fault. For a start, that would merely mean that (if it were a unidirectional RCD) the electronics would remain powered after the trip (which is the situation which the devices normally happily live with for years or decades)and, in any event, the device tripping (hence disconnecting inverter from grid) would presumably very rapidly result in the inverter shutting down, wouldn't it. ... but, as always, maybe I am'missing something'!

The second issue, relating to use of the test button with a unidirectional RCD during export, is the one we have been discussing at length in this and other recent threads - and would seem to be an unavoidable potential risk, albeit far from quantified.

 

[mikehammer69]

I didn’t understand it myself

What that second issue if any could be

 

[JohnW2]

I didn’t understand it myself ... What that second issue if any could be

That makes two of us, then ... but it would be a little worrying (and would make us rather sceptical about anything which came from BEAMA) if the 'experts' at BEAMA were imagining an issue which doesn't actually exist - so, in some senses, I might hope that you and I are both 'missing something'!

 

[ericmark]

Thank you, @mikehammer69 the report states

The product standard for RCBOs states that if it is necessary to distinguish between the supply and load terminals, they shall be clearly marked, for example.

So with RCBOs, we can, it seems, identify between the two types.

The 'typical' residual current circuit-breaker (RCCB) is an electromechanical device, however, electronic RCCBs also exist. RCCBs for consumer units are in the form of a two module-sized device. These devices are not usually marked in and out, and therefore are bidirectional.

So it seems this also covers the RCD, so all we need is like when we get a meter change, and they mark the fuse size on the DNO fuse, we can have the RCD/RCBO type marked on the CU. So from date of this instruction, it will take 5 years for all rental accommodation in England to be marked up.

quite apart from the fact that it would be a bit rich if a buyer did not see this until after he/she had bought to product

This is clearly a problem. We don't tend to read instructions before purchase, however we have the same problem with ovens above or below 3 kW and a large label saying how many kW seems to be enough.

 

[JohnW2]

"The product standard for RCBOs states that if it is necessary to distinguish between the supply and load terminals, they shall be clearly marked, for example."

So with RCBOs, we can, it seems, identify between the two types.

I'm not so sure about that. Most, if not all, RCBOs have a 'flying lead' for the supply-side N (with the load-side terminals marked as such) - so it is clearly possible to determine the intended orientation of installation - but I don't really see how that can be taken to indicate whether or not a device is 'bidirectional' (I suspect that a high proportion of those ion service are not).

You quoted:

"The 'typical' residual current circuit-breaker (RCCB) is an electromechanical device, however, electronic RCCBs also exist. ...."

That statement is surely way behind the times, hence a serious understatement, isn't it? Whilst the earliest RCCBs were, indeed, purely 'electromechanical', I thought that for a very long time virtually all of them used 'electronics'. Is that not the case?
Your quote goes on to say ...

".... RCCBs for consumer units are in the form of a two module-sized device. These devices are not usually marked in and out, and therefore are bidirectional."

I also have serious doubts about that statement. The vast majority of RCCBs in-service in CUs are, indeed, 2-module-sized devices which rarely, if ever,have 'in'and 'out' markings - but it has always been my understanding (reinforced by the diagram usually on them) that they were unidirectional devices. Is that not the case?

So it seems this also covers the RCD, so all we need is like when we get a meter change, and they mark the fuse size on the DNO fuse, we can have the RCD/RCBO type marked on the CU. So from date of this instruction, it will take 5 years for all rental accommodation in England to be marked up.

Hmmm. I'm not at all sure about that, either. If, per what you quoted, the requirement is that (if it is necessary to distinguish between the supply and load terminals), then they should be clearly marked to that effect, I suspect they meant 'clearly marked by the manufacturer', not by an EICR inspector. Apart from anything else, in the absence of any 'manufacturer markings', how is the EICR inspector going to ascertain for certain whether the device needs such a label/marking? Furthermore (as emboldened text above), what you quoted only requires such 'clear labelling/marking' IF the device is unidirectional - so absence of such labelling could either mean that it's bidirectional or else that no-one has yet considered applying a label!!

I think that all of this (IF it were proposed) would need a lot of 'thinking through'!

 

[ericmark]

Sorry, my point is the markings are normally obscured but the CU cover. The "they shall be clearly marked" does not say on the normally visible part, but I suppose one could say not clearly marked if on a part of the device not normally visible?

 

[JohnW2]

Sorry, my point is the markings are normally obscured but the CU cover. The "they shall be clearly marked" does not say on the normally visible part, but I suppose one could say not clearly marked if on a part of the device not normally visible?

The material you quoted said that the product Standard said that (IF distinction is required between supply and load terminals), the terminals should be clearly marked. Those terminals will never be visible when the device is installed in a CU or other enclosure, so I presume the requirement must relate only to the device itself, 'as sold' (not 'as installed').

 

[ericmark]

Good point.

