MSC Software Magazine, Volume X - Winter 2019 by MSC Software

Sustainability
Makes Good Business Sense
Panasonic GM US Army Samsung

Achieve Excellent Virtual Autonomous Metals AM Co-Simulation of a
Indoor Air Circulation Driving & Active Safety Data Management Washing Machine
2020 VISION
Smart Multiphysics Focused CFD
With the 2020 release of MSC Software’s Cradle CFD solutions, multiphysics
focused CFD is now easier than ever. Get the most accurate answers every
time with best-in-class co-simulation toolchains:
• Actran FEA coupling for high fidelity aeroacoustics

• Discrete Element Modeling co-simulation for accurate predictions
• Photorealistic postprocessing to visualize accurate multiphysics
• New turbomachinery vertical application enhances user productivity
• New multiphase cavitation models
Request a Quote
www.mscsoftware.com/cfd
2 | Engineering Reality Magazine

ED I TORIA L solutions that will help solve our our suburban future, with concurrent
sustainability challenges this century. Our developments in smarter technologies
Vice President of Marketing, Keith Hanna, and electrification offering reduced
articulates how CAE will be in the vanguard emissions and noise pollution. As the
Design & of delivering practical engineering solutions

to future sustainable development in his
industry races to make safe autonomous
vehicles reality, I encourage you to
Engineering commentary on page 14. As Ola pointed

out in Las Vegas, this is a real business
read this edition’s excellent ‘Thought
Leadership’ article by Chris Kinser from
Simulation for opportunity for all industries, as well as a

way to resolve some of the biggest and
General Motors in Detroit (page 9) that
describes how our VTD product line
Sustainability most complex problems facing humanity

today.
is central to GM’s virtual test mileage
requirements for safe and efficient self-
driving cars.
At HxGN LIVE, I shared our vision of a
smarter design ecosystem where better This magazine sees several excellent
decisions can be made at the point cutting-edge co-simulation stories from
of design. It is a vision we are making Samsung in Korea concerning washing
real through our unique portfolio of machine design (page 50), Volkswagen
simulation solutions that improve any in Germany for sub-assembly duplication
CAE user’s productivity, make it possible (page 6), and Panasonic in Japan
to enhance your product quality and optimizing fan design (page 86). I would
maximize your manufacturing efficiency. also bring to your attention the fascinating
To make these tools smarter, we have US Army additive manufacturing use of
developed new capabilities in several of MaterialCenter for data management.
our CAE products that provide you with In Aerospace & Defence, Avio also
design-for-manufacturability outputs, cost share their experience, plus we have
optimization in a common environment several automotive stories from leading
and enable product sustainability companies such as Ford, Adient, Audi,
decisions to be made at the design VW, Daimler, Hyundai and Sany in this
phase. packed magazine.
Paolo Guglielmini In this edition of Engineering Reality, Launched recently, MSC Apex Generative
CEO, MSC Software several of our customers share their Design provides unprecedented
approaches to sustainability. Coriolis from productivity benefits for our users in
India (page 44) describes work simulating additive manufacturing and a wide
I
composites with Digimat to lightweight range of other CAE design spaces.
f you haven’t already done so, I aircraft wings that improve aircraft fuel We also have excellent commentaries
would encourage you to find the efficiency and Analog Way in France on Integrated Computational Materials
time to watch Ola Rollén’s keynote (page 78) outlines how it used our Cradle Engineering (ICME), Smart Generative
presentation on YouTube entitled Computational Fluid Dynamics (CFD) Design and additive manufacturing in
“Your Data can Save the World” from software to reduce the thermal effects of this magazine from our in-house experts
our annual HxGN LIVE event in Las Vegas big video display screens in Times Square Roger Assaker, Raj Dua and Hendrik
last June. In it, Ola, Hexagon’s CEO, New York to save electricity. Schafstall respectively. Finally, do check
succinctly outlines why sustainability is out our ‘Tips & Tricks’ article showing
critical to all of us on planet Earth given Additive manufacturing will be co-simulation between MSC Nastran and
the way we are polluting our environment, a transformative technology for AVL Excite (page 84) and the exciting
depleting our natural resources, and not sustainability in the next 20 years, since Artificial Intelligence (AI) and Reduced
recycling enough of what we’ve already its mass commercialization minimizes Order Modelling (ROM) work we are doing
manufactured. Legislation and societal scrappage, enables the lightweighting of that combines our CAE product portfolio
pressures will contribute to rectifying our transportation parts and right-first-time with the machine learning capabilities of
pernicious habits, but it is science and manufacturing benefits as shown by our partner CADLM from Paris (page 27).
engineering that will ultimately underpin stories from the US Army (page 60) and
many of the real-world answers to MBFZ Toolcraft (page 82).
these problems that our children and
grandchildren will inherit from us. There are an estimated 1.3 billion road
vehicles in the world today, the majority
I believe that predictive computer-aided of which contribute to greenhouse gas
engineering (CAE) simulations will play emissions and air pollution in urban areas.
a big part in ensuring we deliver the Most analysts agree that autonomous
renewable energy and transportation vehicles will play an important part in
Volume X - Winter 2019 | mscsoftware.com | 3

14 Table of
Contents
F E ATUR E S
General Motors:
09
Christopher Kinser
General Motors Advances Virtual
Autonomous Driving & Active Safety
Keith Hanna:
14
Sustainable Engineering is Good Business Sense:
Computer-Aided Engineering Can Solve
Global Environmental Challenges
51
Samsung Electronics:
Adams Marc Co-Simulation
of a Washing Machine
60
US Army:
US Army Use of MaterialCenter for Metals
60
Additive Manufacturing Data Management
86
Panasonic Ecology Systems:
Cradle CFD Innovation that Achieves
Excellent Indoor Air Circulation
06
Volkswagen AG:
Increase Agility and Collaboration in Engineering
Development Using MSC Nastran Modules
86 23
Ford Motor Company:
Reduce Need for Physical Prototyping
Using Adams Real Time
CADLM:
27
Enabling Accurate Design Decisions
While Compressing Engineering Timelines
with CADLM Technology

30
Hendrik Schafstall and Raj Dua:
Bridging the Gap Between Design and Additive
Manufacturing Using Smart Generative Design
34
Avio:
A Smart and Connected World: Avio to Ensure Executive Editor
Spatial Launchers Structural Integrity Keith Hanna
Editor
40
Adient: Elizabeth Yeh
Car Seat Manufacturer Cuts Costs and Improves
Quality With Stamping Analysis Tool Copy Editor
Mary Harrison
Robin Wolstenholme
44
Coriolis:
Fast and Accurate Contributing Editors
Additive Manufacturability Analysis Yijun Fan, Raj Dua, Bhoomi Gadhia,
Hemanth Kolera-Gokula, Roger Assaker,
Romain Baudson, Zanlang Yin, Hendrik
47
Daimler Trucks: Schafstall, Dan Marinac, Volker Mensing,
Accelerates Product Development Cycle with Keita Fujiyama
10X Faster MSC Nastran Performance
Design
Dante Fiorini
56
NVIDIA: Elizabeth Yeh
MSC Nastran Speeds Forward
Using Graphics Processing Units
66
Roger Assaker:
10x ICME to Save Time, Cost and MSC Software Corporation
Weight in Materials Simulation 4675 MacArthur Court, Suite 900
Newport Beach, CA 92660
714.540.8900
68
AUDI: www.mscsoftware.com
Multi-Resolution Traffic Simulation for Copyright © 2019 Hexagon AB and/or its subsidiaries. All rights
Connected Car Applications using VIRES VTD reserved. Hexagon, the Hexagon logo, and other logos, product
and service names of Hexagon and its subsidiaries are trademarks
or registered trademarks of Hexagon AB and/or its subsidiaries in
the United States and/or other countries. All other trademarks
74
belong to their respective owners. Information contained herein is
SANY: subject to change without notice. Hexagon shall not be liable for
Catching Structural Fatigue of Concrete Mixer Truck errors contained herein.
at the Design Stage with Modal Synthesis Although every effort is made to ensure the accuracy and
reliability of material published MSC Software does not accept
any responsibility for the veracity of claims made by the
79
contributors or the manufacturers. All material is strictly copyright
Analog Way: and all rights reserved. No part of this magazine may be
reproduced without prior permission of MSC Software. E. & O. E.
Designing Premium Innovative Solutions
for 30 Years Have feedback? Have a story? Submit to:
engineeringreality@mscsoftware.com
82
MBFZ Toolcraft:
Flexibility Through Additive Manufacturing:
How Simulation Supports 3D Prototyping
AVL EXCITE™:
84
MSC Nastran and AVL EXCITE™ Integration
Delivers Better, Faster and More Efficient
Powertrain Simulations

ST RUCT UR E S
Volkswagen AG
Increase Agility and Collaboration
in Engineering Development
Using MSC Nastran Modules
By Dr. Bruns Dr.-Ing. Jürgen Bruns, CAE-Methods, Volkswagen AG
Bhoomi Gadhia, Product Marketing Manager, MSC Software

A
s we all know in the finite element world, traditional FEA model build-up
contains multiple parts and sub-structures which are then organized in
finite element codes (such as ‘include’ files in MSC Nastran) for easy
exchangeability. While using these include files, a strict numbering guideline
needs to be followed to avoid ID collisions between INCLUDEs.
It is required in MSC Nastran that a complete model is built-up without overlapping of ID’s.
This sort of build provides independent INCLUDE files to handle which are easy to reuse
and interchange whenever the vehicle design changes. However, with this method, it is
hard to manage these files as the users must follow an elaborate numbering scheme. This
method needs high discipline, monitoring effort as well because it needs to be adapted
regularly due to new requirements.
Volkswagen has enjoyed a long-standing relationship with MSC Software for many
years. Volkswagen not only uses the advanced technologies from MSC Software for their
analyses, they also drive the requirements for the new developments in MSC Nastran.
One of the most recent developments is “Modules” in MSC Nastran. As computational
resources are becoming cheaper, models are getting larger. These larger models require
efficient model management. With this new technology called “Modules” in MSC Nastran,
whose development was driven by Volkswagen, the model management now becomes
very accessible.
Modules are introduced in MSC Nastran as an assembly management tool; it is a

standalone Bulk Data section that typically contains a part or component of an overall
assembly. Modules are analogous to part super elements without reduction. MSC Nastran
can also create multiple instances of a primary MODULE with a process referred to as
MODULES Instantiation.
MSC Nastran
Figure 1: Traditional Model Build Up

“The modules approach helps us save time and
ensure our high-quality results. We can
use the decrease effort in model build-up,
exchange, as well as pre-post processing for
NVH, Durability, and Nonlinear Analysis.”
- Dr. Bruns, Volkswagen
modules. Post-processing results are

available in the HDF5, F06, punch,
and op2 files, which are partitioned by
module.
So, to conclude, MSC Nastran Modules

have the same characteristics as
INCLUDEs, but without any ID collisions.
Modules reduce the “numbering” work
in the model assembly processes.
Figure 2: Model Build up at Volkswagen Modules enable the automotive OEMs
to easily exchange the models with
other analysis groups (such as statics,
The key benefits of modules are that they reduce reliance on a pre-processor for creating crash, NVH). Finally, users can use
copies of, translating, rotating, or mirroring an existing primary MODULE – this also helps switch variants easily to try out different
reduce modeling time. Additionally, MSC Nastran’s MODULE uses similar definitions as designs in the vehicle development.
some other codes – this allows for quick translating of a MSC Nastran BDF for other
simulations such as crash simulation with PAM-CRASH.
Now, there is also a need to connect these different modules, for which manual, automatic
and semi-automatic methods are available for connecting modules.
Each module has its own ID numbering scheme for grid points, elements, properties and
so on. The ID numbers do not have to be unique across Modules.
“The modules approach helps us save time and ensure our high-quality results. We can
use the decrease effort in model build-up, exchange, as well as pre-post processing for
NVH, Durability, and Nonlinear Analysis,” says Dr. Bruns from Volkswagen.
A module is a standalone Bulk Data section that typically represents a component of

an overall assembly, e.g., an automobile wheel or fender. Modules allow the users to
assemble multiple Bulk Data sections to form a single MSC Nastran input file. Each
module is delimited by the BEGIN MODULE and END MODULE Bulk Data entries.
Just like in part super elements, each module may have its ID numbering scheme
Figure 3: MSC Nastran Module Overall
for grids, elements, properties, and the ID numbers do not have to be unique across Bulk Data Section

Q&A
Autonomous
Vehicle Testing
With Christopher Kinser, General Motors,
Milford, Michigan, USA
E
ngineering Reality Magazine recently interviewed Chris Kinser from General
Motors, the Director of their Global Autonomous Driving Center in Michigan
and an industry expert in rapidly emerging sector of vehicles (AVs). He has
a long history of working on electrification and autonomous driving systems,
and his team in Milford is responsible for vehicle integration of several
General Motors’ advanced technology programs, including self-driving vehicles, as
well as automated driving and active safety technologies. Chris’s expertise in software,
controls systems and vehicle performance integration have been recognized with three
Boss Kettering Awards. Chris holds a Bachelor’s in Electrical Engineering from Kettering
University and a Master’s of Engineering from Rensselaer Polytechnic Institute, USA.

What does your role in General vehicles will be electric vehicles. Not
Motors involve and what is only are electric vehicles better for the
GM’s overall approach to the environment and quieter for city traffic,
fast-emerging Self-Driving but they allow for simpler integration
opportunity? of the advanced technologies required
for the cleanest and safest operation of
I am the Director of the Global autonomous vehicles. For example, an
Autonomous Driving Center at our Milford, all-electric vehicle has a more stable power
Michigan proving ground where I manage source and a faster responding propulsion
a large engineering team. We believe that system that provide it inherent advantages
autonomous technology will play a key role over its internal combustion counterparts.
in our vision of a world of zero crashes,
zero emissions and zero congestion
through the enormous potential benefits it Why did GM choose
holds for society in the form of increased Hexagon/MSC technology
safety and access to transportation. for its Autonomous Driving
strategy?
General Motors is in a unique leadership
position when it comes to developing and We see Hexagon as a company totally
deploying self-driving vehicles in that we devoted to the autonomous sector in its
are the only company to have everything business focus. Hexagon’s combination
from design, engineering, validation, and of sensor and scanning technologies
testing all under one roof. My team is works like Leica cameras, and its simulation
closely with teams all around the country on software suite like MSC’s VTD (Virtual
developing autonomous driving solutions. Test Drive) software, fill many of the
needs of the market. VTD is in the center
of a comprehensive GM simulation
What do you see as the big environment that we have developed with
challenges to Autonomous Hardware-in-the-Loop. We use VTD in
Mobility going mainstream in conjunction with software products like
the next 10 years? CarSim and Simulink (for control systems)
in our real time virtual automated driving
We are in the middle of a fundamental shift vehicle testing environment.
in how people and goods move through the
world. Autonomous mobility will certainly
play a huge part in that and at GM, we will What is your vision for GM
be guided by the needs of our customers. in the autonomous mobility
It is also one of the most difficult challenges space in say 5 years from
for automotive engineering. The biggest now?
challenge I see to Autonomous Mobility
going mainstream is getting all the systems It is still the early days of autonomous
necessary for self-driving vehicles to work mobility and we are excited by the
together seamlessly. Next time you’re opportunities for this technology to
behind the wheel, take a moment to reflect improve the world. In terms of engineering
on all the tasks you are performing to and development, we will continue to listen
drive the vehicle. Working on developing a to our customers and deliver advanced
system that can perform those same tasks mobility solutions that meet their needs.
is the engineering challenge of our lifetime.
That’s why at GM, we believe that a safe
self-driving vehicle should be built from the Which country or countries
ground up with seamless integration of the do you think will go fully
self-driving system. autonomous with cars first in
your opinion?
Will all autonomous cars be I can’t speak to the specifics of timing,

electric vehicles? but we have focused our shared
autonomous development on San
At GM, we believe that all autonomous Francisco and the United States.

