caiyu/openEMS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

PML: try to fix pml working for a finite conductor waveguide

Browse Source 
sigma > 1000 S/m is considered a conductor (not ideal solution)

Signed-off-by: Thorsten Liebig <Thorsten.Liebig@gmx.de>
This commit is contained in:
Thorsten Liebig 2023-03-08 22:10:30 +01:00
parent cb63ab01c4
commit 568cdbdfac
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
FDTD/extensions/operator_ext_upml.cpp
View File

@ -404,7 +404,9 @@ bool Operator_Ext_UPML::BuildExtension()
CalcGradingKappa(n, pos,__Z0__ ,kappa_v ,kappa_i); CalcGradingKappa(n, pos,__Z0__ ,kappa_v ,kappa_i);
nP = (n+1)%3; nP = (n+1)%3;
nPP = (n+2)%3; nPP = (n+2)%3;
if ((kappa_v[0]+kappa_v[1]+kappa_v[2])!=0) // if eff_Mat[1] > 1e3 assume a metal and disable PML to continue a signal layer
// sigma > 1000 S/m is "very lossy" already...
if (((kappa_v[0]+kappa_v[1]+kappa_v[2])!=0) && (eff_Mat[1]<1e3))
{ {
//check if pos is on PEC //check if pos is on PEC
if ( (m_Op->GetVV(n,pos[0],pos[1],pos[2]) + m_Op->GetVI(n,pos[0],pos[1],pos[2])) != 0 ) if ( (m_Op->GetVV(n,pos[0],pos[1],pos[2]) + m_Op->GetVI(n,pos[0],pos[1],pos[2])) != 0 )