[ From the Beginning ] - MSC Software...MSC.Nastran, MSC.Patran and MSC.ADAMS from MSC's portfolio of best-in-class technologies. Under BTCC rules only certain modifications to the

[ From the Beginning ] MSC.Software: Putting Customers One Step Closer to the Simulation Environment of the Future

A few months back, I asked our customers about their needs. The greater number of customers provided a list that began with lower costs, simulations that better represent real life and faster processing. I'd like you to know the list wasn't filed and forgotten. It became the mantra for the development team working on MD Nastran.

MD Nastran is the result of customers' demands and it is already changing how and when companies use simulation. MD Nastran allows customers to transition from single point simulation tools to an integrated enterprise solution in the same environment they have used for years. By combining such best-in-class technology platforms as MSC.Nastran, Marc, Dytran and LS-Dyna into one fully integrated multidiscipline solution for the enterprise, MD Nastran will:

Cut costs substantially with the use of a single model that eliminates the time engineers spend creating models for each different discipline. Better represent real life results through a combination of fully integrated linear, nonlinear and explicit nonlinear simulation coupled with the ability to run such optimization routines as stochastics. Allow use of HPC true 64-bit processing for the use of more simulation and earlier in the design process.

MD Nastran transitions from simulating how a design will perform to how the product will perform.

When compared with bundled single-point simulation tools, MD Nastran can reduce the time-to-solution up to 50% because customers work with a single common data model. Using multiple models for uncoupled discipline analysis with multiple single point tools is time consuming. Plus, increasing commonality through the use of a single model representation for multidiscipline analysis enables higher result accuracy.

MD Nastran has been well received by many of our industries analysts. For example, Donald H. Brown, chairman, Collaborative Product Development Associates, LLC (CPDA), said "With MD Nastran, MSC produces an enterprise class solution that will provide simultaneous multidiscipline system level simulations to accurately represent real life scenarios - MD Nastran obsoletes single point simulation tools."

Charles Foundyller, CEO, Daratech, Inc., said, "By offering MD Nastran as a single solver running linear, non-linear and explicit simulations in parallel, that the company says produces results 50% faster than its competition, MSC.Software is offering the market a powerful, much-needed, speedup that has the potential to greatly enhance inter-discipline innovation and accelerate time-to-market."

Additionally, MD Nastran eliminates the silos of simulation technology that exist within today's enterprise environment, resulting in cost savings from the overall development investment. This allows engineers to develop additional skills, which will further their careers by providing added value.

An enterprise simulation solution optimized for multidiscipline analysis accelerates time-to-market by enabling customers to perform linear, implicit nonlinear and explicit nonlinear all within the same simulation platform. Furthermore, MD Nastran accounts for the interaction across discipline domains to accurately model real life scenarios.

Competitive advantage is the result of maximizing product innovation. MD Nastran enables more design iterations through its support of true HPC 64-bit (ILP) and SMP/DMP (shared memory parallel / distributed

Page 1 of 2

7/27/2009http://www.mscsoftware.com/alpha/print_article.cfm?volume=7&articleId=63

memory parallel) support for large/complex model management and optimization. The 64-bit port of MD Nastran can handle large and the continuously growing models of assemblies with millions of degrees of freedom with ease.

MD Nastran adds a powerful competitive advantage to product development processes that enables customers to innovate industry-leading products while trimming costs. For over 40 years, MSC.Nastran has been the industry de-facto standard and leader in CAE simulation. Now, MD Nastran, the only multidiscipline solution, leads the industry into a future that empowers customers with value-driven innovation.

To get the fully story, read the MD Nastran in this issue.

Page 2 of 2


[ Company & Industry News ] Current News & Events

Spirit AeroSystems Selects MD Nastran for Multidiscipline Simulation Spirit AeroSystems Inc, the world's largest independent supplier of structures for commercial aircraft, has selected MD Nastran for multidiscipline simulation capability with MSC.Nastran, MSC.Marc, and MSC.Patran. MD nastran is a flexible solution for manufacturing organizations requiring single source access to the diverse range of simulation solutions which MSC.Software has to offer. These solutions will help Spirit AeroSystems engineers solve complex product development challenges and evaluate more design variants within a single, unified framework. Spirit AeroSystems, Inc. an Onex company, is the world's largest independent supplier of structures for commercial aircraft. It designs and builds part of every Boeing commercial aircraft currently in production except the 717. "We are pleased Spirit AeroSystems has further invested in our partnership by increasing their investment in our integrated enterprise solutions for their virtual product development needs," John Howaniec, senior vice president, Americas sales operations, MSC.Software. "We take great pride in building and maintaining customer relationships and this new agreement validates our commitment to provide customers with industry leading solutions."

