• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 739 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:

BQ40Z50-R2: QMAX update is not working with OCVFR bit set

BQ40Z50-R1: Failing Learning Cycles due to OCVFR for Chem ID 0x0440

James Sedillo
Prodigy 111 points
Part Number: BQ40Z50-R1
Other Parts Discussed in Thread: GPCCHEM

Failing most learning cycles on 3sp1 packs of ANR26550-m1b cells with a bq40z50-R1. OCVFR remains set on a charge to 10.8V (3.6V/cell).

A solution is appreciated.

James S.

  • 0
    TI__Guru** 105760 points

    Hello James,

    Have you used the GPCCHEM tool to program the gauge with the best chemistry ID? Usually this is a common missed step.

    Also charging to 3.6V may put the voltage in the flat region for your battery, which means the gauge won't take an OCV measurement for the learning cycle.With the low final charge voltage you may not be getting the required 90% change in DOD needed for the learning cycle to complete successfully as well.

    I would recommend reviewing the application note SLUA903, it has a good step by step guide for completing a learning cycle: e2e.ti.com/.../Achieving-the-Successful-Learning-Cycle.pdf

    Sincerely,

    Wyatt Keller

  • 0
    Prodigy 111 points

    Is there any further explanation on OCVFR and "FlatVoltMax." I feel like I'm aiming for an undisclosed target and I don't know why TI just doesn't disclose what this supposed region is as it is a tremendous barrier to success with this product. My battery chemistry is among those listed in the chemistry database so I don't know why the Chem tool would be necessary. Also, given that this is a series-configured pack, how much voltage variation is permitted between the cells at VCT so has not to trigger OCVFR? Furthermore, why is it that at VCT, I can transition to update status 0x05 (indicative a potentially successful QMAX update), but it isn't until a full discharge, and I fail to obtain update status 0x0E, that it becomes apparent my learning cycle as failed?

  • 0 in reply to James Sedillo
    TI__Guru** 105760 points

    Hello James,

    The OCVFR is defined in the chemistry ID, you won't have access to the range which it is set.

    If it is the exact same battery you do not need to run the GPCCEM tool.

    There should be very little cell to cell variation when performing the learning cycle, it should be started with balanced cells. If you are getting the OCVFR from cell voltage variation they are not balanced enough to do a learning cycle.

    The learning cycle needs two OCV measurements to calculate the Qmax parameter of your battery, that is the first updated from 0x04 to 0x05, then it needs to update the Ra table on the next discharge after learning the Qmax update

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 111 points

    In my last experience I achieved the transition from 0x04 to 0x05, but no success on the transition to 0x0E for the Ra table update. With 0x05, I should have established the upper bound or QMax. How can I possibly fail on the 0x05 to 0x0E transition? It seems like 0x05 would be the most difficult to achieve since it is the most sensitive to having a balanced pack.

    On the matter of cell balancing, cell voltage variation at full charge (i.e. at  VCT) on a series-connected battery pack  is inevitable. What is a reasonable cell voltage variation range to work toward to satisfy this device?

    Lastly, If a cell characteristic is submitted to the GPCCEM tool, will a FlatVoltMax value be provided or will it also be witheld information?

    Sincerely,

    James

  • 0 in reply to James Sedillo
    TI__Guru** 105760 points

    Hello James,

    The update status should change from 0x05 to 0x06 first before 0x0E.

    The discharge has to be within a certain range (C/5 to C/10) in order for Ra to update. Make sure to keep the current within that range during discharge in the learning cycle.

    In SLUA903 it explains how update status changes during the learning process.

    The flatVoltMax won't be reported in the GPCCHEM report, but while performing a learning cycle none of the cells should be within the range because it will lead to a failed learning cycle due to the 90% DOD criteria failing.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 111 points

    Wyatt,

    If I monitor the DOD for every cell, and see that one or more cells has a DOD below 90%, will it be obvious that my discharge has been or is likely to become unsatisfactory for learning?

    Also, a tuning question on the optimal pre-learning balancing process... Does the bq prefer a top or bottom balance process? I'm shooting for top balancing as I'm hoping it will gain me a favorable outcome with the mysterious flatvoltmax setting. My cells are LiFePO4 and have this crazy voltage relaxation behavior that decreases soon after a charge.

    And on the OCVFR flag. If this indication sets at any time during the charge relaxation stage, do I fail...or not necessarily? I've seen this toggle on then off in a single relaxation run.

    Thanks,

    James

  • 0 in reply to James Sedillo
    TI__Guru** 105760 points

    Hello James,

    If you are discharging the batteries and they are not balanced the battery pack will most likely not cycle by 90% DOD. Yes if one of the DOD0 values didn't cycle by 90% it will most likely fail the learning cycle.

    It is challenging to work with LiFePO4 chemistry because of the relaxation. I would check to make sure LFP_RELAX is not interfering with your learning process.

    The OCVFR flag just signals when you are in the flat region, it does not mean it will fail but it does indicate it will take 48hrs for an OCV measurement to be taken, which is required for learning process. You can turn this flag off to enable OCV measurements when the DV/Dt condition is met.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 111 points

    Wyatt,

    After manually balancing the cells via top balance and after a discharge, I achieved update status 0x0e on the subsequent charge cycle by toggling OCVFR off prior to the charge as you suggested. I retained the LFP_RELAX enabled.

    Thanks for the help.

    James S.

  • 0 in reply to James Sedillo
    TI__Guru** 105760 points

    Hello James,

    That's great! Let us know if you have any more questions.

    Sincerely,

    Wyatt Keller