The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> Modelling differential/common mode input impedance
https://designers-guide.org/forum/YaBB.pl?num=1031772436

Message started by Hans_Pletzer on Sep 11^th, 2002, 12:27pm

Title: Modelling differential/common mode input impedance
Post by Hans_Pletzer on Sep 11^th, 2002, 12:27pm

I am currently trying to simulate the combination of a lna followed by a mixer using Spectre.

In order to partion the design (and the simulation work) I would like to simulate the lna as standalone, terminated with a T-model of the differential mixer input (btw. the article concerning testbenches for differential circuits is very helpfull indeed) taking the differential mode and common mode input impedance of the mixer into account.

Up to this point everything is fine, but I have some questions regarding the noise analysis of the differential lna. It is no problem to use "noise analysis", since I can specify the output of the circuit simply as node, but I don't have an idea how to apply the s-parameters analysis since it requires the usage of a port component at the output.
This means to me I have to replace the T-model (which consists of complex impedances in general) of the mixer input impedance somehow with a combination of port components-but how should such a combination look like resp. how can I take the imaginary parts of the T-model into account?

It would be great if somebody could help me,

hans :)

Title: Re: Modelling differential/common mode input imped
Post by Ken Kundert on Sep 12^th, 2002, 11:34am

Hans,
What is it that you are trying to gain by performing the s-parameter analysis? What does it give you that the standard noise analysis does not?

-Ken

Title: Re: Modelling differential/common mode input imped
Post by Ken Kundert on Sep 14^th, 2002, 5:15pm

Hans,
Let me ask the question a different way. Is the issue that you want to know how to measure the s-parameters of a differential circuit? That topic is discussed in http://www.designers-guide.com/Analysis/diff-sp.pdf. Or, is the issue that you want to know how to measure s-parameters when there is no obvious resistive load? I don't think I know the answer to that one. It's not clear that s-parameters make much sense in that situation. That is why I asked why you want the s-parameters.

If you are determined to compute the s-parameters, you might consider building a matching network that fits between the LNA and the load port that to the LNA exhibits the impedance of the mixer.

Alternatively, you can build a model for the load that looks like the mixer, and then sample the output of the LNA using a VCVS and use that to drive the load port. In this case, of course, S22 and S12 are meaningless.

-Ken

Title: Re: Modelling differential/common mode input imped
Post by Hans_Pletzer on Sep 15^th, 2002, 10:58pm

Hi Ken,

thank you for your answer. I already have read your article concerning the differential s-parameters and it has been of great help - especially the ideal balun has made "sim-life" easier. The reason why i intended to use s-parameters analysis is that it directly delivers minimum noise figure and Gopt.

I now model the mixer input with an equivalent T-model and now sample the output using a vcvs like you proposed - works fine.

Hans