• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 62 subscribers
  • Views 402 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.

bq27200 calibration

David Durgin
Prodigy 70 points
Other Parts Discussed in Thread: BQ27200, BQ27210

I'm using a bq27200 gauge in a system with a typical load of 500mA; I've selected EDV1 for the C/16 point as per the programming instructions. On occasion the load can be as high as 1.5A, when this occurs EDV1 is reached well before C/16. When this happens during a qualified learning discharge, the gauge re-calibrates to a falsely low C.

Has this issue been encountered by other users? Is there a way to prevent this erroneous calibration from occurring?

Any help/advice will be appreciated.

 

  • 0
    TI__Intellectual 2340 points

    Can you indicate by how much is the capacity being learned lower? What is the nominal capacity for the battery being used?

  • 0 in reply to MikeV
    Prodigy 70 points

    The low C re-learn gets worse with age; in an example I recently witnessed in the lab, the reported SoH dropped from 25% to 6% in one update (when the EDV1 threshold was crossed). This was with a 2+ year old battery at ~1.25A load. In this case the LMD register was reduced by the -12.5% limit.

    The nominal (new) C for this pack is 2400mAh (Moli ICR18650J).

  • 0 in reply to David Durgin
    TI__Intellectual 2340 points

    David,

    On this given battery that is over 2 years old, has the bq27200 been learning the capacity of it through out the 2 years of operation or are you just taking a bq27200 that has your default configuration for a new battery and making it run for the first time with the battery?

    I'm trying to understand if the SOC drop from 25% to 6% is because you are trying to learn capacity of an aged battery but with a starting point based on a new battery or if it is just a matter the increased impedance of the battery due to aging having to strong of an impact.

    The bq27200 does compensate capacity based in varying discharge rates but does not take in account the increase of impedance as it ages. The bq27210 (which is a more updated device from the bq27200) actually has the capability to compensate based on aging.

  • 0 in reply to MikeV
    Prodigy 70 points

    Mike,

    I'm seeing two causes of under-learning C.

    One: EDV1, in the EEPROM, is set to 3.4V based on a calibration/learning cycle of ~660mA (where c/16 is reached at 3.4V); if accessories are connected to this system the load may be 1.5A; as far as I can tell, when EDV1 is reached (at any load) the 27200 simply sets the SOC to 6% on the fly and then places the coulomb count + 6.25% into the LMD register. It gets worse with age & increased Z(i), but this phenomenon causes a significant under estimate of C even with a relatively new battery.

    Two: As the battery ages, the fixed EDV1 occurs earlier and earlier on the discharge curve, again causing an under estimate to be logged into LMD, even at the nominal load of about 660mA.

    Thanks for your help so far; any recommendations will be appreciated...

    DD