I am working in Simlab and have a very simple cylindrical "air" capacitor. It's comprised of:
- Outer housing with ID = 6.00 inches and a thickness of 0.125 inches
- Solid inner conductor with OD = 5.00 inches
that's it, all suspended in air.
I am able to extract s-parameters from the solution parasitic extraction, however when they are plotted, the smith chart represents a series inductor, not a 2 port shunt capacitor.
My setup is as follows:
- Import step file and convert to parasoild via Simlab's import geometry tool
- Select solution: parasitic extraction; 1 MHz - 40 MHz w/ 1 MHz step; Single-ended port type
- assign material to each body: aluminum
- mesh each body. I used EM mesh Tet4 elements with size = 0.25, min size = 0.05, angle = 30 deg, grading factor 1.3. Total element count around 700k (probably too many, but I've tried a dozen mesh sizes, some finer, some more coarse; all give same S-param results)
- assign ports: single-ended, one on each end of the inner conductor; 2 ports in total
- Add the meshed bodies to the solution
- Leave default solver setting except stopping criterion tolerance = 1e-5
- Solve
When I plot the resulting S-parameters, the smith chart looks like an ideal series inductor instead of a shunt capacitor. When extracting capacitance values from the S-parameters, the capacitance is much lower than it should be (for this cylindrical air capacitor, hand calcs give about 70 pF, but S-param extracted capacitance values are around 20 pF.
Im not too sure where I'm going wrong here, but it's a simple case with minimal inputs, so I'm probably missing something very simple.
Please help.
Daniel P.
