Strange issue with solar power production

[Hanno Labuschagne]

Hi everyone

I noticed a strange problem with our solar power system's production in Pretoria today (29 May 2025). The specs are as follows:

• 6 x 545W JA Solar panels
• 5kW Deye hybrid inverter
• 2 x 5.12kWh Dyness DL5.0C batteries

Early on in the day, I noticed that solar production was picking up slower than in the past few days despite clear skies and similar irradiance.

By 11:35, just as our 3kW geyser turned on, the peak generation was only at 2.06kW, despite the battery still being far from full. Yesterday at the same time it was 2.68kW. See images below for comparison:





I'm not sure what other appliance was turned — maybe a microwave or kettle — but when our consumption picked up to over 4kW, the solar production all of a sudden returned to similar levels as yesterday.

Peak was around 2.7kW while yesterday's peak was 2.8kW.

This continued while the geyser remained on and consumption was above 3kW. However, as soon as the geyser went off, production dropped unusually low again.




Despite the batteries still being far from full, production was 2.05kW by 13:25. Yesterday it was 2.55kW at the same time, while the batteries were at a higher state of charge.

I have not changed any settings on the inverter between these days.

The batteries are set to charge up to 75A. The 75A works out to roughly 3.8kW with the battery voltage, much higher than our system supports. That should therefore not be the problem.

Any thoughts or ideas? I suspect a battery issue so have logged it with the installer to troubleshoot. Don't think it would be panels if the production returned to normal with the higher demand.

 

[Hanno Labuschagne]

System has never had this issue in more than a year of running. Only recent change was addition of another 5.12kWh battery but has been working perfectly well for more than a week. Was able to charge up to 100% before 15:00 each day with geyser going on.

 

[My_King]

 

[Flanders]

If it was a gradual decline I'd say check the panels for dust. People seem to underestimate the diminishing results caused by dirty panels - especially winter on the highveld. I cleaned mine 3 weeks ago and they're already coated and peak showing decline of more than 10%.

In your case it seems not to be the case though with the sudden jumps.

 

[Ernabr]

If it was a gradual decline I'd say check the panels for dust. People seem to underestimate the diminishing results caused by dirty panels - especially winter on the highveld. I cleaned mine 3 weeks ago and they're already coated and peak showing decline of more than 10%.

Ours get washed every two weeks for that reason

 

[chrisc]

I was my panels 3 or 4 times a year. I can't say I notice any improvement, but I prefer them clean. The gardener uses a long pole with mop and detergent and then a high-pressure washer for rinsing

Our main problem is aphid residue. More on one side, the higher ones hardly get dirty

 

[Ernabr]

I was my panels 3 or 4 times a year. I can't say I notice any improvement, but I prefer them clean. The gardener uses a long pole with mop and detergent and then a high-pressure washer for rinsing

My husband was told not to use detergent, only a soft brush and water. He bought a Gardena extendable pole with a soft brush. Uses minimal water that way. Oh, he's also installed bird spikes on the top edges to keep the hadedas from cr@pping on them.

 

[thehuman]

Looking at that graphs it looks like batteries refuse to take full charge current.
Maybe try do a full shutdown and power up.

 

[Flanders]

I was my panels 3 or 4 times a year. I can't say I notice any improvement, but I prefer them clean. The gardener uses a long pole with mop and detergent and then a high-pressure washer for rinsing

Our main problem is aphid residue. More on one side, the higher ones hardly get dirty

You're in WC? GP gets dusty AF in winter. It's most apparent on the panels and is a real problem.

I'm actually looking to invest in this to make my life easier in the long run:

Electromann SA

It’s never been easier to get the latest gadgets and gizmos to enhance your home, upgrade your car or get the security and surveillance equipment you need.

electromannsa.co.za

 

[Ernabr]

You're in WC? GP gets dusty AF in winter. It's most apparent on the panels and is a real problem.

