Other Parts Discussed in Thread: BQ28Z610, BQ24295, BQ27510, BQSTUDIO,
I am trying to configure the BQ27441-G1A for a new larger battery (2900 mAh). I have previously had this working with a smaller battery using the default settings.
Based on the quick start guide, it appears that I only need to change 4 register values for this to work.
I followed the pseudo code example in the TRM, section 3.1 (pg 14) "Data Memory Parameter Update Example", to create a bash script (attached) to update these registers.
I have set the 4 parameters using this bash script:
Design Capacity = 2900 mAh
Design Energy = (Design Capacity) * 3.7 = 10,700 mWh
Terminate Voltage = 3200 mV
Taper Rate = (Design Capacity) / (.1 * (Taper Current)) = 98 (.1 Hr rate)
I tested the larger battery with the default values in the BQ27441-G1A, and with the updated values. I am now having problems with the battery gauge accurately reporting the battery percentage. In our tests the battery is jumping from 92% to 100% charged, and then during discharge the battery is dying while still reporting between 17 - 30 percent remaining.
Is there another step to programming the fuel gauge that I missed?
#!/system/bin/sh ####### #README ####### # # Script to modifiy the registers of the TI BQ27441-G1A battery gauge for a new battery configuration # # This script is intended to overwrite the default values of 4 important configuration parameters. # These parameters are: # Design Capacity - Nominal battery capacity printed on the battery label, or found in the battery data sheet. (in mAh) # Design Energy - Is equal to Design Capacity x 3.7 for bq27441-G1A # Terminate Voltage - The minimum operating voltage of your system. This is the target where the gauge typically reports 0% capacity. (mV) # Taper Rate - Should be set to the current threshold in mA below which your charger IC is set to stop charging once it considers the battery to be full. (Taper Rate = Design Capacity / (0.1 * Taper Current)) # # Based largely on a pseudo code example in Section 3.1, pg 14 of the TRM Rev. A (http://www.ti.com/lit/ug/sluuac9a/sluuac9a.pdf) # # # See Criticallink Internal Redmine issue #12557 # # Jarrett Pischera 6/7/2017 # Jpischera@criticallink.com # ################################################################### # User Modifiable Device Configuration Parameters (Decimal Numbers) ################################################################### # Design Capacity in mAh USER_Design_Capacity=2900 #Terminate_Voltage in mV USER_Terminate_Voltage=3200 # Taper Current in mA USER_Taper_Current=256 ####################################### # Device configuration parameters (hex) ####################################### # Design capacity convert to hex and calculate MSB and LSB Design_Capacity=$(printf 0x%X\\n $USER_Design_Capacity) Design_Capacity_LSB=$(printf 0x%X\\n $(($Design_Capacity % 0x100))) Design_Capacity_MSB=$(printf 0x%X\\n $(( ($Design_Capacity - $Design_Capacity_LSB) / 256))) # Design Energy is directly dipendent on the Design Capacity. The factor of 3.7 is from the TRM for the -G1A chips temp_DE1=$(($Design_Capacity * 3)) temp_DE2=$((($Design_Capacity * 7) / 10 )) temp_DE3=$(($temp_DE1 + $temp_DE2)) Design_Energy=$(printf 