Smart solar mppt rs 450/100 not feed in grid when battery is full

DIY
, , , , ,
22

Should I understand that you have access and knowledge of the Venus OS, as long as you’ve quoted that piece of code from dvcc.py?

Anyway, try first to update the Cerbo to the latest firmware and see if it solves the problem.

If not you can always revert to the previous version.

23

Hi not sure if This will help but enabling svs worked in this instance. ( if not already tried)

24

Hi Dave,

in my system the SVS is “Forced off” and cannot be changed for some reason.

Thanks

25

Ok yes some managed batteries do that, sorry it didn’t help.

26

Sorry one other question, when you put say a 3kw load on the system and there is excess pv power available does the RS 450/100 supply that power to the inverter or is it taken from the batteries?

27

Hi Dave,

when there is a load on the output, the DC PV will supply that power if it is available. It is not coming from the batteries. The only problem I see it is that it doesn’t feed in into the grid when it is needed.

Thanks

28

Hi @gpatti

If you don’t want to update to the newest firmware in Cerbo, I believe that your old (3.13) dvcc.py script from where you’ve quoted above, in the zone of the batteries quirks, for the Pylontech quirk it looks like the one below.

Try to modify the lines for the 15S batteries like in the picture below and see if it’s working.
The Pylons limit the CCL to 0 (zero) when they are full so the charger is stopped.
We are trying to work around this.

2
2871×504 141 KB

29

Hi Alex,

just to be sure : to make the change you are suggesting is it ok to just overwrite the python script and reboot the Cerbo GX?

Thanks

30

You just modify the orange circled part of the code, not the whole python script.

I believe that the original part of the code (on Venus OS v3.13, like you have) looks like:

	if charge_voltage > 30:
		# 48V battery (15 cells)
		return (min(charge_voltage, 52.4), charge_current, feedback_allowed, False)

You’ll change only the above portion of code with:

	if charge_voltage > 30:
		# 48V battery (15 cells)
		capacity = bms.capacity or 100
		return (min(charge_voltage, 52.4), max(charge_current, round(capacity/4.0)), feedback_allowed, False)

Don’t forget to make a backup copy of the original dvcc.py file as a precaution.

And yes, after that you reboot Cerbo or for quickness you can just kill the dbus_systemcalc process and the system will reload it with the new modification.

31

Hi Alex

Unfortunately, this change is not solving the problem. I believe this modification may only affect the charging current of the battery. However, from what I understand, that is completely controlled by the battery’s BMS. My suspicion is that something might not be correctly configured in the process (let me know if you agree). The expected behavior would be for the Cerbo GX to send information to the Smart Solar MPPT 450/100 to attempt to regulate the battery voltage 0.4V higher than the final battery regulation. This way, the inverters can use that energy to feed into the grid.

The firmware has been updated but it doesn’t solve the problem either.

I would like to try the following:

immagine
immagine1087×299 18.2 KB

The actual code is:

immagine
immagine726×134 4.55 KB

I’m not entirely sure about the syntax, particularly for the part checking the SOC.

What do you think?

Thanks

32

First try to see if the increase in voltage is solving your problem.

If you are using dbus-spy, see that in the com.victronenergy.system, you have several keys that begin with /Debug/…
Play a little with SolarVoltageOffset and/or VebusVoltageOffset.
The values are in Volts, so be careful and use values below 1. In other words 0.xx.
Just select the key, hit enter and an input prompt should appear in the lower part of the screen.

This should increase/decrease the charge_voltage variable, in other words /Link/ChargeVoltage.

But just increasing the voltage when the charger is current limited is not enough.
So you may also need to allow the charger to “produce”, so, what I’ve suggested with the code above for charge_current should also be applied.
But you can simulate this by tuning the CurrentOffset on the /Debug/… keys.

So, for starters, if I were you, I will leave the scripts alone, not modify them at all, and, when SOC is 100%, I would set SolarVoltageOffset to 0.4 and CurrentOffset to 30 and see if the feed-in starts.
Beware, raising the DC bus voltage could also influence your battery voltage which is tricky…

But also the point raised by the WhiteCitySolar in the message #4 about grid code is a valid one.
If grid code is properly set, then the feed-in should not malfunction…

33

Hi Alex,

ok, thank for your suggestions.
Yes, I forgot to reply to WhiteCitySolar , I have checked the grid code and it is correct.

Regards,

Giuseppe.

34

Hi Alex,

I have tried all possible settings, but nothing works. For some reason, the feed-in from the DC PV is not allowed. What are the possible settings that could prevent the feed-in from the DC PV to the grid?

Thanks