• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 1015 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.

BQ34Z100-G1: golden Image neccessary?

Michael Schmid53
Prodigy 150 points
Part Number: BQ34Z100-G1
Other Parts Discussed in Thread: BQSTUDIO, BQ34Z100

Hello,

we plan to use the BQ34Z100-G1 with 80Ah 18650 LiFePo cells.

As the production process should not be too complicated, we would like to use the IC directly and only change some parameters in the config of the device. These changes will be done by our series microcontroller using I2C, without any bqStudio or computer connection in the test-bench.

Is it correct that the "chem ID" is something that is only in the BQ software and is not written to the device itself somewhere? Do we just leave the register "device chemistry" at "LION" and modify some of the settings like the "design capacity" and "design energy" and the "pack configuration"?

Is this possible, not using any fancy Software, golden image, ...? Would a learning cycle be performed automatically for each Battery once it is used first?

Is there an overview available, which registers we should modify upon first-time usage?

thanks,

Michael

  • 0
    TI__Guru 52115 points

    Hi Michael,

    The bq34z100-g1 is an Impedance Track fuel gauge.  You cannot simply change a few configurations and then expect an accurate state of charge reading. You must choose a proper chemistry id that matches your battery profile in the bqStudio and have it programmed to the gauge device.  It must be done with the help of bqStudio. 

    The learning cycle also needs to be done before a golden image file that can be used for mass production is generated.

    See the following app note that explains how to do the learning cycle.

    4034.How to Complete a Successful Learning Cycle for the BQ34z100.pdf

    Andy

  • 0 in reply to Andy Liu (BMS)
    Prodigy 150 points

    Hi Andy,

    thanks for your reply. We know the reading would initially not be very accurate, but as I understood the device is self-learning. So after a few cycles it should be ok.

    It seems very complicated and time-consuming to program such a golden image on each device in production from BqStudio.

    What exactly are the values that are in this golden image, could we not program them from our microcontroller which is on the same PCB anyways? Then we would not need additional hardware (like this EV2300) and software on the testbench. BqStudio does not seem to be useful for mass production anyways.

    best regards,

    Michael

  • 0 in reply to Michael Schmid53
    TI__Guru 52115 points

    See my comments below.

    We know the reading would initially not be very accurate, but as I understood the device is self-learning. So after a few cycles it should be ok.

    [Andy] No.  If a proper chemistry id is not chosen and programmed to the gauge device,  the state of charge reading will NOT be accruate.

    It seems very complicated and time-consuming to program such a golden image on each device in production from BqStudio. What exactly are the values that are in this golden image, could we not program them from our microcontroller which is on the same PCB anyways? Then we would not need additional hardware (like this EV2300) and software on the testbench. BqStudio does not seem to be useful for mass production anyways.

    [Andy] You don't need to use bqStudio. Instead, you can make use of the FlashStream Parser.

    See this app note. 

     http://www.ti.com/lit/an/slua801/slua801.pdf

    Andy

  • 0 in reply to Andy Liu (BMS)
    Prodigy 150 points

    okay. so for initial cell testing and creating a golden image file, there is no other way than using BQstudio and this EV2300?

    How big are these .fs files, can we store them in the flash memory of our microcontroller, such that we can flash from there?

    After flashing the golden image, we still need to configure all the config registers on the BQ34Z100? Or are they included in the golden image? Still, nobody can tell me what this "golden image" actually contains.

    When do I do calibration?

    I really liked the documentation and datasheets for most TI chips so far, but for this BQ34Z100 is seriously is a mess, you can not get any information from anywhere except for asking here in the forum.

  • 0 in reply to Michael Schmid53
    TI__Guru 52115 points

    Hi 

    okay. so for initial cell testing and creating a golden image file, there is no other way than using BQstudio and this EV2300?

    [Andy]  Yes. 

    How big are these .fs files, can we store them in the flash memory of our microcontroller, such that we can flash from there?

    [Andy] See the example df.fs file below.

    https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/196/0100_5F00_0_5F00_16_2D00_bq34z100G1.df.fs

    After flashing the golden image, we still need to configure all the config registers on the BQ34Z100? Or are they included in the golden image? Still, nobody can tell me what this "golden image" actually contains.

    [Andy] No. The purpose of a golen image is for mass production.

    When do I do calibration?

    [Andy] Before you do the learning cycle, you need to finish the calibration first.  If you use BQ34z400 to determine a ChemID, you also need to do calibration first to make sure your measurement results are accurate. 

    I really liked the documentation and datasheets for most TI chips so far, but for this BQ34Z100 is seriously is a mess, you can not get any information from anywhere except for asking here in the forum.

    [Andy]  I would suggest your take a look a the BQ34z100 TRM.  Check Appendix D.

  • 0 in reply to Andy Liu (BMS)
    Prodigy 150 points

    Thanks for that golden image file. I think a 6kb file we can store in the microcontroller flash and then copy it to the BQ device from there.

    I did not fully understand the calibration topic: each PCB is different because of the resistor tolerances and so on. So if we store the calibration values in the golden image and dumbly flash it to each BQ chip, the calibration will be inaccurate in mass production. Should not each device be calibrated individually after production?

    thanks,

    Michael

  • 0 in reply to Michael Schmid53
    TI__Guru 52115 points

    Hi Michael,

    Calibration data can be included as part of this golden file to cut down on battery pack
    production time. If going this route, it is recommended to average the calibration data from a large sample size
    and use these in the golden file. Ideally, it is recommended to calibrate all packs individually as this will lead to
    the highest performance and lowest measurement error in the end application on a per-pack basis.

    Andy