Degradation of domestic solar panels

JKenH

Title change 17 February 2022, to address what may be a wider issue. Originally I had been under the impression that domestic solar does not suffer significant degradation and my own problem was perhaps an isolated issue (electrical possibly) peculiar to my system. From another comment posted I suspect there may be a more widespread problem with degradation than what I was aware of and invite others to share their experience good or bad with both panels and inverters.
Ideally perhaps you could post statistics to support any degradation after adjustment for annual variation. In fact if you are happy your system is performing without any degradation it would help others like me if you can show annual (or even monthly) output with a note of your area and age of your panels.

if you have individual monitoring of panel output I would be interested to see how it does vary across a roof to establish some sort of norm. Thanks.

Originally posted as 

Difference in output from solar PV strings

I originally posted the first paragraph below on a different thread but then decided to start a separate thread and request some help, please.

I am having a problem with reduced output from my panels installed in 2018. I have panels on 2 roofs each with its own inverter. The panels on each roof are arranged in 2 identical strings. I can see from the app that monitors my inverters that one string on each inverter produces less than the other but the difference can vary from zero to around 25%. Strangely this is happening on both strings. We have tried swapping the strings round to see if the problem is in the inverter but it isn’t. I would like to think it is something simple but it seems it isn’t. 

Below is a screenshot showing 2 readings for one of the inverters taken 6 minutes apart. (It is taken from an email I sent to a guy who had been round earlier that day who said the panels might need cleaning) There is hardly any difference between the 2 strings in the first reading and around 10% 5 minutes later (my inverter stats refresh every 5 minutes or so). There is absolutely no shading issue that could account for the difference. There was some cloud around but the sun was either in or out so in the second reading all the panels were in full sun and in the first in cloud. 

The photo below is just included to demonstrate that the problem occurs at higher outputs and was taken at 12.47pm whereas the ones above were around 5pm both on 9th September.

The shot of both strings producing the same power was an exception. The difference is usually 10-20% but can be more at low outputs. The above stats are for the inverter on my west roof. I have a similar problem with my east roof.

Has anyone any ideas, please?

Edit, photo of the west roof attached taken at 11.13am on 23 August. The chimney being on the north side doesn’t shade the panels.

QrizB

Copied from the other thread (where I'll delete my post):

That is a bit bizarre. Are all the panels nominally identical? It sounds almost as though you've got two intermittently-dodgy panels, one on each of your two bad strings.

Asking the same question as I posed to Cottage_Economy; are they still under warranty and can you get the installer back?

JKenH

Thanks.

Installer is no longer trading. I contacted MCS but was told there was no form of redress. The panels are nominally identical, all installed at the same time. 

QrizB

That's a shame. The installation I purchased last year came with an insurance-backed guarantee, which might prove useful in the event the installer fails. I don't know how long this has been an MCS requirement (it seems to be one now, but wasn't in 2012) so there was a chance you had one too.

The only thing that immediately comes to mind would be to take down the offending strings and test the panels individually to identify the faulty ones (or whatever other fault there is, eg. water in a plug somewhere). That's going to be a pain in the whatsit.

MikeHEQA

If you know what panels are, the VOC of panel on back of panel or look online ie mine are (14 panels split on two roofs) JA Solar JAM60S20-380 V0c - open circuit voltage is 41.62... multiple by number of panels for your max voltage 7x 41.62= 291.34volts of String.  Check online app output if you can.  your max voltage x amps equals your power out P= V x I

Check your inverter output stats for up to 500-550 or 550-600 volts max volts not all inverters are equal.   If all good then with your low voltage by your stats, something is bringing down the string. Test with multimeter, but that means shutting down system and taking apart each panel from string to test every panel individually which is not easy.  As you know shadding will bring down the operating voltage. 

JKenH

I took this photo of the back off the panels when they were installed. Max voltage is 32.0 which tallies with the 216v I am seeing. The 183v on the other string led me to believe that one of the other panels isn’t working. (6x32v = 192 v) but if that were the case then wouldn’t that drag the whole string down. 

My next thought was what if the strings have been wired 6 and 7 to each inverter? I have 14 panels on the west roof and 12 on the east but then why would I see similar watts (and volts) on odd occasions? I know they can’t be wired that way as one roof is in sun when the other in shade so the difference between the outputs of the strings would be far greater.

I know the problem hasn’t been like this from the original installation date as last years output was 5348w compared to 5926 in 2020 (I first noticed an issue around September/October in 2020] and 5885 in 2019. I can also see data from a house 100yds away online and my peak outputs on fine summer days were down when his were similar to the previous year. 

ispookie666

I'm a bit lost. You have differences between the strings which are variable, have I understood this correctly ? 

These are pics from my two system (current house and old house). 

The first pic is from the current house, showing the the front array.  There are no shading and showing variability. The gap does tend to widen at times. 


The second pic is from our old house. Again big difference in panel output and one of them has clearly been falling off for the last 18months



I think you are seeing normal variability and panel deg.

JKenH

Yes, I was interested in your degradation when you first posted it and might even have commented. Perhaps everyone suffers from some degree of degradation and I am only noticing it because I have identical specification strings to compare. SolarEdge does make it a bit easier to monitor if you have a dodgy panel and I wish I had that facility. Also with SolarEdge, if one of your  panels is misbehaving it doesn’t drag the others down with it. Your worst panel is 11% down on your best. I might just have similar degradation on one panel to you but it impacts all the panels on that string.

I would be interested to see how the variation between your panels varies from day to day.

I suppose I should regard the degradation I have as normal but I have got used to reading on this forum that degradation on solar panels is minimal and so was surprised to see around 10% overall. Someone else commented on a different thread a while back that the annual variation in insolation masks any degradation so maybe quite a lot are suffering to a degree and just not aware of it. 

There is a house 100 yards away with solar output posted online that enjoys exactly the same pattern of sun and cloud as I do so it is useful to be able to make a comparison of the year to year change on my system compared to that one.

Edit: My total annual output in 2021 was 90.9% of my 2019 figure and 90.2% of the 2020 figure. The comparable stats for the house down the road were 92.4% and 93.5%. Looking at it on this basis things don’t look quite so bad. 

However my best day in 2021 (41.25 on 2nd June) was however only  90.2% of my best day in 2020 (45.72kWh on 30th May) whereas for the house down the road the 2021 figure was 96.6% of the 2020 figure on the same days. 

QrizB

JKenH said:

I took this photo of the back off the panels when they were installed. Max voltage is 32.0 which tallies with the 216v I am seeing. The 183v on the other string led me to believe that one of the other panels isn’t working. (6x32v = 192 v) but if that were the case then wouldn’t that drag the whole string down.

Have you ever - carefully - checked Voc for your strings? 7 panels should be just shy of 280V DC.

JKenH

QrizB said:

JKenH said:

I took this photo of the back off the panels when they were installed. Max voltage is 32.0 which tallies with the 216v I am seeing. The 183v on the other string led me to believe that one of the other panels isn’t working. (6x32v = 192 v) but if that were the case then wouldn’t that drag the whole string down.

Have you ever - carefully - checked Voc for your strings? 7 panels should be just shy of 280V DC.

I wouldn’t have a clue how to do that.

ispookie666

These are two screenshots a few days apart from the system installed in 2015