• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 259 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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:

BQ28Z610-R1: FullChargeCapacity calculation.

BQ28Z610-R1: FullChargeCapacity is jumping

Daniel Kirsch
Prodigy 30 points
Part Number: BQ28Z610-R1

Hi,

I have charged and discharged few times and I saw some strange behaviors and hope you can help: 

Charging from zero to full:

* you can see that the charger is charging for >6200mAh 
* when EoC is been detected and FCC is dropping and also the Remaining capacity 
* the charger stops (I know it's a bit late here but it's nicer to see) and after a short rest time QMax is been updated. FCC and RemCap is dropping again. 
* FCC and RemCap is now below 6000, but it was charged for over 6200mA!

Discharging from full to zero:

* the FCC is directly after discharge going up 
* after a rest of 5h the FCC increased even further 
* no R value update was done only QMax

Even after few charge and discharges the FCC and RemCap is still jumping. I thing the jump down after charge comes from the option CSYNC that is in default on and the jump up at end of discharge comes from a wrong R value? By looking in the register at the moment when it's full charged and before QMax is updated:

* DOD0 Passed Q is still -6262mAh (I would expected after CSYNC that should be the value for FCC and RemCap)
* but FCC and RemCap is 5892
* even True Full Chg Q is 5892

After Qmax update at fully charged the register looks like this:

So the question are now:

* Why is it jumping? I would expect a more stable value. 
* Could be CSYNC be the problem? Here i expect that DOD0 Passed Q should be the same as FCC and RemCap after charging. Why is it not the case?
* Is the problem that there is no R value update related?

Regards,

Daniel

  • 0
    Prodigy 30 points

    After discharging from full to empty and before QMax update, the DOD0 Passed Q value was 6238. That's close to was is been measured in charge! (Also FCC and RemCap is after the fourth discharge more fitting to this value. But when I look on the discharge value a cycle before, I thing that was more by luck.)

    Here all registers from that moment:

  • 0 in reply to Daniel Kirsch
    TI__Genius 10370 points

    Daniel,

    Please provide a log file and GG file so that we can investigate this issue further. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Prodigy 30 points

    Hello Eric, 

    I send the files to your college Mr. Hares in Germany. I can't put the GG files to a public forum. 

    This night I have done a charge with CSYNC deactivated, the results where nearly the same. With the same drop at the end of charge. 

  • 0 in reply to Daniel Kirsch
    TI__Guru** 103125 points

    Hello Danial,

    Have you performed the learning cycle on the battery and found a matching chem ID?

    If you have, then I would recommend enabling smoothing for your tests. Without the logs we cannot debug much further, only give suggestions.

    Sincerely,

    Wyatt Keller 

  • 0
    Prodigy 30 points

    Hello Wyatt,

    we had few days ago a discussion with Eric Vos, that was really helpful. I'm still checking his recommends. Our problems were:

    * The learning cycle was not fitting to the later application (the load at learning was >x6 higher)
    * The max application load was below C/10, so the R values didn't not update. 

    @Eric: Were you able to check the maximum value for our battery? Currently I have set it to 25. 

    Thanks a lot

    Daniel Kirsch

  • 0 in reply to Daniel Kirsch
    TI__Genius 10370 points

    Daniel,

    Please keep it at C/25 and see if performance improves. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Prodigy 30 points

    Hi Eric, 

    the project would like to go down till C/30. Would that also be possible?

    What would be the impact, if this value were chosen to low? 

  • +1 in reply to Daniel Kirsch
    TI__Genius 10370 points

    Daniel,

    The fear is your resistance update is less accurate due to the delta between OCV and loaded V not being very high. I do recommend running a few cycles to see how the Ra table updates. C/30 is really borderline to working only testing will know

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Prodigy 30 points

    Hi Eric, 

    thanks for the advice. I will give it further. What is concerning me now are the different filter that are quite strict with the default values. Our application use the battery mostly only as backup in case of a power failure. So the aging effect over time is higher than the cycle effect. Is it necessary to reduce the filters or change some other parameter? Or does the gauging already taking care about this?

  • +1 in reply to Daniel Kirsch
    TI__Genius 10370 points

    Daniel,

    Rarely discharged applications are harder to gauge. This gauge is intended mostly for personal electronics that utilize the full range pretty often. The gauge does not have logic embedded to decrease capacity based on idle time. It assumes QMax is learned/updated within a time window that allows the truth to be in sync with the gauge.

    I recommend putting is a "learning" scenario into your application which allows the gauge to update resistance and QMax periodically.

    Thanks,

    Eric Vos