simulia-scn-1311

8/10/2019 simulia-scn-1311

1/24

DANA ACCELERATES

PRODUCT DESIGNWITH SIMULATION

LIFECYCLE MANAGEMENT

The 3DEXPERIENCE Company

SIMULIACOMMUNITY NEWS

#06 NOVEMBER 2013

8/10/2019 simulia-scn-1311

2/24

10

MATRIX

12

DANA

16

G

EORGFISCHER

SIMULIA Community News is published by Dassault Systmes Simulia Corp., Rising Sun Mills, 166 Valley Street,

Providence, RI 02909-2499, Tel. +1 401 276 4400, Fax. +1 401 276 4408, [email protected], www.3ds.com/simulia

Editor: Tim Webb Associate Editors: Kristina Hines, Jessica Nicholson Graphic Designer:Todd Sabelli

The 3DS logo, CATIA, SOLIDWORKS, SIMULIA, DELMIA, ENOVIA, GEOVIA, EXALEAD, NETVIBES, 3DSWYM and 3DVIA are either trademarks or

registered trademarks of Dassault Systmes or its subsidiaries in the USA and/or other countries. Other company, product, and service names

may be trademarks or service marks of their respective owners. Copyright Dassault Systmes, 2013.

in thisIssueNovember 2013

3 Welcome LetterColin Mercer, VP R&D SIMULIA, Dassault Systmes

4 Product UpdateNew Releases of Abaqus 6.13 and Isight 5.8

5 NewsDassault Systmes Acquires Safe Technology

6 Optimization SpotlightVoith Improves Efficiency of Automatic

Transmission Components

8 Strategy OverviewThe Quest for the Best

10 Case StudyMatrix Applied Computing Helps Springfree Optimize

the Backyard Trampoline

12 Cover StoryDana Accelerates Product Design with

Simulation Lifecycle Management

15 AlliancesGRM Consulting

Dynardo

16 Optimization SpotlightGeorg Fischer uses Tosca Structure to

Improve Fuel Efficiency18 Case Study

DNV uses Abaqus to Study Blowout Scenarios

20 Academic UpdateUniversity of Toronto Blue Sky Team uses Abaqus

for their World Solar Car Challenge Vehicle

22 Tips & TricksCalibrate materials using Isight and Abaqus

23 User CommunitySIMULIA Learning Community Update

Contributors: Keith Alexander (Springfree), Guido Quesada (Matrix Applied Computing),

Shane Finneran (DNV), Bernd Whrle (Voith), Frank Popielas (Dana), Roman Brauner

(Georg Fischer), Paul Park and Ahthavan Sureshkumar (University of Toronto), Parker GroupOn the cover: Frank Popielas

8/10/2019 simulia-scn-1311

3/24

Welcome Letter

New Team Members Solidify

Our Vision for the Future

As we near the end of 2013 and turn our attention to planning for 2014 and the coming years,I find this to be an opportune time to share with you whats been happening within SIMULIA

R&D. Our teams have been hard at work to get new releases out to our user community whileimproving the depth, breadth and quality of our products. This year we have expanded our R&Dteam and technology coverage through acquisitions of FE-Design, Simpoe and Safe Technology.

And we are focused on delivering next generation realistic simulation applications inside theDassault Systmes 3DExperience platform. As you can see, 2013 has been a very busy andproductive year!

This issue of SIMULIA Community News(SCN) highlights our growth and leadership indeveloping solutions for process integration, automation, and design optimization. Ourcommitment to delivering these industry-leading solutions began in earnest in 2008 with theacquisition of Engineous Software for the Isight and Fiper technologies. This coincided with therelease of our Simulation Lifecycle Management applications (see page 12). Earlier this year, we

took the next step by expanding and delivering the most powerful optimization tool set availablethrough the addition of the FE-Design team and their Tosca applications (see page 8)

You can read about Voith, a leading specialist for automatic transmissions, and their use ofTosca Structure to produce a lighter and more robust design on page 6. The case study fromGeorg Fischer Automotive illustrates their use of Tosca to reduce component weight by 32%and ensure their product conforms to all required stress loads (see page 16). The case study byMatrix Applied Computing on the optimization of a cleat for a springfree trampoline is yet anotherinnovative example of using Tosca technology with Abaqus (see page 10).

These are just a few of the many user success stories that illustrate the business value yourorganization can gain in leveraging our robust and proven optimization technology. I encourageyou to find out more about this exciting technology on our public website and in the SIMULIALearning Community.

While our team, portfolio and vision have all expanded significantly over the years, our focus

remains exactly the sameto provide our user community with:

Innovative, state-of-the-art realistic simulation technology based on fundamental mechanicsprinciples that capture the physics behavior

High-quality and global technical support

An open platform for integrating and collaborating on simulation processes, performing designexploration and optimization, and managing your simulation data and intellectual property

We are committed to working with you to influence the technology and user experience that isbeing made available through the Dassault Systmes 3DEXPERIENCE platform. The goal isto make our robust, realistic simulation and optimization applications open and accessible tousers of all skill levels through shared best practices, simulation collaboration, and data andIP management. We strongly believe that simulation, deployed in this manner and integratedinto your business processes, will create enormous opportunity and value to you and your

organization.We sincerely appreciate the long-term relationship and confidence that you have in us and ourproducts. Your candid feedback is especially important in helping us to deliver products thatallow you to focus more time on your engineering design challenges. Your use of our productsprovides monumental value to other users in our community as well as our society. We lookforward to your continued commitment to sharing your knowledge and experience with ourglobal community at the 2014 SIMULIA Customer Conference in Providence, Rhode Island nextMay. The location near our Providence R&D headquarters will enable you to meet with many ofour team members during the conference. I look forward to our ongoing collaboration to expandthe access and benefits of realistic simulation.

With best wishes for 2014,Colin MercerVP R&D SIMULIA, Dassault Systmes

3SIMULIA Community News November 2013www.3ds.com/simulia

8/10/2019 simulia-scn-1311

4/24

4 SIMULIA Community News November 2013 www.3ds.com/simulia

Product Update

SIMULIA delivers a scalable suite ofunified analysis products including

Abaqus for FEA, Isight for parameteroptimization and process control, andSEE, which allows all users, regardless oftheir simulation expertise or domain focus,to collaborate and seamlessly sharesimulation data and approved methodswithout loss of information fidelity.

The latest releases of Abaqus, Isight, andSEE provide powerful new capabilitiesand enhancements that will improve yourproductivity, efficiency and performance-based decision making.

Abaqus 6.13 provides provensolutions for linear, nonlinear, andmultiphysics simulation. The releasedelivers breakthrough capabilitiesand enhancements for Performance,Electromagnetic Analysis, Heat Transferand more.

Abaqus 6.13 introduces a new particlemethod that allows users to performan analysis using the Discrete ElementMethod (DEM). This enables themodeling of particulate material behaviorin pharmaceutical, chemical, food,

ceramic, metallurgical, mining, and otherindustries and is well-suited for particlemixing applications. The particles interactgenerally with each other and with allother parts of a model.

Additionally, steady-state flow problemscan now be solved directly, eliminatingthe need to approximate steady-stateconditions using a long-duration transientflow simulation. This enhancementsubstantially improves overallcomputational performance incommon steady-state CFD simulations.

Direct coupling between anelectromagnetic and a thermal or stressanalysis procedure is now supportedthrough the co-simulation capability. Thiscapability allows simulation of problemssuch as induction heating, in which theJoule heat output from an electromagneticanalysis drives a thermal analysis, whilethe temperature output from the thermalanalysis affects the electromagneticfields through temperature-dependentelectromagnetic properties.

New Releases of Abaqus, Isight, and

Simulation Execution Engine Now Available

For More Informationwww.3ds.com/abaqus

www.3ds.com/isight

Isight 5.8 delivers the market-leading, open-desktop solution for simulation processautomation and design exploration. Newcapabilities and enhancements in Isightfocus on Model Building, Components,Runtime Gateway Visuals, and LicenseUsage.

In this release, the new visual CorrelationTable displays a heat map of correlationvalues between selected input and outputparameters. This helps users to quicklyidentify highly correlated input-output pairs.

In previous releases of Isight, it was not

possible to create array graphs by usingmultiple arrays. Isight 5.8 now supportsarray graphs using slices of two differentarrays. For example, given displacementversus time and force versus time data,the user can now easily graph force versusdisplacement. This greatly simplifies post-processing of Abaqus component historyarrays.

SIMULIA Execution Engine (SEE) 5.8enhances the power of simulation byenabling users to distribute and parallelizethe execution of simulations and shareIsight simulation process flows and results.

SEE can be used directly from Isight, acustomizable web interface, or as part of aSIMULIA Simulation Lifecycle Managementdeployment.

SEE provides a new capability to allowusers to end a single work item on thedashboard, as opposed to ending anentire job.

Also in this release of SEE, users cannow determine whether to continue orterminate a job at startup if a requiredaffinity is missing when unsatisfied work-item affinities would cause the execution to

pause and wait for a station to come online.

New co-simulation functionality in Abaqus 6.13

enables multiphysics coupling of electromagnetic

and thermal domains as shown in this example of an

induction-heated fuser. Understanding temperature

distribution and temperature rise time of fusing rollers

is critical for energy-saving considerations of printers.

