Far Field Source File in CST MWS

[hookups1534]

Hello all,

I am trying to do something very simple and cannot seem to get it to work. All i wanna do is make a far field source file for a simple cos^q(theta) pattern. What i need to do is to make the far field source file then use it as an excitation for a I-solver simulation. Please, anyone who understands the formatting of the far field source file respond to this post. The format is not as simple as the following cause i tried that and it did not work:

Re(E_Theta) = cos^q(theta)*sin(phi)
Im(E_Theta) = 0
Re(E_Phi) = cos^q(theta)*cos(phi)
Im(E_Phi) = 0

I got these from the equation for the far field of a cos^q which is:
E(theta, phi) = cos^q(theta) * [sin(phi)* theta_hat + cos(phi) * phi_hat]

This has no imaginary part but when you look at the far field source files generated from CST, they always have an imaginary part, i think that the way that I think that the file expects the theta and phi components is wrong, please anyone who understands the file format or anyone who can help or has alrady made a far field source file ever before. Thanks again guys.

 

[e_m_c]

Hi, I never used this feature in CST, so sorry could not directly help you. But I am using another Method of Moments solver (EMCoS TriD), where it is also possible to specify Radiation Pattern Source. Attached is User's Manual - please take a look on chapter 3.5.11 Pattern source block.

You can see also general format description in attached picture.
If you could generate your needed source in this format I could test how TriD solver can handle such a source.

Is your source frequency dependent? In this case format is just little bit different - you need to specify similar block for each frequency.

 

[hookups1534]

Hello,

Thanks so much for your reply, i cannot tell you how good it feels to talk to someone about this problem i have been ripping my hair out to get this to work for the last 4 days now!!! and its just a stupid cos^q(theta) pattern, I couldn't even get a simple dipole to work and dipole is just sin(theta). Well, what would really help me is if you could generate a file with a pattern of cos^q(theta) in spherical coordinates for just a few theta and phi samples, no more than 5 or so for each. Then i can reverse engineer the way that the file is generated and apply the concepts to my file. My file is the same format as yours except my pattern values for E_Theta and E_Phi are in Re( ) and Im( ) part and yours are just Mag( ) and Ph( ). So, if i have a file with the correct pattern in your tool, then i can just convert the values from polar to re/im and then try the file in CST. Please let me know if you could do that for me. Thanks again for your post, i am hurtin over here with this.

 

[e_m_c]

I'll try to generate required pattern. What are the q values you are interested in? *Sorry, this is not really my specialisation, I might ask trivial questions*.

I took a look on paper 'Realizable Feed-Element Patterns and Optimum Aperture Efficiency in Multibeam Antenna Systems' by K. S. YNGVESSON, J. F. JOHANSSON,
Y. S. KIM, Y. RAHMAT-SAMII, FELLOW - there they consider values of q as a quadratic function of d/Lambda, i.e. K(d/Lambda)^2 - 0.5.

 

[hookups1534]

Thanks so much, we are so fortunate to have people willing to help eachother out like you, thanks again. Okay, so the q values i will need are 0.396, 3.4879, and 8.199 but i only need one of these to start, the easiest will prob be the 8.199. Thanks again, let me know if there is any other information you need.

 

[e_m_c]

See attached archive with matlab codes and Radiation Pattern Source files for EMCoS EMC Studio. Is this what you expected?

What you are going to excite by these patterns?

 

[hookups1534]

Wow, that is exactly what i expected, thanks so much, i really appreciate this. I should be able to go from this file to the file for my program. Thanks again, I will let you know how it goes but once again, thanks so much, I am going to give you a thank you but is there any other way to give you some points or something?

Added after 3 hours 51 minutes:

It worked, my problem was always assuming the imaginary part is 0, when in actuality it will not becuase of the sign involved with the real part. Thanks again, problem solved!!

Also, I really really really appreciate the matlab script to generate the pattern files, thanks again!

 

[e_m_c]

Glad to hear this. Keep the topic updated if you have further issues.