White Papers Abstracts

White paper abstracts and links to the full white papers (PDF format) provide valuable information on the latest stress analysis technologies and techniques. Click here for complete list of white papers.

Optimizing the Performance of NEi Nastran

Abstract:
NEi Nastran has built-in logic that automatically selects the best solver and settings for optimal performance when running an analysis. For most analysts, the default settings are recommended and will provide good performance for a wide range of problems. However, when running very large models or models that take a long time to solve, there are certain settings that may improve solution time.

Click here to view entire white paper.

Dynamic Design Analysis Method (DDAM) Handbook

Abstract:
During WW II, the US and British Navies found that ships often sustained extensive damage to onboard equipment by bombs and torpedoes that had not hit the ships directly, but had detonated nearby. Research into the causes of this phenomenon revealed that the explosions were generating a shock wave under water. It was this large magnitude, fast traveling, exponentially decaying pressure pulse that was causing much of the damage. Further research after the war served to characterize the shock wave more clearly, and the US Navy developed a series of machines to test potential equipment. These shock machines attempted to simulate the loading that shipboard equipment would see in a near-miss shock event, and most equipment intended for Navy service was required to submit to this testing.

Click here to view entire white paper.

Utilization of NEi Nastran's Nonlinear Restart Feature

Abstract:
NEi Nastran’s nonlinear restart capability is examined using a plate with a single load and a nonlinear static solution. The plate is first loaded and assigned to carry out 10 nonlinear increments, and all 10 incremental nonlinear database files (*.TDB) are saved. A restart is implemented at 50% of the load on the same model using the *.TDB import capability – a new feature in NEi nastran V9.1.

Click here to view entire white paper.

Composite Plates with Two Concentric Layups Under Compression

Abstract:
A new concept for designing composite panels with improved performance under compression is presented. In this concept, the panel consists of two different concentric layups. A Rayleigh-Ritz-based approach to model such rectangular panels under compression is presented. The buckling load and the in-plane stresses everywhere in the plate are determined using an energy minimization approach. The results are compared to detailed finite element models and, for special cases, to other published finite element solutions and are shown to be in good to excellent agreement except for cases where twisting-bending coupling (not accounted for by the present method) is significant. As shown by specific examples, the present method can be used to obtain lighter configurations than single-layup geometries. In addition, the method can be applied to plates with rectangular cutouts and plates with terminated stiffeners.

Click here to view entire white paper.

NEi Nastran V9.1 x64 (64-bit Solver)

Abstract:
In a 32-bit operating system such as Windows 2000 or Windows XP there is a limitation of 232 = 4GB of memory that can be addressed. Windows reserves 2GB for the operating system leaving only 2GB for external programs (such as NEi nastran). With the analyst’s ever increasing demand for more detail and finer mesh sizes, some have reached the limit of the 32-bit platform.

With the release of Windows XP x64, engineers now have a direct upgrade path to a 64-bit platform without having to switch to a Unix/Linux based operating system.

Click here to view entire white paper.

Design Optimization of a Composite Wing

Abstract:
A leading international aerospace company developed a new composite wing design manually using standard aerospace structure design procedures. Because this design was unable to meet the performance requirements of the program, HEEDS was subsequently employed to optimize the design. The final optimized composite wing design exhibited a sizeable increase in critical buckling load (80%), a judicious decrease in failure index (30%), a moderate increase in stiffness (15%), and a slight increase in mass (6.6%) with respect to the original baseline design.

Click here to view entire white paper.

Model Comparison for Beams with and without Shear Center Offsets

Abstract:
When defining beam elements in NEi nastran the load is by default applied along the neutral axis of the beam between nodes. However, if you specify shear center offsets on the beam property card, the element is effectively shifted such that the load is applied along the shear center axis of the beam between nodes. Either way, the resulting displacements will not include any rotation about the beam axis. However, specifying an element offset, in conjunction with the property shear center offset, will result in the proper beam displacements including the desired “twist”.

