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.

BQ24133: Enabling charging of battery

Part Number: BQ24133

I’m having a difficult time diagnosing a test unit that doesn’t charge the battery.  Measuring with a current probe at battery terminal, no current going into the battery.  I received this unit from my Verification team.  They relayed to me that the pump did charge the battery at one point, but does not anymore.  He said it had about 50 charge/discharge cycles on the battery (not completely discharged, only to about 60% discharge).

 I did try different batteries, same phenomenon.

 The STAT pin is blinking, which means a couple states

   

Charge suspend: it looks like this is only tied to the temperature qualification of the battery.  Battery is within temperature ranges (per below)

Input overvoltage: Input is 12V.  Well within the adapter voltage (per below)

Battery overvoltage: Battery is 6.8V.  Not an overvoltage condition.  CELL = FLOAT.  Measured at 1.091V

Timer fault:  Do not believe so.  Happens immediately.

(two commas, something deleted?): what was deleted from datasheet?

Battery absent:  battery is definitely connected.

Per section 9.3.9 of the datasheet, I reviewed the status of all of the pins.

 Any ideas?  This is weird.  I don’t want to remove or reflow any components until I can absolutely determine there was an issue with the part.

Thanks,

Daniel

  • Hi Daniel,

    There is also a short circuit protection that will latch off the device if it exceeds our MOSFETs current limits (about 6A). See if replugging the adapter makes the fault go away (section 9.3.19).

    Regards,
    Steven
  • Hi Steven,

    Thanks for the quick reply.  Plugging and unplugging the adapter does not change the state.

    When I look at the current measurement on the input of my adapter, I do see some inrush events exceeding 6A.  They are brief (10 us and 400us).  However, these inrush events occur when the charger is disabled (ISET = 0).  Sometime after 400 us inrush event, the charger is enabled (ISET = 440mV).  After enabling the charger, I do not see the charger transition into the expected charging state (per above conditions, they all seem to be met).

  • Steven,

    I took a few more measurements for thoroughness.  The inrush events occur at the same time with the good working unit.

    I did notice that when I unplugged the battery, the working unit started behaving like the non-working unit.  (see below)

    Plugging the battery back into the working unit:

    However. the non-working unit most definitely has a battery connected to it.

    I reviewed the battery detection flowchart, and everything looks good (I think), except for VBATOWV.  I cannot find any reference to VBATOWV in the datasheet, except in section 9.3.15.

  • Hi Daniel,

    It would be useful to see AVCC, PVCC, SRN and VREF voltage. These can also show if any fault, problem with the battery or LDO noise are occurring.

    As for the VBATLOWV threshold, on the electrical characteristics table it is name as VLOWV (page 8)

    Regards,
    Steven
  • Steven,

    Thanks for the response.  I was not sure about the VBATLOWV because the state table has it listed as VBATOWV.  Clearly a typo, thanks for clearing that up.

    I will get scope shots of the 4 signals.

    Daniel

  • Hi Steven,

     I acquired the waveforms requested on a good unit and the DUT unit for comparison. I did not see anything out of the ordinary. There is a small droop on both the good unit and DUT on PVCC between the input voltage and battery, lasting for about 3ms. Since this was on both units, I’m not sure if it is the cause of the DUT not initiating charge.

    I also grabbed a waveform of the voltages when the battery is not in the good unit. The droop on PVCC is present on good unit without battery.

  • Hi Daniel,

    I would like to see your schematic. I'm thinking that the input current inrush has to be controlled, either through a load switch or through our ACFET turn on. This inrush may be damaging something in the system.

    Also, have you tried to swap the ICs between the working board and non working board? Does the problem follow the IC or the board?

    Regards,
    Steven
  • Hi Steven,

    Attached is a snip of the schematic for the BQ24133.  I can email you the full schematic, if you provide your contact information.  I've also emailed it to my local contact, who may forward it to you as well.

    One thing to note, the TS pin is connected to a 3.3V LDO.  This is so we can monitor temperature during discharge as well (since VREF goes away when operating on battery).

    I do have a capacitor to control the ACFET signal and dampen inrush.  How would this manifest itself with not charging the battery, but all other functions of the device work satisfactorily?

    I did notice that the datasheet recommends 500k or higher between ACDRV and CMSRC, but the eval board uses 499k, so I did as well.

    I have *not* tried to swap the component to another board.  It's a fairly difficult component to remove without upsetting other components in the area.  I wanted to have the swap be the last thing that I would try to isolate the part.

