DP83822HF: IBIS model Ringing issue

Hi Abhishek,

I found a similar incident on E2E. Please check if this helps. https://e2e.ti.com/support/interface-group/interface/f/interface-forum/990080/dp83822hf-issue-with-dp83822-ibis-model

--
Regards,
Gokul.

Hi Gokul,

Thanks for providing the link...!! Here I want to ask you, can IBIS model be tuned/modified for our simulation. 

~Abhishek

Hi Abhishek,

IBIS results depend a lot on the tool being used. I think it should be okay if we can reasonably fine tune the IBIS model to get meaningful results.

--
Regards,
Gokul.

Hi Gokul,

Thanks....!!!

When we look into the IBIS information provided by Texas, we interpret that the second line data of falling edge might be causing this issue as it is resulting in high slew rate.

Can we ignore/comment-out second data points of every falling edge for all the buffer models, where slew rate causing discrete nature and proceed for simulation ?

~Abhishek

Hi Abhishek,

I am not aware if commenting out second data points will solve the issue. We use hyperlynx for the simulations and we can't recreate the issue. So I can't try out fine tuning experiments.

Please check from your side if this tuning works and proceed further if you don't see any issues.

--
Regards,
Gokul.

Gokul,

is it ok to tune IBIS model for our simulation.....!!

Concern here, aren't we ignoring some of the behavioral information of the device under consideration ? 

~Abhishek

Hi Abhishek,

You can tune the model to make it work for your tool. Please send me the tuned file with highlighted changes so that I can review them.

You can do a few checks from your side with tuned model to verify if the following aspects are covered after tuning.

1. No high overshoots/undershoots with 50ohm ideal trace model.
2. Significant undershoot with 40 ohm trace model.
3. Significant overshoot with 60 ohm trace model.

Please let me know after you have the results.

--
Regards,
Gokul.

Hi Gokul,

I am attaching fixed IBIS file of the model. Here second data point of falling edges are commented. I would like to communicated and get response on below points

• Fixed IBIS model with changes made can be included in simulation for replicating the behavior ?
• Can you please verify above mentioned checks for us on the tool you specifically use for simulations?

Also, Please confirm us that we are good to include this fixed IBIS model into our simulation and analysis.

Thanks.

~Abhishek

dp83822_fixed.ibs

Hi Abhishek,

Thanks for the sharing the file. I can do the required checks in Hyperlynx and will get back to you by Thursday.

--
Regards,
Gokul.