We have a big problem. Some EEPROM parameters of the bq27010 fuel-gauge were not set properly during production. (The charge termination taper current threshold is too low. The charge termination voltage threshold is too high.) I don't think it is possible to reflash the EEPROM externally from the battery pack because of the need for 21V programming pulses on a separate pin (not exposed to outside of pack).
It may be possible to program the fuel-gauge RAM variables externally. The values would not stay if a reset occurred, but might work temporarily.
Perhaps the host device firmware could occasionally write the RAM variables to correct values.
This app note gives some hope on the subject: www.ti.com/lit/pdf/slua414
"A user can program 10 EEPROM locations to optimize the setup for the specific battery and host
characteristics. These EEPROM locations are mapped directly to RAM locations 0x76 to 0x7f. These
locations cannot be written by the host except during a special test mode. This test mode can be entered
by writing 0xdd to address 0x6e. This action will allow write access to these RAM locations and
subsequent transfer of the RAM content into EEPROM. The bq27x10 processing is inhibited while 0x6e
contains 0xdd. When EEPROM programming is complete, address 0x6e must be written back to 0x00."
characteristics. These EEPROM locations are mapped directly to RAM locations 0x76 to 0x7f. These
locations cannot be written by the host except during a special test mode. This test mode can be entered
by writing 0xdd to address 0x6e. This action will allow write access to these RAM locations and
subsequent transfer of the RAM content into EEPROM. The bq27x10 processing is inhibited while 0x6e
contains 0xdd. When EEPROM programming is complete, address 0x6e must be written back to 0x00."
Is this possible? Is it possible to test with the eval software?