A new visual Correlation Table was added to Isight tohelp the user to easily identify highly correlated input-

output pairs.

Abaqus release 6.13 can now model braided or

interwoven wire products made from stranded

materials. This enhancement reduces computational

modeling time by orders of magnitude.

8/10/2019 simulia-scn-1311

5/24


For More Informationwww.3ds.com.com/fe-safe

On September 4, 2013, DassaultSystmes announced the acquisitionof Safe Technology Ltd., the technologyleader in fatigue simulation for predictingand extending product durability. Theacquisition expands Dassault Systmes3DEXPERIENCE platform and SIMULIAsrealistic simulation applications to providethe most complete and accurate durability-prediction solution on the market. Safe

Technologys fe-safe software is usedby more than 500 customers worldwide,including General Motors, CaterpillarInc., Cummins Inc., Emerson Climate

Technologies, Honda Jets, Harley DavidsonMotor Company and Hyundai Motors.

Consumers want a product that is builtright. They want a product that lasts.Durability deeply affects the emotionalattachment between brands and theirusers. The 3DEXPERIENCE platformis all about building brand loyalty for acompanys business and the productsit offers. This is why Safe Technology is

such a good fit for Dassault Systmes,said Bernard Charls, President andCEO, Dassault Systmes. Advancedfatigue and durability software solutionsare an essential part of the product designprocess. Safe Technology will enhanceDassault Systmes SIMULIA structuralsimulation to predict and analyze productlife quickly and accurately.

The manufacturing companies that aregaining the most value from fatigue analysishave made a strategic decision to integratefe-safe into their daily design process. Thisis enabling their design and engineeringteams to develop durable products fasterand more affordably. Many of thesecompanies are already using Abaquswith fe-safe. This enables multiple loadingconditions with the cyclic load history, from

Abaqus, to be used by fe-safe to predictthe life of the component. Additionally,several companies are also using Isightand Tosca with fe-safe to accelerate designexploration and optimize their productdesigns for cost, weight, shape and more.

fe-safe also includes an extensive databaseof cyclic material property data and permitsusers to add their own property data fromfatigue tests. fe-safe interfaces to all majorFEA packages. SIMULIA is committed tocontinuing these partnerships to ensurethat fe-safe is open and provides the mostadvanced fatigue technology available for allFEA users.

Safe Technology has been enhancing the

capabilities of fe-safe continuously since theearly 1990s, in collaboration with industryleaders, to ensure that it continuously setsthe benchmark for fatigue analysis software.

Technical firsts include:

Providing multiaxial fatigue methods as astandard capability

Incorporation of Battelle's StructuralStress Method for fatigue of welded joints

High temperature and thermo-mechanicalmethods

Creep fatigue methods

Physics-based approach tocomposite fatigue

Methodology for fatigue of elastomers

Custom Framework module to enableusers to use their own proprietary fatiguealgorithms

Distributed processing capability thatenables users to achieve linear speedincreases

Solution to provide strain correlatedFEA load calculations and turn complexcomponents into multi-channel load cells

There will be many opportunities to learnabout fe-safe in the coming months. Rightnow, you can learn about fe-safe by viewingwebinars and downloading technical papersavailable at the Safe Technology website.

There will also be training classes availablein 2014 and you can meet with fe-safeexperts at several industry conferencesthroughout the year. In addition, you will beable to learn more about fe-safe at the 2014SIMULIA Community Conference to be heldin Providence, Rhode Island.

Safe Technology, the Leader in Durability

and Fatigue Analysis, Joins Dassault Systmes

Achieve real business value by making fatigue analysis anintegral part of your design process

Fatigue of welds are notoriously difficult to predict accurate

This is a weld fatigue analysis of a prototype rear trailing arm

link. Test results to crack initiation: 0.83 repeats of block

cycle test. fe-safe fatigue life prediction to crack initiation:

0.81 repeats Courtesy: Ford Motor Company

Fatigue damage does not necessarily occur at the point

of maximum principal stress. In this fatigue analysis of

a tube yoke suspension component, fe-safe accurately

predicted the real point of failure as confirmed by lab

tests on actual loaded specimens. Courtesy: DANAAutomotive Systems Group, LLC

This is an analysis of a supercharger spring. When the

spring is in operation there is a very complex multiaxial

stress state in the coil and high residual stresses

because of the forming process. This is another casewhere fatigue damage does not start at the point

of maximum stress. Test results to crack initiation:

60,000 cycles. fe-safe fatigue life prediction to crack

initiation: 67,500 cycles. Courtesy: Eaton Corporation,

Vehicle Group, USA

News

8/10/2019 simulia-scn-1311

6/24


Optimization Spotlight

Voith Improves Efficiency of

Automatic Transmission ComponentsLeading specialist for automatic transmissions benefitsfrom Tosca Structure optimization software for

development of its DIWA automatic transmissions.

The increased efficiency sets new standards.

As environmental standards become stricter and consumer demands for vehicleswith greater fuel efficiency rise, automotive suppliers need to support theircustomers with lighter components that will help meet the push for lower emissionsand better gas mileage. Voith Turboa division of Voith GmbH that specializesin hydrodynamic drive, coupling and braking systems for road, rail and industrial

applications and ship propulsion systemsdetermined that one way to help meetthese challenges was to enhance the efficiency of its transmissions.

Voith targeted weight reduction through material savings as a primary route to increaseefficiency. Although their engineers had already manually optimized their transmissiondesigns, they felt more savings could still be identified. To find the optimum geometry ofautomatic transmission components, Voith turned to topology optimization with ToscaStructure software from FE-DESIGNsince May, 2013, part of the Dassault Systmes3DEXPERIENCE technology portfolio under the SIMULIA application.

Photo: Daimler AG

Significant material savings

through weight reduction

with Tosca Structure

30%

Weight Reduced

Planet CarrierOriginal Design CAE Design

Tosca Structure

design approach

8/10/2019 simulia-scn-1311

7/24


For More Informationwww.voith.com

www.3ds.com/tosca

By using Tosca Structure we found the best solution to realize material savings and

increase efficiency in the product devolopment process of our automatic transmissions.

Thus the topology optimization of our planet carrier resulted in a lighter and morerobust design as well as significantly higher production efficiency.

Bernd Whrle,Technical Calculations

Bus Drive Systems, Voith

Optimizing a planet carrier: weight reductionby creating a new design

Tosca Structure is a flexible, modular software system for non-parametric structural optimization, e.g. topology, shape, and/or bead, using finite element analysis (FEA). Voith Turbo used

Tosca Structure to optimize a planet carrier component, part of

the gearing system in its DIWA automatic transmission. Planetarygearing systems consist of one or more outer, or planetary, gearsthat revolve around a central sun gear. The outer gears aremounted on the planet carrier, which is a moveable arm that itselfrotates around the sun gear.

Starting with CAD data of the carrier, the engineers firstused Tosca Structure to define the available design space bysubtracting functional areas and joint spaces to connecting areas.Restrictions placed on the optimization ensured that any design ormanufacturing requirements are kept. In this case the functionalstiffness of the planet carrierrequired to guarantee bearingdurability and equal load on the tooth flankswas maintained.

Tosca Structure then automatically identified areas that did not

contribute to the force flux and removed the materials withinthese areas that were not essential. Because requirements for themanufacturing process were taken directly into account duringoptimization, the design proposal could then be easily transferredto Voith Turbos own CAD system where modifications were madein order to meet further casting parameters.

By using Tosca Structure we found the best solution torealize material savings and increase efficiency in the productdevelopment process of our automatic transmissions, said BerndWhrle, Technical Calculations Commercial Vehicles, Voith Turbo.

Weight savings also provide manufacturing benefitsThe new design of the planet carrier generated significant savingsin material, with a weight reduction of more than 30%. The more

compact size of the revised planet carrier permits an additionalcomponent to be placed in the molding box, allowing the samenumber of castings to be produced with fewer casting processes.

Voiths experience with Tosca Structure shows that even designsthat have already been manually optimized still contain significantoptimization potential. The updated design of the planet

Voith uses SIMULIAs Tosca Structure topology optimization software to

reduce the weight of components such as the planet carrier in their DIWA

automatic transmissions.

carrier maintained the required stiffness and lifetime, achievedconsiderable material and weight savings, and resulted in moreeconomical production.

The topology optimization of our planet carrier resulted in a lighterand more robust design as well as significantly higher productionefficiency, Whrle said.

8/10/2019 simulia-scn-1311

8/24


Strategy Overview

C reating the best design forperformance, weight, cost and othermission-critical factors faster than thecompetition is essential for launchingmarket-winning products. Findingthe best design options requires amultidisciplinary approach with multipletrade-off studies. To reduce the timeand cost of physical testing, companiesare now turning to virtual testing alongwith design exploration and optimizationtechnology to sort through the growingcomplexity and accelerate decision-making.

