This section explains how to step manually through the solution sequence.
You can specify five types of load step options for RSP method analyses.
These options specify how often the tangent matrix is updated during solution. Available options are:
Program-chosen (default)
Full
Modified
Initial-stiffness.
For a 3D analysis, the full Newton-Raphson option is recommended with adaptive descent. To specify Newton-Raphson options, use either of the following:
In addition to defining SOURC36 elements to describe the currents in a model, you have a symmetry reflection option. If you have a current source that exhibits circular symmetry about the global Cartesian Z axis (such as a coil), you can define just one sector of it with SOURC36 elements, using the following:
Then, you can duplicate the current source via the EMSYM command or . You specify Biot-Savart calculations using either of the following:
By default (that is, if you do not specify Biot-Savart calculations), the HS contribution from all selected source elements to all selected nodes is calculated when you initiate the solution process. The program does no further HS calculation during solution.
You use general options mostly in nonlinear static analyses. For a 3D static magnetic analysis, the only general option is the number of substeps (the time step size). See Using the Alternative Solution Option for 2D Static Magnetic Analysis for details on specifying the number of substeps.
A nonlinear analysis requires multiple substeps within each load step, so that the specified loads can be applied gradually to obtain an accurate solution. By default, the program uses one substep per load step. For the RSP method, one substep (default) is recommended.
The nonlinear load step options are:
Number of equilibrium iterations.
This option obtains a converged solution at each substep. The default is up to 25 equilibrium iterations per substep. However, you may have to increase the number, depending on the degree of nonlinearity.
Convergence tolerance.
The program considers a nonlinear solution to be converged when specified convergence criteria are met.
You specify a convergence criterion by specifying a value and a tolerance for that value. Thus, value * tolerance gives the convergence criterion, for example, if you specify 5000 as the typical value of magnetic flux and 0.001 as the tolerance, the convergence criterion for magnetic flux would be 5.0. Ansys, Inc. recommends that you leave the value to default (program-calculated) and set the tolerance to 1.0E-3 for flux-based convergence.
You can base convergence checking on magnetic potential (MAG), magnetic flux (FLUX), or both. For potentials, the program compares the change in nodal potentials between successive iterations Δ Φ = Φ1 - Φi-1 to the convergence criterion.
For magnetic flux, the program compares the out-of-balance load vector to the convergence criterion. In the model does not converge within the specified number of equilibrium iterations, the program either stops or moves on to the next load step, depending on whether you have specified the option to halt an unconverged solution. Convergence based on magnetic flux is recommended for a 3D, scalar potential, magnetostatic analysis.
Mechanical APDL enables you to graphically track convergence norms via its Graphical Solution Tracking (GST) feature, available for both interactive and batch runs of the program. For a detailed description of this feature, see Tracking Convergence Graphically and the /GST command description in the Command Reference.
For details on specifying nonlinear options, see Using the Alternative Solution Option for 2D Static Magnetic Analysis.
To solve the load step, use one of the following:
Repeat Steps 1 through 3 to specify additional load steps, if any.
See 3D Static Magnetic Analysis (Scalar Method) for a discussion of postprocessing results from a 3D static magnetic analysis (scalar potential formulation).
An example command sequence for a nonlinear 3D RSP method static analysis follows:
/solu nropt,full,,on ! Full Newton-Raphson, adaptive descent cnvtol,flux,,le-3 ! Convergence criteria solve ! Solve finish