• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 62 subscribers
  • Views 212 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

BQ27742-G1: Smoothing Engine Algorithm

WY Lin
Prodigy 10 points
Part Number: BQ27742-G1

How are FCC & RM filtered to get smoothened SOC?

FCC and RM get filtered with the same algorithm and frequency?

The SOC jumps mainly come from RM change or FCC change?

As the battery accumulates more cycles, will the filtered data and unfiltered data converge?

  • 0
    TI__Guru** 102355 points

    Hello WY,

    The smoothing engine is very complex and is out of scope to explain it in this thread.

    The essential operation of the smoothing feature is to take any jumps in the true values of FCC and RemCap and time average them so they converge when we would expect without jumping. The smoothed values also can have other features influence them as well and this is why it can get complex.

    Sincerely,

    Wyatt Keller 

  • 0 in reply to Wyatt Keller
    Prodigy 10 points

    Hi Wyatt, 

    Thanks much for the response.

    We are using the RemCap data to do our own calculation and because we have the SmoothEN bit set, we refer to the unfiltered data. 

    However, the sudden drops in the unfiltered data is bothering us, but if we redirect to the filtered data sometimes it's gives an unreasonably high number. 

    Any suggestion on where in BQ27742-G1 we can find a remaining capacity data that do not jump and stays in reasonable range at the same time?

    Thanks in advance.

    WY  

  • 0 in reply to WY Lin
    TI__Guru** 102355 points

    Hello WY,

    This is the behavior of the gauge, we run simulations based on the expected load using load select, Qmax, Ra and other factors to estimate the FCC and RemCap. If the conditions change quickly or if one of the inputs to the simulations changes (high load from previous cycle and load select uses the last cycles load current) then it will cause a jump.

    You can refer to the impedance track application note explaining the operation: https://www.ti.com/lit/an/slua364b/slua364b.pdf

    It will not be possible to back calculate the FCC and RemCap values, they are too complex and from the outside perspective not knowing the gauge internal registers the calculated value will never match the expected exactly.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 10 points

    Hi Wyatt, 

    Thanks much for the speedy response and explanation. 

    Need to bother you with some additional question though..

    Is there a chance that filtered RemCap turns out to be a greater value than the designed capacity?
    we've seen filtered FCC be that way, but we are wondering if the filtered RemCap would be more subtle?

    Also, what would be the most common case for the unfiltered RemCap to have sudden drop? It is to my understanding that RemCap takes measurement every 1 second. 

    Thanks in advance. 

    WY 

  • 0 in reply to WY Lin
    TI__Guru** 102355 points

    Hello WY,

    Yes it is very possible for the filtered RemCap to be higher than the design capacity. Design Capacity should be the value from the cell manufacturer datasheet and is usually a nominal value.

    If the FCC is recalculated the RemCap will also be adjusted with the FCC jump. FCC is not updated every second, there are specific triggers when the gauge will run an FCC simulation. They are outlined in the TRM and the application notes for Impedance Track.

    Sincerely,

    Wyatt Keller