Dassault Systmes is a world-leader indeveloping 3D software applications totransform the way products are designed,produced, and supported. The companyis delivering on a long-term strategyto provide the most powerful realisticsimulation and design optimizationtechnology available to the researchand product development market. The

expanding technology portfolio forrealistic simulation and optimizationincludes embedded simulation withinCATIA and SolidWorks, plus advancedmultiphysics, durability, automation,design exploration and optimizationcapabilities with Abaqus, fe-safe, Isight,and Tosca. Together these applications

provide the most complete, open androbust optimization toolset available. Theyalso form the basis for next-generationrealistic simulation and optimizationtechnologies within Dassault Systmes'ground-breaking 3DEXPERIENCEplatform.

Parametric and Non-ParametricOptimization for Product Innovation

SIMULIA Isight is widely used to integratemultiple cross-disciplinary models andapplications together in a repeatablesimulation process flow. Once defined,they can be executed across distributedcompute resources, rapidly exploring thepossible design space and identifying theoptimal design based on the user-definedparameters.

Tosca, recently acquired by DassaultSystmes, is the leading technology fornon-parametric structural and fluid flowoptimization. Tosca Structure delivers

powerful optimization solutions for designof lightweight, stiff, and durable parts andassemblies. Tosca Fluid enables topologyoptimization-driven design concepts forfluid flow systems. Product designs arecreated directly through the optimizationprocess, driven by the performance goalsspecified. As a result, innovative design

proposals can be optimized earlier in thedesign process.

In addition to robust optimizationcapabilities, openness and flexibility areessential in todays dynamic development

environments. Both Isight and Tosca openlysupport integration with most leadingFEA and CFD solvers and pre-processors,including Abaqus, ANSA, AnsysWorkbench, Fluent, and Star-CCM+. Theresulting geometries from a Tosca or Isightoptimization study can be easily transferredback to the originating CAD program toupdate the final design.

Starting with the Abaqus 6.13-AP release,Tosca technology will be available forAbaqus users through Extended Tokenlicensing. This licensing option alsoprovides access to Isight as well assimplicity, flexibility and cost effectivenessfor small or large teams wanting to leveragemultiple technologies from SIMULIA.

Complimentary OptimizationWorkflowsWhile there are optimization tasks forwhich either a parametric approach or anon-parametric approach fits best, thereare certain optimization tasks where thecombination of Isight and Tosca Structureobtains the overall best result. Possibleworkflows that can be considered are:

Sequential Coupling: Isight optimizesthe parametric part of the design andthe results of this optimization serve asconstrains for a non-parametric Toscaoptimization for a different region of thedesign. Isight also acts as an automationtool to integrate both parametric and non-parametric optimization.

The Quest for the BestOptimization: an act, process, or methodology

of making a design, system, or decision as fully

perfect, functional, or effective as possible...Merriam-Webster Dictionary

COMPLIMENTARY

OPTIMIZATION

WORKFLOWS

SIMULIA Tosca

Structural topology

Non-parametric Optimization Parametric Optimization

Design of Experiments

Structural shape Six Sigma

Taguchi Method Bead

Multidisciplinary

Optimization

Fluid topology Monte Carlo Analysis

SIMULIA Isight

8/10/2019 simulia-scn-1311

9/24


Close Coupling: Isight creates variouslocal optimization tasks driven by global

optimization criteria. For each localoptimization task Tosca provides the bestoptimization solution. Afterwards Isightchooses the best solution according to theglobal optimization criteria.

Sequential Coupling ExampleIn a typical design analysis of anautomotive suspension, a sequentialworkflow is followed. First the design ofthe suspension joint location is completedthrough Kinematic and Compliance (K&C)analysis with rigid multi-body software.

This is followed by the topology designof the suspension components based on

Design optimization is a complex way of saying finding

the right design for business, for customers, and for the

world. The 3DEXPERIENCE platform is all about optimizing

a companys business and the products it offers. Is it whatcustomers want? Can we produce it quickly and effectively?

Is it right for a sustainable future? These are the questions

industry must ask to harmonize products, nature and life.

And these are the questions our 3DEXPERIENCE platform

answers.

Bernard Charls, President and CEO, Dassault Systmes

For More Informationwww.3ds.com/tosca

www.3ds.com/isight

the stress analysis using commercial finiteelement software.

With Abaqus, Isight, and Tosca, thissequential workflow can be combined tooptimize suspension joint location andstructural topology of the componentssimultaneously (see Figure 1). Theanalyses are integrated using both Isightfor suspension joint location, and Tosca fortopology optimization of the suspensioncomponents, as a unified approach. Behindthe scenes, Abaqus is used for both K&Canalysis and stress analysis allowing, forexample, accurate loading from the tires tobe considered.

To optimize the suspension joint locations ofa rear multi-link suspension, Toe and Cambercan be considered as objective functions andthe joint point coordinates of the suspensionlinks set as design variables. The joint pointsof the links are optimized while the Toe andCamber behavior follows throughout theK&C analysis. Once the optimal suspensionjoint points are obtained, the topologyoptimization analysis is automaticallyperformed starting from simple beam shapeswith rectangular cross-section profiles.

This unified and automated approach makesit easy to perform systematic analyses tosimultaneously optimize the design of thesuspension joint locations as well as thetopology of the suspension components.

Business Value of Accelerating

Design OptimizationWith this unprecedented range of integratedtechniques, the potential for optimizationto transform the business impact thatsimulation can provide has never beengreater. There are already many examples ofhow Isight and Tosca technology are used byleading product development companies tosave significant time and cost.

Over time, the business value will grow asthese applications are fully integrated withother SIMULA process applications. Aspart of the 3DEXPERIENCE platform, thesepowerful automation and optimization

capabilities will be easily accessible to allusers of CATIA, SolidWorks, and SIMULIAtechnologies, and will continue to be openfor integration with other CAD, FEA, and CFDtechnologies. This will enable manufacturingcompanies to build on their current softwareand engineering investments and expertise.

When applied as an integral part ofthe overall product development cycle,SIMULIAs process integration, automation,exploration and optimization technology canhelp to maximize efficiency, meet businessobjectives, and get the best, most innovative,

reliable, and sustainable products to marketmuch faster and at lower costs.

Figure 1. Automated Tosca Optimization within Isight.

TOE Link

Create TOELink Script

Create TOE LinkFE Model ATOM Script for

TOE LinkATOM Analysis

for TOE Link

8/10/2019 simulia-scn-1311

10/24


Case Study

Ask a kid if they like trampolines andtheir eyes are sure to light up with

joy. But ask some parents and you mayget an entirely different response. Despitetightened manufacturing standards,covered springs and frames, and safetynet enclosures, some 90,000 childrenare still injured on traditional trampolinesevery year in the United States alone. Still,the equipment is becoming increasinglypopular worldwide.

Dr. Keith Alexander, an associate professorof engineering at Canterbury University,

New Zealand, felt there had to be a saferway to enjoy the trampoline. I alwaysbelieved that, as an engineer, I should bedoing things that benefit people, he says.When his wife nixed his plans to buy aconventional trampoline for his daughterKatie out of safety concerns, he decidedto build a better one. He began by workingin his garage, later moving his research tothe University where help from his studentsplus government and private grants keptthe momentum going.

Analyzing reams of data, Alexanderdetermined that there were three main

impact zones in the classic trampolinedesign (which dates to the 1930s) thatneeded to be arranged so the userwouldnt impact them: the springsconnecting the bounce mat to theframe, the metal frame itself and, ofcourse, the ground.

Many prototypes later (starting from 1989)Alexander arrived at his final, innovativesolution: move the frame below thejumping surface, create a soft edge to themat, and surround the whole thing withtensioned, ultraviolet-resistant netting that

would bounce users back towards thecenter if they went off-kilter. The result wasthe SpringfreeTrampoline (see Figure 1),first produced in 2003 and currently sellingat a rate of some 40,000 per year aroundthe globe.

So if trampolines are your thing, theSpringfree would certainly make a safer,healthy, eye-catching addition to thebackyard. Yet its what you dont seethats most unique: the high-tech cleatsunder the mats edge60 or more pertrampoline, depending on the model. Madefrom fiberglass-reinforced polypropylene

and shaped like a wide letter T, these havea rounded, midline socket into which theballed top of each fiberglass support rod fits(see Figure 2). The lower ends of the rodsconnect to the metal base of the trampoline.

The plastic edge fitting of my design hadalways been the most complicated bit ofthe eight or nine different versions I cameup with, says Alexander. The cleat, whichis inserted into a reinforced pocket in themat, harnesses the downward tension ofeach jump, ensuring that the ball at the topof the rod stays wedged into the socket asit moves up and down.

Once the Springfree was commercialized,word about its benefits began to spread.Parent- and safety-association awardsfollowed, sales bounced upwards andproduction swung into high gear. Thecompanys New Zealand-based injectionmolding supplier, Action Plastics, was keptbusy manufacturing cleats at a rate that hasnow reached some 2.5 million units per year.

As the molding tool used to manufacturethe cleat drew near the end of its lifetimein 2013, we had some pretty significantinducements to build a new tool andwe realized it was a great time to lookat optimizing and improving the cleatcomponent, says Hamish McIntyre,engineering manager at Springfree NewZealand Ltd. Because the cleat was so

complex in terms of loading and mechanicalproperties we decided we needed anengineering consultant to help us out. Welocated Matrix Applied Computing, througha Ministry of Science and Innovation expertsearch program, and selected them basedon their advanced capabilities.

