• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 100 subscribers
  • Views 571 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.

DM6446 USB VBUS_ERR during working

Richard12023
Intellectual 300 points

Hi,

    I got a stability problem: we connect a selfpowered usb device to dm6446 usb host. After the usb connection is established, we transfer heavy data under usb high speed mode. But after some hours(10hours or maybe just 2 hours),  message "VBUS_ERR\n\n" is printed in console and our software detect the usb connnection is broken.

    VBUS_ERR is printed in musb/plat_uds.c:static irqreturn_t musb_stage0_irq(struct musb * pThis, u8 bIntrUSB, u8 devctl, u8 power).  

             if (bIntrUSB & MGC_M_INTR_VBUSERROR) { printk("VBUS_ERR\n\n"); .............}

 

My questions:

  • we need enlarge the usb power supply?  maybe it's a power lost problem under heavy usb transit load.
  • which one to enlarge? self-powered usb device or davinci usb host?
  • 0
    TI__Mastermind 41665 points

    Richard said:
    we need enlarge the usb power supply?  maybe it's a power lost problem under heavy usb transit load.

    A VBUS error would normally indicate that the VBUS voltage dropped below threshold, so a loading problem would be the typical reason, though if you have custom hardware than there may be some other reason why VBUS was sensed to be too low. The situation is particularly strange if your USB device is truly self powered and not drawing on the VBUS rail, as at least on the EVM the VBUS should be very stable if nothing is drawing on it. Before going to actually enlarging any supplies you may want to verify that the VBUS voltage is stable with a multimeter or oscilloscope, particularly if you see any glitches that may be putting it under the threshold.

    Richard said:
    which one to enlarge? self-powered usb device or davinci usb host?

    If the DM6446 is throwing the VBUS error it would mean the VBUS pin was sensed to be too low typically, so most likely the DM6446 side of the supply, though as suggested above you may want to verify the voltages you currently have and analyze the power supplies with a scope first.

  • 0 in reply to Bernie Thompson TI
    Intellectual 300 points

    I found there some usb bug fixes at http://software-dl.ti.com/dsps/dsps_registered_sw/sdo_sb/targetcontent/psp/mv_lsp_1_10/index.html

    Does there any improvement about VBUS_ERR?

     

    Can self powered usb device draw/effect the power stablity through the usb cable? If the self powered usb device has a sudden power consuming, the power supply for usb device hardware unit is drown, in such a case, the self-powered power is low, so usb cable voltage dropping/disturbing case the VBUS_ERR.  Is it possible?

  • 0 in reply to Richard12023
    TI__Mastermind 41665 points

    Richard said:

    I found there some usb bug fixes at http://software-dl.ti.com/dsps/dsps_registered_sw/sdo_sb/targetcontent/psp/mv_lsp_1_10/index.html

    Does there any improvement about VBUS_ERR?

    Not very likely, these patches were primarily to add capability enhancements, and as far as I know they did not really deal with VBUS. Note that those patches are very old, unless you are running on a software stack from a couple of years back these patches will not be of much help, the latest DVSDK 2.0 as of today with current PSP (including the USB drivers) can be found here.

    Richard said:
    Can self powered usb device draw/effect the power stablity through the usb cable? If the self powered usb device has a sudden power consuming, the power supply for usb device hardware unit is drown, in such a case, the self-powered power is low, so usb cable voltage dropping/disturbing case the VBUS_ERR.  Is it possible?

    This depends on your 'self powered' USB device and what the manufacturer meant when they called it 'self powered'. If the USB device is using the 5v VBUS rail for something than there is the potential that it could impact the VBUS voltage, on the other hand if it is truly self powered and leaves the 5v VBUS line essentially disconnected on its end, using its own power supply, than it should not be impacting it.

    What sort of self powered USB device do you have connected? If you believe it to be pulling down the 5v VBUS line under some conditions even though it should be self powered than you may want to contact the manufacturer.

  • 0 in reply to Bernie Thompson TI
    Intellectual 300 points

    Our package is dvevm_1_20, mv_pro_4.0(2.6.10_mvl401-davinci_evm).  I prefer to keep on the current SDK unless there is usb bug fixes for me, you known I don't want to test many modules again.

    Our usb device is a SoC, it does many heavy jobs , and I'm sure it's self powered.  After checking layout and some analysis, we think the power supply is enough, maybe the usb cable is not good connected.

     

    Any suggestion?

    Thank you!

  • 0 in reply to Richard12023
    TI__Expert 7435 points

    Richard,

    Did you happen to apply all the related patche for LSP1.2 kernel that you are having?  If not can you apply the same.

    regards

    swami

  • 0 in reply to swami
    Prodigy 150 points

    Hi Richard,

    We are facing a similar problem. Did you find the solution?

    Regards,

    Sumit