• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 750 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.

DS125BR820: SFP+/QSFP Compliance Question

jack dawson
Prodigy 70 points
Part Number: DS125BR820
Other Parts Discussed in Thread: DS125DF410

Hi, 

   I'm considering using a DS125BR820 device in between an ASIC and a SFP+ cage in a new design. I saw the following app note from TI describing pretty much the same use case I have but I have a question regarding the test methods described in the document. 

http://www.ti.com/lit/an/snla226/snla226.pdf

In the document, it says a PRBS9 pattern was used for all transmit jitter and eye mask tests, however when I look at the SFF-8431 specification, section D.5 states either PRBS31 or a valid 64/66 bit pattern must be used for the total jitter measurement: 

Why did TI choose PRBS9 over PRBS31? I'm mainly concerned that since the transmit jitter specification is rather tight for the SFP interface, the app note might be under reporting the TJ? 

Thanks, 

    Jack 

  • 0
    TI__Mastermind 43795 points

    Hi Jack,

    The application note was addressing DDj components Duty Cycle Distortion (DCD) and Pulse Width Shrinkage (DDPWS).  The PRBS pattern is used for these tests per SFF-8431 section D.3.  The sampling oscilloscope used at the time could not monitor every transition within PRBS31 pattern and determine a Total Jitter measurement down to a 10E-12 error rate directly.

    It is important to note that the DS125BR820's ability to achieve low jitter is primarily determined by the driving ASIC Tx and the channel characteristics.  If the ASIC Tx does not have the best Tx jitter or channel non-idealities are significant using a DS125DF410 retiming device is a better choice.  The retimer isolates some of the channel and all the ASIC Tx characteristics from the SFP+ waveform resulting in a low jitter solution.

    Regards,

    Lee 

  • 0 in reply to Lee Sledjeski
    TI__Guru* 93670 points
    Hi Jack,
    It's been three days and there is no update on this E2E thread. I am assuming, you've been able to resolve this issue. We are closing this case and please re-open if there is a new update.
    Regards,,nasser