Click here to view entire white paper.

NEi Nastran Nonlinear Analysis Handbook

Click here to view entire white paper.

Automated Surface Contact Generation (ASCG) Usage in NEi Nastran

Abstract:
NEi Nastran V9 adds a powerful new feature that allows automatic surface contact generation (ASCG) between discontinuous bodies. This feature is ideally suited for models in which the contact areas are initially touching and where little sliding is expected.

The automated surface contact generation works in three phases. First the user defines parameters such as which elements should be considered for each contact pair generated (the default is all solid and shell elements in the model), what the near tolerance is for objects to be considered in contact with each other, and the type of contact to be generated (i.e., general, welded, bi-directional sliding, or rough). Using these user defined parameters, a region of elements is considered. The default would be all solid and shell elements in the model. The first phase consists of identifying the external surfaces of these elements. For shells it would be the entire element. For solids it would be any exterior surface. The second phase then looks for grid points that are positioned on or near and above (within the user defined tolerance) of each surface. The third phase eliminates invalid contact scenarios by considering surface normals. It is important that surfaces that are to be welded are not offset and that shell normals are oriented properly.

Click here to view entire white paper.

Automated Surface Contact Generation in the NEi Solver Presentation

Abstract:
This presentation looks at a technique internal to the NEi Nastran solver that provides automatic contact recognition and creation directly in the FE Mesh.

Click here to view entire white paper.

Spot Weld Failure and Stress Redistribution Analysis Using NEi Nastran

Abstract:
The spot weld is one of the most common fastening techniques used in various industries like aerospace, automotive, maritime, etc. Often, engineers are interested in predicting the response of an assembled sheet metal structure when there is a progressive failure of one or more spot welds for a given applied loading. Design changes may be required if the structure cannot withstand loading in the event of one or more spot weld failures. But how can redundancy be simulated using finite element analysis software? NEi nastran is capable of simulating this effect via some interesting, simple, and easy to set up techniques which we will discuss in this paper.

Click here to view entire white paper.

Development of a 3D Finite Element Model for Evaluating Insertion Trajectories of Electrode Arrays, Contact Stresses and Associated Damage in the Human Cochlea

Abstract:
A 3D finite element analysis has been used in this study to model the insertion of the single wire (uniform stiffness), the Nucleus standard straight (graded stiffness) and the Contour electrode arrays with different stiffness properties into cochlear scala tympani. The results show the predicted trajectories and contact stresses exerted by the electrode arrays, from which the propensity of damage can be compared with results of published experimental studies. The present work overcomes limitations of the previous 2D finite element analysis, which was unable to predict out-of-plane trajectories and contact pressures on the basilar membrane. Three modes of damage are proposed as a result of high contact pressures exerted by the tip of the electrode array on the lateral wall of the cochlear spiral and on the basilar membrane, and as a result of sliding contact along the length of the electrode array. Results of the model have shown that the graded stiffness array is least likely to cause damage to delicate regions within the scala tympani compared with the Contour and the uniform stiffness electrode arrays. The tip of the Contour array was found to exert greater contact pressures (~0.35MPa) on the lateral wall than that of the graded stiffness array (~0.2MPa). The sliding contact stresses on the basilar membrane generated by the graded stiffness array (~0.09MPa) were found to be smaller than those of the Contour array (~0.6MPa). The likelihood of penetration of the basilar membrane by the tip of the Contour array was comparable with that of the graded stiffness array up to an insertion depth of 13mm from the round window but increased significantly beyond this point. The sliding contact stresses of the uniform stiffness array and stresses exerted at its tip were significantly greater than those of the graded stiffness and Contour arrays and therefore the uniform stiffness array has the highest propensity to cause damage during insertion. The 3D finite element model can help investigate different factors that may influence the likelihood of damage and useful for designing new advanced electrode arrays to improve the success of implantations and control of their final placement to maximize signals to hearing nerves.

Click here to view entire white paper.