General Motors Advances
Virtual Autonomous
Driving & Active
Safety
By Chris Kinser, General Motors
G
Figure 1: Cruise Autonomous Cars eneral Motors operates a total vehicle performance center at the Milford
Proving Ground in Michigan (Figure 2). The Global Autonomous Driving
Center is a subset of this work focused on developing active safety
features like advance park assist, lane keep assist, full-speed range
adaptive cruise, and Super Cruise. This work is guided by GM’s vision of
a future with zero crashes, zero emissions, and zero congestion. The mission of our team
is to provide smooth, capable driver assist systems that delight our customers.
GM’s Approach to Automated Driving
The industry standard scale for levels of autonomy (SAE) is helpful from an academic
perspective when discussing vehicles and their capabilities. However, when we begin
development of a new vehicle or system, we don’t start with a level in mind, but rather
with the use case and a set of features that we believe we can safely implement. It is this
focus on safety that guides us through the process.
General Motors is the only company that has everything from design, engineering
validation, and testing all under one roof. This is more than just designing and building
the vehicle. It also includes everything from in-house security and connectivity systems
to software development and high-resolution mapping. Having everything under one roof
puts us in a unique position to safely develop and deploy autonomous vehicle technology.
Super Cruise
Super Cruise is an advanced driver assistance feature that enables hands-free driving on
supported roads. It combines adaptive cruise control and lane-centering control with a driver
attention system (Figure 3) to allow you to drive with your hands off the wheel and eyes
on the road. Super Cruise is aimed at providing comfort and convenience in long-distance
travel and daily commutes. Customers receive updated maps on a regular basis (Figure 4).
Figure 2: General Motors operates a total
vehicle performance center at the Milford
Proving Ground in Michigan

Safety is
engineered
into every
step in Cruise’s
self-driving
vehicles Figure 3: General Motor’s Super Cruise
including design, Cadillac Driver Attention System
development,
manufacturing, Cruise Autonomous Vehicle
testing and (AV) Program
validation. In May 2016, GM completed the into every step in Cruise’s self-driving
acquisition of Cruise Automation a vehicles including design, development,
Silicon Valley startup with considerable manufacturing, testing and validation.
self-driving software development On a typical day, Cruise autonomous
expertise. Combined with our expertise in test vehicles safely execute 1,400 left
engineering and developing vehicles, our turns and our teams analyze all that data
teams began testing self-driving vehicles and apply learnings. Based on Cruise’s
in San Francisco, CA, Scottsdale, AZ experience of testing self-driving vehicles,
and Warren, MI. By September 2017, we every minute of testing in San Francisco
revealed our first self-driving test vehicle is about as valuable as an hour of testing
built from the start to operate on their in the suburbs because of the complex
own with no driver (1). Safety is engineered decisions being made.
Reference
‘How we built the first real self-driving car
(really)’, Kyle Vogt, Cruise, September 11, 2017
Blog Post: https://medium.com/cruise/how-
we-built-the-first-real-self-driving-car-really-
bd17b0dbda55
Figure 4: GM Super Cruise, before going to

production, required mapping every major
road in the U.S. and Canada

SU STAINAB ILITY
Sustainable Engineering is Good Business Sense:
Computer-Aided Engineering
Can Solve Global
Environmental Challenges
By Keith Hanna, VP Marketing, MSC Software
E
nvironmental concerns over the sustainability of our unique planet and its
complex ecosystems are increasingly dominating our mainstream news
outlets having moved from a fringe topic to be a political priority in the
developed world over the last ten years (1, 2). The advent of environmental
activism such as direct action from the ‘Extinction Rebellion’ movement over
the last year, huge concerns over climate change especially with respect to rainforest fires
in the Amazon, shrinking ice sheets at the Poles and receding mountain glaciers, and the
rise of Green political parties in Europe and the ‘Green New Deal’ in America has brought
the issue very much to the fore. And ever more stringent environmental legislation is
coming from our legislators and parliaments to help address the issue. But what exactly is
‘Sustainability’?
The concept of sustainability is arguably as old as us humans reusing our nearby

resources to survive from year to year, but it has emerged in our time to take its present
form because of the Environmental Movement in the 1970’s with the best definition being
that from a seminal United Nations Report on ‘Our Common Future’ (3) in 1987 by the
Norwegian politician Gro Harlem Brandtland who wrote that:
“Sustainable Development is the development that meets the need of the present without
compromising the ability of future generations to meet their own needs.”
The UN’s response to Brandtland’s work was the subsequent Kyoto and Paris
Environmental Agreements in 1997 and 2015 respectively. These protocols made all the
world’s major nations agree to restrictions in harmful co-called ‘greenhouse gas’ emissions,
along with timetables for countries to move towards ‘sustainable’ energy sources in the
future. Sadly, the protocols have not been entirely successful as several nations have flouted
them, but they have created impetus and a framework for change today.

The Earth and Sustainability – of CO2 goes up into the atmosphere and is
How Bad Is It Really? not recovered or recycled. Rollén states that
71 out of the 100 top polluters in the world
In a presentation at HxGN LIVE in June today are companies or corporations; only
2019, President & CEO of Hexagon, Ola 29 are countries.
Rollén, delivered a sobering keynote to
the audience describing unprecedented Our consumption of difficult-to-
sustainability challenges facing humanity biodegrade products increases year on
and our beautiful blue planet (2). He year - we typically produce 1,000,000
presented some shocking statistics plastic bottles a day in the world and
including the fact that one million species less than half are recycled. By 2029
on our planet are in survival jeopardy today. more than 5 billion plastic bottles will
Indeed, since the dawn of time we have have been made and disposed of,
killed off over half of the species on our and 8 million get into our oceans every
planet including 80% of all sea and land year. Rollén reported that we ingest
mammals. By 2050 we will have trawled all 7,000,000 microplastic particles per year
fish from our oceans if we continue to fish at into our bodies. Indeed, our western
the levels we do today. In the last 120 years world addiction to single-use plastics
alone, we have lost 10,000,000 hectares is contributing to a mountain of plastic
of forestry from the face of the earth; that’s waste that we recycle in Europe and
about 1/10th of the land mass of Canada. North America and then our governments
President & CEO of Hexagon, Ola Rollén, at
ship the waste that cannot be disposed
HXGN LIVE
A big factor in the current ecological crisis of half way round the world for either
on planet earth is the increasing world landfill or, increasingly, it ends up in illegal
population of people and the sustainability burning of plastics in developing countries
challenge that inevitably poses given with lax environmental regulations.
finite resources and a thrust towards
Western living standards across the world. Whatever one’s political viewpoint, it
In 1804 the world’s population passed is self-evident that we need to reach
1,000,000,000 people for the first time. It sustainable usage of the world’s natural
took another 123 years to reach 2 billion, and recycled resources especially with
and then the planet’s population rose to 3 a rising population. Consumerism in the
billion people in 1960, 4 billion in 1975, and developed and developing world alone,
it has now peaked at 7.7 billion today. By urbanization, and an addiction to single
2050 it is estimated that the world will be use fossil fuels, is creating a huge problem
inhabited by 10 billion people and we will that humanity needs to address in the 21st
have an even bigger global middle-class Century if we and our planet are going to
market of consumers. The world’s human survive in a form we are comfortable with.
population has risen dramatically but Indeed, it is stating the obvious to say that
we’ve not scaled sustainably, especially a large number of Asian, North American
when you consider that we are reaching and European cities are at risk from sea
co-called ‘Peak Oil’ in our natural resource levels rising if temperatures rise significantly
exploitation and one-third of all agricultural because they are located on low lying
produce each year goes to waste. coastal plains.
Across the world we emit approximately

2 million tonnes of CO2 each day into the Sustainability in Manufacturing;
atmosphere and this is projected to double A Road Less Travelled, A
by 2030 unless we reverse our behavior Journey Worth Taking
(2)
. We are emitting so much CO2 that a
‘Doomsday Limit’ will be reached by 2050 At the end of the Second World War
according to Rollén where it will be hard Japan’s industrial base was crushed;
to correct for the emissions we make. manufacturing machines were broken, and
Of the 36 Billion tonnes of CO2 related engineers had to work with few materials
greenhouse gases we emit each year, on production lines that were prone to
nearly 35% comes from energy production shortages in the post-war depression.
(for electricity), 25% from agriculture, American academics like Professor
20% from industrial emissions, 14% from Edwards Deming came to Japan and
transportation and 6 % from buildings. In the encouraged Japanese manufacturing
USA alone, 68% of total energy production industry to adopt a wholesale embracing

of ‘Total Quality’ methods in factories. and creating more waste than we can lower cost with safe, efficient operations
Slowly, over two decades Japan pulled handle. So, is there any good news? Well, and longer lifespans. It will also mean
its manufacturing base up to be one continued economic growth according to that buildings and the built environment
of the best in the world. Japanese Rollén doesn’t have to be at the expense will have optimized resources by design,
industry invented the term ‘Just in Time of the planet. Every company today wants less rework and projects delivered on
Manufacturing’ that caused their industries to boost its efficiency, its productivity, and plan, on time, and on budget. In the
to become world class. The Japanese quality outcomes in Smart Factories and Mining Industry, sustainable engineering
word kaizen that represents total quality Smart Manufacturing processes as they will mean minimal environmental impact,
manufacturing means incremental embrace the looming autonomy challenge. safety is prioritized, and operations are
performance improvement within existing And if they can achieve those outcomes run at maximum efficiency. In agriculture
manufacturing processes, and it soon at scale, not only can they sustain growth and farming, sustainability will mean that
led to better products in their Factories for their businesses, but also for the planet inputs are optimized and farmers will
such that by the 1960s their productivity that we all share. In effect, sustainable produce higher crop and livestock yields
was twice that of equivalent Western manufacturing will lead to fewer natural with less waste, at lower costs.
European and North American factories, resources being used, less waste through
with minimal waste and high quality. ‘Just perfect quality, and less pollution through
in Time Manufacturing’ was rechristened zero or no emissions. Design & Engineering
as ‘Lean Manufacturing’ in the 1980s Simulation (CAE) for
by western academics as it came back To achieve sustainability in all industries, Sustainability
into the Western World and was widely engineers will need to increasingly design
adopted. Factories then moved towards sustainable processes, products and Since its nascent use over 50 years
automation in the late 20th Century and we value chains. Sustainable development ago, Computer-Aided Engineering (CAE)
are now seeing a shift towards autonomy in engineering design means embracing has by its very predictive nature been at
in the 21st Century. four context levels: Planet, Society, the forefront of more efficient and high-
Business, and Technology (1). It needs to quality product development. It has also
Like all global manufacturing industries, address problem definition, synthesis of been leading the way in developing new
equipment manufacturers will be potential solutions, analysis of possible environmental technologies, and MSC
increasingly forced to embrace solutions, and then improvements to Software has been a pioneer. We see
sustainability criteria in their product the final solution, all in the context of a this repeatedly because CAE engages
development whether they want to or solution that meets both the needs of with research & development design
not, because of the quadruple pressure people and all specific context constraints processes so that better and more efficient
points of declining natural resources, with regard to the planet and prosperity. products can be designed with fewer
governmental legislation, societal pressures Sustainable design engineers will need to (and sometimes no) prototypes being
and consumer choice. According to Rollén consider and meet constraints on energy, built during the manufacturing process.
(2)
most companies today want business water, food, health, and other basic This means less material waste, low
growth and many talk of sustainability human needs as part of their sustainable scrappage rates and lower energy usage.
however it is defined. But all definitions manufactured product design process. In short, CAE helps to design products
are grounded in the same notion, that Once achieved we will see for instance right-first-time and fit-for-purpose in each
we’re overwhelming the earth’s resources rapid startup of industrial facilities at and every industry it is applied to. CAE is
Adams Wind turbine model during a gust event

