If you have a new account but are having problems posting or verifying your account, please email us on for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact

Intermittent solar PV power drop almost to 0, don't know why?

  • 15-08-2022 3:18pm
    Registered Users Posts: 5

    Hey everyone, I've been a lurker for a while but finally decided to sign up. Nice to meet you all!

    I've been having a few issues in the past few days with my solar PV setup. The power is peaking and troughing massively even with uninterrupted sunlight. I know the power can drop in the heat, and it's been really hot recently, but in some cases it drops by up to 90%, which I don't think is normal.

    Our setup is 5 x Trina Solar 275W panels, on our roof, facing south east, with a 1.1 kWp Sofar 1100 inverter. No battery.

    It started on Wed 10th August. The power got up to about 1 kW, then dropped suddenly and completely stopped for the rest of the day. The inverter continued to report data during that time, but no power was generated. The temperature of the inverter got up to 52C, but that's within its safe range, and it's been hotter than that before, so I doubt that's it. When I got home I didn't notice anything obviously wrong with the setup, but I'm not an expert. There was a bunch of fibreglass wool insulation that had fallen on what I think is a fireman switch for the panels in the attic, which I took off, but I don't know if that's related.

    Since then, we've been experiencing troughs where we can lose 50% to 75% of our power generation, then it recovers, and drops again. This has been in this glorious weather with zero clouds in the sky. Saturday 13th was worst, when the power output dropped to 0. I turned the inverter and panels off and on again, wondering if it was a glitch, and it recovered for a bit, then dropped to zero again. It's like the system can't handle powerful sunlight anymore. Right now, the sun is shining directly on the panels, and the inverter reported 0.00 kW output, then it eventually cycled off (like it does at night when it has no power) and now it's reporting 0.55 kW, which is still lower than it should be.

    I'm at a loss. At the end of July we had the panels and inverter moved slightly to make room for an attic conversion, and I don't know whether that's related. The people that moved them are solar panel installers so I assume it wasn't them that caused this. Also, the system worked fine for nearly two weeks after they moved the panels. I'd write it off as the heat, but a total stop in power generation shouldn't be heat related? So basically, I have no idea!

    Does anyone have any thoughts or insights? Is there something really simple I'm missing? Sorry for the information overload, I wanted to make sure I was comprehensive. Thanks so much!

    Post edited by Deskana on


  • Registered Users Posts: 788 ✭✭✭hick

    sorry to hear of those issues it can be very frustrating. I’m just wondering if there is any possibility of shading? Nearby chimney or trees that could cast any kind of shade even on any 1 panel? If there is any possibility that could take your whole array out, for short and long periods.

  • Registered Users Posts: 1,454 ✭✭✭DC999

    Deffo shade hitting the leftmost panel in that one photo. It's covering half the panel in shade. So you've a shade issue if nothing else. I assume you've no optimisers to help shading? If you're not sure, you'll likely see them in the app you use to check your output.

    And looks like the output drops over the afternoon looking at the graphs (all the graphs show it going from 1kwh, 0.7, 0.5 as the day moves on - bar one where it dropped to zero immediately) - which would appear shade related.

    And does the shade then start to cover more panels at the same time you see the output drop? You can take pics for yourself on a sunny day to see how it moves across.

    Depending on how long you have it, can you compare the output to this time last year, or a similar summer month earlier in the year? Then you can compare if it's the same (and you always had an output issue but only realising now) or this is a new issue, or one that got worse somehow.

    Do you have a pic of where the panels where before you moved them for the attic veluxs? As in was it higher than where it is now so wasn't shaded?

    Others will know more on why 1 graph went from full to zero immediately. And on some of your graphs it goes to zero then back up a few times in a day. I did see other mention something similar could be a faulty connector on another thread. Not sure what thread. Not able to help on that part.

    And also ask the installer that moved it. They should be able to help. Did they also install it? Did they recommend optimisers for the shade or suggest anything else as part of the initial install or move? Don't include the installer names here please, rule is they don't get added.

  • Registered Users Posts: 5 Deskana

    Wow, I didn't know that about shade! I obviously knew that shade on one of the panels would reduce the power output, but I thought it would be proportional (e.g. shade on half of one of the five panels = 0.5 * 0.2 = 10% power loss) and wasn't aware that it could be even more than that. That's good to know. They certainly didn't say that when moving the panels.

    The panels used to be significantly higher, the bottom of the panels were near where the top of the veluxes are now. Up there, there's no way any of them ever got any shade like they do now. I've not tracked exactly how much gets shaded, but I think I've seen it cover the left panel entirely and some of the one next to it. We just took some reduced power output as a tradeoff for the attic conversion, but we didn't think it would be quite this severe.