SHERPA - An Efficient and Robust Optimization/Search Algorithm

Abstract:
Numerical design optimization is now an industry-accepted practice for more quickly identifying designs that meet increasingly stringent performance specifications and cost targets. Rather than manually iterate on design parameters in the hope of finding a design that meets the required specifications, automated numerical optimization algorithms can yield much better designs in much less time. These algorithms work with existing analysis tools, which predict how well a design performs. So the final result of an optimization run is an analyzed model of the best design and its predicted response characteristics. One of the keys to a successful optimization study is the effectiveness of the search algorithm used. This paper provides brief answers to the following questions about optimization algorithms:

• What does it mean for an algorithm to be efficient and robust, and why is it important?
• How do various algorithms compare on these important characteristics?
• What makes the algorithm SHERPA so effective?

Cyclic Symmetry Example

Abstract:
The purpose of this example is to show how to setup a cyclic symmetry model in NEi nastran, and to verify these results against a full model.

This example will use a 15 degree wedge of a turbine model which is 1 of the 24 total blades. One of the new features of NEi nastran V8.4 is the ability to setup cyclic boundary conditions automatically using the case control command CYSYMGEN, which will create the boundary conditions (via MPC equations) using a defined cylindrical or spherical coordinate system.

Click here to view entire white paper.

Modeling an Adhesive Bondline Using NEi Nastran

Abstract:
Noran Engineering, Inc. was approached by The Membership & Strategic Initiatives of The Adhesive and Sealant Council, Inc. regarding creating an adhesive element that would be standard for all commercial FEA packages. The approach was to use a combination of rigid elements and 3-DOF springs to model the adhesive properties. An alternative was introduced by NEi Software, which is to use a single CBUSH element that would replace the rigid/DOF spring combination. In addition to this method, a third alternative was proposed which is to model the adhesive using 3D brick elements since NEi Nastran can handle high aspect ratio brick elements without a loss in accuracy.

Click here to view entire white paper.

Comparison of Torsional Constants and Shear Areas of Beams

Abstract:
This document presents a comparison of the torsional constants and shear stiffness factors (or shear areas) of various cross-sectional beams. The results are obtained from FEMAP, NEi Nastran and MSC.Nastran. The torsional constants are compared with results calculated from finite element models composed of solids, while the shear stiffness factors are compared with data published by Guttmann and Wagner (Reference 2).

Click here to view entire white paper.

Using Enforced Displacements in Nonlinear Analysis

Abstract:
Enforced displacements and rotations in NEi nastran are referenced by an SPCD bulk data entry. Some confusion can arise when using enforced displacements because in most pre/post processors they are created as a “load”, when in fact they are treated as a special type of constraint in NEi Nastran. In addition, an SPCD entry is referenced by a LOAD = card in the Case Control.

Click here to view entire white paper.

NASA LaRCO2 Failure Criteria

Abstract:
A LaRC02 failure criteria is derived from Hashin’s criteria and Puck’s action plane concept, which applied on fiber reinforced polymer (FRP) laminates. LaRC02 criteria doesn't require curve-fitting parameters and it takes fiber kinking into consideration.

Similarly to Puck’s criteria, LaRC02 is based upon several types of failure modes. These modes are matrix failure in compression, matrix failure in tensile, fiber failure in compression, and fiber failure in tensile. LaRC02 requires same inputs as Puck’s or Tsai-Wu’s criteria.

Click here to view entire white paper.

Shock Response Spectrum Analysis via the Finite Element Method

Abstract:
This report gives a method for determining the response of a multi-degree-of-freedom system to a base excitation shock, where the shock is defined in terms of a Shock Response Spectrum (SRS).

A finite element model is used to determine the normal modes and frequency response function of a sample structure. Commercial finite element analysis software is used for this purpose.

The following steps are done outside the finite element software by using programs written in C/C++. The source code for these programs is available from the author by request.

The impulse response function is calculated from the frequency response function via an inverse Fourier transform.

