HFSS vs. Momentum accuracy
Another question regarding run time on HFSS , I also read about it a lot yet when I build my own 3D geometry something Made my simulation slower than when I run others maybe it's due to boundaries on the air box
Near ports I use radiation and other faces finite conductivity, maybe it's a wrong way ...
Maybe my drawn geometry e.g. branch line coupler is not clean enough?
No. Momentum uses a different method to solve Maxwell's equations (Method of Moments) that is more efficient for such planar models. HFSS uses a more generic method.
So in such case it's better to work with Momentum? Yet I have seen difference in results , a frequency offset and different S params magnitudes how would I know which solution is more valid?
In my experience: yes, because it is more efficient and has less chances for mistakes in modelling.
That indicates that one or both models are not accurate enough. This is not a tool problem, it is incorrect use of the tool. All these EM tools require a skilled user for accurate results. From my experience, HFSS and other 3D tools offer more possiblities for making mistakes in the model.
From more than 20 years work in EM simulation: The biggest error source in EM simulation is the user. An experienced user will have very similar results from both tools for planar structures, but it takes time to learn using the tool.
I agree , I so want to not make mistakes, in HFSS the most issue for me is the right definition of ports and boundaries for each situation
In Momentum I'm not sure I'm doing right all the pec box and what if I'm not using it at all around the model? and should I use and draw line for drawing lines or I can draw rectangulares , should I use RF simulation or Momentum , should I use dot port or edged , should I put the port on the edge line or in some dot inside the line etc. many little questions that there no where to find the answers...
I'm not sure what you mean. Momentum uses OPEN boundary condition on the sides. For a PEC boundary at the top or bottom, you can set that boundary in the Momentum substrate definition.
Whatever you prefer, all is possible when drawing in the layout editor: lines with width or polygons or rectangles or ... You are not limited to the models from the palette, so you can draw whatever you like, and the EM solver will analyze that drawn layout.
Momentum RF mode assumes there is no radiation. If you have radiation, using Microwave Mode is required. It is never wrong to use Microwave mode, that is the more general solution. RF mode is faster because it can re-use some data.
Sometimes that makes a difference, so you need to understand the difference:
https://muehlhaus.com/support/ads-ap...edge-area-pins
If a RF Structure has 2D or 2.5D, Momentum ( or equivalent ) is always faster and accurate.HFSS is a General Purpose real 3D FEM Simulator but it's slower and memory/cpu eager and it's not necessary to use it for 2.5D planar structures.
Thank you all, what do you mean 2.5D?