RCD tripping with PV

[Martin Tickner]

I have been working my way through an installation looking for why the RCD trips in light misty rain.
I have eliminated all outside installations the only thing left outside the building are the solar panels.
I know little to nothing about the PV installation. So my question is, is there anything on the roof that can cause the RCD to trip?
For info I think there about 5-6 years old.

Thanks in advance

 

[SparkyChick]

As far as I know, the only thing that should be on the roof will be solar related DC cables running from the panels to the inverter. This should be electrically separated from the AC side of the installation, with faults on the DC side being managed/recorded by the inverter.

How often does the RCD trip? You say in light misty rain... if it's just light misty rain I'd be thinking moisture in the air as opposed to water ingress as the later should manifest in any rain.

Can the RCD be reset once it's tripped or does the client have to wait for a while for things to dry out?

What circuits are on the RCD? What test results do you have for those circuits? Is the board a split load board? What test results do you have for the circuits on the other side? Have you checked for loose connections in the incoming supply? Is it an overhead supply? Do the lights flicker if the installation is put under heavy load? Does the RCD trip at any other time? How long has it been going on? Did they have any work carried out around the same time as the tripping started? Have you asked the client to keep a diary of when it trips (including things like weather, time of day, what's going on, had they just switched something on or off)?

Lots of questions I'm afraid, but this is the kind of conversation I have with any client when I'm called out to nuisance RCD tripping.

 

[Charlie_]

Did you ramp test the rcd?

 

[marconi]

If the inverter is the transformerless type then these can cause nuisance RCD trips. Often, giving the inverter its own RCBO - so it has the full 30mA 'allowance' for earth leakage - can remedy the problem and avoid blacking out the house.

Also check to see if the frames holding the panels or the panels themselves are bonded back to the MET. PV arrays can leak ac or dc current (no insulation is perfect and capacitance exists between mains ac and pv dc conductors) or become electrostatically charged. If the frames/panels are not earthed then when it rains or is misty there can be intermittent earthing of them through the roof and building to terra firma. When these current flow to earth or the static discharges they can alter the balance of the currents flowing in the RCD's coil enough to cause a trip.

Last check the state and placement of any dc pv connectors - they should be out of any pools of water, not clamped to metalwork and properly made so no water or mist can penetrate them.

A further question - what is the earthing system? It's not TT is it?

 

[Charlie_]

It will have nothing to do with the PV..
Ramp test the rcd and ascertain what your normal earth leakage is at..
From there you will be able to suss out what you have left over..
Fluctuations on the quality of earth due to the weather may be causing the rcd to trip

 

[Wilko]

Hi - aside from all our usual causes of RCD trips there are known issues with solar panels capacitance effects. This ref explains it and offers solutions, I hope it helps .
https://files.sma.de/dl/7418/Ableitstrom-TI-en-25.pdf

 

[bill01803]

Solar inverters shouldn’t be on a shared rcd because of their shutdown time.

 

[Gary K]

There should be an isolator between the CU and the pv inverter.
It should take seconds to confirm the PV is not causing the trip.

 

[TJ Anderson]

It will have nothing to do with the PV..
Ramp test the rcd and ascertain what your normal earth leakage is at..
From there you will be able to suss out what you have left over..
Fluctuations on the quality of earth due to the weather may be causing the rcd to trip

You can't categorically say that it is not the PV without testing. The inverter itself could have earth leakage or, if transformerless, could be leaking AC through the array via bonding (roof systems almost never are bonded though), or through the wet building to earth when it rains.

Easily eliminated by switching off PV at AC isolator next time.

 

[TJ Anderson]

Solar inverters shouldn’t be on a shared rcd because of their shutdown time.

Yes, the RCD can trip under fault disconnecting the mains supply, yet the inverter could still feed onto the busbar via its MCB for up to 5 seconds under G83 rules, subsequently keeping the fault supplied.