Fitting a 100ma RCD on Solar PV to solve nuisance tripping.

[Octopus]

Just the circuit to the inverter. Which is on its own 2 way consumer unit.

And then SWA cable - so no rcd required imho

 

[marconi]

https://www.electricalsafetyfirst.org.uk/mediafile/100117576/Best-Practice-Guide-3.pdf

See page 12/20 Para (iv) for advice on RCD protection .

 

[marconi]

From my research it seems many Sunny Boy inverters are transformerless which means that there is usually no galvanic isolation between the PV dc input and ac output. If you tell me the model number of the inverter I will delve further.

I thought of an experiment you could do which might determine whether the leakage current waveform is ac or dc. At the moment the RCD is type B which can detect the waveforms ac 50Hz, ac 50Hz pulsed, Smooth dc, and ac greater than 50Hz into kHz region - see page 7/21 of:

http://www.doepke.co.uk/download/Techpub-16

The type AC RCD commonly used in Consumer Units can only detect ac 50Hz - it is blind to ac 50Hz pulsed, Smooth dc, and ac greater than 50Hz into kHz region.

What you could try then is inserting a type AC 30mA RCD upstream of the current type B and then observing what happens every morning to the type AC one -does it or does it not trip?

Just an idea for your consideration which might cast a bit more light on the problem by giving information on the waveform of the leakage current (as well as my # 15).

 

[netblindpaul]

Don’t forget that a type AC RCD may simply be blinded by DC such that it will not operate at all, as in will not sense AC or, DC leakage.

 

[marconi]

Re #24: What you could try then is inserting a type AC 30mA RCD upstream of the current type B and then observing what happens every morning to the type AC one -does it or does it not trip?

I am not suggesting removing the Type B.

 

[Deleted member 26818]

What would be wrong with removing the RCD altogether?

 

[marconi]

Re# 26 - see #22

 

[Deleted member 26818]

Re# 26 - see #22

Been there, done that.
What would be wrong with removing the RCD altogether?

 

[marconi]

Answer yourself - I will be interested to read your reasoning.

I have provided a reference which explains why it is right an RCD type B is intalled because there is no dc - ac side galvanic isolation. It may well be the manufacturer of the in-use Sunny Boy inverter says one is not required - one for the OP to clarify - though for me the presence of the type B RCD indicates that the installer has read and abided by the manufacturer's instructions.

 

[Deleted member 26818]

Answer yourself - I will be interested to read your reasoning.

I have provided a reference which explains why it is right an RCD type B is intalled because there is no dc - ac side galvanic isolation. It may well be the manufacturer of the in-use Sunny Boy inverter says one is not required - one for the OP to clarify - though for me the presence of the type B RCD indicates that the installer has read and abided by the manufacturer's instructions.

The reference you have provided indicates that where required, B type RCDs should be used. i.e. circuits of locations containing baths or showers, socket-outlets or cables concealed in walls.
There is nothing there stating that the supply to or the output from the inverter requires RCD protection.
The OP says that the RCD protects the circuit to the inverter and that the circuit is wired in SWA.

 

[marconi]

At page 12/20 para (iv) of:

https://www.electricalsafetyfirst.org.uk/mediafile/100117576/Best-Practice-Guide-3.pdf

(iv) Where a microgenerator having a d.c. source does not incorporate the equivalent of a transformer providing at least simple separation between the d.c. and a.c. sides, an RCD installed for fault protection by automatic disconnection of supply or for additional protection (I∆n <_ 30 mA) must be of a type that will operate as intended in the presence of d.c. components in the residual current. (This does not apply where it has been established – such as from a specific written statement given by the inverter manufacturer – that the inverter provides galvanic isolation between the d.c. and a.c. sides that prevents it from feeding d.c. current into the electrical installation.) Note. A Type AC RCD will not fulfil the above requirement. Depending on the level and form of d.c. components, an RCD (where required) will need to be of Type A to BS EN 61008 or BS EN 61009, Type B to IEC 62423, or Type F to IEC 62423. However, in the case of a PV power supply installation, Regulation 712.411.3.2.1.2 of BS 7671 stipulates that the RCD (where required) shall be of Type B.

 

[marconi]

duplicate

 

[Deleted member 26818]

At page 12/20 para (iv) of:

https://www.electricalsafetyfirst.org.uk/mediafile/100117576/Best-Practice-Guide-3.pdf

(iv) Where a microgenerator having a d.c. source does not incorporate the equivalent of a transformer providing at least simple separation between the d.c. and a.c. sides, an RCD installed for fault protection by automatic disconnection of supply or for additional protection (I∆n <_ 30 mA) must be of a type that will operate as intended in the presence of d.c. components in the residual current. (This does not apply where it has been established – such as from a specific written statement given by the inverter manufacturer – that the inverter provides galvanic isolation between the d.c. and a.c. sides that prevents it from feeding d.c. current into the electrical installation.) Note. A Type AC RCD will not fulfil the above requirement. Depending on the level and form of d.c. components, an RCD (where required) will need to be of Type A to BS EN 61008 or BS EN 61009, Type B to IEC 62423, or Type F to IEC 62423. However, in the case of a PV power supply installation, Regulation 712.411.3.2.1.2 of BS 7671 stipulates that the RCD (where required) shall be of Type B.

Yep, as I said.

 

[marconi]

Now can we have your argument and reasons why the RCD can be dispensed with please?

 

[Deleted member 26818]

It’s not needed and It appears to be nuisance tripping.

 

[marconi]

Why is it not needed please?

 

[Deleted member 26818]

Why is it not needed please?

Because.

 

[marconi]

Please provide an argument with reasons using your knowledge of electrical engineering, the wiring regulations, awareness of the law for the safe use of electricity and appreciation of the installation in question.

 

[Deleted member 26818]

Please provide an argument with reasons using your knowledge of electrical engineering, the wiring regulations, awareness of the law for the safe use of electricity and appreciation of the installation in question.

I’ve asked you why the RCD can’t be removed.
All you’ve done is provided a link to a source which states that any RCD in the installation will have to be a B type.
Something that I am well aware of.
You have not given any reason as to why it can’t be removed.
Now (though you’ve not answered my question), you want me to explain in depth why an RCD is not required.
The simple fact of the matter, is that from the information provided, the RCD is unnecessary, it’s not required as far as BS7671 is concerned and at the moment it is causing a nuisance.

Now your suggestion, is to install an incorrect RCD to see if that will or will not also trip.
So if it trips, what then?
If it doesn’t trip, what then?
Who pays for the installation of this unnecessary and incorrect RCD?
Most of these inverters have inbuilt RCD protection and they also show error codes if there’s a fault.
Having an external RCD, is probably preventing the inverter from logging any fault.

 

[Wilko]

Hi - looking at a Sunny Boy installation guide (on this sunny day) and they seem to recommend 100mA rather than 30mA. I can't comment on the requirement to have an RCD, as I just don't know .
http://files.sma.de/dl/5690/SB3300TLHC-IEN094121.pdf