• State Resolved
  • Locked Locked
  • Replies 26 replies
  • Answers 4 answers
  • Subscribers 65 subscribers
  • Views 962 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.

BQ34Z100: Update status is not updating to 05 after charge-Relax cycle

Sudha Nayak S
Intellectual 380 points
Part Number: BQ34Z100
Other Parts Discussed in Thread: BQSTUDIO, BQ40Z50

Hi,

We are using Lithium Ion NMC 3s1p battery with each cell voltage 3.7-4.2V with 2900 mAh capacity. We have got Chem ID as 3288 from GPC tool.

We have programmed this chem ID using BQstudio and we are using BQ34Z100 as guage meter.

We are following the steps for learning cycle as shown below

1. Enable IT
2. Send Reset command (0x0041)
3. 550mA Discharge to empty
4. Rest 5 hours (Important)
5.The [VOK] bit and [RDIS] bit in the IT status() register got cleared  and The Flags[OCVTAKEN] bit did set when a valid OCV reading occurred
6. 1A Charge to full . At the end of charge the [FC] bit in the Flags() register did set.
7. Rest for 2 hours (Important)
8. The [VOK] bit in the IT status() register is cleared  and The Flags[OCVTAKEN] bit did set when a valid OCV reading occurred.

however update status is not updating to 0x05 as expected.

Because of this we are not able to move forward to do discharge cycle.

Please help us to understand what step we are missing because we are following same steps as mentioned in the document as well as those steps are confirmed by TI forum. .It's been more than 2 months and we are still stuck at learning cycle which is gating our further implementation.

Please find the attached .gg file before and after charge relax cycle and log file.

Learning_cycle_30March.csvstart_of_LearningCycle.gg.csvafter_charge_relax.gg.csv

Best Regards,

Sudha Nayak

  • 0
    TI__Genius 10370 points

    Sudha,

    The reason the learn cycle is failing is because QMaxDOD is not satisfying the 90% DOD update needed to get the initial QMax. At this point i am not sure what is causing the delta between QMaxDOD and DOD0. 

    Please discharge the device to empty, Reset the gauge, then wait for [VOK] to toggle. DOD0 and QMaxDOD0 should both update to the same value. After that follow your same procedure and charge to full. You should then see QMaxc DOD update. The delta between the two DOD points should be bigger than 90% 16384 in order to get the first Qmax update. 

    With you only using a 3S battery why not use the bq40z50 device instead of the bq34z100? The bq34z100 uses the resistor divider and gauges on the pack level instead of the single cell level which will impact gauging. In addition with the bq40z series you will get added features like cell balancing. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 380 points

    Hi Eric,

    Thanks for your reply.

    We have checked the log which we have shared with you and in that we can see DOD0 and QMaxDOD0 updated to 16179 value in charge relax cycle. 

    Please find the attached screenshot of the log 

    We carried out the steps as you have mentioned. Discharged the battery to 9V(termination voltage) and did Reset the gauge. [VOK] bit got set to 0, when[OCVTAKEN]bit becomes high, DOD0 and QMAXDOD0 values are 16108 and 16107 respectively.

    Please find the attached BQStudio screenshot

    We did not understand at which point, QMaxDOD is not satisfying the 90% DOD update needed to get the initial QMax ?

    Please help us to understand this so we can carryout learning cycle from the beginning.

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    Intellectual 380 points

    HI,

    Any update on this? This issue is gating our further implementation.

    Please guide us with your valuable inputs.

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    TI__Guru** 105760 points

    Hello Sudha,

    I think this is similar to a new post made, it would be most efficient to keep the same discussion in one thread so we don't have multiple threads with the same questions.

    The DOD varies from 0 to 16384, between the two relaxation periods you must see a change of at least 90% in this value (for example 1000 to 16000 would be ((16000-1000)/16384)*100 = 91.6% change, this would be good for the learning cycle process.

    https://e2e.ti.com/support/power-management/f/power-management-forum/991420/bq34z100-g1-need-to-have-understanding-of-life-cycle-characterization-which-is-not-working-with-the-battery

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Intellectual 380 points

    Hi Wyatt,

    Thank you for sharing the information. I will check the log.

    Sure, I will maintain single thread.

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    Intellectual 380 points

    Hi Wyatte,

    We have procured new Panasonic battery ie NCR18650GA and we found Chem ID for this as 2107 from BQstudio Chemistry selection tab.

    We are using 3S1P battery with 3450mAh capacity.

    We have programmed the Chem ID using BQStudio and started learning cycle.

