• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 19 subscribers
  • Views 777 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.

TXU0104: How do you adjust C_comp as required in the IBIS model?

awseg464ytgaw4gawv
Prodigy 120 points
Part Number: TXU0104

I'm trying to use the IBIS model for the TXU0104 to evaluate the signal integrity of the PCB.

When I read the [Notes] section in the TXU0104.ibs file, it says:

C_comp although listed in numerical order of value are obtained from varying PVT conditions. For PVT conditions for each specific value refer the buffer description under respective [Model]. IBIS model user is responsible for using appropriate C_comp to simulate for Signal Integrity.

How exactly does the user go about "using appropriate C_comp"? Am I supposed to edit the model file? For example, in the desired Model do I have to replace the C_comp values with the commented out values?

| variable                                  typ         min         max
C_comp                                      6.05pF      5.21pF      7.21pF
| C_pullup                                  14.02pF      6.09pF      25.42pF
| C_pulldown                                9.23pF      6.55pF      14.97pF
| C_tri-state                               6.05pF      5.21pF      7.21pF

If so, how do I choose which one to use?

Thanks

  • +1
    TI__Genius 16735 points

    Hello,

    The C_comp value should be selected based on what condition of the signal integrity you are testing. You should be using the pullup C_comp when looking at rising edge performance and pulldown for when looking at the falling edge performance.

    Regardless, the TXU devices are not capable of doing extremely fast data rates. So these small differences should not impact your signal integrity. What would matter more is the loading you are giving to the outputs.

    Regards,

    Karan

  • 0 in reply to Karan Kotadia
    Prodigy 120 points

    Thanks for the information! The datasheet for this device mentioned that the device has very fast edges even into heavy loads and that SI must be taken into account. However, the datasheet didn't mention anything about rise or fall times.

    The simulations with the default C_comp have produced better results than I worried about (considering the comments in the datasheet).

    But just to be clear, if I needed even more accurate results, would I be editing the actual IBIS model itself in order to adjust C_comp? Or is the adjustment done somewhere else?

  • 0 in reply to awseg464ytgaw4gawv
    TI__Genius 16735 points

    Hi,

    The device has 12mA drive strength. The rise and fall times you can hand calculate based on the datasheet parameters.

    This is a good guide on how to do that: https://e2e.ti.com/support/logic-group/logic/f/logic-forum/718814/faq-what-is-the-output-transition-rate-for-a-logic-device

    I don't believe adjusting C_comp will help any further in testing. The way to get the most accurate results would be to order samples or an EVM and test on the bench.

    Regards,

    Karan