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.

BQ35100: EOS Mode => Trend Average

Part Number: BQ35100

Hi,

I would like to know if it's possible to change the number of value for short trend average and long trend average on BQ35100 ?

Example:

Pass short trend average of 50 to 25 values (DF1(25))

Pass long trend average of 100 to 50 values (DF2(50))

Thanks.

  • Hello Jeremy,

    Yes it is possible, there will be a few parameters you will need to change though. Let me talk with the designers and get back to you.

    Sincerely,

    Wyatt Keller

  • Hello Jeremy,

    We recommend using the EOS Detection Pulse Count Threshold and the EOS trend detection % to effectively modify those parameters.

    If Mov_fast > Mov_slow * (1+(EOS Trend detection)/100) and if Pulse counter > ½ Pulse Count Threshold, the gauge will set the EOS signal.

    Hope this helps.

    Sincerely,

    Wyatt Keller

  • Hello,

    Do you have any news about the modification of filters ?

    Thanks.

  • I can pass Short trend filter to 254 and long trend filter to 255 ?

    Will this reduce the number of samples taken in the average?

    Thanks.

  • Hello Jeremy,

    It is very complex to change the number of samples for the averages, this is why we recommend using the  EOS Detection Pulse Count Threshold and the EOS trend detection % to effectively change the operation to match what could be done by changing the sample count.

    Sincerely,

    Wyatt Keller

  • Hello,

    Ok i understand it's very complex but i have 2% to EOS trend detection and 1 to EOS Detection Pulse Count Threshold and it's not sufficient for me because i take one measure per week and i would like to detect the EOS flag before the battery voltage is of 6V (Configuration 2S; 7.3V).

    I can not do better with these two parameters?

    Thanks

  • Hello Jeremy,

    The pulse count threshold should be the number of pulses until your batteries will be drained by 85%.

    What were your results from the test? did the gauge not accurately measure EOS?

    The function of the moving averages is to see when there is a sharp increase in impedance to know if the batteries are close to EOS, if ay 6V there is not a significant change the gauge will not be able to detect it.

    Sincerely,

    Wyatt Keller

  • Why i can't take the pulse count threshold to 1 for have a detection directly when the battery start ?

    Because if i set the pulse count threshold to 85% of used battery, I can't detect a impedance increase before 85% of used battery ? I only have 15% of capacity when the calcul for the EOS Flag start ?

    Thanks.

  • Hello Jeremy,

    The 85% is the suggestion we make because the sharp impedance increase will happen after this point and allow for the detection. Before this point the impedance between the moving averages is very low.

    Was the gauge not setting the EOS flag during your test when the EOS trend detection was set at 2%?

    Sincerely,

    Wyatt Keller

  • Hello,

    Yes my EOS flag is activated at 2% EOS trend detection during my test. But the EOS flag is appears at 96% discharge. Only 4% of the battery capacity remains. I think it's because i have only one sample per week, so i don't have enough samples for a detection before 96%. What do you think ?

    It's not possible to have a EOS detection at 70% ?

    Thanks

  • Hello Jeremy,

    How many samples/pulses are you taking before the battery is discharged? Have you tried it at 1%?

    It will be hard to detect at 70% because there is very little change at that point and there are only a few samples.

    Sincerely,

    Wyatt Keller

  • Hello,

    I have 5 samples (so 5 weeks) before EOS flag.

    I don't tried 1% because i don't want an EOS detection on a variation of the impedance which would not be the true end of life of the battery.

    I will try another strategy.

    Thanks.

  • Hello Jeremy,

    I don't think 5 samples is enough to properly calculate EOS, even if the averages were changed.

    You can try attaching a dummy load that you an pulse with the MCU so the gauge can get more samples during the discharge process, then it may be able to calculate EOS correctly.

    Sincerely,

    Wyatt Keller