Hello,
I have a Multiplus II GX with Venus OS 3.52, additionaly 2 MPPT trackers 150/35. Battery is a DIY-48V 280Ah with a JK Inverter BMS. The BMS is connected via CAN and acts as the controlling BMS for the Victron system.
On the Venus I use a self-developed Python script that reduces the maximum charge current at low or high battery temperatures. I write this values via dbus to: com.victronenergy.settings → /Settings/SystemSetup/MaxChargeCurrent
In the last months, this worked fine. But a few days ago, MaxChargeCurrent was set to 5A, but the system charges the battery with more than 7A continuously from the MPPTs. The Multiplus couldn’t feed into the grid, because the grid power was already near zero watts.
Normally, the MPPTs should reduce their power under these conditions. But at that day, it didn’t work.
I checked the charge current limit in the Venus GUI under DVCC, and there, the 5A limit where visible.
To avoid damaging my battery, I reduced the charge voltage at the DVCC, which resulted in a reduced MPPT power as expected.
Your graph, whilst pretty, shows no actual values for anything.
Do you have other loads (or is the system doing anything else) at the time? Can the be accounted for with system self consumption? Are you sujg anything like SCS or SVS? Is the syste. Battery monitor the JK itself?
Are you using ess with feedback? @grua has a most valid point! Optimised with battery likfe without ? Keep batteries charged all this changes behaviour in the system.
As an FYI most battery control is voltage and current control especially when slowing charge.
@LX:
I have no further DC-components. As the the JK-BMS is the controlling BMS, the Victron system has to regulate on the measurement value of the BMS. In the past, that worked fine, but not on that day.
My DVCC-settings:
It seems that a few minutes after the BMS-current exceeds maxChargeCurrent, the MPPTs left the MPP of around 100V. The voltage then rose to higher values. Maybe they reduced their current but because of an unknown reason, the Victron system regulated the current with a constant offset on maxChargeCurrent of about 2 to 3Amps.
Here the raw data from influxDB without aggregation:
So the charge current limit seems to work in general, but with an offset.
For further information:
The maxChargeCurrent is unstable in this condition, because another self programmed regulator trys to limit the max cell voltage at cold conditions to 3.4V, which worked in general:
I’m not sure, what you mean. Deepl translates this to german like “Is the jk bms not programmed to repair itself?”
Maybe you ask for the protection features of the BMS?
The BMS is only capable to shut charging down at undertemperature or cell overvoltage. The BMS cannot limit the charge current by itself or by CAN-signal for low temperatures or high cell voltages.
Because my battery is located outdoor, I have to take care to limit the charge current at low temperatures without shutting down the system completely in the winter.
So I developed this algorithm in Python.
X-axis is temperature, y-axis is ChargeCurrent.
The limits of the battery cells depends on the SOC, Eve spezifies two curves. My script limits the charge current according to the green curve.
So… I have several system with JK bms
Through DVCC they control the system. The system obeys… I have not had really cold temps to deal with though. On the cooler days, i do see the CCL lower.
My topic is not the battery or the BMS. My topic is, that the victron systems doesn’t regulate the charge current as requestet.
Is this a bug or is there an explainable reason for it?
The word bug can be used when it is a supported battery.
But with the JK BMS is it not on the compatibility list. So there is that. Then you have customised a few things. So it is no longer a stock system either.
Since I personally use JK bms on several systems and haven’t experienced issues. So some things need to be explored. one of them is BMS set up, since that is infinitely customisable and something small can be overlooked.
How is your bms connected to the system, with can? Have you tried other methods? There is a custom driver for other methods, i have tried several and they work better than the JK 2017 CAN.
