44
MSC Software Magazine Volume II | Summer 2012 REALITY simulating Developing New Vehicle Concepts Faster Robotic military vehicle developed in under 18 months using simulation Compressing Product Development Cycles Accurately modeling system performance prior to the prototype phase Engineers Get Design Right The First Time 30% performance improvement using simulation to evaluate design alternatives Mission Critical How Simulation Propelled Pratt & Miller into New Markets Jean-Louis Migeot and Jean-Pierre Coyette, Free Field Technologies (FFT) Interview with Acoustics Experts Motorsports to

Simulating Reality

Embed Size (px)

DESCRIPTION

Summer 2012

Citation preview

Page 1: Simulating Reality

MSC Software Magazine Volume II | Summer 2012

REALITYsimulating

Developing New Vehicle Concepts Faster Robotic military vehicle developed in under 18 months using simulation

Compressing Product Development Cycles Accurately modeling system performance prior to the prototype phase

Engineers Get Design Right The First Time30% performance improvement using simulation to evaluate design alternatives

Mission Critical How Simulation Propelled Pratt & Miller into New Markets

Jean-Louis Migeot and Jean-Pierre Coyette, Free Field Technologies (FFT)Interview with Acoustics Experts

Motorsports to

Page 2: Simulating Reality

MSC UPDATE & NEWS

REALITYsimulating

MSC UPDATE & NEWSIN THIS ISSUE

Leslie Bodnar, Editor [email protected]

Marina Carpenter, Graphic Designer/Assist. Editor [email protected]

Stephanie Jaramillo, Assistant Editor [email protected]

Patrick Garrett, Assistant Editor [email protected]

Lydia Westerhaus, Assistant Designer [email protected]

MSC Software Corporation 2 MacArthur Place, Santa Ana, CA 92707 714.540.8900 | www.mscsoftware.com

5

31

16

35 TECHNOLOGY MATTERS Interview with Acoustics Experts Jean-Louis Migeot and Jean-Pierre Coyette Free Field Technologies (FFT)

Understanding Eye Injuries Numerical modeling and analysis of the mechanism of retinal detachment as a result of blunt impact University of Cassino, Italy

PARTNER SHOWCASE Gathering All the Puzzle Pieces Multi-objective optimization software works in an open environment and automates the entire design ESTECO

Image Based Modeling for Biomedical Implant Design Applying the innovations offered by Additive Layer Manufacturing to solve traditional limitations Simpleware

A Solution Shipbuilders Can Rely On Savings through Idealization of Production Models AVEVA

UNIVERSITY & RESEARCH Creating Future CAE Users Preparing for a Successful Career Srinivas Reddy, MSC Software

A Practical Master’s Degree in FEA Gain knowledge to Better Leverage Advanced Simulation Tools Ingeciber, S.A.

LETTER FROM THE EDITOR

New Possibilities

PRODUCT NEWS IN BRIEF Introducing the New Adams/Machinery New Solution Simulates Gears, Chains and Belts LETTER FROM THE CEO

CAE in the Supply Chain - It’s Time!

MSC UPDATE & NEWS Exchanging Ideas & Experiences Americas, China, EMEA, India, Korea

FEATURE STORY Motorsports to Mission Critical How Simulation Propelled Pratt & Miller into New Markets Pratt & Miller

CUSTOMER SPOTLIGHT Compressing Product Development Cycles Multidiscipline Simulation Helps Bring Gun Turret Drive System to Market Faster General Dynamics Land Systems

Engineers Get Design Right the First Time ITW Delfast Improves Fastener Performance 30% by Using Simulation to Evaluate Design Alternatives ITW Delfast Group

Developing New Vehicle Concepts Faster Robotic Military Vehicle Developed in Under 18 Months Using Simulation BL Advanced Ground Support Systems

SPECIAL: BUSINESS PARTNER SPOTLIGHT Predicting Fatigue Failures with Confidence Virtual Load/Fatigue Process Leading Edge Engineering

Unveiling a Monumental Gold Coin Simulation Plays Vital Role in Creation of Record Breaking 1 Tonne Gold Coin Compumod

TECH TIPS Adams: Creating Flexible Parts Walt Daniel, MSC Software

Patran: Getting Efficient Darrell Sinclair, MSC Software

MSC Nastran: Visualize Glued Contact Cornelia Thieme & Dominick Lauzon, MSC Software

SPECIAL SPOTLIGHT: Helping Aspiring Engineering Students Become Tomorrow’s Workforce MSC Software Works with Students to Solve Real World Problems Caressa Matsuoka, MSC Software

38

40

3

4

7

8

10

24

25

26

14

16

18

20

22

32

34

36

28

30

27

Page 3: Simulating Reality
Page 4: Simulating Reality

w w w . m u l t i c o r p o s . c o m . b r

Multicorpos is a Virtual Product Development company that works with �nite element

analysis, multibody system dynamics analysisand computational �uid dynamics.

0

5

25

75

95

100

Page 5: Simulating Reality

Volume II - Summer 2012 | 3

Letter from

the Ed

itor

Letter from

the Ed

itor

What inspires us to reach for new summits? Often times, it is the idea of new possibilities. In an effort to improve, we seek out opportunities to grow and

advance in both our personal and professional lives. I was reminded of this during my recent visit with Pratt & Miller, a dynamic engineering company located in New Hudson, Michigan.

The impressive team at Pratt & Miller went from developing vehicles under extreme deadlines as a successful designer and builder of race cars to becoming a valued engineering partner outside of racing. As their expertise with virtual testing grew, so too did opportunities for new projects. Recently, the team developed a new military vehicle prototype in just 12 weeks. In our feature story on page 10, you’ll see how their unique capabilities led to business expansion in new markets.

Then there are companies like ITW Delfast Group. The team’s design engineers are achieving 30% improvements in performance in new design concepts. By automating aspects of their analysis process, they’re finding new ways to reduce costs within their company. Check out their story on page 16.

In other customer spotlights, learn how

General Dynamics Land Systems (GDLS) moved from separate simulations to system-level co-simulations beginning on page 14. In this article, you’ll discover how GDLS was able to compress product development cycles with a new approach to virtual testing.

Developing new concepts faster and more accurately is critical to gaining competitive advantage. On page 18, get an inside look at how BL Advanced Ground Support Systems developed a robotic military vehicle in under 18 months – using simulation as their secret weapon!

Please don’t forget to check out pages 38 and 39 to learn how we are helping to prepare engineering students to enter the workforce with free MSC Software Student Editions and on-going educational initiatives that are building a knowledgeable ecosystem of future CAE engineers.

We hope these stories and others inside this issue give you new insights for finding your own “new possibilities” for gaining competitive advantage.

Sincerely,

New Possibilities

Leslie

We hope these stories give you new insights for finding your own

“new possibilities” for gaining competitive advantage.

Page 6: Simulating Reality

4 | MSC Software

PRODUCT NEWS IN-BRIEF

Introducing the New Adams/MachineryNew Solution Simulates Chains, Gears, and Belts

MSC Software recently introduced Adams/Machinery, available inside the Adams/View interface (Adams lets users build and simulate models of any mechanical system with moving parts.) Adams/Machinery was developed to help manufacturers of machinery equipment and other mechanical systems like cameras and power tools efficiently build functional virtual prototypes of components and systems early in the design cycle, before building physical prototypes. The solution includes customized productivity tools for modeling and pre-processing chain, gear, and belt components.

The clean-looking interface includes in-line help and information about components, their connections, and various modeling-fidelity options. Wizards help guide users through model setup, manipulating model parameters, and modeling options.

For instance, in the gears module, users can simulate the backlash of a gear pair in a streamlined fashion using a gear-creation wizard. And in the belts module, users can predict belt tension and loads using pulley and belt wizards. Additionally, in the chains module, users can study contact forces between sprockets and chains.

In the gear-creation wizard, engineers can choose either the simplified modeling method, which neglects friction and quickly calculate the contact force between teeth, or a 3D contact-modeling method to study the backlash based on the actual working center distance and tooth thickness. Adams/Machinery supports a variety of gear types, including spur, helical, straight bevel, and spiral-bevel configurations.

In the belts module, engineers can predict the tension and load of belts in models built using the pulley and belt wizard. Users can also run design studies to find the proper tensioner stiffness to minimize slippage and minimize peak belt loads. Adams/Machinery includes Poly-V grooved belts as well as smooth and toothed belts.

The chains module lets engineers study contact forces between sprockets and chains and examine how the change of contact forces affects the overall mechanical system’s performance. The software supports involute and roller chains.

The new modules make engineers more productive. The table below shows how much time can be saved using the new modules as opposed to creating components directly using the traditional method (i.e., using Adams/View alone to build them).The gear module lets users perform such operations as calculating backlash.

Page 7: Simulating Reality

Volume II - Summer 2012 | 5

Traditional Method

Using Adams/Machinery

Time saved

Base model creation (frame and shafts)

0.25 day 0.25 day 0

Building the gear box 1.0 day

(requires external geometry and limits modeling method options)

0.25 day

(external geometry unnecessary; more modeling method options)

0.75 day

Building the belt system 5.0 days 1.0 day 4 days

Building the chain system 5.0 days 1.0 day 4 days

Post-processing (defining desired output)

0.5 days 0 days 0.5 days

Total time 11.75 days 2.5 days 9.25 days

The belt and chain systems which once took even proficient Adams/View users a week to finish can be completed in one day using Adams/Machinery. The total time saved for this speed-reducer model using new Adams/Machinery is over 9 days, which represents a significant cost savings for engineers, which translates to cost savings for the company.

In summary, Adams/Machinery provides engineers with an approachable, easy-to-use, high-fidelity modeling solution that can drive quality up and costs down. This first release of Adams/Machinery includes high-fidelity simulation capabilities for gear, chain, and belt systems. In addition to handling chain, gear, and belt systems, future releases of Adams/Machinery are planned which are expected to include modules for bearings, cables, and electric motors. u

How gear components look in Adams/Machinery.

The chain drive was built using the sprocket and chain wizard.

The pulley-belt system was built in Adams/View using the new belt module.

Adams/Machinery was used to build a speed-reducing mechanism, which consists of a gearbox as well as a chain sprocket and belt-pulley system.

Page 8: Simulating Reality

2012 AEROSPACE USERS SYMPOSIUM

September 18-19, 2012

The Westin Washington Dulles Airport2520 Wasser TerraceHerndon, Virginia 20171

Learn more: www.mscsoftware.com/aero2012-reg.com

2012 INTERNATIONAL GROUND VEHICLE USERS SYMPOSIUM

October 23-24, 2012

The Dearborn Inn, Marriott Hotel20301 Oakwood BoulevardDearborn, Michigan 48124

Learn more:www.mscsoftware.com/vehicle2012-reg.com

2012_UserConf-FP2.indd 1 8/21/12 10:59 AM

Page 9: Simulating Reality

Volume II - Summer 2012 | 7

CAE in the Supply Chain

It’s Time!Expect an explosion in the flow of mission critical behavioral models through the supply chain as one of the new realities in engineering.

It has well been written over the past 20 years that product development continues to be global. There is just no large company in the world that can rely on the entirety of its product being designed with one team sitting in one location for an entire design cycle. Skill shortages, outsourcing, and suppliers scattered around the world are a few reasons. If you read the myriad of articles written on how to smooth the process, the main topic is about geometry and how to make it flow through the process.

But simulation is beginning to take center stage. Parts and assemblies not only need to fit, they need to perform over the long term. Every CFO who has to account for expensive warranty returns and recalls may not know what the solution is, but knows how important product quality and performance is to the business.

I hear many auto and aero OEMs expressing concern because they cannot receive reliable simulation models from their suppliers or no models at all.

If a supplier provides CAE data today, it is generally static pictures or documents. This was the same as the 70’s and 80’s when drawings of geometric shapes were sent from supplier to the OEM. In the simulation world, static performance data such as frequency response functions, strength allowable, transfer functions, and a myriad of other plots, do not enable the OEM to utilize the data effectively in the simulation of the suppliers’ components effects on the overall system.

As simulation becomes prevalent in design from the system to detail, from conceptual to final, a new way of working is a must and the technologies to enable this new work method are within sight. The new way will enable the supplier to instrument their design and deliver the behavioral representation along with the static data.

The computational/behavioral representation enables an OEM to create a systems simulation model that can be used to select vendors or to refine the performance criteria for the component. Virtual component deliverables enable a bi-directional exchange that lets the systems integrator make the best component decision and lets the supplier tune a component optimized for the system’s requirements.

Many suppliers and OEM’s worry that their technology or “know-how” or unpatented IP information will leak to their competitors through this CAE model sharing. To enable this

kind of exchange, CAE systems need to allow exchange of behavioral representations at various levels of fidelity and with an ability to restrict access to detailed IP in the component model.

What is required is a virtual part data model that contains the behavioral representation in detail, but hides that

detail with a more “block diagram” view of the component. This data model only exposes the

“instrumented” simulation model that contains the required inputs/outputs to allow the virtual component to be part of the OEM assembly and to respond to the “true” boundary conditions that are derived from the systems simulations. This exchange could also allow “sensors” to monitor the behavior of key results anywhere in the component.

In the initial stages of a virtual design, there will remain (even in this new method) the same kinds of approximations and ranges of error as do today with the physical prototypes. Eventually, as the decisions become more refined, the data will become more accurate and the behavior more precisely simulated. Ultimately, test validation will still occur…but design will be largely (and ultimately completely) determined by simulation.

Going forward, expect a new and better way of working and expect MSC Software to deliver the tools to make it happen. u

Letter from

the CE

OLetter fro

m the C

EO

Dominic GallelloPresident & CEO

MSC Software

Going forward, expect a new

and better way of working and expect

MSC Software to deliver the tools to make it happen.

Page 10: Simulating Reality

8 | MSC Software

MSC UPDATE & NEWS

Exchanging Ideas & Experiences

MSC AMERICAS Top 5 Reasons to Attend MSC Technical Workshops & Seminars:1. 93% of attendees surveyed said

they would recommend to a colleague.

