We have an installation with two SFK 300Ah LFP batteries, and overall they have been working well.
These two batteries are being monitored by a Victron SmartShunt, and the shunt installation is correct. When monitoring shunt current, it always shows the combined current of both batteries, so we are confident the shunt is accurately capturing total current flow.
The issue is with the state of charge over time. The last time the batteries were fully charged, they were then taken off charge and left to run overnight. By the next reading, the SOC reported by the SFK batteries was 61%, while the Victron SmartShunt was showing 74%.
That is a pretty large discrepancy. Do you have any thoughts on why there would be such a difference in SOC between the batteries and the Victron shunt? Than You
Post a screenshot of your SmartShunt settings. Key ones for lithium that can affect SOC are the capacity setting, 1.05 or lower Peukert coefficient and 99% charge efficiency factor.
And yes, first of all the smart shunt settings should be correct, this is what I wrote. But if I can’t find such information for a BMS, or even which type of BMS it is, then I have that feeling, that this might not be given in any datasheet or manual I found, because no one ever measured or validated this.
OGPS
(Ed @ Off-Grid Power Systems - offgridps.com)
6
SFK uses (currently) JBD BMS’s with their own custom firmware on them and the shunt resistors are fairly sensitive and accurate. I would expect them to stay within 1-2% SoC readings.
OP: make sure you didn’t accidentally turn on the battery heaters using the switch on the lid. That will most definitely cause a large deviation in SoC. Though, this time of year I doubt that’s the issue.
The smart shunt will not match the BMS reading. The external shunt can not track internal use by the battery such as heaters, usage by the bms and other things. This is normal, we recommend relying on the BMS shunt as it will be more accurate than the external shunt.
No, not really. The state of charge (SOC) isn’t the result of a direct measurement of a physical quantity, but rather is estimated more or less accurately from multiple measurements and assumptions (e.g., the assumption of a Peukert value). And all measuring devices have their inherent inaccuracies. A deviation like yours is perfectly normal.
My JK inverter BMS is notorious for delivering quite inaccurate SOC readings. Last night I got low-voltage alarms even though less than 300 W were being drawn from my battery, and it was still showing over 50% SOC. In reality, it was probably already down to about 10%.
The SOC calculations of the Victron SmartShunt are generally considered relatively reliable. Much more reliable, in any case, than those of my JK inverter BMS.
The most important thing is never to interpret a measurement as a representation of absolute truth. All measuring devices have one thing in common: they invariably display incorrect values. The only difference is how far they are from the truth.
And then you have to understand that the SOC is not a physical quantity that can be easily determined. It is an electrochemical quantity and can only be measured with reasonable accuracy by destroying the battery.
What is extremely relevant for us is that the unavoidable errors in SOC determination accumulate over time, as long as the battery is not reset to a somewhat arbitrarily set 100% SOC by a defined full charge. The SOC is, without exception, always only a rough guideline. For example, you should always include the voltage and the charging or discharging current when discussing your SOC. If, like mine last night, the battery shows 50% SOC but only 50V, then it’s clear: the SOC is far from reality.
OGPS
(Ed @ Off-Grid Power Systems - offgridps.com)
11
That is the correct answer for SFK batteries. The same is not true for most of the Amazon batteries. And as Tom says, it’s true of the JK-BMS, too. But, the JK is fine if the batteries can reach 100% every couple of weeks.