I'm actually looking to invest in this to make my life easier in the long run:

Electromann SA

It’s never been easier to get the latest gadgets and gizmos to enhance your home, upgrade your car or get the security and surveillance equipment you need.

electromannsa.co.za

We're in Mpumalanga. Just asked my husband why he preferred the brush. "Grit gets trapped on a squeegee and it becomes like a scouring pad, which is the last thing you want on your panels. the scratches will trap more grit that will be all but impossible to remove."
Wash brush, cleaning brush for sensitive surfaces, rotatable, PET bristles

 

[leon.davibe]

Hi everyone

I noticed a strange problem with our solar power system's production in Pretoria today (29 May 2025). The specs are as follows:

• 6 x 545W JA Solar panels
• 5kW Deye hybrid inverter
• 2 x 5.12kWh Dyness DL5.0C batteries

Early on in the day, I noticed that solar production was picking up slower than in the past few days despite clear skies and similar irradiance.

By 11:35, just as our 3kW geyser turned on, the peak generation was only at 2.06kW, despite the battery still being far from full. Yesterday at the same time it was 2.68kW. See images below for comparison:

View attachment 1823968

View attachment 1823970

I'm not sure what other appliance was turned — maybe a microwave or kettle — but when our consumption picked up to over 4kW, the solar production all of a sudden returned to similar levels as yesterday.

Peak was around 2.7kW while yesterday's peak was 2.8kW.

This continued while the geyser remained on and consumption was above 3kW. However, as soon as the geyser went off, production dropped unusually low again.

View attachment 1823972

View attachment 1823973
Despite the batteries still being far from full, production was 2.05kW by 13:25. Yesterday it was 2.55kW at the same time, while the batteries were at a higher state of charge.

I have not changed any settings on the inverter between these days.

The batteries are set to charge up to 75A. The 75A works out to roughly 3.8kW with the battery voltage, much higher than our system supports. That should therefore not be the problem.

Any thoughts or ideas? I suspect a battery issue so have logged it with the installer to troubleshoot. Don't think it would be panels if the production returned to normal with the higher demand.

Some systems with coms will pick up tha max charge rate bassd on amount of batteries seen as online and if one battery is off-line

It may reduce the max charge

The battery may accept charge

Just that the coms failure means the inverter works as if only one battery is online even if both are available it thinks one is and reduces the amps maybe ?

 

[PhireSide]

Some systems with coms will puck up tha max charge rate bassd on amountnof batteries seen asnonline and if one battery as off-line

It may reduce the max charge

The battery may accept charge

Justbthatbthe coms failure means thje inverter works asnif only one battery is online even if joth are available it thinks one is maybe ?

 

[Koosvanwyk]

Hi everyone

I noticed a strange problem with our solar power system's production in Pretoria today (29 May 2025). The specs are as follows:

• 6 x 545W JA Solar panels
• 5kW Deye hybrid inverter
• 2 x 5.12kWh Dyness DL5.0C batteries

Early on in the day, I noticed that solar production was picking up slower than in the past few days despite clear skies and similar irradiance.

By 11:35, just as our 3kW geyser turned on, the peak generation was only at 2.06kW, despite the battery still being far from full. Yesterday at the same time it was 2.68kW. See images below for comparison:

View attachment 1823968

View attachment 1823970

I'm not sure what other appliance was turned — maybe a microwave or kettle — but when our consumption picked up to over 4kW, the solar production all of a sudden returned to similar levels as yesterday.

Peak was around 2.7kW while yesterday's peak was 2.8kW.

This continued while the geyser remained on and consumption was above 3kW. However, as soon as the geyser went off, production dropped unusually low again.

View attachment 1823972

View attachment 1823973
Despite the batteries still being far from full, production was 2.05kW by 13:25. Yesterday it was 2.55kW at the same time, while the batteries were at a higher state of charge.

I have not changed any settings on the inverter between these days.

The batteries are set to charge up to 75A. The 75A works out to roughly 3.8kW with the battery voltage, much higher than our system supports. That should therefore not be the problem.

Any thoughts or ideas? I suspect a battery issue so have logged it with the installer to troubleshoot. Don't think it would be panels if the production returned to normal with the higher demand.

What is the charge current shown on the LiBMS screen of the inverter. Likely that the BMS is throttling max charge amps?