“Sustainable
Development is the
development that meets
the need of the present
without compromising
the ability of future
generations to meet
their own needs.”
-Gro Harlem Brandtland

also ‘upfront’ in every product design and structural, acoustic, fluidic, manufacturing alter the nature of the mechanical design
manufacturing process (and concurrent and mechanism design. In short, CAE process by embedding sustainability related
throughout the process) in the world. simulation has resulted in leaner power concepts of design exploration into the early
This means it can be used to design out generation operations and reduced design phase for manufacturing. I also think
undesirable performance features and to carbon footprint by design. As the cost of that this century there will be opportunities
avoid negative environmental impacts at scalable sustainability keeps decreasing for new socially conscious, environmentally
the early conceptual design stage when in real terms, improving industrial and friendly, sustainable, high-tech engineering
Return on Investment (ROI) is highest. manufacturing processes will therefore products and companies all producing
And CAE can conceptualize as yet have an impact on sustainability big time. outputs conceptualized and designed
untested manufacturing processes and in PLM (Product Lifecycle Management)
new products, and therefore is a very cost In terms of sustainability and acoustic software driven by CAE simulations.
effective and safe way to test ‘what if..’ simulation, our Actran product has been
scenarios without hazardous consequences used to simulate exotic materials and
to both people or the environment. their acoustic impacts, eg composites Towards Next Generation
and other new materials. We see it used Sustainability in Aircraft
In real terms this can be seen most by to assess drive-by noise for trucks, trains Design with CAE
the work done in the Power Generation and cars, and in Green Building Codes for
and Renewables Energy sector over noise sustainability in simulations. MSC Present aerospace forecast trends indicate
the last two decades where CAE Nastran and MSC Apex have addressed that there will be 34,000 civil aircraft in the
has been at the vanguard of new sustainability issues in analysis of world by 2036, a 60% growth on today
technologies with predictive simulations. mechanical structures with the use of FEA with 40% being replacement aircraft. This
MSC Software products such as MSC simulation to provide generative design demand in air travel is going to happen
Nastran, Adams, Cradle CFD, Digimat, for eco-friendly lightweight structures. because of more affluence in Asia. It is
Simufact, MaterialCenter, Actran and They have also been used to optimize also worth pointing out that electric battery
Marc have all been helping to design manufacturing processes for reduced powered (and hybrid) aircraft engines, plus
radically new and innovative energy waste and environmental impact and to aircraft with solar panels on their wings
generation processes, nuclear reactors, extend product life via durability and total that result in low emissions are being
next generation ships, planes, trains, cost of ownership (TCO) simulation in designed by CAE tools today before they
spacecraft and automobiles. CAE has order to support development of smart are flight tested. Additive manufacturing
been used extensively over the years to connected autonomous ecosystems. of specialist parts, particularly in defense
simulate renewable technologies such applications, is becoming more common.
as wind and wave/water turbines, and In summary, I believe that the need for Internal parts of aircraft can be 3d printed
zero emission solar panels for optimal environmental sustainability will significantly but increasingly we will see more critical as

well as non-critical parts being additively Moreover, the engine’s acoustics are engine cars totally by 2030. Major
manufactured especially with more 40% lower than other aircraft engines cities like London are banning all diesel
sustainable materials. in its class thus minimizing acoustic engine cars from the city center today
pollution – all thanks to CAE. Boeing to minimize dangerous particulate
The new Airbus A350-1000 and and Airbus aircraft airframes won’t be emissions. Norway will phase out
Boeing 787 Dreamliner are coming to made of aluminum. Instead, exotic new conventional cars by 2025, followed by
an airport near you. They are the most titanium composite materials are being France and the United Kingdom in 2040
technologically advanced passenger employed that are stronger and lighter. and 2050, respectively. India has said
aircraft in the world and these next These interestingly have an added benefit that no new gasoline or diesel vehicles
generation aircraft have been extensively that the atmosphere inside the cabin can will be allowed by 2030 (if economical).
designed by CAE. They are eco-friendlier be more humid than ever before such that
than their predecessors. The A350’s passengers will breathe air with a density In order to achieve emission reduction
wings actually structurally deform during and humidity typical of that at 6,000 feet targets in Europe, the electrification of
flight especially at altitude where the rather than 40,000 feet thus leading to powertrains for all OEM vehicles will be
aircraft slightly transforms its flaps to less hair static and itchy, skin rash than required. Indeed, the advent of electric
fine-tune its center of gravity depending ever before! powertrains inside cars will remove a
on its weight at that point in time as fuel lot of complexity in manufacturing with
is consumed. Doing this optimizes its ensuing sustainability benefits. Typically,
flight lift-to-drag ratio and decreases fuel Sustainable Automobiles Hit a conventional internal combustion
consumption over conventional planes The Road thanks To CAE engine has 1,400 parts yet an electric
significantly. The Airbus’ engines will be vehicle powertrain will only have 200, so
the most powerful ever, generating a The automotive industry is going through in principle manufacturing will be easier.
maximum thrust of 97,000 lbs (over a unprecedented changes due to tighter Electric vehicles are projected to be 54%
quarter of the thrust of the NASA Space emission regulations and autonomous of all new car sales in 2040 (5).
Shuttle)! The net effect is its range of mobility disruptions. In the European Union
8,000 miles means it will travel from alone, we will see new CO2 regulations in All sorts of lightweighting technologies
London to Perth Australia in one flight. In 2020 where an emissions target of 95g are also being employed as standard in
addition, optimal fan performance in the of C02 / km will be in place for all new the manufacture of our road cars today.
design of the aircraft’s engines, coupled passenger car fleets, a 27% reduction Composites were once the preserve of
with other efficiency improvements compared to today. This target will be Formula 1 cars but are becoming more
designed by CAE simulations mean that coupled to an excess emission premium for common in saloons. And, of course,
the A350-1000 will use 25% less fuel manufacturers failing to meet the emissions the major disruption that autonomous
than any other commercial aircraft. This standard of 95€ for every g/km of excess vehicles will bring to the ground
will reduce carbon dioxide emissions emissions per vehicle. In Denmark, the transportation industry in the next 5 to
and overall traveler carbon footprints. government is banning internal combustion 10 years will also lead to sustainability

benefits such as near zero emissions, and we see a wide variety in every product economic driver in some industries
better acoustics and longer vehicle we touch and use, both natural and man- and application areas.
lifetimes and more usage because most made. Material resources on our planet
will be electric vehicles. are under immense pressure because they Every single CAE simulation in the
are finite and under intense pressure from world today requires good material
In terms of MSC Software solutions overmining and under-recycling according properties to produce accurate
for automotive sustainability, we are to Victoire de Margerie from the World results. Sustainability in engineering
seeing extensive exhaust and drive- Materials Forum (5). She has identified simulation predictions therefore
by environmental noise predictions several bottlenecks citing the concern that requires proper material utilization
and coupled fluid-acoustic simulations 60% of the world’s cobalt is used to help and optimization throughout a
of exhaust silencers as well as make electric batteries non-inflammable manufacturing process. And every
particulate emissions modeling. Our but it is sourced from the Democratic CAE simulation should require a
Digimat, Simufact and MaterialCenter Republic of Congo in Africa, one of the fully traceable simulation supply
lightweighting solutions involving world’s most unstable countries, and chain with trackable components
composites and additive manufacturing there is no alternative today with only 25 throughout the product’s lifecycle
are best in class and NVH from MSC years of known reserves left in the world. and its circular economy from
Nastran and Actran are well respected. When we consider Smartphones, there conceptual idea to recycling.
Increased use of VTD for environmental are 2.7 billion units in the world today,
virtual testing of ADAS systems and a tenfold increase over 10 years ago. To address materials sustainability,
autonomous vehicles will ensure the All smart mobile phones use rare earth MSC Software’s MaterialCenter and
safety of today’s and tomorrow’s vehicles metals for their screens and 90% of all rare Digimat productlines are part of a
and right-first-time deployments. earth production is controlled by China. unique strategy for a comprehensive
And many nations see this as a national 10x materials solution that covers
security concern. The market leader, virtual material development,
We Are Living in A Material Apple, has created a generous buy-back virtual materials testing, advanced
World and There Is A smart phone program to aid in recyclability (multi-scale) and standard material
Sustainability Issue of its phones but also to ensure that it modeling serving all major FEA
has access to enough rare earth metals codes, inclusion of the effects of
It is safe to say that materials are so it doesn’t become a business risk. additive manufacturing, addressing
everywhere in the world that we inhabit; Sustainability is therefore becoming an automatic fiber placement and
Integration with iPoint in MaterialCenter for Compliance Check

the effect of defects, ensuring digital approach allows users to make decisions
continuity and a material-centric digital- for a more responsible material design at
twin, embracing artificial intelligence, a very early stage in the product creation
and finally ensuring materials compliance process – the precondition for a truly
and sustainability. As an example, circular economy and sustainable supply
MaterialCenter provides a complete chains.
solution for the current and future needs
of aerospace materials data & process
management. Designed to manage the Summary and Conclusions
complete materials workflow as a single
point of entry for all materials related In the 21st Century sustainability is
activities, it guarantees that engineers use becoming an overriding priority in the
a consistent source of approved materials globalized economy across all industry
derived from traceable integrated processes sectors and manufacturing operations.
thus resulting in improved simulation fidelity, Social, environmental, and economic
reduced data loss and the elimination of trends are driving the sustainability agenda.
tedious manual data management activities. And this convergence of existential need
This produces the 10x productivity gains and technological opportunity will require
our customers see. sustainable engineering design solutions
to improve product performance, product
In terms of sustainability and compliance quality, product manufacturability and
of materials, Hexagon/MSC’s e-Xstream ultimately product and material recyclability.
Engineering team is working with Computer-aided engineering simulation
iPoint GmbH in Germany, a pioneer when done well will result in minimal
in the digital circular economy, to offer waste in manufacturing production which
sustainable materials management means no recalls, no scrappage and no
and simulation solutions as part of our rework. Sustainable design, thanks to
highly productive and unique Integrated CAE simulation, will therefore help to drive
Computational Materials Engineering economic benefits for all our customers,
(ICME) solution. Our partnership with as well as address mankind’s ever-growing
iPoint Systems also permits a direct environmental concerns for our planet and
integration of environmental compliance its finite resources. At both Hexagon and
check, and comparison of the properties MSC Software, we have a simple mantra
of both physical and simulation data that embodies our sustainability vision …
sets in MaterialCenter 2020. Material Do Well to Do Good. We believe that it’s
environmental and social compliance good economic sense to do sustainable
is a complex landscape and a critical engineering and we believe that our
aspect for many organizations. Direct children and grandchildren will thank us for
integration of iPoint’s Compliance it… and this will happen without humanity
Application (iPCA) with MSC Software’s being driven back to the Stone Age.
MaterialCenter 2020 ensures robust and
immediate environmental compliance
checks providing the unique capability of References
combining in-house corporate materials.
1. “Engineering for Sustainability - A Practical
This software partnership allows us to offer
Guide for Sustainable Design” by G. Jonker,
robust and accelerated environmental Elsevier Press, 2012, ISBN 044453846
compliance checks and is a unique 2. “Your Data can Save the World”, Ola Rollén
capability combining corporate materials Keynote Speech, HxGN LIVE 2019, Las
Vegas, USA https://www.youtube.com/
management and compliance with the
watch?v=tYWuwVJylVQ
latest information on legislation like REACH
3. “World Commission on Environment and
and RoHS which permits comparison Development, Our Common Future”,
of properties from both physical and Brundtland, G.H. United Nations Report
simulation data. The combined integrated A/42/427, 1987
4. “Towards Sustainable Design in the Sports
offering provides material engineers, design
& Leisure Industry”, Hanna, R.K & Subic, A,
engineers, CAE analysts, and compliance International Journal of Sustainable Design
specialists the full capabilities needed 1(1), January 2008
in their respective domain of expertise 5. “Sustainable Growth Thanks to Using
with access to information needed from Materials Smarter, Less & Longer”, Victoire
de Margerie, ICME Conference, Bordeaux,
the other disciplines. This frontloading France, October 1st 2019

FEAT URE

Ford Motor Company
Reduce Need for
Physical Prototyping
using Adams Real Time
By Tony Ge, Ford Motor Company

Eric Pesheck, Paspuleti Rahul Naidu,
Michael Collingridge, Hemanth
Kolera-Gokula, MSC Software
T
he Automotive industry is under constant pressure to address market
demands in resource-constrained engineering environments. Stiff
development timelines are compounded by shrinking physical prototyping
budgets. The Ford Motor Company is focused on providing its customers
with a vehicle that is world-class in both drivability and comfort, which
are critical attributes that impact a customer’s perception of the vehicle. These critical
attributes are in turn impacted by the transmission shift strategy.
The traditional approach to transmission calibration at Ford involved physical testing of

full vehicle prototypes with extensive test schedules which was both time consuming and
expensive. In collaboration with MSC software and its services group, Ford developed a
Hardware-in-the-loop testing process to reduce the need for physical prototyping.

Hardware-in The Loop Testing (HIL)
Ford has been a longstanding user of generation of the FMU, the model
Adams vehicle dynamics models and representation that conforms to the
with the release of Adams Real Time (RT) FMI co-simulation standard
in 2017, Ford identified an opportunity to • Integration with HIL test bed. The
leverage its existing Adams knowledge base FMU was then ported onto the
further downstream in the development HIL platform and the model was
cycle for vehicle testing and calibration. calibrated to test data prior to
performing hardware tuning.
Together, MSC and Ford have implemented
a combination of physical components Since the real time Adams model was
and Adams Real Time models in a HIL derived from an existing vehicle model
test environment to evaluate the shift that was already being used within Ford
quality of multiple vehicle platforms. For for vehicle development, the model
each test set-up, a physical engine and development overhead was minimal.
transmission prototype was connected to Generation of the real time model from the
a virtual Adams model of the vehicle. The existing full-fidelity Adams model involved
implementation consisted of three phases: multiple model reduction strategies.
• Model conversion for real time For this use-case, it was necessary to
analysis. Existing Adams vehicle include leaf springs in the real-time model.
models were used to create Since the pre-existing beam-based leaf
corresponding Adams Real Time spring model was too complex for real-time
models that met the real time analysis, it was replaced with a 5-link model
platform/hardware requirements with equivalent behavior. A suspension
platform/hardware requirements subsystem simulation was run on a beam
• Model setup. The model was prepped leaf spring model to obtain target values
for integration into the test bed. This for the spring characteristics. The model
involved creation of I/O channels, tuning was assisted by executing a DOE
tuning of the RT solver settings and that fine-tuned the hardpoints and bushings
Speed
Torque

Figure 1: Simulation Correlation with
Test data at 50 % pedal input (VSPD
Vehicle Speed, OSS-Output Shaft
Speed)
Figure 2: Simulation Correlation with Test

data at 75 % pedal input ( VSPD-Vehicle
ACC Speed, OSS-Output Shaft Speed)
to capture the right behavior in the 5-link Given this input torque, the Adams model Figure 2 shows the output shaft speed and
model such as primary frequencies, static reacted and provided a driveshaft speed the longitudinal acceleration response. The
preloads, vertical rates and wind-up rates. response to the dynamometer which then vehicle response in between and during
The Adams Real Time model had 150 enforced this speed on the engine and gear shifts showed good agreement. The
DOF and preserved the topology and transmission. The model also provided agreement was consistent across all pedal
parametrics of the original model. Elements the longitudinal vehicle acceleration as positions pointing to the predictive nature
such as hardpoints, joints, springs, the vehicle changed gears, which could of the Adams model. Calibration engineers
dampers and bushings were maintained then be related to a ride comfort index. then used the HIL setup to gauge and
and could be modified. The benefit of this The ability to use a combination of vehicle tune transmission parameters to attain the
is that the model could capture higher dynamics models and hardware to desired behavior.
frequency characteristics in the system simulate vehicle behavior provided Ford an
responses and different configurations opportunity to create both more efficient The HIL dyno testing process has helped
could be quickly explored. and more comprehensive testing programs the Ford team meet its test objectives
while reducing vehicle prototypes. with fewer physical prototypes and in
The Adams-Real Time model was then a lab environment that facilitates the
exported as a Functional Mock-up Unit The HIL realtime predictions were validated tuning process. The same engine and
(FMU) and ported onto a HIL platform against vehicle test data to confirm correct transmission prototypes can be tested
for execution. In future implementations, vehicle response to known input torques. with multiple vehicles under multiple
modeling parameters such as hard Figure 1 shows the velocities of the actual loading scenarios.
points, bushing characteristics etc. will vehicle speed versus the model.
be exposed in the FMU for run time
tuning. This will allow Ford engineers to
make changes to parameters and tune
model response without having to make
changes to the original Adams model and
export a new FMU. What are the 5 Myths of MBD? Download our Free
Whitepaper: www.mscsoftware.com/5myths
The physical torque at the output of
the transmission was instrumented and
applied to the Adams model as an input.

