Hello everyone,
I observe every day that my Victron system continuously charges my battery.
But at 90% it jumps to 100%?
Is it correct that way?
With my electric car, it takes just a long time to go from 80% to 100%.
is this normal or do I have a problem?
As a battery I have Pytes V5. I load via Mppt and DC.
I’m not familiar with this particular battery brand, but after a quick Google search, I discovered that it’s a lithium-iron phosphate (LFP) battery. Based on this, I assume that the state of charge is being calculated by the battery itself and then transmitted to the Victron system over CAN.BMS. If that’s the case, I would suspect that the internal BMS of the battery is not properly calibrated.
Is this battery new or have you had it for a while? If it’s new, this behavior wouldn’t be surprising and will normalize over time. However, if you’ve had it for a while and frequently charge it to 100% to balance the cells, then this behavior is a bit unusual.
Hope that helps.
@Floki, Check this post for some potential solutions: Shunt / battery monitor jumping to 100% or reporting a high SOC when the Volts are low and/or the BMS has disconnected - #2
Maybe wrong settings on the shunt. Watch this video. https://youtu.be/mEN15Z_S4kE?si=8OVpcerIonesbXro
@Floki
Did you program the system per the guide?
https://www.victronenergy.com/live/battery_compatibility:pytes#minimum_battery_sizing
The widget you show is not soc but is voltage. So an interesting one to jump up like that. A more useful comparison is voltage to SOC to check if target programmed (or bms requested) voltage is met and kept for the absorption time.
The second one to check is the bms requested CCV. There should be a widget if comms are working.
FYI i have no personal experience with Pytes. These are tips that is use with other brands.
Hello Alexandra
Yes, my settings are according to the instructions - I compared them again.
As a newbie, I’m not allowed to upload more than 2 pictures.
Here is a compiled one including SOC.
Yesterday morning it was constantly charging to 90%. Then it jumps to 100. Then I unloaded (auto) in the next charging phase again to 90% and then jump to 100%.
So based on all the feedback, it’s not normal and I have a problem?
Then I contact my salesperson. Maybe he set the values ​​for the Pytes batteries incorrectly?
Thanks Ledlight
For me the Pytes are the battery monitors.
Does everything run automatically and you can’t set anything else?
Thanks Jeremy
I didn’t install a shunt. BMS occurs directly through the phytes. Will tell my supplier. Maybe he did something wrong with the battery and volt settings
Thanks Peter
Yes, the battery is new.
Hi @Floki
My Pylontechs are doing almost the same. Look below.
But what I have found, looking at the charging curve, is that somehow the time spent in 90% is consistent with a time the SOC is evolving to 100% on the same curve it evolved up until 90%. Just that the SOC is not reported until near 100%.
Also, regarding at the consumed amphours value, when it hits 90%, all the consumed amphours are already put back into the battery.
That lead me to believe that somehow, that period from 90% to 100% is an “absorption like” phase inside the battery.
I know you have Pytes, but, in my opinion, don’t worry, as many aspects could be shared among Chinese manufacturers, including firmware and how the firmware behaves.
Alex
My Pylontechs also do this, at 90% SOC the charge current is reduced during the final absorption and balancing phase and the SOC stops increasing and then it jumps up to 100%. There is nothing you can do about this as it is in the BMS.
So as I and others said, since the batteries are new I wouldn’t worry about it for now. The BMS inside the battery will calibrate over time to produce smoother SOC curve with more precise values.
The chart shown was from the GX. If you chart a widget from the BMS itself (usually 512 ID), is it the same?
If so, drift happens. And drift on new batteries can be worse due to balancing.
In any case, it is BMS generated, and if it doesn’t settle with time and decent charging (at full SOC), then I would speak to the manufacturer as they really need to adjust the BMS profile and firmware.
If your comment was intended for me… If not sorry.
When you use a connection between battery BMS and Cerbo, these two are identical, as inside the Cerbo’s scripts
there is a direct copy: System SoC <= Battery SoC.
See below on my system, where System’s SoC seems to be hidden, but it’s there, below the Battery’s SoC.
More of that, I’ve observed that if you charge the battery slowly, then that jump is not so obvious if at all…
Comment was for the OP. On occasion the BMS is not set as system monitor.
High charge rates will push up the battery voltage which can convince a lazy BMS to synchronise.
This is the nature of multi-stage charging and LFP batteries. The jump up in this case coincides with the charger kicking in, pumping in bulk amps with voltage rising in result. LFP cells have a very flat discharge/charge curve. This means very small changes in resting voltage can result in very large differences in SOC. To really dial it in you need to factor in cell drift and internal cell resistance for your specific cells. Even after that, you’ll get erroneous results during times of high charge and discharge.
Quite usual behavior with new batteries, that is just the BMS SOC synchronizing to 100% as it sees the battery voltage going up at the very end of the charging phase. You’ll see that from time to time, but with synchronized cells after the first couple of charges it should become more accurate.
