Our PV has, after 8.5 years had developed an intermittent issue with 'isolation resistance'.

After it's rained the inverter (SMA 3600TL) stops generating and says 'isolation resistance - check generator'- I presume this means that the inverter is detecting earth leakage.

After a period (a few hours) of dry and sun the system starts working again which we have been advised suggests that the problem is water ingress.

Given that we have 1.5 years of insurance left we are keen to get this fixed but thought there must be relevant wisdom from people who have solved this sort of issue before on GBF.

To complicate matters our original installer has ceased trading so we'd be looking at a claim on the insurance scheme (iwi.biz).

What does a long lasting fix to this problem look like - (i.e. are we replacing a few exterior connectors or all wiring?) and how painful is the insurance process?

TiA

John

First I would ascertain if it is covered by the insurance.

Then is the fault repeatable with a hose pipe - it makes fault tracing much easier if you don't have to wait for rain!

If it is covered by the insurance get a professional to undertake the repair (probably with agreement from the insurance co.)

If the insurance co. don't want to know either a professional fixer or diy but eventually you (or the Mr. fixit) will need to get on the roof to isolate the sections/panels and fix the problem.

As an after thought - if the fault is repeatable with a hose pipe then perhaps the easy way to isolate the fault down to a few panels would be to hose a section at a time.

I am suspecting a Slightly damaged wire, connectors are very reliable

Check the insurance policy wording before you do any work as they may not cover systems they have been worked on by others.

Update for those interested or with similar problem in future.

Applied to the insurance company a couple of weeks ago. Typical of insurance companies they don't make your life easy - I needed to apply to them for a specific form before even making a claim and there was a 30 day time limit on making the claim from the first event (so don't hang around if you have a problem).

After that I got contacted by a (surprisingly helpful risk management company processing my claim) informing me that both the insurers and the underwriters had gone under, but that any remedial work was 90% covered by the Financial services compensation scheme and suggesting a solar repair expert with attached quote.

Said Solar guy took two hours to tighten all the connectors on the roof this morning replacing one that was corroded and poorly attached.

If this doesn't fix it then it gets expensive (scaffold and disassembly) so glad I went down the insurance route as they are clearly on the hook if more action is needed. I'll find out next time it rains heavily.

After the initial claim this was actually surprisingly straightforward.

Posted By: tonyTightening connectors, were they the push and click ones

Presumably the bits which needed tightening were the caps on the back which need to be pretty tight otherwise water gets in and corrodes the contacts. That's far more likely than faults in the wire.

Posted By: jms452Solar guy suspected that the fitters has just hand tightened the screw on caps on the back of the connectors rather than using a tool to get them fully tight. This was on the tails connecting the inverter to the factory made connectors on the panels themselves. Apparently one of the wires came out when he was retrieving it to check if it was tight enough and it was corroded inside.

The connector he replaced was very similar to the one in Ed's photo.

He also tightened all the screws in the DC isolator's wire clamps as apparently these can work loose over the years due to temperature cycling (but ours seemed ok).

I was quite surprised by quite how tight my electrician made the screws in our sockets. He said this was because they would all become looser over time due to thermal effects. So I've made a point of doing things up extra tight ever since, and checking them after a year or two where possible. I also believe his theory about CU fires, which he reckoned were mostly due to the removal of the second screw on connections, so changing the boxes to metal to contain the fire were a pretty brain dead 'solution' to the problem. Reinstating the second screws to eliminate the problem being a more sane approach, IHHO.

I think there is another factor involved with the fires in CU. When It first came to my attention a few years ago now I was suspicious that it could be connected (no pun intended) to the roll out of smart meters. I managed to find on line the London fire Brigade stats on fires in consumer units and compared them to the roll out stats of smart meters. To me there was a distinct connection as meter installation increased so did CU fires. The reason I conjecture is that the installers of meters are only trained in installing meters and not allowed into CU. Consequently as they disconnect the meter tails and remake them in the new meter the very stiff 25mm tails waggle about and disturb the connection in the CU which as already mention are only held by one screw now in modern CU. Additionally the meter tails have changed certainly in my previous house the tails were multi-stranded fine strands which easily compressed and formed a larger surface area connection beneath the 2 securing screws. Now the tails are larger diameter strands do not compress as easily and add in the variability of the hardness of the copper between makes you now have a perfect situation for a fire in a CU. Changing the regs to metal CU does not prevent the fire just helps to contain it. It was the wrong solution to the problem as a consequence of lack of proper analysis of why the fires were occurring.

👍