• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 33 subscribers
  • Views 817 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.

SAS-3 Training and Cascading DS125BR401A and DS125BR111 Devices

mipmapped
Prodigy 20 points
Other Parts Discussed in Thread: DS125BR800A, DS125BR111, DS125BR401A, DS125BR820

I am considering using both DS125BR401A (or DS125BR800A) and DS125BR111 in a chain for SAS-3 (12Gb/s) to compensate for cable and multiple board losses. DS125BR111 would be on a backplane next to the SAS drive (TX and RX). DS125BR401A would be on a board at the end of the cable connected to the controller (also for TX and RX). There are board/connector losses between the two redrivers. Would SAS-3 link training work in this implementation? Do you see any other issues of concern?

  • 0
    TI__Mastermind 29655 points

    Hi mipmapped,

    The parts you have mentioned are all ideal for use with SAS-3 link training and OOB signaling. I'd like to add that the DS125BR111 and DS125BR820 are the latest repeaters we have released that feature a super-linear output driver specifically for link-training purposes in order to preserve subtle precursor and postcursor Tx FIR passing through the repeater. In addition, the DS125BR401A is a viable option for bidrectional SAS-3 signaling, and the DS125BR800A can be used when you need to compensate for a particularly long amount of attenuation loss (>30 dB).

    So far your application seems to work. Some questions to help us understand what to recommend:

    1. Can you provide a block diagram of the topology you plan to use and the estimated amount of loss you expect before and after the repeater? The primary area of concern before proceeding further is identifying whether the repeater you are using is appropriate for the amount of expected loss before and after the part.

    2. Are there any constraints to where the repeaters can be placed? Please allow at least 3-5 dB from the Tx output of the repeater to allow for natural attenuation of any residual distortion or over-EQ that may be present from the repeater driver.

    3. Is the repeater in a closed (fixed, endpoints never change) or open system (user can swap in any SAS drive he/she chooses)?

    4. Are you definitely planning to cascade any repeaters from endpoint to endpoint? We typically do not recommend this if one repeater will suffice. The addition of a second repeater along the signal path may make link-training more difficult due to the signal passing through more signal conditioning blocks, where the additional amount of potential distortion the signal experiences in the second repeater ends up doing more harm than good on preserving the link training pre/postcursors.

    Thanks,

    Michael

  • 0 in reply to Michael Lu (Santa Clara)
    Prodigy 20 points

    Hi Michael,

    Can I personal message you to continue this discussion?

    Alan

  • 0 in reply to mipmapped
    TI__Mastermind 29655 points

    Hi Alan,

    Sure thing. Please reach me at m-lu@ti.com.

    Regards,

    Michael