How To Use PCHScan
NOTE: The program described here is available only to Goddard Civil Servants and Contractors.
The program PCHSCAN was written by J. Bolognese and M. McGinnis to assist in determining whether a model passed the unit enforced displacement check. It is located on "apollo" in directory /home/jeff/bin. Type pchscan to run the program.
-
The user must use the "PUNCH" option on all results output requests but does not require a sorted bulk data file. For example:
DISP(PUNCH) = ALL
- The program requires that all displacements be in the same coordinate system as that which was used to apply the unit displacement. This does not have to be the basic coordinate system.
- As noted above this will require separate runs for checking displacements and all other properties on model with multiple output coordinate systems.
PCHSCAN does not require that all 6 DOF be checked in one run, or that the subcases be in any particular order. The most common arrangements of subcases are available as menu picks or the user may define their own sequence.
PCHSCAN works by reading through the punched results and disregarding all acceptable results. In most cases the expected values are zero but few models will produce perfect zero values. In practice, and in PCHSCAN, the following values are considered to be acceptable. The user may adjust these filter values by altering the parameters shown in the table. They are easily found in the beginning of the source code for PCHSCAN.
TABLE 1: PCHSCAN "Zero" Values
Result Type | Acc. Value | Parameter |
SPC FORCES | 1.0E-10 | SPC_TOL |
DISPLACEMENTS | 1.0E-06 | DISP_TOL |
GRID POINT FORCES | 1.0E-05 | GFB_TOL |
ELEMENT FORCES | 1.0E-01 | ELF_TOL |
ELEMENT MOMENTS | 1.0E+00 | ELM_TOL |
Displacements are reported if they vary by greater than +/- DISP_TOL. Those displacements, rotations, forces and moments which exceed the acceptable values are reported to the user. Since a single node or element has multiple possible errors of each type (the six DOF or various stress components) the quantity of normalized error is used to report errors. It is simply the average error for those quantities checked. It should be used to alert the user to those areas which are the likely source of the problem.
Sandra Irish
Back to the Enforced Displacement Page
Back to the FEM Validity Checks Page
|