• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 338 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.

LMH1297: Setup for use at FibreChannel rates

David Ehlers
Prodigy 45 points
Part Number: LMH1297
Other Parts Discussed in Thread: LMH0397

Does the device require any special setup when being used for rates other than SMPTE? We are trying to use it at the 4x FibreChannel rate of 4.25Gbps, which doesn't match any of the reclocker rates.

Sometimes the device requires several resets (performed by power-cycling the board) before an error-free output is observed. Additionally, performance improves when the board has been running for a few minutes. We're not sure if setup or signal integrity is the greater factor.

Thanks!

  • 0
    TI__Guru* 88905 points

    Hi David,

    LMH1297 does not lock to the 4.25Gbps FC. Also, CTLE on the 50 ohm side is centered to boost signals at 6GHz. Could you please let us know your application for the LMH1297 device or a block diagram if possible.

    Regards,, Nasser 

  • 0 in reply to Nasser Mohammadi
    Prodigy 45 points

    The required protocol is based on FibreChannel. Note that we are getting it to work - it's just a problem getting it started after power on, with the sensitivity focused on the RX side. Are you suggesting that it should not work?

    Thank you.

  • 0 in reply to David Ehlers
    TI__Guru* 88905 points

    Hi David,

    LMH1297 hasn't been tested or characterized at 4.25Gbps and this is the concern I have.

    1). I am assuming you are using 75-ohm coax cable between LMH1297 CD and LMH1297 EQ. What is the cable reach between LMH1297 CD and LMH1297 EQ?

    2). Do you see a better performance if you put both LMH1297 CD and LMH1297 EQ in CDR bypass mode?

    Regards,, Nasser

  • 0 in reply to Nasser Mohammadi
    Prodigy 45 points

    1) We are using 25' of 1694A.

    2) Equalizer OUT_CTRL is set to equalizer bypassed, reclocker enabled. This produces errors at power on and eventually stabilizes after warming up and received several board-level resets. Our "original setting."

    Setting OUT_CTRL to enable the equalizer and reclocker always produces errors.

    Setting OUT_CTRL to bypass the reclocker (equalizer setting doesn't seem to matter) allows for an error-free start up, but destabilizes as the design runs ("warms up"). This seems like the opposite behavior of the "original setting."

    I understand your concern with 4.25Gbps. It just seems like we're close to success. Is there another device we should be considering besides the 1297?

    Best regards,

  • 0 in reply to David Ehlers
    TI__Guru* 88905 points

    Hi David,

    I would have expected CDR bypass would be our best option.

    You can try using LMH0397 which is pin compatible with LMH1297. LMH0397 equalizer gain frequency is closer to the 4.25Gbps versus LMH1297. In either case, please disable CDR or put device in CDR bypass.

    Regards,, Nasser

  • 0 in reply to Nasser Mohammadi
    Prodigy 45 points

    Thank you for the recommendation. However, I don't see any indication in the LMH0397 data sheet that it can operate beyond 2.97G. Also, I'm not sure what you mean by "equalizer gain frequency."

  • 0 in reply to David Ehlers
    TI__Guru* 88905 points

    Hi David,

    LMH0397 3G SDI Nyquist frequency(1.5GHz) is closer to your 4.25Gbps Nyquist than LMH1297 12Gbps data rate. If you like, please send me a block diagram with some description of your application so we can close this case. Please below note email address: Nasser.mohammadi@ti.com.

    Regards,, Nasser