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.

BeagleBone Black TPS65217C unexpected behavior

Other Parts Discussed in Thread: TPS65217

I'm getting random reboots in a BeagleBone Black (BBB revision A5B) with the LINUXEZSDK-BONE v06.00 and powered by an external AC/DC adapter (no USB cable connected).

I have two identical systems under test and I got one reboot each 2 to 3 hours but some days both systems work without reboots for 10 to 15 hours.

I have captured a random reboot with the oscilloscope (see attached image).

    Ch1 → VDD_3V3B (P9.4)
    Ch2 → VDD_5V (P9.6)
    Ch3 → SYS_5V (P9.8)
    Ch4 → SYS_RESETn (P9.10)

The random reboot is produced by 1 second SYS_5V fall therefore seems like the TPS65217C has detected a fault.

Monitoring the TPS65217C registers I have found an unexpected behavior:

watch -n 0 i2cdump -y -f 1 0x24

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: e2 3f 01 01 b1 80 b2 01 00 00 88 00 7f 0c 18 11    ????????..?.????
10: 05 06 09 38 26 3f 7f 00 03 15 5f 32 40 20 00 00    ???8&??.??_2@ ..

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: e2 3f 01 01 b1 80 b2 01 00 00 8c 00 7f 0c 18 11    ????????..?.????
10: 05 06 09 38 26 3f 7f 00 03 15 5f 32 40 20 00 00    ???8&??.??_2@ ..

The USBI flag in the INT register is toggling indicating USB power status change and the USBPWR flag in the STATUS register in toggling indicating USB power present/not present. Of course, as mentioned bellow the BBB in powered only by an AC/DC adapter (no USB cable connected), therefore the USBPWR flag in the STATUS register should be always 0.

As USB detection comes from VBAT comparison I guess that VBAT can be the problem. In the BBB the BAT, BAT_SENSE and TS pins are left independently floating. As in the figure 18 (power connection for battery-less 5V only operation) of TPS65217C datasheet both BAT and BAT_SENSE pins and left floating but connected together I have tested this configuration in the BBB but incorrect USBPWR flag behavior is still present. I have also tried to connect USB power input to GND as in figure 18 but same behavior.

Other test I have do are:

- Disable USB power
    i2cset -y -f 1 0x24 0x01 0x2f  → USB_EN=0 at PPATH register.

- Disable battery charger
    I2cset -y -f 1 0x24 0x04 0xb0 →  CHG_EN=0 at CHGCONFIG1 register.

In both cases the incorrect USBPWR flag behavior is still present

Can your suggest me any workaround?

Do you believe that the unexpected behavior of USB detection can be the cause of random reboots?

Best regards.

  • Hi Antonio,
     
    A similar issue was reported sometime ago. I will forward this to the factory team to check if there is an explanation already.
  • I can confirm the TPS65217C of my BBB has the same behavior.

    I patched the tps65217 driver with irq handling in the TI BBB android kernel. The PMIC_INT was issued every 2 seconds which is caused by the USBI flag in the TPS65217C interrupt register, although there is no USB connected.

    [  217.095367] tps65217_irq: USB power status change
    [  217.259338] tps65217_irq: USB power status change
    [  219.096801] tps65217_irq: USB power status change
    [  219.256103] tps65217_irq: USB power status change
    [  221.094177] tps65217_irq: USB power status change
    [  221.262084] tps65217_irq: USB power status change
    [  223.095611] tps65217_irq: USB power status change
    [  223.259033] tps65217_irq: USB power status change
    [  225.097045] tps65217_irq: USB power status change
    [  225.255859] tps65217_irq: USB power status change
    [  227.094024] tps65217_irq: USB power status change
    [  227.262908] tps65217_irq: USB power status change
    [  229.095489] tps65217_irq: USB power status change
    I paste part of the log above. The actual timing of USBI status change is at a inteval of 0.16, 1.84 seconds interval. I wonder if this is contributing to the random reboot of BBB running TI android image.
  • Guys,
     
    Thanks a lot for your input. The factory team is investigating this issue and your information really helps. We will update this thread once a solution is reached.
  • Antonio Cebrian said:

    I have captured a random reboot with the oscilloscope (see attached image).

        Ch1 → VDD_3V3B (P9.4)
        Ch2 → VDD_5V (P9.6)
        Ch3 → SYS_5V (P9.8)
        Ch4 → SYS_RESETn (P9.10)

    The random reboot is produced by 1 second SYS_5V fall therefore seems like the TPS65217C has detected a fault.

     

    This appears to be related to USB OTG probing that occurs with the Linux 3.2 kernel. You may want to monitor USB0_VBUS to see what happens during a reset.

  • Biser Gatchev-XID said:
    Guys,
     
    Thanks a lot for your input. The factory team is investigating this issue and your information really helps. We will update this thread once a solution is reached.

    Hi  Biser,

       My customer also faced this issue when they are in PP stage, so could you please share the final solution for this case ASAP?

        Thanks!

        Yaoming

  • Hi Yaoming,

    The root cause is the USB OTG driver pulsing the VBUS supply.

  • Hi Sir :

    I have the same issue on my board.
    Could you provide the solution to this issue?
    Now, I disable USB driver support to test this issue.
    Should I modify OTG driver or modify my hardware circuit?

    Best regards,

    Marcus
  • Yuenjune,

    Take a look at the post on the Google Beagleboard group:

    https://groups.google.com/d/msg/beagleboard/xPxzYyNsA78/XeyX43MODkUJ

  • Hi Lei :

    Thanks for your reply!

    I will test your solution.

    Best regards,

    Marcus