Home/Products/Ansys Lumerical/MODE
ANSYS LUMERICAL
Three solvers: FDE, varFDTD and EME. From mode analysis of a waveguide cross-section to propagation analysis of planar integrated photonic circuits and long tapers, it efficiently handles the problems where 3D FDTD becomes expensive.
Overview
FDE (finite-difference eigenmode) solves the waveguide cross-section as a two-dimensional eigenvalue problem, giving effective index, mode profile, dispersion and propagation loss. Because no propagation calculation is needed, it is extremely fast.
varFDTD reduces a 3D structure to a 2D effective index distribution and analyzes the in-plane behavior with 2D FDTD. It suits planar structures where coupling between different vertical slab modes is weak, and handles propagation in every in-plane direction without assuming an optical axis.
EME (eigenmode expansion) divides the structure into sections of uniform cross-section and connects the modes of each. It handles bidirectional propagation. Once the mode decomposition is done, the propagation length can be changed without redoing the mode calculation, which makes length sweeps and optimization efficient.
How MODE’s solvers are arranged
01
Evaluates the eigenmodes of a waveguide cross-section, with effective index, loss and dispersion.
02
Efficient 2.5D analysis of in-plane propagation in planar integrated photonic circuits.
03
Sweeps propagation length efficiently without redoing the mode calculation.
Where it fits
For some problems another method solves faster and more accurately. The following is a guide to choosing.
For structures where coupling between different vertical slab modes is strong, or where out-of-plane radiation and propagation matter, the effective index approximation in varFDTD may not be appropriate. In that case consider EME or 3D FDTD, depending on the problem.
Use MODE when you want to
Consider another method
Choose between MODE’s three solvers and the solvers in other products according to the problem and its scale.
Typical applications
Edge couplers and spot size converters
Inputs and outputs
Set the cross-sectional geometry, materials, wavelength and device structure, and evaluate mode characteristics, propagation characteristics, S-parameters and more.
INPUT
| Cross-sectional structure | Layer stack, width, thickness, materials |
| Wavelength | A single wavelength, or a sweep range |
| Number of modes | How many eigenmodes to compute |
| Propagating structure | The in-plane layout for varFDTD, or the change in structure along the propagation direction for EME |
MODE
Waveguide analysis
OUTPUT
| Effective index | n_eff and group index for each mode |
| Mode profile | Field profile on the cross-section |
| S-parameters | Transmission and reflection between ports |
| Loss | Propagation loss, bend loss, coupling efficiency |
Analysis workflow
In MODE, what matters is choosing the solver that suits the problem and correctly setting up and confirming the mode or propagation characteristic you are after.
01
Build the cross-section of the waveguide or fiber and assign materials.
02
Use FDE for cross-sectional modes, EME for long propagation such as tapers, and varFDTD for planar propagation in in-plane devices.
03
Take a region large enough to contain the evanescent components, then set the number of mesh cells or the maximum mesh step.
04
In FDE, set Search near n / in range and Number of trial modes appropriately, and confirm from the mode profiles that the mode you want is in the Mode List. For a wavelength sweep, select the mode of interest and use Track selected mode to follow the same mode.
05
In FDE, check effective index, loss, mode profile and dispersion; in varFDTD, field distribution, transmission, mode coupling and S-parameters; in EME, transmission and reflection between ports and the S-matrix. Where useful, FDE waveguide characteristics or varFDTD and EME S-parameters can be used in a later stage such as INTERCONNECT.
Frequently asked questions
Questions we are often asked before adoption.
How closely do varFDTD and 3D FDTD results agree?
There is no single figure for agreement. varFDTD reduces a 3D structure to a 2D effective index distribution, so it is particularly effective for planar structures where coupling between different vertical slab modes is weak. Under those conditions it can agree well with 3D FDTD. Where coupling between slab modes is not negligible, or where out-of-plane propagation and radiation matter, the varFDTD approximation may not be appropriate, so verify with 3D FDTD where needed.
When does EME have the advantage?
It suits structures with long propagation distances, such as long tapers, spot size converters, MMIs and periodic devices. Once the mode decomposition is done, the propagation length can be changed without redoing the mode calculation, which makes length sweeps and optimization efficient.
How does this differ from the beam propagation method?
BPM in general uses the slowly varying envelope approximation and presumes a dominant propagation direction. varFDTD assumes no optical axis in-plane and handles propagation in every direction, while EME analyzes propagation including reflection through a bidirectional eigenmode expansion. Which of these solvers to use depends on the problem, for instance high index contrast or long waveguide structures.
Can I import a beam from Zemax?
Yes. Beam data from Zemax OpticStudio can be brought into MODE via the ZBF format. In FDE you can evaluate the overlap with a mode and the power coupling, and in EME the ZBF data can be imported directly into a Port. MODE results can also be returned to OpticStudio as ZBF.
We can advise on everything from simulation setup to judging whether a result is sound. Hands-on training is also available.
Support →
Training →
You do not need to share everything. Whatever you are able to tell us is enough.
The demo uses a structure close to your own. We can also give you an idea of the computational scale and the time it takes.