• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 33 subscribers
  • Views 715 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.

DP83867E: XI/XO interface model

Michael T Schneider
Expert 3790 points
Part Number: DP83867E


I downloaded the ibis model for DP83867ERGZ. But, the ibis model does not have any information for the 25MHz XI, XO interface. We need to do simulation for this interface in order to convince ourselves that we meet TI spec. Can you help us to get the model for this XI/XO interface?

 

For this interface, I used FC7BACBMI25.0 crystal in series with a R1 = 50 ohm resistor as the XO source termination resistor – see figure below. The 50 ohm can be other value, depending on simulation result. CL1=CL2 = 30pF.

  • 0
    TI__Mastermind 18425 points
    Hi Micheal,

    The IBIS model does not support simulation of the XI/XO interface. Looking at the datasheet for FC7BACBMI25.0 and it looks like the load cap values for the crystal is 20pF. As a general rule of thumb, capacitance on each singal pin of crystal should be about twice the load cap value in the datasheet. It would be 40pF in this case. I would recommend changing CL1/CL2 to 36pF each in order to account for parasitic capacitance added by traces and IC package. Series resistance of 50ohms is ok.

    -Regards
    Aniruddha
  • 0 in reply to Aniruddha Khadye
    TI__Expert 3790 points
    Stray capacitance was assumed = 5pF. So, CL1=CL2 = 2*(20 [for xtal] – 5 [for stray]) = 30 pF. Make sense?

    We think there is a failure mode with this XI/XO interface, that is, the PHY may work at a harmonic of the crystal frequency. How can we model this interface so that we know the PHY will only work at the crystal frequency, not at the harmonic?
  • 0 in reply to Michael T Schneider
    TI__Mastermind 18425 points
    Hi Micheal,

    Since we are working on this issue over email, I will close this thread for now.

    -Regards
    Aniruddha