• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 346 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.

BQ40Z50-R2: RSOC Stays at 100% too long

Vincent Praplan
Prodigy 40 points
Part Number: BQ40Z50-R2

Hello,

We are working with BQ40Z50-R5 gauge, and we are trying to optimize it for a new battery. There is one problem we are facing:

When we start discharging, the RSOC will stay at 100% for a long time because RemQ is higher than FCC. What I don't understand, is that after charge, passed charge is 2980mAh, but FCC will only be updated to 2880, so 100mAh less. After battery rest, RemQ is lowered to FCC value, and when we start discharge, RemQ will take the higher value back: 

I tried running a few cycles (with 100% RA change allowed) to make sure RA tables and Qmax are updated, but the behavior is still the same. Temperature during charge and discharge cycles is only varying a few degrees. You can find the config file here : bmon.gg.csv

Is it expected the FCC is not increasing more ? And can you help us avoid RSOC staying at 100% for a long time ?

Thank you in advance for your help.

Vincent Praplan

  • 0
    TI__Mastermind 37761 points

    Hey Vincent,

    Can you also share the entire log file. Will need to analyze the entire log to see have a better picture of what's going on.

    Regards,

    Nick Richards

  • 0 in reply to Adrian
    Prodigy 40 points

    Hi Nick,

    You can find the log file here : Full Log

    Best regards,

    Vincent

  • 0 in reply to Vincent Praplan
    TI__Mastermind 37761 points

    Hey Vincent,

    Thank you for sharing the log file, I will need some time to review this. I should get back to you by tomorrow. 

    Regards,

    Nick Richards

  • 0 in reply to Adrian
    TI__Mastermind 37761 points

    Hey Vincent,

    I looked over the data and I believe what you are seeing is normal. It its much more important to look at RemCap and FCC, these vales were equal to each other after charge. I would recommend trying another test where there is a larger discharge current and see if you seeing the same results. We are suspecting that the low discharge is causing SOC to stay at 100% for longer.

    Regards,

    Nick Richards 

  • 0 in reply to Adrian
    Prodigy 40 points

    Hi Nick,

    Thank you for looking at the logs.

    There is something I don't get. From my understanding, a higher discharge current should in fact reduce the FCC estimation, so the gauge should stay even longer at 100% (if the charge energy is the same). Can you explain why you think the low current might be the issue ?

    And in fact a higher discharge current would not help here, because our real discharge current will be significantly lower. For the test I'm speeding up the discharge with 300mA, but the real use case will be more between 30-60mA discharge....

    Today I did another cycle where I lowered the charge current (2000 -> 600mA) and taper current (400-> 50mA). The passed charge was higher, but the FCC jumped almost to this same value. With the same 300mA discharge current, the gauge did not stay longer at 100%. Unfortunately I made a wrong manipulation and the current was not logged, but you can see the other values : Lower CC cycle

    Could it be the high taper current that makes the gauge think the FCC cannot be that high ?

    I will continue making some tests, also with higher discharge current to see the difference.

    Best regards,

    Vincent

  • 0 in reply to Vincent Praplan
    TI__Mastermind 37761 points

    Hey Vincent,

    Our thought process was that there might not have been enough past charge with the small discharge to trigger the impedance track simulation to decrease SOC from100%. However, I reviewed your gg file again and gained new insight. You stated that with the same discharge current, the gauge did not stay longer at 100% and updated SOC more appropriately. In your settings, I see you have load select was set at 0 (Average I last Run). This makes sense since the last test run was at the -300 mA discharge, since the new test was also run at -300 mA, the Impedance Track algorithm was able to run accurate simulations for the new test (thus better SOC updates). We usually recommend having load select set to 1 (Present average discharge current), as this does not depend on the previous discharge cycle.

    Regards,

    Nick Richards

  • 0 in reply to Adrian
    Prodigy 40 points

    Hi Nick,

    I did other cycles before, also at -300mA. So for the two cycles I shared, the "Average I last Run" was already the same at 300mA. The difference must come from something else. I think it's better not to change now, but after the problem is solved, I will try your suggestion of "Present average discharge current".

    I was wondering if it could come from the end voltage of the 2nd cycle that was significantly higher, with the lower taper current. While the battery was in this EOC state, at rest, I tried loading other chemID from the GPC list (I made sure to have always the same RA values). After the gauge reset, some chemID did give a higher FCC estimation than others. I'm running new cycles with one of these other chemID, and in fact the gauging looks better with shorter time stuck at 100%.

    I thought a different chemID would only change SOC but not FCC, that it is linked to RA tables. Is it expected FCC can change too ?

    Then I'm also wondering if my Ra coefficients are not too high. (They are much higher than the ones from the chemID I tried, and when I changed them, the FCC was dropping a lot) I will try to make new learning cycle to see if I get different values. 

    Best Regards,

    Vincent

  • 0 in reply to Vincent Praplan
    TI__Mastermind 37761 points

    Hey Vincent,

    SOC, FCC, and Remaining Capacity are all correlated to each other. Yes, FCC is also expected to change when the ChemID changes.

    When you refer to Ra coefficients, are you specifically talking about the Ra table, these values should not be manually changed.

    Regards,

    Nick Richards

  • 0 in reply to Adrian
    Prodigy 40 points

    Hi Nick,

    For the RA coefficients, I had the impression they were quite high. So I did manually set them to a lower value, and made a few discharge cycles for the gauge to update them again. After two cycles with RA update, the coefficients were about 20-30% lower than what I had before. It's like if the gauge will only increase the coefficients when needed, but not decrease them ? (or less easily)

    And now, with the freshly updated coefficients, and the new chemistry, the overcharge is significantly reduced, less than 2% of the total battery capacity, which is ok for us !

    Best regards,

    Vincent

  • 0 in reply to Vincent Praplan
    TI__Mastermind 37761 points

    Hey Vincent,

    That's great to hear. With more cycles, the accuracy should hopefully continue to improve.

    Regards,

    Nick Richards