Analysing a coaxial to rectangle waveguide transitions

[sandruska]

coaxial-to-waveguide transition design

I am Analysing a coaxial to rectangle waveguide transitions and the results of the parameters S11 and S21 aren't corrects.
I am using rectangles for aproximating the coaxials and I'm using like excitation, lumped ports (internal to waveguide). Waveguide material is vacuum, and I haven't assigned any Boundary to the design. With this, the results of parameters S11 and S21 in the range of 11-15 GHz are wrong in the order of -10dB, why?. (please, see the design)

Thank's for all.

 

[studyEda]

Could you provide the dimensions of the coaxial and rectangular waveguide?

 

[asdfaaa]

Actually I didnt get much information about your setting of excitation and lunmped port, I can only base on what I guessed from ur figure.
And also you can connect the coax pin out with coaxial line, then can use wave port outside the waveguide, we can catch all information about the mode transition.
Best Regards,

 

[asdfaaa]

sry, figure missed

 

[itaifrenkel]

Hi,

I do not have enough points to read your document. I assume you are using HFSS. First, I would recommend to use Waveports with Driven Modal solution. Make sure you set enough modes for the port, otherwise you would get incorrect results.

Second, why are you approximating coaxials with Rectangles ? HFSS deals with true surfaces easily. You can even control the surface approxiamtion using mesh operations..

Third, I do not understand your boundary conditions ...
From the HFSS menu click Boundary Display (Solver view). Make sure the boundary conditions are as you expected. The default outer boundary condition is PEC (Perfect Electric Conductor).

Itai

 

[asdfaaa]

Hi,



Second, why are you approximating coaxials with Rectangles ? HFSS deals with true surfaces easily. You can even control the surface approxiamtion using mesh operations..



Itai

I am trying to answer this question to u. ^_^
when the size of the waveguide is electric large in terms of wavelength, the shape of the coaxial pin has less influence on the mode field distribution inside the waveguide. it's function is to provide the excitation.
And in HFSS mesh process, you can see the aspect ratio of the coaxial pin (if you use cylinder) is very bad. it will crazyly increase the number of unkonwns(time and memory consuming). So we can see lots of this kind of substistution of cylinder with rectangular (cubic) object with nice accuracy of simulation.
Best Regards,

 

[sandruska]

sry, figure missed

I'm sorry but I don't understand the figure. Where I would draw the port?. I need to draw the excitation in the rectangles of the figure (aproximations to coaxial).