A time history is synthesized to satisfy the SRS. The response time history of the structure is then calculated via a convolution integral using the synthesized time history and the impulse response function.

This approach is referred to as the synthesis method in this report. An advantage of this method is that the impulse response function can be used for numerous time history inputs. There is no need to rerun the finite element analysis for each input case.

Click here to view entire white paper.

Tips for Running Large Models in NEi Nastran

Abstract:
When running very large models in NEi Nastran, memory limitations often arise usually under the error: FATAL ERROR: S1110: INSUFFICIENT MEMORY AVAILABLE TO CONTINUE. The NEi nastran Reference Manual has the following explanation for this error:

Cause: This error results when a block of memory cannot be allocated and is usually caused by one of the following:

• Not enough virtual memory available.
• An attempt to allocate past the operating system 32-bit memory limit (usually 2 GB).

Action: Increase your virtual memory setting (see Virtual Memory in Windows Help). If this does not correct the problem, set the Model Initialization Directive, RAM, to a smaller value. Most versions of Windows have a 2GB limit for a single process. This limit can be reached when the RAM directive is set above 900MB. If neither of these actions helps, consider selecting the PCGLSS or VIS solver on the DECOMPMETHOD directive.

The goal of this paper is to explain in more depth the different strategies to overcome the memory limitations.

Click here to view entire white paper.

Using the RESULTLIMITS Feature in NEi Nastran
Instructions on how to determine the limits (max/min) of a set of elements for any output vector

Abstract:
The purpose of this document is to describe in general how to use the RESULTLIMITS feature in NEi Nastran. This feature allows the user to request from NEi nastran the result limits (max/min) of a given set of elements.

Click here to view entire white paper.

Using the MODALDATABASE Feature in NEi Nastran

Abstract:
The purpose of this document is to describe in general how to use the MODALDATABASE feature which is new to NEi Nastran V8.2. The model being used is located in the example folder of the installation directory and is titled trrsbed.nas.

Click here to view entire white paper.

NEi Nastran Results Neutral File Format
Examples of the displacement and elements results format

Abstract:
The NEi Nastran Results Neutral File system is used for intermediate and final results storage in all NEi Nastran solutions. Appendix A, Results Neutral File Format, in the NEi nastran Reference Manual contains additional information and provides the column number definitions for each file. Additional information is provided below for the displacement and element results file formats.

Click here to view entire white paper.

Connecting Dissimilar Element Types (Utilizing Contact Surface Elements)
Instructions on saving the modal data of a normal modes analysis

Abstract:
Contact surface elements were developed in NEi Nastran for use in nonlinear analysis, but can also be used in linear analysis (linear statics, normal modes, buckling, etc.) as weld elements to connect dissimilar element types and shapes. The default contact element property settings will automatically provide adequate penalty stiffness at the weld interface. An adaptive algorithm automatically adjusts penalty values for each contact segment based on surrounding stiffness.

Click here to view entire white paper.

Performance Effects: Skew and Aspect Ratio for CTETRA Elements
A discussion on the effects of skew and aspect ratio on the 10 Noded Tetrahedron Elements

Abstract:
The 10 noded CTETRA element is very popular in FEA analysis today due to its versatility. Autotetmeshers that are used to generate these elements can create highly distorted elements (high face skew angles and aspect ratios). This report looks at the effect the element shape has on accuracy. Figure 1 shows a cantilever beam made with 127 elements and 323 nodes. One end of the beam is constrained but allows unrestricted poisson’s expansion. The other end of the beam uses an averaging element (RBE3) to distribute a point load over the beam tip. The length of the beam is varied to change element skew angle and aspect ratio. Element skew angle and aspect ratio are defined in Section 2. The results are presented in Section 3.

Click here to view entire white paper.

