• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 646 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.

AM5728: SATA signal multiplexing

Brian Hammerbacher
Prodigy 245 points
Part Number: AM5728
Other Parts Discussed in Thread: TUSB9261, HD3SS3212

I am looking at implementing a mux of the SATA signals from the Sitara AM5728, a TUSB9261, and a M2 SATA drive.  I was looking at using HD3SS3212IRKSR.  I wanted to check and see if there would be any issues as the MUX, and TUSB9261 and the SATA drive would be powered at times when the Sitara is off.  If this is an issue, are there and suggested MUXes that may alleviate the problems?

Thanks,

Brian

  • 0
    TI__Mastermind 27860 points
    Hello Brian,
    I can't speak to the suitability of this particular switch for SATA in general, but I would caution against muxing *any* of the high-speed interfaces present on AM5728. The parasitic and timing influences on the the signal pairs can be problematic for these interfaces given their high speed and (relatively) small signal swings. Please refer to SPRAAR7 Appendix A - Table 4 for device-specific high-speed implementation guidelines and requirements.

    As for your power question, please refer the caution notice in section 5 of the AM572x TRM stating that all IO cells are NOT fail-safe compliant and as such should not be externally driven in absence of their IO supply.
  • 0 in reply to -DK-
    Prodigy 245 points
    Hello,
    Maybe I posted this in the wrong section, that mux is specifically made to mux USB 3.1, SATA etc. I have used it already on another project to mux SATA between two M2 drives and a TUSB9261 device.

    I understand that the AM572x IO cells can not be driven, my question really is since the MUX in the situation with the AM572x being unpowered would not be driving the AM572x, they should be floating from my understanding of the mux, would that be sufficient to not drive the IO. This may be a question that would be better answered by a someone with knowledge of the HD3SS3212 Mux which is also a TI part. Do I move this to a different section?

    Thanks
  • 0 in reply to Brian Hammerbacher
    TI__Mastermind 27860 points

    Hi Brian,

    Yes, HD3SS3212 questions would need to be addressed in a different forum (a different organization inside TI supports this device). I'll move this thread for you.

    Please do note that my cautions on muxing AM572x high-speed signals remain. From a processor perspective, we do not recommend it.

  • 0 in reply to -DK-
    TI__Mastermind 43795 points

    Hi Brian,

    From the perspective of the MUX (HD3SS3212) there is no problem with powering down the Sitara processor.  Since the HD3SS3212 is powered, it will still pass any signaling  based on the select logic signal level.

  • 0 in reply to Lee Sledjeski
    Prodigy 245 points
    Lee,
    I think what I'm trying to find out is what state are the IO PADS in which are not selected. So if I have the sitara unpowered and the ones that would be connected to the Sitara are not selected on the HD3SS3212, are those IO PADS high impedance or are they driven by the HD3SS3212?

    Thanks,
    Brian
  • 0 in reply to Brian Hammerbacher
    TI__Mastermind 43795 points
    Hi Brian,
    Any unselected pad would be hi-z (open switch).
    Regards,
    Lee