    Thanks,

    Daniel

  • Daniel,

    I don't see anything wrong with this schematic.

    Can you disconnect the system from the charger? Also, connect TS to our VREF through R88, remove R1. See if only the charger has the issue, or something in the system is causing it.

    I think swapping ICs will give us more clues whether the IC is damaged or the board is damaged. Measuring continuity on the pins will also provide data on whether ESD damage was involved.

    Regards,
    Steven
  • Steven,

    I moved the NTC reference from the LDO to VREF.  It did not change the behavior of the circuit (still not charging, STAT toggling).

    I removed the battery charger IC and replaced it with a new one.  The circuit behaved as expected.

    Can you provide some guidance as to what resistance measurements I should perform on the IC?  AVCC to gnd? PVCC to GND?

    I also want to get a working unit, remove it's charger IC, and replace it with the faulty IC to see if it follows the IC.  However, I do not want to do that if you would like to obtain the unit to do further testing on it.  Please let me know.

    Thanks,

    Daniel

  • Steven

    Below is a table of a few measurements.  It doesn't immediately look like an ESD issue.

    Your guidance is appreciated.

    Daniel

  • Hi Steven,

    To confirm it is IC centric, the new bq24133 IC was removed from the test unit and the DUT IC placed back on.  The behavior returned (no charging, stat blinking).

    What are the next possible steps here to check the IC?

    Thanks,

    Daniel

  • Hi Daniel,

    These continuity measurements are useful.

    Verify the continuity between the following:
    1. PVCC and PGND
    2. PVCC and SW
    3. BTST and SW
    4. BTST and REGN
    5. SW and PGND

    If any high current or high voltage situation may have occurred in the power path, these may show some damage.

    Regards,
    Steven
  • Hi Steven,

    Attached are measurements requested.  I don't see anything out of the ordinary.

  • Hi Daniel,

    This is definitely puzzling.. How many units failed from how many in total?

    Regards,
    Steven
  • Steven,

    We had 27 charge/discharge cycles on this DUT with the failed IC. on the 28th charge cycle, it would not charge the battery anymore.

    I had a lot of about 30 units with multiple charge/discharge cycles, combined run hours of over 2500+. But this one unit had a failure. I cannot figure out why. It's definitely IC related, since replacing the IC fixed the issue.
  • Daniel,

    Can you provide the sequence on how you do the charge/discharge cycles? How much discharge current is used and at what voltage you allow the battery to discharge? How does the SRN and SW waveform look like on the issue unit w/ no battery connected?

    It might be that it is not even initializing the battery detection routine, which means that maybe some other pin is not at the right voltage to initiate charging, or that the IC got damaged internally at some point.

    This device has been on the market for 8 years, this is the first time I've heard of a failure like this.

    Regards,
    Steven
  • Hi Steven,

    For the initial testing we were running, we would charge the battery up to completion, and then run the device for about 45 minutes.  This timeframe would bring the battery down to about 50% state of charge.  Then we would plug the device back in and charge it until completion, and repeat the cycles.

    The battery did not get down to the under voltage threshold on the batteries PCM.  The discharge current is about 0.9-1.2 Arms.

    I will obtain a waveform of SRN and SW with no battery.

    Daniel

  • Hi Steven,

    Below are the requested scope traces:

    I took waveforms of a good unit first (needed to wait while technician re-attached bad IC to the DUT).

    Next I took the same setup (original DUT with bad IC and captured without and with a battery):

    I did notice that there was a very brief time (about 600us) where the SW pin was active.  I zoomed in on that below:

    Then I noticed that this activity repeated approximately every 1.5s

    This coincides with the 1.5-s delay is complete after initial power up.  Something is causing the device to continually try to complete it's power on sequence.  However, the anomalous behavior does not appear when we replace the IC with a new one. 

    What in the IC could be causing the power on sequence restarts if the voltage levels outside the device are valid?

    Thanks!

    Daniel

  • Hi Daniel,

    The 1.5s delay is actually the battery detection routine running, but it seems that it cannot detect the battery properly, maybe SRN pin damage.

    There is nothing notable on the schematic or IC behavior that can help root cause this as is, or at least is not as obvious. All other pins have the correct voltage to start charging and there is no continuity problems on the IC pins. This shows that there is something damaged internally in the logic circuit pertaining to the battery detection routine. The problem follows the bad IC, so that rules out the board and schematics. If you would like, you can submit an FA request through our official process. Please contact the field team to submit the request.

    Regards,
    Steven