• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 788 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.

Part selection help - high speed differential 1:8 mux (2.5GHz)

Shlomi Bakish
Prodigy 30 points
Other Parts Discussed in Thread: SN65LVCP408, TS3USB3000, TS3USB3031

Hello

I'm trying to find a solution for muxing of CSI signals. This is for ATE board.

I have 2 lanes + 1 clock (all differential) TX port  and 8 x (2 lanes + 1 clock) RX (differential) ports.

So I need to be able to do a loopback from the TX port to each of the RX ports, which means 1:8 (3 pairs).

The CSI2 D-PHY is 2.5GHz 200mV ptp differential signal (it has also a SE 10MHz 1.2V part, changing on the fly)

RF relays are not a good solution because of their size,

TS3USB3000 is a good solution if it was SP8T. I could use it to build a mux, but that means the signal will travel through 4 such switches in series, I'm not sure
what will be the impact on signal integrity.

Unit cost is not a problem. 

  • 0
    TI__Mastermind 43795 points

    Hi,

    The closest solution I can think of for this application would be a 8x8 crosspoint switch.

    I do not know if it could interface directly with the CSI signals and something would need to be done for the 1.2V SE signaling as well.

    The part number is SN65LVCP408

    Regards,

    Lee

  • 0 in reply to Lee Sledjeski
    Prodigy 30 points
    thanks for your answer.
    The device you recommended is used to enhance the signal integrity from what I can understand.
    I need the connection to have minimal impact on the signal, for better or worse. I need to catch failing devices, and pass good devices,
    so if the signal output is marginal and the mux is enhancing it - I might pass a bad unit.
    I can live with minimal degradation and losing some good units.
    Any other ideas ? what about the option of using multiple 1:2 switches like the TS3USB3000 ?
    It's very small, so no problem with board space, but how bad will it be to pass such a signal through 4 such switches one after another?
  • 0 in reply to Shlomi Bakish
    TI__Intellectual 1165 points

    Hi,

    I think you would only need 3 levels of switches when using SPDT. Please consider TS3USB3031, and using only the two USB paths as they are exactly the same, while the MHL path in TS3USB3000 and TS3USB3031 has different specification than the USB paths.

    I can not answer you how much your signal will be degraded by 3 TS3USB3031 in series, because this will depend on more parameters (layout, general signal strength)

    Best regards

    Cevin

  • 0 in reply to Cevin
    Prodigy 30 points

    Thanks ! seems I missed the difference between the USB and MHL paths, so the  TS3USB3031 is indeed better, while using only 2 usb poles.

    I do need a 4th level of muxing - because I need to be able also to open the loop, i.e - not connect to any RX PHY, instead the NO poles will be routed to ATE digital channels
    for DC measurements.

    I will try to get the IP vendor to review the  TS3USB3031 when placed 4 in series.