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.

BQ24266: Stopped charging and system went down

Part Number: BQ24266

We have seen intermittent failures with the bq24266 not charging the battery, and the system fails.  The input voltage should be utilized when the battery level gets low, but we are seeing the Vout falling lower than the input voltage and the battery voltage dropping also.  For example, last night a-a system went down. The Vinput was stable at +5V, but the battery voltage had dropped to 2.86V and the Vout was also at 2.86V.  The part did not charge the battery and the Vout did not get the necessary voltage boost as expected.  After removing the +5V and reapplying power, the system started to charge again and the Vout was back to its desired voltage.  Any direction is appreciated, we are shipping units and concerned with the random failures we are seeing.

  • Allan,

    Sorry to hear that. Can you send your schematic? As long as VIN is applied, the buck converter should provide output to SYS, limited only by the USBx pin settings. How are the USBx lines being pulled up/down?
  • Jeff,

    Thank you, it is a nice part when it works!  Wish it had pull-ups on the USBX lines, so I tie them High/Low as needed.

    Below is the schematic used for the power system.  Vinput ranges from 9V to 60V, but the external switcher drops it down to +5V to feed the system power.  I have the charging current set to 500mA with normal drain at 300mA except when the cellular radio is transmitting.  Recharge time depends on cellular activity.

    Allan

  • 1057.Schematic Prints.pdfLooks like it did not attach my schematics.

    Allan

  • Allan,

    I do not see any obvious issues with the schematic. My only concern is that input current limit is set to 500mA. If the load at SYS, including battery charge current, exceeds ISYS = 5V*500mA*92%/4.2V = 0.547A then VSYS will droop. Can you capture V(IN), V(SYS), I(SYS) and I(BAT) waveforms on a scope when this event occurs, so maybe trigger on SYS falling?
  • Jeff,
    Is it input current limit or battery charging current?? Also do you know if the IUSB lines are internally pulled high so if I need to make a change can I simply have the high line as open?
    Allan
  • Allan,

    The 500mA input current sets the maximum power available to the system (first priority) then the battery (second priority). If ISYS is higher than expected, then the battery may never charge.

    The IUSB pins are not internally pulled up.
  • Jeff,

    So if I understand this correctly, the maximum current is defined by these lines?  So if nominal current is ~300mA, but I get pulsed current draws of ~800mA, and have the IUSB lines conifgured for 500mA, then the output current will somehow fault?  The battery will charge when the current draw is less than 500mA, giving the battery what's left...correct?  When the current exceeds the IUSB specified amount, what happens?  There does not seem to be a fault line, does it simply turn into a constant current and drop the voltage to maintain the current requirements or does it have the battery provide the boost needed.  What happens when the current returns to a less than 500mA state, does it return to normal?  This still does not explain why the chip was in a under voltage state (~2.8V) when it should have been ~4V, almost like the power path on the main in to out was open.

    So still trying to understand the fault the we have seen once in a while.

    Thank you,

    Allan

  • Allan,

    The IUSBx lines set the maximum input current that the buck converter can draw from its input source, essentially turning the converter into a constant current source as you mention. If the system load, including battery charge current, exceeds the buck converter output as limited by the IUSBx lines, then the battery provides the remaining load to the system. In this state, the system voltage drops to the battery voltage. So if the battery was depleted to 2.8V then the system voltage drops from the minimum system voltage setting to 2.8V. Alternatively, if the load step is very fast and very large, the battery's internal resistance as well as trace and connector resistance could result a large initial droop even if the battery was at a higher voltage. There is no fault as the charger's powerpath architecture was designed to do this. When the load current is reduced, the battery stops supplementing the system load and returns to accepting charge current.
  • Jeff,

    Thank you for the help.  We have designed a board around the BQ25606 that bypasses the 24366 issues.  The new part has two lines, the ILIM and the ICHG.  Just to confirm, the ILIM is the chips power-path  current limiter, and the ICHR is the current limiter for the battery. 

    I have set the VSET=4.208V

    I wanted the battery charged at 500mA but the chip to pass ~1A. 

    I would have the ILIM resistor set to 511 ohms resulting 935mA Power-path limit.

    ILim=KLim/RLim where KLim=478, RLim=511

    Battery charging current:

    IChg=KIChg/RIChg where KIchg=677, RChg=1.35K

    Am I on the correct path?  I rushed the boards to be made because I have to get the system shipped, and so far the testing is good.  Any pointers from the expert is appreciated!  I have four products that we are redesigning the new part into so want to make sure it is correct.

    Thank you,

    Allan

  • Hi Allan:

    Your resistance selection is correct. To make ILIM work, D+/D- pins need to be floating. 

    Regards,

    Hong

  • Hong,

    Thank you, I understand.  I am not using USB for this application, just using the part attached to +5V to support charging and switching the battery into our system.  I have verified the charging is good, now I am battling the system disconnecting both the battery and the input voltage for 100mS, causing our system to reset.  The 100mS event is not a gradual disconnect, but a hard disconnect/reconnect (after 100mS).  Adding a large load capacitor to the output does not help "bridge" the interruption so were perplexed as to what is happening.  The project that is supposed to ship today is on hold...not good.  We switched over to the bq25606 because we would have better control over the charge current/system current and it is working great...except this 100mS event.  I was hoping Jeff from TI would chime in but have not heard from him in the last couple of days.

    Allan

  •  And the design.

    Allan

  • One other tidbit, when running off battery only, the output is smooth and clean, no 100mS event. So it appears to be something internally is causing the 100mS issue. Changing VSET did no have any effect on the length of the 100mS event. When the fault occurs both PG and STAT led's blink 3 times in unison 1 second period, then PG is stable and Stat indicates the battery is fully charged. Any suggestions appreciated.
    Allan
  • Hey Allan,

    From your waveform, is VSYS dropping or this another voltage? Under what condition is that voltage dropping (battery only, vbus applied)?

    I also don't understand why the OTG pin of the charger is connected to the ID pin of the USB port?

    Can you also provide load details on VSYS?


    Regards,
    Joel H
  • Yes, the VSYS. VBus and Battery causes this situation. The USB connector is not used, just shown on the drawing, so it has no effect on the operation. I am closely watching the circuit over the weekend, we made a new batch of boards and will see if it is a manufacturing issue. Thank you for your follow-up.
    Allan
  • Hey Allan,

    So if the USB connector is not used, what is OTG pin tied to? Or is it floating? If it is floating, please do not leave like this and pull to GND if not using the OTG function.


    Regards,
    Joel H