While trying to run a parameter sweep in CST, I observed that for one particular value of the paramater I a getting some strange S21 characteristics. There are certain suden spikes in the curve at a particular frequency.
Could anyone clarify why this is happenning? Note that this happens here for one particular value of a parameter (gap = 0.13). For gap = 0.1 and 0.16 things seems okay. Is there any problem in my simulation procedure? Awaiting your valuable suggestions.
PS: I am trying to design a transition here from a planar patch structure to a Dielectric Waveguide.
S2(1),1(1) represents transmission to 1st mode in Waveguide port from Microstrip line port and S2(2),1(1) repesents that to the 2nd mode.
Hi!
I am quite sure, but the problem would have arisen because of improper selection of port, for that purpose first form the outer cylinder (vacuum) of radius slightly greater than your assigned radius and then select edge of outer cylinder for assigning ports. Second, what I saw, that your excitation is been given at some other line (microstrip line), not at your assigned port (may be, I am wrong), this can also create some problems. One more thing, gaps which you defined, have some physical significance in terms of electric field, magnetic field, fringing field effects and wavelength. So, it is also a problematic case, that's why you would be getting spikes. I will run your problem by own, then I will be able to discuss exact solution of your PS.
All The Best,
Ms Dwidevi,
Thank you so much for your suggestions. I am trying to check by varying my port dimensions too but have not received any positive outcome yet. Note that the 2 modes used at the dielectric waveguide port are also staying constant with frequency so that should not be a problem.
I would await further updates from your side when convenient and meanwhile keep trying to get an answer to the problem myself.
Hi Halder, in my opinion the problem doesn't seem to be related to port classification. Is has more to do with mode coupling and the gap size. To analyze this effect in more detail, analyze the fields at the frequency of the spikes. Plot fields at any normal frequency and comparison between the two will give an idea of what's happening.
Possible approaches you can try :
1. Make a SMOOTH transition between the feedline and rectangular patch.
2. Round the edges of the patch.
Cheers
Hi Halder,
Please try the following:
1- Try circular patch not rectangular as you do.
2- make inset feed to ensure all power delivered to the patch. I think there mismatch between the feeding line and the patch.
Thanks
Mr. Qureshi,
Thanks for your reply. I shall put field monitors and try to check the fields at the particular frequencies. Also shall check results by blending the edges.
Could you please elaborate a bit about the term "smooth transition" though? I did not understand the same fully.
And it seems like there is something related to the 2 modes itself as suddenly there's a slight dip in transmission to one mode and a sharp rise in the other according to the simulations.
- Badges
- Science topic