• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 34 subscribers
  • Views 705 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Question about using LMH0302 SDO for two smpte outputs

Mitchell Kary
Intellectual 630 points
Other Parts Discussed in Thread: LMH0302, LMH0344

Hi,

I have an application where I have a LMH0302 which needs to drive two SMPTE outputs. There is no guarantee that there will be a smpte monitor connected to both SDO and SDO_BAR at the same time. In some cases, SDO will be connected, in other cases SDO_bar may be the only output connected and finally in some cases they will both be used simultaneously. Will that be a problem?

Also can I assume that if both monitors are being used then the image will be identical on both even though one is generated from the compliment output?

Thanks,

Mitch

  • 0
    TI__Genius 13475 points
    Mitch
    Ideally, the two outputs will see the same load, so when only one monitor is connected, the unused output will be terminated with a 75 Ohm terminator - but we live in the real world where that rarely happens, and in reality, it is OK - there will be a slight degradation in signal quality if the outputs are unbalanced, but it is very commonly done, and there should be no bad consequences.
    Yes, you can connect two monitors to the outputs, and you will see the same image on both. SMPTE use an encoding scheme called NRZI, in which a '1' is encoded as a transition within a bit cell, and a '0' is encoded by the lack of a transition within the cell - the result is that you can invert the signal, and it will have no effect upon the data being sent or received.
  • 0 in reply to Mark Sauerwald88474
    Intellectual 630 points

    Mark,

    Perfect! I hooked up the TI SD3GAEVK into my Tek WFM smpte analyzer and saw that the eye diagram for the unloaded case was in fact a bit degraded relative to terminating the second output as well.

    I think I'm good to go on that issue. In my application the smpte input to the LMH0344 will coming from another PCB within 8 inches of the PCB which contains the LMH0344. Should I bypass the eq?

    Also I noticed that on the evaluation board R1/L1 are physically on top of each other presumably to optimize the 75 ohm routing. I would prefer not to have to do that. Is there an alternate suggestion TI could make for optimizing the 75 ohm routing for the R1-L1parallel combination like perhaps using 0201 packages next to each other?

    I know that optimizing the return loss is not easy in general, but I was hoping that TI might have had another board implementation you could point me to that didn't rely on physically putting them on top of each other.

    Thanks,

    Mitch

  • 0 in reply to Mitchell Kary
    TI__Guru* 93670 points
    Hi Mitch,

    1). LMH0344 Bypass or normal operating mode should work(given you have just 8 inches of PCB). Given such a short distance, bypass could be better.

    2). Yes you can use inductor and resistor next to one another. We have other boards using this scheme. Please take a look at 3GDAIII
    www.ti.com/.../SD3GDAIII;tisearch=Search-EN-Everything

    Regards,,nasser
  • 0 in reply to Nasser Mohammadi
    Intellectual 630 points

    Nassar,

    The pages (ie page 16) that show layout information in the snlu112.pdf appear corrupted. Are they supposed to look like the following?

  • 0 in reply to Mitchell Kary
    TI__Guru* 93670 points

    Hi Mitch,

    I cannot read the information you have in your email. In any case, i have attached the layout file for this board. You can use Allegro Free Viewer to read this file.

    Regards,,nasser

    SD3GDA3_R2_305-PD-12-0650_PCB_081412-6.brd