• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 34 subscribers
  • Views 935 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.

DS80PCI402 / IBIS-AMI Model

Toshiyuki
Mastermind 7075 points

Hi,

Our customer has some questions about DS80PCIxxx IBIS-AMI Model.

*Q1

Does this model have package model, e.g. RLC ?

*Q2

Which do you recommend transmission impedance 100Ω or 80Ω ?

*Q3

Does this model have PVT(process, supply voltage, and temperature) parameters and junction temperature?

*Q4

What is the value of jitter in this model?

Best Regards,

Kato

  • 0
    TI__Expert 3175 points

    Hi Kato San,

    We are looking into this post and will respond soon.

    Regards,

    TK Chin

  • 0
    TI__Intellectual 1985 points

    Hi Kato-san,

    1. There is an s-parameter model for the package included with the IBIS-AMI model release folder, called repeater_pkg.s4p.
    2. This device can work with either 100 Ohm or 80 Ohm. The low-frequency differential input and output impedance is designed to be ~100 Ohms, however I know that many PCIe components (connectros, etc.) are designed to be 80 Ohms, so whether you design your traces to be 80 Ohms or 100 Ohms will depend on the overall channel and its associated components.
    3. The model is for nominal process/voltage/temperature corners as it is meant to be a general indicator of device and system performance. Also, since we use a bipolar silicon germanium process, we do not have as much performance variation with PVT as you would see from a traditional CMOS process.
    4. Residual deterministic jitter will be modeled in this device when the equalization is applied during the simulation. It is not modeled with separate jitter parameters. Therefore, there is no fixed value for jitter in the model, as it will be a function of the channel and the applied equalization.

    Hope this helps.

    Casey

  • 0 in reply to Casey Morrison
    Mastermind 7075 points

    Casey - san,

    Thank you for the reply.

    Please let me ask some questions about your reply for Q4.

    Residual deterministic jitter is defined in the data sheet.

    So, does the IBIS-AMI model include the value of residual deterministic jitter?

    Best Regards,

    Kato

  • 0 in reply to Toshiyuki
    Mastermind 7075 points

    Casey - san,

    Residual deterministic jitter is defined in the data sheet.

    So, does the IBIS-AMI model include the value of residual deterministic jitter?

    I'm looking forward to your reply.

    Best Regards,

    Kato

  • 0 in reply to Toshiyuki
    TI__Intellectual 1985 points

    Hi Kato-san,

    Yes, practically speaking it does. Residual deterministic jitter, in the context it is presented in the datasheet, is nothing more than ISI which has not been compensated for by the equalizer. This can come about as a result of under-equalization or over-equalization, and therefore it is captured by the model.

    Regards,

    Casey