How can I determine the reason why two of my SmartSolar chargers regularly switch off when battery is full (ie. around noon), even when the inverter is not yet inverting at its limits (see screenshot?)
The 150/100 (connected to VE.Can) works almost always
But both 100/20 (connected to VE.Direct) are switched off
Expected behavior: MPPTs should gradually be limited or switched off when inverter is inverting at full power, or grid feed in limit (8500kW) is reached.
The 150/100 seems to be working correctly, the question is why the 100/20 behave differently.
All three have identical battery settings and are set to BMS (Pylontech VE.Bus) control.
System: Quattro-II (like MP2) 5000/48 with ESS, Fronius 5-1 at AC-Out, Hoymiles at AC-In
Three solar chargers (SmartSolar)
Batteries: Pylontech
Yes, indeed, my mistake when writing, fixed it. The installation was two years ago. However, terminology doesn’t solve the issue, unfortunately… why would that happen? Incorrect logic by BMS? Should I override it by disabling BMS control? Is there a log where I can find more information why output of those two chargers is zero?
What is the PV-Array voltage of the MPPTs showing (Detailed stats) short before they turn off? They possibly drop to something bellow 52,87+5V due to current temperatures and therefore stop producing?
So, your Panels eventually simply not deliver enough voltage when the battery is full and their temperature is at maximum during noon?
Generally speaking, all arithmetic check. 3702-943+2490=5249. So around 5231.
So, those MPPTs are off because there is no need for more power.
You will say that the feed-in limit (8500) is not reached…
Indeed, but take a look at the grid voltage: almost 255V.
At the moment, there may be a “dynamic” limit in place, because of a too high grid voltage.
The first wouldn’t be correct - of course I have a “need” for power, I want to sell as much as I can. So the limit should be feed charging capacity and feed-in limit - not loads.
The second point, yes, I thought about it. BUT the chargers and the BMS controlling them are all DC side? So it would be the inverter somehow limiting inverting, and communicating this to the MPPTs. That should be visible in some parameter on the VRM advanced panel I guess? I thought about the BMS charge and discharge limits, but that seems to be only for the batteries (basically zero A when battery is full). That said, I had a similar problem with Hoymiles Microinverters AC-side switching off, but I changed the grid profile and that fixed it. So this would need to be a setting in VE.Configure?
110% of Un of 230V would be 253V. So quite possible that this is the reason. I will try to set it to 115%
But still interested where charging limitations by the inverter can be read in the detailed logs.
I think the reason is here. The documentation states that +5.1V are required for the mppts to work. When the Battery is reaching 100% (You may want to add the battery voltage graph as well) it will not be a constant 52.8V but have some ups and downs, usually even short spikes, when loading stops.
So that is VERY close to running into issues with to low voltage beeing delivered from the panel.
A test to verify if this is true may be to limit your Charging voltage to 49V, 50V. This will basically not charge your battery to 100%, but would give your mppts 2 or 3 more volt difference to keep running during noon.
Or - as dumb as it sounds - shade one PV-Array with a parasol or something for a day and see if that MPPT is able to continue power output during noon because the panels stay cool and voltage higher.
Which is part of the cause. The hotter the panels get, the lower their voltage will be. If you look at the voltagecurve of a regular day, you’ll see that they start / end high and go to their minimum during noon, when they get most sun and heat up:
Also keep in mind that you have a battery voltage of ~ 52 on one side, but ONE MPPT providing a charge voltage of ~ 58V. If that MPPT is (most likely) closer to the failing mppts, they will experience a higher voltage level “there”, thus having a lower voltage difference than towards the actual battery voltage which most likely is measured by the BMS directly.
Maybe if you just browse through the history, looking for not so sunny days or even rainy days, you will see that they kept working during noon?
Very interesting, thank you. I did not realize it was 5.1V + Vbat, I just had this 5.1V in the back of my mind. It would indeed explain why only the small MPPTs are affected.
Here the battery voltage vs current of the small chargers:
Trouble started at 11:20, the battery voltage was around 52.5V, the PV voltage around 59.6V. That would be above 5.1V, but not much. And there could be some error margin due to the logging interval.
I will limit the charging voltage as you suggested and see how it works out tomorrow.
Well, even if this is not nice, you apparently have enough PV-Power during hot & sunny days, so having the mppts shutdown is not a huge loss.
When days get colder, voltage drop during noon will not be that high anymore and you can utilize the full power throughout the day again.
I’ve checked the MPPT Calculator, finding the right panels for a MPPT 100 with an 48V System is a tricky thing. It’s got to be enough in series to reach >60V during hot days. (your current issue), but voltage also has to be low enough to not exceed 100V during cold days.
You could add another panel of the same kind, and exchange the mppts with a 150V Model. (expensive option)
Maybe - depending on your current system - using shorter / thicker Battery cables could be the cutting edge to reduce voltage drop on the battery side.
Using the next size in PV cabling could reduce the voltage drop of the PV Array.
Eventually you could just replace the panels with some that have 1-2V more (if that then matches the 100V constraint)
But all Options cost some money, tho. Maybe you should first monitor on how much days a year this issue occurs. If it’s 20-30 days during Peak-Summer-Time…
Probably you could resolve this by managing your battery smart: Only charge to ~65% in the morning, and when noon is passed, charge to 100%, so panels have cooled down by the time your battery voltage gets to high.
Yes, the last option is probably the easiest. It can be easily done in NodeRED.
But still, one thing: the Technical Specification clearly says “1b) The PV voltage must exceed Vbat + 5V for the controller to start. Thereafter the minimum PV voltage is Vbat + 1V.”
I do not see a difference of less than 1V which would trigger a shutdown. It could be there, like from a passing cloud or something, but not in the charts as I see them. The MPPTs definitely do report the PV voltage they see, which would already contain temperature-specific issues. And they also should report the battery voltage.
One way to rule out that its the BMS or inverter telling the MPPT to switch off would probably to switch off external control in the MPPT settings
just came into my mind: Every 10th minute (so at :00 :10 :20 …) the MPPT is doing some sort of MPP-Validation, going through a broader range of voltage / current combinations.
My own monitoring software shows that, the VRM stats don’t.
So, during that, they may fall bellow that 1V limit and then turn off.