Of particular interest to McIntyre wasMatrixs familiarity with realistic simulationusing Abaqus finite element analysis (FEA)software from SIMULIA, the DassaultSystmes 3DEXPERIENCE application.

Engineering a Safer BounceMatrix Applied Computing helps Springfree

optimize the backyard trampoline

Figure 1. A Springfree trampoline is

put to the test by an enthusiastic user.

Figure 2. Closeup of the innovative support structure of

a Springfree trampoline shows how white rods connect

via cleats to the upper, soft-edged jumping surface.

Lower end of rods attach to the secure metal frame

that sits on the ground. Protective netting not shown.

8/10/2019 simulia-scn-1311

11/24


I had worked closely with FEA teams atAirbus in the U.K. and knew how powerful

Abaqus was, he says.The fact that Matrix used Abaqus wasone of the things that drew me to thembecause Id seen the software in actionon advanced aircraft design.

As Guido Quesada, CAE Design Engineerat Matrix, recalls, We were already familiarwith Springfree trampolines and knew itwas a very technical product. Sometimesa potential customer calls with somefunny toy or crazy idea thats hard to takeseriously. This was definitely not the casewith Springfree. From the start, we knew

we had a significant challenge with thiscleat design project.

There were two main drivers of thechallenge, according to McIntyre: Wewanted to improve the strength, theultimate load capability, of the cleat. Andwe didnt want to introduce significantadditional cost into the component. Thequantity we buy each year is expandingalong with our sales growth. Just a couplecents less for materials per cleat endsup as a significant cost savings. We saw

this as an opportunity to tweak thecomponent and get the most out of it.

And so the tweaking began, starting withsome good old-fashioned field research.Both Matrix principal engineering analystDon Campbell and Quesada took informal

walks around their neighborhoods,watching people jumping on Springfreetrampolines in order to visualize whatwas happening to the cleats as the matswent up and down and the rods bent intandem.

Then it was back into the physical testinglab, where a device was built to apply theload from a rod to the center of a cleat,supporting it at the sides like the socketwould. SolidWorks CAD models weremade of the rod and cleat componentsset up in the testing apparatus. The CADmodels were meshed with Abaqus FEA,

and data captured through the physicaltests served as input for the computersimulations (see Figure 3) that predictedhow changes in cleat geometry wouldaffect strength and durability.

Right from the outset we worked closelywith our moldmaker [Sean Dryer of ActionPlastics] to ensure that what we came upwith was manufacturable, says McIntyre.Matrix imported Moldflow software resultsinto their Abaqus models to incorporatethe effects of the orientation of the glass-

fibers in the cleat and the residual stressfrom the manufacturing process itself.

The final cleat design the team arrived at(see Figure 4) uses slightly less materialoverall and has two ribs added on one side,to provide extra support where needed and

to distribute stresses more evenly. Its nowappearing in new Springfree trampolinesworldwide.

After the redesign project was complete,Quesada took a further look at the finalcleat geometry using SIMULIAs Abaqus

Topology Optimization Modeling (ATOM)tool, which automatically finds the optimumconfiguration (weight as well as shape) of acomponent while confirming performancewith FEA to ensure strength and durabilitytargets are met.

Our previous workflow, involving detailed

CAD modeling, mold simulations, nonlinearsimulation and simplified modal analysiswas ultimately highly successful, he says.

But when we applied ATOM to the samedesign, the process became much fasterand effective with automatic exploration ofthe optimal geometry. I was very surprisedto see that ATOM helped spot features Ididnt think of, including a third rib. It alsosaved us a lot of CAD work since the

ATOM procedure uses the same referencegeometry during each optimization round.

A new four-rib design, which enhancesthe automated optimization by putting

in additional engineering and designjudgment, is being considered for the nextgeneration of trampolines.

For More Information

www.matrix.co.nz

www.springfreetrampoline.com

www.3ds.com/atom

Figure 4. Abaqus FEA model

of cleat optimized by hand

shows the two ribs that were

regarded as most effective. This

version is now in production on

new Springfree trampolines.

Figure 3. Abaqus FEA analysis of rod ball movement within cleat socket.

"When we applied ATOM

to the same design, the

process became much

faster and effective with

automatic exploration of

the optimal geometry. I was

very surprised to see that

ATOM helped spot features

I didnt think of."

Guido Quesada, CAE Design Engineer,

Matrix Applied Computing

8/10/2019 simulia-scn-1311

12/24


Cover Story

The recent revival of the U.S. automotiveindustry is a success story on many levels.Bridge loans granted by the government

enabled some automakers to restructure.Tightened fuel-economy and pollutionstandards spurred new R&D. An improvingeconomic climate released pent-up demandfor new carsand this time Detroit wasready with fuel-efficient, eye-catching modelsthat brought buyers back to the showrooms.

The Big Three (Ford, GM and Chrysler) wereon the path to recovery.

As the automakers business picked upagain, so did that of their suppliers. DanaHolding Corporationa U.S.-based, Tier 1global supplier of axles, driveshafts, sealingand thermal-management products, off-

highway transmissions, and service partssaw a strong turnaround in revenues andmargins. While their customers recovery

was certainly a major contributor to theseresults, credit for Danas rebound also goesto an evolution in mindset that helped keepthe company on track through tough times.

We underwent a cultural change at Danafrom a mainly cost/manufacturing-drivencompany to an engineering-driven one,says Frank Popielas, senior manager ofadvanced engineering in the Dana Power

Technologies Group and head of CAE(computer-aided engineering) for Dana. Thismeant a shift in focus, from how to controlcosts and manufacture efficiently, to howto innovate. Obviously all these need to

be integrated. But if you focus solely oncosts, product quality will go down. As anengineering-driven company, we look at

how to improve a product from a quality andfunction perspective. Innovation, supportedby the right engineering tools, made ourcompany more competitive.

Many automotive suppliers struggled duringthe recession, and unemployment rates rose.However, since Dana had already developedsubstantial in-house CAE and high-performance computing (HPC) resources,the company made a point of retainingtheir design engineering teams. During thedownturn, we kept our focus on CAE, saysPopielas. We knew that, in the long run, theinvestment would be worth it.

Founded in 1904, Ohio-based Dana now has 24,000

employees in 26 countries. The companys global

brands include Spicer drivetrain products, Victor

Reinz sealing products and Long thermal products.

Danas Engineers Team Up to Accelerate Product

Design with Simulation Lifecycle ManagementMajor automotive supplier realizes up to 25 percent time-savings in their

computer-aided engineering projects

8/10/2019 simulia-scn-1311

13/24


Engineering return-on-investmentwith CAEIndeed, CAE has proved invaluable to thecompany. Danas products include a vastrange of gaskets (cylinder head, exhaust,

intake manifold, etc.), cam covers, andheat exchangers; mechanical and electricalcomponents; driveline components andassemblies (axles, driveshafts, gear boxes,etc.) and more. Materials range frommetals to rubber to plastics to fiber-based.Production processes include casting,injection molding, heat treatment, forming,magnetic pulse welding, etc.

Because we make pretty much everythingaround the automotive powertrain anddrivetrain, our portfolio involves hugecomplexity, says Popielas. To ensure

quality when designing our products, weneed to look at everything that can impactthem, including stress, strain, fatigue,molding, gas, oil and cooling flow, air andoil separation, thermal distribution and, ofcourse, their complex interactions. CAE isthe toolkit that supports the developmentof our products in the engineering space.Simulation enables us to verify and validatevirtuallyproduct functionality.

While real-world testing remains the ultimateproof of that functionality, Danas extensiveuse of CAE has enabled the company todo less and less physical testing in recent

years. Simulation speeds up the productdevelopment process, captures datathat can be used to optimize the productand gives our engineers more freedomto innovate, says Popielas. Innovationis critical for us, but it still has to be cost-effective. Our CAE resources help minimize,or even neutralize, many time-consumingtasks of the past, such as creating drawings,prototyping, and going through extensivephysical testing for each design iteration.CAE takes out costs across the board forus.

Danas CAE arsenal is extensive (seeFigure 1). Among the tools are Abaqus,the companys longtime finite elementanalysis (FEA) solver for realistic simulation;Hypermesh, Abaqus/CAE and Simlab forpreprocessing; StarCCM+ and FlowVisionfor computational fluid dynamics (CFD);MoldFlow for molding simulation; and fe-safefor fatigue. Isight is used for optimizationtasks such as Design of Experiments (DOE).

While continuing individual componentanalysis, Dana has also made the step intosimulating subsystems, complete systems,and global models. As the company

transitions to full systems engineering in thevirtual world, we expect to add even moresoftware codes in the future, says Popielas.

Helping CAE tools work together withSimulation Lifecycle ManagementAs the simulation process at Dana matured,the challenge for the engineering groupshifted from how to accurately predict real

product performance, to other pressingissues: How to more effectively connectsimulation with the rest of the business anddecision-making processes. How to improvecollaboration with both the customer andamong Danas global engineering resources.How to improve the management of thegrowing volume of data generated by thesimulation process, the approaches and theIP created by the CAE analysts.

