• State
  • Locked Locked
  • Replies 9 replies
  • Subscribers 63 subscribers
  • Views 1297 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.

Problem with average Time to Empty & SOC / LEAD Acid Battery

LOIC REBOUSSIN
Expert 1620 points

Hi,
I have an optimized firmware with my Lead acid battery.
In fact, I learned to BQ optimized cycle, Chem ID... And everything is good.

Now, I want to verify gauge in my application. So I charged my battery and everything are OK.
But after that, when I discharged my battery under ~ C/2 I have a big problem:
Average time to Empty and SOC is too high at the end of discharge. See this below:


The discharge of battery are stopped under 8V ( < Vcell terminate voltage)

you can see that SOC = 20% and  average time to Empty = 22min, it's a big problem for knowing cut of battery  !

Can you explain why ?



Thank you,

Best regards,

Loïc

  • 0
    TI__Guru 73830 points
    Loic
    Can you provide the log file and an srec file exported from the pack?
    Tom
  • 0 in reply to ThomasCosby
    Expert 1620 points

    Tom,

    thank you for your help,

    Please find attached files,

    Best regards,

    Loïc

  • 0 in reply to LOIC REBOUSSIN
    TI__Guru 73830 points

    Loic

    I updated the Ra table to fit your discharge rate and changed the Load Select to improve performance. You can try this srec file.

    Tom

    https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/196/0100_5F00_0_5F00_16_2D00_bq34z100G1_5F00_modifiedTI.srec

  • 0 in reply to ThomasCosby
    Expert 1620 points
    Tom,
    Thank You,
    I am going to re-try with this configuration. But I what does not include what I missed not to obtain the good result ....
    Loïc
  • 0 in reply to LOIC REBOUSSIN
    Expert 1620 points

    Tom,

    I have re-try with your Firmware "0100_0_16-bq34z100G1_modifiedTI.srec" but It didn't works properly :

    After this test, I have exported parameters and I have found that "Number of series" parameter = 1 instead of =6

    Maybe it's this paramter that cause this error?

    Loïc

  • 0 in reply to LOIC REBOUSSIN
    TI__Guru 73830 points

    I had changed the Number of Series Cells to 1, because it is easier to work with on the simulator and I forgot to change it back. That would cause gauging problems oon a real pack. Can you change the parameter? PbA cells are more difficult to gauge than Li-Ion cells, because they have a non-linear characteristic to the series rresistance. So, sometimes we have to update the Ra table to match the impedance at the discharge rate used in the application. Hopefully, that will help. 

  • 0 in reply to ThomasCosby
    Expert 1620 points

    Tom,

    Thus, to have no this problem, I presume that I have to discharge the pack under current rate of application (Instead of C/5) during Qmax and resistance learning ?

    I make a reference to SLUA334B procedure.


    Loïc

  • 0 in reply to LOIC REBOUSSIN
    TI__Guru 73830 points

    That would provide a more accurate resistance table in this case, but there is also a change in DOD requirement that can restrict Qmax updates. The change in DOD requirement is 50%, for PbA cells, during the optimization cycle and 37% after that. You may not meet the 50% requirement at a higher discharge rate. The Mathcad program that i use does not have this restriction.

  • 0 in reply to ThomasCosby
    Expert 1620 points

    Hi Tom,

    Thank you for your help.

    I have a better result now :

    In future application, I will apply a different current rate during Qmax and Ra Tables Update (Learning status) with : I(application) <  I(Learning) < I(50% Dod restitued)

    I retrieved the log file and with your application report "Accuracy and Verification Methodologies for Advanced Fuel Gauges" (SLUA684) I found :

    1- An average time to Empty error between  -4,5min < ATTE_Error < 6.07 min (first 8 minutes log exclude)

    2- A SOC error between -2.57% < SOC_Error < +6.46%

    3- A Remaining capacity error  between -11.77% < Qrem_error < +5.46%

    what is the criteria for knowing if results are in the good range ?

    Best Regards,

    Loïc