• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 34 subscribers
  • Views 1304 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.

LVDS buffer for 10m twisted pair cable

Nicolas Bruneau
Prodigy 490 points
Other Parts Discussed in Thread: DS92001, DS90CP22, DS10CP152, DS90LV001, DS92LV010A

Hi,

I'm looking for a LVDS buffer with fanout 1:1 or 1:2
The goal is to drive a 40MHz clock or data over a 10 meter twisted pair non shielded cable.

Particularly good feedback would be to have the devices part number that can support those requirements as well as some info on how to calculate the maximum cable distance this buffer can drive and under which conditions. 
App notes, field data, simulations are more than welcome :-)

Thanks for your time and your support,
-nico

  • 0
    TI__Mastermind 43795 points

    Hi Nico,

    Looking up attenuation curves for CAT5 UTP wiring gives an attenuation of 15dB/100meters @ 40 MHz.  Since this is just 10 meters, the attenuation is limited to about 1.5dB.  This is well within the capabilities of an LVDS buffer to receive.

    Here are a couple devices to consider.

    1:1  DS90LV010 or DS92001

    1:2 DS90CP22 or DS10CP152

    If you are using a RJ-45 connector, I would avoid using the pair of signals which is typically routed on pin 3-6.  Instead I would use pins 1-2 and/or 7-8.

    Best Regards,

    Lee

  • 0 in reply to Lee Sledjeski
    TI__Prodigy 490 points
    Hi Lee,
    Thanks for your fast reply.

    Is there a typo on DS90LV010 ? Is it not DS90LV001 instead?

    PS: What is the fundamental difference between DS10BR and DS90LV and SN65 products? cause they look very similar except the frequency range they address.
  • 0 in reply to Nicolas Bruneau
    TI__Mastermind 43795 points

    Nico,

    The DS90LV010A works better at low speeds.  It has a slower edge rate to accomodate topologies with multiple loads like backplanes.  The DS90LV001 can work at higher speeds - it has more bandwidth and a faster output edge rate.  It is primarily designed to work in a point - to - point topology.

    Variations of LVDS devices have been developed since the early 90's.  Over this long period of time specific devices and families have been tailored to meet new demands and speeds.  This has resulted in LVDS branching out into BLVDS, MLVDS and other derivative technologies which solved new system problems.  Since they all share common history many of the specifications look very similar. I would say that speed is definately a primary difference among the components and the root cause is the system topology they are designed to work in.  Backplane technologies drive multiple loads and have slow edges while cable technologies drive single loads and the edge rates tend to be faster.

    Regards,

    Lee

  • 0 in reply to Lee Sledjeski
    TI__Mastermind 43795 points
    Sorry ... DS92LV010A
  • 0 in reply to Lee Sledjeski
    Intellectual 450 points

    for RJ45 why 3-6 pair should be avoided. What implication will be there this pair i being used??

  • 0 in reply to Dexter
    TI__Expert 3175 points
    Lee is on vacation, let me jump in to give some explanation:
    For RJ-45, there are 4 differential pairs, and the standard pin-arrangement for ethernet is 1-2, 3-6, 4-5, 7-8.
    RJ-45 is not designed for high speed use, there is no isolation between adjacent pair. Expecially for bi-direction links (ie one pair TX, one pair RX). For improved result, it is suggest not to place adjacent pairs at 1-2, 3-6, if possible, put them at 1-2 and 7-8 will give better crosstalk reduction.
    regards,TK Chin