The distance between the PML boundary condition and the waveguide

afshiinafshiin Member Posts: 7


I have a question. 

I want to know how much the distance between the PML boundary conditions and the waveguide?

I wanted to know the optimal distance for simulation. for example FDTD, VarFDTD and FDE. 

 If the distance is large, the volume of the simulation area increases. Therefore the simulation time increases. Otherwise, if the distance between the boundary conditions and the waveguide is small, there is a possibility of error in the results.

Please guide me. 


  • ShelleyShelley Posts: 15Member

    If the figure you draw contains substrate, it may be easy to understand. Generally speaking, the distance between PML boundary and structure is half a wavelength at least. In addition, the PML boundary below should cut off substrate. FDE should be big enough to avoid mode cut-off.

  • afshiinafshiin Posts: 15Member

    Thanks for your answer.

    Yes, according to your answer, I have put the substrate in the figure. Now I have specified the distance in the figures. Please determine that they are drawn correctly?

    Besides, I have two questions. 

    1- If the wavelength range,(1530 nm - 1560 nm), is considered instead of a wavelength, should be the distance from the boundary condition to the waveguide at least half the wavelength greater? 

    For example in wavelength range (1530 - 1560), distance <= 1560/2.

    2- Also, in the figure related to the cross-section in the coordinates (y-z), if the upper boundary condition is the metal type, what should be the distance boundary condition metal to the waveguide? Is it possible to consider the PML boundary condition for metal? 

  • kjohnsonkjohnson Posts: 218Ansys Employee

    Hello @afshiin ,

    This topic is discussed in the post here:

    Here is the relevant section:

    Simulation span

    The span of the FDTD region should be set such that PML boundaries are a half wavelength away from the sides of any geometry objects in the simulation. “Wavelength” here refers to the longest wavelength in the source spectrum, taking into account the refractive index of the material between the object and the boundary. Exceptions to this rule include substrates, cladding, or any other objects that are supposed to extend beyond the simulation region (for example, the ends of input/output waveguides). For example, in a simulation of a simple straight waveguide in the Z direction:

    XY cross-section

    • Substrate extends through the X max/min and Y min boundaries.

    • X max/min and Y max PML boundaries are at least a half wavelength away from the sides of the waveguides.

    XZ cross-section

    • The ends of the waveguides extend through the Z max/min boundaries.

    To answer your questions:

    1) You should consider the longest wavelength in the spectrum when setting the simulation span.

    2) The rule of thumb for metal boundaries should be the same as the rule for PML boundaries.

    Also, note that these are just guidelines. The final simulation spans should be determined using convergence testing.

  • kjohnsonkjohnson Posts: 218Ansys Employee
Sign In or Register to comment.