• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 886 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.

DS125MB203: Reference design for DS125MB203 and any issues with interfacing SFP+ and QSFP+ transceivers?

Narendra Reddy1
Prodigy 40 points
Part Number: DS125MB203
Other Parts Discussed in Thread: DS100MB203, DS110DF410, DS125DF1610, DS110DF111

Hi,

I am looking for a reference design for DS125MB203. Is any one has previews experience in using this part and faced any issues with interfacing SFP+ and QSFP+ transceivers?

Thanks

Narendra

  • 0
    TI__Mastermind 19235 points

    Hi,

    We unfortunately did not release a QSFP+ or SFP+ reference design for DS100MB203. My input below:

    • The DS100MB203 should be fit for use for QSFP+/SFP+ host applications for optical media cases. This repeater has more than enough EQ to compensate for the module host channel insertion loss specified by SFF MSA. The only limiting factor for this case is the output random jitter for the switch ASIC-SerDes interfaced to the DS100MB203, as the repeater does not retime the signal
    • If passive direct attach copper cables need to be supported I would instead recommend to use a retimer such as TI part number DS110DF410

    Cordially,

    Rodrigo Natal

    TI HSSC Applications Engineer

  • 0 in reply to Rodrigo Natal
    Prodigy 40 points

    Thanks for the info.

    I have attached my block diagram.Could you please let me know do you see any issue with QSFP+ to QSFP+ or SFP+ to SFP+ interface with direct attache cables like shown in the attachment?Do we need to use any additional device on the path additional to 1:2 mux for my application?

    Thanks

  • 0 in reply to Narendra Reddy1
    TI__Mastermind 43795 points

    Hi,

    Are you also using the 2:1 Mux path in the "Right to Left" direction?

    What is the approximate attenuation in the channel between the ASIC and the MUX?

    What is the approximate attenuation in the channel between the Mux the the SFP+/QSFP+?

    Will the interface utilize an equalization training algorithm?

    Regards,

    Lee

  • 0 in reply to Lee Sledjeski
    Prodigy 40 points

    Hi Lee,

    Please see my comments inline...

    Are you also using the 2:1 Mux path in the "Right to Left" direction?

    Ans:Yes the Mux is in both directions.Our requirement is just switching between 10G and 40G path so TI suggested to use this Mux.

    What is the approximate attenuation in the channel between the ASIC and the MUX?

    Ans: Design is not yet started but I assume the trance length in between  2-3 inches  

    What is the approximate attenuation in the channel between the Mux the the SFP+/QSFP+?

    Ans: Design is not yet started but I assume the trance length in between  2-3 inches  

    Will the interface utilize an equalization training algorithm?

    Ans:Right now don,t have much information.Let us know your suggestions based on the above trace lengths whether it is really required or we can go ahead with defaults values.

    Thanks

    NR

  • 0 in reply to Narendra Reddy1
    TI__Mastermind 43795 points

    Hi NR,

    With the passive DAC cables the total channel attenuation is unknown and that makes the redriver mux difficult to implement.  As Rodrigo has said a retimer is preferred in this type of solution.  The DS110DF111 is a relatively small device which has a can be programmed to have a mux-fanout function.  Otherwise, the DS125DF1610 product would work well if many signal paths are involved.

    Regards,

    Lee