• State TI Thinks Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 2094 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.

BQ4050EVM-561: Relative State of Charge accuracy problems prevent charging

Donovan Sydow
Intellectual 300 points
Part Number: BQ4050EVM-561
Other Parts Discussed in Thread: BQ4050, GPCCEDV, BQ24773

Hello,

We have been evaluating and using the bq4050 in reference designs, and have noticed over time the Relative State of Charge (RSoC) always becomes inaccurate over time. We found that we couldn't rely on this value from the fuelgauge to actually measure our state of charge and have instead started to use voltage as it is accurate. After further testing in issues related to charging, we have also noticed that the bq4050 uses this RSoC internally to determine when to adjust current to charge the battery. So in our case even though they are calibrated to have a max voltage of 16800 and a capacity of 10800mah, this RSoC will incorrectly configure itself to set itself at a absolute charge state of 110% at lower voltages, effectively crippling the ability to charge the battery to full capacity. So far, we've only been able to bring the previous RSoC back to a calibrated setting after physically unplugging the bq4050 from the battery and ensuring no current flows to it. Considering this requires manual intervention on a sealed device this isn't feasible to do under normal operating conditions. If it isn't completely isolated from power and just unplugged from the battery, the RSoC then goes from 110% to something like 1%, which is still an inaccurate reading. This appears like it is a fundamental firmware problem. My questions are is there a way to reset this RSoC via i2c communication that forces it back to its programmed calibration? Secondly, since the RSoC isn't reliable, is there a way to disable the charge current from relying on this to manage the battery charge level? 

  • 0
    TI__Genius 14985 points
    Hi Donovan,
    What process did you follow to configure the bq4050 for your battery? Did you use the GPCCEDV tool to adjust your CEDV coefficients to match your battery discharge characteristics?
    Regarding the decoupling of charging from the SOC level, you should disable the [SOC_CHARGE] bit in the Charging Configuration dataflash parameter.
  • 0 in reply to dMax
    Intellectual 300 points

    I can't say what method was used to calibrate it, as we received the bq4050 solution with the battery from an engineering firm loaded with the calibration. I would say they probably used your recommended and documented method. 

  • 0 in reply to Donovan Sydow
    Intellectual 300 points
    So I just checked out our configurations and calibrations and we already have the SOC_CHARGE bit in the Charging Configuration dataflash already set to 0. I can tell you in testing and in the field, that they all seem to use the RSoC to handle current and charging even though it is disabled.
  • 0 in reply to dMax
    Intellectual 300 points
    SOC_CHARGE is already disabled, but the bq4050 still uses RSoC to manage charge current.
  • 0 in reply to Donovan Sydow
    TI__Genius 14985 points
    Hi Donovan,
    Are you seeing the requested ChargingVoltage( ) and ChargingCurrent( ) changing depending on SOC, or do you mean that the gauge is turning off the CHG FET and blocking charge current when it reports 110%?
    There are other options which you might need to disable, including the [CHGFET] and [CHGSU] configuration bits in the FET Options dataflash parameter.
    For example, when GaugingStatus() [FC] is set AND FET Option[CHGFET] = 1, the CHG FET turns off.
  • 0 in reply to dMax
    Intellectual 300 points
    Currently we don't have a lot of features implemented in our application that communicates to the bq4050. I can see that we have Temperature(), Voltage(), Current(), AverageCurrent(), RSoC, ASoC. Unfortunately it turns out what I thought was RSoC measurement was actually ASoC. I can measure the Current() is zero, even though the ChargeCurrent() is set to positive value, and MaxChargeVoltage() is set at 16800 via the bq2477x, which is higher than the voltage measured from Voltage() on the bq4050.
  • 0 in reply to dMax
    Intellectual 300 points

    I don't see these as options or set via the calibration data we've saved and flashed on these. PCHG_COMM (Bit 0): Precharge FET selection 1 = CHG FET is the only thing set for Fet Options. 

    Gas Gauging calibration:

    Gas Gauging FC Set Voltage Threshold 4200
    Gas Gauging FC Clear Voltage Threshold 4100
    Gas Gauging FC Set % RSOC Threshold 100
    Gas Gauging FC Clear % RSOC Threshold 95
    Gas Gauging TC Set Voltage Threshold 4200
    Gas Gauging TC Clear Voltage Threshold 4100
    Gas Gauging TC Set % RSOC Threshold 100
    Gas Gauging TC Clear % RSOC Threshold 95

    what does one do, when the RSoC or ASoC thinks the charge level is 100%, when in actuality it is 60-70% or even less. All cell voltages were under 4000. 

  • 0 in reply to Donovan Sydow
    Intellectual 300 points

    So I currently have one in this current state. Battery voltage is 15566, RSoC, and ABSC both show 0, current readings from the bq4050 is 0, even though the bq24773 driving charge is configured for 16800 mv, and output at 896 ma for current. Would love to get a chance to troubleshoot this further if you have any ideas or other things to check, so we're not constantly having to dismantle our designs just to get charging to work.

  • 0 in reply to Donovan Sydow
    TI__Genius 14985 points

    What are the settings for SOC Flag Config A and SOC Flag Config B?  

    The bits in there control the options for settings and clearing the FC and TC bits, which you might have an effect on your FETs.

  • 0 in reply to dMax
    TI__Genius 14985 points
    Hi Donovan,
    Hopefully your issue is closed and we will close this thread. If you have more questions, please feel free to start a new thread.
  • 0 in reply to dMax
    Intellectual 300 points

    Oh I thought I was on ignore, for whatever reason I wasn't receiving updates to this issue until you just posted. I'll check those flags, hopefully I still have this one in the same state and it didn't run the battery down.