CST2013: Special Frequency Domain Solver Parameters 

Simulation: Solver Start Simulation Frequency Domain Solver Specials

Linear solver frame

Solver type: This option allows you to specify whether an iterative or a direct linear equation system solver should be used. In general, the direct solver is recommended for relatively small problems with multiple excitations. The iterative solver is advantageous for large problems. As a default, the frequency domain solver automatically chooses the solver type.

Use Helmholtz Equation: This option is available for the general purpose frequency domain solver with hexahedral mesh only. If activated, the Helmholtz equation is used when running the solver. This usually leads to a faster convergence of the solver, especially for low-frequency problems.

Low frequency stabilization: This option is available for the general purpose frequency domain solver with tetrahedral mesh only. If activated, it may lead to a faster convergence of the solver for low frequency problems.

Use accelerated recalculation: If activated, the general purpose frequency domain solvers store a subset of the results from frequency samples calculated so far to reuse this information for an accelerated recalculation of additional frequency samples in subsequent solver runs. Because a direct equation system solver does not profit from a good initial guess of the solution, this mainly applies to the iterative solver. However, you may try to calculate some frequency samples using the direct solver and continue with the iterative solver, thereby taking into account solutions from the direct solver to speed up the simulation.

Use new iterative solver: This option is available for the general purpose frequency domain solver with hexahedral mesh only. It is recommended to use the new iterative solver unless convergence problems are encountered.

Maximum number of iterations: Iterative solvers can be forced to terminate after the given number of iterations by enabling this option.

Solver order frame:

Allows to specify whether the general purpose frequency domain solver with tetrahedral mesh uses a first, second, or third order method. Second order is the default. If the structure is geometrically complex, first-order is an alternative. Higher order allows to achieve accurate results with less mesh cells and eventually less memory consumption than lower order, if the structure contains electrically large voids rather than many geometric details. For a given mesh resolution, higher order will provide even more accurate results.

Whenever the solver order is changed, for instance from second to first order, the resolution of the initial mesh and some parameters in the adaptive mesh refinement dialog should be adapted. These settings can be applied automatically. For a change back to a previously active order, the corresponding settings can be restored. A query will be displayed in either case. Furthermore, if a higher order is selected, some mesh and solver settings may be relaxed, if only a comparable level of accuracy is required. A prompt will ask if relaxed defaults for higher should be applied. Answer no only if you wish to increase the solver's accuracy by using higher order.

Variable: If this option is activated, the general purpose frequency domain solver with tetrahedral mesh is allowed to use variable order for each tetrahedron, rather than constant order throughout the calculation domain. The solver order's upper limit is then given by the order selected in the drop down combo box right above the Variable check box. Enable this option if the structure contains electrically small details as well as large voids. The solver then will assign an initial distribution of the solver order to the tetrahedrons, which eventually is changed in the course of the adaptive mesh refinement.

Curvature... It is recommended to increase the curved element order, which is a property of the mesh, if the solver order is increased. The button directly opens the special tetrahedral mesh properties where the Curved element order can be changed.

Materials frame

Activate the check button Fit as in Time Domain to consider the fit procedure of the tangent delta settings in the Material Parameter Conductivity Dialog using a dispersive Debye model correspondent to the time domain. A general or conductivity dispersion fit of tabulated data will also be considered as in the time domain solver. With this button deactivated, the constant fit option is realized as a constant tangent delta over the complete frequency range while a dispersive fit results in a linear interpolation between the defined loss angle values. Likewise, all other tabulated material properties are linearly interpolated.

The setting affects also the treatment of broadband surface impedance materials, i.e. ohmic sheet, tabulated surface impedance and corrugated wall (see the Material Overview for details). With this button activated the simulated surface impedance will be computed accordingly to the same fitting scheme used for the time domain solver. Otherwise a linear interpolation scheme of the data will be applied.

The setting also affects the treatment of dispersive materials if the model simplification is configured so that lossy materials are treated as lossfree.

Calculate static H-field for Ferrites allows to use an inhomogeneous static biasing H-field for ferrites. Set up a single project with material definitions and sources for the low and high frequency simulation. The static field is calculated automatically by the magnetostatic solver before the frequency domain solver run. This is supported by the frequency domain solver with tetrahedral mesh. See the Material Overview for details. An example can be found here: Biased Circulator.

Activate the check button Calculate power loss to include into the 1D Results tree folder information concerning the losses over materials. Both volume and surface losses will be extracted. Enable Store per solid to include a subfolder sorting losses by solids.

Open boundaries frame

The solver's realization of the open boundary can be chosen from the drop down box. Solver default is the recommended setting. Open boundaries for unit cells are implemented as a Floquet mode port, which is not listed here. For the usual open boundaries however, either simulation speed and low memory consumption (SIBC), or the low artificial reflection of the perfectly matched layer (PML) can be favoured.

With tetrahedral mesh, the default is the standard impedance boundary condition (SIBC.) It shows low artificial reflection for plane waves that impinge on the open boundary perpendicularly (the field solution of course is not necessarily a plane wave, but might be considered as the superposition of various plane waves at different angles.) For "plane wave" angles closer to grazing incidence, the PML provides lower artificial reflection, but at the cost of higher memory usage and more demanding linear equation systems, which require more time to solve.

With hexahedral mesh, the open boundaries are usually realized with a PML, but a free space SIBC, which is valid for vacuum background material may be chosen as a low-memory or low-frequency alternative.

The check button Add space before mesh generation increases the distance to the SIBC open boundaries beyond the visible bounding box of the structure (tetrahedral mesh only.) It is recommended to enable this option or to choose PML if farfield monitors have been defined. For weakly radiating and electrically small structures this additional space might not be necessary. If the option is off, please add space manually to ensure that the open boundary is not too close to the structure.

Waveguides frame

Activate the check button Electric shielding for all ports to let the boundary of all waveguide ports be treated as a perfectly shielding (PEC) wire frame.

The check button Ignore losses can be enabled to simplify the port mode calculation: all lossy materials and boundary conditions for waveguide ports will be treated as loss-free during the port mode calculation. This flag can be used independently of the treatment of lossy materials in the 3D solver run.

The option Port mesh and 3D mesh match only applies to the tetrahedral mesh and is disabled by default. The waveguide port mode solver operates on the planar mesh of the waveguide port and applies an adaptive port mesh refinement for this mesh. By default, the port mesh is separated from the volumetric mesh, and an overlap calculation is used to map the port mode solution onto the boundary of the volumetric mesh. However, if you enable Port mesh and 3D mesh match, the port mode solver's mesh adaptation directly refines the surface mesh of the 3D mesh, so that no overlap calculation is required. It is recommended to let port mesh and 3D mesh match whenever the overlap calculation fails due to geometric tolerance problems.

OK

Accepts the input and closes the dialog.

Cancel

Closes this dialog box without performing any further action.

Help

Shows this help text.