    The issue is definitely new to the past few weeks. I found data from some days last year where we had uninterrupted sun, and it was a solid curve of power. I'd say the problem is definitely new.

    The installer that moved it isn't the same one that installed it when we moved in. The panels were already there when we bought the house. The new installers did say that we'd expect some reduced power output from shade, but they said it'd be roughly proportional to the area shaded. I've been back on to them but not gotten very far with them yet, hopefully they can come and check that it's all working. Maybe it is as simple as a loose connector.

    I don't know whether that's because there's optimisers on there already but there's something else wrong, or whether they were just wrong about that. I found some notes from that installation but they're pretty vague, I can't find any mention of optimisers but I don't know whether that's because they didn't put them in, or just didn't mention them.

    Thanks for your help so far!

  • Registered Users Posts: 5 Deskana

    P.S. I removed the name of the installer from my original post. I'm sorry, I wasn't aware of that rule. Thanks for letting me know.

  • Registered Users Posts: 1,454 ✭✭✭DC999

    Did the panels come with a new build house? If so, there won't be optimisers I would guess. Would have been the least expensive they could setup based on other threads from people here that got them.

    Can you safely reach the shaded one from the skylight? Hard to tell from pic how far below they are. Could power it all off and disconnect the one that is shaded. So stays bolted to the rail but just remive the cables so connevted to nothing. Others here would need to tell you how to safely do that. Then you have 4 panels. See if the issue occurs. If not, it's shading at least.

    You only need optimisers for the ones that get shade. Costs from 40 each quid to buy - but need them fitted. When I got mine they had to lift all the panels to fit them as they didn't have them on install day. So wasn't something they could just cable tie on easily. But maybe others here know if you can for them more easily while the panel is connected.

    Anyways..annoy the company as you're doing that moved them and ask them to look at what it could be too. If you change stuff they might say you're at fault.

  • Advertisement
  • Registered Users Posts: 5 Deskana

    Following up on this in case anyone’s reading along or finds this thread in the future and wonders what happened.

    There is definitely a faulty wire. The system totally died for a few days generating no power. The inverter was off and the fireman switch wasn’t lit. The attic company foreman was around and tugged on one of the cables in the attic and the fireman switch flickered, came back on, and the system came back to life. So that needs fixing either way.

    However, I do think there is a shade problem as well. Now that the system is back alive, we’re seeing the same drops in the afternoon that correlate with the corner of my neighbour’s roof shading the left panel. It is possible that these power drops are also caused by that same loose wire, because semi-loose wires can give you all sorts of weird and whacky behaviour, but it seems unlikely given that it aligns with the shade.

    I’m coordinating with the attic company and solar guys but it’s slow going. At least, in the mean time, we’re getting a bit of power generation…

  • Registered Users Posts: 2,004 ✭✭✭jkforde

    "I’m coordinating with the attic company and solar guys but it’s slow going."

    I'd be a lot more assertive with them, their f!!k up, they need to sort it asap, highlight the fire risk and it might light one under them!

    🌦️ 6.7kwp, 45°, SSW, mid-Galway 🌦️

  • Registered Users Posts: 1,454 ✭✭✭DC999

    Yeah, and send them an email each (separate email I'd suggest) so there is a formal written trail of your concern, risk of injury + damage and a question mark of liability should anything happen. Phone calls or voice messages are hear say, even text messages likely are if they delete them. Tell them both you expect them to both to come to the house on the same day + same time and you're asking them to stay until it's resolved

  • Registered Users Posts: 5 Deskana

    Overdue update...

    The electrician came back and confirmed there was a loose connector between the solar panels and the fireman switch, which he fixed. He doesn't know anything about solar PV though, so he had no idea about anything else.

    Reading the manual for the specific solar panels I have, it says each panel has bypass diodes in the junction box. Unfortunately, the manual isn't more specific than that, it doesn't say how many, how they're connected, etc. I don't know whether or not that's a substitute for additional diodes or not. The solar guys quoted me around €900 to install bypass diodes on the system and said it's expensive because they "have to get the scaffolding back out", so I think they're just cowboys. I'm pretty much done with them.

    There's still something weird going on with the system. Take this graph from today. It's been sunny all morning. Over the space of an hour, the power started to drop, probably due to the shade on the leftmost panel increasing. As an experiment, I power cycled the system, and now it's consistently generating energy. If you don't power cycle it then it generally recovers on its own, but power cycling seems to speed up the process. I have no idea why power cycling would solve the problem.

    I'm not really sure what to do at this point. The system is only 1.1 kWp anyway. I suspect spending more money on it to stop it from occasionally misbehaving is unlikely to be worth it, especially since when it does have those power drops, it eventually recovers on its own.