To address these needs Dana turnedto Dassault Systmes and the SIMULIASimulation Lifecycle Management (SLM)solution. Dana engaged in an in-depth

evaluation of SLM tomeasure capabilityto their specificneeds, and begandeploying SIMULIAs

SLM in the summerof 2012. SLMenables a companyto define andmanage simulationmethods, modelsand procedures(scenarios). Whenyou get into virtualengineering, in orderto not waste yourinvestment you haveto have a tool inplace that manages

data, process anddevelopment, saysPopielas.

Historically, CAE at Dana was the purviewof individual experts who would select fromamong multiple software tools to performeverything from design data preparationto simulation execution to results analysis,storing both inputs and solutions mostlyon their local hard drives. This madecollaboration difficult, and coordinatinglarger projects a major challenge.

Communication about design changes was

also an issue, with the experts sometimesrunning simulations on outdated data files.The situation was further complicated byDanas growth strategy of dispersing teamsglobally in order to keep closer contact withtheir geographically diverse customers.

An individual product validation can involvethousands of gigabytes of information overtime, says Popielas. When you generateas much data as we do, especially in theform of such complex simulations, youneed to be able to assess what you have,know where it came from, and track it overtime. SLM is enabling us to do all of that.

As they began deploying SLM, Dana tooka bottom-up approach that helped theprocess evolve logically. This was not atop-down, enterprise-wide deployment,says Popielas. We considered all thedifferent processes we wanted to connectvia SLM and looked at where it madesense to start.

An important aspect of the implementationprocess has been the creation of guidedtemplates within SLM that capture bestpractices and standard method for

Figure 1. Sample set showing the variety of different CAE software programs

Dana uses for simulation during the design and development of its products.

The companys deployment of SIMULIA SLM provides an open environment

supporting multiple technologies so they can use applications that best supporttheir needs, as well as collaborate with internal teams, customers and their

own suppliers.

When you generate as much

data as we do, you need to

be able to assess what you

have, know where it came

from, and track it over time.SLM is enabling us to do all

of that.

Frank Popielas, senior manager of

advanced engineering, Dana Power

Technologies Group and head of CAE

Continued

8/10/2019 simulia-scn-1311

14/24


simulation work on core products. A nextstep Dana is working on is the creationof fully automated templates. Thesestandardized templates are in the frontend, with data loaded by non-CAEengineers and simulations running in thebackground, says Popielas. Our usersdont need to have knowledge of thespecifics of the CAE software they areusing, just access to a menu of provenshortcuts that help them accomplish theirgoals. If additional information is neededduring the course of a product validation,they can use SLM to quickly locateinput from our physical test labs. Thisautomation enables our future-vision for fullvirtualization of engineering as it is beingimplemented in all areas.

As deployment proceeds throughout Dana,Popielas says, SLM is helping us manageeverything so much more effectively. Wecan easily store data and find it again,literally saving weeks of searching. In thepast we wasted a lot of time hunting for

information. SLM is not just archiving, itsan environment where you always haveworking access to all the information youneed.

Openness and live visualization arekey to collaborationIn the future, SLMs open platform willalso allow the Dana team to shareand exchange relevant dataamongthemselves, with their automotivecustomers, and with their own suppliers.

Our user base is very diversified, saysPopielas. We need to be able to work with

any software tool out there. Within SLM weeasily generate connectors through whichwe can line up to the different softwarepackages, which can be CAE tools, aPDF, or even an internally written script,anything with an executable.

SLM process management capabilitiesmake such teamwork smoother byproviding consistency and repeatability.

Were seeing 20-25 percent time savingsover our previous methodology, Popielassays. We can more readily identify thesweet spot for cost-versus-performancethat generates profitability.

For More Informationwww.dana.com

www.3ds.com/SLM

As all this complexity runs automaticallybehind the scenes, what the Dana teamsappreciate most is the visual interfaceeveryone works from when accessing SLM(see Figure 2). The user sees an onscreenturntable, on which sit 3D representations

of the various jobs in progress on any givenpart or assembly. By clicking on a particularimage, the user can identify and navigate toall relevant simulations being worked on bywhoever is involved in that particular aspectof the products development. As the userperforms whatever tasks are required thatday, all changes (and their history) areupdated automatically, and are accessibleto everyone authorized to access theproject. Remote teams can all be on the(user secured) Live Simulation Review inorder to collaborate, in real time, on modeldevelopment.

This is where I think SLM is the leadingtechnology on the market, saysPopielas. The methodology goes handin hand with what I like to call iCAE.

The understanding of the importance ofcollaborative computer-aided engineeringis deepening as the use of simulationbecomes increasingly widespreadthroughout industry. If you want to drawa lot of different disciplines into theengineering space together, you have toprovide a virtual environment that is visuallyintuitive. The ability to collaborate this wayis a key reason why we went with SLM.

The future potential of iCAEFull implementation of SLM at Dana isexpected to be complete in the near future,and the companys engineers are alreadyanticipating how the software will enablethem to further their exploration of the

potential of iCAE.

Upcoming enhancements include deployingtemplates into engineering spheres suchas manufacturing. We want to integratemanufacturing steps into simulation, saysPopielas. Understanding the physics ofthe production process will further improveproduct quality, as will optimizing the layoutof manufacturing stations. The remotevisualization offered by cloud computingis still on the horizon, notes Popielas, butmany benefits of iCAEare already available.

When you start taking advantage of these

technologies, you have so much to gain,says Popielas. We are talking double digits,always. Its extremely cost-effective whenyou do it right.

Figure 2. Live Simulation Review screenshot of a cylinder head gasket (CHG) project shows the turntable

format in SLM (from SIMULIA, Dassault Systmes) through which users can access all aspects of a product

development project and collaborate with remote teams in real time.

"We can easily store data

and find it again, literally

saving weeks of searching."

Frank Popielas, senior manager of

advanced engineering, Dana Power

Technologies Group and head of CAE

8/10/2019 simulia-scn-1311

15/24


Alliances

Meeting structural requirements forstiffness, strength and durability canoften lead to the introduction of additionalreinforcements, adding mass and costto a design. Automotive Body in White(BIW) design is a typical example of thischallenge.

As a SIMULIA alliance partner since 2009,GRM has provided topology optimizationto many Abaqus customers through its

TruProducts family. These tools give usersan efficient and cost effective solution to

optimize structures for stiffness, stress andplastic strain. TruProducts modules include:

TruForm Topology Optimization

TruPly Composite LaminateOptimization

TruSize Shell Gauge and TopometryOptimization

TruShape Topography and FreeShapeOptimization

Robust Design Optimization (RDO) usingComputer Aided Engineering (CAE)parameter-based optimization techniqueshas become increasingly important inmeeting product performance requirementsas well as cost and time-to-marketchallenges. One of the main challenges

of using CAE parameter-based RDO isthe parameterization of the geometry. Inaddition to CAE-based geometry variationsusing mesh morphing or shape functions,leveraging CAD-based geometry variations,to accelerate design optimization, providespromising potential.

To investigate CAD-based geometricvariation, an automated process flowbecomes necessary starting with aparametric geometry update followed byall required CAE simulations and resultextraction. An example of this processwas presented by engineers from Robert

TruProducts Optimization Enables Part

Reduction in Composite Automotive

Body Design

Robust Design Optimization using CAD-

based Parametric Geometry

Case StudyFor one auto OEM, the turret design ofa new, lightweight vehicle was creatingissues for torsional stiffness and strengthrequirements. While meeting therequirements could be achieved throughthe addition of a localized reinforcement,this would have resulted in adding massand cost, compromising the objectives of

Bosch at the 7th Weimar Optimizationand Stochastics conference. In theirapplication, the parametric CAD modelwas created in CATIA, the FEA modelingwas in Abaqus, and RDO was performedusing optiSLang from Dynardo. optiSLangprovides capabilities for performingsensitivity analysis and investigating designoptimization and robustness.

To support this process, BOSCH initiatedthe development of a plugin by theSIMULIA services team in Germany. Theplug-in, named Optiqus, creates allnecessary interfacing files to automatethe CAD/CAE parameter updates, andto interface with optiSLang. Boschhas published several examples ofsuccessful applications using parametricCAD-based modeling and Abaqussimulations for product optimization andtolerance investigation using sensitivity

studies, optimization and robustnessevaluations. These capabilities of CAE-based parametric modeling and processautomation enable RDO technologiesto be implemented as standard virtualprototyping processes for optimizingproduct performance and robustness.

the lightweight body design. TruProductswas successfully applied to the turretdesign in two stages; first, to optimize thepressed form of the turret top; and secondto minimize thickness in a Sheet MoldingCompound (SMC) composite material

option.TruShape Pressed Form

Shape OptimizationConsidering the existing steel material,the form of the turret top was optimizedto minimize the peak stress under abuseloading and maximize the stiffness.

Allowing reinforcements not deeper than5mm to be added to the design, themaximum stress was reduced to 50% of itsinitial value and the stiffness was increasedby 280%.

TruPly SMC Composite

Material OptimizationFurther to the development of the steel part,the introduction of a lightweight, compositeSMC material was applied, allowing forfurther mass reduction. TruPly was used toautomatically determine the SMC materialthickness distribution to meet the strengthand stiffness requirements.

