Coaxial wave guide

[zeljkoT]

Hi

I would like to adapt the test case "mgdyn_wave_td" from a rectangular to a coaxial wave guide.

Therefore I have following questions:

• Perfect Electric Conductor BC:
  What is the Elmer definition for a perfect electric conductor (n x E = 0) boundary condition? From the test case I understand it should be:
  

  Code: Select all

  a = real 0.0
  a {e} = real 0

  right?

• Input Port Definition:
  
  I would like to define the excitation to be a radially decaying field on the input port surface oscillating with cos(1*t).
  How can I access the coordsX, coordsY and coordsZ mesh components in MATC to define the Ex, Ey and Ez components of the input port field?
  
  Following expression creates such vector field in Paraview for a 2D disk source:

  Code: Select all

  1/(ln(0.107/0.04)) * 1/mag(coordsX*iHat+coordsY*jHat) * (coordsX*iHat+coordsY*jHat) / mag(coordsX*iHat+coordsY*jHat)

  Evec.jpg

  (183.09 KiB) Not downloaded yet

• Ouput Port Definition:
  On the output port, I would like to define the BC to match the wave impedance of the coaxial structure Zw (no reflection):
  
  Zw = sqrt(L' / C') = sqrt( 196.7e-9 / 56.54e-12 ) = 58.982 ohm
  
  What is the corresponding BC?
  

  Code: Select all

  Electric Damping Coefficient = Real 58.982

  correct?

Thank you very much in advance for help.

Cheers

[kevinarden]

I modified the test case to use coordinates x, y, and z.
No mag function in MATC, may consider writing a user function.

case.sif

(2.12 KiB) Downloaded 19 times

[zeljkoT]

Thx for your reply.

It helped me make progress and it's looking quite good already.

Just to ensure everything is covered (for others facing similar issues):

• Perfect Electric Conductor BC:
  BC for perfect electric conductor is
  

  Code: Select all

  a = real 0.0
  a {e} = real 0

• Input Port Definition:
  
  As suggested in your response, the input port can be defined using MATC. The magnitude function can be reformulated to be used in MATC.
  
  In this example, the field is defined to be radial in the y-z plane, and the wave travels in the x-direction. The center of the input port for this example is assumed to be at (0,0), but if it's located elsewhere, it should be accounted for in the port equation.

Question:

The excitation is an electric field oscillating with cos(1*t). The very first half wave (tip of traveling wave) is oscillating weirdly but I don't understand why. Any suggestion for initial conditions?

1.png

(742.88 KiB) Not downloaded yet

Thx in advance.
Cheers