0x%X\\n $temp_DE3) Design_Energy_LSB=$(printf 0x%X\\n $(($Design_Energy % 0x100))) Design_Energy_MSB=$(printf 0x%X\\n $(( ($Design_Energy - $Design_Energy_LSB) / 256))) # The terminate voltage in mV (hex) Terminate_Voltage=$(printf 0x%X\\n $USER_Terminate_Voltage) Terminate_Voltage_LSB=$(printf 0x%X\\n $(($Terminate_Voltage % 0x100))) Terminate_Voltage_MSB=$(printf 0x%X\\n $(( ($Terminate_Voltage - $Terminate_Voltage_LSB) / 256))) # The taper rate, assuming that the charger taper current is 256 mA. Following the formula above. Taper_Current=$USER_Taper_Current Taper_Current_15=$(( ($Taper_Current + ( ($Taper_Current * 15 ) / 100 ) ))) Taper_Rate=$(printf 0x%X\\n $((($Design_Capacity * 10 ) / $Taper_Current_15))) Taper_Rate_LSB=$(printf 0x%X\\n $(($Taper_Rate % 0x100))) Taper_Rate_MSB=$(printf 0x%X\\n $(( ($Taper_Rate - $Taper_Rate_LSB) / 256))) ################# # I2C Information ################# i2c_bus=0x0 i2c_chip_addr=0x55 ################################# # Register Modification Procedure ################################# # UNSEAL the device by writing Control() (0x00 and 0x01) with the UNSEAL key 0x8000. ie Control(0x8000) i2cset -y -f $i2c_bus $i2c_chip_addr 0x00 0x00 i2cset -y -f $i2c_bus $i2c_chip_addr 0x01 0x80 i2cset -y -f $i2c_bus $i2c_chip_addr 0x00 0x00 i2cset -y -f $i2c_bus $i2c_chip_addr 0x01 0x80 # Enter CONFIG UPDATE Mode SET_CFGUPDATE. ie Control(0x0013) i2cset -y -f $i2c_bus $i2c_chip_addr 0x00 0x13 i2cset -y -f $i2c_bus $i2c_chip_addr 0x01 0x00 # Confirm that CFGUPDATE mode has been set sleep 1.5 echo "Confirm bit 4 is high for CFGUPDATE:" i2cget -y -f $i2c_bus $i2c_chip_addr 0x06 # Ensure block data memory control by writing 0x00 using the BlockDataControl() command i2cset -y -f $i2c_bus $i2c_chip_addr 0x61 0x00 # Access the State subclass (0x52), which contains the configuration parameters, using the DataBlockClass() command i2cset -y -f $i2c_bus $i2c_chip_addr 0x3E 0x52 # Write the block offset location using DataBlock() command. For data 0 to 31 use 0x00, for data 32-41 use 0x01 i2cset -y -f $i2c_bus $i2c_chip_addr 0x3F 0x00 # Read the 1-byte checksum using the BlockDataChecksum() command old_chksum=$(i2cget -y -f $i2c_bus $i2c_chip_addr 0x60) # Read the Design Capacity bytes staring at 0x4A (offset = 10; from TRM) old_Design_Capacity_MSB=$(i2cget -y -f $i2c_bus $i2c_chip_addr 0x4A) old_Design_Capacity_LSB=$(i2cget -y -f $i2c_bus $i2c_chip_addr 0x4B) # Read the Design Energy bytes starting at 0x4C (offset = 12; from TRM) old_Design_Energy_MSB=$(i2cget -y -f $i2c_bus $i2c_chip_addr 0x4C) old_Design_Energy_LSB=$(i2cget -y -f $i2c_bus $i2c_chip_addr 0x4D) # Read the old Terminate Voltage starting at 0x50 (offset = 16; from TRM) old_Terminate_Voltage_MSB=$(i2cget -y -f $i2c_bus $i2c_chip_addr 0x50) old_Terminate_Voltage_LSB=$(i2cget -y -f $i2c_bus $i2c_chip_addr 0x51) # Read the old Taper Rate starting at 0x5B (offset = 27; from TRM) old_Taper_Rate_MSB=$(i2cget -y -f $i2c_bus $i2c_chip_addr 0x5B) old_Taper_Rate_LSB=$(i2cget -y -f $i2c_bus $i2c_chip_addr 0x5C) # Write the new Design Capacity i2cset -y -f $i2c_bus $i2c_chip_addr 0x4A $Design_Capacity_MSB i2cset -y -f $i2c_bus $i2c_chip_addr 0x4B $Design_Capacity_LSB # Write the new