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.

DS90UB949A-Q1: No or sporadic lock to deserializer DS90UB948-Q1

Part Number: DS90UB949A-Q1
Other Parts Discussed in Thread: TMUXHS4212,

Greetings,

we are encountering a problem with our display driving solution.

We are using the DS90UB949A-Q1 to convert HDMI to FPD-Link III which is then sent to a remote Deserializer connected to a Display. In order to be able to switch between multiple video sources we are using the TMUXHS4212 as shown in the picture below:

Our boards were all tested when they first arrived and worked without any problems. After sitting on the shelve for some time only one of the boards works as intended by being able to still reach the locked status with the remote Deserializer and transmitting video data.

Some other boards will work after repeatedly engaging the "Digital RESET0" of the DS90UB949A-Q1 chip. But even though the picture is being transmitted flawlessly register 0x0C[0] will show that there is no link. Also register 0x5A[7]  will show no FPD_LINK_RDY status but 0x5A[6] shows that FPD-Link III is being transmitted.

On the rest of the boards we are sometimes able to reach the state as described above but no picture will be displayed.

The boards were tested with different cables and different displays but to no avail. The hardware design follows the EVM example when it comes to passives like CMC and caps.

My question to you is if you have any idea what could cause this sporadic behaviour, especially after the fact that the boards were initially working fine and there were no changes made to them. Could the mux cause problems in terms of signal integrity even tough we payed careful attention to the correct design of the transmission lines?

If you have any approaches to find the cause of this error i would like to hear from you.

  • Hi Ulrich,

    To eliminate the switching source of the mux, did you try to connect directly from 949 to 948? Which type of CMC are you using? Murrata? I noticed that the CMC is very sensitive. Can you try to probe across the pin if there is any short by probing pin 1-2, 3-4, and 1-4?

    Aaron

  • Hi Aaron,

    thank you for your fast input.

    The issues seems to lie with the MUX. After removing it and bridging the pins on one of the boards that wasn't working anymore, we could immediately get a connection between the 949 and the 948 with a clear picture on the display.

    Do you have any experience with the TMUXHS4212 in FPD-Link III switching applications, or would it be better if I closed this thread and opened a new thread specific to that device?

    Ulrich

  • Hi Ulrich,

    I'm glad that we were able to find the root cause. I haven't use this device before, but I can look into this. Regarding the ticket, whichever you are comfortable with, if you think resolve your issue, you can mark resolve to close this ticket. Or you can open a new thread address to MUX team.

    Aaron