AI / MACHINE L E A R N I N G

Enabling Accurate Design
Decisions While Compressing
Engineering Timelines
with CADLM Technology
By Kambiz Kayvantash, CEO, CADLM

Hemanth Kolera-Gokula ,Fabio Scannavino,
Manuel Chene, Raoul Spote, MSC Software

D
ynamic processes can be less computational overhead. Reduced In a recent proof-of-concept, Adams, the
described using differential Order Models (ROMs) provides an Multi-Body Dynamics (MBD), and Lunar, the
equations, and the solution opportunity to create a virtuous Real-Time supervised Machine Learning solution from
to these equations can loop between Design and Operations CADLM, were used to create Reduced
provide insight into the with real time information sharing. It also Order Models (ROMs) of vehicle behavior.
nature of these processes. However, provides an opportunity for simulation The simple scenario involved a vehicle
the simulation of such equations utilizing software providers like MSC to truly accelerating on a straight road encountering
computational techniques, such as finite democratize engineering simulation across an obstacle in its path. The goal was to
element or finite volume methods can the product life cycle in a scalable manner predict any interference with the obstacle
become computationally very expensive without compromising on model fidelity. using real-time reduced-order models. The
or for some industrial problems unfeasible. process occurred in two distinct phases. A
The modeling of optimization problems In the past decade, much effort has been design space for exploration that spanned
and Multiphysics phenomena in practical made to develop various methods of two dimensions, vehicle velocity, and
engineering applications is often very model reduction. MSC and CADLM have mass was defined in Adams. The process
challenging as repeated numerical partnered to develop model reduction then involves creating a dataset in Adams
simulations are required. A remedy is a approaches for a variety of engineering that spans the design space. Lunar then
simplification of the physics-based model problems while remaining agnostic to consumed this dataset and created a purely
but that relies on the experience and the underlying physics type. Using this data-based model that emulated the high-
intuition of the engineers. Another avenue approach, the engineering simulation fidelity representation in Adams.
is reduced-order modeling, a mathematical community can tailor the level of model
approach serving to overcome high fidelity to the underlying simulation intent. The transient predictions from the ROM
computational costs of the simulations. For example, reduced-order surrogates are in line with the Adams results both in
The primary goal is to approximate the of high-fidelity models can be used to terms of the trend and magnitude with the
large-scale problem by a much smaller explore the design space and execute added advantage of real-time predictions.
one, which yields somewhat less accurate computationally intensive, vehicle reliability, The caveat being that the ROM is valid only
results but can be solved with considerably and optimization tasks. within the bounds of the design space and
Real-Time Reduced Order Models

Figure 1: ROM Lunar analysis of an Adams Car Model
Training the ROM Lunar Model to create a ROM Validation between ROM and Adams results Prediction based on validated ROM
any explorations beyond these bounds

Figure 2: Marc-Cradle analysis of an Diaphragm Valve Model
would require additional ROM training.
CADLM capabilities can be leveraged

across the MSC portfolio and are agnostic
to the physics type. Reduced order
representations of both systems and
component can be created. MSC is also
looking to leverage ROMs to extend its
industry leading capabilities in the area
of multi-physics. As an example, ROMs
simulating highly non-linear Fluid-Structure
interactions have been created from
time intensive Marc-Cradle simulations.
The application goal was to optimize
the displacement sensitivity of a flexible
membrane with respect to various inflow
velocities and membrane mechanical
properties. This involved the costly
computations and co-simulation between
Marc and scFlow. A Reduced Order Model
was constructed based on only eight
completed runs and employed within an
optimization loop. The solution was found
by LUNAR within seconds and a final
co-simulation was conducted in order to
validate the optimized solution.
CADLM technology allows simulation users

to make effective and precise engineering
decisions while compressing design
timelines. The ability to create traditional
high-fidelity models and extract reduced
order surrogates from them provides a
simulation user the opportunity to tailor the
model fidelity to the simulation intent and
conduct real time optimization or reliability
studies.
For More Information: www.cadlm.com

Bridging the Gap Between
Design and Additive Manufacturing
Using Smart Generative Design
By Hendrik Schafstall, CEO Simufact

Raj Dua, Product Manager,
MSC Apex Generative Design

For MSC Software and Hexagon,
Generative Design is an initiative to
provide a tool that will help customers
design concepts unimaginable by the
human mind.
W
ith the release of MSC Apex Generative Design, MSC Software is
now offering an entire digital solution from the design to the final
validated part for all materials. Connecting design solutions like MSC
Apex Generative Design to virtual manufacturing simulation with What is
Digimat AM or Simufact Additive, the design can account for the Generative Design?
engineering and production phase challenges earlier in the product development phase.
As a digital twin, the virtual manufacturing simulation is used to identify the best printing Simply stated, Generative Design is a
process and to optimize the orientation of the part and the build process. Furthermore, process of automatically generating
the outcome of the additive manufacturing process chain can be used for the validation several design concepts that satisfy
of the “real” geometry, while accounting for the residual stress distribution and the local a set of user defined objectives,
deformation under real load conditions using MSC Software’s design validation solutions criteria, and constraints. Generative
such as MSC Nastran or Marc. The end-to-end process enables engineers to make sure Design can be accomplished in
their optimized designs are validated for manufacturability and performance. many ways depending on what
criteria and constraints have been
defined by the user. For example,
if a user defines a set of structural
loads and boundary conditions
that a part must withstand as
criteria, an upper stress limit as
a constraint, and an objective of
MSC Apex minimizing mass, a method known
Generative Design as Topology Optimization (which
many of our MSC Nastran users
are very familiar with) can be used
to generate a number of design
concepts that satisfy the given
criteria and constraints. However,
Generative Design is more than just
Topology Optimization. For instance,
a user may want to know what the
best way is to package a number
of electronic components in a given
space in order to minimize the
gap between all the components.
Generative Design can help answer
that question. For MSC Software
and Hexagon, Generative Design
is an initiative to provide a tool to
our design customers that will truly
act as a companion and help them
think of design concepts that are
unimaginable by human mind.

Why Do We Need are seen as two major constraints in wide Each optimization always leads to a
Generative Design? adoption of AM for mass production. geometrical and mechanical correct design
Therefore, there is a need to account and that can be used for manufacturing. In
optimize for the total manufacturing costs addition to the geometric side, the user
and print time while designing parts for AM. must also understand the cost and
With MSC Apex Generative Design, we are feasibility of using AM for this design
focusing not only on optimizing the parts for candidate. With MSC Apex Generative
AM, but also optimizing the process for AM. Design, our goal is to allow users to
We believe that it is only after we bridge the specify manufacturing related constraints.
gap between design and manufacturing For example, if the goal is to minimize
that we can see AM become a sustainable the cost of 3D printing, then MSC Apex
manufacturing method. Generative Design will automatically check
each design candidate for: (a) amount of
material required for the part, (b) volume
Bridging The Gap of support structure required for support
and heat dissipation in the AM machine, (c)
MSC Apex Generative Design is being cost of removal of support structures and
developed as a first-of-its-breed tool machining for desired surface roughness,
The first release of MSC Apex to bridge the gap between design and (d) costs related to maximizing the number
Generative Design has been manufacturing. Our goal is to automate of parts printed at one time on a build plate,
released to assist design engineers the process of Generative Design with etc. These checks are performed in the
create organic topologies that can user intervention only required for defining background using MSC’s Simufact Additive
be manufactured using 3D printing, the objective, criteria, and constraints technology for metal parts and Digimat AM
i.e. Laser Powder Bed Additive for design space exploration. MSC Apex technology for polymers. At the end of the
Manufacturing. Technologies Generative Design will then account optimization routine, MSC Apex Generative
such as Topology Optimization for how the part fits within the overall Design selects the candidates that meet
are being reinvigorated thanks assembly, how it redirects loads to other the specified criteria and summarizes
to advancements in Additive parts of the assembly as its stiffness them. The design engineer can then export
Manufacturing. It is widely accepted changes, and most importantly MSC the selected design(s) in CAD format and
that Additive Manufacturing has Apex Generative Design accounts for perform further checks, for example, for
the ability to manufacture virtually manufacturability – all automatically while buckling, fatigue, and nonlinear for part
any topology. As a result, the generating several design candidates that performance or decide for one design
industry has seen a rise in the all meet the user’s defined expectations. based on additional reasons such as of
number of tools that allow creation Many Generative Design tools in the dirt problems or just aesthetic ones. The
of organic topologies via concepts market today allow users to minimize design engineer may also choose to send
such as Topology Optimization the mass subject to a stress constraint. the part to a manufacturing engineer to
and Generative Design. However, The tool then solves a mathematical perform further checks on manufacturability
if you have ever tried to 3D optimization problem and produces one or via Simufact Additive and/or Digimat AM.
print any “organic” topology more design candidates. Although these Users will be able to perform any geometry
that resulted from the Topology candidates often are more ideas for a modifications needed using the geometry
Optimization algorithm, you have part and visualizations of how the forces editing tools in MSC Apex. Eventually, MSC
probably realized that AM is not flow through the design are. A proper Apex Generative Design is able to perform
very forgiving, and an unrevised Generative Design tool needs to produce these checks during the optimization
Topology Optimization result often is directly printable designs that can be used process automatically as well.
far from feasible. Despite its unique without any need for manual rework of
ability to manufacture virtually geometry defects. This is what MSC Apex
any topology, AM still has many Generative Design will deliver. Speed Is Crucial
limitations today. Issues such as
shrink lines, cracking, overheated In order to evaluate several design
zones, etc. have kept AM from candidates in a time effective manner,
replacing other manufacturing it is necessary to have a finite element
methods. These issues were not solver and an optimization engine
as pervasive when 3D printing was that can take advantage of the latest
only used for prototyping. However, computing technologies for extremely fast
they become prevalent when performance. With MSC Apex Generative
using AM for production parts, Design, we have done exactly that. We
especially primary or secondary have completely rewritten the FE solver
structural parts for Aerospace or and the optimization engine to scale on
Automotive industry. Today, cost of Figure 2: Topology Optimization of a GE Engine multiple GPUs and CPUs. The ability to
manufacturing and time of printing Bracket using MSC Apex Generative Design explore design space in a time efficient

Figure 3: Design for Additive Manufacturing using MSC Apex Generative Design
manner ensures that the design process is as possible while keeping into account the full potential and benefits of AM, users
not a bottleneck and thus allows our users the boundary conditions, constraints and need to be able to produce designs that
to make decisions solely based on design optimization goal. Thus, several design are specifically validated for AM. With MSC
criteria. Only a complete examination of the candidates are produced and directly Apex Generative Design, we are developing
design space with a variety of results, and in verified in the background using Simufact technologies that validate manufacturability
a short time, leads to the best results. Additive for metals or Digimat AM for in the Generative Design process. As such,
plastic products. While selecting the right the optimization engine only produces
candidate and iterating the manufacturing geometry candidates that have been
Demonstrating The Potential simulation, the perfect design in terms validated for AM.
of manufacturability, weight and costs is
To bring evidence on the potential of MSC selected. As a last step in the virtual world, Finally, after printing the part with your 3D
Apex Generative Design and to show this design finally gets a last validation printer of choice, Hexagon metrology’s
its usability, a wheel carrier of a formula with MSC Nastran for FE qualification state-of-the-art scanners can verify the
student team is considered to demonstrate and back again in Adams to ensure the accuracy of the simulations and compare
a use case. Due to its very complex load correct stiffness and behavior in the overall the “as-built” part to the “as-designed”
cases and a high demand on lightweight assembly. Thus, an optimal design was part. This allows for genuine “First Time
design, it is the perfect fit for demonstration. found that was printed and successfully Right” 3D printing. Time and cost are two
Furthermore, there is a lot of experience used in this year’s formula student season. of the major constraints in wide adoption of
in optimizing this part, as this race series AM today. Typically, with MSC Software’s
officially is an engineering competition that Generative Design solution we find that we
requires to develop a new race car each Summary and Conclusions can cut the time and cost of simulations
year. Other MSC tools such as Adams and by x10. Furthermore, most importantly,
MSC Nastran have been used for this part MSC Software’s MSC Apex Generative with our bridge to manufacturing, we find
in the past for optimization. Design is bridging the gap between design that we can get closer to “First Time Right”
and additive manufacturing. Additive 3D printing. MSC Apex Generative Design
As shown in Figure 3, the development Manufacturing has come a long way technology is here to make the design and
process starts with retrieving the loads by since its inception and is changing the development for AM smarter and more
a multi body simulation based on Adams manufacturing landscape. In order to realize sustainable.
Car. Hereby, the overall suspension is
engineered, including all coordinates for
the connection points, as well as the acting Read an overview in our
forces. This information is used to set
up the optimization model and define its
goals. Therefore, a “design space” as big
Read an Overview in our Previous Issue for More
as possible is added (shown as translucent
Information on Hexagon and MSC’s End-to-End Solution:
material). In this case the overall inner space
www.mscsoftware.com/Engineering-Reality-Summer-2019
of the rim minus the installation space for
wishbones and braking system is selected.
Running the optimization, this material
in the design space is reduced as much

ACOUST IC S
A Smart and
Connected Avio to Ensure
World:
Spatial
Launchers
Structural
Integrity
W
hile the race to Mars
is making headlines,
another ongoing
space race that will
shape the digital era
and the way the world is connected is
getting less attention: satellites.
Satellites are essential for a lot of services

that we use everyday. In the last decades
the Low Earth Orbit (LEO) is becoming
increasingly crowded due to the growing
market demands. In order to meet the
low latency and gigabit connectivity
requirements of native 5G networks which
enable the digital era and connect the
world by delivering broadband access,
there is a need to develop a universe of
multi-orbit satellites.
Operating closer to Earth, they offer

lower latency (delay in a round-trip data
transmission) than any satellite orbit.

