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.

BQ27546-G1: GPCRA0 issue

Part Number: BQ27546-G1
Other Parts Discussed in Thread: GPCRA0, GAUGEPARCAL, GPCCHEM

Hi,

from https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1113669/bq27546-g1-soc-jump-during-relax-discharge-relax-test

this thread will concern Ra0 and GPCRA0 issue.

Before running the test I want to make sure something.

1.How could I know that Ra0 I got is a proper one? is there a range of Ra0 for this ChemID? What if getting samples of Ra0 and find average?
2.the limit of discharge rate is?
3.gg file sent to GPCRA0 should be pulled before or after the test?
4.is it ok to log temparature with one decimal place?

  • #1: You'd have to check, if the source data (the current, voltage, temperature log file) was done exactly as described in GPCRA0. If this is correct, then the result from GPCRA0 will be correct.

    #2: The minimum load is approx. C/10. Below this, the current and voltage info usually isn't high enough resolution to calculate Ra accurately. The max. load should not exceed the limit of the cell datasheet. It should be a system high load because part of the GPCRA0 process is the thermal model (T rise and T time constant) and that is best measured when the cell heats up like in the end application.

    #3: It should be before the test (=after your first programmed the ChemID (it needs to be a non-modified ChemID (non-learned)).

    #4: Yes.

    The instructions for GPCRA0 must be followed very carefully (esp. the charge and relax times). Also, make sure that this is done with the cell in the pack installed in the end application for best thermal parameter results.

  • Hi, After running the test, my Qmax is 98838 which exceed the maximum number.

    this is my source file toGPCRA0_02.zip

    and this is the report toGPCRA0_02-report.zip

    I did let the thermistor be underneath the cell. I have a box cover the board, but It is not a thermal box.

    I set AC to 28 degree celcius. relax time are 2.5 hrs and 5.5 hrs for charging and discharge. discharge rate is C/5.

  • How did you identify the ChemID (1122). Was this via GAUGEPARCAL?

    1122 is for a 10000mAh cell, hence the Ra tables are going to be *very* different from a 90mAh cell.

    Please confirm: You want to use a 90mAh cell?

    GPCRA0 interprets the current info in your log file as Amps and not mA because the values are so small. The gauge uses integer math and with a 90mAh cell, you'll reach the limits what the gauge can reasonably resolve.

    You'll have to use current scaling to make this work well. Change all capacity related parameters with a scaling factor of 10 (e.g. Design Capacity = 900mAh, Qmax = 900mAh, Design Energy = 3420mWh etc.). Then change the sense resistor to one that's 10x bigger.

    I re-submitted your data to GPCRa after scaling current x10 to test if GPCRa picks this up correctly and it does:

    Optimized Impedance Track parameters:

    Qmax, mAh : 988

  • Thank you so much for your answer.

    I use GPCCHEM to identify CHEMID, I know that 1122 is for 10000mAh cell, but it is what the GPC told me. So can't help using the ID.

    I DO want to use a 90mAh cell and I know that BQ27546's minimum support capacity is 100 mAh, but It must be BQ27546.

    So the next thing I need to do is to change parameters in gg file and resend zip file to GPCRA0 to find a new Ra table, isn't it?

  • What I recommend is changing the sense resistor to a value that's 10x higher than what your use now and then modify the configuration so that the gauge thinks it's a 900mAh cell (=10x current scaling). This allows the algorithm's integer math to work accurately.

    You'd ideally re-measure the cell behavior (=repeat the relax/charge/relax/discharge log file generation) because you'll get much better data for the current in mA from the gauge.

    It may be acceptable (but not ideal) to just do what I did, edit your log file to multiply the current column data by 10 and re-submit to GPCRA0.

  • As I remember the sense resistor is limited to 5-20m ohm? I just want to make sure it 's ok to higher up.

    for the change of sense resistor, It is only for the test or also need to implemented in final device?

    well,after the gauge think it is 900mAh, do I need to re-identify ChemID?

  • The gauge measures voltage over a sense resistor. As long as the voltage is within the specified limits in the coulomb counter, there is no limit to the sense resistor. There are practical limits (esp. taking the integer resolution of several gauging results into consideration) but it's incorrect to say that the sense resistor must be between 5mOhm to 20mOhm.

    You will need to implement the higher sense resistor also in the end application.

    You do not need to re-identify the ChemID.

  • I have got new Ra table and Qmax. but Once I set them in data memory and reset. design capacity dropped to 972 and state of health drop from 90 to 76. is it because of the cell?

    ps. the scaling factor is 12.75

    here is the report  toGPCRA0_06-report.zip

  • You need to also configure load prediction according to your current scaling. GPCRA0 will report cell Ra based on the scaled current, hence the gauge will calculate a higher voltage drop than "reality" and this must be taken into consideration with the load prediction settings (Load Select).

    SOH is a function of Qmax and cell resistance and SOH Load I so this also must be set correctly, otherwise the gauge will calculate diminished SOH.

  • update, SOH is 100 now. thank you