David Jones Appointed Vice President of Corporate Services David Jones, an experienced high technology enterprise executive, will global services, technical support, documentation and product services. "I am pleased to announce the appointment of David Jones as VP of corporate services for MSC," said Glenn Wienkoop, president and chief operating officer, MSC.Software. "In this role David will manage global services, technical support, documentation and product services. His primary focus will be driving continuous improvements and profitability on a corporate basis. David's proven track record in high technology, his leadership in the services industry, and his successful enterprise experience will ensure MSC's continued success in our services business. I am pleased to welcome a proven industry leader to MSC." Bill Weyand, chairman and CEO, MSC.Software stated, "As MSC completes the final stage of building our global executive management team, I am pleased to welcome David who brings worldwide expertise in enterprise business and added-value services, and completes a strategic part of being the leader in VPD Solutions." David Jones joins MSC.Software from Sun Microsystems where he held the position of chief technologist for the global data center practice in the client solutions organization providing leadership and definition to the solutions brought to market as well as providing support to strategic global projects. Prior to this role, Mr. Jones held several critical leadership positions at Sun including principal engineer, US data center solutions director and chief architect of the Western field organization. His valuable experience in the Industry is diverse and includes parallel processing research for a major oil company, clinical trial systems for life sciences, as well as research for the USDA. Mr. Jones has 20 years of experience in architecture, engineering and IT management with degrees and credentials in chemistry, computer science, and computer engineering from the University of Irvine and California State Polytechnic University.

Triple Eight Race Engineering Extends Winning Partnership with MSC.Software Triple Eight Race Engineering who, together with VX Racing, are now five times consecutive winners of the British Touring Car Championship (BTCC), has further extended its supply partnership with MSC.Software. MSC.Software first partnered with Triple Eight Race Engineering at the start of the 2004 season in order to maintain their competitive advantage by significantly upgrading their simulation capability. With a broad portfolio of simulation solutions at their disposal, designers at Triple Eight began the complex task of designing, verifying, and optimizing the structural and control elements critical to the car's racing performance. Following championship winning seasons in 2004 and 2005, the current expansion of this partnership continues to strengthen the competitive development of the team, further extending the use of

Page 1 of 3


MSC.Nastran, MSC.Patran and MSC.ADAMS from MSC's portfolio of best-in-class technologies. Under BTCC rules only certain modifications to the race car configurations are possible. Using a combination of flexible and rigid multi-body dynamics techniques, engineers at Triple Eight Race Engineering use MSC's industry leading solutions to optimize for strength and weight, targeting a stiffer, lighter structure but without compromise on reliability. System level dynamic simulations are used to enhance suspension design, both predicting and automatically coupling load data into the detailed FEA calculations. Such integration of FEA and multi-body dynamics technology has already allowed engineers at Triple Eight to design the torsional and bending behaviour of an innovative rear suspension system, the performance of which contributed significantly to the championship success of the 2005 car. "It is very difficult in motorsport to find technical partners who are prepared and able to work at your speed and level," said John Morton, chief designer, Triple Eight Race Engineering. "With MSC.Software we have found a winning combination of technology and expertise, a partnership which enables our team to continue to compete at the very highest level." Amir Mobayen, senior vice president, EMEA, MSC.Software stated, "With a history of successful partnerships in touring car, endurance racing, and Formula 1, MSC.Software has developed an inherent understanding of the special demands of motorsport applications. In addition to providing on-track competitive advantage for our racing partners, these collaborations are the proving ground for simulation technology and process improvements which are of direct benefit to MSC customers across a range of industries." Work has already begun for the coming season, with additional focus on the mechanical dynamics of the suspension and other structural and handling systems of the 2006 car. The 2006 BTCC season commences at Brands Hatch on April 9, and Triple Eight Race Engineering and MSC.Software are looking forward to another year of highly competitive performance.

Tata Motors Selects MD Nastran to Drive Innovation Tata Motors, India's largest automobile company, has invested in MD Nastran with such multidiscipline simulation functionality as MSC.Nastran and ADAMS for improving processes to drive innovation. "Tata Motors is committed to being a leader in the extremely competitive Indian automotive market and to expanding into the global market," said Mr.T N Umamaheshwaran, chief technology officer, Tata Motors. "The MSC.Masterkey licensing system allows us to have a flexible and valuable utilization of MSC solutions for NVH, vehicle dynamic, durability and vehicle development. We selected MSC's tightly integrated products including MSC.ADAMS, MSC.Nastran, MSC.Actran and MSC.Fatigue to develop better products, faster." "India is an important growth market in the global automobile industry and is providing us with multiple opportunities to deliver the industry's best and most tightly integrated suite of solutions to accelerate the development of higher quality, more innovative products," said Dr. Christopher St.John, senior vice president, Asia Pacific operations. "Our Indian automotive customers deploy our solutions quickly, these emerging companies recognize the value of VPD and understand the solutions will deliver the competitive advantage to outpace more mature organizations in the race to deliver innovative products, faster." Jack Varney Appointed Vice President Global Learning and Development Jack Varney, a senior enterprise sales training executive has been appointed to lead global learning and development activities at MSC.Software. "Adding a senior global sales training executive to our team will further the goal of selling large enterprise software solutions to our global accounts," commented Glenn Wienkoop, president and COO of MSC Software. "Jack's first priority will be the development and implementation of a world-wide sales training curriculum to support MSC's transition to enterprise sales activities. His focus will be on value-added ROI sales skills, executive relationship building and the selling of software solution bundles aligned with MSC's new product offerings." Mr. Varney has over 20 year of marketing and sales training knowledge with both technology and enterprise software firms and was most recently the director of global sales training at Electronic Data Systems. He also held senior sales training and marketing positions at Standard Register and Pitney Bowes.