Figure 1: Vega Launcher - Mission profile and stages overview - www.arianespace.com
Also Earth observation programmes such as ESA Copernicus aim to to achieve a global,
continuous, autonomous, high quality, wide range Earth observation capacity. This
will provide accurate, timely and easily accessible information to help understand and
improve the management of the environment, understand and mitigate the effects of
climate change, and ensure civil security. In this horizon, Low Earth Orbit (LEO) satellites
are essential. Their proximity to Earth also means that launching such type of satellites is
cheaper and requires less fuel than other types of satellites operating at longer distance.
The Avio VEGA program is a European Space Agency (ESA) program for missions in LEO
and the associated Vega launcher is the ESA’s satellite launch vehicle designed to send
small satellites into LEO. Since its maiden flight in February 2012, of the small, flexible
launcher, Vega successfully placed more than twenty payloads into orbit. In 2014, at the
Ministerial Conference of ESA Member States, the dawn of the next phase for the Vega
launcher was approved, leading to the new configuration of the VEGA launcher called
VEGA C, offering improved performance with a maiden flight scheduled for 2020.
Among the multiple technical challenges overcome by the Avio technical teams, one of
them was to ensure the integrity of the complete launcher, its structure and components to
ensure a safe environment for the payloads. Indeed, any space rocket launcher is subject
to severe external-pressure loading generated by the mixing of the rocket-engine exhaust
stream with the ambient atmosphere. The acoustic load might be critical to the proper
function of the vehicle and its components as induced structural vibrations and internal
acoustic load can lead to malfunctioning of electronic and mechanical components. Figure 2: Vega and Vega C Launchers

Hexagon | MSC Software
Solutions
Avio’s engineers used Avio’s engineers used Hexagon CAE

solutions MSC Nastran and Actran to
Hexagon CAE solutions predict the vibro-acoustic response of the
MSC Nastran and Actran complete launcher structures at lift-off. The
computational approach is straightforward
to predict the vibro- and uses MSC Nastran to efficiently
acoustic response of the compute the mode shapes and related
complete launcher eigenfrequencies of the launcher structure

while Actran is used to model both the
structures at lift-off. acoustic environment and acoustic load.
The VEGA Launcher Upper Part
The Avio VEGA and VEGA-C launchers

are made of multiple stages, each having
a critical role in the success of the launch.
Each part of the launcher has its own
function during the mission and the
payload fairing, also called Heath Shield,
installed on the last stage of the launcher
is responsible of the payload protection. It
is an 7.8-meter-high – 2.6-meter-diameter
shell made of composite CFRP sheets with
aluminum core that protects the payload
from the thermal load, aerodynamic fluxes,
Figure 3: Numerical Analysis Workflow acoustic environment and contamination.
Fairing first Fairing second Fairing higher

bending mode breathing mode order mode
Figure 4: Vega Fairing at ESA ESTEC Large European Acoustic Facility Figure 5: Eigenmode shapes extracted with MSC Nastran
during the Upper Composite Acoustic test

Experimental tests were conducted in environment at launch and numerically acoustic cavities were modeled. The diffuse
ESA ESTEC Large European Acoustic created by superposition of multiple acoustic random field was applied on the
Facility in qualification flight configuration plane waves was applied directly on the surrounding external fluid domain. The main
in order to assess the complete structural structure. This quick and useful numerical benefit of this kind of simulation is that all
response of the launcher upper part strategy provides an accurate insight of the environments, external and internal, are
during the lift off phase. During test, the the structure vibrations at locations where accurately modeled but the large amount of
standard payload was replaced by a structure and acoustic do have weak degree of freedoms involved in the analysis
microphone tree measuring the fairing interactions but it is not able to represent require very demanding solution in terms of
interior acoustic environment. In addition, the internal sound pressure level inside the RAM capability and computational time.
multiple accelerometers were located on acoustic cavity and its interaction with the
the structure. internal structures. The simulations are performed using finite
element model up to 2000Hz in order to
An MSC Nastran Finite Elements model In order to overcome this limitation a explore the high frequency capabilities
of the entire upper stage structure was second solution explored the simulation of this kind of analysis and the different
created by Avio’s engineers, allowing them of the internal acoustic cavities. The finite approaches have proved to be suitable
to compute the structure mode shapes element models of the cavities were added for different phases of a developing
and related eigenfrequencies. The data to the entire structures and two possibilities strategy for new launchers. Also possible
acquired by the acoustic test were used were investigated: a hybrid solution with simulation improvements have been
both for the correlation of the VEGA upper the structure as modal components and identified as, for example, the use of
stage and for the evaluation of different the cavities as physical components; and the Non-Parametric Variability Method
simulation solutions provided by Actran a fully modal solution with both structural (NPVM), a non deterministic approach
software. This study was performed in parts and acoustic cavities as modal in modal frequency responses through a
order to make a tradeoff between the components. The latter solution proved to Monte-Carlo solution framework included
different simulation opportunities provided be more flexible and versatile considering in the Actran software.
by Actran and the required computational also the computational time. The diffuse
time with a view to developing new acoustic random field was obtained
launchers. As a first step, and in order to through random plane wave superposition The New VEGA C Launcher
quickly assess the general behavior of the applied directly on the external surface of
structure submitted to the exterior acoustic the structures. After the validation phase using
load, the fairing interior and exterior the VEGA experimental test, the
acoustic environment was not modeled. As a final simulation, a very detailed acoustic simulation methodology was
A Diffuse Sound Field (DSF) excitation reconstruction of the test configuration implemented in order to evaluate the
with frequency dependent intensity was performed. In this case also the vibro-acoustic response for the new
level corresponding to real acoustic external environment and the internal VEGA C structures. In particular the
Figure 6: External and Internal acoustic Figure 7: Vega Fairing - Vibration levels comparison
domains with the relative microphones and
measurement points.

Figure 8: Vega C General data
analyses were performed for all the

most sensitive parts of the launcher
with respect to the acoustic loads.
The analysed structures are all the
interstages between the solid rocket
motors and the VEGA C upper part in its
different configurations.
A multistage rocket is a launcher vehicle

that uses several stages, each of which
contains its own engines and propellant.
The different stages are separated by
interstage structures that are designed
to allow the connection of the different
stages and their disconnection during
the separation phases. These structures
are very sensitive to the vibro-acoustic
environment because they contain a lot of
electronic equipment needed for the flight.
Figure 9: Vega C - Interstages
The same combination of MSC Nastran
and Actran were used to evaluate the
vibro-acoustic response of the structures
as reported in Figure 10 where the the overall stiffness of the launch system. Both Vega and Vega-C launchers provide great
acceleration spectral density is reported flexibility of mission thanks to the different offered payload adapters allowing to carry up
as an example for the 315Hz 1/3 octave multiple payloads at a time according to different configurations. Vibrations at the payload
band frequency. adapter are strongly impacted by the fairing cavity and related resonances. Accurate
predictions of the payload adapter levels of vibrations require us to consider the acoustic
A payload adapter is a technologically environment as illustrated on Figure 11 where we report the foreseen configuration of
advanced structural element that is able, the Upper Stage of the VEGA C launcher. The internal payloads are considered coupled
despite its very low mass, to withstand with the launcher structure and the acoustic cavities in the analyses. As example is the
important flight load and contribute to different configurations reported in Figure 12 it is possible to recognize(central figure) also

the future space RIDER (Space Reusable Integrated Demonstrator for Europe Return) that Testing New Modeling
is a planned uncrewed orbital spaceplane aiming to provide the European Space Agency Capabilities
(ESA) with affordable and routine access to space. Its expected maiden flight is 2022.
In addition to the above FE approach,
Avio’s engineering team also performed a
SEA analysis of the fairing structure using
Actran Virtual SEA approach. Relying on
the previously created FE models, a SEA
model of the fairing structure was created.
The Virtual SEA approach implemented
in Actran does not require access to any
experimental or analytical expression to
build up a SEA model. It is a very efficient
and affordable technique to extend the
vibro-acoustic analysis of existing FE
models to higher frequencies without SEA
expertise requirements. Furthermore, as
the Virtual SEA approach relies on existing
low frequency FE models, obtained SEA
results are valid at low/mid frequencies
and a smooth transition exists in-between
mid and high frequency results.
Comparison between available

Figure 10: Vega C - Interstage response map measurements and Actran Virtual SEA
results on the fairing structure were
made. In particular, the average vibration
levels over different structure areas were
analyzed. Very good matching in-between
measurements and Actran Virtual SEA
results were observed. It demonstrates the
potential of this new approach to tackle
such analysis and will further be used for
the future launcher structural analysis.
About AVIO
Avio is a leading international group

engaged in the construction and
development of space launchers and
Figure 11: Vega C - Payloads configurations solid and liquid propulsion systems
for space travel. The experience and
know-how built up over more than
50 years puts AVIO at the cutting-
edge of the space launcher sector,
solid, liquid and cryogenic propulsion
and tactical propulsion.
Avio operates in Italy, France and

French Guyana with 5 facilities,
employing approx. 1,000 highly-
qualified personnel, of which
approx. 30% involved in research
and development. Avio is a prime
contractor for the VEGA program
and a sub-contractor for the Ariane
program, both financed by the
Figure 12: Vega Fairing - Virtual SEA analysis European Space Agency (“ESA”)

M ANUFACT URIN G
Car Seat Manufacturer

Cuts Costs and Improves Quality With
Stamping Analysis Tool
By Michael Walters, Advanced Manufacturing Engineer, Adient, Michigan, USA

A
dient is an automotive to continually innovate its offering and Adient wanted to test the feasibility of
company that uses manufacturing processes. With increasing design changes before parts reached
FTI Formingsuite incentives for manufacturers to deliver the production floor to avoid splitting,
COSTOPTIMIZER more fuel-efficient products, Adient’s necking, wrinkling, and other stamping
professional sheet Michigan-based team opted for a new defects. With the company’s previous
metal simulation software to efficiently stamping analysis tool to help design process, a drawing would be signed off by
communicate design changes and reduce lighter car seats while improving processes stamping engineers, but the assessment
rework. to save time and reduce material waste. of its feasibility was largely unstandardized
and based on the engineer’s experience.
According to automotive manufacturer Every year Adient supplies seats and The tooling shops would produce a form
Adient, 1/3 of all cars feature the components for more than 25,000,000 simulation, and if issues were found,
company’s components. As such, cars. But the company’s influence in it would be sent back to the stamping
there’s an inevitable pressure on Adient the automotive market isn’t limited to its engineers, who would in turn send it back
overwhelming representation in today’s to the product engineers.
vehicles. It also places an incredible
responsibility on the company’s designers Besides causing inefficiencies and
AI17 Front Seat Conversation Mode
and engineers to lead the way for contributing to lost time and materials,
tomorrow’s vehicles. this process didn’t maximize the
team’s capabilities for innovation, as
More and more, the automotive industry past experience didn’t offer the full
is under regulatory pressure to boost fuel understanding designers needed to
efficiency and cut emissions. As such, the assess different forming capabilities and
lightness of cars is an increasing focus potential problems that might occur using
for automotive manufacturers, particularly new alloys and materials.
given the growing prevalence of electric
vehicles and consumer concern about
how far these vehicles can travel before Optimizing Manufacturability
needing to recharge. To drive greater fuel
economy in petrol and electric vehicles, Adient’s partnership with FTI stretches
Adient is continually working to make back into the company’s history before
seating even lighter. Since around 2010, its emergence as a spin-off of Johnson
Adient has reduced the weight of its seats Controls, which saved time and money
by between 20 to 30 percent. The current by acquiring a range of FTI solutions,
goal: to produce seats weighing only 10 including the material cost reduction tools
kg (22lbs) or less by 2020. COSTOPTIMIZER, FASTBLANK, and
FASTFORM Advanced.
To achieve this goal, the team at Adient
explored the forming capabilities of higher Adient opted for FTI’s stamping analysis
strength steel and hybrid materials, as tool FORMINGSUITE COSTOPTIMIZER
well as lighter and thinner structures and PROFESSIONAL to evaluate the
mechanisms to lower weight by up to manufacturability of designs, performing
25 percent. But the company’s previous fast blank size calculation, blank overlap
method of determining manufacturability detection, springback analysis and
of new designs was causing problems. determining thinning strain limits.
Using COSTOPTIMIZER PROFESSIONAL,

More and more, the designs are evaluated according to
the automotive three key factors: Forming Limit Diagrams
(FLD), thinning strain, and thickening strain.
industry is under For Advanced Manufacturing Engineers at
regulatory Adient, the FLD is absolutely crucial. FLD is
pressure to boost everything. It’s different for every material.

It doesn’t matter what the thinning or
fuel efficiency thickening strain is; if there are points
and cut emissions. above the FLD major strain line, the part
will fail.

COSTOPTIMIZER PROFESSIONAL’S user-friendly capabilities for interpreting these FormingSuite really allows us to build on
three factors has enabled Adient to drive improvements in the use of materials and that, increasing confidence in our analyses.”
product design. For example, by giving designers precise insight into the percentage
and location of thinning strain occurring in a mounting foot for a seat rack, Adient could “Since introducing FTI, Adient now
increase the part’s thickness from 2.2 to 2.5 mm while reducing thinning strains from 32 requires a passing grade for forming
to 18 percent. simulation on all stampings prior to being
released for production. Forming-related
The digitization and simulation of part manufacturability has been invaluable for Adient in problems during die tryout have been
gaining a greater understanding of parts. “When looking at a blank it might appear normal, practically eliminated, which in turn saves
but the software makes it easy to identify characteristics that you might not perceive with money on the bottom line. In 2018 alone,
the naked eye, like for example blank overlap where flanges are too tall in the corner,” says I performed approximately 1,800 feasibility
Michael. “Previous processes were very much based on drawing upon experience and simulations”, adds Walters.
Countermeasures – The Communication Key
Original Geometry
Use the “Compare Geometry” function to overlay the 2 versions. Revised Geometry
The Counter Measures page is where you can tell the Product Engineer what needs to be changed in order to make the part formable.
Remember that the part must pass all 3 criteria: FLD, Thinning, and Thickening. This part passes Thickening, but fails both FLD and Thinning.

Tolerance Negotiation –
Using Springback Analysis
Knowing how the part will be located

in the check fixture is key to getting a
Countermeasures meaningful springback analysis.
In this example, I have left the B1-

1. Thinning strains
exceed the maximum B2 and C datums unfilled, as these
allowable - risk of will be pinned in the fixture. The
splits. Decrease
surface below these datums will
depth of bead.
create the A1, A2, & A3 datums and
will be clamped on them.
It was clear from the analysis that the the software’s intuitive countermeasures
bead was set too deep into this material page enables users to communicate to the
at the ends (490XF, 2.5 thick) product engineer important changes that Now the results look more realistic!
need to be made. “We now have a stronger
By simply reducing the depth of the bead at communication process and can ensure Here’s where the tolerance
the ends, the part now passes all 3 criteria. that designs that have issues don’t go to negotiations begin.
production,” says Michael. “By catching and
FORMINGSUITE’S communicative fixing problems early on, significantly fewer Let’s say the Engineer wants a 1mm
capabilities has helped Adient avoid making parts are coming back from our tool shops profile around the hole at the top
parts that are unfeasible, saving time and and we’ve seen an encouraging reduction of the curl. You can easily prove,
resources and reducing bottlenecks. Using in waste and reworks.” with this data showing 1.87mm
springback that it is not possible –
you need at least a 2mm profile.
Or take the forward-most hole

in the side of the bracket. If the
With a more streamlined manufacturability assessment process, Adient is firmly in the driver’s Engineer wants a 0.5mm profile, it
seat to continue solidifying its position as one of the leaders in automotive seat production. may not be possible. Better to ask
for at least 0.75mm profile.

