• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 807 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.

Original question:

TUSB8041: TUSB8041 doesn't work with USB3.0

TUSB8041: TUSB8041 USB3.0 does not recognize the device

Louis Chang2
Prodigy 65 points
Part Number: TUSB8041


Hi Sir,

I have a design of hub USB 3.0 with the TUSB8041I.

The host controller (upstream side) is directly provided by the CPU, downstream is 4 port USB3.0 Type-A connector.

The CPU side is replaceable. When the CPU uses the Intel Broadwell-DE platform, USB 3.0 functions normally. When using Intel Skylake-D CPU, it cannot recognize USB3.0 devices but can recognize USB2.0 devices.

I have tested that the upstream USB 3.0 signal can normally identify the USB 3.0 device, while the downstream the eye diagram measurement result is a pass.

Would you do me a favor?

  • 0
    TI__Guru 60870 points

    Hello Louis,

    As a debug test, when the TUSB8041 is connected to the Intel Skylake-D CPU, can you try toggling USB_VBUS to the hub to see if the hub then connects at USB 3.0 speed?  Also, if you are using a Windows based operating system, can you use usbview.exe or USB Device Tree Viewer utilities to see if the hub upstream port is connected to the CPU or not?

    Thanks,

    JMMN

  • 0 in reply to JMMN
    Prodigy 65 points

    Hello JMMN,

    I tried to toggling the USB_VBUS pin of the TUSB8041 briefly, and then plug the device in, but I still ca n’t recognize the USB 3.0 device
    On Skylake-D CPU, Superspeed USB Hub is not detected (No device is connected) on USB Device Tree Viewer, but on Broadwell-DE CPU, it can be detected.

  • 0 in reply to Louis Chang2
    TI__Guru 60870 points

    Can you share a schematic? 

    Regards,

    JMMN

  • 0 in reply to JMMN
    Prodigy 65 points

    Hello JMMN,

    Schematic as attached.

    USB_Hub_TUSB8041I.pdf

  • 0 in reply to Louis Chang2
    TI__Guru 60870 points

    Hello Louis,

    Is the EEPROM used in the design?  If so, can you share the file with me?

    You can send it over direct message, I have sent you a friend request.

    Regards,

    JMMN

  • 0 in reply to JMMN
    Prodigy 65 points

    Hello JMMN,

    I have reserved EEPROM, but I didn’t use it.

  • 0 in reply to Louis Chang2
    TI__Guru 60870 points

    Hi Louis,

    What OS is being used?  Can you send me a screen shot of Device Manager?  We have not seen issues with any of the intel platforms in the past.

    Regards,

    JMMN

  • 0 in reply to JMMN
    Prodigy 65 points

    Hello JMMN,

    I use the Win10 Pro 64bit. 

    Attachment is the result I see in UsbTreeView.

    The picture of Skylake-D_USB is the test result of Upstream, Skylake-D+Hub_USB is the test result of Downstream. Broadwell-DE+Hub_USB is the test result of Downstream after replacing the CPU platform.

    UsbTreeView_Hub.zip

  • 0 in reply to Louis Chang2
    TI__Guru 60870 points

    Hi Louis,

    Thank you for the information, it looks like the Skylake platform is going to compliance mode on the USB 3.0 port where the hub is connected.  This occurs because the host port is detecting the rx terminations of the hub before the hub is ready to communicate with the host, so the LFPS polling fails and the host port enters compliance mode.  There are a couple possible solutions:

    • Increase the GRSTz timing of the hub to keep the host from detecting the rx terminations of the hub, by removing the pull-up on GRSTz (rely on the internal pull-up) and increasing the capacitor value.  
    • Or since he hub will not respond to polling until USB_VBUS is high, adjust the USB_VBUS timing to make sure that USB_VBUS is going high before the host accesses the hub.

    Regards,

    JMMN

  • 0 in reply to JMMN
    Prodigy 65 points

    Hello JMMN,

    I applied for TUSB8041 EVM for testing but found no problems with USB 3.0.

    After confirming the schematic, I found that I only placed the AC coupling on the CPU TX side, but on the RX side did not place AC coupling. After adding AC coupling to the RX side, USB3.0 can work normally.

    As for why the Broadwell CPU is not abnormal, it is probable that the CPU has processed the signal.