SigFit - Interpolation




Results from SigFit may be interpolated to an array or a new finite element mesh. Arrays data may also be interpolated to the finite element mesh as input for analysis in SigFit.

 


A very powerful feature of SigFit is interpolation of finite element results to either an array or to another finite element mesh. Interpolated arrays are often used as the analysis prediction format to compare to optical test data. Arrays are also used to describe finite element deformation predictions which are not well represented by a reasonable set of zernike polynomials such as quilting in a lightweight mirror. Interpolation to a new finite element mesh is often used to apply optical test masks or to limit analysis to only those regions of each optical surface that are used by the ray bundle from a specific field point.

Data to Interpolate
To perform interpolation the user may select one of the Interpolation Data picks as shown in the form to the right. The choices to interpolate include unmodified displacements as read from the finite element results, displacements after modification with the radial displacements, displacements after actuation through adaptive control, and optical path difference due to temperature or stress optic effects.

Output Format
The output format is then chosen from four array formats or a user supplied set of grids. Available formats include Zygo/CODEV/OSLO, OASIS, and ZEMAX (formatted and unformatted).

Displacement Direction
The user can chose to interpolate displacements oriented parallel to the optical axis or normal to the optical surface. This choice is very important since the array formats for different optical analysis tools require one of these two choices.

Interpolation Polynomial Order
If nodal rotations are available in the finite element displacements then the user may choose cubic interpolation. However, if nodal rotations are known to be constrained, such as in a solid element model, linear interpolation may be chosen.

Array Definition
For array output the array is defined by the number of points on each side of the array, the starting locations in the X and Y directions, and the ending locations in the X and Y directions. Finally, for array output the user may chose to write visualization files so that the interpolated array may be viewed without the use of the optical analysis software. The array is written as a finite element model and nodal result file consitent with the user selectable finite element software (NASTRAN, ANSYS, or COSMOS) and post-processing software (MSC/PATRAN, FEMAP, COSMOS, or ANSYS).

New Grid Set Definition
For interpolation to a new set of grids the user simply specifies the file in which the new grids are contained and the range of grid IDs for which interpolation should be performed as shown in the form to the right. Interpolation output to a new set of grids is formatted in the result output format of the user specified finite element software. This allows the user to subsequently perform any of the analysis capabilities on the newly interpolated data as if it had originally come from the finite element analysis.


In addition to allowing arrays as an output format, SigFit also allows the user to describe optical surface deformations with arrays as an input format. This is useful for importing test data into SigFit to fit Zernike polynomials or evaluate how correctable it is with adaptive control.

After selecting the array data file, the user defines how the array is formatted by the form shown to the right. Vizualization files similar to those written for output arrays may be requested.


Sigmadyne, Inc. 803 West Ave. Rochester New York
Copyright © 2003 Sigmadyne, Inc. All rights reserved.