Question about using NPORT instance in Cadence

[ngox]

Hi, I've been trying to use NPORT instance in Cadence in order to try out some S-parameter files that I have. I cannot see anything wrong with the way I'm setting up the simulation but the resulting transfer functions I get after simulation are not correct at all. The S-parameter file I am using an actual .s4p taken directly from measurement so I don't think that should be an issue either. Any help would be greatly appreciated! Thanks!

 

[BigBoss]

The source which you have used is wrong.It's a AC source and cannot be used in s-parameter simulations.
You have to use "psin" or "port" element instead of it.These components have internal source impedance ( e.g. 50 Ohm by default) and you should define this component in s-parameter simulation setup ( if you do 1-Port s-parameter simulation, you should add this in port definition).
In additional to, you may use "psin" or "port" element for other ports and if you add these elements in s-parameter simulation setup, you can do 4-Ports s-parameter simulation without adding extra resistors.

 

[ngox]

The source which you have used is wrong.It's a AC source and cannot be used in s-parameter simulations.
You have to use "psin" or "port" element instead of it.These components have internal source impedance ( e.g. 50 Ohm by default) and you should define this component in s-parameter simulation setup ( if you do 1-Port s-parameter simulation, you should add this in port definition).
In additional to, you may use "psin" or "port" element for other ports and if you add these elements in s-parameter simulation setup, you can do 4-Ports s-parameter simulation without adding extra resistors.
View attachment 109262View attachment 109263

Thanks for responding. I tried to use psin as well as port but it still seems a bit weird when I plot the s-parameters. I ran the sim using sp and selected all 4 ports, and then tried plotting the results using the main form. I'm expecting at least one of these to have some kind of lossy profile as this is just s-parameter of backplane traces. None of these look right though.

 

[BigBoss]

Thanks for responding. I tried to use psin as well as port but it still seems a bit weird when I plot the s-parameters. I ran the sim using sp and selected all 4 ports, and then tried plotting the results using the main form. I'm expecting at least one of these to have some kind of lossy profile as this is just s-parameter of backplane traces. None of these look right though.

Have you checked the format of the s-parameters in dialog box of n-port component ??
Is it Touchstone ? If it's so, you should select Touchstone format in dialog box of n-port.
Your s-parameters may be so ?? Could you checked them with another simulator for instance ADS ??