This is comparison of two test cases: mgdyn_harmonic_wire + mgdyn_harmonic_wire_impedanceBC2


Data from 3D wire model (mgdyn_harmonic_wire + f=1.0e6):

Variables in columns of matrix: ./s.dat
   1: boundary int: current density im e 3 over bc 3
   2: boundary int: current density re e 3 over bc 3
   3: boundary mean: av re over bc 3
   4: boundary mean: av im over bc 3
   5: res: eddy current power
   6: res: electromagnetic field energy

   1.644427986500E-002  -8.871722206343E-004   5.542000000000E-005  -3.541809650383E-027   4.903526786027E-008   4.557897127464E-013

Data from impedance boundary layer model

Variables in columns of matrix: ./g.dat
   1: boundary int mean: surface current re 3 over bc 1
   2: boundary int mean: surface current im 3 over bc 1
   3: res: eddy current power
   4: res: electromagnetic field energy
   5: res: surface current power

Three densities with Element Divisions 1(3) = 8, 16, 24

 -2.733644555514E+000   2.478774650999E+000   0.000000000000E+000   4.304331957857E-013   4.445897435804E-008
 -2.720890739035E+000   2.425768289278E+000   0.000000000000E+000   4.306177181125E-013   4.338292942813E-008
 -2.654393876621E+000   2.505441808254E+000   0.000000000000E+000   4.318517226198E-013   4.349864525289E-008


Power dissipation:
s(5)/g(5) = 0.90667

Imaginary current (R=0.001):
s(1)/(2*pi*R*g(2)) = 1.0558

Real current:
s(2)/(2*pi*R*g(1)) = 0.051652


Conclusions:
1) Imaginary current difference ~5%
2) Power dissipation difference ~10% (quadratic)
3) Results are robust with mesh resolution
4) There is something fishy with the real current densities...




 
