BQ34Z100-G1: SOH Calculation Mismatch with the actual state of Battery

Hi Anvin,

Please allow me sometime to verify some answers.

Regards,

Evan

Hi Evan,

Sure, Awaiting for your answers. Please feel free to ask for any clarification meanwhile.

Regards
Anvin

Hi Anvin,

Sorry for the delay.

1) SOH is the ratio between predicted FCC/Design Capacity. What "battery capacity usage" are you referring to?

3) Predicted FCC is the FCC at room temperature and SOH current rate which is the highest discharge rate for your battery.

2) We can fix this by using the GPCRA0 tool it will update private parameters that should give you better SOH.

GPCRA0 Application software & framework | TI.com

Regards,

Evan

Hi Evan,

Thanks for the revert.

     1) The learning cycle was completed for the full Capacity of the battery and the Design Capacity was configured accordingly for this learning cycle. i.e. 31.7Ah for the battery voltage from 58.8V to 35V. The Battery usable capacity is limited to 92~95% Depth of Discharge on-field application by adjusting the OV and UV in the BMS i.e. the capacity used on-field application is 29.2Ah from 58.1V to 42V . When I mentioned "battery capacity usage" I was referring to the battery capacity used for on-field application. 
      To be more concise with the question, After flashing the same golden image on the fresh battery packs and using only 92% of the Energy from the battery for the on-field application i.e. (29.2Ah/31.7Ah) starting from the 1st cycle to 250 cycles and upwards, the SOH is always between 92% to 95% and is never 100% and there is no significant degradation of Battery Observed through SoH. Is the SOH Calculation getting affected by not using the entire capacity for the on-field application?

Option A
3) Is SOH current Rate same as the SOH Load 1 under Configuration Data? The SoH Load Current is set -15125 to be scaled upto the max discharge current of variable load of 45000mA. 45A is the application's Peak current, 40A is the Max current and the average current is around 15A, if the average current is to be set as SOH Load 1, then I believe the configured value should be scaled down by 3 and should be 5050mA. Let me know what you think of this SOH Load 1 value as shown below:

2) I'll try out the GPCRA0 tool and let you know.

Option B
Also, I was thinking about using the Load Select values to user mode by setting it to 6 and shift the Cell Terminate voltage to align with the UV cutoff i.e. 42V and update Reserve capacity to 2.5Ah (31.7Ah-29.2Ah) to accurately derive the SOH to 100%. However, I was wondering if these values for Cell Terminate voltage and Reserve capacity should be scaled according the voltage scaling factor of 2 and capacity scaling factor of 3 for these registers below?

FYI, I've scaled down the Voltage values by 2 and the current and capacity values by 3 and the Energy value by 10 in the configuration Data flash to fit inside Gauge Register bandwidth. PFA.
Data Flash.gg.csv


Let me know between Option A & Option B, where the SOH inaccuracy could be fixed or if it is not the root cause of the SoH inaccuracy at all.

Regards
Anvin

P.S. I was trying to interpret the SOH Calculation with block diagrams as shown below, let me know of any corrections.

Hi Anvin,

Let's go with option A first. Please let me know if things improve after this.

1) This should not affect SOH

3) Yes, all current and capacity parameters should be scaled. (3.1 in doc. linked)

This document may be useful for you.

bq34z100-G1 High Cell Count and High Capacity Applications

Please allow me sometime to confirm some things with my team about your block diagram.

Regards,

Evan

Hi Evan,

Thanks for the update.

I've started testing with Option A, how many cycles of charging and discharging should be completed for the ideal corrected and accurate results?

1) Thanks for the clarification on this.

3) Thanks I'll recheck all the parameters and let you know if any discrepancy is observed. (As discussed above, SOH Load 1 user Configured value was incorrect since it was configured for the peak current of the load, whereas it should have been configured for the average current load).

Sure, I'll await for the confirmation about SOH Block Diagram. This will help in understanding the SOH Calculations more easily!

Regards
Anvin

Hi Anvin,

Sorry for the delay, your block diagram looks you have a good understanding of SOH.

Regards,

Evan

Hi Evan,

Thanks for the confirmation on the block diagram.

There is progress after making the changes.

After changing the SOH Load 1 value from max current of load to the average current load value and completing 1 cycle of charge & discharge the SOH bumped to 96%  from 91%

FYI, this was with resistive load. What can cause the SOH to not hit 100%?

NOTE: LOAD mode:1 and Load Select:1

Regards

Anvin

Hi Anvin,

Can you change your Load Mode to 0 and Load Select to 1 and see if things improve. These settings should be more optimal for your application.

Regards,

Evan

Hi Evan,

Load Mode 0 is for constant current, my actual application load on-filed is variable current. The resistive load is only for temporary bench level testing.

Are you suggesting me to permanently switch to load mode 0 i.e. constant current for the variable current load application on-field or only for the temporary testing on bench level?

Regards

Anvin

Hi Evan,

Thanks for the confirmation. I'll make the amendments accordingly.

Regards
Anvin