Andrew Meyer Appointed Vice President of Corporate Marketing Andrew Meyer, an experienced enterprise software executive, will lead MSC.Software's marketing organization as vice president corporate marketing.

Page 2 of 3


"As MSC further strengthens its senior management team, I am very pleased to announce Andrew Meyer, a highly seasoned enterprise software executive, as our new vice president of corporate marketing," said Bill Weyand, chairman and CEO, MSC.Software. Andy's track record of enterprise software accomplishments bring a new level of expertise to his role at MSC and will drive our objective of becoming the market-driven, leading enterprise VPD solutions provider. His wealth of marketing knowledge and strategic insight of the enterprise software industry further expands the collective skill set of the new MSC executive team and I am pleased to have a proven industry leader join us in this important corporate role." Mr. Meyer has over 20 years of senior marketing and management experience with software and technology-based companies. He joins MSC from Worksoft where he served as vice president of marketing. During his tenure, he successfully led product strategy, product management as well as marketing communications. Prior to Worksoft, Andrew spent four years at Websense as their vice president of marketing where as a senior member of the management team they became the dominant leader in their market - growing revenue from $17 million in 1999 to more than $110 million in 2003. His tenure also included the introduction of a business model that led to one of the most successful IPO's of 2000. Mr. Meyer holds a master's degree in business administration from the University of New Orleans and a bachelor's degree in mechanical engineering from the Georgia Institute of Technology.

"With more than twenty years of senior management experience in the software industry, Andy will be responsible for the company's worldwide marketing program and will focus on delivering the next generation of enterprise analysis and simulation management solutions to market," said Glenn Wienkoop, president and chief operating officer, MSC.Software. "Andy has the experience, breadth and vision that will help drive the marketing organization to new heights as we lead MSC to become a world-class provider of integrated enterprise VPD solutions. I'm confident that this addition will accelerate MSC's market leadership and further capitalize on the solid momentum our business has demonstrated. Andy is a strong and very welcome addition to MSC."

Page 3 of 3


[ Technical Matters ] MD Nastran The Multidiscipline Simulation Game Changer

The biggest issue facing today's manufacturers is cost reduction. Excessive defects in product development and variation in materials and manufacturing processes generate unexpectedly high costs. MD Nastran evaluates product performance with coupled system level simulations that consider the interaction between disciplines before the design decisions are made.Working from a common model, errors are drastically reduced and stochastic optimization minimizes the risk of variation and cataclysmic combinations that can't be determined by any other process. Ultimately, MD Nastran is changing the way the game is played to minimize changes after design decisions, time to manufacture, time to market and costs. Often these costs are the result of engineering change orders (ECO) after making design decisions and extraordinary warranty costs because defects were not identified and corrected earlier in the development process. By the time a design is approved or released for

manufacture, when a defect is discovered with a process, part, subsystem or product, it can take weeks or months to respond. By the time the change reaches the shop floor many parts or products may be involved. Single-point simulation tools identify design defects after the design decisions have been made. MD Nastran identifies defects in the product, much earlier in the process and saves a lot of time and costs.

Multidiscipline Simulations A good example of the need for multidiscipline simulations includes racing boats that begin to fly. This phenomenon is attributed to failure to simulate the interaction between airflow and structural deformation. Single point solutions allow multidiscipline analysis, but do not consider the interaction between disciplines. Only MD Nastran considers interaction (coupling) between disciplines, uses a common model, runs stochastics optimization and takes advantage of the speed of 64-bit HPC processors. Simulation of physical phenomena with real life results requires an accurate representation of the complex interactions between key disciplines. Even with recent advances in modelers (pre and post processors), computing power and automated capabilities, discipline specialists still manually simulate the complex inter-discipline interactions as discrete

analysis steps. Within a given discipline, analysis steps can be time consuming. However, assessing large volumes of analysis data to determine how to hand-off results from one discipline to another is inevitably orders of magnitude more tedious, subject to human error, compromises simulation accuracy and often is unrepeatable.