 

[mikehammer69]

The Beama article states devices not marked as directional are bidirectional
Applies to the below

 

[JohnW2]

The Beama article states devices not marked as directional are bidirectional

That might be true in the future, possibly even true in terms of items being sold now (although I rather doubt it), but that surely is not true historically (i.e. in relation to the millions of RCCBs currently in-service in UK CUs), is it?

On the basis of what I have always understood to be the arrangement of the test buttons (as illustrated by the diagram that most of them bear), it has historically always been my understanding that they were 'unilateral', despite the absence of any marking related to 'directionality'.

I think most of us are agreed (unless we are all 'missing something') that any RCD will function satisfactorily as an RCD regardless of 'which way around' it is wired (i.e. which way current is travelling through it)- the only potential issue we have thought of relates to what happens, or may happen, when the test button is pressed.

 

[mikehammer69]

I did say “applies to below” which you didn’t quote

 

[JohnW2]

I did say “applies to below” which you didn’t quote

You did, but I didn't really understood what that meant. The 'below' doesn't seem to give any indication about 'dates' (i.e. whether it is talking about in-service or future devices).

 

[ericmark]

Wiring Matters Issue 99 - March 2024 also refers to Bidirectional protective devices.

This in turn gives links to the BEAMA Technical Bulletins, and it does seem if nothing to say it is unidirectional with arrows, load, in, out etc. Then considered as being bidirectional.

The only problem is one needs to remove CU cover to see those markings, so not really any good for plug and play.

It seems V2G compatible charger needs a G99 to be accepted, so what ever system is used to allow power into the grid, and also reading the German stuff, they have to inform there equivalent of the DNO to export.

If a EV socket can have a special V2G compatible, why not a BS 1363 socket having a S2G compatible. (solar to grid)

 

[JohnW2]

This in turn gives links to the BEAMA Technical Bulletins, and it does seem if nothing to say it is unidirectional with arrows, load, in, out etc. Then considered as being bidirectional.

It does say that - but, hardly ever (in the last ~40 years) having seen one with such markings, I'm coming to wonder whether (despite what I previously thought) significant numbers of 'unidirectional' ones actually exist, or have ever existed??

I've just looked at one of my 'spares' on my shelves. Since I don't recall ever having had to replace an RCD (MCCB) in any of my CUs, I suspect that it is probably well over 20 years old, and it is certainly typical in appearance to what I recall of virtually all those which I've seen. It bears absolutely none of the markings you mention. The nearest to a 'marking' of any terminal is (as you can see in photo) an "L" adjacent to the top left terminal (close to the test button) - that presumably being the terminal into which the supply-side L would normally be connected...

The diagram on the side is interesting ...

... although none of the terminals on the device are actually numbered, the fact that they are referred to as 1/2 & 3/4 at one end of the device (top of diagram) and 2/1 & 4/3 at the other end (bottom) seems to suggest/imply that it is meant to be 'bidirectional'. More to the point, the third pair of contacts on the trip mechanism (for the 'test circuit') (one of the methods I've mentioned for addressing the 'test button issue') mean that (provided the device operates when the test button is pressed - see below) the potential 'test button issue' will not arise, in which case I presume it can be regarded as 'bidirectional'. The diagrams below show the current paths whilst the test button is pressed, both before and after the device trips (if it ever does!).

In creating those diagrams, I was reminded of the fact that almost all the methods of addressing the potential 'test button issue' (to make devices 'bidirectional') I have mentioned would probably 'fail' (in the event of prolonged depression of the test button) if (presumably due to 'being faulty') the device did not trip when the button was pressed. In fact, the only approach I've mentioned that would still 'protect' in the absence of a trip would be some sort of direct (mechanical or electronic) restriction of the 'test current' to a very small duration. However, maybe those designing these 'bidirectional' devices felt that it didn't matter that they could 'become unsafe' if the device didn't trip when the test button was pressed - since in that situation one would hope (but could not be certain!) that the device would be replaced, anyway??

Whatever, if standard devices were essentially 'bidirectional' more than 20 years ago, it would seem unlikely that there would have subsequently been any (seemingly 'detrimental') changes to producing 'unidirectional' ones.

The only problem is one needs to remove CU cover to see those markings, so not really any good for plug and play.

Indeed so - which is why I'm not sure that any of these discussions or 'statements' about the characteristics/markings/whatever of RCDs are actually very relevant to the issue of 'plug-in solar inverters'.

If a EV socket can have a special V2G compatible, why not a BS 1363 socket having a S2G compatible. (solar to grid)

I suspect that there is no reason - but the whole idea of 'plug-in-solar' would presumably be that it could be plugged into any 'ordinary' BS1363 socket, isn't it?

 

[mikehammer69]

This video talks about a second issue with newer rcds. It appears to be an issue affecting the electronic components when rcd trips.

Also mentions the test button issue