Agilent Momentum : parameters extraction

[chromatik]

agilent momentum

Hi, all !

A new question for you, I'm trying to find a way out of L, C and maybe R extraction of a structure designed in Momentum.
I got a flat spiral coil, and I'd like to obtain its specs, but without using Momentum RF.
The reason is I heard (and read) that Momentum RF is suitable for high frequency (few or more GHz), out of this range the results wouldn't be that realistic.
And I'm working in MHz scale, so...

Thanks a lot for your precious future help

Regards

 

[Abhishekabs]

momentum agilent

No one uses ADS momentum generally for MHz simulations.

 

[chromatik]

pi equivalent model from s-parameters extraction

I do
I agree with you, nobody with judgment uses Momentum for that kind of work... but I have to work with the tools I have
Sometimes, decisions bothers you and you have to face it up

 

[Abhishekabs]

momentum s parameter

Try to get help of agilent support team.
Thay can guide you.
Also You can try NEC for loop simulations.
there is free student version of sonnet 11 available. try that

 

[Joe_User]

momentum vs sonnet

As Abhishekabs mentioned, you can use the free Sonnet Lite. It can do one or two things for you:
1) If the spiral coil is simple enough, you can enter the dimensions and material properties to get S-parameters. If you have already done this in Momentum, and you trust the results, then no need to use Sonnet Lite for this.
2) Once you have the S-parameter results, you can ask for an equivalent PI network. You can use the S-parameters from Sonnet Lite, or the S-parameters from Momentum.

 

[chromatik]

momentum rf

Thanks guys
Joe_User> when you say "ask for an equivalent PI network", I was running Momentum simulation, and I don't know how to do it. If you can give me a hand with it, that would be really cool !
Thanks again !

 

[Joe_User]

agilent ads momentum coil simulation

To get an equivalent PI network for you momentum simulation using Sonnet Lite:

1) First you need to output an S-parameter file from momentum. If you already know how to do this, then skip to the next step. The procedure to do this is different for different ADS versions, and there are probably lots of ways to do this, but here is one: open a data display window (found under the Window menu) and then open the data file tool (under the Tools menu of the data display window). When done, you should have an S-parameter file (*.s2p) on your hard disk. Make note of this name and location.

2) Next open Sonnet Lite. Select Project | View Response | Browse for project. At the bottom of this window you will see a drop list called "Files of Type:". Select "Touchstone Files (*.snp)" from the list. Then browse for the s-parameter file created in step 1. This will create a plot of dB[S11] vs frequency in the sonnet graph. You should verify that this plot looks correct to you.

3) Next, select Output | PI model file. This opens a window with some options and a big text window that shows the PI model(s) corresponding to these options. The format is a spice netlist. Sonnet Lite creates a PI model based on pairs of frequencies, so if your S-parameters had two frequencies, you will see one PI model. If you had 101 frequencies, then you will see 100 PI models. I suggest you look at the first few and see if they make sense and if they are consistant. I have noticed that the model tends to break down at higher frequencies. Never trust a PI model from any software tool since it has to make the assumption that your S-parameters really do fit a PI model. If they don't, you will get a non-physical model.

4) If the model looks sensible, then just use it. If it doesn't, or if it is empty, then you will need to play around with the options. In particular, click on the "Model Options". These settings allow you to set thresholds for various components in the PI model. For example, Rmax is set to 1000 ohms. This means that anytime the PI model requires a resistor above 1000 ohms, it is hidden from the output format. The Sonnet manual says that this is done because these models can get pretty complicated and you may not want dozens of 1 megohm resistors to ground in your model since this just clutters up the model.

Hope this helps. I think I got everything right. I have used this quite a few times, and most of the time it works well, but other times I get junk (probably due to my S-parameters not fitting a PI model).