2. Presenters are experts in the field, consistently ranking high (5/5) on feedback surveys.

3. The events are FREE!

4. All attendees receive a take home evaluation license of MSC software products that include the latest tools and technologies.

5. The events provide a valuable networking opportunity with attendees representing a wide range of industries and companies.

Over the past six months, the MSC Americas Team has connected with approximately 700 workshop and seminar attendees. A few of the FREE technical seminars and workshop titles have included:

• Composites Technology Day with MSC Nastran (workshop) This workshop demonstrates the capabilities within MSC Nastran for advanced composite material designs. The hands on approach walks attendees through basic shell composites modeling, solid composites, progressive ply failure and delamination and optimization

of composite structures and simulation of manufacturing processes. This workshop has been featured in 6 different cities.

• Acoustic Theory & Numerical Simulation (seminar)

• Hyperelastic Materials Characterization for Finite Element Analysis (workshop)

Register for a workshop and seminar today! For a full list of upcoming events in the Americas, please visit www.mscsoftware.com/Events-Americas.

MSC CHINA Aero/Astro User Conference 2012MSC China connected with over 400 customers at the MSC Aerospace and Astro User conference in Xian and Beijing May 29—June 1! Customers in the aerospace, astro and national defense industries attended to get a first look at MSC’s 2012 roadmap and interact with domain experts on topics such as composites, Aeroelastic applications, Superelement Technology and CFD. Conference participants enjoyed learning about a special customer application spotlight from Airbus and agreed that the event as a whole was both valuable and productive.

Auto Seminar in ChangchunChangchun is one of the biggest auto production centers in the country. This region is considered a pioneer in the automotive industry by actively adopting CAE in design and analysis. MSC China engaged with the technologically advanced automotive culture through an industry focused seminar on June 13th. Attracting more than 70 OEM and supplier attendees, the event was a clear success.

Teaming Up with UniversitiesMSC China continues to connect with students through key initiatives. MSC Campus Roadshows conducted at Universities in Beijing, Shanghai and Xian throughout March and April were successful at introducing students to new versions of MSC Software products and providing them with internship opportunities through MSC’s intern program.

MSC China is also sponsoring 7 Universities at the 2012 China Formula SAE competition, taking place in October. With over 40 teams competing, MSC China was eager to provide students with the competitive edge by providing special trainings to improve their knowledge of the software quickly. Stay tuned for 2012 winning results!

MSC EMEA Business Partner Summit 2012The 2012 EMEA Business Partner Summit took place in Istanbul, Turkey on May 3rd and 4th. Eddy Fadel, Channel Business Director EMEA, and Kais Bouchiba, Vice President MSC EMEA, opened the Summit, to welcome 57 MSC Business and Technology partners. Participants gathered to learn useful information from Channel Management, the MSC Technical Team and Marketing but most importantly to share their experiences.

Conference highlights included a special customer spotlight presented by Mr. Orkun Tarkci, Research and Development Manager for Sarsilmaz

NEWS

Page 11: Simulating Reality

Volume II - Summer 2012 | 9

Silah. Mr. Tarkci’s paper described the successful usage of MSC solutions for the development of light firearms. Other highlights included a special segment on Actran and conference workshops that provided partners with increased working knowledge and better sense of the capabilities of MSC products.

Awards honoring 5 winners were presented at the end of the conference:

• Top Performance 2011 – CSOFT, JSC, Russia

• New Logo Champion 2011 – Eksen, Turkey

• Fastest Growing Partner 2011 – Magic Engineering, Romania

• Major Service Project 2011 – InSumma, Netherlands

• Best Academia Focus – Esteq Engineering, South Africa

Italy Roadshow 2012In June, the MSC Software Italian team organized three days of conferences and workshops in collaboration with industry companies and three important universities, Politecnico di Milano, Politecnico di Torino and Università La Sapienza di Roma to provide solutions to pivotal questions. Each event was centered on a specific industry including machinery, aerospace and automotive and hosted a large mix of engineers, key users, students, teachers and researchers.

Speakers from industries and universities presented case studies and real life examples of how CAE solutions and MSC products optimize and simplify their engineering processes from the earliest stages of design. Attendees were also informed about the latest state of the art CAE Technology and future developments in the Machinery, Automotive and Aerospace industries to meet present and future technological challenges.

MSC France brings Industry and Education TogetherMSC is dedicated to developing strong strategic partnerships between industry companies, education and MSC Software solutions. To support these efforts, MSC France has developed interactive initiatives to build and strengthen these relationships.

MSC France is successfully organizing several networking events to provide opportunities for meaningful interaction between these groups. Stakeholders meet to exchange ideas, experiences and needs. Universities are invited to present their courses and students to meet with the companies attending the events. Industry attendees look forward to meeting students for open internship positions within their organizations. To encourage student internships at industry companies, MSC is now offering free licenses to all companies hiring students.

If you are a company seeking to infuse up-and-coming talent with valuable CAE experience into your organization by way of student internships, contact MSC France at [email protected]. Use Subject Line: Industry + Education Opportunity.

MSC KOREA Korea User Conference 2012The MSC Software 2012 Korea User Conference was held in COEX Intercontinental Seoul, Korea on June 8th. Now in its 23rd year, the conference has become the oldest as well as the largest simulation and analysis event in Korea. The event is renowned in the region for providing the opportunity for engineers to network with peers to exchange valuable information, ideas and experiences.

Three key takeaways from the 2012 conference recognized by attendees were 1) More Technical Presentations 2) A special University track and 3) a “Best Technical Presentation” award to 8 presenters. Each of the four tracks consisted of 9 exceptional technical presentations given by 36 industry experts. The Dynamics track was remarkably popular, with attendees acknowledging the value in the exploration of Dynamics, currently a big trend in the region.

The University track was a first time, special event for students and researchers to network with industry professionals at the conference. Many university students and professors enthusiastically participated in the event and offered MSC Korea some great suggestions for future events.

Conference attendees chose the 8 “Best Technical Presentation” winners. The winners were thrilled to receive a lucky gold key and celebrated recognition from their peers. To download the outstanding 2012 Korea User Conference presentations by industry experts, please visit: www.mscsoftware.com/Korea-User-Conf.

MSC INDIA NVH Symposium 2012

“Mute that Unwanted Noise” was the theme for MSC Software NVH Symposium, held in 3 cities across India through February. A total of 211 participants from 76 companies attended the event to make it one of the most successful events of 2012. With the recent addition of Actran Acoustic Simulation in the MSC Product suite, users can stay within the MSC product suite for end-to-end NVH simulation (with MSC Nastran & Actran). Users learned techniques to integrate MSC Nastran and Actran through the

“Actran for Nastran” module, which gives users access to the best of both Nastran NVH Solutions and Actran Acoustics.

The event also provided a great platform for participants to discuss domain challenges with product experts, Stefan Thynelius from MSC Sweden and Julien Manera from FFT Belgium. The Actran Case Studies presented at the event were especially exciting for participants. Most importantly, the symposium provided attendees with a good understanding of how companies around the world are using advanced simulation technology to design better and faster. According to a recent Indian industry report, the focus on Durability, Safety & NVH is increasing amongst industry segments such as Automotive and Aerospace. MSC Software is poised to equip analysts with the right tools and technology to achieve their objectives efficiently.

The India User Conference 2012 is taking place in Bangalore on September 13-14. Register today, www.mscsoftware.co.in.

Page 12: Simulating Reality

10 | MSC Software

FEATURE STORY

Pratt & Miller learned how to develop vehicles under tight deadlines and get them right the first time as a highly

successful designer and builder of race cars. In 2005, the company created an Engineering Services Division to bring the same skills to industrial customers. The company found a niche developing showcase vehicles, fully engineered working prototypes, for defense contractors, under deadlines as short as a few months. Vehicle dynamics simulation using MSC Software’s Adams software plays a key role by making it possible to evaluate and optimize the performance of critical vehicle subsystems long before prototypes and even detailed CAD models of the vehicle are available. Recently, the company created a prototype of a new wheeled military vehicle in only 12 weeks. The ability to develop showcase vehicles so quickly has helped the Engineering Services Division increase its revenues by a factor of 100 and its engineering staff by 122 people in just 7 years.

Roots in RacingFounded by Gary Pratt and Jim Miller in 1989, Pratt & Miller focused exclusively on racing during the company’s early years. The Pratt & Miller team played a key role in eight consecutive GT1 manufacturer and team championships for Chevrolet and Corvette Racing in the American Le Mans Series together with 7 class wins in the 24 hour LeMans, the words most prestigious sports car race. The company also implemented Cadillac’s 1st factory race program that delivered manufacturers’ and drivers’ championships, changing the public perception of GM’s premium brand. In addition, Pratt & Miller-built Pontiacs and later Camaros have earned team, manufacturers’ and drivers’ championships in the Grand-Am Rolex Sports Car Series.

In 2005, Pratt & Miller decided to diversify into other industries, creating the Engineering Services Division. In 2008, when difficult economic conditions caused a downturn in corporate

How Simulation Propelled Pratt & Miller into New MarketsPratt & Miller | Based on an interview with Jesper Slättengren

Motorsports to Mission Critical

Page 13: Simulating Reality

Volume II - Summer 2012 | 11

This type of fast turnaround has enabled us to develop a thriving

business supporting defense contractors and other wheeled

vehicle manufacturers.

Page 14: Simulating Reality

12 | MSC Software

support for racing, this became an important division. “In motorsports, you have to have fast turnaround, and there is no room for errors,” said Jesper Slättengren, Manager Modeling & Simulation for Pratt & Miller. “It was normal for us to develop a race car in six months compared to three or four years typically required by major automobile OEMs. We achieved fast turnaround through our expertise in the use of advanced computer-aided tools and processes. We felt that these same skills might help us in the engineering services business.”

The division targeted the automotive OEMs and their suppliers and defense agencies and contractors specializing in wheeled vehicles. The latter turned out to be an important niche for Pratt & Miller. Defense contractors are typically given hundreds of pages of specifications for a new vehicle. They must submit lengthy written proposals for their proposed solution and often a showcase vehicle as well. The proposals have tight deadlines and the contractors, who are used to working on huge projects that extend for years or even decades, often aren’t able to turn around a complete vehicle in that timeframe. Now they have the option of hiring Pratt & Miller Engineering Services to do that work.

Simulation Enables Development of a Fully Engineered Military Showcase Vehicle in Only 12 WeeksMSC Software’s Adams vehicle dynamics simulation, which has long been a key

component of the company’s race car development, has turned out to be equally critical to the success of these projects. “With showcase vehicles, there is no time to build and test,” Slättengren said. The Engineering Services Division uses Adams/Car to quickly build and test functional virtual prototypes of complete vehicles and vehicle subsystems. Slättengren appreciates Adams/Car’s specific functionality for automotive development, such as modules for chassis, tire, driveline, and driver simulation. “Adams/Car is stronger in automotive-type modeling than any of its competitors,” he noted.

Working in the Adams/Car environment, Pratt & Miller engineers simulate vehicle performance under actual road and off-road conditions, performing the same tests their prototype vehicles will eventually face in a test lab or on a test track. With Adams, however, these tests are performed in a fraction of the time, which is particularly important in situations such as prototype vehicle development, when components such as springs can have lead times as long as 8 weeks. “There is no way you can physically test 4 or 5 different types of springs when your timeframe to deliver the prototype is only months,” Slättengren said.

In addition to using Adams/Car to simulate the performance of different vehicle configurations, engineers in Pratt & Miller’s Engineering Services Division use the software to generate loads for finite element analysis (FEA). “A good Adams model is absolutely critical for

FEA,” Slättengren explained. “Without it engineers would be working in the dark.”

Blank sheet to Prototype in 12 weeks

To illustrate the value of Adams simulations in situations where prototype vehicles must be designed and built extremely quickly, Slättengren offers the example of a project his division recently did for a defense contractor. The contractor wanted to respond to an RFP issued by a branch of the military. In addition to the paper bid, they needed to submit a prototype vehicle that met the project’s requirements. Believing the timeframe was too short for them to create the prototype in-house, they hired Pratt & Miller Engineering Services to do that work, giving them a deadline of 12 weeks.

The military had drawn up a list of several hundred specifications for the vehicle, including how many occupants it must hold, the turning radius, minimum speed over half-round obstacles of various heights, occupant impact limits over different off-road profiles, minimum lateral acceleration during cornering (i.e. 0.5g to 0.6g without wheel liftoff), and so on. Tire size, power train and transmission were specified by the contractor.

Pratt & Miller engineers used Adams/Car to design the front and rear suspension during the first few weeks of the project, long before there was a complete CAD model of the vehicle. One Adams analyst and two to three designers worked full-time on the front suspension while a similar team focused on the rear suspension. Adams simulations were used to evaluate and tune the suspension according to the specifications and also to supply loads to the designers for use in their FEA analyses.

The real-time collaboration between the Adams analysts and designers resulted in a rough suspension design at about three weeks into the project. “The suspension was about 90% done from a topology

A good Adams model is absolutely critical for FEA. Without it,

engineers would be working in the dark.

Page 15: Simulating Reality

standpoint,” said Slättengren. “Some suspension components would still change because they needed to be weight-optimized, but we knew where everything would connect.”

From there they created a full vehicle model in Adams to evaluate the loads on the driver and passengers, making sure the impacts fell within the specified range. “Doing a traditional iteration of springs, dampers and anti-roll bars can take many, many weeks, or even months because there are a lot of conflicting requirements,” explained Slättengren. To speed up the process, his group used HEEDS optimization software from Red Cedar Technology to set up a series of Adams analyses that automatically simulated spring, damper and anti-roll bar combinations through a range of sizes and properties. By using HEEDS to automate the optimization process, the team accomplished in a weekend what would have taken months by hand.

Five weeks into the project, the majority of the Adams simulation was done and all of the vehicle’s key performance variables were set. Adams was only needed after that point to

re-optimize springs and dampers whenever the mass distribution of the vehicle changed. The next week was taken up by detailed design and creating component drawings. By week six the company had started building the frame. At week 8 they put in the suspension. The remaining time was devoted to body work.

Business Expansion

“We think this is the fastest an engineered vehicle has ever been designed,” Slättengren concluded.

