• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 64 subscribers
  • Views 3220 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

BQ27510. CC Gain, CC Delta. Computation.

Andrew Strukov
Prodigy 120 points
Other Parts Discussed in Thread: BQ27510, BQ27500, BQ27510-G2Hello TI! I use bq27510. I send commands to BQ from AVR with using I2C. The value of sense resistor is 10 mOm. How can I compute value for CC Gain and CC Delta WITHOUT using Evalution Software? What are the dependencies between CC Gain and AverageCurrent, CC Delta and NominalCapacity? How can I get right value for CC Gain and CC Delta in format F4 from decimal format. The matching between CC Gain and AverageCurrent, CC Delta and NominalCapacity does not give me accuracy in computations. Thanks, Andrew.
  • 0
    TI__Intellectual 2100 points

    EVSW does not write these values directly.  Most people have been doing this through BQeasy, so the documentation is a little old.

    Take a look at App note SLUA440: www.ti.com/litv/pdf/slua440

    This app note is for the bq27500, but the calibration command sequence is the same for the 510. 

    Cheers,

    Charles

  • 0 in reply to Charles Herder
    Prodigy 120 points
    Charles, BQ didn't enter in calibration mode after sending command or it's another reason. At all infrormation from SLUA440 doesn't work. There is no information about calibration NominalCapacity! How can I calibrate NominalCapacity? Thanks, Andrew.
  • 0 in reply to Andrew Strukov
    TI__Intellectual 2100 points

    How do you know the BQ gauge did not enter calibration mode?  Note that you cannot enter calibration mode if the gauge is SEALED. Check the SS bit of the ControlStatus register to determine this. 

    The Nominal Available Capacity is also located in the DataFlash and can be written to directly if you want to set it yourself.  This value is calibrated automatically when you perform a learning cycle. 

    See the following application note for Learning Cycles: http://www.ti.com/litv/pdf/slua544

    Let me know if this helps,

    Charles

  • 0 in reply to Charles Herder
    Prodigy 120 points
    SS bit is clear. The calibration commands don't work. BQ is staying in normal mode. DesignCapacity is located in the DataFlash. I mean NominalAvailableCapacity. "NominalAvailableCapacity( ): 0x0c/0x0d This read-only command pair returns the uncompensated (less than C/20 load) battery capacity remaining. Units are mAh." CC Delta changes the speed of changing NominalAvailableCapacity(see app note slua458 page 38) then battery is charging or discharging. I get zero in NominalAvailableCapacity early then battery is discharging to terminate voltage. It's my problem. How can I calculate CC Delta? I ask you help me solve it. Andrew.
  • 0 in reply to Andrew Strukov
    TI__Expert 4050 points

    CC Gain:   data flash raw data 0x7F71205C, EVSW reading with: 10.124

    Raw data (0x7F71205C) -> first two byte is for exp gain. The exp is 0x7F -> 127, so gain:  (127-128-24) = -25 -> 2^(-25)

     

    The rest is the low address to high address hex read from gauge and it is big endian. Bit 7 of the MSB is used for the sign bit, but in the conversion the mantissa is considered to be shifted so the bit 7 would always be a 1, to preserve the most possible significant bits in the mantissa.  So after determining the sign here, the implied high bit must be set back to one for the conversion to be correct.

     

    0x71205C OR with 0x80 = 0xF1205C -> convert into DEC and times 2^(-25) = 15802460*2^(-25) = 0.47095

    CC Gain  = 4.768/x = 4.768/0.47095 = 10.124  (matches!)

     

    CC Delta:   data flash raw data 0x940898C0, EVSW reading: 10.147

     

    Raw data (0x940898C0) -> first two byte is for exp gain. The exp is 0x94 -> 148, so gain = (148-128-24) = -4

    0x0898C0 OR with 0x80 = 0x8898C0 -> convert into DEC and times 2^-4 = 559500

    CC delta = 5677445/x = 5677445/559500 = 10.147

     

    Board Offset: data flash raw data 0xF6.  EVSW reading : -0.27

    F6 -> convert to DEC = -9.

    Board offset = x*0.48/16 = -9*0.48/16 = -0.27

     

    CC Offset:  data flash raw data 0xFA89, EVSW reading with new *.encr = -0.67

    0xFA89 -> convert to DEC = -1398.

    CC offset = x*0.00048 = -1398*0.00048 = -0.67

  • 0 in reply to Ming Yu
    TI__Expert 4050 points

    Please use updated calculation in attached document

    How EVSW Display the Raw Data V1.03.pdf
  • 0 in reply to Ming Yu
    Intellectual 450 points

    This doesn't make sense! I have implemented this calibration data reading on my system. At the time I used the "new" format, but when I used the calibrated values on EVSW and used the generated golden image on my board and read it again, I discovered that EVSW actually uses the "old" format.

    EVSW version 0.9.71

    BQ27510-G2