• State Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 7 answers
  • Subscribers 64 subscribers
  • Views 627 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.

BQ34Z100-G1: Calibration

Jeffrey Wright
Expert 2875 points
Part Number: BQ34Z100-G1
Other Parts Discussed in Thread: BQ27520-G4, BQ34110, BQ34Z100, BQSTUDIO, BQ78350, BQ34Z110

Hello;

It is my understanding that there is no documentation for calibration of the 34z100. However, an FAE told me that slua640b would be a suitable guide, since the gauge is similar to the BQ27520-G4.

However, there's reference to reading 0x79, 0x7A...etc, which is not listed in the 34z100 TRM or datasheet.

Where can I find information on how to do host calibration current for the 34z100?

  • 0
    TI__Guru** 102955 points

    Hello Jeffrey,

    You can follow the steps in the following thread: e2e.ti.com/.../635852

    I would recommend calibrating a sample set of gauges and upload the averaged calibration data.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Expert 2875 points

    The 34z100 TRM is all but useless. I found that the 34z110's TRM is much better, and have implemented the calibration using the commands listed, which are not listed in the 34z100 TRM. This needs corrected.

    Q: Once asserted, how do I clear the Alert from the gauge?  The 34z110 lists commands for this; e.g. 3.2.45 PIN_ALERT1_RESET: 0x0069 "This MAC subcommand directs the bq34110 device to deassert the ALERT1 pin."  Are these commands also implemented on the 34z100, but not listed?

  • 0 in reply to Jeffrey Wright
    TI__Guru** 102955 points

    Hello Jeffrey,

    Both gauges run very similar functions and have the same base.

    The ALERT will reset when the condition is no longer present for the bq34z100, the alert functions are different for the two gauges.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Expert 2875 points

    For some reason, this page will not allow me to ask a related question. Therefore I'll ask in here.

    Why does the offset calibration take so long? I thought my driver was bad when it takes approx 30 seconds to complete, until I tried it with Studio...which says it could take up to a minute?? For CC gain cal, my driver uses the hidden analog conversion counter (0x79) to average 8 samples which takes 8 seconds.

    So rather than use the built in offset calibration command, I assume I can simply use the same technique as CC gain, and use the analog conversion counter, and then write the offset to Flash. Is there a reason this should not work?

  • 0 in reply to Jeffrey Wright
    TI__Guru** 102955 points

    Hello Jeffrey,

    I'm not sure the exact reason why it takes so long for bqStudio, it's recommended to average multiple cycles so bqStudio is probably averaging a couple calibration cycles. Usually time is not an issue for calibration since the calibration values are usually averaged from 20 or so boards and used for the final golden image.

    You can code your own calibration routine if you want to implement it during production on all boards.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Expert 2875 points

    Will you please confirm whether both board and cc offset calibration are necessary/recommended on the 34z100. I note that the appnote SLUUBF7A for the 34z110 explicitly states: 

    5.6 Board Offset

    "...Calculating the Board Offset also calculates the CC Offset; therefore, it is not necessary to go through
    the CC Offset calibration process if the Board Offset calibration process is implemented...."

  • 0 in reply to Jeffrey Wright
    TI__Guru** 102955 points

    Hello Jeffrey,

    You can calibrate for just the board offset, that should be fine.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Jeffrey Wright
    TI__Mastermind 19195 points

    Hi Jeffrey,

    Check out the evm userguide: 

    Best regards,

  • 0 in reply to Wyatt Keller
    Expert 2875 points

    Hello again. 

    When calibrating CC offset, I'm getting large values, both with my host calibration routine, and with Studio. Pasted below is the CC offset number for the EVM as well, and it's also in the same range as my application board. However, the EVM's Board Offset number is quite low, but my board's is showing 80 to 100 counts.  Is this large CC offset typical? It seems way high.

    What is the difference between Board Offset and CC offset, since both are required to compute the CC gain.????

  • 0 in reply to Jeffrey Wright
    Expert 2875 points

    Hello;

    I've read every document I could find, including that one. It is of little use. I have an EVM and have been using it to confirm my host's configuration driver and calibration. However, since the bq34z100's TRM lacks critical information, as I've noted, I've been using the 34z110's TRM. However, Matt Sunna told me that he's uncertain as to just where these two gauges differ in operation, and exactly just how they're the same. Nonetheless, they do have the same calibration commands (despite them being hidden on the 34z100, not mentioned in the TRM), since I can calibrate cc gain using them.

  • 0 in reply to Jeffrey Wright
    TI__Guru** 102955 points

    Hello Jeffrey,

    The offset can vary depending on your layout and trace lengths, so that wouldn't be too worrisome as long as it is reading accurately. The calibration should be the same for both gauges, I can look into why it wasn't added to the TRM.

    CC offset is to deal with the internal calibration of the Coulomb Counter, the board offset is for the external from traces. In the end they are both an offset value for the gauge to adjust the readings.

    As I mentioned earlier, usually for production the average of multiple completed board calibrations will be used in the final file to program the gauges, so the time it takes for calibrating each gauge is not usually an issue.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Expert 2875 points

    Ok, thanks Wyatt;

    Trouble is, my client's BMS's all implement host calibration at initial board power up, and a 40 second Board Offset calibration time is a big issue. I may have to simply skip this calibration and use an empirical value. I understand no one knows just why this takes so long, but it would be nice know. I'll assume that many of your older gauges are similar, but I do know that the 78350 does not include board calibration, and the CC conversion rate is much faster, making CC gain and offset Cal much quicker... 

  • 0 in reply to Jeffrey Wright
    TI__Mastermind 19195 points

    Hi Jeffrey,

    Try this document: 

    Best regards,

  • 0 in reply to Nick Brylski
    Expert 2875 points

    This gauge is a bit different from the 34z100, and similar to the bq78350, which is used in one of my client's BMS, and for which I wrote the driver. As I said, I was directed to the bq34z110 TRM and am leveraging it since its a much better document. My frustration has been with the 34z100's TRM, which is wholly inadequate. Please suggest to your management that it's in desperate need of updating.