“This type of fast turnaround has enabled us to develop a thriving business supporting defense contractors and other wheeled vehicle manufacturers. Our Adams expertise has been the cornerstone of the success of our Engineering Services Division. In our division, we have about 60 years combined Adams experience, which is probably more than any other North American consulting team outside of MSC. This expertise has helped us grow from a standing start in 2005 to eight-digits in revenues this year.”

Please visit www.prattmiller.com for more information. u

Page 16: Simulating Reality

CUSTOMER SPOTLIGHT

Compressing Product Development CyclesCo-simulation enabled GDLS engineers to accurately model the performance of the complete system prior to the prototype phaseGeneral Dynamics Land Systems | Based on an interview with Zhian Kuang

The gun turret drive on a combat vehicle presents a very complex design challenge.

When the vehicle travels over rough terrain, the gun turret drive compensates for the vehicle’s motion and keeps the gun pointed precisely at its target with 99.5% accuracy. In the past, General Dynamics Land Systems (GDLS) engineers used separate simulations to evaluate different aspects of the gun turret drive design, such as the rigid body structures, flexible bodies and control system. But engineers were not able to evaluate the performance of the gun turret drive as a complete system until they built and tested prototypes.

In the last few years, GDLS engineers have begun using a multidisciplinary-based co-simulation process to model the operation of the gun turret drive system while taking into account all of the key physics involved in its operation. The centerpiece of this simulation

effort is the use of Adams dynamics software to model the rigid bodies, nonlinear joints and contacts in the gun turret drive. Adams was selected because of its nonlinear contact capabilities. The accuracy and relatively few assumptions required by this approach provide more accurate simulation predictions and reduce the time required for troubleshooting physical root causes, resulting in a significant reduction in time to market.

Need to Consider Multi- Physics in Design ProcessGDLS builds a variety of combat vehicles such as the Abrams M1 tank, Stryker mobile gun system and MRAP blast- and ballistic-protected personnel carriers. Designing these products for optimal performance requires consideration of a wide range of physics including rigid body structures, flexible bodies,

suspension systems, nonlinear body to body contact, nonlinear large scale deformation, thermal, electrical, electromagnetic, fluid, and others.

GDLS has developed the capability to model each of these physics individually and single-physics simulations are performed frequently during the design process. But each of these simulations is heavily dependent on other physical processes outside its scope which makes it necessary to make assumptions that have a negative impact on accuracy. Traditionally, designers working on actuators, controllers and associated electronic circuitry have to wait for mechanical hardware to be procured and tested before tuning their systems to meet mechanical requirements. This process is normally the controlling factor for the delivery leadtime of a new product.

Overview of Co-Simulation ProcessThe new co-simulation process cooperatively solves the dynamic and nonlinear contact behavior of the mechanical system interacting with the discrete behavior of the digital motor/controller system. The co-simulation process is initiated and controlled from the MATLAB/Simulink environment using the “Adams plant model” (which accounts for all rigid body dynamics and flex body dynamics). Co-simulation allows the control system model to process discrete models using the ode4 Runge-Kutta integrator based on the variables received from Adams. At the completion of each time step, Simulink sends its output to Adams and waits for the Adams solver to calculate the solutions to its set of variables using the GSTIFF integrator. Upon solving

14 | MSC Software

Page 17: Simulating Reality

Volume II - Summer 2012 | 15

the state variables, Adams sends its data over the PIPE communications line to Simulink and the process advances to the next time step.

Control models were simulated in Simulink. The control system was designed for two modes of operation. Inertial stabilized mode stabilizes the weapon in space with respect to perturbations in pitch and yaw. Non-stabilized mode controls the gun in elevation and azimuth in the local reference frame of the vehicle. Control algorithms were designed to include compensation for both the rigid body dynamics along with the bending modes of the gun/cradle system.

Modeling of Mechanical SystemsThe gun turret drive CAD model was imported into Adams including all of the assembly tolerances required for the final product release. In complicated machinery it’s very common for cumulative tolerances to cause problems that aren’t identified until prototype testing. The Adams model overcomes this problem by incorporating the tolerances of the individual components and providing 3D redundant constraints that incorporate the impact of the cumulative tolerances. This makes it possible for engineers to determine the impact of tolerances on product behavior and investigate the impact of tightening or loosening tolerances prior to the prototype phase.

In this highly complex weapons system, the ability to account for nonlinearities is critical to accurate simulation. The key advantage of Adams is that it accounts for the nonlinearities

in this system through its ability to model nonlinear on/off contacts, large displacements associated with part deformations and nonlinear materials.

Mechanism analyses are done in Adams to determine which bodies can remain as computationally economical rigid bodies and which need to be converted to computationally more intensive flexible bodies. For example, if the first mode frequency of a component is well above the frequencies likely to be experienced in operation then it is normally modeled as a rigid body. The first mode for the assembly consisting of the gun, breech block and adapter is within the range of interest in free-free modal analysis and also when constrained by assembly conditions so it was modeled as a flexible body. On the other hand, the first mode of the cradle is outside the bandwidth of interest when constrained for assembly conditions so it was modeled as a rigid body.

DC gear motors were used as actuators to control the elevation of the gun. Each motor was mechanically modeled in Adams including the motor brake, rotor, stator and a geared output shaft. GDLS electronic circuitry for motor controllers was modeled in Simulink. Torque commands from the Simulink control system model are assigned to the output shaft. The output shaft gear engages a sector gear that drives the gun assembly in the reference frame of the turret with the motion profile incorporating the assembly tolerances.

Comparison to Measured DataThe frequency response of the Adams plant was compared to measured data. The simulation was run in assembly mode incorporating the full range of potential design positions of all parts based on their tolerances. The percentage difference between the simulation predictions and physical measurements for the frequencies of the first five modes were respectively 3%, 18%, 10%, 10% and 17%. Allowing the system to settle with gravitational loading closed up some tolerances and reduced the difference for the four and fifth modal frequencies to 0.1% and 0.25%. It’s important to note that these simulation predictions were generated on the first cut

analysis without tweaking the model to match the measured data. Adjusting the details of the joints at the attachment points affected the frequency of modes 1 and 2, allowing engineers to understand how joints and contact forces influence the frequency response of the mechanical system. GDLS engineers compared the performance of simple bushings, 3D contacts, classic joints and forces to understand what each component contributed to the responses measured in the prototype.

Detecting and Troubleshooting JammingOne of the most critical considerations in the gun turret drive is the potential for jamming, a condition in which side loading on rotating components causes the shaft to flex and increases the frictional force between the shaft and bearing. In some cases, the frictional force can rise to a level that stops the shaft from rotating. Components subject to jamming are normally modeled as flexible bodies to increase simulation accuracy.

One of the key benefits of Adams is its ability to identify jamming prior to the prototype phase through the use of nonlinear contacts in which friction varies depending on the loads on the shaft and bearings and other factors. In this application, Adams identified a jamming condition even though the designer was certain that the shaft would not jam. Later when the prototype was built it was determined that jamming did occur in the original design. The early identification and understanding of the problem saved a considerable amount of time and money in troubleshooting.

A key advantage of Adams is that relatively few assumptions are required for the plant model. The accuracy and relatively few assumptions required for the multiphysics approach enables GDLS engineers to understand, negotiate and trade off both upstream and downstream requirements with much greater visibility to their impact on system requirements than was possible in the past. Knowing the impact of subsystem performance based on physics enables the system integrator to play a more active role in requirements and cost control tradeoffs as opposed to be solely driven by suppliers’ perspectives as often occurred in the past.

In summary, multiphysics co-simulation requires fewer assumptions and is easier to use for troubleshooting physical root causes than other computer aided engineering methods. Co-simulation of the gun turret drive enabled GDLS engineers to accurately model the performance of the complete system prior to the prototype phase. The ability to identify problems and evaluate potential solutions and to understand the effect of component specifications on system performance helped significantly compress the product development cycle. This is why GDLS engineers called the model a ‘Virtual Machine’ built by using Adams. And the virtual machine is producing cost savings, especially in product R&D processes. u

With Adams, the early identification & understanding of the

jamming condition in the gun turret drive saved a considerable amount of time and money in

troubleshooting.

Page 18: Simulating Reality

16 | MSC Software

CUSTOMER SPOTLIGHT

The ITW Delfast group designs and produces engineered plastic and metal

fasteners for the automotive industry. The plastic fasteners typically are secured by a clip snapping into contact with a serrated shaft. These fasteners present a difficult design challenge because of the complexity involved in multiple contacting bodies undergoing large deformations with sliding contact. In the past, an experienced analyst performed finite element analysis, but this was expensive and time limitations meant that only the more difficult designs could be analyzed. In the cases where there wasn’t time to analyze the design, it was often necessary to modify the tooling at an average cost of $1500.

To address these challenges, the company developed a method that enables design engineers with little or no computer aided engineering (CAE) background to perform the analysis and produce good results. Providing design engineers with analysis capabilities makes it possible to analyze nearly every new design. The engineers are typically able to improve the design by evaluating two or three alternatives to the original design. The new approach helps engineers get the design right the first time, eliminating the need to remake the tooling.

The ITW Delfast Group includes divisions in Brazil, China, France, Germany, Italy, Japan, Spain, Sweden and the United Kingdom. Its customer base includes all the major automotive original equipment manufacturers (OEMs) as well as Tier 1 and Tier 2 suppliers. The group’s product line includes trim clips, hole plugs, hot melt plugs, brake and fuel line routing clips, watertight fasteners, sound seal screws and multi-blow products. The ITW

Delfast Group is a unit of ITW which has 825 decentralized business units in 52 countries that employ approximately 60,000 people.

Complex Analysis Challenge Figure 1 shows a finite element analysis of a typical fastener produced by ITW Delfast. The company’s plastic fasteners nearly all require nonlinear analysis because components, particularly the clip which is shown in green with a fine mesh in Figure 1, undergo large deformations. Additional analysis challenges include the need to incorporate friction and plasticity into the analysis and address rapidly changing contact conditions as the clip snaps into contact with the shaft. In addition, complicated combinations of boundary conditions and load sequences are often needed to address the varying loads the fasteners will experience during assembly and in use.

The company has design teams in each of the countries where it operates and these teams work closely with customers to develop a large number of custom designs to address new applications. In the past, Kristian Ostergren, Head of Design for ITW Sweden, was the only person capable of performing these complex analyses. His time was occupied with other responsibilities so that he could only perform analysis on a few of the more difficult applications. “The result was that in the past most of our custom products were designed using the build and test method,” Ostergren said. “Sometimes, the performance of the initial design was unsatisfactory. In that case we had to change the design and remake the tooling.”

In the applications where the company was able to perform analysis, it was usually possible to improve the performance of the product and nearly always to meet the design requirements with the initial prototype, eliminating the need for retooling. Ostergren wanted to provide design engineers with a tool that would enable them to perform nonlinear analysis despite their limited analysis background. “I looked into CAD integrated solutions that are designed to provide non-CAE savvy users with the ability to perform analysis,” Ostergren said. “But I discovered that these tools do not provide the capabilities to perform the complex nonlinear analysis required for our products. In addition, our divisions use a number of different CAD systems, which would have made it difficult to implement a CAD integrated solution.”

Engineers Get Design Right the First TimeITW Delfast Improves Fastener Performance by 30% Using Simulation to Evaluate Design AlternativesITW Delfast | Based on an interview with Kristian Ostergren

The template has captured,

implemented and a large extent automated our analysis best practices and

put them into the hands of our design

engineers.

Page 19: Simulating Reality

Volume II - Summer 2012 | 17

Capturing Best Practices in a TemplateOstergren identified MSC Software’s SimXpert as a tool with the potential to capture the CAE process and analysis best practices into a simple tool for use by design engineers. SimXpert enables expert analysts to develop templates that cover all stages in the simulation process including modeling, job setup, solving and reporting within one integrated workspace environment. SimXpert templates are more than just a macro in that they provide full access to database objects and methods, support custom actions written in Python and offer high level loops and conditional branching. Templates can be designed by recording a sequence of operations or by dragging and dropping operations into the template workspace.

SimXpert templates can utilize linear and advanced non-linear, static and dynamic structural analysis based on the complete solution set provided by MSC Nastran. They can also predict loads and analyze system motion including flexible bodies based on MSC Software’s Adams capabilities. MSC Nastran can also be used to perform large deformation, highly nonlinear, short duration transient analyses for structural impact and coupled fluid-structure interaction problems.

Ostergren and MSC worked together to develop a template that provides a high level of automation while enabling users to interact with the analysis in order to ensure

that it accurately represents the current design. The template fully automates the process of defining and naming parts and properties, generating symmetry constraints, defining contact bodies, load set generation, analysis setup, job submission and report generation. The template can be run either step by step in interactive mode or in semi or full automatic mode.

Ensuring a High Quality Mesh The template simplifies the generation of a good mesh which is the key to a robust analysis and accurate results. The template checks the dimensions of the part and suggests the element sizes to be used in the different regions. These regions are color coded as shown in Figure 2. The user can then fine tune the element size selections such as by refining the mesh in regions where contact or high stress concentrations occur.

The template then performs an automated process to adjust the locations of the nodes to improve element quality. Materials are added by entering the modulus, stiffness, strain limits at yield and break, and Poisson’s ratio from the data sheets. The template offers an efficient way to manage and share material data between different users by making any new material automatically accessible. The entire reduced set of setup parameters is collected in a single window as shown in Figure 3. Robust defaults are provided for analysis and contact settings.

The template makes it fast and easy to change geometry, materials or boundary conditions and reanalyze the new design. The user checks off whether he or she would like to change the material, geometry, contacts and loads and boundary conditions and then the template is automatically re-run, stepping

through the process of making the desired changes.

Iterating to an Optimal Design The template also assists the user in maintaining the structure of the analysis directory tree. Each analysis for a particular application is stored in a directory with a unique filename. This approach simplifies the creation of a report comparing the different design alternatives shown in Figure 4. The report compares key analysis results for each of the selected cases. The deformed geometry of each design alternative is compared side by side. Graphs showing the load history overlay are provided for each design alternative.

