Hi,
I have noticed that the SoC drift / ghost discharge is still present in 3.70 when peak shaving is enabled. When PS is disabled (‘only above min soc’), the issue disappears. I see that this issue has been reported in the past on multiple BMS brands (01/2025). It seems that the BMS is reading a ghost negative current caused by something that is happening on the DC bus (ripple,..). The battery voltage does not drop, which proves that the measured current is not real and SoC should not drop.
In this 7 hour period (PS active, min soc reached) the voltage does not drop so no real power draw. BMS is measuring a phantom -1.7A draw which causes the calculated SoC to drop:
On days where the battery is at min soc for a long time, this drift can be as high as -10% which makes it very difficult to manage the system correctly. When PS is disabled, the issue disappears and the accuracy of the SoC from the battery is very high. I have been making sure the battery charges to 100% every couple days so the BMS recalibrates SoC to 100%, but this should not be necessary and makes it impossible to for ex. lower the charge current above 90% because the SoC value is way to low after a couple of days. This puts unnecessary stress on the batteries.
Any input or tips would be much appreciated, thx.