EIGRL

Bulk Data Entry

EIGRL – Real Eigenvalue Extraction Data, Lanczos Method

Description

Defines data required to perform real eigenvalue analysis (vibration or buckling) with the Lanczos Method.

Format

(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
EIGRL	SID	V1	V2	ND	MSGLVL	MAXSET	SHFSCL	NORM

(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
EIGRL		0.1	3.2	10

Field	Contents
SID	Unique set identification number. No default (Integer > 0)
V1,V2	For vibration analysis: Frequency range of interest For buckling analysis: Eigenvalue range of interest See comments 3, 4, and 10. Default = blank (V1 < V2, Real, or blank)
ND	Number of roots desired. See comments 3 and 4. No default (Integer > 0 or blank)
MSGLVL	Diagnostic level. Default = 0 (Integer 0 through 4 or blank)
MAXSET	Number of vectors in block or set. Default = 8 (Integer 1 through 16 or blank)
SHFSCL	For vibration analysis: Estimate of the frequency of the first flexible mode. For buckling analysis: Estimate of the first eigenvalue. See comment 9. Default = blank (Real or blank)
NORM	Method used for eigenvector normalization. See Comment 2. MASS: Eigenvectors are normalized to the unit value of the generalized mass (this is not a valid option for linear buckling analysis). MAX: Eigenvectors are normalized to the unit value of the largest displacement in the analysis set. MAXT: Eigenvectors are normalized to the unit value of the largest translational displacement in the analysis set. Default = MASS for normal modes analysis Default = MAX for linear buckling analysis (MASS or MAX)

Comments

1.	In vibration analysis, the units of V1 and V2 are cycles per unit time. In buckling analysis, V1 and V2 are eigenvalues. Each buckling eigenvalue is the factor by which the prebuckling state of stress is multiplied to produce buckling in the shape defined by the corresponding eigenvector.

In vibration analysis, eigenvectors are normalized with respect to the mass matrix by default. In buckling analysis, eigenvectors are normalized to have unit value. NORM = MASS is not a valid option for linear buckling analysis. If NORM is set to MASS for linear buckling analysis, OptiStruct automatically switches to MAX normalization.

3.	The roots are found in order of increasing magnitude: that is, those closest to zero are found first. The number and type of roots to be found can be determined from the following table. In vibration analysis, blank V1 defaults to -10.

V1	V2	ND	Number and Type of Roots Found
V1 V1 V1 V1 blank blank blank blank	V2 V2 blank blank blank blank V2 V2	ND blank ND blank ND blank ND blank	Lowest ND or all in range, whichever is smaller. All in range Lowest ND in range [V1, + ¥] Lowest root in range [V1, + ¥] Lowest ND roots in [-¥,+¥] Lowest root. Lowest ND roots below V2 All below V2

Number and Type of Roots Found

blank

Lowest ND or all in range, whichever is smaller.

All in range

Lowest ND in range [V1, + ¥]

Lowest root in range [V1, + ¥]

Lowest ND roots in [-¥,+¥]

Lowest root.

Lowest ND roots below V2

All below V2

4.	The Lanczos eigensolver provides two different ways of solving the problem. If the eigenvalue range is defined with no upper bound (V2 blank) and less than 50 modes (ND < 50), the faster method is applied.

5.	Eigenvalues are sorted in the order of magnitude for output. An eigenvector is found for each eigenvalue.

6.	In vibration analysis, small negative roots are usually computational zeros, indicating rigid body modes. Finite negative roots are an indication of modeling problems. If V1 is set to zero explicitly, V1 is ignored. It is recommended that V1 not be set to zero when extracting rigid body modes.

7.	MSGLVL controls the amount of diagnostic output during the eigenvalue extraction. The default value of zero suppresses all diagnostic output. A value of one prints eigenvalues accepted at each shift. Higher values result in increasing levels of diagnostic output.

8.	MAXSET is used to limit the maximum block size in the Lanczos solver. It may be reduced if there is insufficient memory available. The default value is recommended.

9.	A specification of SHFSCL may improve the performance of a vibration analysis. It may also be used to improve the performance of a buckling analysis, especially when the applied load differs from the first buckling load by orders of magnitude.

10.	AMLS and AMSES eigensolvers require that V2 be specified.

11.	This card is represented as a loadcollector in HyperMesh.