• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 34 subscribers
  • Views 546 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.

TLK10232 LVDS driver can be interface to REFCLK0/1 input?

Kikuo Mita
Intellectual 2515 points
Other Parts Discussed in Thread: TLK10232

Hi Team,

Please advise me answring to a question bellow.

If LVDS (ANSI-644-A) compliant driver can drive the REFCLK0 or REFCLK1 inputs?

Referring the datasheet (page 98) the “Differential input voltage for REFCLK0/1 must be
from 250 mV to 2000 mV p-p.

In contrast, the LVDS driver output pulse its differential amplitude is 247 to 454 mV p-p.
It is  3mV smaller than REFCLK0/1 input specification.

I wonder if you can change datasheet specification extending the minimum differential
amplitude to 240 mV p-p or much less from 250 mV to accept SNASI-644-A LVDS input?

Mita

  • 0
    TI__Mastermind 20820 points

    Hi Mita,


    Sorry for the delay in response.

    I think the 3mV difference between the LVDS spec and the TLK10232 input is negligible and should not be considered as a reason to not use the part. Can your customer safe guard their design so that they are not at the bottom 1% of the specified voltage range? Is there a particular reason why their voltage is so low?

    I can say though that we will not be updating that spec in the data sheet as data sheet changes at TI require a major effort. Please let me know what other questions you have?

  • 0 in reply to Michael Peffers
    TI__Intellectual 2515 points

    Michael-san,

    Thank you for your feedback.

    The reason why I (or customer) is that most of clock buffer hires LVDS output.
    The LVDS electrical specifications is speechified in ANSI-664-A.
    Referring to this standard the minimum differential amplitude is 247 mV instead of 250 mV.
    Each time customers design their board using LVDS clock buffer, we may get same question
    over and over.
    LVPECL may be an option instead of LVDS buffer, however, LVPECL is power eater compered to
    LVDS.
    So I asked to revise the datasheet as long as 3 mV difference is negligible.

    Or, can you provide customer a side letter which states that TLK10232 Ref Clock input can
    receive ANSI-664-A LVDS compliant input.

    Mita.

  • 0 in reply to Kikuo Mita
    TI__Intellectual 2515 points

    Michael-san,

    I recall my request.
    I gave up recommending use LVDS driver for REFCLK0/1 inputs.


    The 247 mV of the minimum output amplitude of LVDS driver is based
    on the 100 ohm +/- 1% termination resister.
    LVDS driver outputs larger than 250 mV differential voltage against 100 ohm
    Termination resister, and 247 mV for 100 ohm -1%.

    Referring to the TLK10232 datasheet, REFCLK0/1 inputs has 100 ohm (typ)
    On chip termination resisters respectively. However, the datasheet does not
    Specifies the minimum value of the on-chip termination resister value.

    Assuming the +/-10% of validation on resister value, the input amplitude may
    be as low as 225 mV. It is far smaller than the 250 mV of minimum input level.

    Mita