This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Problem with learning cycle bq34z100

Other Parts Discussed in Thread: BQ27200

Hello,

 

We have built a 16s9p solution that uses the 34z100 for fuel-gauging.  The pack is designed around Iron Phosphate cells (please email for specifics).  The device is calibrated and programmed with default values.  We needed to "fool" the current parameters to be x5. 

The programmed design capacity is 6000mAh, which is a factor of 5 lower than the actual capacity.  We believe we have the appropriate chem ID loaded, and the 34z100 is properly calibrated (with current reduced by a factor of 5, so the sense resistance is actually 5x higher than actual).  When we try to perform a learning cycle, the reported capacity never exceeds 200mAh and the resistance table will not update. 

 

Any help (asap) would be appreciated.  I would prefer to share files offline via email.

 

Thanks,

Joe Perino

Nova Battery Systems

  • As a followup, as I discharge the battery pack the count of passed charge keeps resetting and never exceeds about 280mAh even though at this point in time at least 5000mAh have passed through the sense resistor on discharge.  On charge, the count seems correct at -5800mAh passed charge.

     

    --Joe Perino

  • Joe,

    Passed Charge will get reset to 0 when the device sets a DOD point. Could you send a log file showing this anamoly, so that I can try to determine what is triggering the event?

    Thanks, 

    Tom

  • Hello Tom,

     

    Thanks for the response.  Actually, the passed charge was resetting during actual charge and discharge.  Every time the passed charge reached 280mAh, it would start over.  However, I started with a fresh board (thanks for sending me the default .senc file yesterday now I can reprogram the board I was working with originally), reloaded the chemistry profile, programmed our default .gg file, and the problem is now fixed.  The x5 factoring is working just fine for us, as the reported capacity is 1/5 the true capacity and the passed charge now gets counted properly on both charge and discharge.  And yes it resets to zero on rest.  I was able to achieve both Qmax and Resistance table updates.  I wonder if the problem may have been that the chemistry profile was not properly loaded?

     

    With the 34z100, is it sufficient just to set the chemistry ID in the .gg file, or does it actually have to be programmed into the device?  And in order to make sure it gets programmed with the default .gg constants in one step, can I write a golden .senc file read out from a device with learned data (with update status and cycle count reset as per slua334b), since .rom files cannot be used with this device?

     

    Thanks,

    Joe P.

  • Joe,

    The chemistry data needs to be loaded into the device before the gg file. The gg file contains the Ra-table, but it does not contain the OCV table and other parameters. You can save a golden senc file and load it into other packs. It contains all parameters. You will still need to calibrate the new pack to achieve the best accuracy. The next bwEVSW release will allow you to export and import .rom files to the device.

    Tom

  • Hi,

      I am slightly deviating from the actual topic, sorry about that.

    I am using BQ27200 (bqjunior) as a fuel gauge for my application. I want to conduct the learning cycle test on the BQjunior. Can you provide me with some manual or application note on how to conduct this test.

    Thanks an advance