For More Informationwww.grm-consulting.co.uk

For More Informationwww.dynardo.de/fileadmin/

Material_Dynardo/dokumente/

Presentation_Power_Unit.pdf

(German language)

CAI

Start

Abaqus/CAE Abaqus/Solver

Optiqus Catia

Plugin Dynardo (ETK)

Output parameter

from Abaqus

Definition objective

& Constraints

Input parameter=

Designtable

*_run.bat

optiSLang

*.pro

*.odb

-CatProduct

-CatPart

CATIA V5

Workflow of sensitivity analysis of a power unit.
http://www.dynardo.de/fileadmin/Material_Dynardo/dokumente/Presentation_Power_Unit.pdfhttp://www.dynardo.de/fileadmin/Material_Dynardo/dokumente/Presentation_Power_Unit.pdfhttp://www.dynardo.de/fileadmin/Material_Dynardo/dokumente/Presentation_Power_Unit.pdfhttp://www.dynardo.de/fileadmin/Material_Dynardo/dokumente/Presentation_Power_Unit.pdf

8/10/2019 simulia-scn-1311

16/24


Optimization Spotlight

As the auto industry worldwide workshard to improve fuel efficiency andmeet increasingly stricter CO

2emission

standards, almost every conceivable wayto make vehicles lighter in weight is onthe digital drawing board of automotiveengineers.

Such is certainly the case at GeorgFischer Automotive AG, the Swiss-based manufacturer and developer of

high-performance casting componentsfor the industry. As a key supplier thatcasts more than 600,000 metric tons ofiron, aluminum, and magnesium into 100million components yearly, the companyis addressing weight through advances indesigns, materials, and processes.

The benefits of such lightweightingstrategies are clear and align with thecompanys sustainable mobility mission.

A U.S. Department of Energy report saysthat reducing vehicle weight by 10 percentcan increase fuel economy between six toeight percent.

Designing the weight outFor GF Automotive, an essential part of thedevelopment of chassis components is theuse of modern simulation and optimizationsoftware methods. Engineers at the firmhave employed these computer-aidedengineering (CAE) tools for years, achievingsignificant savings in product developmenttime and upgrades in quality. The company

currently relies on Tosca Structure, with itstopology and shape optimization modules, forreducing component weight while ensuringthe product conforms to all required stressloads.

Reducing component weight requires aholistic approach to product developmentthat combines engineering expertise withcomputer-based methods. Tosca Structurefrom FE-DESIGN (which has been part ofDassault Systmes SIMULIA since May, 2013)is a flexible, modular software system fornon-parametric structural optimization. Itsmethods include topology, shape, and beadoptimization that couple with industry standardfinite element analysis solvers. GF Automotiveuses the software on a wide variety of vehicledesign projects in an effort to save weight.

Designing for a Lighter FutureGeorg Fischer Automotive uses Tosca Structure

optimization software to design lightweight

components for improved fuel efficiency

A lighter steering knuckle, redesigned by

Georg Fischer Automotive using Tosca

Structure software for optimized structure

and weight, is in large-series production in

the Audi A3 and other vehicles.

Figure 1. GF Automotive optimized steering knuckle (left) and location in wheel suspension system (right).

8/10/2019 simulia-scn-1311

17/24


Optimizing a steering knuckle:As light as possibleyet safeIn an automotive suspension system, thesteering knuckle serves to attach the wheel

to other suspension components (Figure1). GF Automotive engineerslooking toeliminate weight from this component,while keeping the structure strong andsafeperformed an optimization exerciseusing Tosca Structure software.

Through the use of Toscawe get designsbest adapted to loading, for lightweightand notch stress-free components in i ronand light metal castings," said RomanBrauner, former project engineer in productdevelopment at GF Automotive.

The essential starting point for a successful

high-volume (or series) design is to firstdefine the appropriate design space.For the steering knuckle this includedconsideration of the axle, the wheel, andthe kinematics of the suspension assembly.

The components product specificationswere then transformed into mechanicalrequirements for the analysis.

Optimization to reduce weight began withthe Tosca Structure topology module. Theanalysis created a lightweight, stiff andstrong design concept considering allrelevant operating load and load-abuse

cases simultaneouslywhich is especiallyimportant for a dynamically stressedcomponent such as a steering knuckle.

The softwares shape optimization modulewas next used to eliminate an additional100 grams of weight through modificationsto the parts surface geometry.

Considering manufacturability andmaterial choices for mass productionFor a supplier that relies heavily on castingsuch as GF Automotive, topology andshape optimization are perfectly suited totheir manufacturing methods because ofthe inherent design freedom of the process.

Castings flexibility allows sophisticatedoptimized lightweight designs to be mass-produced cost effectively and with highquality.

The benefits of the optimization process are

many. For one, the new design concept canbe easily transferred from Tosca Structure toa CAD model with little added interpretationeffort.

Tosca Structures fully automatedoptimization process also provides resultsquickly. In the case of the steering knuckle,the software enabled engineers to easilyarrive at a manufacturable design byconsidering critical manufacturing processesand stepssuch as de-molding directionsand clamping points for mechanicalfinishingright in the optimization setup.

Value can be further added by selectivelycombining the improved design with theright material and a revised manufacturingprocess. GF Automotive engineered anew type of cast iron, SiboDur, developedin-house. This material is characterizedby better properties at break, tensilestrength*, and fatigue strength**. Giventhese properties, the lighter optimizeddesign geometry complies with all safetyrequirements for the part.

In addition to aluminum, innovative cast-iron materials are the trend, said Brauner.

In each case, the aim is to use only the

minimum amount of material required forthe safe function of the component.

32% lighter component yieldsreduced CO

2

emissions and winsan awardFollowing optimization results (Figure 2), thesteering knuckle, which initially weighed4.39 kg (9.7 lbs), is now 2.98 kg (6.6lbs)a 32 percent savings. Since eachchassis requires two components, the totalreduction is 2.82 kg (6.2 lbs) per vehicle.When looking at projected installation in 1.6million vehicles, weight savings from thiscomponent alone would cut an estimated11,600 tonnes (12,787 tons) of CO

2

emissions over the life of the part.

The new lighter steering knuckle has been

implemented in large-series productionsince 2012. Its being used by VW in theirMQB platform, e.g. in the Golf VII and inthe Audi A3, among others. The design hasalso received the prestigious environmentalkoGlobe, awarding pioneering innovationsfor sustainable mobility, in the categoryof raw materials, materials, and processoptimization.

"Tosca Structure allows us to developour products efficiently and thus fulfillsthe steadily increasing demands of ourcustomers for consistent lightweight

designs, said Brauner.*Tensile strength: The resistance of a material toa force tending to tear it apart, measured as the

maximum tension the material can withstand

without tearing.

**Fatigue strength: The maximum stress a material can

endure for a given number of stress cycles without

breaking.

For More Informationwww.georgfischer.com

www.3ds.com/tosca

Figure 2. Toscaoptimization of a steering knuckle. Optimized design with 32% weight reduction.

32% WEIGHT

REDUCTION

32%

5.0

4.0

3.0

2.0

1.0

0.0

Weight [kg]

Initial design

Optimized design

"Tosca Structure allows us

to develop our products

efficiently and thus fulfills

the demands for consistent

lightweight designs.Roman Brauner, former project

engineer in product development

at GF Automotive

8/10/2019 simulia-scn-1311

18/24


Case Study

On the evening of April 20, 2010,Deepwater Horizon suffered a blowout

while drilling in the Macondo Prospect,an area in the Gulf of Mexico 40 milesoff the southeast coast of Louisiana. Theplatform caught fire; two days later, it sank.Numerous attempts were made to sealthe well, but oil continued to spew intothe Gulf until July 15, when a temporarycap was put in place. Relief wells thenpumped concrete into the area underneaththe wellhead, and the well was deemedpermanently sealed in mid-September.

Investigating the causeof the blowoutDuring the initial event and in the weeksafter Deepwater Horizon sank, numerousattempts were made to activate theblowout prevent (BOP) stackinitiationof the emergency disconnect sequencefrom the failing rig, automated dead-mancircuitry on the BOP stack, even remotelyoperated vehicles working directly on thestacks autoshear function, one miledown. Nothing stopped the flow of oil.

Even while the oil industry was working toshut down the well and contain the spill,

the Departments of Interior and HomelandSecurity signed an order to begin aninvestigation. On September 1, DETNORSKE VERITAS (U.S.A.), Inc., Columbus,Ohio office was awarded a contract todetermine the performance and possiblefailure modes of the BOP stack. DNV pulled

together an expert team of 40 individuals inforensic investigation, materials specialists,BOP operation, systems controls, andcomputer modeling from its Columbus,Ohio, Houston, Texas, and Hvik, Norwayoffices to address all issues involved in thismultidisciplinary investigation.

Finite element analysisas a forensics toolShane Finneran, project engineer and teamlead in the Computer Aided Engineering(CAE) Group at DNVs Materials andCorrosion Technology Center, Dublin,Ohio, was a key player in the investigation.Finnerans group faced the problem ofquickly constructing and testing numerouscomputer models of the stacks mechanicalcomponents. They turned to Abaqus FEAsoftware from SIMULIA, part of DNVsinvestigative toolkit for over a decade.

