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.

TUSB2046B: Windows/Linux operating systems do not detect the HUB. It does not work

Part Number: TUSB2046B


Hello to the whole community. I need your help for my circuit with the TUSB2046BVFR.

In the picture you will see my circuit for a Linux system and also windows. Unfortunately this circuit does not work. It is not recognized by the system.


I also read in the forum the different problems of the other members and apply the solutions and modifications to my system without success.

Does anyone have any ideas so I can fix this urgent problem?

 

Thank you


 

  • Hi,

    Does anything appear in device manager when connected to a Microsoft host? Can you check levels on DP0 / DM0? Also does the crystal start?

    Regards,
    JMMN
  • Hi,
    thanks for your feedback. Nothing appear in device manager when connected to a Microsoft host. The Crystal could start und the Level on DPO/DMO was approximately 3,1v.
    I also pullup the reset pin with 1.5k resistor to 3.3V and 1uF tied to gnd.

    Best Regards,
    Ludo
  • Hi Ludo,

    Only DP0 should be pulled high, DM0 should be low if it is powered on and not enumerated. Can you check that the DP / DM lines did not get swapped in layout?

    Regards,
    JMMN
  • Hi JMMN,

    i checked the System and the DP / DM lines did not get swapped in layout. I pulled up the reset pin with 1.5k resistor to 3.3V and 1uF tied to gnd. Is it ok ? EEGATA/GANGED is floating. should i pull it down or up ?

    Regards,
    Ludo
  • GANGED can be pulled down. Can you confirm that the upstream port has 15K pulldowns installed on DP / DM (or they are incorporated internally)? What is the status on the SUSPEND pin?
  • Hi,

    Today i pulled GANGED down  and checked the oscillator. I only pull down alls the 4 Downstream port with 15K. should i pull the upstream also with  15k ?

    I checked my system Today and windows could detect the device bot Linux. Tomorow i will check the system because the Downstream port doesnt work.

    The device show USB 1.1 on Windows (like in the picture ) is it correct ??? because the other Hub was showing usb 2.0  

  • Hello JIMM,

    the device could work correctly on windows. But i still have this Problem in Linux.

    Do you know how i can fix this problem ?

    Thanks. 

  • The TUSB2046B is a full speed hub (USB 1.1), so it is correct for it to report that in the descriptors.

    Do the downstream ports work in Windows?

    Most USB host controller have internal 15K pulldowns and thus external pulldowns are not needed between the host and the upstream port of the hub, but I just wanted to confirm that this was the case.

    Which Linux kernel are you using?

    Regards,
    JMMN
  • Do the downstream ports work in Windows?

    yes all downstream ports work correctly in windows.

    There is no pulldowns between the host and the upstream port in my circuit. Ther is only a 1.5k pullup. the Voltage on D+ was arrond 0.9v while Pullup. and D- was 0.8v. Is it correct ?


    I m using Linux-imx version 4 based on Mainline with an IMX6 processor.

    Regards,
  • Do the downstream ports power on in Linux? The debug statements are not very helpful. If the host is not active (suspend mode), I would expect DP to be greater than 2.7V volts and DM to be lower than 0.8V.

    Regards,
    JMMN
  • Yes the downstream ports power on. But i still have no communication between the Upstream and the Host.

    DP and DM are not connected to each other. If the module is off I have arrond 3V on DP and about 0V on DM.
    When the module starts I get DP(0.9V) and DM(0.7V).
  • Ok, physically the DP/DM connections seem fine. Can you check to see if there is an update to the Linux Kernel? We don't see the same issue with TUSB2046B on other Linux builds.
  • Hello JMMN,
    many thanks for your support and your contribution. it was very helpfull. Today I got my circuit running with Linux.
    There was more on the host side. The OC line from the host wasn't done correctly on the circuit board so the pin was active all the time (activ low) and the host didn't respond to the hub.

    I bridged that and now the hub is working properly.

    Regards,