• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 32 subscribers
  • Views 636 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.

LMH0303: SDO output undershoot/ringing on 75R load

Matteo Silotto
Prodigy 110 points
Part Number: LMH0303
Other Parts Discussed in Thread: LMH0397

Hi all,

we are using LMH0303 to convert 3G-SDI 100ohm differential signal to 75ohm single-ended output.

We have an issue with the output signal (bad eye diagram).

As first we had suspicions that an impedance mismatch could generate those problems, so now we have removed most of the output components (L-R) from the original configuration and leave only a simple 75R on both outputs.

This is the actual configuration:

Components are placed near the cable driver, with short traces.

This should be the best impedance matching on LMH030 output but, even with this configuration, the waveform presents a ringing that creates undershoot on the center of the eye diagram.

What could be the reason for this behavior? 

  • 0
    Prodigy 110 points

    This is the PCB layout, with the indication of mounted components and oscilloscope probe contact points:

  • 0 in reply to Matteo Silotto
    TI__Guru* 88875 points

    Hi Matteo,

    It is best to have return loss passive network installed. This provides optimum impedance matching through 3GHz bandwidth.

    Once you have return loss network installed, it would be a good idea to do return loss measurement to make sure it is compatible with SMPTE requirements. Additionally, this checks passive or imperfect pcb layout as well.

    Reviewing your pcb layout, I think there are some improvement we can make:

    1). Probe Signal: Assuming you have 75-ohm controlled impedance, this pad may indtroduce parasitics degrading your return loss performance. Normally we make sure there is void(no GND or VCC under these pads).

    2). Also, we put two 75-ohm pullups in parallel with once another. this provides a better magnetically balanced impedance.   

    Please attached note LMH0303EVM return loss, output eye diagram, and pcb layout for your reference. Also, it maybe a good idea to consider another device - LMH0397 - since this part requires just a single 4.7uS cap on the output(there is no return loss network).

    Regards,, Nasser 6136.SD303_600137-002 REV-B.ZIP

    Regards,, Nasser 

  • 0 in reply to Nasser Mohammadi
    Prodigy 110 points

    Hi Nasser,

    As you indicated, we remounted the return loss network and have made many other measurements in this condition.

    We are not far from obtaining the right output signal so we hope that with a PCB improvement, with your suggestions, it should become compliant with the specification.

    Thanks for your quick response.

  • 0 in reply to Matteo Silotto
    TI__Guru* 88875 points

    Hi Matteo,

    Thanks for the update. I am going ahead and close this thread. Please open up a new case if there is an issue.

    Regards,, Nasser