• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 1066 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.

Original question:

AFE OCD/AFE protection work early problem

AFE calibration and OCD/SCD accuracy?

Chen Alex
Prodigy 55 points
Other Parts Discussed in Thread: BQ76930, BQ78350-R1

Hello,

Our project is using BQ76930 AFE, from the data sheet, in section:

7.3.1.1.2.1 Optional Real-time Calibration Using the Host Microcontroller

it mentions that we need to do calibration for improve accuracy. 

Is it necessary to do the calibration? How to implement the calibration?

Does the calibration impact the SCD/OCD accuracy?

Thank you.

 

  • 0
    TI__Guru* 81034 points
    Hi Alex,

    I just answered this same question on another thread. Are you working with Frank?

    SCD and OCD are based on hardware comparators and calibration cannot impact these. Calibration can improve the current measurements reported by the BQ76930 to the gas gauge (like BQ78350-R1) or microcontroller.

    "The bq76930 is calibrated at the factory so all devices will have consistent measurements within the datasheet specification. However, measurements can also be affected by board parasitics. The bq78350-R1 companion gauge or a microcontroller can do calibration to correct for any error due to the board. This is described in detail in Section 12.2 of the BQ78350-R1 TRM (Technical Reference Manual): www.ti.com/.../sluubd3d.pdf"

    Best regards,
    Matt
  • 0 in reply to Matt Sunna
    Prodigy 130 points
    Dear Matt
    Thanks for replying .
    I have anther question, in the data sheet page 38.
    It indicate that "Short circuit in discharge delay setting. A 400-µs setting is recommended only in systems using
    maximum cell measurement input resistance, Rc, of 1 kΩ (which corresponds to minimum internal cell
    balancing current or external cell balancing configuration)."
    What does this means? should i set scd delay to 400 us to be more accuracy?
    thanks
  • 0 in reply to Frank Hsu94
    TI__Guru* 81034 points
    Hi Frank,

    Someone also asked this question yesterday: e2e.ti.com/.../745663
    The delay setting should not be related to the threshold setting and would not explain the issue you are seeing. You can try to experiment with this, but I do not think it will affect the result.

    Best regards,
    Matt
  • 0 in reply to Matt Sunna
    Prodigy 130 points

    Matt Sunna said:
    Hi Frank,

    Someone also asked this question yesterday: e2e.ti.com/.../745663
    The delay setting should not be related to the threshold setting and would not explain the issue you are seeing. You can try to experiment with this, but I do not think it will affect the result.

    Best regards,
    Matt

    Dear Matt 

    Morris is our EE guy, in fact after setting SCD delay to 400us, we can get more accuracy SCD threshold value .

    Before setting delay to 400us, we set to 70us, the result as the following

    SCD Threshold(mV) No. 17 PCB (delay set to 70 us)
    　 SCD happened (mV) error %
    44 34 22.73%
    67 52.8 21.19%
    89 68 23.60%
    111 83.1 25.14%
    133 113 15.04%
    155 128.2 17.29%
    178 147.1 17.36%
    200 158.4 20.80%

    After setting scd delay to 400 us the value as the following

    SCD Threshold(mV) No. 17 PCB (delay set to  400us)
    　 SCD happened (mV) error %
    44 37.7 14.32%
    67 56.7 15.37%
    89 75.6 15.06%
    111 98.2 11.53%
    133 113.1 14.96%
    155 147.2 5.03%
    178 158.5 10.96%
    200 173.5 13.25%



    You can see the result, after setting delay to 400us, the error will be small than setting to 70us 

    But it still didn't achieve to 10%,  Do you have any other suggestion or direction can tell us to clarify this issue?

    BTW we are following  the document TIDUC43A"20-Series Battery Management Module Reference Design" to develop our BMS system

    So we will use two I2C to read/write AFE register, it is the big difference between us and EVM board, Is it possible that read/write operation to affect SCD accuracy?

    Thanks  

  • 0 in reply to Frank Hsu94
    TI__Guru* 81034 points
    Can you try the experiment with a different sense resistor value? Or maybe instead of applying a voltage across the sense resistor, can you try using a known load current - this might help with any error if the voltage is not applied directly to the input filter (solder and trace resistance)?

    Matt