Engineers sometimes carry information by hand or force the information from motion in a static manner to impact the FE representation of a system. MD Nastran connects them so the information is live, e.g. they are in an open loop environment. Whether it is linear, nonlinear, motion, CFD or explicit dynamics, MD Nastran allows disciplines to work together, rather than simply communicate with each other. Working together implies they provide correct engineering and mechanical feedback to each other at exactly the right time. Moving beyond traditional multi-physics systems, discipline chaining/integration between multi-body motion and FEA facilitates simulation capabilities that allow enterprise-wide multidiscipline simulation to drive design early in a product cycle, such as external system loads spectrum definition. The same is true with integrated FEA and CFD analysis.

MD Nastran

Page 1 of 4


MD Nastran combines such best-in-class technology platforms as Nastran, Marc, Dytran and LS-Dyna into one fully integrated multidiscipline simulation solution for the enterprise. MD Nastran's multidiscipline coverage has increased since 2001, beginning with traditional best-of-class structural statics and dynamics; and then expanded with such core disciplines as state-of-the-art implicit/explicit nonlinear and multi-body motion. Such industry-specific disciplines as crash, NVH, acoustics, aeroelasticity and metal forming support all the disciplines necessary to meet the needs of all manufacturers, i.e. results that very closely represent real world behavior. By precisely addressing the physics, chaining/integrating key engineering disciplines and bridging traditional computational models MD Nastran represents the

physical continuum with much higher accuracy.

MD Nastran resulted from four years of intensive development, strategic acquisitions - particularly of ADAMS motion and Marc nonlinear analysis technologies, and important partnerships. The latter include LS-DYNA, which provides nonlinear dynamics for the embedded Nastran crash capability, and both Fluent and CD-adapco CFD for fluid-structure interaction (FSI) capabilities. FEA and CFD disciplines meet at the surfaces of solid structures and each may provide loads and boundaries for the other. However, manually inputting such data from one discipline to another prevents consideration of interactions and keeps engineers using each discipline separately. This ends up costing a lot of time. By using third party MpCCI, MD Nastran benefits from that program's inter-communication backbone to couple FEA and CFD capabilities.

MD Nastran offers a number of industry-specific capabilities made possible by its inter-related disciplines. These include noise, vibration and harshness (NVH) and acoustics for automotive design, crash and passenger safety, drop analysis, and thermal management from thermal conduction to 3D radiation problems.

NVH studies make the benefits of MD Nastran obvious. An ADAMS model simulates the car on a bumpy road, indicating how the irregular surface impacts the noise and vibration of a particular vehicle. MD Nastran makes it possible to turn the ADAMS model into a mathematical representation integrated with a full Nastran NVH model. Engineers use the same model for simultaneous simulation of the acoustics for the passenger cavity - resulting in an integrated study of true NVH characteristics on a real road, complete with defects and bumps. The loads generated by the NVH simulation can also be used later for crash simulation.

In another example of the coupled disciplines in MD Nastran, automotive engineers who have run an ADAMS simulation of a suspension system can go into Nastran to assess the life of the A-arm with the suspension data forming an essential part of the analysis.

Common Model MD Nastran can reduce the time-to-solution up to 50% when compared with bundled single-point simulation tools because customers can now work with a single common data model in place of multiple models for uncoupled discipline analysis using multiple single point tools. Based on a system level model - single data model with multi-physics representations - MD Nastran allows everybody working on a design access to the same data. This does not mean use of a single model across every discipline, but that extractions from a common model are used to create representations of systems for simulation with common loads and constraints. The system level models created with rigid body elements can be imported into MD Nastran. However, not every simulation problem is solved with a single equation. It requires a number of equations working with

Page 2 of 4


common data to provide the most realistic simulation possible. A finite element (FE) model is an abstraction of a physical system that can be loaded and conditioned to study various disciplines. The type of disciplines that must interact with each other guide and determine whether analysis is simultaneous, integrated, staggered or loosely coupled.

A system level model that combines motion studies with FEA and CFD offers many applications for multiphysics. For example, MD Nastran removes the barriers between linear and nonlinear analysis for the aerospace industry, which is focused on the use of composite materials. MD Nastran recognizes loads that impose strain values over 3-4% and automatically performs nonlinear analysis.

Optimization Users can run optimization loops at various levels of simulation. Shape and topological optimization operate within each discipline, and variability (stochastic or probabilistic) optimization determines the robustness of the design. Engineers can look at system equations and determine variability at all levels - especially where material properties and manufacturing processes have a great deal of variability. MD Nastran's unique optimization sequence allows the combination of such varied engineering events as static analysis and NVH, etc.

64-Bit High Performance Computing Ten-years ago, modeling rivets in an aircraft FE wing model was not even a consideration; today it is commonplace. More sophisticated and complex simulation models with unbounded model and analysis data set size increasingly have become a necessity. At the other end of the lifecycle, engineers simulate entire vehicle systems prior to manufacturing. Even before considering discipline integration, it is clear that model size and complexity remains unbounded.

Taking into account inter-discipline simulations places an even greater burden on computational resources and demands, fueling the continuing demand for computational optimization. To this end, MD Nastran continues the leadership in high-performance computing established by its predecessor products with its first mover port to true 64-bit (ILP), continued investment in solver optimization, SMP/DMP (shared memory parallel / distributed memory parallel) support and enhancements to our superelement technique for large/complex model management and optimization capabilities.

