• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 1437 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.

BQ27621-G1: Accuracy for small capacity batteries

Oscar Aurin
Intellectual 290 points
Part Number: BQ27621-G1
Other Parts Discussed in Thread: BQSTUDIO

We're currently using a BQ27621-G1 gas gauge for a space constrained, low-powered device. The battery is 45mAh LiPo and average currents are ~5mA. The gas gauge is configured for the low capacity according to the scaling method.

This has worked fairly well in our tests, but we have had troubles getting the SoC up to 100% (it usually stops between 90 and 100%). This isn't much of an issue - we just consider anything >90% as fully charged. However, we are now receiving reports from customers where the SoC stops at 60%, even after charging the battery for more than 6 times the charge time. The battery is more or less brand new. We haven't been able to confirm that battery is at full capacity, but also see no reason why it shouldn't be.

Considering the road ahead, is it suggested to keep using the BQ27621 for this type of application and is there then a way to improve or diagnose the accuracy, or are the other more suitable gauges? (E.g. bq27426 or bq27220)

  • 0
    TI__Guru* 76230 points

    Hi Oscar,

    If it wouldn't be too much of a hassle, i would recommend switching to either of the gauges you mentioned. To  resolve your current  issue, you will have to obtain this failing units and extract a gg file, and perform a full charge, discharge cycle while logging with bqstudio to help us debug what could be wrong. 

    thanks

    Onyx

  • 0 in reply to Onyx Ahiakwo
    Intellectual 290 points

    Hi Onyx,

    Thanks for your reply! I'll see if we can get hold of the unit and extract the file, but it wasn't designed for it, so it may be problematic.

    The reason for switching would be that they are better suited for low-powered applications than bq27621?

  • 0 in reply to Oscar Aurin
    TI__Guru* 76230 points
    Hi Oscar,
    Yes that is correct, they are less of a hassle to deal with than the bq27621 and ofcourse they have low power consumption.
    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Intellectual 290 points

    Follow-up question; when the battery is completely discharged, its PCM will kick in and disable the battery, leaving the battery gauge unpowered, as well as the MCU. When the MCU starts up, it will read the flags register and check the ITPOR flag. If its set, it will perform configuration of the gauge (design capacity, design energy, terminate voltage and taper rate) and then we issue a soft reset. We have BIN connected to ground through a 10k resistor, as the battery can not be removed from the device.

    Is this sufficient, or do we need to do something else, like issuing a BAT_INSERT at start-up? I found no mention of this in the manual - only that "the host must inform the gauge of battery insertion and removal with the BAT_INSERT and BAT_REMOVE subcommands", but since the battery is not removed or inserted - I'm not sure how its supposed to work?

    Thanks,

  • 0 in reply to Oscar Aurin
    TI__Guru* 76230 points
    Hi Oscar

    Since you have the battery embedded, set Op Config [BIE] = 1. that way the gauge always detect a battery is inserted.

    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Intellectual 290 points

    Hi again Onyx,

    Thanks for the reply and sorry to bother you again, but as I review this issue, some more questions arises...

    I read the quick-start guide and the reference manual and found a difference I'm curious about (since we followed the quick-start guide, and I believe the TRM to be correct).

    In section 3 in the QSG; reading the design capacity (as an example) is done by reading register 0x4A. In section 3.1 in the TRM however, the design capacity is read from register 0x43. Is this a typo in the QSG or am I missing something?

  • 0 in reply to Oscar Aurin
    TI__Guru* 76230 points

    Hi Oscar,

    The TRM should be the correct one. Pls check it for confirmation.

    thanks

    Onyx

  • 0 in reply to Onyx Ahiakwo
    Intellectual 290 points

    Once again, thank you for your answer.

    I checked through the QSG to see what effect it had, and a curios/serious side-effect is that it changes "Voltage at Charge Termination". For us, it meant changing voltage at charger termination from the default 4190mV to (an unreachable) 4242mV. For the sake of completeness, I'm including a table of the incorrect addresses in QSG as well as some screenshots.

    We're testing it as we go, but I think the accidental modification of the V at CHG term register was the culprit to never reaching 100% capacity.

    Design parameter QSG address Correct address QSG address modifies
    Design capacity 0x4A 0x43 Terminate voltage (LSB) + Undefined
    Design energy 0x4C 0x45 Undefined
    Terminate voltage 0x50 0x49 Undefined
    Taper rate 0x5B 0x54 V at CHG term (LSB) + AVG I Last Run (MSB)