hfss meshing
and i assign a special mesh operation (element length 1mm) for the trench only
and let the other objects in the design meshed with default scheme
but i'm not sure that if this is wrong or true for having more accurate results
can any one help me on how the mesh affect the design and what is the appropriate choice for accurate result (make a special mesh operation in some objects or use default meshing for all objects)
my design is attached for more detail.
thnks for ur interest
Hi walaa,
the meshing will definetly have an effect on your design. Restricting the mesh in certain sensitive regions of your design will usually improve the accuracy of your results, so long as its done correctly. One way you can check how good your mesh is, is by plotting the convergence of S11 (using adaptive pass instead of setup1). You should notice the value of S11 converge after a sufficient number of passes.
As a side note I noticed you used a lumped port to excite the microstrip. You might want to try using a waveport but read the HFSS manual to figure out the appropriate size of your waveport. This will also require you to place one side of your airbox in the same plane as the waveport as waveports cannot be placed inside the radiation boundary.
dear YoungEng
thnks for ur help
i was try to use waveports and also consider the appropriate size of port according to manual but i notice that lumped port is more easier for this problem because as i change dimensions of waveport (according to range in manual) it gives me different reult for each case .
so i want to ask u :does any problem exist for using lumped port for microstrip?
also i want ur help for if i use discrete sweep what is the appropriate solution frequncy must be assigned (the frequency at which the antenna was designed or the max frequncy of the sweep range) , as i know for using fast sweep we assign the solution frequency same as designed freq and also assign it in the middle of sweep range.
thnks again for ur interest
lumped ports can be used for microstrip, usually when the microstrip is "inside" your model, however if your ports are at the periphery then waveports is probably a better choice. This being said if you change your waveport size but keep it within the general bounds given to you by the manual then your solutions should not changed by much. If you do see a substantial change in your results then you should check and make sure you have run enough adaptive passes so that your results have converged.
When using the discrete sweep, the solution frequency should be the highest frequency in the band.
You can place the Waveport inside the radiation boundary aswell but in that case you have to use a PEC cap in order to save it by getting short circuiting!