Design engineers view color plots to see the strain concentrations. Typically they will redesign the fastener to spread the strain over a large area of the part in order to reduce strain values. Once they have reached a point where the analysis shows that strains are within acceptable limits, analysts often evaluate the possibility of using a less expensive material. “During this iteration process, our design engineers can typically make a 30% improvement in performance compared to their initial design concept,” Ostergren said.

“This helps to achieve a larger margin of failure which in turn avoids damage due to misuse and makes the product last longer.”

Design engineers were provided with a training session that focused on the critical aspects of the analysis process. “The template makes it easy to generate results – the training was designed to ensure that design engineers can generate good results,” Ostergren said. The training helped to build an understanding of the analysis setup and the solver. The importance of generating high quality elements during the meshing process was explained. Critical interrogation of the results was emphasized in order to identify potential errors.

“The template has captured, implemented and a large extent automated our analysis best practices and put them into the hands of our design engineers,” Ostergren concluded. “The results have included substantial reductions in analysis time, improved design performance, and reduced prototyping and manufacturing costs. Furthermore, by using the template, our users are gaining confidence and competence to increase complexity and run new analyses beyond the template.” u

Figure 1: Analysis of typical ITW Delfast fastener

Figure 3: Setup parameters are contained in a single window

Figure 2: Color coded surfaces help users visualize current mesh settings

Figure 4: Report compares performance of selected cases

Page 20: Simulating Reality

18 | MSC Software

CUSTOMER SPOTLIGHT

Robotic Military Vehicle Developed in Under 18 Months Using Simulation

BL Advanced Ground Support Systems (BL) specializes in developing vehicles

used by air and ground forces. In the past, when the company relied on outside consultants for simulation support, it found that considerable time was wasted in communications and waiting for simulation results. Building the internal capability to do multibody dynamics and multidiscipline simulations with MSC Software’s Adams and SimXpert has been key to developing the capacity to design vehicles to its own specifications that can later be configured to meet a range of specific customer requirements.

The company is developing a robotic vehicle platform called the BLR that will handle a wide range of military missions without requiring a human driver. While previous robotic vehicles have taken 6 to 7 years to develop, the company is on track to finish the design of the BLR robotic vehicle in well under 18 months. “Simulation gives us the ability to evaluate the performance of many different design alternatives early in the design process and select the best design in less time at a lower cost than would be required using conventional design methods,” said Ronen Veksler, Analysis Department Manger for BL Advanced Ground Support

Systems. “The BLR’s extraordinary capabilities and compressed development time is a direct result of the huge number of simulations that have driven the design process.”

BL Advanced Ground Support Systems is a privately held company that has 30 years experience in designing, developing and producing ground support equipment for fixed wing and rotary wing aircraft, vehicles for ground forces and mixer feeders. The company was recently selected by Lockheed Martin through its subsidiary PDI/BL International to design and manufacture a weapons loader for the F-35 Joint Strike Fighter. BL customers include the Israeli Air Force, Israeli Ministry of Defense, Israeli Ground Forces, Israel Aerospace Industries, Israel Military Industries, Rafael, Soltam, UK Ministry of Defense, Italian Ministry of Defense, Lachish Industries and Elbit.

Developing Internal Simulation CapabilitySimulation has long played a key role at BL Advanced Ground Support Systems by providing the ability to evaluate the performance of alternate design concepts prior to the physical prototype stage. Up to

two years ago, the company worked with outside consultants that provided simulation services. “The problem with this approach was that a considerable amount of time was required to communicate the design concept to the external consultant and then there was a long wait while they performed the simulation and communicated the results back to us,” Veksler said. “It normally took weeks to find out how well our design performed.”

“The BLR robotic vehicle is a very ambitious project and it was obvious from the beginning that it would require an enormous simulation effort to develop a vehicle that could deliver world-class performance on each of the missions that we envisioned for it,” Veksler said. “If we had continued with our previous practice of using outside consultants, simulation would not only have been a serious bottleneck, it would have been impossible to design this vehicle.”

“To address this challenge we set about developing our own internal analysis capability,” Veksler said. “We had considerable experience in working with simulation results so we had a good idea of what we were looking for. Adams is the de facto standard in vehicle engineering because of its ability

BL Advanced Ground Support Systems | Based on an interview with Ronen Veksler, Analysis Department Manager

Developing New Vehicle Concepts Faster

Page 21: Simulating Reality

Volume II - Summer 2012 | 19

to model every aspect of the design process. MSC Nastran is the leader in structural analysis. MSC Software now provides both of these tools and others within the SimXpert environment behind a single user interface that allows teams to share data, models, results and best practices. With our own internal analysis capability, we have reduced simulation turnaround time from weeks to hours.”

Developing a New Robotic Ground VehicleBL’s new robotic ground vehicle is unusual in that the project was initiated by the company itself rather than by a customer request for proposal. The company defined aggressive specifications for a vehicle that can be produced in a range of configurations with wheels or tracks and can carry different payloads including weapons and surveillance systems. A key requirement is the ability to maintain speeds of 50 kilometers per hour over extremely rough terrain. The vehicle is designed to cross over steps 3 feet high and other challenging obstacles that will allow it to operate in nearly every potential battlefield around the world.

Before beginning simulation, engineers defined the basic configuration of the vehicle. They decided to use a trailing arm suspension

because it allows for large travel by each wheel. The powertrain consists of hydraulic hub motors for each wheel with skid steering. BLR used what it calls a ladder chassis consisting of a space frame made of rectangular tubes.

The simulation process started with examining potential suspension components and geometry. “We defined the basic suspension components as rigid bodies in Adams without going into a lot of detail,” Veksler said. “We used code to define the skid steering. We ran the initial design over test tracks including alternating sine and co-sine roads. We defined a number of mission profiles consisting of turns and steps and hill climbs. We used this very basic model to make basic high-level design decisions such as the type and position of the bearings used to support the trailing arms and the power requirements for the hub motors.”

Moving into Detailed DesignThe detailed design began while BL engineers were at the final stages of determining concept geometry using early analysis stages.. The de-featured CAD model was exported from the computer aided design (CAD) software into the SimXpert multidiscipline simulation environment. SimXpert was then used to automatically mesh the model, create special elements as needed and define loads and boundary conditions. This process is nearly completely automated and can be accomplished so quickly that dozen of design iterations can be simulated in a single day. With the CAD geometry associated to SimXpert, it was easy to utilize the detailed geometry in the analysis to take the flexibility of critical structural components into account in the vehicle simulation. For example, by converting the trailing arms to flexible bodies, BL engineers were able to determine the impact of the trailing arm geometry on the vehicle performance. Likewise, components’ structural strength and dynamic properties can also be easily determined.

“Without a complete understanding of the design’s behavior, critical failures can be overlooked,” Veksler said. “The ability to easily deploy multiple solvers on a single design within the SimXpert environment is crucial to our work. SimXpert gives us the ability to run linear FEA, nonlinear FEA, multi-body dynamics, thermal simulations, crash tests, virtually any type of simulation we need with a minimum of additional work.” For example, one of the vehicle applications involves carrying a firing system. Engineers incorporated the firing event as a load using a force vs. time graph and used transient response analysis to understand how the vehicle would react to firing.

Support is CriticalSupport provided by both MSC Software and its Israeli business partner MSI is crucial to BL’s success. “Having a knowledge-base as vast and as professional as MSC and MSI helps us improve our analysis abilities and is a strong factor that differentiates us from our competitors,” Veksler said. “Knowing that we have strong professional backing allows us to rely more on our analysis results and produce a better first article every time.”

“Simulation helps encourage innovative design methods because engineers can easily explore alternative design concepts in very little time or expense,” Veksler concluded.

“Simulation also provides detailed diagnostic information that helps us understand why a design is performing as it is. With SimXpert we spend less time translating and fixing CAD data, meshing, reworking models, and creating the same plots and charts over and over. This means that our engineers can devote more time to developing new vehicle concepts and bringing them to market faster than our competitors.” BL is nearing the completion of the detailed design of the BLR robotic vehicle and will soon begin production of the prototype. u

BLR’s extraordinary capabilities & compressed

development time is a direct result of the huge number of simulations that have driven the

design process.

BLR Concept Drawing (featuring tracked wheels)

Adams Analysis using a flexible trailing arm on a Sin&Cos ground

FEA Analysis carried out on the trailing arms and chassis following an Adams simulation

Motor selection simulation - inclined slope with “gravel like” traction

Adams Analysis using a flexible trailing arm on a Sin&Cos ground

Step climbing simulation (incorporated skid steer controller - motor speed

state variable control)

Page 22: Simulating Reality

20 | MSC Software

Predicting Fatigue Failures with Confidence Virtual Load/Fatigue ProcessLeading Edge Engineering, an MSC Software Business Partner | Wayne Tanner

BUSINESS PARTNER SPOTLIGHT

Many companies already use FEA to analyze their structures, but

typically using a set of static or inertial limit loads, and modal analysis to validate their products. While these loads usually work well as screening loads and have proven successful at eliminating catastrophic failures, they typically don’t do an adequate job of simulating and predicting fatigue failures. Increasingly, there is a need to do a better job of predicting fatigue failures early in the development process, and developing solutions without the expense (in terms of time and dollars) of multiple physical prototype iterations. Predicting fatigue failures can significantly reduce long term warranty costs, and be used to optimize the structure.

This article is intended to educate readers on the basic components of a virtual load/fatigue analysis process, discussing some of the assumptions behind them.

Loads/Fatigue Process OverviewA virtual load/fatigue analysis process uses an Adams Multi-Body Dynamics (MBD) system model to calculate the response and loads of the system as it operates through a virtual set of events or duty cycle. These loads are then exported to a MSC Nastran FE model to recover the stresses in the component of interest. These stress histories are then used to calculate the fatigue life of the components with the appropriate fatigue algorithms for welds, parent material, etc. This process is shown in Figure 1.

While a virtual load/fatigue process is intended to test a system prior to building a prototype and performing physical testing, it needs to be correlated with test data. The ultimate goal is for the load/fatigue process to augment (not eliminate) the physical testing process. There are some parameters such as damping and tire interactions

which need to be determined from physical testing. For this reason, the virtual process needs to be developed in coordination with testing, and correlated back to the physical test. During each successive development cycle, more confidence is developed in the MBD system model, and improved model parameters are developed. Not only is confidence developed in the virtual process, but it can be used to simulate events that are difficult to setup and test, information (loads, stresses, strain, etc.) can be gathered from any point on the system, and complete built/test cycles can be eliminated.

Duty CycleDeveloping an accurate duty cycle is a key component to accurately predicting loads and fatigue. A well-defined duty cycle is important for physical testing as well as assisting in developing a good understanding of how your customer uses your products. The duty cycle is a series of events (roading, loading, digging, hauling, etc) that comprises the typical loading that a system will see throughout its lifecycle. It is important to include not only severe events, but also events that may seem non-damaging. Each event is then assigned a percentage of time experienced during a typical fatigue unit (hour, day, mile, etc).

Over time, the make-up of the duty-cycle will change. It is important to document these changes, and understand how they impact the design of your product.

MBD Model SetupTo accurately predict loads from a MBD system model, the model must consist of flexible components so that it can respond dynamically to input loads. Using Flex Body components allows the structure to respond both statically and dynamically, providing accurate load prediction.

A Flex Body component replaces the rigid component in the Adams MBD model with a Finite Element model. The flex body uses the theory of modal superposition to represent the deformation of the structure by a series of mode shapes. The common question that is asked is “How many modes do I need to accurately represent my structure?” The real answer to this question is “it depends”, but the general rule is you need have modes that are 2X times as high as the highest frequency in which you are concerned about. For example, if you have frequency content of up to 200Hz in your loading input, you generally need to include up to 400Hz of modal data. Using Residual Vectors and other techniques we can reduce the number of modes necessary to represent the motion of your structure.

Once the MBD model is complete, it can compute loads at all attachment points for the component of interest, as well as stresses in a flex component. Adams can calculate nodal stresses on flex body structure, but these stresses are not preferred to use for fatigue calculations. These stresses are generally used to identify hotspots and high stress/loading time points.

While a virtual load/fatigue process is intended to test a system prior to

building a prototype and performing

physical testing, it needs to be correlated

with test data.

Page 23: Simulating Reality

Volume II - Summer 2012 | 21

There are two basic methods for stresses from the MBD model to calculate fatigue lives. The first is using a Quasi-Static method; the 2nd is using a Modal method.

Quasi-Static Stress Recovery MethodThe quasi-static method (also known as linear superposition) makes the assumption that the structure is behaving statically. This method is conceptually easier to understand, but caution needs to be taken to verify this assumption. This assumption is usually valid if the frequency content of the loading is less than 1/3 of the first mode of the structure.

Using this method, the loads at each attachment point (and the acceleration loads) are exported for each time point of the simulation. For each load attachment point on the structure, a corresponding “unit loadcase” is solved in the FE model. The unit loadcase analysis will typically use inertial relief to restrain the model for each subcase which has a single unit load.

The fatigue analysis then scales each unit load by the load value for that load attachment point for each time step. Then for each time step in the analysis, all of the stress states for each scaled unit loadcase are summed to get a complete stress state for that time point. Once the complete stress state for each element at each time point is calculated, the fatigue analysis then calculates the fatigue life of each element over the entire load cycle. The advantages of this method are that it is computationally inexpensive, minimizes file sizes, and is conceptually simpler to understand. The disadvantages are that it may not be accurate

when the natural frequencies of the system are close to the frequency content of the loading.

Modal Stress Recovery MethodThe modal stress recovery method (commonly referred to modal superposition) is similar to the quasi-static method, except the unit loadcases are replaced with modes, and the load time histories are replaced with the modal responses for each mode shape.

To perform modal stress recovery, Adams will write out a .mdf file which can be used in a modal transient (SOL 112) restart of the original normal modes (SOL 103) database that was used to create the flex body (.mnf file). From the modal transient analysis, MSC Nastran will calculate the stress at each element through the duration of the analysis. However, this results database can be extremely large. To minimize file size, typically the SDISP card is used in MSC Nastran to output the modal participation data to a punch (.pch) file. The fatigue analysis then multiplies each modal response with the appropriate mode and sums this product across all modes for each point in time. This then produces the stress history for each element of the structure for the entire time history, which is used to calculate the fatigue life of each element.