The ambitious scope of MD Nastran means that it must handle very large problems - and to make very large simulations fit within today's time constraints, the program runs on both 32-bit and 64-bit computer cluster environments. MD Nastran is especially optimized to run on the 64-bit supercomputing environment. With the 64-bit enabled MD Nastran, MSC.Software offers the best scalable simulation platform and leads the industry in the migration from 32-bit to 64-bit architecture. Actual speed will depend on the kind of parallelization used for computing, and for really huge problems, may require hundreds of 64-bit processors.

For example, with MD Nastran aircraft engineers can conduct the analysis of an aircraft wing load that includes hundreds or thousands or individual rivets. Studying a single rivet in such an application requires complex nonlinear elastoplastic analysis. In the past, such an analysis required more computer power than most engineers could access. Today's computers enable full wing simulation, with all the rivets - resulting in savings to companies, compared to physical tests, in the order of half a billion dollars for a new generation aircraft.

Conclusion It is a given that manufacturers need to perform interoperable multi-disciplinary analyses on growing models (parts and assemblies) to continue to satisfy the high demands of their consumers. MD Nastran's complete multidiscipline coverage allows manufacturing customers to address a broad set of true multi-discipline problems with higher accuracy and reliable performance predictions. Additionally, MD Nastran utilizes multi-disciplinary interactions to solve the large and continuously growing models of assemblies with millions of degrees of freedom efficiently and quickly via HPC 64-bit processing. The power of MD Nastran puts customers one step closer to the VPD environment of the future.

Page 3 of 4


Safe Harbor Language This document contains forward-looking statements, including all statements relating to the features, benefits, capabilities and performance of MD Nastran and other MSC.Software products. These statements are subject to risks and uncertainties that could cause actual results to be materially different than expectations. Such risks and uncertainties include, but are not limited to, changes in technology, the end-user computing and analysis environment, implementation and support that meet evolving customer requirements, general industry trends and the impact of competitive products. Furthermore, information provided herein, which is not historical in nature, are forward-looking statements pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. All such forward-looking statements are based largely on management's expectations and are subject to and qualified by risks and uncertainties that could cause actual results to differ materially from those expressed or implied by such statements. The Company undertakes no duty to update any forward-looking statement to conform the statement to actual results or changes in the Company's expectations.

The MSC.Software corporate logo, MSC, and the names of the MSC.Software products and services referenced herein are trademarks or registered trademarks of the MSC.Software Corporation in the United States and/or other countries. NASTRAN is a registered trademark of NASA. All other trademarks belong to their respective owners. © 2006 MSC.Software Corporation. All rights reserved.

Page 4 of 4


[ Case Studies ] Simulation Solution Closes Physical Test Gaps by Joseph A. Schudt, Prasad Kodali, Hyung-Joo Hong, Vivek Chidambaram, Glenn Babiak, Robert L. Geisler General Motors Corporation

PROCESS The general procedure whereby measured data is supplemented, adjusted or extended is initiated with an assessment of the current measured data and the measurement vehicle. This data is assessed for accuracy and completeness. The focus is on identifying launching points for supplemental studies intended to extend the dataset. Then gaps between the measured dataset and the needs of the validation (or other associated activity) task must be identified. It is these gaps that the VPD tools will be used to fill. For each gap, a capability assessment of the simulation solution must be made. If the solution has gaps in that ability to simulate an event or condition, the analysis may need to be balanced or augmented with test data. Therefore, simulation must identify such additional information as shock characterizations, bushing measurements or full vehicle finite element models that must be acquired to perform a needed simulation. Existing measured data should be used as much as possible as a basis for assessments of design changes or variation.

MEASUREMENT CORRELATION AND MODELING REQUIREMENTS To the extent that it is possible the analytical simulation vehicle should be based upon the test vehicle. Deviations from a simulation of the test configuration are better representations of the actual changes to the vehicle configuration. We have found certain issues to be critical in matching the performance of measured vehicles. These issues include tires, vehicle characteristics, vehicle trim height (stance), mass, bushings, stabilized bars, flexible body structure, other flexible components, jounce bumpers and shock absorbers / active suspension controls. All of these issues must be capable of simulating test events and any deficiencies understood and quantified. Dimensions, force and moment curves have a big impact on results attained with simulation and therefore an engineer’s judgment will be very important to understand model fidelity. Additionally, many of these issues are interconnected and by changing one, another is affected, for example larger mass or tire size will affect trim height.

