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.

TUSB211: TUSB21x not working as desired - Is this a valid application scenario?

Part Number: TUSB211
Other Parts Discussed in Thread: TUSB216, TUSB212

C1: Single Board Computer with USB OTG

C2: 1m Micro-USB to USB-A Cable

C3: PCB with TUSB21x on it

C4: 5m USB-B to USB-A Cable

C3: PC or Laptop

In our application scenario (as seen above) we do not see any improvement when using a PCB with TUSB21x (C3) instead of just using a USB-A to USB-B coupler.

In fact, results are slightly worse compared to a plain coupler. Our goal is to get 5m USB Cables (C4) working between a PC (C5) and our housed device but the maximum working cable length for C4 is 2-3m.

We started off with a TUSB211 then changed to TUSB212 as we thought it could be DC offset issues (which wasn’t the case) and finally we changed to TUSB216 because we wanted to try it the additional RX sensitivity possibility. For all TUSB21x we have systematically tried the different settings (if supported by the respective IC) on AC boost, DC boost and RX sensitivity without success.

For some reason we can’t see why the signal quality does not seem to get better in any case.

We could provide schematics and PCB layout via a non-public way…

Unfortunately, we do not have eye-diagram measuring equipment to provide further details (which makes it hard to judge I guess...)

Questions are:

  1. Is the TUSB21x (TUSB211, TUSB212, TUSB216) intended for the setup described?
  2. As we are running out of ideas – does TI have some ideas which we may not have thought of? Wink

Any help would be much appreciated! Thanks in advance

  • Hi,

    The scenario description is a valid setup for these devices. From your description is sound as if the redrivers are not entering the correct modes during testing.

    • Are you doing eye diagram testing? If so can you share the eye diagrams using TUSB212 when RST is high vs low?
    • What test procedure are you following to capture the eye diagram of your system?

    I would also check the state of the CD and EN_HS pins (if possible) to get an idea on the state of the device during testing. Feel free to direct message me your schematic through E2E and I review. Lets stick to 1m testing for now so that we can ensure that the eye is being properly compensated.   

  • This issue could be resolved by E2E direct help (big thanks @Malik Barton57) and some further investigations and tests with the setup. The findings are now presented here for documentation.


    There were 2 main problems in our setup:


    First, we have used a very bad quality “1m Micro-USB to USB-A Cable” (C2) for most of the tests we did. Its shield was not connected from one plug to the other. Replacing that one with a quality cable made a first huge step closer to the goal.


    Second, we have found out (with the good cable) that in our setup a TUSB21x redriver shouldn’t be used at all. Even the TUSB21x lowest boost settings caused over-compensation of the USB signals in our setup as the signals WERE GOOD ALREADY. This could be the case if one has a strong USB PHY driver (like we had).
    We didn’t have measuring equipment, so we had to find this out the hard way. ;) Should be more obvious if one has an appropriate scope to see the signals.


    With a good cable (C2) and omitting the TUSB21x we achieved the goal to get 5m USB cables (C4) working.

    CONCLUSION (TL;DR):
    - Always check quality of used cables before checking everything else (like schematic, PCB, etc) for the hundredth time Wink
    - Depending on the application scenario it could be that a redriver isn’t needed at all and could possibly over-compensate (worsen) an “already good USB signal” (which was the case here).