The advantages of this method are that it will account for resonant (dynamic) effects of the structure, and without storing the entire time history of stresses for the entire model.

The disadvantages are that is conceptually more complex, and requires MSC Nastran to compute the modal participation file.

Fatigue AnalysisOnce the stress histories are calculated for the entire duty cycle, there are several different algorithms that may be used to calculate the fatigue life. It is generally recommended to use strain life algorithms to calculate fatigue life for parent material, as this method is capable of accounting for overloads and load sequence effects.

To calculate fatigue life of welds, there are several methods available, but all are based on empirical stress life calculations which typically classify the welds to identify the appropriate stress-life curve to use. Other advanced methods have also been developed which eliminate this classification.

SummaryImplementing a virtual loads/fatigue process requires expertise in several advanced disciplines (Multi-Body Dynamics, FEA Dynamics, Fatigue Analysis, and Test Correlation). This is a process which may take several simulation/test/correlation cycles to build confidence and develop models with the appropriate level of sophistication. Successful companies usually have one or more engineers who are tasked with championing this process and build confidence and experience with implementing this system for their products, working through the challenges of correlating models and possibly changing the culture of their product development process to focus on fatigue and loads vs. single loadcases and hotspot stresses. uReferences:

Adams Training – ADM710 Flex Body Dynamics and Modal Stress Recovery using Adams

Patran Training – PAT318 Durability and Fatigue Life Estimation Using Patran

Virtual Loads/Fatigue Process Overview

Quasi-Static Stress Recovery Method

Frequency Response Function

Modal Stress Recover Method

Page 24: Simulating Reality

22 | MSC Software

Unveiling a Monumental Gold CoinSimulation Plays Vital Role in Creation of Record Breaking 1 Tonne Gold CoinCompumod, an MSC Software Business Partner | By Peter Brand

Compumod is pleased to announce that its Finite Element computer

simulation has been used to assist in the design and planning of the recently unveiled record breaking one tonne gold coin cast by The Perth Mint.

Weighing a massive one tonne of 99.99% pure gold, the monumental coin measures nearly 80cms wide and more than 12cms deep.

Prior to the casting, Compumod was engaged by The Perth Mint to create a computer simulation of the planned pour in order to assess the mould and mould fixture’s integrity to ensure they did not deflect to a point whereby the critical dimensions of the coin were affected.

Compumod undertook this work using MSC Software’s Marc nonlinear software. The software enabled Compumod to undertake a Nonlinear Transient Finite Element Analysis (FEA) of the pour. Once completed, Compumod was then able to accurately predict temperature distributions and deflections of the mould and mould insert during the pouring of the 1,000 kg of 1,300 deg C molten gold into the mould.

Peter Brand, Technical Director of Compumod said, “At Compumod we have undertaken many interesting types of projects but this was definitely a one off! Due to the amount of gold being poured and its temperature, we had issues not just with the heat transfer and the variation in material properties of the mould and insert, but also the hydrostatic pressure of the gold itself in the mould. I am pleased to say that our analysis confirmed that the as designed mould and inserts were up to the task and we are proud to have been in a small way associated with this record breaking casting.”

Phillip Kruger, Services Manager for The Perth Mint said, “I engaged Compumod for this task as we had never before cast such a large coin and were concerned about the adverse effects of the forces and temperatures involved in 1 Tonne of molten gold. Compumod’s results helped to reassure us that the design we had commissioned for the mould and inserts was up to the task and would produce the high quality gold coin we have today.”

Exclusive behind-the-scenes footage of the creation of the coin can be seen at 1tonnegoldcoin.com or youtube.com/user/perthmintbullion.

For more information on any Compumod products or services please email [email protected] or visit www.compumod.com.au. u

Compumod’s results helped to reassure

us that the design we had commissioned for the mould and inserts was up to

the task and would produce the high quality gold coin we have today.

BUSINESS PARTNER SPOTLIGHT

Page 25: Simulating Reality

PRINCIPLES OF FATIGUE ANALYSISCourses on Fatigue and FEA Available!

By Dr. Neil Bishop(Author of the NAFEMS Book, Finite Element Based Fatigue Calculations)

LOCATIONS AND SCHEDULE - 2012

September 3-5 NAS319A, B & C Birmingham, UK

September 18-20 NAS319A, B & C Gothenburg, Sweden

October 1-2 NAS319A & B McLean, VA, US

October 3-5 NAS319A, B & C Wichita, KS, US

October 8-10 NAS319A, B & C Ann Arbor, MI, US

October 11-12 NAS319A & B Toronto, Canada

Oct 31 – Nov 2 NAS319A, B & C Rio de Janeiro, Brazil

November 13-15 NAS319A, B & C Paris, France

November 27-29 NAS319A, B & C Munich, Germany

February 5-7 2013 NAS319A, B & C Torino, Italy

- Mechanical Engineer, ITT Geospatial Systems

The instructor is an industry renowned expert. Terrific seminar.

Instructor Background: Dr. Neil Bishop has been teaching fatigue life estimation techniques and dynamics to undergraduates, postgraduates and design engineers for the last 20 years. His teaching style reflects the

philosophy that fatigue life estimation techniques (and FEA) should be accessible to all design engineers, not just the specialists. He recently co-authored the NAFEMS publication “Finite Element Based Fatigue Calculations”. He recently directed an R&D project for the US Air Force dealing with acoustic fatigue.

REGISTER TODAY: www.mscsoftware.com/fatigueworkshop

2012_Fatigue_Workshops-FP4.indd 1 8/22/12 11:45 AM

Page 26: Simulating Reality

24 | MSC Software

Adams/ViewFlex has MSC Nastran under the Hood

Adams 2012 introduced a new capability called ViewFlex for creating flexible parts directly inside of Adams/View. This module allows creation of Modal Neutral Files with either solid or shell elements. There are meshing options and a basic set of element types. In general the part is restricted to one isotropic material such as aluminum or steel. ViewFlex is perfect for those situations when you need to make a bar or frame flexible in an Adams model but don’t have access to a full-fledged FEA solution.

There is no mystery about the underlying FEA tool that creates the MNF--it’s MSC Nastran! ViewFlex includes a subset of MSC Nastran adapted for creating Adams/Flex parts. You do not need an MSC Nastran installation or license, just Adams/Flex and the ViewFlex license feature. You can even examine the MSC Nastran files after the MNF has been generated.

There is a detailed example in SimCompanion KB8020371, “Adams/ViewFlex Flexible Body Generator.” Follow the instructions to create the flexible part. There are two major steps for preparation: 1) creating the mesh and 2) selecting the attachments (to define interface nodes).

When preparation is complete ViewFlex creates the MSC Nastran input file, runs the job, and then swaps the flex body into the Adams model.

Most users will mesh existing geometry to replace rigid parts with flexible ones. There are also options to create and mesh extrusions as well as to import an existing mesh. There are settings for element type and order. Element size can be specified or set to be automatically determined.

The attachment options include MSC Nastran RBE2 (rigid) and RBE3 (averaging) elements. There are tools for selecting nodes in the vicinity of the attachment point by distance or manual pick. When preparation is complete, ViewFlex creates the MSC Nastran input file, runs the job, and then swaps the flex body into the Adams model.

After creating a flex part with ViewFlex, look in the Adams working directory for the MSC Nastran files. If you don’t recall the location, you can find your current working directly for Adams by going to File->Select Directory. You will see a file such as PART_2.mnf, the Modal Neutral File. You’ll also see files like PART_2.bdf (MSC Nastran bulk data file, the input deck) and PART_2.f06 (MSC Nastran output file). You can load the input and output files into a text editor and check the model definition details as well as output messages.

The first model image is from an Adams/ViewFlex part that was created from an existing rigid rectangular block using solid tetrahedral elements. In the input file you’ll see MAT1 and PSOLID cards along with a large number of GRID and CTETRA cards. In the Adams/View window you can see an interface node marker along with a spherical joint icon.

The second model image is from a complex part that was imported from a Parasolid file. For a first pass at creating an MNF, the element sizes were set to the part thickness. The attachment point was defined as an RBE3 that averages the displacement of the 20 nodes closest to the center. The ViewFlex (i.e., MSC Nastran) solution time to do the normal modes solution, constraint modes solution, and MNF creation was just a few seconds on a laptop computer. A more detailed mesh for this part along with a second attachment point and more modes took just seven seconds!

By Walt Daniel Sr. Technical Representative, MSC Software

MSC Software TECH TIPS

Creating Flexible Parts inside AdamsAdamsTM

Page 27: Simulating Reality

Volume II - Summer 2012 | 25

Transparent Display

Have you ever wanted to set up the display in Patran so that part of your model is transparent? You can, with the appropriate commands. First, set up groups for the parts of the model you want to be transparent, and set group display mode (under Display / Entity/Color/Label/Render).

In Display/Shading, we can set transparency for the whole model. In order to apply it only to certain parts, we first need to use another option: Display/Named Attributes. Create a new Named Attribute called Transparent:

Now set the transparency in Display/Shading. This is stored in the current Named Attribute. To set it for individual parts of the model, go to Utilities/Group/Group/Named Attributes. Here we can assign the “Transparent” attribute to the required groups - and that’s it!

Quick RBE Creation

Patran 2011 introduced a new utility for quickly creating RBE2s and RBE3s. Starting from the selection of geometry faces/edges, or element free faces/edges, it will quickly create an RBE “spider”, automatically generating the reference node (or optionally using an existing node).

Access it under Utilities / FEM-General / RBE2/RBE3 Spider.

For example, using the “Element Free Edge” option: simply put a selection box around the required free edges, and there’s your RBE!

If you’re using an older version of Patran then 2011, this tool can easily be retrofitted to that version. Contact your local MSC technical representative for details.

By Darrel Sinclair Senior Technical Consultant, MSC Software

Getting Efficient with PatranPatran MSC Fatigue

Page 28: Simulating Reality

26 | MSC Software

Glued contact in MSC Nastran is a powerful tool to join dissimilar meshes. It saves a lot of meshing effort: the meshes for different parts need not be congruent. You only need to define a contact body for each part and set the contact to “glued” in the contact table. This process can be accelerated by the automatic contact table, which is available in SimXpert. So you save time on the modeling side, but on the other hand you need more time for post processing. How can you verify that all parts are correctly connected?

MSC Nastran outputs the contact status at the end of the job. Contact status = 1 indicates a tied node (slave), contact status = 0 indicates a retained node (master). No data means that the node is not in contact. But the contact status is normally available only at the end of the analysis. For a large job, you might wait a few hours to find that the contact status is not as expected, and you need to improve the contact definition. How can you see this already before the analysis, and get more insight? There are different variants.

Print Internal MPC equations

Glued contact is internally represented by MPC equations. The case control command NLOPRM MPCPCH=BEGN will output these MPC equations to the punch file. You can examine them in a text editor, or import them into your preprocessor to view graphically which nodes are connected. This requires a Nastran job to be run, but it’s possible to use a dummy job which contains only the glued contact.

Contact Status Plot Before the Analysis

If you use SimXpert as pre-/postprocessor, you can find a new tool in version 2012: Advanced Tools – Contact Status. It creates a dummy Nastran job which only calculates glued contact, attaches the result file and shows a fringe and scalar plot of the contact status. You can modify the settings of this plot in the results menu. The scalar plot, which uses only markers, gives the clearest picture. Note that with the default contact settings in MSC Nastran, the slave node connects to a master surface (element), and all of the master surface’s nodes are shown to be in contact. So you will often see markers on two master nodes per one slave node. The contact status tool makes use of the NLOPRM MPCPCH=BEGN method described above, and also the MPC equations are automatically imported into the database.

Intermediate Output for Nonlinear Job

The contact status tool is intended for glued contact, which is calculated by MSC Nastran before the actual analysis begins. If you work not with glued but with touching contact in a nonlinear SOL400 analysis, there is also a method to see contact results before the job has finished. The case control command NLOPRM OUTCTRL = INTERM gives you a separate op2 file for each increment of the analysis.

By Cornelia Thieme Technical Support Coordinator, MSC Software

By Dominick Lauzon Lead Application Engineer, MSC Software

Visualize Glued Contact

Example model with 4 contact bodies

PCH file with MPC equations

MPCs displayed on model

Contact status plot in SimXpert – 2 display variants

Page 29: Simulating Reality

HELPING ASPIRINGENGINEERSBECOME TOMORROW’S WORKFORCE

With 2,000 in attendance and 150 teams in the competition, Formula SAE is one of the biggest university events of the year, bringing future engineers together with the local engineering community. As a sponsor of the event, MSC Software was able to be a part of the magic that took place in Brooklyn, Michigan.

Beginning in 1978, the Formula SAE competition challenges student design teams to develop a small Formula-style race car. Each student team designs, builds, and tests a prototype

based on a series of rules to ensure safety of the vehicle. The FSAE competition is a worldwide event, the biggest of them taking place in Michigan. MSC Software was a proud sponsor of the event in a number of ways, the biggest being through the student teams themselves. A number of the teams actually used Adams/Car to rapidly create a full dynamic model of their student competition vehicle and simulate complex dynamic events. As Karthik Krishnan (Graduate student at

Colorado State University) told us, “Knowing Adams/Car enabled me to beat out my competition and also allowed me to get the prestigious job I wanted with Pratt & Miller Motorsports.”

To wind down from the competition, MSC Software also hosted a networking reception with nearly 100 individuals from industry. “The reception allowed attendees to enjoy the VI Grade car simulator that worked with Adams. The food was complimented by all and the evening rain presented us with

a rainbow over the Michigan International Speedway event site. It took 30 additional minutes to encourage everyone to leave – difficult task as everyone was very much enjoying the evening.”