M AT ERIA LS
Fast and Accurate

Additive Manufacturability Analysis
By Yvan Blanchard, Coriolis Composites

Anthony Cheruet, e-Xstream Engineering
T
his article focuses on the design optimization of complex 3D composites
structures made by additive manufacturing processes.
There are commercial CAD-CAM software solutions for detailed offline

path programming, but there is a growing need for innovative tools and
methodologies for trade off studies very early in the design stage. A new
innovative solution has been developed on top of the CATFIBER© software, allowing both
designers and stress engineers to quickly analyze complex double-curved geometries. It
also includes a variable stiffness approach with tow-steering, and structural analysis of the
manufacturing defects using Digimat© software.
Design Analysis Framework Description
Today, more and more CFRP structures are manufactured by automated processes such
as fiber placement robotic systems (Figure 1).
The design trade-off analysis can be done on a simplified quasi-isotropic laminate (with full
plies), in order to just analyze the surface curvature impact, independently of plies shape.
But thickness effects, material excess, staggering rule, part productivity rate, could not be
correctly estimated.
The engineering and manufacturing requirements may quickly interfere, and a difficult
compromise between feasibility, strength and cost needs to be found, especially with
double-curved layup surfaces.

them test many combinations, such Structural Strength Analysis
as material width, maximum number
of tows by course, and maximum To verify the structural strength of the
fiber deviation angle. The solution optimized Layup design proposal, it
should be able to analyze complex and is important to have a quick and easy
representative laminates such as large transfer of all the as-manufactured
aerospace panels with double curvature composite properties onto a structural
and hundreds of plies. mesh used for sizing purposes. This
mapping is done using the Digimat
An automatic ply splicing algorithm, Platform© and concerns the transfer of
based on both engineering and
manufacturing requirements, allows
to quickly and easily generate a
manufacturable design proposal. This
algorithm also uses a patented rosette
transfer feature, allowing steered-path
propagation and then variable-stiffness
modeling (Figure 2).
Ply course centerlines and splice cuts

are first computed, then the ply boundary
is filled with (tow) strip surfaces. This
allows us to capture all the process and
material specificities, such as triangle
Figure 2: Steered-paths on wing skin surface.
of gaps, tow overlaps, minimum course
length (MCL), and minimum distance
between tow cuts (Figure 3).
The design analysis system is also

able to compute and output several
Figure 1: Automated Fiber Placement robotic manufacturing cost indicators, to help
system (Coriolis Composites) designers sort the manufactural design
proposals (number of courses by
sequence, number of tow cuts and drops,
buy-to-fly ratio).
Moreover, such analysis systems are not
able to reach a good compromise between The design analysis tools were
performances, level of details and results implemented on top of Coriolis
accuracy. The expected computation Composites CATFIBER© Offline
time is a few seconds to a few minutes so Programming solution, through a
that several analysis runs can be done in dedicated infrastructure made of scripts
parallel within a few hours only. (Python or Visual Basic) and a Microsoft
Excel© spreadsheet. This allows us to
Figure 3: Plies build-up with tape
Designers and stress engineers need easily launch several background runs courses covering.
robust tools and methodologies to help from a very light user interface.

the as-manufactured fiber orientation, References
the exact location of the gap and
Wu KC. “Design and analysis of tow-steered
resin-rich area. Using Digimat, such
composite shells using fiber placement”.
information can be transferred In: Proceedings of the American Society for
automatically to the Finite Element Composites 23rd Technical Conference.
Model used for sizing activities by Memphis, TN, USA. 2008, doi: 10.4271/2012-
01-0082
Stress Engineers. In this case, Digimat
Fayazbakhsh, K., Arian Nik, M.,Pasini, D.,
can generate the composite layup Lessard, L., Defect layer method to capture
command cards with the local as- effect of gaps and overlaps in variable
manufactured fiber orientation for stiffness laminates made by Automated Fiber
Placement, doi: 10.4271/2012-01-0082
each ply. In addition, the effect of the
G. Gonzalez Lozano, A. Tiwari, C. Turner
gaps on the local stiffness can be A design algorithm to model fibre paths
handled in two ways, depending on the for manufacturing of structurally optimized
reconsolidation process. The first one composite laminates. In: Composite Structures
(2018), doi:10.1016/2018.07.088
considers that the gaps have an effect
on the local thickness while the second
one considers that the gaps are filled by
resin and affects the local fiber volume
fraction of the composite. Using a
micro-mechanical model of the material,
this local variation of the fiber content
Digimat can is computed at each Gauss point of the
generate the FE Model (Figures 4 & 5).
composite layup
command cards Figure 4: Thickness map analysis
(quasi-isotropic laminate) with tow gaps
with the local as- capture with Digimat© software.
manufactured
fiber orientation
for each ply.
Figure 5: Local stress field per ply and effect of gaps computed by Digimat©
For Details About CADFiber Standalone©, or CATFiber©

for CATIA Solutions: www.coriolis-composites.com
For Details About Digimat©:

www.mscsoftware.com/product/digimat

Daimler Trucks
Accelerates
Product Development Cycle
with 10X Faster MSC Nastran Performance
By Tim Rzesnitzek, Senior Analysis Engineer,

Daimler Trucks North America
Bhoomi Gadhia, Product Marketing Manager,
MSC Software
D
aimler Trucks manufactures and sells commercial trucks and buses as well
as performs services and maintenance of their products. Daimler strives for
efficiency and innovation, and the Daimler Trucks North America development
team is applying computer-aided engineering (CAE) techniques to the truck
development process to help the company achieve these goals.
Truck manufacturers must ensure the safe carriage of payloads while simultaneously light-
weighting their vehicles to improve fuel efficiency and associated emissions, and the ability
to predict the fatigue life of the vehicle components is extremely important. Predicting
fatigue life accurately in the early stages of the design and development cycle enhances
product life, reduces testing and prototype costs and accelerates time to market.
MSC Nastran is the crucial link in our durability analysis workflow to perform finite
element analysis. Our team has selected MSC Nastran because we conducted a market
evaluation and found its performance was consistently 10X faster compared to competing
finite element software products.
Workflow:
Multibody Simulation Time-domain Finite Element Analysis

Fatigue Analysis
Forces Extraction Loading Stress/Strain
Figure 1: Durability Analysis Workflow at Daimler Trucks NA

Proving Ground Duty Cycle
Figure 2: Simulation of all durability events of physical test according to duty cycle
We perform multibody simulation in order

to accurately replicate the dynamic and
non-linear bushing behavior and to capture
the road input by using sophisticated,
elastic tire models together with 3D road
profiles. All durability events of the physical
test program are simulated in full length
according to the duty cycle. Different duty
cycles are used dependent on the vehicle
type and the targeted customer profile
(e.g. highway vs. extreme duty application).
Finite element methods are used in MSC

Nastran to compute the stress distribution
in the vehicle substructures, like chassis,
cab or hood. In the FE simulation, the Figure 3: Substructure of a truck - with input loads from MBS simulation
substructure is excited in all interface
locations with the loads retrieved from the
Multi-Body simulation.
The stress results of the FE analysis are

used to calculate damage for almost every
metal component and selected plastic
parts. Finally, we receive a distribution of
the combined damage for a complete duty
cycle as well as the damage contribution
of every durability event.
Figure 3: Simulation of sub-structures with input loads from MBS simulations

The FE analysis of a particular substructure
is approached in two steps. At first, we
perform a normal mode analysis in order to In the second step we run a transient modal analysis for every durability event, which is
compute real modes, residual vectors and set up as a restart run based on the normal mode analysis. A ‘restart’ is a logical way
modal stresses. In addition to the usual of continuing from a previous run without having to start from the beginning. In dynamic
result data we keep the MASTER/DBALL analysis, the calculation of normal modes is, in general, the most expensive operation in
and IFPDAT files. This analysis has to be terms of computation time and cost. Using the restart option, the normal modes do not
executed once for a specific model. have to be re-calculated for every durability event. This does not only save computation

SOL103 SOL112
• Computation of • Restart based on SOL103

eigenmodes, residual • NO re-computation
vectors and according Keep DBALL
MASTER IFPDAT of eigenmodes
modal stresses
• Computation of modal
• To be performed once participation factors only
for a specific FE
• To be performed for every durability event (track)
time but also ensures consistency

regarding the modes of the eigenmode
analysis and the participation factors of
the several durability events.
The main result of the second step is the

computation of the participation factors
for every durability event. Additionally,
we also export the interface loads which
are applied to the FE model as well as
the displacements of selected nodes.
The interface loads and the nodal
displacements are used to animate the
time-dependent deformation in response
to the applied interface loads.
Once the transient response analysis

is performed, we use MSC Nastran
Figure 4: Transient Response Analysis on Substructure of a Truck with Restart
results to derive stress time-history
by superposition of modal stress
and participation factors in nCode.
E-N analysis for metal parts and S-N
analysis for plastic parts is performed
for every durability event. Through
these analyses we achieve an accurate
prediction of the combined fatigue
damage according to the duty cycle.
This completes our workflow.
Solution Time Using MSC Figure 5: Time reduced using MSC Nastran
Nastran and Competitor Finite
Element Solvers The validation of a full vehicle incorporates the analysis of the chassis, cab and hood
structure. By performing these analyses with MSC Nastran we reduce the overall
When we compared the solution time for turnaround time by a couple of days. With that, the use of the MSC Nastran solver is
normal modes analysis using MSC Nastran crucial to perform our durability analyses.
and the competing finite element solver we
had been using, we observed that MSC
Nastran can solve this model in 5 hours Summary
whereas our previous finite element analysis
product took over 50 hours to perform the The Durability process at the Daimler Trucks North America has been outlined above. The
same analysis using the same machine durability process has enabled identification of high stress locations more effectively. We found
environment. Normal Modes Analysis (SOL the overall process to be nearly ten times faster than the previous method. With fewer and
103) is 10 times faster using MSC Nastran. smaller intermediate files, the bookkeeping is also much simpler. We have successfully validated
this methodology based on several parameters and it has exhibited a good level of correlation
The performance issues using the former with our physical test data. The Daimler Trucks North America team see several opportunities
finite element solver were mainly related to in running similar simulations in the near future. MSC’s robust offering of MSC Nastran solver,
the modal stress computation of second continuous support from field team as well as continuous investments in the development has
order solid elements. made a significant difference in simplifying and speeding up our durability analysis.

C O -S IMULAT IO N

Adams Marc
Co-Simulation of a
Washing
Machine
By Seung-Oh Kim, Samsung Electronics and
Hemanth Kolera, MSC Software
JD Power & Associates ranked Samsung such as water levels, temperatures,

#1 in Customer Satisfaction in 2019. loads and tub motion. The rotary motion
of the drum creates a flow of detergent
A washing machine is an essential and water which cleans the load and
component of our day-to-day activities. also extracts the water during the drying
Advances in washing machine design phase of the cycle. The tub is suspended
have included increases in capacity and inside a housing using a suspension
spin speed. The industry-wide machine system, designed to limit the propagation
weight and cost reduction initiatives of vibrations from the tub to the housing
conflict with these design initiatives. and from the housing to the floor.
Reducing vibrations in washing machines Modeling and analysis of the washing

is a challenge. During spin cycles, machine dynamics is a considerable
S
significant imbalanced centrifugal forces challenge as the washing machine
amsung Electronics is are generated, which then result in comprises of several highly nonlinear
one of the world’s largest vibration and shaking. Elimination of such elements such as friction dampers,
electronics manufacturers vibrations is critical to the design of silent rubber feet, rubber bushings etc.
and has a presence in over washing machines and the best possible
80 countries and employs user experience. Samsung has devoted resources to
around 308,745 people. As of June 2018, develop methods that can model and
Samsung Electronics’ market cap stood A modern washing machine is a complex predict washer dynamics. The model-
at US$325.9 billion. The 2018 Which? multi-body system comprising a drum, based insights are then used to develop
Awards recently announced Samsung suspended to a tub with bearings in one technology that can reduce vibrations in
Electronics as the leading Large Appliance or both ends. Operation of the washing the washing machine. Samsung’s roadmap
brand for its range of home appliances. machine is actively controlled based on for washing dynamics analysis starts
For Front-Load and Top-Load Washers, sensors that gauge operating parameters with first developing credible models that

Tub
Front Frame
3
Figure 1
can accurately capture the fundamental The washing loads are transmitted can accurately model the system,
physics. This involves accurate modeling from the gasket to the springs damper, providing realistic boundary conditions
of the spring, damper and diaphragm suspension system and then to the main for the non-linear components. A co-
characteristics and material properties of frame. simulated model would simulate much
the elastomers. At the next level, analysis is faster than a complete model in non-
used to optimize the design for noise and linear FEA without compromising on the
vibration. The ultimate goal is to mimic the Adams-Marc Co-Simulation accuracy. Current diaphragm simulation
actual laundry exercise. approaches at Samsung that did not
We used an Adams-Marc co-simulation leverage co-simulation between Adams
Front loading washing machines have capability to include geometrically and and Marc did not capture the actual
a large grey gasket or diaphragm seal materially nonlinear structural behavior systems dynamics and had limited value
around the door. This gasket is what in multibody dynamics (MBD) simulation. to the design process.
keeps the water from flowing out of the Any Adams model and any Marc model
washer when in use. One side of the can be used in co-simulation with this The Adams models of the washing
diaphragm is connected to the frame and tool. Post-processing is done separately, machine dynamics include models for the
the other side to the tub. Besides leak Adams results in Adams, and Marc results spring, the damper and the diaphragm.
proofing, the gasket provides vibration in the Marc postprocessor. The spring is modeled as a preloaded
insulation between the frame and the tub spring element with tensile force, while the
and also prevents the washing machine A co-simulation between an MBD damper is modeled as a two-stage non-
from jamming (See Fig.1). A cross- solution like Adams and a Non-linear linear damper element. The diaphragm is
sectional view of the washer is shown. FEA model such as Marc provides modeled as a bushing element, and the
Section 1, provides the seal between the several benefits. The non-linear FEA can primary stiffness is calculated based on
door and the glass. There is no contact accurately describe non-linear behavior the Marc analysis (Figure 2).
between section 3 of the diaphragm and of flexible components, including plastic
the rotating drum. The gasket design has deformation, non-linear material, large The Marc model captures the nonlinear
to ensure that there is no contact or wear deformation of the components, buckling, material properties of the rubber diaphragm
in the middle section 2 of the gasket. self-contact etc., while the MBD model using empirical models. Various constants
Figure 2
Spring
Diaphragm
Damper
Assy Frame Front Assy Tub & Drum Assy Main Frame

