Author |
Topic |
|
rick
2 Posts |
Posted - Dec 22 2023 : 10:56:21
|
Should special attention be paid to the selection of reference points in the dielectric region? For a closed region, selecting the reference point inside the area can yield accurate results. If chosen outside the region, the results would be inconsistent. tmpref = DotProd( ( (**itc)->m_clsDielRef3DPoint - newpanel->m_clsVertex[0]), newpanel->m_clsNormal ), |
|
Enrico
543 Posts |
Posted - Dec 23 2023 : 16:59:34
|
Yes, there is no notion of topology. All dielectric panels are tested one by one against the reference point to define what is 'in' and what is 'out' or, if you want, given the plane on which a panel lies and the corresponding two half spaces in which the plane divides the space, which half-space contains the reference point.
This is the reason why in FasterCap you can define a reference point on a per-panel basis, so you actually need only to know the panel normal and you can easily calculate its reference point - valid for that panel only! - for when the panel is used to represent part of a dielectric interface.
You also have the option to split the dielectric interface in multiple parts and define a reference point for each of them. If the parts are convex, defining a reference point can be done in a unique way.
Best Regards, Enrico
|
|
|
|
Topic |
|
|
|