Batrium SOC - reported value

batteryshed77

New member
Joined
Nov 12, 2020
Messages
15
I'm having another issue which I'm hoping is a simple setting somewhere from being fixed - but is leaving me perplexed.

I have a 600Ah pack, at 43.6V nominal (12s Li-ion @ 3.63V/cell based off the datasheet) that's 26kWh.

I'm regularly charging my pack between what should be ~20% - 90% capacity, but Batrium will indicate charge/depletion much quicker than it should (a least based on the SOC values it broadcasts) By this I mean SOC goes from 100% -> 0%, meanwhile my batteries are still above nominal voltage.



1614705701179.png

Here are my shunt settings. I've set a capacity of 600Ah, but you can see a 'daily cumulative' charge of 928Ah but a Daily e-meter of only 15.7kWh

This broadly aligns to what I'm seeing, it seems to think I've put in 928Ah (40.4kWh @ 43.6V) whereas in reality I've only put in 15.7kWh (360Ah @ 43.6V)

Why does Batrium think that 15.7kWh is 928Ah? It is clearly measuring the correct amount of energy (15.7kWh), knows the current, knows the voltage - so why 928Ah? I'm pulling my hair out.
(For black and white confirmation, I currently charge my pack at up to 108A for 4 hours per day. This is the only charging window, and charging tails off for the last hour - so I'm confident 360Ah / 15.7kWh is correct & 928Ah is not.)



Does this matter? Well not at the moment, but I'm looking to implement a level of control based on SOC - and at the moment that isn't viable without a reasonably accurate SOC.
 
Does this matter? Well not at the moment, but I'm looking to implement a level of control based on SOC - and at the moment that isn't viable without a reasonably accurate SOC.

Can't help you with Batrium, just want to say in general, that's exactly why I am basing the main controlling always on direct Voltage readout.

This method came a little bit out of fashion because of LFP chemistry (there with its flat curve and resting behavior, you need to include a lot of brain power into the prediction), but other chemistries are so much easier to SOC-predict.

Your Batrium does obviously calculate something wrong, and there will surely be someone who can tell you what is not correct, but as well with other Coulomb meters, they tend to drift if you do not hit physical limits (100% or 0%) at each cycle. (It's like you, trying to walk from your home to workplace with closed eyes and just counting steps. And now say you are not walking all required steps each days, just change direction back and forth. After a few days, you won't be able to tell where exactly you are on your way). So they need to be corrected from time to time, and that again can only be done by a direct voltage readout translated to an SOC.

I don't know what level of control by SOC you will be looking for, but I highly recommend going back to basics and calculate it from a voltage reading.
 
My experience (with Batrium SoC) is consistent with @RolandW's. I'm using 18650, so the voltage curve is sufficiently variable to manage things by voltage. @daromer swears by Batrium SoC - and some have suggested 'electrical interference' as a cause of weird Shunt info... but I've never been able to track this down.
 
Back
Top