• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 33 subscribers
  • Views 1618 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.

LMH0303: Input termination query

Roger Wilson
Prodigy 80 points
Part Number: LMH0303
Other Parts Discussed in Thread: LMH0387, LMH0344

In the LMH0303 datasheet, centre-tapped termination is shown for the SDI differential inputs. Is this really necessary? Would a simple 100R resistor between the two differential lines work just as well (and be easier to layout on a PCB). My application involves driving these inputs from a Xilinx FPGA GTX transceiver outputs no further than 3 or 4 inches away so a good signal quality should be available.

Thanks,

Roger.

  • 0
    TI__Mastermind 29655 points
    Hi Roger,

    The purpose of a center-tapped termination for the SDI differential inputs is to help defend your input signal to the LMH0303 from common mode noise. While this is highly advised to ensure that your SDI signal is as clean as possible going into the LMH0303, it is not a hard requirement for the LMH0303 to operate, and if you can assure that the signal quality is clean, a 100-ohm resistor across the input polarities will also work.

    I would still advise for robustness that you implement the split 50-ohm terminations with center tap to GND so that you can allocate a pad to include this center-tap resistor if necessary. You can then choose later to populate or not populate it based on results at the time of evaluation.

    Regards,

    Michael
  • 0 in reply to Michael Lu (Santa Clara)
    Prodigy 80 points
    Hello Michael,
    Thanks for your reply. It was just that in the datasheets of other devices e.g. LMH0387, a simple 100R resistor is shown. I was wondering if there was some deeper significance to the LMH0303 showing the split termination whilst others didn't. Thanks, Roger.
  • 0 in reply to Roger Wilson
    Prodigy 80 points
    Just spotted another point. In the datasheet, the input filter has a 3.9uH inductor across the 75R resistor. The SD3GDAEVK evaluation kit design uses a 6.8uH inductor. Is there some significance to this? Which is the "right" value to use please? Thanks, Roger.
  • 0 in reply to Roger Wilson
    TI__Mastermind 29655 points
    Hi Roger,

    The choice of the inductor is dependent on the output return loss present in your overall system, so this is a system-tuned parameter. For the SD3GDAEVK, the ideal value was found to be 6.8 uH. The output return loss varies with different inductor values and there is a trade-off between return loss margin (compared to the SMPTE limit) and the rise/fall time (slew rate). Higher inductor values provide more return loss margin at the expense of a slower slew rate. It is best to find the appropriate balance when tuning the inductor value to meet your design requirements.

    For now, I suggest starting with 6.8 uH, and then try progressively smaller inductors, each time checking to see whether the resulting return loss and slew rate meet your expectations.

    Regards,

    Michael
  • 0 in reply to Michael Lu (Santa Clara)
    Prodigy 80 points

    Hi Michael,

    Sorry I forgot to say I'd moved on to the LMH0344 with this question relating to the input filter from the cable to the equaliser.

    Roger.

  • 0 in reply to Roger Wilson
    TI__Mastermind 29655 points

    Hi Roger,

    The same concept will apply for choosing the inductor for the Return Loss network on the LMH0344.

    Regards,

    Michael