• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 4 answers
  • Subscribers 63 subscribers
  • Views 778 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.

BQ40Z80: Learning cycle fail because of design capacity (24v 150000 mah)

Rudraradhya P
Prodigy 180 points
Part Number: BQ40Z80
Other Parts Discussed in Thread: GPCCHEM, , BQ40Z50

Hi

We are trying to generate the Golden file for 24V 15000 Mah Battey and all time the learning cycle is failing due to design capacity or for any other reason.

For the same batteries with the configuration of 24V 2600Mah (1p6s) Golden file generated with no issue. 

We observed that the Full charge capacity is reduced to half of the Qmax we setting, pls let us know what the issue.

Chem Id is as below and log file is  attached

Thank you 

Pramod

19 11 2020 (2 trail).log19 11 2020(2trail time 11.30).log

  • 0
    TI__Guru** 105760 points

    Hello Pramod,

    How did you select the chemistry ID? Can you share the GPCCHEM report?

    We usually reference SLUA903 or the learning cycle guide on the bq40z80 product page for these issues.

    From looking at your logs, the learning cycle is not completing because you are not getting a 90% DOD change, this could be due to selecting the best chem ID or simply not cycling the voltage the full battery specification range.

    www.ti.com/.../slua848

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 180 points

    HI Wyatt Keller,

    The selected Chem ID is working good for other battery configuration like 6s1p and 6s2p, as I mentioned earlier we are able to generate the Golden file for all other combination with selected Chem ID.

    But Not for 6s6p, do you mean that for every battery configuration we have to change the Chem ID. ?

    for 6s6p, why the Full charge capacity is reduced to half of the Qmax we setting. 

    Regards

    Aradhya

  • 0 in reply to Rudraradhya P
    TI__Guru** 105760 points

    Hello Aradhya,

    You may need to set the cell interconnection resistances, or make sure all the connections are secure. The same chemistry ID should be ok as long as the voltage range they are using is the same as the other applications (not de-rated voltages)

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 180 points

    I Didn't understand cell interconnection resistance. can you please explain.

    Also I tried 1s6p, 2s6p and 3s6p combination.

    -For 1s6p Full charge capacity was 15000mah,

    -for 2s6p full charge capacity was 126000 mah and

    -for 3s6p full charge Capacity was 110125 mah 

    As we increase the series connection the Full charge capacity is reducing and due to this the learning cycle is failing.

    May I know the reason for reduction in Full charge capacity as we increase series combination.

    Regards

    Aradhya

     

  • 0 in reply to Rudraradhya P
    TI__Guru** 105760 points

    Hello,

    The cell interconnection resistance may drop the FCC value depending on how well the cells are connected in series.

    You can reference the Impedance Track app note for more information regarding the FCC updating: e2echina.ti.com/.../Theory-and-Implementation-of-Impedance-Track-Battery-Fuel_2D00_Gauging-Algorithm.pdf

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 180 points

    HI Wyatt Keller

    The Above solution didn't worked, we have observed that in our 6s6p configuration all the series cells are not getting discharged to full(Only Port 6 is getting discharged for capacity )  and hence may the learning cycle fails. Attached are a couple of our log files. can you please analyze and let us know what is the issue.

    Regards

    Aradhya

    log 14-12-2020.zip

  • 0 in reply to Rudraradhya P
    TI__Guru** 105760 points

    Hello Aradhya,

    Are you using a 15,000mAh pack or 150,000mAh pack? In your previous posts they keep switching.

    It still looks like you are not cycling by 90% DOD. Please follow the app note for doing a learning cycle for the bq40z80: www.ti.com/.../slua848

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 180 points

    Hello Wyatt Keller,

    we made the learning cycle to  90% DOD but, the Learning cycle fails because of the below reasons

    1.FCC will get decreased by 2000 mah(while discharging FCC=14000 mah ,after the fully discharged FCC =11500mah) 

    2.we charged the battery to 100% but fails to set the FC(fully charged)  flag because of   CTOS(charge time out) flag enabled.

    Thank you 

  • 0 in reply to Rudraradhya P
    Prodigy 180 points

    HI ,

    We made one more attempt with changing the cells, changed the cells from ICR_1865026J to LG_INR18650_M26. With LG_INR18650_M26 these cells, we are able to generate the golden file for 4S1P. But with the same cells, the cycle is failing for the 6s6p configuration. I have attached the log files. Can you please let us know what may be the issue?

    Regards

    Aradhya3_Jan_2021.zip

  • 0 in reply to Rudraradhya P
    TI__Guru** 105760 points

    Hello Aradhya,

    From looking at the logs I'm not sure I understand the learning cycle, your Vbat and Vpack don't correlate and your cell 5 seems to be way out of balance with the rest of the cells. At the start of the learning cycle the cells should be balanced.

    It looks like the DOD change is right at the borderline 90% value as well.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 180 points

    Hi Wyatt Keller,

    I corrected the  Vbat and Vpack by calibrating the Vpack Voltage. Also, I removed Cell5 and made a set up of 5s6p. Attached is the log file and the gg file. The FUSE bit has got set this time, do I have to short the fuse pin to the ground, or do I have to update any settings? Please suggest.

    I am using the BQ40Z50 EV board for the learning cycle. Attached is the log and gg file.

    Regards

    Aradhya

    07-01-2021(4).zip9-Jan-2021.gg.csv

  • 0 in reply to Rudraradhya P
    TI__Guru** 105760 points

    Hello Aradhya,

    You should perform the learning cycle with the PCB for your application so the trace/connection losses can be seen by the gauge during learning.

    If you are not using the fuse you can disable it. If it's not used it should be driven to VSS as stated in the data sheet.

    You're using the bq40z80 EVM for all the testing right now?

    Sincerely,

    Wyatt Keller