BQ34Z100: Autocalibration

Hi Jose,

Thank you for the response.
Just a couple quick follow-up questions:

1) If auto-calibration has begun, it must go through the whole 16 second process, is that correct?  In the system this is installed in, a load is applied and the auto calibrate is not aborting.

2) Is there a current threshold that would allow for an auto-abort?

3) How often should a calibration be performed?

Thanks,

Kevin

Hi Kevin,

Is CAL_EN bit set in Pack Configuration Register? This will enable calibration mode.

Kevin Knicker said:

What are the effects of exiting? It seems like the CC Offset won't update, but how much can that value change?

When you send the exit calibration command, you are technically overwriting the CC without adding offset. This is why CC offset won't update. Exiting calibration mode will set the Calibration Mode flag off, and calibration offset flag off.

Kevin Knicker said:

2) Is there a current threshold that would allow for an auto-abort?

I wasn't too clear in the first reply. If (initial voltage - current Voltage) > calibrate offset  voltage drop (this should be 0V by default). This means the load current has likely increased considerably. Therefore, offset calibration will be aborted.

A workaround here will be to manually send the exit calibration command, and make sure CAL_EN is zero in Pack Configuration Register.

Regards,
Jose Couso

Hi Jose,

CAL_EN is off in the Pack Configuration Register.

Our customer is stating that the auto-calibration is starting and goes through the entire 16 second process and isn't aborting with a 1A to 1.3A load applied upon startup.  

I suggested the exit command as a workaround.

If overall accuracy isn't affected long term by sending the exit command, do you recommend a frequency in which a subsequent calibration should be performed?

Thank you,
Kevin 

Hi Kevin,

Kevin Knicker said:

If overall accuracy isn't affected long term by sending the exit command, do you recommend a frequency in which a subsequent calibration should be performed?

I recommend to do Autocalibration every ten hours in temperatures between 5 to 45 C, and when current is less than sleep current. 

Also, what is the version of the device? R1 or R2?

Regards,
Jose Couso

Hi Jose,

The part is the bq34z100-G1.  Is it possible that the auto-abort isn't working because the voltage drop isn't greater than the calibrate offset voltage drop

Thanks,
Kevin 

Hi Kevin,

Yes, this is the condition for the gauge to abort autocalibration. 

Check with the customer if the voltage drop is greater than 32 mV

Calibrate offset voltage drop by default is 32mV

Regards,
Jose Couso

Hi Jose,

Yes, the voltage drop is greater than 32mV and the Auto-abort isn't working.  

The 16 second calibration still takes place.  In the datasheet for the bq34100, which the bq34z100 is based upon, there is a mention of a specific 16 second delay that has to do with the I2C lines being held low.

"During normal Gas Gauge Operation when the I2C clock and data lines are low for more than 5 seconds and Average Current is less than Sleep Current in mA, then an automatic CC Offset calibration is performed. This takes approximately 16 seconds and is much more accurate than the method in Calibration mode."

Can you clarify if that should state the lines are idle instead of low?  The I2C lines are never held held low in the application, so clarification as to whether the conditions are stated correctly in the datasheets and TRM would be helpful.

Thanks,
Kevin 

Kevin,

Kevin Knicker said:

Can you clarify if that should state the lines are idle instead of low?

The i2c lines don't need to be low for the autocalibration.

I will double check with FW team if Cal Abort is 32mV. I am not 100% sure.



Regards,
Jose Couso

Hi  Jose,

I appreciate you looking into this.

It would nice to have clear definitions and values for all of the variables listed in the Autocalibration section of the datasheet.

This would include:

• Bus Low Time and if the I2C lines need to be low or idle.  The datasheet states low, but you has said idle.  If so, that datasheet will need to be corrected.
• Autocal MIn Current.
• Delta Voltage
• Delta Temperature
• Autocal Time
• Cal Abort

What we seeing is that there is a 16 second Autocalibration taking place when applying a load.

We want to be able to read the current from the bq34z100-G1 when applying the load but we cannot because it reports zero until it completes and it is causing problems in the system.

