Gavin286

I think I'm going to decrease the charge rate of the batteries so they take almost the whole day to charge, and see if this happens over the course of the next few days. Note also that it only seems to happen during hot days - if that's also another possible contributing factor.

Thanks for your input. I preferred P1000's comment as it was better news I know for a fact that my undercounter lights in the kitchen pulse when this is happening, so I'm sure it's also affecting every electrically connected device which is the entire house. What would the process be to check (disconnect and re-connect) the MC4 connectors? Will it be enough to just turn off the PV switch on the inverter?

Should I post some pics of the settings? The inverter is exactly the same as a Deye if that helps...

No sir, not that I can tell. The batteries are always charged fully before 12 midday and it happens after that. Which setting should I check or rectify?

Thanks for your input, @I84RiS. I also suspected this to not be normal or good for the system overall to be so erratic. Frankly, the installer and I are not on good terms any longer after I had to force them to remove 2 panels from my setup due to the voltage being too high (as per the flipping spec of the system) and him outright ignoring them. Perhaps someone on the forum here would be able to assist me with a site visit, or recommend someone that could just go over the system to see where the issue might be. Of course I don't expect this to come free of charge and will happily pay for the assistance!

The load is stable with a pool pump and normal loads (no washing machine) of about 1.5kw. This may fluctuate by 10w or so, but not close to 1000w which the PV supply seems to do. Again, it's only in the afternoon that this happens. Between midday and say 4pm. It's an 8kw Magneto (Deye) inverter with 18x 455w panels, 9 per string in series. My nominal voltage is around 385v that max's out at ±405v. A picture says 1000 words so a video must say even more! Here is a link to the video I took: https://youtu.be/9KSCOp9ZOCg

Hi everyone, I have a weird issue that I'm not sure exactly how to classify or interpret it. Now, I can't speak for this happening every day as I am back at work and it only seems to happen during the peak heat of the day, but my PV supply seems to fluctuate wildly with supply. This is after batteries are charged. The PV will be supplying the load of lets say 2kw, then in a second it will drop to 20w, then a few seconds later, 1.5kw, then again a few seconds later 600w. This causes the source of power to fluctuate quickly between PV, battery and grid to supply the load which is stable. I have a video clip of this happening but I'm not sure where to upload it to for someone to view. I'd like to find out why this is happening, and how to rectify it, as surely its not good for the electronics to be constantly between the various sources of power? Thanks so much!

My 2c on this and from direct experience. I have an 8kw Magneto inverter (which is Deye). I have (had) 20x 455w with 49.5 VOC panels over both mppts in series. I ran into issues with over voltage and random irregularities with heat and current. Since removing 1 panel from each mppt I no longer have issues and run nominally at between 380v and 420V with 9 panels per string in S. Moral of the story, stick under the mppt listed voltage of 425v max. My electrician INSISTED that the installation would be fine since he ignored the 425v on the mppt spec and only looked at the 500v max rating of the PV supply overall. This 500v is for spikes from CEE.

Agreed. So far I haven't required a load of more than 6kw at a time, so there is good future expansion, but at a cost of perhaps too many panels. Oh well, its done already Thanks for the reply!

Hey folks, So just an update on this issue if anyone runs into the same or similar issue. My inverter was set to max charging current of 100a which is fine since I have 2x 5.1kwh batts that can both charge at 100a each. I have discovered that by dropping that max current to 50a, I do still get a voltage increase (however not a spike anymore), and the PV supplies the load straight away. Note also that my charge time has only increased by roughly 20 minutes! Previously it would bottom out to ±100w and then take about an hour to slowly ramp up to match the load requirement. I assume with my 0.5% knowledge of electricity, perhaps this fits into the laws of electricity IE voltage / amperage etc have to be friends to work nicely together And today the electrician is coming to remove 1 panel from each string down to 9 per string in S. This should drop the voltage down by 45v or so to keep within spec of the 425v max recommended per mppt.

We have a guard hut outside our property and their network switch got fried as well. Not sure if its linked to the issue I experienced, as they are connected via ethernet, as well as shared mains power (municipal feed only, not through inverter). I think we did get lucky with the AC and DC surge protection we had installed last week! I did read on another forum where a guy had his brand new and sealed Netgear equipment taken out by lightning storms - I assume through EMP type interference. Insane!

Morning everyone. This past Saturday we had a lightning strike. A few seconds before the zap, there was an audible crackle from the plugs / wiring. The inverter reported a DC over_current event and reset itself after a few minutes. Earth leakage did not trip. My network switch was the only electrical device that was destroyed. Note that its a PoE switch and does have devices externally mounted (cameras and wifi access point). Can't confirm if the linked devices are dead. I do have AC and DC Phoenix Contact surge protectors installed and are still functional. 1. How do I protect the replacement switch / devices in the future? The plug is earthed and allegedly working correctly 2. Why did earth leakage not trip? 3. How can I confirm that the surge protection was installed correctly? Thanks!

I for one am also weary of these installers who think they know it all, and refuse to listen to recommendations even by the supplier. IE MY electrician.

Thanks for pointing that out. I'm still experimenting as I go along, so I'll look into that. I'm also treading carefully with load shedding, so its all a juggle to maintain charge / usage / savings in one go...

Thanks so much for the interest! We are now running 20 panels on one string each into the 2 MPPT's, but we now have voltage issues - seems like 1 panel too many. The electrician is insisting its not the case, and the suppliers are only now trying to resolve the issues. Hopefully this week we'll have a final resolution. RE the battery run time - the 2 batts don't quite get us through a full night so I did set a top-up to occur between 23:30 and 01:30.

Don't say the T word with Sonoff tagged in the thread 😆

Good call. Its cost only a few extra R to run the neutrals, so even if you NEVER use them, they're there for possibly future use.

Lets hope you ran those neutrals beforehand...

Thanks. I wonder how much I could sell the panels for...

I have 10 on one roof separated by a small elevation, and the other 10 on the top side. All 20 are on the same side (North facing). The issue is that the 1 set of 10 is on a roof that has a vaulted ceiling and there is no access to the under side of the tile. The other 10 are in a traditional ceiling that is easy. The PV only supplies max 6kw which is my demand. I have to turn stuff on to use the power generated. We got extra panels to compensate for cloudy / rainy days, but losing 2 panels I don't think will make that much diff. FYI both strings are in S so I would assume its a matter of un coupling 1 panel on each string and bob's your uncle.

So an update on this: I've managed to get the suppliers to confirm that the voltage is too high. The electrician is still in denial and I'm basically having to force him to re-string the PV. Now here's the thing: Is it difficult to split the 2 sets of 10 panels into 2 strings each? As in labour / cost intensive? I was thinking I would just remove 1 panel from each string and sacrifice ±700w and either return or sell the 2 panels. the VOC of each panel is 49.5

They are wired 10 panels in series, 10 per mppt...

Just out of interests' sake - what happens if I don't split the strings and I continually hit the max V?

Thanks for the help guys. I assumed it was 500v max. I'll get the electrician to split the strings into 4 - 2 per mppt.

My panels are the 455w units