Figure 3
in the empirical model are calculated using can be executed in Marc to gauge the (v18.1) is established using MSC CoSim
material test results. The workflow for natural frequencies and the diaphragm v1.6. The MSC CoSim engine has been
capturing the nonlinear material properties is and estimate the level of vibration. developed in order to provide a co-
as shown (Figure 3). Non-linear static analysis can be used to simulation interface for the direct coupling
predict the equivalent stiffness and any of different solvers/disciplines within a
Material data is gathered using tests such contact. The primary stiffness calculated Multiphysics framework. The .ADM and
a simple tension, planar tension, Eg. from the Marc model feeds into a rigid .ACF files from Adams and the .DAT file
Biaxial Tension and Shear. The gathered Adams Multibody dynamics model which from Marc are imported into the CoSim
data is processed to extract various can predict the dynamic acceleration/ interface.
model coefficients required to characterize displacement and the load transfer to
non-linear material models such as the frame. Besides computing the loads The Adams analysis conditions included
Mooney, Ogden etc. The data fitting can on the washer frame, an Adams-March an unbalanced laundry load of 500g
be prioritized to capture the dominant co-simulation can be executed to run which was used to trigger the vibrations.
strain and strain levels. The characterized transient dynamics simulation to evaluate Maximum motor RPM was set to 1300.
material models are then used to run deformation, contact and buckling. Two Marc cases were evaluated, on two
various analysis types in Marc as shown different diaphragm geometries, with and
in Figure 4. Linear harmonic analysis The coupling between Adams and Marc without side ribs.
Figure 4
Step 1: Step 2: Step 3: Step 4:

Rubber Material Test Data Reduction Curve Fitting Check Dominant
(Simple Tension, Planar Tension, (Cycle Selection, (Mooney, Ogden …) Strain
Eq. Biaxial Tension & Polynomial Fitting, (Select Strain Level)
Simple Shear) shift)

Co-Simulation has
helped Samsung gather
design insight at both
the system and
component level.
Side Rib
Figure 5
It can be seen from the co-simulated

results (Figure 5) that the tub acceleration/
displacement increases, while the frame
front force decreases compared with the
bushing model.
Design decisions can be made based on

direct comparisons of simulated metrics
such as total strain and contact normal
forces for the two gasket designs.
Exploration of the washing machine

dynamics via co-simulation has helped
Samsung gather design insight at both
the system and component level. The next
steps in Samsung’s co-simulation journey
involves more in-depth simulation-based
exploration for vibration reduction and
robust design using the most optimized
set of Co-sim analysis conditions.

AEROSPACE SOLUTIONS
SMARTER AIRCRAFT ENGINEERING FROM CONCEPT TO REALITY
STRUCTURAL AND MULTIBODY MATERIALS MODELING

DYNAMICS SIMULATION Allows composite material and structure
Enables lightweight structural design modeling.
and multibody dynamics analysis.
ACOUSTIC SIMULATION FLUID AND THERMAL DYNAMICS

Helps reduce aircraft noise and Improves flight efficiency and optimizes
analyze vibrations. passenger comfort.
WWW.MSCSOFTWARE.COM/AEROSPACE

MSC Nastran
Speeds Forward using
Graphics Processing Units
By Niveditha Krishnamoorthy,
Development Relations Manager, NVIDIA
Bhoomi Gadhia, Product Marketing
Manager, MSC Software
A
nalysts in today’s environment are working on increasingly larger and
complex models. With advances in electric car manufacturing and quieter
electric motor engines, high-frequency Noises and Vibrations are becoming
more prominent, which makes these calculations exponentially more
expensive.
As computing technology improves, the industry is leaning towards analyzing real-world

scenarios more accurately with the use of Finite Element (FE) Analysis to support these
high-fidelity models.
Current trends in High-Performance Computing (HPC) are moving towards the use of
multicore processor architectures to achieve speedup through the extraction of a high
degree of fine-grained parallelism from the applications. This hybrid computing trend is
led by NVIDIA Graphics Processing Units (GPUs), which have been developed exclusively
for computational tasks as massively parallel co-processors to the CPU. Today’s GPUs

can provide memory bandwidth and of enterprise-grade, high-performance Why Use GPU?
floating-point performance that are systems, and software, with high-value
several factors faster than the latest services to help manufacturers implement Most customers ask why they should
CPUs. To exploit this hybrid computing MSC Nastran throughout the value buy a GPU. It is generally beneficial when
model and the massively parallel GPU chain and product lifecycle. Many of models are mainly composed of solid
architecture, application software will these systems are already certified by elements, and the simulation is dominated
need to be redesigned. MSC Software and deliver unparalleled by matrix factorizations, not like in Normal
productivity, performance, and flexibility. Modes Analysis (SOL 103), where there
MSC Software and NVIDIA engineers are lots of modes, and FBS dominates.
have been working together on the use GPUs have been a significant part of Furthermore, each user should evaluate
of GPUs to accelerate the sparse direct HPC for decades, and MSC Nastran whether multiple CPU cores can give
solvers in MSC Nastran for the last has been on the market for more than them the same benefit as a GPU while
several years. We will dive deeper into 50 years. There had been attempts to keeping in mind that GPU usage can free
the recent GPU computing developments utilize GPUs in MSC Nastran in the past, up some of the CPU cores for use by
in MSC Nastran in this article, including which have not led to much success. other applications/users.
the support of NVH solutions with One of the reasons was the architecture
Fast Frequency Response (FASTFR), of MSC Nastran, which relied heavily on A primary use of GPU is to offload
and Matrix Multiply and Add (MPYAD) the disc storage of the data and having massive computational tasks from CPU.
modules development. Representative a minimum memory footprint. Another GPUs will not be useful for small models
industry examples will be presented to reason was the memory limitations and that operate on a small amount of data,
demonstrate the performance speedup the overhead caused by the data transfer do a large number of I/O operations, or
resulting from GPU acceleration. between the host and the GPUs. In the iteratively solve small sets of equations.
last few years, GPUs have accelerated Every time a GPU is invoked, it comes
The resulting rapid CAE (Computer-Aided performance – NVIDIA has added more at a fixed cost of, typically, a fraction of a
Engineering) simulation capability from cores and memory specifically for the FE second, 0.1 – 1 sec. So, if you are trying
GPUs has the potential to transform applications. This has helped specifically to use a GPU to compute a product of
current practices in engineering analysis with NVH simulations. two 100 x 100 matrices, which takes 0.01
and design optimization procedures, seconds on the CPU, you will pay a 0.1 –
enabling manufacturing OEMs to deliver To accommodate the GPU usage, 1-second penalty for just using the GPU
their products to market at a much faster MSC Nastran has evolved to maximize before you even start transferring the data
rate than is currently possible with CPU- memory usage, and the GPUs have between the host memory and the GPU.
based solutions. become increasingly faster, and their on- This will not benefit your simulation.
board memory has reached the tens of
Working jointly with NVIDIA provides gigabytes mark, they are finally ready for However, if you are dealing with a large
MSC Software a comprehensive portfolio successful symbiosis. model that requires operations on
CPU GPU
Leveraging the GPUs

for dynamic analysis
using MSC Nastran
2019 at Volvo opens = SPEED UP
for substantial savings
in run time (up to 50%
faster) for the runs
extending to high
frequency.

matrices with dimensions of the order steps. Also, the schematic of a CPU
of 10000 x 10000 or more, then the with an attached GPU accelerator is
GPU will come in very handy. In fact, for such that CPU begins/ends job, GPU
dense matrix operations, every additional manages heavy computations. This gives
GPU is roughly equivalent to having an unparalleled productivity and performance
extra CPU with 20 cores. By splitting with NVIDIA GPUs.
the workload between the CPU and one
GPU, you can achieve approximately a
factor 2x speedup. When splitting the Examples
workload between the CPU and two
GPUs – factor 3x speedup, and so forth. When you deal with sparse matrices
For sparse matrix operations, every GPU and sparse BLAS operations, the work
is roughly ten times faster than the CPU, partitioning logic is simpler, since all work
so all sparse BLAS operations should be will be handled by GPUs only. In case
handled by GPUs alone. you are using one GPU, a typical large
sparse BLAS operation will be about 3x
Similar to the CPUs, GPUs have multiple – 10x times faster. Since most of these
Figure 1 cores that can be used for parallel operations are scalable, the more GPUs
computations, and they also have parallel you use, the greater the speedup. With
“streams” that can perform parallel data two GPUs, you get 6x – 20x speedup,
transfers between the memory of the host and so on. Specific speedup values are
machine and the GPU memory. Every time strongly dependent on the hardware, on
a GPU is invoked in a computation, all the the size of the input matrices, and the type
input matrices need to be copied from the of operations.
host memory into the GPU memory, and
the resulting matrix needs to be copied Figure 1 represents an example of
from the GPU back to the host. By using performance improvement for dense-
multiple streams in parallel, the cost of sparse matrix multiplication by adding one
these data transfers can be significantly or two GPUs. Input matrix parameters:
reduced. However, it would be best if you matrix A 56,000 x 192,000, 100% dense;
considered that each GPU stream could matrix B 92,000 x 192,000, 1.5% dense.
only communicate to a dedicated CPU Blue bar: 20-core Intel Xeon CPU; red bar:
core. So, when you are performing some single NVIDIA Tesla K40 GPU; green bar:
computations involving dense matrices, two NVIDIA Tesla K40 GPUs.
precise balance must be reached between
how many CPU cores are included in Figure 2 represents performance
calculations, and how many are dedicated improvement for a dense matrix
to communicating with the GPU streams. multiplication benchmark: 50,000 x 50,000
Figure 2
dense matrices. Blue bar: 20-core Intel
Rule of the thumb is: the number of CPU Xeon CPU; red bar: 20-core Intel Xeon
threads/cores should always be larger than CPU + two NVIDIA Tesla K40 GPU; green
the number of GPUs you are trying to use. bar: 20-core Intel Xeon CPU + NVIDIA
For example, in a scenario where you try Tesla K40 + Quadro GV100 GPU; purple
to compute something using two GPUs bar: 20-core Intel Xeon CPU + two Nvidia
and just four CPU cores, the workload will Tesla V100 GPU.
be divided between the GPUs because
the number of the CPU cores is too small It is possible to use several different GPU
to compete with the GPUs. In a scenario models on the same host system, with
where you try to use two GPUs and 20 varying sizes of memory and efficiencies.
cores, four cores will be assigned to each Every time Nastran uses GPUs for BLAS
GPU for parallel streaming of data, and operations, it will check the elapsed times
the remaining 12 cores will be used for the on each GPU and CPU and balance the
computation. workload in a way that would minimize the
elapsed time across all devices. This way,
MSC Nastran GPU acceleration is the fastest GPU/CPU gets most workload,
user-transparent, meaning jobs launch while the slowest GPU/CPU gets the least
Figure 3 and complete without additional user workload.

The current implementation of GPU GPU; green bar: 20-core Intel Xeon CPU +
support is using CUDA Toolkit by NVIDIA two NVIDIA Tesla K40 GPUs.
and only works with NVIDIA GPU devices.
The threshold matrix size is set to 4000 Figure 6. Performance improvement for
for dense matrices and 500 for sparse model from Fig.4 (2000 frequencies, 73,600
matrices. Matrices are partitioned among eigenmodes) by adding two GPUs of
the CPU and GPUs and, depending on the different types. Blue bar: 20-core Intel Xeon
memory availability, may be processed in CPU; red bar: 20-core Intel Xeon CPU +
one or multiple passes on each device. two NVIDIA Tesla K40 GPU; green bar:
20-core Intel Xeon CPU + two NVIDIA Tesla
Figure 4
Below, we shall consider several examples V100 GPU.
from the automotive industry to highlight the
usage of GPU accelerate the solution. All
models are SOL111 with varying number Key Takeaways
of degrees-of-freedom (DOF), frequency
range, number of eigenmodes, and load GPUs allow speeding up large
cases. Note how the performance improves computationally intensive MSC Nastran
with an increasing number of frequencies operations by factors between 2 and 20,
and eigenmodes. The reason for this is compared to the elapsed time on CPU
that most of the computationally intensive alone. Depending on the percentage
operations are proportional to the number of these operations in the total elapsed
of eigenmodes to the third power and time of the entire simulation, the results
the number of frequencies. This is where obtained can be several times faster than a
the GPU makes the impact. In all the simulation that is run only on CPUs.
simulations shown below, the GPUs were
actually used only in a couple of modules. The larger the simulation, the larger
Yet they managed to cut down the entire the speedup. For the same model, the
simulation elapsed times by 15-35%, or frequency response analysis over the
between 0.4 – 4 hours. frequency range of 1500 Hz and 50,000
Figure 5
modes improves from 11.3 hours to
Figure 3 shows an example of SOL111 7.2 hours by adding 2 GPUs. With the
performance improvement by adding a frequency range increased to 2000 Hz and
single GPU. The model: 20.5M DOF, 2050 73,000 modes, the elapsed times on CPU
frequencies, 30,000 eigenmodes, 4 right- alone and with two Tesla K40 GPUS are
hand sides. Blue bar: 20-core Intel Xeon 30.3 and 18.1 hours, respectively.
CPU; red bar: 20-core Intel Xeon CPU +
NVIDIA Tesla K40 GPU. Newer GPU models are several times
faster than older models. For dense BLAS
Figure 4 represents the car-body model operations 5-year old Tesla K40 model
with 43.5M DOF provided by Volvo Car is roughly equivalent to adding an extra
Corporation used for benchmarking. CPU to the host machine. Adding a newer
Quadro GV100 or Tesla V100 model is
Andrzej Pietrzyk, Method Development approximately equal to adding CPU that
for NVH CAE at Volvo Car Corporation, is 2-3 times faster. It is possible to split the
says, “Leveraging the GPUs for dynamic workload dynamically between different
analysis using MSC Nastran 2019 at Volvo GPU models and get the best performance
opens for substantial savings in run time out of all available devices. For sparse BLAS
(up to 50% faster) for the runs extending operations adding a Tesla K40 GPU leads
to high frequency. This is definitely a highly to a speedup of 10x on average. Adding
promising development path for MSC a Tesla V100 leads to a speedup of ~20x Figure 6
Nastran.” – 30x compared to the elapsed times for
the same operations handled by Intel MKL
Figure 5. Performance improvement for kernels on CPU.
model from Fig.4 (1500 frequencies, 50,000
eigenmodes) by adding one or two GPUs. Acknowlegdgement: We would like to thank
Blue bar: 20-core Intel Xeon CPU; red bar: Illia Silin who contributed greatly towards
20-core Intel Xeon CPU + NVIDIA Tesla K40 this article.

