Author Topic: S parameters of a connector  (Read 747 times)

0 Members and 1 Guest are viewing this topic.

Offline RFbeginnerTopic starter

  • Newbie
  • Posts: 8
  • Country: de
S parameters of a connector
« on: August 15, 2019, 08:50:45 am »
Hello everyone,
I am desperately trying to find the Scattering parameters of a connector. For that purpose, I designed two lines, one of length L and the other of length 2L.
First idea was to use the T parameters to find the connector’s S parameters and the idea was as follow:
Tmeasured_L = Tc Tline Tc
Tmeasured_2L = Tc Tline Tline Tc
And then: [Tmeasured_2L] * [Tmeasured_L]-1 = Tc Tline Tc-1 = Ta
                    [Tmeasured_L]-1 * Ta = Tc-1 Tc-1
and then invert everything and square root… this is the matlab code but the result cannot be correct

Code: [Select]
%% S Parameters 


DataL = read(rfdata.data,'S2P-L.s2p');
Data2L = read(rfdata.data,'S2P-2L.s2p');

freq = DataL.Freq;
SPL = extract(DataL, 'S_PARAMETERS',50);
SP2L = extract(Data2L, 'S_PARAMETERS',50);

%% S to T

TPL = s2t(SPL);
TP2L = s2t(SP2L);
%% size
size(TPL)
length(TPL(2,2,:))
%%

for i=1:length(TPL(2,2,:))
   
   
    TA(:,:,i)=TP2L(:,:,i)/TPL(:,:,i);

    TB_1(:,:,i) = TPL(:,:,i)\TA(:,:,i);

    TB(:,:,i) = inv(TB_1(:,:,i));
   
    TC(:,:,i)=sqrtm(TB(:,:,i));

end

SC = t2s(TC);

so next idea was to try and find every parameter at a time and I already determined the complex propagation constant, both alpha and beta and seems ok but kinda lost on how to find S11 and S22.
Any ideas?
thanks

                             
                   
 

Online rfeecs

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: us
Re: S parameters of a connector
« Reply #1 on: August 15, 2019, 06:26:43 pm »

First idea was to use the T parameters to find the connector’s S parameters and the idea was as follow:
Tmeasured_L = Tc Tline Tc
Tmeasured_2L = Tc Tline Tline Tc


To start with, this doesn't look right considering the connectors are reversed at the input and the output.

This is similar to the "Thru Delay" method of test fixture de-embedding.  If you have a network analyzer with an "adapter removal" function, it has built in math to do this as well, provided you have the appropriate calibration kits. 

This problem has been solved and there are many papers describing the messy algebra.  Here is one:
https://s3.amazonaws.com/academia.edu.documents/37546242/new_procedure_774.pdf?response-content-disposition=inline%3B%20filename%3DA_NEW_PROCEDURE_FOR_SYSTEM_CALIBRATION_A.pdf&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWOWYYGZ2Y53UL3A%2F20190815%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20190815T182108Z&X-Amz-Expires=3600&X-Amz-SignedHeaders=host&X-Amz-Signature=c21ed26ee4903fd8deb689f2819df37c30ce04e1db73ce2bf7adf7b9d2d58276

Here is another:
https://ieeexplore.ieee.org/document/76424
This one is paywalled.  But it has a long list of references.
« Last Edit: August 15, 2019, 06:30:31 pm by rfeecs »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf