• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 33 subscribers
  • Views 1076 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TMDS181: TMDS181

Oscar Zuniga
Prodigy 120 points
Part Number: TMDS181

Hello, 

I am currently working on implementing the TMDS181RGZT HDMI Retimer to one of our designs and we're coming across some issues. At first, the system was performing well and the video quality at 4K, which was low before the retimer was implemented, was clear. However, after some time, we were not getting any output from the retimer and the system stopped working. We have checked the voltage rails VCC and VDD and both came out at 3.3V and 1.2V, respectively. We are running the retimer through pin strap mode and we believe we have all of the pin straps set up correctly. We did notice one thing, which we want to address to TI:

We are using a voltage supervisor to turn on the power rails of the retimer. Vcc would come up first, Vdd would come up second 10 ms later, and OE would go high 10 ms after Vdd comes up. The datasheet power up requirements have Vdd/Vcc coming up first, followed up by Vcc/Vdd with a max wait of 200 us. OE follows up 100 us later. The ambiguity sort of comes from section 8.3.2 Operation Timing, where it says if OE is held low until VDD and VCC become stable, there is no rail sequence requirement. Does having 10 ms delay between the two rails and the OE cause issues on the chip itself? We have tried running a test with Vdd and Vcc coming up at the same time and then having OE come up, but that didn't work either. Any help would be appreciated. 

Please let me know if you need the pin strap configuration in case that would help you out in any sort of way. 

Thanks

  • 0
    TI__Mastermind 23815 points
    Hi Oscar,

    The timing between Vcc and Vdd doesn;t matter as long as OE is low; after both Vcc and Vdd are stable you just have to wait at least 100us before OE leaves the low state.

    Are you using the EVM?
    Is this a source or sink application?
    What's the pins strap configuration?
    5V_PWR, HDP, and DDC are working fine?
    Have you tried different sources or displays?
    Have you changed the IC?
    Could you share the content of internal registers?
    Do you have scope captures you can share?

    Regards
  • 0 in reply to Moises Garcia
    Prodigy 10 points
    Hi Moises:
    1. We've used the EVM successfully in our Sink application.
    2. We then incorporated the TMDS181 into a small board for a sink application.
    3. Pin straps:
    SWAP/POL - NC
    I2C_EN - PD
    SIG_EN - PU
    PRE_SEL - PD
    EQ_SEL/A0 - NC
    A1 - NC
    TX_TERM_CTL - NC
    We've tried using a couple different monitors.
    The situation is we power up a board with the TMDS181 and the system works fine. After a few hours of testing which includes several on/off cycles and the video is no longer regenerated by the TMDS181. This has happened on two systems now. Once a TMDS181 board stops working it doesn't recover. We are not writing or reading the I2C registers but are using pin strapping instead. We just tested a third board and it initially works just like the first two. We'll share more data later today.
  • 0 in reply to Moises Garcia
    Prodigy 120 points
    Hi Moises,
    Are you using the EVM?

    We did use the Eval board and have had no known issues with it.

    Is this a source or sink application?

    Sink application.

    What's the pins strap configuration?

    SWAP/POL - NC
    I2C_EN/PIN - LOW
    SIG_EN - HIGH
    PRE SEL - LOW
    EQ_SEL - NC
    A1 - NC
    TX_TERM - NC

    5V_PWR, HDP, and DDC are working fine?

    Here's what we are doing with the lines.

    We're using the DDC_I2C lines to control a camera. The DDC_I2C lines go into the sink side of the retimer only. We have the DDC_I2C_SRC pull to ground (just like in your applications on the datasheet)

    HDP bypasses the HDP_SRC on the retimer and goes to the sink side only (same as the sink application).

    5V_PWR is working fine.

    Have you tried different sources or displays?

    Yes.

    Have you changed the IC?

    No.

    Could you share the content of internal registers?

    We're not using I2C, only pin strapping, so this wouldn't apply.

    Do you have scope captures you can share?

    Sorry, no scope captures. What captures would you think would be beneficial for you? We can try and get those for you (pending on our company policies, unfortunately).
  • 0 in reply to Oscar Zuniga
    TI__Guru 60870 points
    Hi Oscar,

    Can you try dropping to a low resolution and see if the system works? I would like to see if the TMDS181 is still operational in redriver mode.

    Regards,
    JMMN
  • 0 in reply to JMMN
    Prodigy 120 points
    HI JMMN,

    No, the TMDS181 is still not operational after we drop the resolution to 720p@30 fps. Does that mean the chip is damaged if the redriver does not work as well?
  • 0 in reply to Oscar Zuniga
    TI__Guru 60870 points
    Hi Oscar,

    If the device was working in redriver mode, it would tell me there was an issue with the retimer circuitry on the RX. In this case, it does appear that the devices are becoming damaged.

    Regards,
    JMMN