Already anxious for next year’s FSAE event, please visit www.mscsoftware.com/fsae or contact [email protected] if you would like to learn how to gain a competitive advantage using Adams/Car. u

By Caressa Matsuoka, MSC Software

MSC WORKS WITH STUDENTS TO SOLVE REAL WORLD PROBLEMS

SPECIAL SPOTLIGHT

Page 30: Simulating Reality

TECHNOLOGY MATTERS

InterviewAcoustics ExpertsJean-Louis Migeot & Jean-Pierre CoyetteJean-Louis Migeot and Jean-Pierre Coyette are the co-founders of Free Field Technologies (FFT), the Belgian company that develops the Actran acoustic CAE software product and that has joined MSC Software last Fall.

Simulating Reality met them in their Mont-Saint-Guibert office to learn more about noise, acoustics and the use of Actran in industry.

SR: Why is acoustics so important to engineers today?Jean-Pierre: We are all living in a noisy environment: in our office, at home, while travelling. We are constantly exposed to pleasant and less pleasant sounds generated by every single object that surrounds us. At home our dishwasher, vacuum cleaner, hair dryer and washing machine combines with traffic noise flowing through the window to create a constant noise background. Our car is a real

“noise factory” and we hear contributions from subsystems as diverse as the powertrain, the air conditioning unit, the tire and of course the high pitch noise of the air flowing around the car. Our office is buzzing with noise as well: from computers, printers, telephones, copiers and colleagues.

Jean-Louis: And yet …did you notice that much progress has been made in quietening all these devices? Sure, the overall noise level is still high but mainly because we have more cars, more fans, more machines of all kinds, but each individual sound source has been dramatically reduced over the last ten to twenty years. Try driving a 1970’s vintage car, running a vacuum cleaner from the 1960’s or flying aboard a Caravelle, and you’ll measure the progress made by acoustical engineers!

SR: True, we can all think of the many noise sources around us we’d like to quiet! But is this public annoyance sufficient to explain the major investments made by many corporations in acoustical engineering?

Jean-Louis: This fight for lower noise levels has in fact been driven by three complementary forces: noise regulations which have become more and more stringent in all industries and all countries, market analysis which has shown that the acoustic quality of a product is an important sales argument and, in more specific cases, the identification of noise as a source of vibrations that can lead to failure. Acoustical engineering was thus progressing in order to meet customer requirements, respect norms and standards and just meet resistance constraints.

SR: Noise has thus become a design attribute on par with stress, fatigue, crashworthiness, fluid dynamics or thermal efficiency? Jean-Pierre: Exactly, noise has evolved, over the last two decades, from something that was “fixed at the end of the design process” to just another design attribute integrated in the concurrent engineering process: noise had to be designed, controlled, understood

…and it had to follow the “right first time” philosophy. Engineers in search for innovative noise reduction solutions needed to perform simulations, and acoustic CAE tools started to appear in the early 1990’s.

Jean-Louis: Acoustic CAE tools like Actran are now mainstream and are part of the daily work of engineers across all industries and regions: automotive, aerospace, defense, railway, home appliances, energy, audio and communications. Acoustic simulation tools are fully integrated in the overall CAE process, connected upstream with CAD and mesh generation tools, and in CFD tools for aero-acoustic predictions and structural vibration analysis tools for vibro-acoustics prediction.

SR: Has Actran permeated in all industries? Jean-Pierre: The transportation industry (automotive, aerospace) is a primary user of acoustic simulation technology and they had a pioneering role in acoustic CAE in the late 1980’s. Actran is for instance used in applications as diverse as powertrain noise radiation analysis, optimal design of air intake and exhaust lines components, prediction of airborne and structure-borne noise from tires, absorption and insulation analysis of individual trim components, vibro-acoustic analysis of layered windshields, air-conditioning noise analysis, transmission or aero-dynamic sources through side windows and, in combination with MSC Nastran, prediction of the global vibro-acoustic response of the trimmed vehicle body. The use of Actran is not restricted to the vehicle OEM but is now very common among the many layers of suppliers.

Jean-Louis: Actran and acoustic CAE also contributes to reducing noise in many other products. Recent projects performed by Free Field Technologies include refrigerator pump noise reduction, acoustic resonance in burners, prediction of the noise generated by an industrial centrifugal fan or predicting the impact noise from the rain on a steel tile. And let’s not forget audio equipment from loudspeakers, telephones, hearing-aid devices or microphones.

Jean-Pierre: Also, now that much progress has been made in quietening many products, sound quality issues start to appear: the focus is no longer on simply reducing the noise level but in improving its quality, in making a product “sound right”. The sound must be

withAn

28 | MSC Software

Page 31: Simulating Reality

Volume II - Summer 2012 | 29

pleasing, discrete yet audible, in line with the image that the product carries. Actran is for instance used to understand a camera shutter noise and in shaping the associated spectrum to meet given specifications.

SR: To complete our interview, could you describe a recent, real-life, user case? Jean-Louis: Well, a very loyal customer of ours, Alenia Aermacchi, recently published an interesting case study. Alenia Aermacchi has been a research partner and an Actran user since 2003 when they collaborated with Free Field Technologies in the MESSIAEN project, a European-funded research programme on “Methods for the Efficient Simulation of Aircraft Engine Noise” also involving Rolls-Royce, Airbus, Liebherr Aerospace and Turbomeca.

Jean-Pierre: The company from Venegono Superiore in northern Italy is engineering and producing aircraft engine nacelles which are lined with advanced materials aimed at absorbing part of the noise generated by the engine fan (see figure 1). The material is made of a thin resistive cover layer backed by resonators. One of the Actran modules, Actran TM, is ideally suited to predict the efficiency

of such liners and is the reference acoustic simulation tool of the aircraft engine industry. TM indeed features a special source model well suited to represent the engine fan (duct modal basis), includes a convected wave propagation operator to take into account the effect of the background flow on the propagation of the sound waves and has a unique capability for handling the liner and its interaction with the sound waves and the grazing flow.

Jean-Louis: The modelling process used by Alenia Aermacchi involves three steps: (1) creation of an axisymmetric acoustic mesh of the air inside and outside the nacelle (infinite elements are added on the edge of the finite element mesh to represent far field radiation), (2) the background flow is calculated and (3) the propagation of selected components of the fan source is computed (see figure 2). Simulation results have been successfully compared with measurements taken in the AneCom test chamber (figure 3).

SR: Thank you Jean-Pierre and Jean-Louis for this quick introduction to acoustic CAE and an interesting use case! u

Figure 1: Acoustic panel installed inside the nacelle

Figure 2c: Example of propagation of an acoustic mode

Figure 4b: Acoustic measurements in the CMDIa section

Figure 2a: Acoustic mesh

Figure 3: Assembly of the physical prototype of the fan in the anechoic chamber

Figure 4c: Acoustic measurements in the far field

Figure 2b: CFD field imported in the acoustic mesh

Figure 4a: Acoustic measurements in the CMDIb section

Page 32: Simulating Reality

30 | MSC Software

TECHNOLOGY MATTERS

Understanding Eye Injuries Numerical Analysis and Modeling of the Mechanism of Retinal Detachment as a result of a blunt impactUniversity of Cassino, Italy | By Prof. Nicola Bonora, Professor of Mechanical Design and Construction Machinery

The detachment or tearing of the retina in the human eye as a result of a

collision is a phenomenon that occurs very often. The activity in question is aimed at understanding the actual processes of dynamic deformation taking place in the human eye when subjected to blunt impact. For this project a FE model was developed in MSC Software’s Dytran solution starting from 3D measurements of real human eyes. The results of the model were then compared to measurements with respect to the deformation at different times and to the residual velocity of the projectile during the rebound phase.

IntroductionOne of the most recent activities of the Construction Machinery Research Group of the University of Cassino is to understand in detail the phenomena that lead to the occurrence of ocular lesions (such as during combat operations) following an approach based on experimental physics and constitutive modeling based on MSC Software simulation products, with particular reference to MSC’s Dytran explicit code. This activity has been carried out in collaboration with the Ophthalmic Hospital in Rome, Italy.

Over 60% of all eye injuries are caused by blunt impact, i.e. impacts with objects of various kinds that do not cause a perforation

of the globe. The clinical manifestations of such injuries is quite different and includes retinal rupture, choroid rupture (the tissue between the retina and the sclera), retinal tear and retinal detachment, macular holes and dialysis (see Figure 1 for the primary structures constituting the human eye).

Although the clinical phenomenology of such injuries has been accurately described already, the mechanisms responsible for the lesions of the retina and choroid associated with blunt impacts have not yet been fully understood. Among the theories proposed to explain the mechanism of damaging internal structures of the eye during a blunt impact, the most widely accepted one is called “vitreous chord pulling-traction.” According to this theory, during the compression of the eyeball in the direction of the impact, the expansion of the sclera in the orthogonal direction generates a critical stress in the internal structures favored by the viscous action of the vitreous, the jelly-like fluid contained in the main eye socket. This mechanism would explain why the retinal tears occur mainly at the vitreous base, i.e. in the vicinity of the circumferential junction between retina and choroid and in the vicinity of the macula.

Based on evidence of a patient who, despite having undergone the removal of the vitreous, had a clear macular hole resulting from a blunt impact, it was decided to investigate this phenomenon in more detail in order to validate the various hypotheses of damaging mechanism with the help of a MSC Software Dytran simulation. In particular, the behavior of a human eye subjected to blunt impact as a result of impact with the ball of steel has been simulated. The model, validated through a comparison of the simulation results against measured data

available in literature, was then used to study the intensity of the dynamic stress waves produced during the impact, in order to assess the primary source of retina failures.

The Simulation ModelThe computational model of the ocular globe was generated starting from an average size human eye represented with the help of MSC Software’s Dytran. The code was selected between the different explicit solvers available on the market mainly because of its advanced fluid-structure interaction capabilities. Assuming as symmetry plane the

The computational model of the ocular

globe was generated with the help of MSC Software’s Dytran.

Figure 1: Primary structures of the human eye: C) cornea, S) sclera, L) crystal, R) retina B) ciliary

body; M) macula, V) glassy, H) vitreous base.

Figure 2: 3D mesh of the eye - each color identifies a different substructure.

Page 33: Simulating Reality

Volume II - Summer 2012 | 31

meridian section of the globe that contains the longitudinal axis, a half-eye model has been created that includes all substructures and tissues that could potentially affect its dynamic behavior: cornea, sclera, aqueous and vitreous humor, crystalline, ciliary body and zonules. The retina was modeled as a thin layer connected to the sclera with constant thickness of 0.2 mm. The related 3D mesh, shown in Figure 2, consisted of 6912 brick elements. The spherical projectile with a diameter of 4.5 mm was modeled as a rigid body. The configuration used for the simulations is that of a normal impact, in which the bullet hits the apex of the cornea in longitudinal direction at a speed of 62.5 m/s (Figure 3).

The Case in ExamGiven the complexity and variability of the physical and mechanical properties of all biological materials, that are strongly dependent by the hydration of the tissues and on how they are stimulated during the characterization tests, the related identification has been performed using the experimental results obtained by Deloria et al. in 1967 through a non-penetrating impact test of the human eye. In the present case, the experimental values of corneal apex displacement caused by the

penetration of the projectile as a function of time have been used as reference.

Additionally, the methodological principle of “Occam’s Razor” has been adopted as constitutive model to describe the behavior of all tissues. This principle enables to minimize the number of parameters necessary to describe the phenomenon with the requested accuracy.

The description of the tissues is based on a linear elastic material model (that neglects all visco-elasto-plastic effects) and on linear state equations. The vitreous has been modeled with viscoelastic fluid with damping. Crystalline and choroid have been described through linear state equations. The model parameters have been then identified through a reverse calibration process that uses as objective function the experimental measurements produced by Deloria et al. Figures 4 and 5 show the experimental values compared with the optimized response of the numerical model.

The ResultsA careful examination of the results obtained through the simulation show that most of the retinal ruptures, which occur as a result of a blunt impact, are located in the macular and in the vitreous area, but very rarely affect the equatorial area. In order to fully understand the simulation results, the pressure values have been extracted in three points of interest. Figure 6 shows that the pressure waves, generated by the impact, propagate in the eye and are reflected as traction waves, thus affecting mainly the ocular fundus and the macula. The speed of propagation of the waves in the eye is much higher than the speed of the bullet, therefore the peak values of tensile pressure (about 0.6 MPa) were observed within 0.05 milliseconds after the impact

when the ocular globe is not yet affected by large deformations (Figure 7). Shortly after the impact (at time = 0.1 milliseconds), the sclera starts to be affected by large deformations due to the penetration of the projectile. This causes the macular area to be subjected to compression, while the vitreous area is mainly subjected to traction. The pressure in the equatorial area is much lower than in other areas, and this confirms that there is a lower risk of rupture of that portion of the retina.

Conclusions & Future DevelopmentsThe preliminary results of the project indicate that the laceration of the retina mainly occurs due to the tension resulting from the reflection of compression waves in the moments immediately following the impact, and not necessarily due to the deformation of the whole eye. The availability of a reliable and validated model for the simulation enabled the research team to understand in detail the pathogenesis of the blunt impact phenomenon, which is particularly difficult to reproduce in a controlled and instrumented manner through physical tests in the laboratory. Practical applications of this study are to be found especially in the military industry, for example in the design of advanced security systems for personnel and for helicopter pilots in the event of a crash landing.

By Prof. Nicola Bonora, University of Cassino, Italy, Professor of Mechanical Design and Construction Machinery; in collaboration with Luca Esposito, University of Cassino, and Tommaso Rossi, Ophthalmic Hospital of Rome-ASL RME u

Figure 4: Comparison between the experimentally measured displacement and the numerical simulation

Figure 5: Evolution of the velocity of the projectile: comparison between arrest time and residual velocity of the projectile

Figure 6: Evolution of pressure in three representative points in the macular area, the vitreous area and in the equatorial zone (negative pressure values

represent traction)

Figure 7: Propagation of waves of pressure in the ocular globe 0.03 milliseconds after the impact

Figure 3: Layout of the simulated impact in longitudinal direction

Page 34: Simulating Reality

32 | MSC Software

Simulating reality has become an essential part of any design and engineering

process with the aim of avoiding expensive and time-consuming physical prototype testing. After identifying variables and objectives, the main difficulty lies in finding the best compromise among optimum designs in order to produce the ideal solution, taking into account different constraints and parameters. Unfortunately, many of such input parameters are completely opposite in their nature, which raises a multi-objective dilemma: how to consider the full range of every single (sometimes conflicting), parameter without prejudicing one or the other? Is it possible to manually identify the best combination of variables and obtain top results without neglecting any detail?