Understanding such dynamic loads as when a car runs over a pot hole or travels down a rough road and their effect on performance is crucial during the development of a new car. However, physically testing all configurations and assessing design robustness through all levels of variation is time and cost prohibitive, as well as manpower intensive and requires costly equipment and track time. We have found multi-body dynamic simulations with such simulation solutions as MSC.ADAMS provide a time and cost efficient method, which supplements test information and closes the gaps between physical test variables. A limitation of road load measurement is a test vehicle represents only a single sample of the total number of

vehicles that will be produced. Typically, a test vehicle is subjected to off-nominal build conditions, which may impact the measured loads. Examples of such conditions include vehicle trim heights (stance), off-nominal bushing and jounce bumper stiffness and variation in tire performance. Other variables may affect test results, such as vehicle set-up (tuning, trim heights, etc.), worst-case loading (axle mass, tire performance, etc), transducers and weather. Since many tests are run far in advance of producing the subsystems and vehicle, understanding design variation is crucial for assessing vehicle performance.

Page 1 of 2


MODELING ASSUMPTIONS Such multi-body dynamics solutions as MSC.ADAMS are employed to simulate these events, although other options are possible. Model fidelity must meet the needs of the event being considered. This implies a greater level of detail must be employed as events become more severe, for example pot holes. Potholes and bumps can be some of the most severe events that are simulated to supplement measured vehicle data. In many cases substantial structural deflection occurs in the chassis and vehicle body structure. It is also possible for wheels to exhibit deflection (occasionally plastic) during these events. Tire models should comprehend wheel rim groundout.

EVALUATION OF RESULTS Assessing changes to the design requires a solid baseline model. For comparison purposes, the baseline analytical run should be correlated to available measured data. This simulation provides the basis for all other comparisons. A variety of ways can be used to compare simulation results with measured vehicle loads. Direct comparison of time histories may be made or additional statistical tools can be employed. For moderate to severe events concerned with peak loads, the criteria by which a simulation is considered correlated must be precise. The peak loads for these events are often the crucial factor, but peaks at different interfaces occur at different times and the phasing must also be matched. It is possible that the measured vehicle experienced complicating effects, e.g. shock cavitations, during the event, which must be accounted for or understood in the context of the extended data set. For such rough road events that are longer in time than pot holes, it may be more appropriate to make a comparison based on total damage. Since the path taken cannot often be exactly determined or simulated, and the primary purpose of these loads is to drive high-cycle fatigue failure, statistical assessments are appropriate. Low to moderate severity potholes are a good measure of model to test vehicle correlation. Severe pothole events can be more problematic in that they may introduce significant structural deflection into the body and chassis components. This structural deflection of the body and chassis represent substantial energy storage during the event. For potholes, time histories are compared directly for wheel force, wheel acceleration, and other component loads. The load phenomena early in the event are often primarily due to tire / road interaction, while later phenomena more likely are due to downstream components such as bushings, shock absorbers and other vehicle structure. Additionally, pothole events often give rise to substantial design or “strength” loads and accurate simulation is extremely important in assessing the impact of vehicle tuning changes and platform bandwidth, e.g. differing tires, shocks, stabilizer bars, etc.

ASSESSMENTS A crucial aspect of making judgments is developing an understanding of the variation inherent both in the production vehicle and in any measurement activities used as a reference. The analytical model is a useful tool in this regard, and we have employed it to discern sensitivities and comprehend variation. In fact, often we conduct simulations to understand the effects of trim height, tire pressure, and shock valving on loads generated during pothole events. These parameters may show significant variation over the life of a vehicle; understanding that variation leads to understanding the variation in the event based loads.

CONCLUSION We are increasingly supplementing measured road load data with data generated with simulation solutions. By using analytical tools to simulate durability events, we have demonstrated success in supplementing measured road load data. Our continued activity is to further leverage the capability of the simulation loads prediction activity while minimizing the cost and timing impact of vehicle testing. We expect the use of these techniques to expand in the future and the capability of the simulation activities supplementing measurement will grow to meet our needs.

Page 2 of 2


[ Case Studies ] Simulating Underwater Explosions 3D Analysis Required to simulate Pulsating Bubble, which Creates the Jet Stream that Causes Extreme Damage to Submerged Structures by Arjaan Buijk, Senior Manager Solver Development Peiran Ding, Development & Application Engineer MSC.Software Corporation

Often a requirement before commissioning a ship, detonating a nearby underwater explosion to determine its effects is expensive and frequently impractical. Computer simulation has proven a much lower cost and practical alternative for providing crucial engineering information to improve the design. However, models based on 1D or 2D explosive shockwave theory or from experiments performed without a structure fail to consider the 3D fluid-structure interaction of near-field underwater explosion (UNDEX) loading problems. However, a new and unique multiple-material Euler solver with Adaptive Multiple Domains has been integrated with MSC.Dytran, allowing 3D modeling and simulation of near-field UNDEX loading conditions, enabling an efficient process that provides a better understanding of the complex UNDEX loading events. Background An underwater explosion creates a three-dimensional pulsating gas bubble, containing the gaseous products of the explosion. causing a high-velocity bubble jet that is extremely efficient in producing damage to nearby