2-D Orthotropic Material Analysis

Abstract:
The model shown in Figure 1 is a 20 inch by 40 inch plate with a 4 inch diameter hole in the center of the plate. It is 0.1 inches thick and constructed of a generic 2-D orthotropic material. The property information is shown in Figure 2 and the material values are shown in the Figure 3. Fourteen elements were defined along the length of the model, and seven elements along the height. The hole was defined to have 18 elements around its circumference. This model will be used to show the importance of correctly using material directions in orthotropic material analysis. This same procedure is used to align material angles for composite/laminate analysis.

Click here to view entire white paper.

Transitioning Bending Elements to Solids

Abstract:
NEi nastran allows for the transitioning of shell elements to solids using the RBE3 averaging element. The RBE3 element generates the MPC equations so that bending moments in bar and shell elements are transformed into couples in the adjacent solids. Figure 1 shows the transition of a bar and plate into a solid hex element and the corresponding RBE3 definition.

Click here to view entire white paper.

Performance Effects: In-Core vs. Out-of-Core Solutions
A discussion of the effects on analysis time for In-Core vs. Out-of-Core solutions

Abstract:
When an analysis model has a factored matrix size larger than the amount of physical memory present in the computer, the VSS solver in NEi Nastran must go “out-of-core” and use virtual memory to solve the model. Generally, an out-of-core solution is not as fast as an in-core solution. The purpose of this test is to observe the effects of forcing an in-core solution by adjusting the RAM Directive Setting for models with matrices larger than the amount physical memory present.

Click here to view entire white paper.

Extracting Forces from Enforced Displacements

Abstract:
This example will display how to extract forces from enforced displacements. The model shown in Figure 1 is a 2.0 x 10.0in. plate with a thickness of 0.1in. and will be used for the analysis.

Click here to view entire white paper.

Performance Effects: Sparse Direct and Iterative Solvers
A Performance Assessment of NEi Nastran's New Sparse Direct and Iterative Solvers

Abstract:
NEi Software has recently added two new solvers, Vector Sparse Solver (VSS) and Vector Iterative Solver (VIS), to its general purpose finite element analysis engine, NEi Nastran. One solver uses a direct approach while the other uses an iterative Pre Conjugate Gradient (PCG) approach. Both solvers are fully sparse and store and operate only on non-zero matrix elements. This paper looks at the effect these solvers have on the performance of NEi nastran for various finite element model and solution types. In many cases performance has increased by a factor of 10, thus allowing jobs that took days to be solved in minutes.

Click here to view entire white paper.

Interpreting NASTRAN BAR and BEAM Results

Abstract:
Bar/beam element output consists of strain energy, stresses, and forces. Stresses and forces are output at the bar/beam end points and interior points (if specified). The sign convention is shown in Figures 1-3.

Click here to view entire white paper.

Genetic Algorithms for Optimal Design using NEi Nastran

Abstract:
The modified ESO algorithm was applied to the well known problem of the design of shoulder fillets in a stepped bar as shown in Figure 1. This is a classical problem investigated by many researchers because of its importance in traditional mechanical design. Most of the work done to date considered maximum stress as design criteria. Earlier work in this field used conventional mathematical programming methods in order to minimize stress in the fillet profile (Kristensen and Madsen, 1976, see Reference 2). Pedersen and Laursen (1983, see Reference 3) used linear programming to derive optimal fillet shapes for various loadings. A comprehensive work on shape optimization for different fillet configurations can be found in Waldman et al. (2001, see Reference 4). They used another popular heuristic method biological algorithm and showed that stress concentration factors obtained were considerably lower than those reported in the literature using sensitivity based methods. However, little attention has been paid to date to consider damage tolerance design that allows for the presence of initial cracks along the fillet profile. It is now established that stress optimized shapes may differ considerably from the fracture strength optimized shapes. Hence, whilst a fillet with a stress optimized profile may be lighter, it may suffer from premature fracture or fatigue failure.

Click here to view entire white paper.

Troubleshooting NEi Nastran Nonlinear Analysis

Abstract:
The following steps should be used to diagnose problems when running any nonlinear analysis:

Click here to view entire white paper.

