I think you can measure this by comparing it's performance witha vector network analyser between two OMTs with the OMTs joined directly together. You should be able to see the equal amplitudes in the channels with 90 degrees phase difference introduced (as well as a large common phase).

we feel you must first evaluate the OMT performance for it insertion loss port to port isolation and phase imbalance in between the ports. As these data needed to externally eliminate the OMT effect on the measured value of you dual polarizer ( It is a simple mathematical steps). After getting the OMT DATA on hand then we need go to antenna test range facility for further your dual polarizer evaluation. If you are intrusted in evaluation of correct axial ratio (AR) and XPD (cross polar discrimination) phase difference. Because you may intrusted in knowing 3dB axial ration angle, we need position angle up which received power drop up to 3dB with respect to received power at boresight (same time rotate the range antenna at appropriate speed). This measurement we need to repeat for both the ports separately ( While taking the measurement at port one the second port must be kept terminated). At this point you have finished the axial ration evaluation. For evaluation of polarizer port isolation first we need to keep a leaner polarized range antenna and for DUT (your dual polarize feed) align one of the port to received for its maximum power recaption. note down the received power and without disturbing the physical setup condition measured the output power form second port. The difference between these to reading will give the port to port isolation value. now for cross polarisation discrimination evaluation, at initial while taking isolation measurement we have kept LP range antenna and one of your dual polarizer port position matched for its maximum power receptions, plot the this received power level for one 360 degree pattern. At this time pl. do not disturb the any physical condition of setup except care fully rotate the range antenna to it 90 degree position now take once again 360 plot for power output from your dual polarizer. this plot should be plotted on same graph sheet on which you have plotted first 360 degree pattern. you may see there is power level deference between these two plot. this power level difference gives you XPD data for one the port of your dual polarizer. Now repeat the same measurement setup for second port. By these steps you can evaluate your dual polarizer.

Malcolm White and Shrivastav a.k.

First thank you for your response. I still try to measure the polarizer correctly. In my research I found a blog post. You may see it via link --> https://www.ainfoinc.com/blog/blog-how-to-measure-the-axial-ratio-of-the-circular-polarizer.html

In there the method is similar as Malcolm White said. And it looks more simpler and quicker to do for me. You both may see the pictures of my measurement setup. I put two 4 port OMT's back to back. Each OMT has 2 TX and 2 RX ports for vertical (V) and horizantal (H) polarizations. In this first setup I did the measurements S21-VV (TX-V to TX-V) and S21HH (TX-H to TX-H). Then I put the polarizer between the OMT's as in picture at 45degree alinged and measure S21-VV and S21HH again. So it's like in blog post ILV1 ILH1 and ILV2 ILH2 I assume. Then I took amplitude division in linear scale;

Delta-V=ILV2/ILV1 and Delta-H=ILH2/ILH1

So if it's true so far Delta-V and Delta-H would be my polarizer's vertical and horizantal results in linear scale. Then I also took their amplitudes division to find Axial ratio --> AR= Delta-V/Delta-H

Then using this formula I calculated cross polar discrimination --> XPD=20*log10((AR+1)/(AR-1))

The results I had from those calculations in Matlab seems so perfect but also unrealistic beacause I've never seen a polarizer that has these kind of perfect XPD results. So I beleive I had made a mistake because I also couldn't see 90 degree phase difference between measurements. But I can't find where. If you have any idea please tell.