Capacity estimation BQ20z65-R1

Hi Fredrik,

Did proper learning with Qmax as well as resistance updates occur? Impedance track algorithm takes into consideration the measured temperature when reporting capacity so variations if significant  temperature changes occurs  is expected. This is so because capacity of batteries are temperature dependent. Your comment has limited info available and there has to be more information to go by for us to be able to help you. It will be good to know the state of the gauge before 21 was sent. It is preferable to send a reset command (41) instead of 21 if it has been initially sent. Log files and an explanation of what you are trying to do will be helpful.

thanks

Onyx

I am trying to obtain correct capacity estimation/battery-low warning when the temperature changes. I am using the evaluation board for bq20z65. I connected a load and changed the temperature but the remaining capacity remains the same. Attached is a logfile.

Fredrik,

Since you are using an EVM, I am assuming you have not carried out a proper learning cycle of your cells. That is a necessity for optimal gauging accuracy.

secondly, Rem cap simulations  are triggered by the following

• Initialization (reset   or POR)
• Start of charge
• Charge Taper   Termination
• Start of discharge
• Every grid update during discharge
• If temperature has  changed by more than DOD at EOC Temp Delta since the last RemCap simulation
• Every Relax Rem Cap Update Time (usually 5 hour) after OCV in relax mode
• Qmax updates

I am not sure how familiar you are with these terms but I have attached an app note that explains how IT algorithm works.

 

Thirdly, why do you think that the reported capacity is inaccurate? Do you have another device that is able to accuratly track the cells' capacity for comparative analysis?

 1072.slua364[1]_theory and Imp of IT battery fuel-guaging algorithm.pdf

thanks

Onyx

Thank you for the information. I still have a few qustions about the following discharges. A discharged of a battery (with Li+ cells) was done in -24C ambient temperature, attached is a log file v5. First with a load of 5 ohm, after 25 seconds changed to 10 ohm and RunTimeEmpty was updated. On sample 377, RemCap and FullChargeCap suddenly increased and RunTimeEmpty was updated, what caused the increase (battery status was not changed)? Another discharge was done with a load of 10 ohm, logfile v6. When the RunTimeEmpty turned 0 the discharge continued, than the load was disconnected/connected and RunTimeEmpty and FullChargeCap was updated but not RemCap, see sample 1522 in file v6. Is that correct? (battery status change: TDA was unset) Discharge continued, a time after RunTimeEmpty was 0 the RunTimeEmpty and FullChargeCap was updated automatically (Only INIT+DSG set), sample 1887 in v6, what caused the update of RunTimeEmpty and FullChargeCap?

hello Fredrik,

The explanation for the FCC jump in V5 file is in one of the points I stated in my earlier post. I had mentioned rem cap is simulation is triggered at every grid point update during discharge. That point were the change occurred is one of the resistance grid points. You most likely experienced a jump because of the load select you chose. I would recommend testing a different load select option and see which has little or no jumps at such extreme temperatures. My guess is that you are currently using the default setting which is average current. the average current at the start of discharge which is when the first rem cap simulation would have occurred will be different from the average current when another rem cap simulation occurs at a grid point hence the jump. Bear in mind that the performance of the guage especially at such extreme temperatures is highly dependent on your settings. Did you choose the correct chem id for your cells? Did you carry out a successful learning cycle. Are your cells going to be used under such extreme temperatures? I believe you understand that the capacity of batteries generally reduce with reduced  temperature.

 

thanks

Onyx