Performance Comparison: VSS versus PCGLSS Solvers
Variations in performance created by using different solution methods and solvers

Abstract:
Seven shell/solid models will be analyzed using different NEi Nastran solution methods. The model files are described below.

Click here to view entire white paper.

Performance Effects: Speed vs. Accuracy

Abstract:
This test will determine the settings for optimum speed when using the iterative (VIS) solver in NEi Nastran. These settings have an effect on accuracy as well as speed; therefore, the point at which accuracy falls off is of note. The model FORK1.NAS was used for this test, and it’s shown in Figure 1.

Click here to view entire white paper.

Performance Effects: Varying the SPARSEMETHOD Setting
A discussion of the effects on analysis performance with variations in the SPARSEMETHOD setting

Abstract:
The purpose of this test is to determine what, if any, effect manually altering the SPARSEMETHOD setting (found under Program Controls, in the Options menu in NEi nastran) has on the solution, and the solution times. This test was performed with several different models in order to obtain a broader summary of the effects.

Click here to view entire white paper.

Performance Effects: Varying the Ram Directive Setting
A discussion of the effects on analysis performance with variations in the RAM directive setting

Abstract:
Below is the model used to test the RAM setting directive. This example is used to determine the effect on the solution time due to the varying of the RAM directive setting (under Memory Management) in NEi Nastran.

Click here to view entire white paper.

Eigensolver Performance Review

A detailed review of the performance of the enhanced PCGLSS V8.2 Eigensolver

Abstract:
A new Preconditioned Conjugate Gradient (PCG) based eigensolver has been added to NEi nastran V8.2 which offers dramatically faster performance for large Eigenvalue problems (over 100,000 DOF). The solver is based on the same technology as the current PCGLSS solver available in V8.1 for linear and nonlinear statics. The new eigensolver uses a blocked Lanczos method and requires less memory than the current subspace solver.

Figure 1 shows a solid model meshed using 5 different mesh densities. An aluminum material was used with the base of the tube fixed. Table 1 gives the model data for each test case. All Eigenvalues from each solver matched exactly to the precision of the output. Figure 2 suggests that for models over 70,000 DOF the Lanczos solver will be faster with the performance ratio increasing dramatically for models over 250,000 DOF.

Click here to view entire white paper.

NEi Nastran Compatibility with Warped Elements

Abstract:
This document displays the compatibility that NEi Nastran has with warped quad element models. In the following document two NAFEMS benchmark problems, both containing warped elements are analyzed using NEi Nastran and the results compared to NAFEMS benchmark results. The two problems examined are the hemisphere point loads problem and the cylindrical shell patch problem.

Click here to view entire white paper.

NEi Nastran Fracture Evaluation
A technical discussion regarding NEi Nastran's Fracture Evaluation capabilities
A comparison of total run time vs. solution accuracy for the VSS and PCGLSS solvers with variations in the MAXSPARSEITER & SPARSEITERTOL settings

Abstract:
Several commercial finite element programs currently use the older generation of 8 noded and 6 noded isoparametric to determine the fracture strength of structural components. When using these elements it is necessary to move the mid-side nodes to the ¼ points. In this fashion an accuracy greater than 1% is achievable if the displacement field calculated for at nodes placed at ~1/15th of the crack length “a” from the crack tip is used to determine the stress intensity factor. When using NEi nastran similar accuracies can be obtained using the CQUADR element. However, in this case it is not necessary to move the midside nodes.

Click here to view entire white paper.

Performance Effects: Varying the FILEBUFFERSIZE & NFILEBUFFERSIZE Settings
A discussion of the effects on analysis performance with variations in the buffer size settings
A discussion of NEi Nastran's compatibility with warped quad element models

Abstract:
This document displays the findings of the effect on a model’s run time that is incurred by varying NFILEBUFFER and FILEBUFFERSIZE. The sample models that were used for this test are shown below and are all located in the NEi Nastran installation folder.

Click here to view entire white paper.

Click here for complete list of white papers.