• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 209 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.

DS90UB948-Q1: HDMI re-render causeses Black screen at deserialiser output

William Taylor
Prodigy 80 points
Part Number: DS90UB948-Q1
Other Parts Discussed in Thread: TFP401

Hi friends at TI,

                        I have one of our tried and tested designs hooked up to a OEM screen - 1280x640 resolution which has DS90ub948-Q1 de-serialiser in single lane mode thus providing lvds outputs on CLK1, D0-D3 pairs.

My unit has a DS90ub925 serialiser. I feed external hdmi source into good old tfp401 then the RGB/SYNCS/CLK to the 925. Works no problem, nice clean stable picture.

Then, when I select a certain function on the external device supplying the hdmi to my unit, the screen goes to black even though the serialiser still thinks it is passing correct RGB - I can see the syncs, clk, pixel data etc. into the serialiser and its general status register 0x0C seems to indicate happy conditions for the link to the screen.

When I probe the lock pin at the 948 it is HI, but probing the the lvds outputs I see the CLK pair correctly, but I also see what looks like the clock on the D2 pair. The other 3 pairs seem quiet.

While I can't control the external hdmi source(it might be doing something incorrecctly when rendering a new page), is there something I can detect fron the 948 by reading a register to tell me what is happening or has this glitch in the hdmi side caused the 948 to lock up?

Rebooting the screen recovers and shows the picture that should be there.

Rebooting the 925 also recovers and again the correct picture is shown suggesting the 948 is not locked.

My unit sits there happily showing good link status and what it thinks is a valid HDMI source.

Is there a way for me to detect anything useful from and then trigger the 948 to come out of this state, or has the illegal hdmi input caused its outputs to be stuck in this mode.

Thanks and best regards

Billy

  • 0
    TI__Expert 8280 points

    Hi Billy,

    Does a digital reset on the 925 or 948 recover the video?

    William Taylor said:
    When I probe the lock pin at the 948 it is HI, but probing the the lvds outputs I see the CLK pair correctly, but I also see what looks like the clock on the D2 pair. The other 3 pairs seem quiet.

    When are you probing the lock pin? LOCK is acquired when there is active video being sent from the SER to the DES. If there is a momentary stop in video I would expect the LOCK pin to go LOW at some point.

    Best Regards,

    Gil Abarca

  • 0 in reply to Gil Abarca Ramos
    Prodigy 80 points

    Hello Gil, thanks for getting back to me.

    I probed the lock pin at the 948 with the black screen and it shows constant HI level.

    I will try a digital reset but my problem is, in order to do that I need to detect the problem exists. At the moment the general status at the 925 says the link is fine. Is there a flag bit in the 948 I can read to say there is a problem?

    I did do this test.

    I set the 925 to toggle between normal run mode then 925-test signal mode, switching every 2 seconds between the hdmi source then blue test signal.

    While showing the blue screen i am able to select the item that normally makes the screen go black then when it switches back from blue to the hdmi source

    the screen is now displaying what I expect to see. 

    So this concludes that the erroneous instant in the hdmi signal never got passed through to the screen during the blue test test signal.

    I hope that makes sense.

    I do believe the source of the problem comes from the HDMI signal but could it cause the 948 to get stuck in this mode?

    Just a thought too, I see section 8.2.2.2 AV Mute Prevention talks about the UH948 version. Does any of this apply to the UB948 since this section is in both the UB948 datasheetand the UH948 datasheet.

    Thanks again Gil.

    Best Regards

    Billy

  • 0 in reply to William Taylor
    TI__Expert 8280 points

    Hi Billy,

    William Taylor said:
    Just a thought too, I see section 8.2.2.2 AV Mute Prevention talks about the UH948 version. Does any of this apply to the UB948 since this section is in both the UB948 datasheetand the UH948 datasheet.

    Yes, this also applies for the UB948.

    We do not have a register in either the 925 or 948 that might detect a glitch from the HDMI source. I would recommend you to consult the TFP401 product line to check if this device is capable of detecting an incoming glitch from the source. Alternatively, you can consult the HDMI source vendor to help debug the glitch that the source is sending.

    Best Regards,

    Gil Abarca

  • 0 in reply to Gil Abarca Ramos
    Prodigy 80 points

    Hey Gil, just seen your message and can now confirm the HDMI source is doing something during blanking. I have set the RGB DE gate bit, Register 0x04[4] in the 925 and it has resolved the issue. So thanks for your assistance.

    Best Regards

    Billy