1. IF The bandwidth is less you can see the angle of S12 and S13 in place of mod of S parameters. But that is not a very good way of seeing the results. So the best way is to write formula and take the difference between the phases of S21 and S31.

Sum up the path distance, for instance the path distance between port 1 to port 3 is lambda/4 + lambda/4. one complete cycle of lambda(wavelength) = 2*pi. thus phase delay = 2*pi/2 = pi = 180 degree.

note: I am assuming the dimension to be lambda/4 X lambda/4 here.

Shailendra Singh @ Thanks .

In HFSS, There is two options. One function is ' arg' and another is 'cang_deg'. which one we should consider. It shows different result regarding 'arg' and 'cand-deg'.

Hello,

Manually, it can be calculated using formulas. Some papers are attached for help.

Thanks,

Hello Debraj, thanks for your question. I think it should be very simple in HFSS to check phase difference between two ports (assuming you are interested in S parameters see my answer below):

1) Setup the simulation environment and use a frequency sweep in the analysis setup (use any: fast , interpolating or discrete).

2) once simulations are finished, go ahead to results-> right click-> create modal solution data report-> Rectangular plot -> select S parameters of interest -> select ang_deg (or ang_rad) and plot your interpreted results. you can put markers and see the difference.

3) if you want it to store a variable of phase difference define an output variable and assign the value-> for example: Ph_diff = ang_deg(S21) -ang_deg(S31) and plot using step 2) choosing your defined variable.

I hope this helps.

Cheers,

Thanks Mrs. Dwivedi and Mr. Singh.

One function is ' arg' and another is 'cang_deg'. which one we should consider. It shows different result regarding 'arg' and 'cand-deg' .

I have checked as per your instruction. I have designed a simple wilkinson divider with equal arm. It should be 0 degree phase shift in between out puts. but it shows 180 degree phase shift. Do you have any reason why it shows 180 degree phase shift ?

Hi Debraj, I would suggest to try and first find out if you are using the right method to plot your results. Try to use ang_deg (I am not sure what cang_deg is! ) . Please share your results by attaching a screenshot so that we can have a look.

Here is the attached pictures.

Thanks Debraj, from your results it seems that the output ports actually have a phase difference of 180 degree at center frequency around 1 GHz. I did not go in detail for your HFSS design. Did you check if the (dB/mag) is behaving correctly?

I would suggest you to read the original paper from Wilkinson and/or perform a circuit simulation to confirm your design works. you can see some examples here (https://www.microwaves101.com/encyclopedias/wilkinson-power-splitters). I am not sure which part is missing, but your device is working more like a hybrid coupler. Please check your resistor is correctly characterized in HFSS (I do not know how to do that, you can search on google).

Hi Aditya Singh,I want to check phase shift provided by one unit cell(metasurface) in HFSS. I used PEC PMC boundary and wave port as radiation. In your formula you used S31 but in my case only 2 port exists.

Rajib Khandokar Hello, thanks for your question. I think it should be straight forward and you should be able to check the S21 , just use port 2 instead of 1. I am not sure, what you are looking for. If this does not answer your doubt please share more details so that I can help!

Thanks again!

First you plot the phase result for S21 and S31 in magnitude, here I am considering port 1 is input port. When you will plot S21phase at that time in that equation bar you subtract it angle of S31 and you will get the plot for difference between phases of port 2 and port 3.

Aditya Singh ,Thank you for your reply. I am designing metasurface with 5 layers. Can you tell me which boundary and excitation should I use?I used PEC_PMC with wave port. How much should be deembed distance?

Rajib Khandokar In my view, after assigning the boundaries and excitation, in the wave port deembed setting should be right at the layer where you expect the incident wave to enter the 5-layered surface and while the other port should be deembed where the wave leaves the surface. In this way we can characterize the structure's response i.e. the total attenuation, phase shift it offers or the dispersion diagram as we mostly are interested in propagation inside the structure

Aditya Singh ,Thank you. Can you please help me answering below questions:

1.Which boundary condition should I use?

2.Which excitation should I use?

I used PEC_PMC as boundary and wave port as excitation. Is it right for this layer design?