The load current doesn't appear to be enough to abort the Autocal either.  The voltage drop created is greater than 32mV, but I have only seen the Autocal abort when applying a much larger load current, and the voltage drop is in the 400mV range.

I have attached a couple of logs that should help.

• Autocal_Abort_90W_Load shows that a 90W load will cause an Autocal abort.
• CCA_Bit_ON shows that once the Calibration process has started, the CCA Bit stays on until calibration is complete and no current is reported.

A better understanding of what could be triggering the Autocal would help and that can be gained by knowing the values above.

Thank you,

Kevin

Data Log

TimeStamp,Qmax Time,Voltage,Current,Control Status,CC Offset
2022-09-30 09:17:41.526,298,23177,0,0x0813,-1459
2022-09-30 09:17:42.030,298,23177,0,0x0813,-1459
2022-09-30 09:17:42.612,298,23177,0,0x0803,-1459
2022-09-30 09:17:44.566,298,23058,0,0x0803,-1459
2022-09-30 09:17:46.209,298,23044,0,0x0803,-1459
2022-09-30 09:17:46.702,298,23030,0,0x0803,-1459
2022-09-30 09:17:48.658,298,23009,0,0x0803,-1459
2022-09-30 09:17:50.288,298,23002,0,0x0803,-1459
2022-09-30 09:17:50.778,298,22995,0,0x0803,-1459
2022-09-30 09:17:52.833,298,22981,0,0x0803,-1459
2022-09-30 09:17:54.471,298,22967,0,0x0803,-1459
2022-09-30 09:17:56.551,298,22960,0,0x0803,-1459
2022-09-30 09:17:58.174,298,22953,0,0x0803,-1459
2022-09-30 09:17:58.718,298,22953,0,0x0803,-1459
2022-09-30 09:18:00.658,298,22939,-1222,0x0803,-1459
2022-09-30 09:18:02.257,298,22939,-1222,0x0803,-1459
2022-09-30 09:18:02.753,298,22925,-1222,0x0803,-1459
2022-09-30 09:18:04.735,298,22925,-1222,0x0803,-1459
2022-09-30 09:18:06.432,298,22918,-1222,0x0803,-1459
2022-09-30 09:18:08.381,298,23030,-1222,0x0803,-1459
2022-09-30 09:18:10.064,298,23051,-1222,0x0803,-1459

Data Log

TimeStamp,Qmax Time,Voltage,Current,Control Status,CC Offset
2022-09-30 08:54:27.188,295,23170,0,0x0013,-1459
2022-09-30 08:54:29.350,295,23170,0,0x0803,-1459
2022-09-30 08:54:30.976,295,22757,0,0x0003,-1459
2022-09-30 08:54:31.468,295,22757,0,0x0003,-1459
2022-09-30 08:54:33.456,296,22666,-4124,0x0003,-1459
2022-09-30 08:54:35.099,296,22638,-4128,0x0003,-1459
2022-09-30 08:54:35.577,296,22638,-4128,0x0003,-1459
2022-09-30 08:54:37.610,296,22596,-4135,0x0003,-1459
2022-09-30 08:54:39.299,296,22561,-4143,0x0003,-1459
2022-09-30 08:54:41.297,296,22540,-4146,0x0003,-1459
2022-09-30 08:54:42.963,296,22512,-1785,0x0003,-1459
2022-09-30 08:54:43.436,296,22512,-1785,0x0003,-1459
2022-09-30 08:54:45.396,296,22855,0,0x0013,-1459
2022-09-30 08:54:47.020,296,22855,0,0x0013,-1459
2022-09-30 08:54:47.545,296,22855,0,0x0013,-1459
2022-09-30 08:54:49.533,296,22855,0,0x0013,-1459

Hello Kevin,

Thanks for the log files. I am looking into this, and I will get back to you when I have a solid answer. I may need a bigger log describing more of what's going on, but I will let you know.
 
Regards,
Jose Couso

Kevin,

Can you ask the customer to export the srec file under golden image tab?

Regards,
Jose Couso

Jose,

Sure, can I send it to you privately?

Thanks,
Kevin