LMH0397: LMH0397: Return Loss Simulation

Hi Diego,

Unfortunately, the person who responded to that E2E 4 years ago is no longer at TI, and so I do not know what he was referring to by IBIS-AMI return loss simulation schematic. I have been able to see the return loss from the existing IBIS AMI file 

If you click on the check/view s-parameters button that will allow you to view return loss for the inputs and outputs.

There are s4p files for the IN0 input and the IO input in the data folder of the LMH1297 download.

Diego Shinji Kawaguchi said:

My understanding is that the IBIS-AMI model for the LMH1297 also supports the LMH0397.
Does this "result of the IBIS-AMI return loss simulation schematic" similarly support the LMH0397?

This is a common question because the LMH0397 has the same simulation model as the LMH1297. To simulate for the LMH0397 use the schematic in the cable folder of the download:

Notice that the bit rate is 2.88 Gbps.

Best Regards,

Nick

Hello Nick,

Thank you for your response.

I have tried connecting the components as shown in the attached image, but I'm not entirely confident in the configuration, particularly with the final termination.

Could you please verify if this setup is correct?

Looking forward to your reply, and thank you again for your support.

Best regards,
Diego

Hi Diego,

Hope all is well. In this case, I am confused why the S-parameters from the IO.s4p file are not enough. Is my understanding correct that with the model you've attached above you are trying to see the sum result of the two blocks?

See this image again. The check/view s-parameters functionality should allow for this to be easier.

Best Regards,

Nick

Hi Diego,

Hope all is well. In this case, I am confused why the S-parameters from the IO.s4p file are not enough. Is my understanding correct that with the model you've attached above you are trying to see the sum result of the two blocks?

See this image again. The check/view s-parameters functionality should allow for this to be easier.

Best Regards,

Nick

Hello Nick,

Apologies for the delayed response.

To address your question, when I reference the check/view s-parameters for IO.s4p, I notice that there are S12 values.
This suggests that the final return loss values are influenced by the termination.
As a result, I kept the package model (rx_io3.s4p) connected to account for this variation in return loss.

I have two questions regarding this:

1. Why is it acceptable to remove the package model (rx_io3.s4p) from the simulation?
2. If I were to remove the package model (rx_io3.s4p), is it still acceptable to use a 75Ω termination for the simulation?

For your reference, when I tried using either a short or open termination on IO.s4p (as shown in the attached image), the return loss did not meet the SDI standards.

Looking forward to your clarification, and thank you again for your support.

Best regards,
Diego 

Hello Nick,

Thank you for your clarification and support throughout this discussion.

I will continue using the 75Ω termination with the IO.s4p file as you suggested. 

Have a nice day!

Best regards,
Diego