    1)Enable IT_Enable

    2)Reset

    3)Discharge the battery till it reaches 9V (ie Termination voltage)

    4)Relax for >5Hrs

    5)Charge the battery till 12.6V (FC=1)

    6)Relax for >2Hrs

    We were expecting VOK bit to get clear and update status to update to 0x05, however VOK bit got cleared but Update status didn't update to 0x05.

    When battery  was in relax state (after full charge ie FC=1 and VOK =1) we  have observed, when OCVTAKEN bit got set, FC bit became zero and Full Charge capacity value jumped from 897mAh to 900mAh and remaining capacity became 0. After some time VOK bit became zero, however update status didn't update. We are not understanding what is cause for this kind of behavior.

    Since this is standard battery and we have Chem ID for this from BQStudio and we are following same steps as mentioned in the documents as well as from TI experts, still we are not able to complete the learning cycle.

    We have attached .gg file and log file. Could you please go through it and let us know if we have wrongly configured any parameter and which is stopping us from completing the learning cycle.

    Please help us to figure out the issue so we can complete the learning cycle at the earliest because it is gating our further implementation.

    ChemID_2107_gg.gg.csvlearning_cycle_2107_charge.log

    Thanks and Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    Intellectual 380 points

    Hi,

    Any update on the above issue?

    Its been more than 3 months and still we are not able to complete the learning cycle. We have created different threads and following the same steps suggested by TI experts. However we are ending with incomplete learning cycle.

    Please help us to solve this issue and to complete the learning cycle.

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    Intellectual 410 points

    Dear Wyatte, 

    It would be very helpful if we get a reply and help on the issue of LifeCycle that we are facing working with BQ34Z100. 

    We have invested significant amount of time, effort and resources both in terms of Board development, Evaluation kit procurement and everything trusting Texas Instruments Support and Technology. But time and again we are hitting bottleneck and with not enough support and help. 

    Thanks and warm regards

    Dharin

  • 0 in reply to Dharin Shah
    TI__Genius 10370 points

    Dharin/Sudha,

    Something is wrong with your log file. For update status to go to 0x05 you must get a qmax update. The first QMax update requires you get 90% change in DOD between your two rested points. DOD is a value that ranges from 0 to 16384 based on the chemID. You log file has this value as 14k-36K which cannot happen.

    The log file seems to be exported using v1.06 while the gg file is v0.16. Please re-run the test using the correct version of bqStudio so we can correctly analyze the log file. bqStudio should auto-detect when opened if you have the right one.

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 380 points

    Hi Eric,

    When we are opening the log file and gg file, we are observing same FW version as 0.16, Please find the attached images.

    Even when we are reading FW_VERSION from BQStudio it is showing 0x0016. 

    Since we are using standard battery with Chem ID from BQStudio and following same steps as mentioned in the document, then what might be the reason that is causing Qmax value out of range?

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    TI__Genius 10370 points

    Sudha,

    It is not an issue with the ChemID. This seems to be a mis-alignment in the data logged. Instead I can use the QMax_DOD which seems to be correct (although I am hesitant to trust it). 

    The issue you are having is as I described above. You are not getting the 90% change in DOD needed for the first QMax update. At fully empty your DOD is 14944 and at fully charged you get 1840. 14944-1940 = 13004. 13004/16384 = 79%. 

    To solve your case. First discharge either to a lower voltage or at a smaller rate. When you rest DOD should be as close to 16k as possbile. Then when you charge you need to charge to a higher voltage or taper to a lower charge current.. This first cycle needs to exercise the battery against the battery specs and not the application. The first discharge after you successfully get a QMax update needs to be done a a c/5 to c/10 rate. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 380 points

    Dear Eric,

    Thank you for the detailed information.

    We will test again and get back to you with additional information found with above experiment.

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    Intellectual 380 points

    Hi Eric,

    We tried to bringdown the discharge voltage to 8.7V and relaxed it >5Hrs, however DoD values were close to 15k.

    Today again we will try to bringdown the discharge voltage to 8..1V and will check DoD values.

    We had doubt that if we bringdown discharge voltage to very low for several times while testing, does it damage the battery?

    Thanks and Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    TI__Genius 10370 points

    Sudh,

    You can lower the dsg rate and keep discharging until it gets lower. FYI 15K is enough as long as on the charge you get the battery to very full again. Remember the delta between DOD points needs to be bigger than (2^14)*0.9. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 380 points

    Hi Eric,

    Thank you for the confirmation.

    We discharged the battery till 8V and relaxed it for >5Hrs and we have seen DoD values as 16024. Then we charged the battery till 12.6V and once charging current reduces to taper current level we will stop charging and let it relax for >2Hrs, however when charging current was dropping , Full charge capacity value jumped from 869mAh to 930mAh and SOC changed from 100% to 0% because of this FC Status changed to 0 from 1. So update status didn't update to 05.

