Other Parts Discussed in Thread: BQSTUDIO, GPCRA0, GPCCHEM
I am using the BQ40Z60EVM-579 eval kit with a 4Series-2Parallel LiFePO4 battery pack, each individual cell is rated at 3.2V, 3200mAh. There is no manufacturer branding.
I have been unsuccessful in getting the “Update Status = 0x04” to increment after several cycles. From my understanding, this is preventing the system from Updating “QMax”, the “Ra Tables” and “Cell Balancing Time”.
From other forum posts I believe the cause is a poor “Chem ID” match for this battery pack. (Currently set to closest match: ID 0400 with 10.39% deviation and a number of warnings)
An external battery analyser was used to log "time,V,I,Temp" values for an individual cell bank, consisting of 2 parallel cells.
To limit the number of iterations to get a better Cell ID match, I would appreciate some assistance with the following questions: (Any other comments are also welcomed and I may well be barking up the wrong tree.)
1. How does logging against parallel cells affect Chem ID calculation? I have not noticed any settings to compensate for parallel cells when submitting logs to the “Gauging Parameter Calculator”. (Currently the parallel cells are spot-welded together, but I can separate them if need be.)
2. How important is the stability/accuracy of the Temperature measurement in the logs for Chem ID calculation? Due to the long period required for logging a cycle, this process typically stretches over a 24h period with day/night ambient temperatures having up to a 20degC variance impact on the logged temperature. Should this be compensated for? (Though I guess this is only used for temperature to impedance lookup tables and not for cycle temperature profiling.)
3. What “Chem ID deviation percentage” is typically acceptable for a workable Chem ID match?
4. How important is it to log the maximum swing of the cell voltage for Chem ID matching?
I am considering using BQStudio with the BQ40Z60EVM-579 eval kit for future Chem ID logging attempts. It seems to have more stable readings than the external analyser we use, which seems to have about 1mV of noise on readings. The eval kit is currently setup with some safety margins and especially charge termination is based on some interpretations made by the gas gauge, which causes the battery to not always terminate charging at the extremes. (This could be fixed with some settings changes if need be.)
5. Is it okay if rest periods in the logs for Chem ID matching stretch beyond the setting of the “REST” register or the recommended 2h and 5h rest periods? Currently the process is driven manually and logs can overrun these state changes, especially when run over night. Or can these over runs in the logs be safely redacted?
6. Is there any benefit in submitting logs with multiple charge-discharge cycles to the “Gauging Parameter Calculator” or is one discharge cycle good enough?
Thank you
