• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 369 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.

BQ78350: CEDV Gauging Parameter Calculator (Deviation High)

Min Hsiao1
Prodigy 30 points
Part Number: BQ78350

Hi

The data that I uploaded keep on getting the "The following errors were reported by the calculation engine: Accuracy Error: Deviation is so high that it is most likely due to anomaly in the data. Please check that data files have recomended format, units and test schedule" error message. I'm not sure if there's some weird formatting thing that I did wrong. Can you take a look at my data set?TI CEDV.zip

  • 0
    TI__Intellectual 1090 points

    This data-set includes extremely high discharge rate data for high rate files, with whole discharge taking only 200 sec and temperature increasing to 80-100C. This looks like a highly abnormal operation for the battery that will damage it. Gauging is also not designed to operate under transient conditions, so fit of CEDV equations can not be good at the same time for the low rate (normal condition) and super-high rate (transient conditions). I was able to get good accuracy for the low rate files with settings
    FitMaxSOC%=9
    FitMinSOC%=3
    LearnSOC%=7
    But high rate data is just too variable due to transient effect and drastic temperature increase, so error remains very high. 

    I woudl also try to decrease terminate voltage below 3.3V, as for extremely high rate applications even 3V is usually to high and 2.7V much more common to get
    more run-time, if the device allows it. However with this extreme discharge rate even going to 3V did not help, so I would recomend to rerun the test under more typical continious discharge condition at high rate, rather then the extreme peak current case.

  • 0 in reply to Yevgen Barsukov
    Prodigy 30 points

    This IS our application use case tho. What is the consequence of using the existing data? Is it gonna be worse than using some random default parameter values?

  • 0 in reply to Min Hsiao1
    TI__Intellectual 1090 points

    If you have to go with this data, I would refit it with below settings:

    FitMaxSOC%=9
    FitMinSOC%=3
    LearnSOC%=7

    This will give good accuracy for the low rate case discharge, but accuracy for high rate discharge will still be bad.

    To get better accuracy for high rate discharge, I would rerun it with a lower rate. You mention that this is an application case but is it the worst case scenario that is rare? If so, rerun it with a more typical scenario.

  • 0 in reply to Yevgen Barsukov
    Prodigy 30 points

    ok, i'll resubmit with those settings.
    Is the problem primarily with the high temp rise? or varying temp rise between different starting temp? or varying temp rise between different load profile?
    This is pretty much our application. I picked the loading as an average, the transient is 2-3 times that number.

  • 0 in reply to Min Hsiao1
    Prodigy 30 points

    basically trying to understand how the values are derived and what the short coming is.

  • 0 in reply to Min Hsiao1
    TI__Guru** 105760 points

    Hello Min,

    With the high change in temperature and the big difference in current from high rate to low rate, it is not possible for the CEDV parameters to accurately predict both conditions. As Yevgen said you will need to prioritize one discharge rate for accuracy, but one rate will not be as accurate.

    Sincerely,

    Wyatt Keller