 

[leon.davibe]

View attachment 1824025

Cleaned up my typos

Some systems determine charge amps based on batteries online

ie lets say 50A per battery
And thus max 100A because 2 are online coms cable gets some corrosion and connection fails

So the battery is on-line accepting charge but inverter thinks it is off-line

Then reduces max charging amps
When you add load the inverter accepts more solar since it has a place to dump it

Edit : unllug and plug the coms cable a few times , as that scrubs the copper pads/pins. See if problem dissapears

 

[Hanno Labuschagne]

Cleaned up my typos

Some systems determine charge amps based on batteries online

ie lets say 50A per battery
And thus max 100A because 2 are online coms cable gets some corrosion and connection fails

So the battery is on-line accepting charge but inverter thinks it is off-line

Then reduces max charging amps
When you add load the inverter accepts more solar since it has a place to dump it

Edit : unllug and plug the coms cable a few times , as that scrubs the copper pads/pins. See if problem dissapears

Will give that a try, thank you. The system did show charging up to 60% before the production faded in the evening as usual. Would it maybe not have stopped reporting at 50% if just one battery was detected? BMS screen also shows it detecting both batteries with maximum charge at 75A.

Also, is it normal for the two batteries to have different SoCs?

 

[B-1]

Will give that a try, thank you. The system did show charging up to 60% before the production faded in the evening as usual. Would it maybe not have stopped reporting at 50% if just one battery was detected? BMS screen also shows it detecting both batteries with maximum charge at 75A.

Also, is it normal for the two batteries to have different SoCs?

View attachment 1824059

No, that normally means there is a problem.
How often do they spend some time at full charge?

As was suggested shut everything down in the correct sequence then start it all up. Configure them to stay on 100% charge for a day so things can equalise and then see if things go back to normal.

 

[leon.davibe]

Will give that a try, thank you. The system did show charging up to 60% before the production faded in the evening as usual. Would it maybe not have stopped reporting at 50% if just one battery was detected? BMS screen also shows it detecting both batteries with maximum charge at 75A.

Also, is it normal for the two batteries to have different SoCs?

View attachment 1824059

If both show as on-line likely the problem isn't coms failure related

i don't see any red light so assume none of the batteries iare in protection mode preventing

So can't think of another scenario that would cause it not to charge at full speed at the moment

Sometimes batteries can drift appart a bit and balance out at the top

The lights can be deceptive as 49% and 51% will be 2 lights and 3 lights respectively

Edit: the soc is calculated based on batteries online

So if one is online at 50% it would be 50% of 5kwh. And if two is online both at 50% then it would still show 50% it would just be 50% of 10kwh

 

[RedViking]

Have you tried unplugging the electric car, some people forget they have one.

 

[PsyWulf]

Will give that a try, thank you. The system did show charging up to 60% before the production faded in the evening as usual. Would it maybe not have stopped reporting at 50% if just one battery was detected? BMS screen also shows it detecting both batteries with maximum charge at 75A.

Also, is it normal for the two batteries to have different SoCs?

View attachment 1824059

Can you change over to Voltage mode rather than BMS and pull their stats?

 

[Hanno Labuschagne]

No, that normally means there is a problem.
How often do they spend some time at full charge?

As was suggested shut everything down in the correct sequence then start it all up. Configure them to stay on 100% charge for a day so things can equalise and then see if things go back to normal.

Typically at 100% for about an hour or two each day.

Seems to be back to normal today. Did a full shutdown of the inverter and disconnected the battery and then restarted everything — following the proper order of steps one way and the other, of course.

It seems to be picking up as on previous days this morning. Already at 740W by 07:45, nearly identical to Wednesday. Yesterday it was at 500W at the same time.

Wondering if maybe some battery protection did kick in yesterday as I did go very low on SoC.

These Dyness batteries have a 90% DoD so I set the minimum to 14% just to be safe. Hit that level yesterday but today only went down to 17%.

The newer battery seems to discharge slightly faster than the old one, which I've heard can be an issue with batteries in parallel.

Going to move that minimum SoC up to 20% for some time for an additional buffer.