• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 4 answers
  • Subscribers 63 subscribers
  • Views 432 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.

BQ24172: Battery removed during charging does not cause STAT pin to start blinking again.

Jackson Wilson
Intellectual 330 points
Part Number: BQ24172

Our application has a removable battery. We have 0 ohm connected to TTC (disable safety timer and charge termination -- our MCU is controlling charge termination).

The steps are:

  1. Power up BQ24172 with no battery connected.
  2. STAT is blinking.
  3. Install depleted battery.
  4. STAT goes low (charging).
  5. Remove battery.
  6. STAT stays low.  

This does not seem to match the documented function for STAT pin (DS pg. 5 -- STAT pin: "Blinking at 0.5 Hz when fault occurs,
including charge suspend, input overvoltage, timer fault and battery absent.")

The only way to get the STAT pin to blink again seems to be to remove power from the BQ24172 or toggle ISET low/release.

Can you confirm this behavior?

Thanks,

Jackson

  • 0
    TI__Guru**** 166626 points

    Hi Jackson,

    You have disabled termination (and battery detection) with TTC=GND.   Even though you have removed the battery, the charger has no way of knowing that and continues to charge the output capacitor up to VBATREG, stops outputting charge current but doesn't terminate, so STAT stays low.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Intellectual 330 points

    Jeff:

    Thanks for the reply.

    If TTC had a 0.022uF cap to ground (for example) what would be the expected behavior of STAT pin if the battery were removed while charging (STAT low)?  Would STAT return to toggling right away?

    Thanks,

    Jackson

  • 0 in reply to Jackson Wilson
    TI__Guru**** 166626 points

    Jackson,

    Yes. Termination would be enabled and STAT would then blink as the charger entered battery detection.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Intellectual 330 points

    Jeff:

    I have just completed a test with TTC capacitor = 0.01uF, using our production battery, and I do not see the behavior you describe. 

    The steps are:

    1. Power up BQ24172 with no battery connected (V+, thermistor, ground all disconnected).
    2. Observe 0.5Hz toggling STAT pin.
    3. Connect partially discharged battery to charger (V+, thermistor, and ground all connected).
    4. Observe STAT pin go low, and see charging current as expected (1A, in our case).
    5. Disconnect only battery V+ pin (thermistor and ground still connected).
    6. No change to STAT pin (stays low).

    Can you please suggest next steps to troubleshoot this?

    Note: We are using the BQ24172 to charge a 7.2V NiMH pack, as discussed in previous posts (however, I have also tested this with a Varta Li-Ion and seen the same unexpected behavior):

    https://e2e.ti.com/support/power-management/f/196/t/883392?BQ24172-NiMH-charging-solutions

    https://e2e.ti.com/support/power-management/f/196/t/922479

    https://e2e.ti.com/support/power-management/f/196/t/915362

    Thanks!

    --Jackson

  • 0 in reply to Jackson Wilson
    Intellectual 330 points

    Jeff:

    A little further detail:  with a 0.01uF TTC cap, STAT does begin to toggle about 1 hour after removed battery V+ from charging circuit.  At 5.6min./nF, that is what would be expected for the safety timer expiration.

    So it seems that battery-removed detection is not working while the BQ24172 is in charging mode and the TTC safety timer is not expired.

    --Jackson

  • 0 in reply to Jackson Wilson
    TI__Guru**** 166626 points

    Hi Jackson,

    I am currently testing a BQ24172EVM on the bench and if I remove that battery, the battery output capacitor charges up, then terminates, then the output capacitor discharges significantly, then recharges, just like as shown in d/s section 9.5.13, which causes the STAT pin to blink at that rate, not necessarily 1Hz.  The frequency will depend on the output capacitance on the output.  This only happens if the TS pin is fooled into thinking there is a thermistor.   If there is a significant amount of capacitance at the output, the battery detection algorithm can be defeated.

    If there is no thermistor then the output fails close to zero and STAT blinks.

    Regards,

    Jeff 

  • 0 in reply to Jeff F
    Intellectual 330 points

    Jeff:  

    I think I found the issue.  Our voltage divider from VBAT to FB is 1M/294K, to get a high enough regulation voltage to ensure that we always charge our NiMH pack in CC mode.

    But, we also have a 100K/0.1uF low-pass RC filter from FB to an MCU pin, so we can monitor battery voltage.  This RC seems to have been interacting with battery detect timing (Figure 16), preventing it from generating the battery absent (1.45 Vlow) signal.  Removing the 100K between FB and 0.1uF/MCU pin has resulted in expected battery-removed behavior during charging (STAT toggles).  We will need to revise our RC low-pass.

    I am running some charge/discharge cycles to confirm that this change doesn't have any other side-effects, but it looks good so far -- the circuit is charging, and immediately detects removing the battery positive connection.

    Hopefully this makes sense and seems to you like a plausible explanation for what I am seeing.

    Thanks,

    Jackson

  • 0 in reply to Jackson Wilson
    TI__Guru**** 166626 points

    Hi Jackson,

    Yes, an RC delay could be causing the issue.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Intellectual 330 points

    Thanks Jeff, I think we know what the issue was and how to address it now.

    --Jackson