• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 98 subscribers
  • Views 388 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.

DM365 DDR2 IBIS Simulations

rmiller
Prodigy 90 points


Hello All,

I am trying to verify SI performance on a layout for the DM365 processor.  While trying to get the simulations set up for the DDR2 interface,  I have ran across some issues that I think are rooted in the DM365 IBIS model.  Has any one come across any of these issues, and if so how did you handle them?

1) There are many available models for the DM365 DDR2 pins; however, only a few of them are defined as I/O buffers.

2)   I have simulated using both the NOSR0_NOTERM_1P8 and the SR0_NOTERM_1P8 models.  Both of these however seem to have incorrect thresholds defined (2V VinH on a 1.8V level signal).  Am I missing something, or is this just a mistake?

3) Many of the models (such as the two I mentioned above) have non-monotonic I-V curves for pullup/ pulldowns.   I manually eliminated the regions that were non-monotonic (since they were outside the range of IC operation). 

Are there any other pitfalls to look out for in this particular IBIS model?

Thanks,

Rick

  • 0
    TI__Mastermind 27585 points

    1) Choose the ones defined as I/O buffers for simulation

    2) The thresholds may need to be corrected in the models. However they are not useful for signal integrity purposes, which is all these models can be used for for DDR interfaces.

    3) The non-monotonicity in some curves is a known issue and is being fixed. But it shouldn't affect the simulation results.

    Jeff

  • 0 in reply to jc-ti
    Prodigy 90 points

    JC,

    Can you elaborate on your response to item 2?  Why are the thresholds not useful for SI purposes? It seems like they are useful in determining how much crosstalk a particular layout may experience, and whether or not that crosstalk will cause issues.

    -Rick

     

  • 0 in reply to rmiller
    TI__Mastermind 27585 points

    You are right, it can help determine if there are any ripples near the logic thresholds. In the case of a discrepancy between the IBIS model and the datasheet values, always use what is found in the datasheet.

    Jeff