structures. It is not the shock waves that damages nearby underwater structures but the expansion and contraction (pulses) of the gases in the bubble. As the bubble expands, the inertia of the outward moving water causes the gases in the bubble to exceed pressure equilibrium with the surrounding water. When the gas pressure falls substantially below the ambient pressure of the water the outward motion stops. Then the higher pressure of the surrounding water reverses the motion at a nearly exponential rate. Again, the inertia of the inward flow of the water causes the bubble to compress and exceed the pressure equilibrium with surrounding water. When the bubble reaches its minimum size, its pressure is several hundred atmospheres. At this point, it is as if a second explosion occurs and the process repeats. The bubble continues through several pulse cycles until it dissipates. The pressure-time history (see fig. 1) reflects the low gas pressure during the phase where the bubble is large and it shows the pressure pulses emitted from the bubble near its minimum size. The period of the bubble pulsations is very long compared with the shock wave portion of the pressure-time history of an explosion. In fact, it is long enough for gravity to affect the bubble. Since a bubble has great buoyancy, it migrates to the surface. However, it does not float up like a balloon, but shoots up in steps.

Modeling The model of the submerged structure was simplified by leaving out detailed features for demonstration purposes. The structure is a cylinder that could represent the pressure hull of a submarine. It was modeled with Lagrangian shell elements, incorporating both a plasticity model, as well as a failure model. This method of meshing is used for structural components that are subjected to large deformations; allowing the dimensions, deformed geometry and residual stress state to be monitored with a high degree of precision.

Figure 1: Pressure waves and bubble phenomena of UNDEX.

Page 1 of 5


The cylinder is 0.6 m long with a diameter of 1.0 m. The covers were modeled as rigid bodies with an appropriate mand center of gravity. If any element exceeds the failure criterion, the structure will fail. Since the boundary of thefinite volume domain is provided by the shell elements of cylinder, once shell element fail flow takes place between inner Eulerian domain and the outer Eulerian domain, a grload is applied to the whole model (see Fig. 2).

MultiplEuler domainwere ufor theinside cylindethe surrouair, waand explosAdditiosince tmodel includewater aexplosmulti--materi

Euler solver was required. The Eulerian method of meshing is used for bodies of fluids and gas, as well as complex material flow. A fast general coupling was used to simulate the interaction between the Lagrangian mesh and Euler mesh. Tcoupled Euler-Lagrange analysis allows combined structural and gaseous fluid simulation. With Lagrangian simulation, the mesh deforms and the material follows the mesh, while the Eulerian gas/fluid analysis keeps tmesh stationary as the fluid and gases flow through the mesh. This short-term transient analysis involves structural parts and/or computational fluid dynamics (CFD) parts. CFD solver uses an Eulerian approach and employs a finite volume method to discretize the governing equatioThese equations are the conservation laws and are integrated in time by a first-order explicit dynamic procedThe Euler mesh is stationary in the space, while the fluid is allowed to flow through the mesh. CFD simulationprovide fluid velocity, pressure fields and other variables. In simulations with fluid-structure interaction, the finside a finite volume domain is bounded by a surface that represents the interacting structure. This surface called a coupling surface and enables the fluid to exert a force on a deformable structure. The unique Adaptive Multiple Euler Domains technology allows single and multi-material Euler solvers. MultipEuler domains are automatically generated around coupling surfaces, and each Euler domain automatically aditself when the coupling surface moves and deforms. The Euler materials can be either hydrodynamic or haveshear strength. Material in- and out-flow can be defined, as well as flow between the Euler domains across por open areas in the coupling surfaces. When the structure fails, the Euler materials flow through the holes, fsurfaces and ruptured areas. This Adaptive Multiple Euler Domains technology allows very efficient modeling UNDEX. By applying the conservation law of mass, the materials of the Euler element are transported from thdonating element to the receiving element. These Euler elements can contain up to five materials in one element. In this case, there are two materials – air and water. With an earlier version, air and water were transported together as percentages. In the latest version of MSC.Dytran, the materials common to both elements are transported out of the donating element and if no common material is left in the donating elemeany remaining materials are transported as well. This approach minimizes unphysical mixing and preserves material interfaces between air and water as much as possible.

Figure 2: Positions of cylinder, water surface and explosive.

Figure 3: Outer Euler domain and its enclosing surface.

