Burrs on the calculation result curve

[eexhliu]

hello, I run the code 'trl.m' for my data, but it appears many burrs in the calculated curve. the curve is centered around the DUT with pad and line simulation line and it don't show any calibration effect.

[la3pna]

How long is your line? is this on real or simulated data?

[la3pna]

if this is simulated data, is it correlated? ie same datapoints. Same applies to real VNA data, it needs to be the same frequency points for TRL to work in this matter.

[eexhliu]

It is simulated data. "T" is a zero length line, "R" is two short line.(the code is used for open and I modify it like below,"L" is a long line with 60-100 degree phase difference to "T". I think they all satisfy the TRL calibration requirement. By the way, would you please tell me how can I show the figure to you? I don't find any like "insert picture button" , thank you.

%%%% the modified code%%%%%
R1a = (w1-b)/(a-w1aca);
R1b = (w1-b)/(w1aca-a);

% An short is used for the reflection measurement. Use this information to
% chose the sign of a

if R1b>0 %because reflect coefficient of terminal one with the other terminal shorted is -1
a = a;
as(n) = a;
c = ca*a;
end;

if R1b<0
a = -a;
as(n) = a;
c = ca*a;
end;

[la3pna]

I will need to see your files, please upload them into a github repo: http://kbroman.org/github_tutorial/pages/init.html

Both files and plots would be good.

[eexhliu]

I'm really sorry I didn't see your reply because I'm busy with some other thing.
I didn't know much about code and I just sometime use MATLAB to process some data in my daily work.
so I don't know how to perform the code in the instruction you give me. If convenient, would you mind give me some other contact way such as email to send related files to you? Thank you very much for your help.