BQ27320:the capacity of battery display error

yunyu hua

Other Parts Discussed in Thread: BQ27320, BQSTUDIO

our product use bq27320 fuelgauge to manage battery,but after having some monkey tests,the capacity sometimes was 1% ，Actually battery voltage was 4.25v.we test fuelgauge Vcc pin and bat pin，the result was ok. If the battery was removed and inserted again,the capacity was ok, our had 100pcs samples and 4 samples occure this error.do you have some sugestions?

over 7 years ago

0 Benjamin Lough over 7 years ago

TI__Mastermind 26225 points

Hi Yunyu,

Could you provide some detail about your test procedure and when does capacity drop to 1% during the test?

Best Regards,
Ben Lough

0 yunyu hua over 7 years ago in reply to Benjamin Lough

Prodigy 70 points

we install cbmstree test app into our tablet and insert adapter to avoid power down during the test

0 yunyu hua over 7 years ago in reply to Benjamin Lough

Prodigy 70 points

I catch some logs and find i2c fail,can you have some idea to debug it?

i2c-msm-v2 78b8000.i2c: NACK: slave not responding, ensure its powered, Invalid slave addr, I2C transfer failed,:
msgs(n:1 cur:0 tx) bc(rx:0 tx:3) mode FIFO slv_addr:0x55 MSTR_STS:0xd1343c8 OPER:0X00000090
2c-msm-v2 78b8000.i2c: NACK: slave not responding, ensure its powered, Invalid slave addr, I2C transfer failed,:
msgs(n:1 cur:0 tx) bc(rx:0 tx:3) mode FIFO slv_addr:0x55 MSTR_STS:0xd1343c8 OPER:0X00000090
2c-msm-v2 78b8000.i2c: NACK: slave not responding, ensure its powered, Invalid slave addr, I2C transfer failed,:
msgs(n:1 cur:0 tx) bc(rx:0 tx:3) mode FIFO slv_addr:0x55 MSTR_STS:0xd1343c8 OPER:0X00000090
bq27320-fuel-guage 4-0055:Failed to write BQ27320_REG_CONTROL cmd

0 Benjamin Lough over 7 years ago in reply to yunyu hua

TI__Mastermind 26225 points

Yunyu,

Are you able to communicate with the device at all or is this intermittent? Also is there anything else on your I2C bus? Your log suggests the master is not finding the gauge on the bus even though the slave address is correct. I would start by probing your data lines and confirming you are seeing start and stop sequences.

If this is an intermittent case, I would also suggest looking to see if the watchdog timer tripped.

Best Regards,
Ben Lough

0 yunyu hua over 7 years ago in reply to Benjamin Lough

Prodigy 70 points

No,if happended this issue,we could not communicate with the device again and no matter we reflashed software it still hoded on,unless we removed battery(fuelgaugepower off ) and inserted battery(fuelgauge power on).
There are two devices on this i2c bus,gauge address is 0x55,charging IC addredd is 0x00.we tested,
host comand is ok,but no respond on data line at 0x55 address.
On the other hand,we guess whether it is into rom mode,but not. Now ,we have no any idea to analysis
the case!

0 Benjamin Lough over 7 years ago in reply to yunyu hua

TI__Mastermind 26225 points

Hi yunyu,

Apologies for the delayed response. When you remove and re-insert the battery pack, it will take an OCV measurement when the device detects the battery insertion. This could be why you sometimes see capacity at 1% but is corrected when you remove and re-insert the battery. If you want to send me your gg.csv, I can look to see if everything looks ok.

Could you also send me some oscilloscope pictures of your SDA and SCL lines to analyze?

Best Regards,
Ben Lough