First tell me, will you get lumped/waveport in practice to give the excitation. The answer is simply no. Then why are you thinking about lumped/wave port. Dont use any one.
Now, the question is, then what to do. The answer is very simple. In practice, we use SMA/coaxial probe finally to give the signal. Then draw SMA/coaxial probe. At the end, you have to draw waveport, not lumped port. Therefore, in all cases, we have to use SMA connector/coaxial prove with waveport. There is no place of lumped port
The use of only lumped port/waveport using simple rectangular 2D surface, may give results, but there is no guarantee that the results are correct. If your assumptions are correct as per HFSS manual (depending on, thickness of surface, er etc), you may get results.
Hi, modeling of SMA coaxial connector which is 50 ohm, has been recommended. also, exciting of the whole structure with wave port for considering higher order modes is better.
Yes, this PDF is best to learn hfss, irrespective of any topic. Please open the pp. 200 ( PDF page no) and try to design probe fed microstrip patch antenna... SMA connector is there...
This manual is corresponding to HFSSv10 manual... If you use higher version of hfss, little variation may be observed to find the actual path, but almost 95% same...my friends did not face any problem in HFSSv17
Which port is better when solution frequency is of order of Terahertz range (where antenna size is much smaller than the available SMA connector size)?