Page 2 of 5


To apply the conservation laws for Euler elements that are only partially inside the coupling surface, the boundary of that part of an Euler element inside the coupling surface must be determined. This boundary consists of the interfaces between Euler elements and the intersection of the coupling surface with the Euler element. These intersections are called polpacks (polyhydron packets). The conservation laws are applied to that part of the Euler element inside the coupling surface. The surface integrals are computed by summing across the Eulerian interfaces and the polpacks. Flow from one Euler mesh to the other takes place through porous shell elements that are common to both coupling surfaces. The pressure computation in elements containing a mixture of air and water is based on one of the thermodynamic equilibrium principles, which amounts to pressure equilibrium. In an Euler element with both air and water, there is a distinct pressure inside the air and a distinct pressure inside the water. Although masses are fixed during the pressure computation the volume of air and the volume of water are not fixed. They are adjusted iteratively until the water pressure equals the air pressure. The cylinder consists of shell elements that deform under stresses and support failure models. An explicit finite element solver is used for the shell dynamics and an explicit Euler solver is used for modeling the material inside and outside the cylinder. The interaction between these two solvers takes place in two steps. First, the mass in the Euler elements exerts a pressure load on the cylinder surface. These loads constitute the boundary conditions for the finite element solver, resulting in new grid point accelerations and velocities for the cylinder. The failing elements are determined from the updated plastic strain or updated stresses of the shell elements. Finally, the cylinder grid points are moved with the new velocities. Second, the cylinder grid points move so the Euler mesh has a new boundary. Consequently, the volume of mass in each element may change. Since density is mass divided by the volume of the mass, both densities and pressure change. To model the fluid inside and outside of the cylinder, two Euler domains are used. The outer domain uses the cylinder surface as part of its boundary. The material is outside the cylinder surface. The contents inside the cylinder are modeled in the inner domain. The inner domain also uses the cylinder surface as part of its boundary. Both Euler domains use the cylinder surface as part of their enclosure (see Fig 3 and Fig 4).

The outer boundary of the outer domain is given by a sufficiently large fixed box. Pressure at the outer boundary is set to the hydrostatic pressure. This behaves like open boundary. The Euler mesh contains the water and the air on the top of the water. The density of water is 1000 Kg/ m3. The bulk modulus is taken as 2.2E9 Pa. Water hydrostatic pressure is defined starting from 1.0E5 Pa at the surface and increasing with depth. A minimum pressure of zero is defined for the water, so that if a portion of water had negative pressure, all of the water would flow out of that region, creating a void. The density of air is 1.1848 Kg/ m3. The ratio of the heat capacities of the gas is constant at 1.4. Specific internal energy is taken as 2.14E5 Kg-m2/s2. Initial air pressure was set to 1.0E5 Pa. The explosive TNT is created in this Euler mesh, too. The density of the explosive is 1700 Kg/m3 and the mass is 0.445 Kg. The specific internal energy is 4.765E6 Kg-m2/s2. The explosive can be modeled by a JWL (Jones – Wilkins – Lee) or IG (Ignition and Growth) equation of state. However, if the explosive were assumed to be a ball, its radius would be of only 0.04 m.

Thus, a finer mesh had to be created to simulate the smaller ball. In this simulation, the explosive was defined as a compressed hot gas (m=1.4). The mass and specific internal energy were chosen as those of the explosive charge. The radius of the gas ball is assumed to be 0.1m and the density is adjusted to 105 Kg/m3 to keep equivalent mass of the explosive. Then a very fine mesh for the explosive is unnecessary, allowing CPU time to be reduced. The initial air pressure of the hot gas was calculated to be 2.0E8 Pa. The inner domain is initialized by air. The surface presents the outer boundary of the domain. The meshes of the outer and inner domains do not coincide. The element size for each domain is the same as 0.1m in this simulation.

RESULTS AND DISCUSSION

Figure 4: Inner Euler domain and its enclosing surface.

Page 3 of 5


The simulation was carried out 0.5 seconds for the duration of the formation and collapse of the first and second bubble until the third bubble begins to form. The CPU time for running this model is 34 hours on a Windows 2000 3.06GHz. The first bubble comes in touch with the structure and collapses (see Fig. 5). This causes a bubble jet that damages the cylinder. Then the second bubble collapses on the cylinder and the bubble jet increases the damage.

The effective stress and plastic strain are plotted on the deformed and damaged shapes of the cylinder (see figs. 6a and 6b). The first and second bubble jets emitted near the bubble minimum radius are clearly seen in the plot. Since the explosive is defined as a compressed hot gas, it may affect the shock wave characteristics. However, the bubble formation and collapse are modeled correctly together with bubble pulse loading. Since damage is caused by the rapid expansion and contraction of the gases in the bubble and not by shock waves, an evaluation of shock wave effect was not made.

Figure 5: Isosurfaces of material fractions

Figure 6a: Effective stress plotted on the deformed shapes of the cylinder.

Page 4 of 5


The impact of the shock wave and subsequent gas bubble causes the cylindrical structure to deform and fail, allowing water to flow into the cylinder (see Fig. 7). The unique "Adaptive Multiple Euler Domains" technology of MSC.Dytran makes it possible to model efficiently. Although experimental data is not available, the simulation results confirm expectations.

Figure 6b: Effective plastic strain plotted on the deformed shapes of the cylinder.

Figure 7: Velocity distribution of fluid.

Page 5 of 5


Documents

[ From the Beginning ] - MSC Software...MSC.Nastran, MSC.Patran and MSC.ADAMS from MSC's portfolio of best-in-class technologies. Under BTCC rules only certain modifications to the