Nowadays these issues are faced by several companies who are forced to invest in engineering consultancy and infrastructure, often with very time consuming efforts. With this in mind ESTECO has developed a software aimed at making product engineers’ job easier. modeFRONTIER is a multidisciplinary and multi-objective design optimization platform, capable of integrating design and simulation tools into a single IT environment. ModeFRONTIER automatically restricts the number of best possible solutions on the basis of the input data and places them on the tradeoff curve known as the Pareto Frontier. The Pareto Frontier focuses on efficiency and offers a spectrum of optimal combinations of the objective function. A great advantage of modeFRONTIER is that it can be coupled with any software (CAD, CAE or general application tools), whether commercial or in-house, and it enables the simultaneous use of a number of such software

packages even on different machines. It also includes a wide range of post-processing features allowing the user to perform very detailed statistical analysis and data visualization, and facilitate his/her decision. In certain cases it is not possible to predict all factors that might influence the performance of a design; however, modeFRONTIER allows the user to carry out a robust design analysis and to identify the most stable solution even when applying minor perturbations of input variables. modeFRONTIER can be applied to basically any industrial sector: from aerospace, automotive and biomedical to the design of general consumer goods.

modeFRONTIER is directly coupled with MSC Software’s Adams Multibody Dynamics Simulation software, largely used in the automotive industry. The combination of MSC’s Adams Multibody Dynamics Simulation and modeFRONTIER allows the design engineer to exploit industry innovation while maintaining top performances of vehicles. Let’s take as an example Ducati Corse, Ferrari, AUDI and FIAT, all of whom reached invaluable results by integrating modeFRONTIER and Adams.

Ducati Corse was faced with the challenge of designing a better performing racing motorcycle. Their starting point was to obtain experimental data by means of an on-board GPS-INS data acquisition system in order to measure parameters such as torque on the handlebar, steering angle and suspensions deflection. Those data were used to subject the vehicle body to a simulation in Adams and the final objectives were to improve steer torque and rear damper speed. By coupling modeFRONTIER with Adams, Ducati managed to reach the best combination of

not less than 19 control parameters (such as bike suspension characteristics, inertial properties, geometry and so forth) and obtain reduced saddle “kicking” and a significant improvement as regards both objectives throughout the entire racetrack. [Ref. MSC EMEA 2004].

Ferrari’s ultimate goal was to create a new, quicker and more stable sports car. They chose to focus on several crucial points, such as the reduction of design development and model testing, gaining knowledge regarding vehicle dynamics before building the first prototype and assessing the robustness of the selected model. Ferrari used the Adams/Car software to develop a full vehicle model and integrated it with modeFRONTIER to automate time-consuming tasks: devising the process workflow, modifying input parameters, defining constraints and producing a synthesis of results. In regard to this particular model, the objectives were to minimize understeer and maximize traction considering certain constraints, such as oversteer, maximum steering wheel angle and maximum distance from road centerline. modeFRONTIER managed to bring forth an optimized neutral solution, to which robust design criteria were applied taking into account small variations of input parameters that might occur “on the road”. Therefore, the new model was not only faster and more stable, but also most importantly – entirely feasible.

Audi focused its main attention on speed, more precisely on the minimization of lap time of their race car. Since aerodynamics plays an important role in this respect, it is safer and more cost-effective to “try-out” different parameter combinations in the virtual environment. The input variables

Gathering all the Puzzle PiecesMulti-objective optimization software works in an open environment and automates the entire designESTECO, MSC Technology Partner | By Maja Engel

PARTNER SHOWCASE

Page 35: Simulating Reality

Volume II - Summer 2012 | 33

were the coordinates of the fundamental points of the diffuser and the inclination and height of the rear wing of the vehicle and each design was evaluated in 12 different positions to measure the actual aerodynamic forces. modeFRONTIER tackled a real-life multidisciplinary optimization problem by successfully integrating third party and Adams software packages. It managed to achieve an improvement of the lap time of 2,36 seconds, increase vehicle speed in different parts of the track and reduce fuel consumption, while reliably solving each process integration issue and design optimization problem.

FIAT used the customized version of Adams/Car, called MB-SHARC, to run simulations and monitor key parameters representing handling and ride-comfort performances of a small commercial passenger car (FIAT 500),

taking into account also the robustness of the solution. The optimization, performed with modeFRONTIER, concerned the suspension mount characteristics and involved an initial DOE allowing users to select important input variables and representative objectives and constraints. The use of modeFRONTIER saved time and reduced the efforts spent normally on daily continuous modifications of the models and multiple analyses. It also helped acquire a complete understanding between all inputs variables and vehicle performances and to obtain the Pareto frontier containing an optimum set of solutions from different conflicting aspects [Ref. NAFEMS Benchmark Magazine, April 2009 edition].

modeFRONTIER is and remains one of the best-performing existing multi-objective optimization software packages. It allows the

user to work in an open environment and automates the entire design and simulation process (even if a number of other software packages are involved) in a single workflow, generating a number of solutions that the user must choose from on the basis of subjective criteria. Coupled with MSC’s Adams software, modeFRONTIER opens a series of new possibilities: engineers don’t have to hassle with combining input variables one by one trying to find what is best, performing real and costly simulations and working separately on each software package to account for all the different aspects in designing a vehicle. modeFRONTIER can gather all the pieces of the puzzle, interlock them and fit them together leaving the decision maker facing only the final design choice. u

Coupled with MSC’s Adams software, modeFRONTIER opens

a series of new possibilities.

Page 36: Simulating Reality

34 | MSC Software

Limitations in Implant DesignJoint replacements are becoming a more common procedure as our ageing population’s expectations of being active into later life exceed the life expectancy of natural joints. These procedures are most frequently performed on the hip and knee, although the number of other joints now being treated, such as the shoulder, elbow and ankle, is increasing. No matter the location, the process typically involves the removal of the patient’s diseased or damaged joint and replacement with metal, plastic or ceramic bearing surfaces attached to a metal support structure which must be fixed into the surrounding bone.

Bone has evolved to be a complex composite material, a stiff dense shell encapsulating a honeycomb structure with anisotropic behaviour thanks to an intricate substructure. The introduction of a ‘lump’ of metal can have a dramatic impact on the load path from the joint, as the relatively stiff metal transfers the load and shields the surrounding bone from the stress it would otherwise see. The biological response of this unloaded bone is for it to be resorbed, so this area reduces in density and weakens. This is linked to the most commonly observed failure mechanism, with the potential of reduced implant fixation, pain and even bone fracture.

A SolutionA suggested solution to this problem has been to reduce the stiffness of the implant, thus lowering the differential between the metal and bone. This could be done by changing the component material; however there are extremely stringent rules and limitations on the materials which can be implanted into the human body. A lot of work has been done

to optimise the external profile of implants; therefore an approach to reducing the stiffness is to make the component semi-solid or hollow. Traditional manufacturing techniques have not been appropriate for creating such a structure, either hollow or with an internal micro-architecture, but recent developments in additive layer manufacturing or 3D printing using metals have opened up many new possibilities.

Using Simpleware and Marc to Generate and Analyze Novel Implant ConstructionsProfessor Mark Taylor, from the University of Southampton in the UK, and Simpleware have been collaborating to investigate how lowering femoral implant stiffness through design can influence load transfer through the femur following hip replacement surgery.

Simpleware provides a world leading solution for the conversion of 3D image data into high quality surface and volume meshes for 3D printing, CAD, FEA and CFD. The software was used to generate several implanted models of a femur, with externally identical implant designs but internally either being solid, hollow or containing a microstructure.

The ScanIPTM module of Simpleware was used to generate the femur model from clinical CT scans of a healthy male. Once a 3D representation of the femur was produced it was modified using 3D editing tools to simulate surgical procedure, i.e. removal of the femoral head. Simpleware’s +CAD module was then used to import and position a basic CAD implant into the image (Figure 1).

Once positioned, the implanted CAD model was used to create three instances for this

initial investigation; model A with a solid stem, model B with a hollow stem and model C with an internal structure. The internal details of models B and C were made using Simpleware’s Internal Structures Wizard. This tool allows the user to interactively select a unit cell shape from a library and use this to fill any volume, defining the cell unit size, its volume fraction and an encasing shell thickness.

Simpleware’s robust and flexible meshing algorithms were used to mesh the model geometries. Simpleware has been developed to segment, reconstruct and mesh the complex and arbitrarily complex structures that can be captured from volumetric imaging modalities such as CT, µCT and MRI. Therefore meshing the intricate structure of the internal micro-architecture was possible (Figure 2). Each mesh produced had matched nodes and elements at the interface between bone and implant, with high mesh quality suitable for direct use in MSC Software’s Marc nonlinear simulation software.

Simpleware exports an input file, not just a mesh; this allowed the material properties of the femur to be directly assigned from the image. There is a linear relationship between the grey level in the image data and apparent bone density, and researched empirical relationships between density and modulus. These factors were simply typed into Simpleware allowing the effect of the inhomogeneous material properties of the femur to be incorporated into the later analysis through the assignment of element-wise modulus values (Figure 3).

The Findings...The models were all imported directly into Marc for analysis, where identical boundary

Image Based Modeling for Biomedical Implant DesignApplying the innovations offered by Additive Layer Manufacturing to solve traditional limitations Simpleware, MSC Technology Partner | By Rebecca Bryan

PARTNER SHOWCASE

Page 37: Simulating Reality

Volume II - Summer 2012 | 35

conditions and loads were applied; simulating the peak forces acting through the femur during normal walking. The results metric chosen was equivalent von Mises stress and plots of this are shown through the implant and bone for the three configurations.

The results show that the conventional, solid stem represented by Model A transfers the majority of the load through the stem and to the bone surrounding its lower end. The stress in the bone above decreases away to very low levels, particularly on the medial side (left side in the image). It is possible that this shielding could result in bone loss around this area, potentially destabilising the implant.

The hollow stem, Model B, and structured stem, Model C, show an improved stress distribution in that there is load transfer more evenly down the length of the implant. This

would hopefully make bone resorbtion less likely and reduce the related complications.

However, the hollow stem is an extreme example and may not be practical in service due to the possibility of buckling. The introduction of an internal microstructure was simple to design within Simpleware and the model used in the analysis can be sent directly to a 3D printer to be made. The structure reduces the weight of the component as well as its stiffness compared to the solid version. Traditional methods would not be able manufacture this design, however direct metal laser sintering machines can build a design like this layer by layer.

As a preliminary study this shows the potential for new manufacturing techniques to address long standing problems. The combination of Simpleware and Marc has been able to test

the theoretical ideas suggested for improving implant design. The workflow established will allow future studies to interrogate this concept much further, as the flow from image to CAD model integration to internal structure addition to solution was straightforward and robust. Further work will be able to extend the scope of the study, for example comparing different internal structure designs and densities and pushing the analysis to include further measures which can predict the ‘health’ of an implant such as micro-motion between it and the surrounding bone. u

For more information please visit www.simpleware.com

Dr Rebecca Bryan, Simpleware Ltd.

Professor Mark Taylor, University of Southampton, UK

Figure 1: Simpleware +CAD module, showing positioning of the femoral implant into the segmented and edited femur model.

Figure 2: Example image of meshed implanted femur with introduced internal structure. Highlighted zone shows matched nodes and elements across part interfaces.

Figure 4: Plots of equivalent von Mises stress for the three models analysed; Model A – solid

stem, Model B – hollow stem and Model C – stem with internal structure.

Figure 3: Plot of the modulus distribution through the femur, automatically assigned from the

image grey level.

The combination of Simpleware and Marc

has been able to test the potential for new manufacturing

techniques to address long standing

problems to improve implant design.

Page 38: Simulating Reality

36 | MSC Software

AVEVA is a world leader in engineering and design software for the marine

industry and has many customers in common with MSC Software, so the release of a new interface between our two solutions is good news, both for AVEVA and MSC, and great for many shipbuilders and their design agents. We spoke with Stéphane Neuvéglise, Head of Business Management – Marine Systems at AVEVA, to learn more about the new product and its significance.

Finite Element Analysis (FEA) is an essential tool in ship design, not only to verify the structural strength of a hull, but also to meet increasingly stringent vibration criteria for qualities such as passenger comfort, environmental noise or the stability of weapons or radar platforms. But the man-hours required for FEA are costly. Worse, tight delivery schedules frequently make it a critical path activity and can require that construction starts before final approval of the analyses by the Classification Society, incurring an element of programme risk. Add to this the highly skilled nature of idealising a hull structure into an efficient mesh for analysis and it’s clear to see the benefits to be gained by streamlining the FEA cycle.

The interface is called AVEVA Hull Finite Element Modeller, and is actually much more than just a file converter; it is a model converter which incorporates a great deal of specialist know-how for idealising real hull structure into an efficient mesh. Mr Neuvéglise explained that hull structural design is a unique discipline, with many rules and principles not found in other steelwork applications. AVEVA’s hull design modules are intelligent and automatically apply specialist marine design practice to the various structural elements as the designer creates or repositions them.

Hull Finite Element Modeller takes this one step further by applying intelligence to the idealisation process. It applies default mesh parameters and idealisations which accurately translate the majority of the true design intent at the first pass and allows the structural analyst to adjust these as necessary to fine tune the mesh, for example to increase mesh density around local ‘hot spots.’ Filtering can be applied to ignore the many small holes and cut-outs characteristic of hull structures, speeding up the subsequent analysis. Once the optimum idealisation has been configured, Hull Finite Element Modeller generates and

exports a session file using Patran Command Language (PCL) for execution in Patran. This file contains all the commands required to create simplified geometry of the original hull model, together with material and element property definitions, within a Patran database.

