• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 63 subscribers
  • Views 604 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: Learning Cycle fails

Thierry Ligeon
Prodigy 50 points
Part Number: BQ34Z100-G1
Other Parts Discussed in Thread: GPCCHEM, BQ34Z100

Hello,

I have a problem regarding the BQ34z100-G1 Fuel Gauge Device.

I am not able to successfully complete a learning cycle.
The learned status stays 0x04, even when correctly following the SLUA903 document.
I was able to use this device correctly in the past, however that was using a Nickel based chemistry.

Currently I am using a 30Ah 6S Li-Po based battery, but it is advertised as a solid state Li-Ion.
I ran a GPCCHEM cycle, which correctly returned multiple suggested usable chemistries.
However, using both the recommended chemistry (0x0508) as the secondary chemistry (0x2588), I am not able to complete the learning cycle.
The pack is sufficiently rested as shown by the OCVTAKEN, RUP_DIS and VOK, The FC bit gets set, but still the learned status won't increase from 0x04.


I've included a log of a dsg-rel-chg-rel-dsg cycle, which may provide some insight into the issue that I am experiencing.
Also included in the file are the results of the GPCCHEM tool.

Before the log cycle, I did forget to set the scaled bit but this shouldn't effect the operation of the device.
I'm using a scaling factor of 7.0, also applied to the configuration and gas gauging parameters.
This file is also included.

If you are able to provide me with any clue of what i'm doing wrong, it would be perfect!

(Side note, I've tried 5 failed cycles with the 0x0508 chemistry, and 1 with the 0x2588, which didn't help.  Also tried modifying taper current, which either resulted in no FC detection, or same learning status as 0x04)

Thanks in advance!

Thierry


30ah_2588_log.rar

  • 0
    TI__Mastermind 19195 points

    Hi Thierry,

    Can you try searching our forum for "bq34z100-g1 update status 0x04". You may find the solution for your issues.

    Best regard,

  • 0 in reply to Nick Brylski
    Prodigy 50 points

    Hello Mr. Brylski,

    Thank you for your reply!

    I've browsed multiple topics regarding this issue (also about non fuel gauge devices which need a learning cycle), however most of the BQ34Z100 learning cycle related issues are Pb related.
    Suggestions on this specific issue didn't result in a success.
    I've tried adjusting the taper current, changing design energy, different suggested chemistries, regulating the cell temperature but all with no success,

    I will keep on browsing the forum and running cycles to try different solutions.

    Thanks in advance,

    Thierry

  • 0 in reply to Thierry Ligeon
    Prodigy 50 points

    So I tried another learning cycle with the fast convergence algorithm disabled (FConvEN = 0).
    This also didn't result in a learned status update.

    So now I wondered if the DOD > 90% condition was met. According to the logs posted above, it should have.
    Those logs were run with a Cell Termintate voltage of 3300mV.

    After this, I wondered if lowering the terminate voltage to 3000mV would make a difference.
    First, I reflashed a clean image, set the chemistery back to the suggested 0508, updated all the parameters, did both current and voltage calibration again.
    After this, I discharged the pack to 100mv below the termination voltage of 18.0V.
    Then the relax phase started, followed up by a FIRSTDOD = 1 after some time.
    Now i've waited 6 hours, and no OCVTAKEN occured, and RUP_DIS is still = 1 ( So no QMAX update if i'm correct.).

    The pack looks sufficiently rested as far as I can see.
    Is it appropriate to send a IT_Enable command again to force an OCV update? or let the pack relax even longer?

    The logs of the current run are also attached.
    Thanks in advance,
    Thierry30ah3v_0508.rar

  • 0 in reply to Thierry Ligeon
    TI__Mastermind 19195 points

    Hi Thierry,

    For learning cycle the discharge rate should be  between C/5  to C/10. 

    Best regards,

  • 0 in reply to Nick Brylski
    Prodigy 50 points

    Thanks for your reply,

    Does this discharge rate also apply for the initial discharge?

    In the SLUA903 this initial discharge current isn't specified.
    I can understand this discharge rate matters during the RA table updates when moving from learned status 0x05 to 0x06

    Thanks in advance!
    Thierry

  • 0 in reply to Thierry Ligeon
    Prodigy 50 points

    So in the end I managed to achieve a successful learning cycle. 

    I switched back to the original suggested chemistry of 0508.
    Then I tried setting the cell terminate voltage at 3000mV.

    The first run I tried didn't want to get a initial DOD.
    The second attempt (Discharged to 100mV below total cell termination pack voltage), Did get a DOD after about one hour.
    After that, the learning cycle went as expected and completed successfully.

    I took the guideline of never discharging a cell lower then 3300mV too much by heart.
    So for this specific pack, discharging to a cell termination voltage of 3000mV was completely fine, and succeeded the learning cycle.

    Thanks for the help!
    Thierry