Guidance for plug in solar for Electricians

Of course its is unlawful to place these plug in devices on the market without conforming to the requirements of the Electrical Equipment (Safety) Regulations as amended and the EMC regulations with the manufacturer, importer or distributer testing the equipment for conformity with relevant product standards and issuing a  Declaration of Conformity (DoC), or a technical file if a novel device, and marking the equipment UKCA or UKNI. The government have issued an interim  product standard  so people should be looking for a DoC indicating conformity to that standard.

Yes it is a plug in device just like a TV or lawn mower. However it has a subtle and important difference with plug in PV it can be both a supply and a load hence the need for bi-directional RCDs.

JP

My TV is a Class II double insulated appliance, that does not back feed electricity into the circuit which it is supplied from. 

John Peckham said:

Yes it is a plug in device just like a TV or lawn mower. However it has a subtle and important difference with plug in PV it can be both a supply and a load hence the need for bi-directional RCDs.

If BS 7671 does not deal with anything that plugs in to an installation, how should the need for bi-directional RCDs be assured?

Chris

You are quite correct BS 7671 does not deal with things that get plugged in to an installation, that is for legislation and product standards.

BS 7671 does however deal with the consequences of plugging things in to an installation such as EVs, appliances with a DC component and AC appliances with high AC leakage currents.

The need for bi-directional RCDs came in with the emergency Amendment 3 as a safety consideration came to light that could not wait for a full amendment. It was put in for fixed solar installations in accordance with 712. The same risks apply to plug in solar.

JP 

Related to this topic, there was an interesting divergence between the German standards for plug-in solar and British standard which DESNZ developed, on the maximum allowed DC capacity of the solar PV modules.

The German standards imposed a limit on the PV module capacity of 960W peak (DC) when the 800W inverter was connected using a regular Schuko plug. The connected PV module capacity could only be raised to between 960W - 2000W peak (DC), connected to the 800W inverter,  when a special plug/socket is fitted, typically the Wieland RST20i3, I understand normally this would be done by an electrician.

Whereas the UK standards allow an 800W inverter with PV modules of upto 2000W peak (DC) straight into a regular 13A plug/socket.

So the German standards appear to be rather more cautious, only allowing a lower PV module capacity which presumably results in a lower average (mean or median) current injected by the inverter via a regular, uncontrolled socket into the existing circuit. To raise the PV module capacity, which then presumably raises the average current injection into the circuit requires someone, electrician or other, to go through the process of checking the suitability of the connected circuit before installing the special plug/socket, which also then restricts which circuit the inverter/system can be connected to.

I have already spoken to three customers who are considering purchasing plug in solar. All I have been able to say so is that according to the regulations I am working to at the moment I don't believe its possible to make the installation compliant and safe.

It may be over the top, but I am not changing my advice until there's clear advice. If did anything I woukd be leaving myself every exposed if something went wrong.

By the way Napit already guide that type ac rcds should be coded as a c2 when electronic loads are  ir likely to be connected.

John Peckham said:

It was put in for fixed solar installations in accordance with 712. The same risks apply to plug in solar.

John, I am with you there.

I cannot see how the imposition of a plug-outlet and socket-inlet changes the physics, or indeed, the risk.

However, the politicians have made up their (collective) mind.

John Peckham said:

The need for bi-directional RCDs came in with the emergency Amendment 3 as a safety consideration came to light that could not wait for a full amendment.

The trouble is that it was an emergency amendment to solve something that was never really a problem.

As mentioned above, my rooftop solar install has a 30mA type AC RCD, that isn't bidirectional. According to the instructions for my original inverter, it should have been a 300mA RCD, but they obviously just used what came in the consumer unit.

That old inverter had a problem that only got worse over time. If there was a storm overnight, then the sun came out in the morning, then it would trip the RCD. This happened so often that I got into the habit of crawling into the meter cupboard to check it every time it rained overnight.

Eventually that crummy inverter failed and got replaced. But that RCD still works fine, despite being totally the wrong type. It never burned out like it's "supposed to" when fed by an inverter.

Anyone feel confident enough to do a PAT inspection and test on a plug in PV unit, and then certify it as safe to use and attach a sticker to it with your contact details on it?

Well I probably would, assuming it looked like mounted for no water ingress, and mechanically OK, albeit I'd skip any hi-pot test. But then for work I'm fairly used to signing off on the manufacturing end of CE marking and similar def-stan paperwork for things I have designed.  My outlook on 'reasonable levels' of product safety is perhaps a bit different.
Being more than half outdoors I'd suggest a short ( 1 or 2 year) retest interval.
Mike