Users commonly run FEA twice; once in the early stages of the design to weed out any major problems and later once the design is sufficiently detailed for a final approval submission. By automating much of the laborious idealisation work, the new interface saves a lot of cost and time; customers estimate that it can halve the effort required, shaving around 1000 man-hours off a typical project. This is not only a direct economic benefit; it can take FEA off the critical path and bring forward Classification Society submission, reducing the risk of having to rework fabricated hull structure. And because it makes FEA so much easier to perform, it encourages more extensive analysis and optimisation to increase design quality and/or to achieve further hull cost savings.

Mr Neuvéglise was very complimentary about AVEVA’s relationship with MSC, citing the excellent training and support provided to AVEVA’s technologists when developing and

A Solution Shipbuilders Can Rely OnSavings through Idealization of Production ModelsAVEVA, an MSC Technology Partner | Based on an interview with Stéphane Neuvéglise

Hull structure shown in AVEVA Marine The same hull structure after idealisation, ready for mesh generation

PARTNER SHOWCASE

Page 39: Simulating Reality

validating the new interface. The company had also worked closely with an offshore platform builder, the lead customer for the new Patran interface. Working with selected customers during development helps AVEVA to ensure that its solutions support industry working practices and enables testing on real ‘warts and all’ project data. This focus on practicality has resulted in a solution which delivers power and ease of use for everyday needs, while avoiding the configuration complexities of trying to cater for every possible situation. “Shipbuilders are very practical people,” said Mr Neuvéglise.

‘They understand the principle of diminishing returns and, above all, they want tools that

they can rely on. On several occasions, ideas for minor refinements were rejected by the customer as not justifying the effort or the increased complexity. Simple is good in this industry.’

“Everybody who has used the product has been enthusiastic about it,” Mr Neuvéglise went on. “Shipyard managers appreciate not only the direct savings, but also the reduction it can make in programme risk. Yard capacity is finite and a delay on one project can have a knock-on impact on others. The industry is also suffering from skills shortages in all areas, so this goes some way to mitigating that.”

We wondered whether automating so much of a skilled process would devalue the work of the professional analysts. “Quite the opposite,” said Mr Neuvéglise. “Everybody likes more powerful tools to work with and manually idealising hull structures can be very laborious and repetitive. Now analysts can concentrate on the really interesting and important parts of the work while the software does the heavy lifting for them. They love it!”

With around half the world’s shipbuilders being users of Patran, engineers could come to love AVEVA Hull Finite Element Modeller as well! u

Finally, the same idealised structure transferred into Patran

Join the MSC EMEA Business Partners Community!Expand your business opportunities by representing one of the ten original software companies and the worldwide leader in multidiscipline simulation.

Henk Viljoen, Managing Director - ESTEQ Engineering

Visit www.mscsoftware.com/partners or contact [email protected]

As an MSC reseller, you always know you’re not alone, but that practical help is around the corner when assisting customers to tackle technology challenges and drive innovation.

Now analysts can concentrate on the

really interesting and important parts of the work while the software does the

heavy lifting for them. They love it!

Page 40: Simulating Reality

38 | MSC Software

UNIVERSITY & RESEARCH

Engineering Simulation has become a critical, if not integral, part of the

design and development process at all forward looking companies, big and small. Students entering the job market are expected to be familiar with simulation technologies and have proficiency in Computer Aided Engineering (CAE) solutions, which would also give them an edge in the competitive job market.

Nonlinear finite element analysis is a fast growing segment of CAE, with companies looking for engineers to fuel innovation in the fast moving market. Because of its versatility, nonlinear finite element analysis is used across major industries like automotive, aerospace, electronics, shipbuilding, energy, biomedical, machinery and more, to solve problems with manufacturing, product performance and failure. As companies increasingly try to save costs, take advantage of new materials and design products with tight tolerances, they see a strong need for efficient nonlinear solutions to answer questions like:

• How does the nonlinear stress-strain relationship affect the performance of the product?

• Is the structure going to buckle, if so where? How does it behave after undergoing buckling?

• How does the product perform under in-service conditions, under varying environmental conditions? How will it perform under very hot or very cold conditions?

• Does the product continue to perform effectively during its entire life?

• Does the product fail? If so, under what conditions? Would it lead to a catastrophe?

Marc, the dedicated nonlinear FEA solution from MSC Software and the world’s first commercial nonlinear FEA solver, is used across industries to analyze and improve a wide range of products and processes, including:

• Elastomeric products - seals, boots, belts, hoses, tires

• Biomedical devices - stents, defibrillators, prosthetic implants, hospital beds

• Electronic components – solder, circuit boards, connectors, actuators

• Manufacturing processes - sheet metal forming, glass forming, welding

• Machinery components – gears, springs, bushings, conveyor belts

• Consumer products – bottles, cans, appliance components

• And more…

Marc 2012 Student Edition is designed to provide students with a great introduction to nonlinear finite element analysis. The software is available for free to registered students, giving access to most of the same capabilities available in the commercial version. Its 5000-node limit is large enough to model and solve problems that provide good insight into mechanics of structures. This helps reinforce the concepts learned in the classroom,

Creating Future CAE UsersGetting Prepared for a Successful CareerBy Srinivas Reddy, MSC Software

Boot Simulation Gear Contact Analysis Tube Bending Analysis

By learning Marc, students do not just learn how to use an FEA tool, but gain a

deeper understanding and appreciation of

nonlinear mechanics helping them to be the innovators they want

to become.

Page 41: Simulating Reality

while developing critical simulation skills required for a career in product design and development.

Whether students are studying manufacturing processes or learning to apply principles of mechanics to product design and development, Marc is the perfect solution for getting ready to simulate the real world behaviors. With Marc, users can:

• Model a wide range of materials metals, rubbers, plastics, shape memory alloys, concrete, powder metals, soil and more

• Understand failure of materials through multiple damage and failure models implemented in Marc

• Solve the toughest problems with the help of automatic remeshing

• Analyze interaction between bodies and study the contact pressures and forces

• Study multiphysics behavior when coupled with nonlinear structural response accurately

This powerful, versatile solution is also easy to learn, so students do not have to spend countless hours to gain the working knowledge in using it. Various resources available also help students in improving their knowledge of nonlinear FEA, which include:

• Several “getting started” videos that walk the user through the user interface,

orienting them through the steps of model creation, analysis and post-processing. This is done with simple textbook examples, so that students can compare the results with theoretical solutions. These examples are available online at the following link: www.mscsoftware.com/marc-videos

• Training notes providing step by step examples and deeper exposure to the software so that students are prepared to create more complex models and learn about the intricacies of nonlinear FEA These notes and examples are provided to the students with installation.

• Extensive documentation with theoretical background and examples helping users to gain in-depth knowledge and be ready to simulate reality.

In this challenging job market, the Marc Student Edition gives students the advantage they need and technological expertise required to succeed. By learning Marc, students do not just learn how to use an FEA tool, but gain a deeper understanding and appreciation of nonlinear mechanics helping them to be the innovators they want to become.

Current students can register and download it for free at www.mscsoftware.com/ student_edition. u

Solutions for:

Services:

Structural AnalysisCFDExperimental Stress AnalysisExperimental Fluid DynamicsTest&MeasurementVehicle DynamicsNVHShock

Training and Consultancy ServicesTesting Services

Advanced Engineering Solution

ww

w.ex-en.com

.tr

EKSEN ENGINEERING

MSC’s Business Partner for more than 10 years!

Popular User Manuals Available

View on iTunes & Google Play to buy and Download books.

MSC Nastran eBooks

Page 42: Simulating Reality

40 | MSC Software

UNIVERSITY & RESEARCH

A Student’s Perspective

Hao Lam is an Engineering Analyst at Vacco Industries, a wholly owned subsidiary of ESCO

Technologies Holding Inc. VACCO Industries is the leading supplier of fluid components to the space and defense industry worldwide, as well as Multi-Fab manufacturing services. Here is how he describes his job:

“My job is to perform calculations solving engineering problems which help the design engineers effectively design products to meet the standard demanded by the customers. The engineering problems can sometimes be difficult to solve and require a wide range of experience and knowledge to understand it and analyze it correctly.”

When asked why he wanted to enroll in the International Masters in FEA program offered by UNED, he replied: “I always seek ways to improve myself and allow myself to be exposed to new techniques and strategies than I can learn to become a better analyst. Recently, I have been involved in using MSC analysis software Marc to solve a nonlinear problem, and I had tremendous success with it, and that is how I became interested in MSC software’s, and that also led me to enroll in the Master FEA program developed by Ingeciber and UNED

In today’s world, knowledge has become the key to moving forward. No longer

is a bachelor’s degree sufficient in most cases to help people get to the positions they desire. In engineering it is no different. While engineering is constantly advancing in the types of products developed, physics is (mostly) the same as it was 100 years ago. But as companies continue to develop more complex and sophisticated products, the ability to visualize and simulate is constantly advancing leaving many engineers with a dilemma; should I get a masters in engineering degree and improve my theoretical understanding of engineering, or should I get an MBA and go down the management track? While this will always be a challenge and will need to boil down to individual goals, in this article we would like to offer a third option. A practical Masters’ degree in FEA. While this degree does not yet offer the ability to continue to a PhD, it does enable you get both the knowledge and experience to better leverage advanced Simulation tools in developing your products.

The idea behind this program started in 1994. Ingeciber a developer and distributor of CAE tools approached UNED (UNIVERSIDAD NACIONAL DE EDUCACIÓN A DISTANCIA) Distance Learning University in Spain with a proposal. Create a distant learning program with the objective of preparing specialists in the use of Finite Element Method (FEM) and CAE Simulation for practical professional application. By bringing together Specialists in FEM (University professors and Ingeciber Senior engineers) they have designed an elaborate study program that has finally evolved into the International Masters in Theoretical & Practical Application of Finite Element Method and CAE Simulation offered globally today.

More than 2,500 postgraduates have participated in this Master’s since its inception, contributing to the prestige and recognition the program has developed over the years.

Committed to excellence in teaching, research and innovation, the Partnership between Ingeciber and UNED is actively engaged with industry, governments and universities across the world in tackling some of the major challenges of today and tomorrow.

UNED was founded in 1972. Its goal is to provide university access to postgraduate students who are working or cannot devote several hours a day attending a regular university or who live too far away from any university campus and finally to some students with many years of experience who still have the desire to keep on improving their skills.

In its 40 years UNED has already trained well over 1 million students and is widely recognized as the largest University in Spain. This year UNED has:

• Over 250,000 students split between 9 Faculties, 2 Engineering Colleges, 49 Postgraduate Master’s and 44 PhD Programs

• A faculty over 7,000 Professors and Tutors• Associate Centers in 15 countries around

the world• 2 TV stations (Canal UNED 24 hours) and

a Radio station• And is prominently available in the social

media: Facebook, YouTube, Twitter and LinkedIn

Ingeciber, S.A. was founded in 1986. The partners, who are all still involved in the company as part of the executive team, invested both their money and their minds to make Ingeciber a reality. Using their broad knowledge of analysis and Computer Aided Engineering, they have grown the company to what it is today. Starting as an engineering

Ingeciber, S.A | By Gonzalo Gardeta

Gain knowledge to Better Leverage Advanced Simulation Tools

A Practical Master’s Degree in FEA

Page 43: Simulating Reality

to become an expert in finite element analysis using extensively MSC software’s Patran and MSC Nastran”. So now that he has been in the program for about three months, here is what Hao has to say: “The UNED FEA Master program powered with MSC Software’s simulation products is a well-structured online program with a great team of staff that is always available to support students’ needs. For the past two months, I have had great success learning Patran and MSC Nastran. The Professors are very helpful and responded to my questions in a timely manner. I am enjoying this program greatly”. Finally we asked Hao, if he felt the course would help him in his current and future engineering roles: “The level of difficulty of the course has given me very interesting challenges. I have learned new techniques to solving finite element problems, and also to compare to what I had learned before I enrolled in this program. It is a great learning experience. The structure of the program also demands me to balance my daily time between my job and school. After completing this program I will be able to apply my knowledge to my job on a daily basis, and at the same time enhance my career to a higher position”.

We would like to thank Hao for joining the Master’s program and for sharing his thoughts with us. Good Luck with your degree.

services organization, Ingeciber quickly captured the eyes of many prestigious CAE software companies, and has partnered with them in various endeavors. Today Ingeciber, S.A represents MSC Software in Spain and is an expert in solving complex engineering problems by using simulation and the application of the Finite Element Method, Boundary Elements, Finite volumes, as needed.

What started as an engineering services organization, followed by representation of third party software programs, eventually expanded to the application of technology in the design processes and mechanical, thermal, structural, fluids, acoustics, mechanical systems calculation. The Skilled Engineers of Ingeciber, S.A. bring a diverse

background of experiences and education from the Mechanical, Automobile, Civil, Aeronautic and Naval Engineering fields. These experiences coupled with an acute understanding of the market needs, led to the development of CivilFEM, Ingeciber’s flagship product for addressing complex FEA in the Civil engineering field. CivilFEM’s new release coming out in Q4 2012 is powered by MSC’s Marc software allowing it to address challenging nonlinear problems other codes in this market find difficult to address.

MSC Software has partnered with UNED to bring students enrolled in the FEA Masters Program the latest CAE technology available to provide the skills and training crucial to their success in the industry. u

Page 44: Simulating Reality

For more information, visit: www.mscsoftware.com/50years

SAVE THE DATE2013 USER CONFERENCES

DATES & LOCATIONS AmericasTuesday / Wednesday May 7/8

EMEASweden - Monday / Tuesday - May 13/14 Germany - Tuesday / Wednesday - May 14/15 France - Wednesday / Thursday - May 15/16 UK - Thursday / Friday - May 16/17 Italy - Tuesday / Wednesday - May 21/22 Russia - Wednesday / Thursday - May 22/23 Turkey - Monday, May 27th

JapanThursday, May 30th

APACKorea - Monday / Tuesday - June 4/5China - Wednesday / Thursday - June 6/7India - September (Date TBD)

MSC Software will celebrate its 50th anniversary in February, 2013. Please join us in celebrating at the 2013 User Conferences.

The events will take place in May and June. Location details coming soon.

*Dates and locations are subject to change.

AD_MSC_50th-A4-1.indd 1 8/13/12 4:29 PM