    Please find the attached log.

    Please help us to understand, what is the reason for this kind of behavior. Why Full charge capacity is suddenly changing?learning_cycle_29_4.csv

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    Intellectual 380 points

    Hi,

    Any solution on above issue?

    Best Regards,

    Sudha Nayak 

  • 0 in reply to Sudha Nayak S
    Intellectual 380 points

    Hi Eric,

    What might be the reason for Full charge capacity value jumping from 869mAh to 930mAh and SOC changed from 100% to 0% ?

    Looking forward to hear from you at the earliest.

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    Intellectual 380 points

    HI,

    Please find the attached log file screenshot, where we can see SOC changed from 100% to 0%.

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    Intellectual 380 points

    Hi Eric,

    We are charging the battery at 1A. IS this is the reason for sudden drop in SOC? Should we reduce the charging current?

    Can Higher temperature cause this kind of behavior?

    Please help us to understand why SOC is suddenly dropping to 0%.We have already invested so much of time in learning cycle and still not able to complete it. This is blocking our further implementation.

    Best Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    Intellectual 410 points

    Dear Eric, Wyatt, 

    Could we get a reply to the issue raised here? 

    We have come to a point where we tried multiple options:

    1. Changed Battery itself to go with a battery with a known ChemID so as not to have ChemID related issue. 

    2. Reduced Discharge Rate. 

    3. Reduced Discharge Voltage point so as to ensure that we get DoD of the order of 16K. 

    4. Also tried to get full DoD range of 90% 

    But we are getting stuck with one "more" bottleneck and Life Cycle Characterization is a multiple days effort. So every time we run across an issue, we are back by 4-5 days and need to re-start everything. 

    Could you please help us with how we address the current issue and be able to move forward and close this exercise?

    Thanks and warm regards

    Dharin

  • 0 in reply to Dharin Shah
    TI__Genius 10370 points

    Dharin/Sudha,

    A few notes I want to point out for others who read this thread

    1) FCC is irrelevant until you have a learned battery (update status 06 or 0E) Please do not look or refer to it

    2) DOD change of greater than 90% is step 1. If DOD after discharge is not close to 16384, you either need to discharge to lower voltage or smaller rate. 

    3) If DOD after charge is not close to 0 (it should not be 0), you need to charge to a higher voltage or taper further

    In your log file there is a major issue I have already pointed out which does not make sense. Your DOD column is bigger than 16384 which is impossible in the gauge. Something about your setup is corrupted. I will take this offline tomorrow with you and schedule an engineer to work with you to debug. I suggest you start with reprogramming the base firmware from TI.com. Something inside the IC is not correct or communication is bad and we are looking in the wrong place. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 380 points

    Hi Eric,

    As I have mentioned, we are seeing sudden drop in SOC ie 100% to 0% because of that FC bit is 0. Can we continue charge relax cycle without FC bit is set to 1?

    We will reprogram the base firmware and start learning cycle again, will get back to you on this.

    it would be great help if we have discussion with an Engineer to discuss on where we might be going wrong. We are open to WebEx, Microsoft team or any screen sharing session where we can do screen share and discuss on the issues because we are completely clueless on how to solve this issue and move forward .

    Best Regards,

    Sudha Nayak

  • +1 in reply to Sudha Nayak S
    TI__Genius 10370 points

    Sudha,

    Yes you can continue to charge/discharge just ignore SOC/FCC/FC/ect.. 

    The E2E team is working on setting up a meeting.

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 380 points

    Hi Eric,

    We have installed latest version of BQStudio and carried out learning cycle, now we have got update status as 0x06.

    After that we completely charged the battery ie SOC as 100%. It shows Full Charge Capacity as 3291mAh, however when we disconnect the battery and reconnect it, the SOC is dropping to 77% because Remaining charge capacity is dropping to 2530mAh.

    We couldn't able to understand this why there is sudden change in remaining capacity if we reconnect the battery.

    Please help us to solve this issue.

    Thanks and Regards,

    Sudha Nayak

  • 0 in reply to Sudha Nayak S
    TI__Genius 10370 points

    Sudha,

    Disconnecting and re-connecting the battery will power cycle the device and rely simply on reading the battery voltage to establish a new SOC. If your battery has any sort of load on it then the voltage would be reduced and calculate a lower RemCap. 

    Please provide a log and GG file for this latest issue. Also i am going to close this ticket since the learn cycle issue is resolved. Please start a new one for better tracking. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 380 points

    Hi Eric,'

    Thank you for you help.

    Sure, I will create new query for change in SOC when disconnect and reconnect the battery.

    Best Regards,

    Sudha Nayak