• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 98 subscribers
  • Views 564 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.

Sitara IBIS Model

Ned Dempsher
Intellectual 260 points

Hi  TI,

I have a question on the IBIS model TI has provided me for the I2C models.

There is a model selector for these pins, e.g. Selector 9, but none of the selected models are open drain which is the I2C requirement.

Also, the falling edges, even for the slow slew options, are very fast for I2C signals causing signal integrity headaches.

Is there an open drain I2C model available for these pins which also reflect slower fall times?

Thanks TI,

Ned  

  • 0
    TI__Guru 63075 points
    I assume you are asking about AM335x since your other recent questions have been related to AM335x. Most of the AM335x IO terminals can be configured for other functions. Many of the other functions would not work as expected if the IOs which can also be configured to operate as I2C we 100% compliant to the I2C specification. For example, the output buffers would be too slow for many functions if these IOs were compliant to the I2C slew rate requirements. Another example is these IOs not being true open-drain. If the IOs were true open-drain they would not meet the requirements of the other pin multiplexing functions. Therefore, push-pull IOs are configured to operate as open-drain when the I2C function is selected.

    It may be helpful if you read Advisory 1.0.26 in the AM335x Silicon Errata.

    Regards,
    Paul
  • 0 in reply to peaves
    Intellectual 260 points
    Thanks Very Much Paul,

    I have two questions for you:

    1. Our part number is AM3354BZCZA100 which means, I think, that it is a revision 2.1 part according to pg. 5 of the errata which means that it isn't subject to the unexpected high-level pulse when transitioning from the low to high impedance states that is covered in Advisory 1.0.26??

    2. But - I guess it is still safe to assume that the faster than normal I2C falling edges as modeled in your IBIS are still real??
    Ned
  • 0 in reply to Ned Dempsher
    TI__Guru 63075 points
    1. You are correct, the issue described in Advisory 1.0.26 was resolved in the latest silicon revision. I just pointed you to this advisory becasue it explains how the push-pull output is configured to operate as open-drain when selecting the I2C function.

    2. Yes

    Regards,
    Paul