FEA provided DNV with a rapid, accuratemethodology to simulate and evaluate thelikelihood of proposed scenarios, manyof which would have been impractical toassess with physical testing.

It would be extremely difficult and generallycost prohibitive to run physical tests underthe same conditions that exist two milesunder the ocean, re-enacting a blowoutscenario, says Finneran. FEA canprovide the means to perform extensivesimulations of many types of damageranging from simplified deformation andbuckling to post-buckling deformation and

shearingwith realistic pressure and forcemeasurements built into those simulations.

A 200-page report published by DNV(available at http://www.bsee.gov/uploadedFiles/DNVReportVolI.pdf) explainsin great detail the reconstruction processand test methodology used, as well as theirfindings.

1. The BOP stack, together with severalpieces of drill pipe, was raised from thewell site and transferred to a holdingfacility at NASA-Michoud.

2. The team cleaned, photographed, and

cataloged the stack, disassemblingwhere necessary to get at its innerworkings.

3. Hydraulic fluids and metal sampleswere taken for analysis, the stackscontrol mechanisms and theirbatteries, actuators, and solenoidswere tested, and the casing and blindshear rams, variable bore rams, andupper and lower annular preventerswere all visually inspected and 3D laserscanned.

4. DNV then used the laser scans toconstruct as-is 3D CAD models of the

damaged equipment, especially in thearea of the BSR, the blind shear ram,which was the only ram on the BOPdesigned to cut the drill pipe and sealthe well bore.

5. SolidWorks software, also fromDassault Systmes, was used toconvert the original CAD files of the ramcomponents into simplified surfacesfor use with Abaqus FEA, enabling theteam to simulate and virtually test theirtheories of what happened.

What went wrong

During normal operations, the huge BOPstack sits on the sea floor directly abovethe wellhead. Its two halves together areas tall as a six-story building and weighover 400 tons. The top half, called a LowerMarine Riser Package (LMRP) contains twoelectronic control pods together with a pairof donut-shaped sealing devices called theUpper and Lower annular preventers. Thepreventers seal around the drill pipe andregulate upward fluid flow.

The lower half of the BOP stack containsa series of shear and bore rams, and thevalves and hydraulics that actuate them.

FEA Aids Deepwater Horizon Failure ForensicsDNV uses Abaqus finite element analysis to study blowout scenarios

The Transocean drilling rig Development

Driller II in the Gulf of Mexico May 20, 2010.

Courtesy of the U.S. Coast Guard.

8/10/2019 simulia-scn-1311

19/24


For More Informationwww.dnv.com

www.3ds.com/abaqus

Depending on the circumstances, theserams serve to choke the well, or kill it ifnecessary. The BOP stack is a complexdevice, and its job is critical: controlpressure within the well, keep the drill pipecentered, and sever and seal the pipe in

the event of an uncontrollable blowout. TheDeepwater Horizons BOP stack failed onall three counts.

On the night of the accident, shortly after aroutine leak-off test of the Upper AnnularPreventer, manifold pressures on the drillingplatform increased five-fold. Oil and gasthen began spilling onto the platform deck,and the rig caught fire. Personnel aboardthe Deepwater Horizon activated theEmergency Disconnect Sequence (EDS) inan attempt to shut down the well, but theywere unsuccessful. The rig was abandoned

a half an hour later.How it happenedDNVs analysis, supported by FEA data,indicates that the primary cause of failurewas the BSR, or blind shear rams inabilityto cut the drill pipe and seal the wellbore.

The BSR is like a big pair of scissors, withopposing blades designed to cut the drillpipe and seal its end in a last-ditch effortto kill the well. FEA clearly showed thatthe shear ram should have been able todo its job. The required force was in themiddle of the test and calculated datarange for a nominal, centered pipe model,

and there was nothing out of the ordinarywith the strength of the drill pipe itself. TheBSR would likely have worked, except forone thing: the pipe wasnt where it wassupposed to be.

When activation of the blind shear ramsoccurred during ROV intervention, theannular preventers were already sealedaround the drill pipe at the top of the BOPstack. The variable bore rams were sealedaround its lower half. The pipe should havebeen in the center of the well bore, whichis where the blind shear ram expected it to

be. However, physical evidence from themarkings on the recovered portion of thedrill pipe indicated that the pipe was at theside of the wellbore instead.

The DNV engineers postulated a reason:One of the pipes tool jo ints was positionedjust below the upper annular preventer,which was closed on the drill pipe. Whenupward pressure became high enough dueto the forces caused by the blowout, thepipe had nowhere to go and the upwardforces (compression forces on the drill pipe)caused the drill pipe to bucklepushingthe pipe to the side of the wellbore. The

blind shear ram tried to cut the pipeoff-center, unsuccessfully. This jammed aportion of the drill pipe between the ramblock faces, preventing the rams from fullyclosing and sealing the wellbore.

FEA supports theories of whyAs DNVs investigation into the proposedsequence of events deepened, FEAcontinued to support the teams theories.

They started with a buckling analysis todetermine what the necessary load wouldhave been to move the pipe to the edge ofthe casing. The FEA models indicated thatthe applied pressures would have beensufficient to force it out of position.

From there, it was necessary to determinetwo other things: would the pipe havesheared had it been in the correct position,

and what forces were needed to shearit off-center? Once again, FEA modelsprovided the needed insight (see Figure1). First, using a combination of the sheardamage parameters, in conjunction withthe known elastic and strength values ofthe material, our FEA data validated theestimated pressures needed to shear thepipe under a center-load condition, saysFinneran.

Next, the engineers simulated the situationseen on the Deepwater. Our shearinganalysis showed that, with the pipeoff-center, the rams would have been

prevented from closing fully, says Finneran.This allowed flow past the blind shear ramsand caused significant erosion of the blindshear ram components and wellbore.

We determined that, under blowoutconditions of this well, it was possible forthe drill pipe to buckle between the UA(Upper Annular Preventer) and U-VBR(Upper Variable Bore Ram), says Finneran.

The DNV hypothesis was supported by theiroff-center BSR shearing FEA model, whichprovided results consistent with thosedamaged pipe segments, as well as theerosion damage seen between the rams,wellbore, and recovered drill pipe.

Moving ForwardNow that FEA has been shown to be highlyuseful in helping identify the causes of theDeepwater Horizon failure, Finneran would

like to see it used to lessen the likelihoodof such events re-occurring in the future. Ibelieve finite element analysis can be avaluable tool in validating the capabilities ofboth new designs and existing equipment,he says. It allows you to incorporate anumber of scenarios and situations thatare difficult, if not impossible, to analyzethrough experimental testing.

Finneran points out one challenge still to beovercome: incomplete raw material data.Because some of the material damageparameters of the actual materials used in

Deepwaters BOP stack were unavailable,he and his team had to substituteequivalent materials with known values intheir analyses.

The industry as a whole would benefitfrom developing material-specific data thatcould help them to further fine-tune theFEA of their designs, he says. This couldenable the development of models that canverify the ability of existing and future BSRsto shear drill pipe under high-pressureblowout conditions. Given the appropriatematerial data, this kind of numericalmodeling definitely has the ability to assess

the extreme pressures and conditions thatmay be seen during a blowout event.

Whatever the outcome of this testing,regulations are expected to change, saysFinneran. Several alternatives are beingconsidered within the industry, such asincluding redundant BSRs as a secondarybackup, and pursuing more robust designscapable of cutting and sealing under mostany situation presented.

Figure 1. Abaqus FEA models showing the

interaction between the drill pipe and the blind shear

ram (BSR). Top image shows the complete shearing

of the pipe when properly centered. Bottom image

shows what happened when the blade did not fully

engage the pipe because the pipe was not centered

in the wellbore. The lateral forces due to buckling

below the BSR pushed the pipe off center and

caused the closing rams to deform the pipe rather

than shearing it.

8/10/2019 simulia-scn-1311

20/24


Academic Update

Sun and Simulation Power-up Solar-Car PerformanceUniversity of Toronto Blue Sky team uses Abaqus FEA to ensure

robust design for their World Solar Challengerace vehicle

One week across the Australian Outbackin a sleek but cramped vehicle poweredonly by the Sun: Would you be up to it?

The Blue Sky team from the Universityof Toronto was: they finished 8thin theBridgestone World Solar Challenge (WSC), aunique race founded in 1987 that now takesplace every other year. This years event,held October 6-13, included 40 teams from

22 countries. As usual, the competitionstarted in Darwin on Australias north coastand ended 3,021 kilometers (1,877 miles)later in the southern coastal city of Adelaide,splitting the island continent approximatelyin half.

The student-built, solar-powered vehiclesentered in the contest were strangely-shaped, sleekly aerodynamic, and cutting-edge: arguably the most efficient electricvehicles on the planet, according to therace website. Averaging speeds of between60-90 kilometers-per-hour (37-56 mph),cars completed the course in 33-48 hours,

with driving limited to daytime (from 8 a.m.to 5 p.m.).

Success in this noteworthy event dependedon the relationship of power to drag. Inpursuit of the optimal ratio, teams worked tofind a winning combination of aerodynamicprofile, lightweight composites, photovoltaicadvances, vehicle dynamics, and more.

