CST waveguide port dimensions
时间:03-26
整理:3721RD
点击:
Hey guys,
I have a planar microstrip structure on a piece of substrate and am simulating it within CST microwave studio.
There is a ground plane on the back and a strip on the top which the middle pin of a SMA connector would connect to.
From reading the CST help literature I assume that the waveguide port should be 6 to 10 times the width of the microstrip on top and should be 5 times the substrate height from top to bottom. I have also made the height a little bigger at the bottom so that it covers the ground plane.
The important dimesions are as follows:
Width of microstrip on top - 1mm
Thickness of substrate - 0.8mm
Thickness of microstrip - 0.03556mm
Therefore the box should be 6mm across and 4.03556mm high (starting at the bottom of the ground plane) right?
I have checked the port mode and it matches the fundemental mode example shown in the help literature.
I only want to check this is right as I have noticed the dimensions of the port can change the results as obviously I want accurate results.
Many Thanks!
I have a planar microstrip structure on a piece of substrate and am simulating it within CST microwave studio.
There is a ground plane on the back and a strip on the top which the middle pin of a SMA connector would connect to.
From reading the CST help literature I assume that the waveguide port should be 6 to 10 times the width of the microstrip on top and should be 5 times the substrate height from top to bottom. I have also made the height a little bigger at the bottom so that it covers the ground plane.
The important dimesions are as follows:
Width of microstrip on top - 1mm
Thickness of substrate - 0.8mm
Thickness of microstrip - 0.03556mm
Therefore the box should be 6mm across and 4.03556mm high (starting at the bottom of the ground plane) right?
I have checked the port mode and it matches the fundemental mode example shown in the help literature.
I only want to check this is right as I have noticed the dimensions of the port can change the results as obviously I want accurate results.
Many Thanks!
Yes, a good check is to see what the electric field is doing at the port. Does the electric field vectors/or 2D plot look sensible? Another check is to increase the dimensions of the port and compare the results/impedance to what you got before, is the error small, if so the first port size is fine.
In my opinion, I'd opt for a port to be a little larger than to be on the small side. Just lookout for warnings when the simulation runs to make sure that there are no other modes other than what you expect.
Also for microstrip, the bottom of the port must be in contact with the ground plane but does not need to go past it.