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 Initialization Conditions - OCV, loaded, charging?

Other Parts Discussed in Thread: BQ27621-G1

Hi all.  We're using a BQ27621 fuel gauge on a single cell lithium-ion battery.  We've found that we get vastly different SOC values from the fuel gauge if it is initialized at no load, charging, and 1C load conditions.  Judging from the forum post here ()  it appears that we should initialize the BQ27621 under no load to get the most accurate readings.  However, from the datasheets, we find quotes like:

"The fuel gauge estimates charge and discharge activity by monitoring the cell voltage. Cell impedance is computed based on estimated current, open-circuit voltage (OCV), and cell voltage under loaded conditions." page 9, bq27621-g1 Data sheet

"The total battery capacity is found by comparing states of charge before and after applying the load with the amount of charge passed. When an application load is applied, the impedance of the cell is measured by comparing the OCV obtained from a predefined function for present SOC with the measured voltage under load. Measurements of OCV and charge integration determine chemical SOC and chemical capacity (Qmax)." page 7, bq276121 Technical Reference manual sluuad4c.pdf

Which is correct?  If we want accurate values, should we initialize the BQ27621 with the battery under load, or with no load?  How much does the accuracy differ in either situation?  Should we never initialize it while the battery is charging?

Thank you very much for your help.

  • To add to the background information, and pose another question on top of the existing question -- in the diagram on page 36 of the Technical Reference manual, the fuel gauge is able to figure out which mode it is in: charging, discharging, or relaxation. Can the BQ27621 automatically scale its SOC estimate based on the mode it finds itself in during initialization?

    Note - I've tried letting the fuel gauge settle for 60 seconds, then 120 seconds, before querying its SOC again, to see if it will adjust the estimate over time. Unfortunately, the estimate remains the same as it was when initialized.
  • Hello Yoo Hsiu Yeh,

    The best way to get the most accurate initialization is to have the gas gauge measure a relaxed voltage. With system side based gas gauging, we have features such as IR compensation on Impedance Track and DVC based gas gauging which will help you with initialization when there is current flowing.

    A note: both impedance track and DVC based gas gauging will converge to 0 at the right point even if you do not have a good initial estimate.

    To answer your question: you will want to initialize the battery SOC estimate under no load conditions ideally or light load conditions.

    When you let the gas gauge settle (I think you are letting the battery voltage settle to a steady value), make sure you also send a soft reset command (0x0042).

    Thanks,
    Kang Kang
  • Thank you Kang Kang. Can I ask you a few more questions?  We have two potential use-cases where the user will probably connect the system-side gauge while the battery is charging or while it is under load.

    Is there any setting I can change to correct the error in the State-of-Charge from initializing during charging or under large load?  I was wondering if setting the User Current Rate or changing the Load Model to follow the average EffectiveCurrent would help fix the error.  Or should I send the soft-reset command more often?

    Also, would it be acceptable to switch to Config_Update mode while the fuel gauge is running?  Does this interfere with fuel gauge simulation?

    Thank you very much for your help!