The team and the carIn the summer of 2011, the 50 studentsof the Blue Sky team began design on theschools seventh-generation solar car (B-7)and their entry in the 2013 WSC. They had

a wide range of backgrounds and abilities:from first-year through graduate level andfrom virtually every engineering discipline.

We dont discriminate among those whocome knocking at our door, says AhthavanSureshkumar, chief engineer for the team.Were looking for people who are willing tocommit their time to get the job done.

According to Paul Park, the teams

managing director, Everyone is somewhatclueless when they join. But they all comeout at the end with really strong technicalknowledge.

Specifications for this years most popularChallenger class were many: for the firsttime, four wheels were mandatory; therewere also stringent stipulations about drivervision combined with upright seating.

These new regulations posed a very difficultdesign challenge for us, mainly in the areaof aerodynamics and driver packaging,says Blue Sky chief advancement officer

Tiffany Hu. In recent years the World SolarChallengehas been shifting towards morepractical designs, which is very exciting.

Abaqus opens virtual windowinto performance

Aventec, Canadas leading provider ofDassault Systmes (DS) 3DEXPERIENCEtechnology portfolio in nearby Markham,Ontario, recommended that Blue Skyuse DS SIMULIAs Abaqus finite elementanalysis (FEA) for realistic simulation andCATIA for CAD-based virtual productdesign and collaboration.

This was the first generation car on whichwe used the Dassault Systmes platform.Not only did it help us collaborate betterand improve work flow tremendously, itwas critical to us being able to designand deliver the vehicle so quickly, saysSureshkumar.

According to Park, CATIA allowed us tofully integrate the systems in the vehicleand gave us the flexibility to design a world-class aerobody, while SIMULIAs Abaqushighlighted potential problems early in thedesign process and helped us avoid over-

design.

Abaqus FEA was employed to predicthow evolving design iterations for specificcomponents and systems would behaveunder real-world conditions. The team,which had never used realistic simulation,focused their analyses on optimizingthe composite aerobody and chassisstructures for weight and strength.

In the previous car, student engineers hadrelied solely on simple beam calculations(torsional and bending stiffness) to determinehow their composite monocoque chassis

design would perform. From there theyapproximated both the number of compositelayers and whether the structure couldwithstand specific loading.

This imprecise method ignored certainfactors, though, such as stiffnesscontributed by the roll cage and theaerobody itself. The result was anoverdesigned and heavier vehicle: everylayer of carbon fiber contributed anextra 10 kg, a significant increase whentarget weight was approximately 200 kg(approximately 440 lbs). Without FEA, we

had difficulty identifying how the chassiswould actually behave. So we played itsafe, says Park.

To start the analysis, CATIA models wereimported into Abaqus. There were surfacemodels for both the lower aerobody,containing the chassis, and the upperaerobody, which encompassed its ownsubstructure. Aerobody components areoften analyzed individually in automotiveapplications because they are designedto separate during a collision. There werealso solid models for both the roll cage(including front and rear roll bars) and the

8/10/2019 simulia-scn-1311

21/24


For More Informationwww.blueskysolar.utoronto.ca

www.3ds.com/abaqus

suspension reinforcement bars (connectingboth left and right suspensions).

All models were meshed in Abaqus. For theaerobodies, Abaqus automated this processaround the complex curves and holes thatare an integral part of their surfaces. For thesolid models, a Python script in the softwarewas used to define mesh elements andmake updating easier as the design evolved.

Tie constraints were next defined in the loweraerobody model between the mesh andsurfaces of the chassis sandwich panels,which were part of the model (see Figure 1).

Python scripting capabilities in Abaqus alsohelped automate the previously repetitiveand time-consuming task of manuallyapplying loads to the models. Each loadcase involved calculating, further defining,and then reapplying reaction forces at fivesuspension clevises (connection points).Now parameters could be changed, nodesets redefined, and loads reapplied in amatter of seconds. This method made thedetermination of contact regions much moreaccurate than was previously possible (seeFigure 2).

Other features in Abaqus further enhancedaccuracy and efficiency. Establishedlayup definitions in the software provided

a convenient way to try out varying plyschedules for the chassis and to fine-tune stiffness and strength for the upperand lower aerobodys surface layups andsandwich-board construction. With the aidof FEA, the team was able to minimize B-7sweight, using only half as much carbonfiber as in the previous car. Abaqus helpedus determine how much stress our lighterdesign could take, says Park. It highlightedpotential problems early in the designprocess.

While simulation of varying ply scheduleshelped with weight-minimization, Abaqus

also provided valuable mass estimates thatthe team used in controlling the vehiclescenter of gravity. The team further determined

through theory, testing, and calculations thatweight, aerobody shape, tires, and wheelbearings were the most critical factors whenlooking to reduce drag and enhance vehiclespeed.

FEA also provided design insights that wouldhave been missed in prior years. Looking atloading, for example, the student engineerslearned that a .75G braking scenario placedas much stress on the car as a 6G bump(see Figure 3). Equally surprising to the teamwas the fact that the front and rear roll barscontributed little to the chassis torsional

rigidity, while the inherent stiffness of theaerobody (especially the driver fairing) playeda major role.

It was far beyond our capabilities to considerall of these various structures together without

Abaqus, says Sureshkumar. The softwareallowed a much more integrated analysis andmodeled external loads extremely accurately.

Abaqus gave us confidence in the complexdesign decisions and tradeoffs we made.

The sun and simulation: a winningcombination

As race day dawned in Darwin, designdecisions had all been made, compositelayups completed, decals applied, test driveslogged. After countless all-nighters, thegleaming B-7 stood poised, reflecting thespring sun, ready for the starting gun. It was aremarkable achievement for a volunteer group

of students, 19 of whom had been chosen totravel to Australia to support the car.

How the race was run depended on morethan just sunny skies. Theres a remarkableamount of strategy, says Park. Tacticsdepend not only on the weather, but onroad conditions, other teams that impedeyour speed and, yes, even the occasionalkangaroo crossing your path. We go as fastas possible but take every precaution to makesure that two years work doesnt break downin the middle of the Outback.

While more than half of the entries in the

competitive Challengerclass did break downduring the race, B-7 didnt, placing secondamong North American entries and achievingthe Blue Sky teams goal of a spot in the Top10. The B-7 completed the 3,021 km/ coursein 45 hours, 38 minutes with an averagespeed of 65.7 km/hr.

The race was the climax of an incredible two-year engineering journey, notes Sureshkumar.

Figure 1. This screen shot from Abaqus of the lower aerobody model (including

the chassis) identifies where structural components are constrained to the chassis

surfaces (orange and pink). Point masses are also constrained to the chassis

(purple lines).

Figure 2. Loads were applied to the suspension hard points in the lower aerobody

model (yellow dots) as indicated in this screen shot from Abaqus.

Figure 3. These screen shots (side view of the lower

aerobody chassis) show the resulting deformation

from a 6G front bump of the sandwich panel chassis

without the aerobody skin (top) and with it (bottom),

using the same deformation upper bound. This

illustrates how the rigidity of the aerobody coupled

with the sandwich panel results in an order of

magnitude decrease in deformation.

8/10/2019 simulia-scn-1311

22/24


Tips & Tricks

For More InformationLogin to view video in the

SIMULIA Learning Community:

https://swym.3ds.com/#post:19810

Material calibration is the process of deriving material behaviors like elasticity

and plasticity from sets of material test data. The material behavior ischaracterized by the value of the material constants and the underlyingmathematical equations. In a simple workflow shown here (see Figure 1), materialcalibration would mean finding out the values of the constants A, mand nthatwould match the curves made from the experimental data.

Isight, with its ability to drive simulations, data comparisons and seamlesslyencapulate workflows for optimization, is indeed well suited for material calibration(see Figure 2).

Using Isight and Abaqus for a Simple

Material Calibration Workflow

Heres What You Do:1. Start with the default model in Isight Design Gateway.

2. Drag and drop the Abaqus Component into the workflow.

3. Configure the Abaqus component to parse the input file, depicting thematerial test and choose the material parameters to be varied.

4. Likewise configure the Abaqus component to parse the corresponding odbfile and choose the relevant output parameters to compare with the materialtest data (e.g. streses and strains).

5. Drag and drop Data Matching component.

6. Configure Data Matching component to read-in the material test dataas Target data set.

7. Map the outputs from Abaqus component to Data Matching componentand read them in as Simulation data set.

8. Choose some appropriate data comparison metric(s) (e.g. Absolute AreaBetween Target and Simulation Data) as parameter(s).

9. Replace the process component to be Optimization component.

10. Choose an appropriate optimizer in Optimization component, and configurethe material parameters to be design variables and the data comparisonmetrics to be objective(s).

11. Run the workflow.

ImagecourtesyofAxe

lproducts.

Optimization

The Brain

The SimulatorThe Comparator

Data MatchingAbaqus

Figure 2

Figure 1

10 -60

1

2nominalstress(MPa)

3

4

5

6

7

8Strain 10%

Strain 50%Strain 100%

10 -4 10 -2 10 0 10 2 10 4

Documents

simulia-scn-1311