Good morning, all. Need some help. I have three 5kw Axpert Invertors making up a 15kw parallel setup. 2 of are connected to 9x325watt panels and one is connected to 12x325watt panels making it a total of 30x325watt = 9.75kw. This setup has been running for over 2.5 years producing on average 7kw and has gone up as high as 9kw. But as of late i'm struggling to see 5kw from the PV. 

I have looked for damaged or burn marks, i have washed my panels to no success. If i measure at the invertors it still gives me the 118V input as the panels are rated 38.8V but the string of panels only generated highest lately 1.5kw as displayed on invertor.

 

What else could be the problem

I think your best bet might be to test the individual strings of 3 panels where they join (combiner box?) and perhaps unfortunately each panel individually. The problem with testing strings in parallel is if one of the strings is faulty and produces 0V, you will still measure the same voltage across all of them at the inverter.

Thank you, will start doing that. 

 

37 minutes ago, WayneE said:

Thank you, will start doing that. 

 

FYI I recently had a panel fault which was not easy to find as it was still producing half the voltage, turned out to be a melted contact in the junction box on the back of the panel.

They're not called "solar" systems for nothing. Everything degrades in the sun - mainly due to UV. So connectors will corrode, at the back of the panel and the bayonet type connectors. Even wire itself can degrade over time, as moisture creeps in and starts attacking the individual strands in your wire.

But I've been wondering about fault finding a panel arrays too. Especially on panels mounted flat on roofs. I'm surprised someone hasn't come up with a solution for this yet. Or maybe someone has?

4 hours ago, WayneE said:

But as of late i'm struggling to see 5kw from the PV. 

Was this a sudden drop or did it deteriorate over time?
Had a similar issue with my 2 paralleled strings. Suddenly only got half my normal. Since then I have had to take out 3 MC-4 connectors already where the two strings were joined. I suspect just low quality connectors used by the installer. One had actually melted 

This was my issue on a 4 month old panel. Looks like a hot connection at the negative ribbon from the panel. I measure 40V on both ribbons and 20V if I move the negative probe below the melted part. I tested the diodes and they are fine. Judging from the positive ribbon there was not much love put into assembling it. Luckily the panel is still under warranty so I will be returning it soon.

3 hours ago, zsde said:

I suspect just low quality connectors used by the installer. One had actually melted 

Very likely a poor connection forming a DC arc. Was the connections crimped with a MC4 crimper? That is as close as you will get to starting a fire.

4 minutes ago, iiznh said:

Was the connections crimped with a MC4 crimper?

I shall assume it was. But as said, I had already removed another 2 that had been arcing before this one. They all connected to the Y splitter/combiner.
Just cut them off and replaced them with ferrules for now.

 

On 2022/12/20 at 12:31 PM, zsde said:

Was this a sudden drop or did it deteriorate over time?
Had a similar issue with my 2 paralleled strings. Suddenly only got half my normal. Since then I have had to take out 3 MC-4 connectors already where the two strings were joined. I suspect just low quality connectors used by the installer. One had actually melted 

Even low quality MC4 connectors will not melt. Only poor crimping or loose connections will cause small arching and heat which will cause melting of even a decent MC4 connector.

3 hours ago, I84RiS said:

Even low quality MC4 connectors will not melt. Only poor crimping or loose connections will cause small arching and heat which will cause melting of even a decent MC4 connector.

I'm not so sure. If you find 2 mating MC4s with no resistance going into each other before the seal and lock they will for sure generate heat. I have found some heating up and changed one at a time per polarity to find the bad side. Temp was measured with a contactless thermometer. This remains the easy way to find hot connections long before arcing takes place. 

17 hours ago, Scorp007 said:

I'm not so sure. If you find 2 mating MC4s with no resistance going into each other before the seal and lock they will for sure generate heat. I have found some heating up and changed one at a time per polarity to find the bad side. Temp was measured with a contactless thermometer. This remains the easy way to find hot connections long before arcing takes place. 

Agreed, if it does not fit tightly the same issue will occur.