Design Energy i2cset -y -f $i2c_bus $i2c_chip_addr 0x4C $Design_Energy_MSB i2cset -y -f $i2c_bus $i2c_chip_addr 0x4D $Design_Energy_LSB # Write the new Terminate Voltage i2cset -y -f $i2c_bus $i2c_chip_addr 0x50 $Terminate_Voltage_MSB i2cset -y -f $i2c_bus $i2c_chip_addr 0x51 $Terminate_Voltage_LSB # Write the new Taper Rate i2cset -y -f $i2c_bus $i2c_chip_addr 0x5B $Taper_Rate_MSB i2cset -y -f $i2c_bus $i2c_chip_addr 0x5C $Taper_Rate_LSB # DEBUG Read the changes to make sure they worked echo "the old MSB and LSB of the Design Capacity, read from the register are:" echo $old_Design_Capacity_MSB echo $old_Design_Capacity_LSB echo "MSB and LSB of the Design Capacity, read from the register, should be 0x0B and 0xB8" i2cget -y -f $i2c_bus $i2c_chip_addr 0x4A i2cget -y -f $i2c_bus $i2c_chip_addr 0x4B echo " " echo "The old MSB and LSB of the Design Energy, read from the register are:" echo $old_Design_Energy_MSB echo $old_Design_Energy_LSB echo "MSB and LSB of the Design Energy, read from the register: " i2cget -y -f $i2c_bus $i2c_chip_addr 0x4C i2cget -y -f $i2c_bus $i2c_chip_addr 0x4D echo " " echo "The old MSB and LSB of the Terminate Voltage, read from the register are:" echo $old_Terminate_Voltage_MSB echo $old_Terminate_Voltage_LSB echo "MSB and LSB of the Terminate Voltage, read from the register:" i2cget -y -f $i2c_bus $i2c_chip_addr 0x50 i2cget -y -f $i2c_bus $i2c_chip_addr 0x51 echo " " echo "The old MSB and LSB of the Taper Rate, read from the register are: " echo $old_Taper_Rate_MSB echo $old_Taper_Rate_LSB echo "MSB and LSB of the Taper Rate, read from the register:" i2cget -y -f $i2c_bus $i2c_chip_addr 0x5B i2cget -y -f $i2c_bus $i2c_chip_addr 0x5C echo " " # Calculate the new Checksum # The maxvar variable deals with the issue of taking the modulus of a negative number in bash # (number of bytes changed * 0x100) + 0xff maxvar=$(((0x10 * 0x100) + 0xff)) temp_var1=$((($maxvar - $old_chksum - $old_Design_Capacity_MSB - $old_Design_Capacity_LSB - $old_Design_Energy_MSB - $old_Design_Energy_LSB - $old_Terminate_Voltage_MSB - $old_Terminate_Voltage_LSB - $old_Taper_Rate_MSB - $old_Taper_Rate_LSB) % 0x100)) temp_var2=$((($temp_var1 + $Design_Capacity_MSB + $Design_Capacity_LSB + $Design_Energy_MSB + $Design_Energy_LSB + $Terminate_Voltage_MSB + $Terminate_Voltage_LSB + $Taper_Rate_MSB + $Taper_Rate_LSB) % 0x100)) dec_chksum=$((0xff - $temp_var2)) new_chksum=$(printf 0x%X\\n $dec_chksum) # Write the new Checksum to BlockDataChecksum() i2cset -y -f $i2c_bus $i2c_chip_addr 0x60 $new_chksum # DEBUG Read the changes to the checksum ot make sure they worked echo "The old checksum was:" echo $old_chksum echo "The new calculated checksum is:" echo $new_chksum echo "Read the saved checksum from the register" i2cget -y -f $i2c_bus $i2c_chip_addr 0x60 # Exit CFGUPDATE i2cset -y -f $i2c_bus $i2c_chip_addr 0x00 0x42 i2cset -y -f $i2c_bus $i2c_chip_addr 0x01 0x00 # Confirm that CFGUPDATE had been exited by polling Flags() register until bit 4 cleared sleep 1.5 echo "Confirm bit 4 is low, CFGUPDATE has been exited:" i2cget -y -f $i2c_bus $i2c_chip_addr 0x06 # Reseal the device i2cset -y -f $i2c_bus $i2c_chip_addr 0x00 0x20 i2cset -y -f $i2c_bus $i2c_chip_addr 0x01 0x00