M AT ERIA LS
US Army Use of
MaterialCenter for
Metals Additive
Manufacturing Data
Management
Based on an Interview with the United States

Army Combat Capabilities Development
Command Armaments Center
T
he United States Army Combat Capabilities Development Command (CCDC)
Armaments Center is the US Army’s primary research and development arm
for armament and munitions systems. It is a leading defense facility for Additive
Manufacturing (AM) of Metals and is located in New Jersey. Armaments Center
has been investigating AM for a number of years now with programs aimed at
exploiting the novel capabilities of additive manufacturing. The facility has a number of AM
systems at their disposal including a laser powder bed fusion EOS M290 machine that prints
in Steel (4340/4140/17-4), Inconel, and Cobalt Chrome; and an E-Beam system, an ARCAM
A2X machine that prints in Titanium, Inconel, and Cobalt Chrome. In addition, there is access
to a wide range of support and testing equipment for powder synthesis (Plasma Reactors,
High Energy Mills), post processing (HIP, Heat Treatment, Surface Finishing), machining in
a full machine shop (EDM, CNC, etc.), testing (Tensile, Charpy Impact, Hardness), and part
characterization (Scanning Electron Microscopy, Particle Size Analysis, X-Ray Fluorescence &
Diffraction, Oxygen/Nitrogen Analysis).
Armaments Center is interested in using AM equipment to prototype, develop, and fabricate

metal parts via a layer by layer powder bed laser sintering process. AM has the potential to
provide a wide range of design flexibility over traditional manufacturing methods allowing for
rapid prototyping, part weight reduction, novel part design, reduced time to product, and
overall manufacturing flexibility. The benefits of AM include a reduced logistics footprint and
time-to-field for replacement parts, manufacturing options to reduce single point failures, and
creation of novel and improved part designs for reduced weight while meeting or exceeding
performance requirements. In turn AM results in a manufacturing process for providing parts
on rapid response, and on-demand basis.

Armaments Center has identified six qualified for part acceptance for use
practical areas of interest for additive in armament systems meaning design
manufacturing technologies in the US and manufacturing process data
Army (see Figure 1): required to support repeatable additive
manufacturing production must be
1. Novel Materials: Novel powder defined. Several examples of additively
synthesis for Non-COTS materials, manufactured armaments produced are
2. Rapid Prototyping: Multiple build shown in Figure 2.
iterations on the same build plate for
design optimization. Small runs for
prototype testing,
3. Replacement Parts: Investigating
component replacements which
match properties but can be delivered
in an accelerated timeframe,
4. Novel Designs: Investigating novel
weapons systems components with
designs difficult or impossible with
traditional machining,
5. Rapid Fielding: Investigating Additive
Technologies to overcome the
challenges of bringing metals additive
to the field, and
6. Process Monitoring: Working to
develop custom In-Situ Monitoring
Hardware which can be retrofit on
existing equipment.
For AM benefits to be fully realized, Figure 1: Additive Manufacturing Areas of

processes must be developed and Interest to the US Army
Figure 2: Metals AM Build Examples for the US Army at CCDC Armaments Center

What is the L-PBF and the constraints for AM. Today,
Quality Strategy? widespread adoption of the AM process
is limited due to a combination of these
CCDC Armaments Center at Picatinny, NJ, many challenges. Finally, with respect
have focused on additive manufacturing to utilization of Digital Product Data in
process development for AM materials and AM, a system for controlled electronic
parameters along with the development data management and sharing must be
of Quality Assurance provisions and implemented - software types used and
requirements to develop manufacturing digital file control must be set prior to
guidelines for robust and reliable new manufacturing initiation.
build L-PBF components. To do this, test
components for new build demonstration
and testing were selected. The team CCDC Armaments Center
wanted to establish a process for Additive Manufacturing
qualification and certification of AM Benchmark Demonstration
components, then transition the process
on to internal government facilities and the A CCDC Armaments Center goal is to
AM industry with a manufacturing guide. qualify powder bed fusion AM technologies
In addition, the US Army wanted to share as a viable alternative manufacturing
knowledge of the additive manufacturing process to fabricate armament systems
process and create a knowledge base of components. To do this, multiple areas
AM products aligning to their roadmap. in the total manufacturing process need
developmental efforts addressed to them to
be able to produce an accepted additively
What are the Challenges to the manufactured component. An additive
Figure 4: Selection of some of our 325 AM
Benchmark samples and parts Use of Additive Manufacturing manufacturing benchmark testing method
in the Military? was devised using 4340 Steel Powder
due to its chemistry, particle size, and flow
First and foremost, part acceptance for US characteristics. AM processing parameters
DoD (Department of Defense) applications were developed focusing on energy density
relies on a process for qualification and ranges, and a DoE (Design of Experiment)
certification. However, the relationship was established that looked at the following
between AM materials properties, parameters over 325 samples that were
processing parameters, and component fabricated (see Figure 4):
performance are extremely complex,
and complicated further by unique part 1. Laser Power
geometries. There is also an extremely large 2. Scan Speed
pool of materials and AM equipment to 3. Hatch Distance
choose from, raw materials must be readily 4. Energy Density Range
available and trusted to manufacturer or
internal specifications, processing condition The resultant AM parts were evaluated
windows must be defined to ensure part based on microstructure, density,
quality, In-Situ Monitoring technology must porosity, and hardness. Their mechanical
be utilized and improved upon, and a properties were compared to wrought
recognition that technology advancement steel after stress relief, quench and
might introduce previously unforeseen temper heat treatment.
manufacturing variables.
For each mechanical test specimen, four
It is fair to say that AM standards are still identical consecutive builds were printed
in development. There is a clear need to assess process reliability. The team
for continuing collaboration between focused on variations in location and
academia, industry, government agencies, orientation within and across builds, while
and others to push standards adoption. collecting tensile, hardness, density, and
Moreover, with respect to design for toughness data. The team normalized
Figure 5: Benchmark Steel 3d Printing Test AM there is a need to educate and samples with heat treatments per the
showing Locations of Parts of the Base Plate inform part designers of new principles AMS 2759 standard.

Figure 6: Benchmark AM Printing Sample Mechanical Property Results for the different print Figure 7: Six AM Printer Machine results for the
locations in Figure 5 same set of parts being printed the same way
AM Benchmark Results and • Equipment chosen included an will occur. Complex data sets can be
Lessons Learned EOSM290, ProX320, SLM, and the generated from even a single build.
EOSM280 Hence, data storage solutions are needed
The tests indicated that parts printed • 4340 steel powder was procured from where process monitoring solutions
in the XY direction had 12% higher a single lot to minimize variance require large file storage spaces and
elongation values than parts built in the Z • A manufacturing guide was written bandwidth. In effect, AM processing
direction. Ultimate Tensile Strength (UTS), and disseminated to all participants pedigrees are required. There is a need
Density, and Hardness values matched outlining all major aspects of the for historical records of print builds to
wrought steel properties. The parts manufacturing process exist for data tracking and analysis to
printed at Location 2 (top left of Figure • The aim of this round robin test relate back to field performance without
5) had the lowest mechanical properties was to observe variance in material duplicating efforts, allowing teams to learn
(~9% less) of all builds. Build locations 2 properties as a function of orientation from mistakes or successes. To do all this
and 4 had Z oriented tensile data with the and plate location across equivalent raises big questions over IT infrastructure
lowest values (see table in Figure 6). This and different equipment types, with issues. If there is no uniform software and
was because gas flow worsened when the same or equivalent process network system across different branches
the machine’s filters were nearly full. In parameters. and centers, then it will be difficult for
addition, many AM process conditions approvals and data sharing to happen
needed to be taken into account such as with additively manufactured parts.
powder coverage, build plate material/ AM Engineering Simulation
condition, recirculating gas filtration, Digital Data Storage Challenges
gas flow rates, part orientation, and part MSC’s MaterialCenter as the
location on the underlying build plate and In particular, given the sensitive nature of AM Data Management Solution
these parameters must be controlled military parts, data security is critical in AM at CCDC Armaments Center
for consistent AM part mechanical – how is digital data adequately protected
properties. Hence, a manufacturing in additive manufacturing? How is data To overcome digital data challenges of
plan with defined operating windows is sharing implemented especially if different additive manufacturing, a software solution
needed to ensure parts are consistently network security protocols exist, where is necessary for traceability, storage,
made to specification. cloud-based solutions are not widely and analysis of simulation material data.
adopted? Moreover, in terms of data Armaments Center used MSC Software’s
classification where data aggregation could MaterialCenter (Figure 8) and developed
The Effect on the AM raise the classification, there is a need for an additive manufacturing schema to
Benchmark Tests on Using a controlled system. Invariably, different enable the storage of all printer machine
Different Machine Types formats occur across a wide variety of parameters along with corresponding
OEM machines for metals AM with no material properties. It utilizes M/S Excel
To check for the effect of different additive standardized software or file format. integration in order to map and import
machines, six AM commercial machines custom templates. The data collected is:
were chosen to print the same parts in a In terms of data organization, a unified file
“round robin” demonstration of variability structure does not in general exist. With • Machine Information
(see Figure 7): AM, large amounts of data generation • Part Data (CAD/STL/MAGICS Files)

• Starting powder properties
• Machine Build Parameters
• Build Layout and Orientation
• Laser Parameters
• Post Processing
• Metallographic Analysis
• Mechanical Testing Data
The data stored using MSC Software’s

MaterialCenter involves a flexible schema
for different applications, an automated
method for input of material process
information, data analysis to compare
and contrast properties and understand
how to optimize them, and finally allowing
traceability of test data. Figure 8: MSC MaterialCenter in the center of the entire AM Material Lifecycle workflow
Capturing the Entire AM at CCDC Armaments Center - this is

Material Lifecycle shown schematically in Figure 11 where
the integration between Windchill and
Figure 8 shows the data tracking process MaterialCenter for additive manufacturing
for additive manufacturing and Figure 9 is shown pictorially. The benefits of
depicts the two parts of the AM process: this system are that it is always up-
to-date for version control tracking; it
1. “Left of Test” where manufacturing leads to less duplication of efforts and
inputs that are used to create a part therefore reduced costs; it is a common
or specimen are captured. This side file system for traceability, file security, This additive manufacturing ePDM
of the test leverages MaterialCenter’s historical storage, etc; it provides for workflow helps to deal with the
Work Request, Pedigree and a better collaborative environment in management of data issue and can
Process features. MSC Software’s order to coordinate efforts; it allows for become a US Army Standardized
MaterialCenter tracks the test quicker fabrication of hard to replace Tool for Lifecycle support when
specimen from raw material through parts with standardized file systems printing a component. It provides
the complete specimen build process and organization; it yields common data confidence through validation to
(Figure 10). The team tracked the models for standardization and validation; the end user of part performance.
materials & the environment, Batch/ it is a single source of data for linkage Everything is in the chain from raw
Specimen numbers and the Part between systems; and it provides true material, files, machines, and post
Inspection. lifecycle configuration management for treatment and it is validated to
2. “Right of Test” where Material products additive manufacturing. In short, for perform as designed. In terms of
are tracked from test to export. us to make Additive Manufacturing as Data Capture, automated processes
“available” as traditional manufacturing to feed into the data management
Finally, PTC Windchill was chosen as the techniques, materials and process data, system were enabled. In terms of
ePDM system for this application and it must be linked to part data using this On-Demand Manufacturing, the
the overall central data system for AM enterprise ePDM approach. team has qualified and authorized
personnel with access to the data.
In terms of Predictive Modeling,
knowing how a part will perform
before printing is invaluable. In terms
of Cooperation & Data Sharing, it
will lead to the saving of money and
time by building on the most up-
to-date work. And, lastly, in terms
of Data Analysis, the team can
optimize process parameters via
statistical modeling and understand
the relationship between key AM
process parameters.
Figure 9: Data management applied to additive manufacturing process

Future Focus of Additive
Manufacturing in the US Army
The US Army CCDC Armaments Center

is aiming at integrating material data
management with other enterprise
software, eg. PLM, and collaboration with
other services such as the US Air Force,
US Navy, etc. Ultimately, this approach
can be expanded to other manufacturing
processes and the capture of legacy
manufacturing data with the creation
and storage of new data libraries. It is
also looking at new materials systems
(functionally graded materials, novel
alloys, hybrid materials), the fielding of AM Figure 10: Overall Flowchart of Additive Manufacturing Data
parts and AM systems for on-demand

Battlefield manufacturing, a wide range
of qualification & certification of materials,
processes and parts via additive Reference
manufacturing, and advanced fabrication
“Army Efforts in Metals Additive Manufacturing & Data Management”, R. Carpenter, SME Smart
integration with sensors and electronics.
Manufacturing Series – Additive Manufacturing, 07 June 2018
Figure 11: PTC Windchill and MSC MaterialCenter integration ePDM system for Additive Manufacturing Data

10x ICME
to Save Time,
Cost and Weight in
Materials Simulation
M
aterials are everywhere in the world that we inhabit, be they polymers, By Roger Assaker,
rubber or metals all the way through to exotic materials like composites, CEO at e-Xstream
foams and ceramics, and they underpin every engineering application
engineering
in all disciplines of computer-aided engineering today. Developing and
integrating new engineering materials into a product can be a very
Chief Material
expensive and risky endeavor requiring numerous years of research & development and Strategist, MSC
millions of dollars invested in prototypes and testing. We at Hexagon and MSC Software Software
have developed what we call our 10x Integrated Computational Material Engineering
(ICME) Solution to address this very issue. The 10x ICME Solution was developed
in collaboration with global equipment OEMs and material suppliers and partners,
and is applicable to the latest advanced materials, reinforced materials and additive
manufacturing materials. If you currently face a materials challenge, Hexagon and MSC
Software is ready to partner with you to make your product a success. With 10x ICME
we aim to make your process less costly, more productive than other alternatives, and
first-time-right to address both the commercial, environmental and engineering challenges
commonly found in the materials development process. Each individual solution within 10x
ICME contributes to the whole, and they are integral to successful outcomes.
HxGN Smart Factory
CAE CAM
CAD/PLM ERP/CAD/PLM Neutral CAD/PLM

1. Virtual Material Development – Many contemporary materials are made of multi-
phases, but existing engineering simulation tools typically are unable to model at the
micro scale where many material properties are determined. Different mixtures, or
combinations of material constituents, at the micro scale can be simulated with our
10x ICME Solution.
2. Virtual Material Testing – Developing a new composite material for instance

can take 2-3 years of testing and millions of dollars. The process usually involves
testing numerous coupons. To compress this time and lower costs we have
created a solution to perform the coupon tests via simulation software to the same
level of accuracy.
3. Materials Lifecycle Management – After you complete expensive material testing,

your valuable material data must be stored, and sometimes it is in a simple PDF or Excel
file, which can become difficult to track on your systems. We have created a solution to
store, manage and distribute material data within a company, enterprise wide.
4. Material Exchange – Materials provided by material suppliers are usually proprietary.

To allow for this in your supply chain, we have created a capability that makes the
supplier’s proprietary material models available in an encrypted manner and allows for
subsequent engineering analysis without giving away IP.
5. Compliance and Sustainability – An increasingly important and critical aspect of

materials development is ensuring environmental compliance and sustainability of
the end product. To this end, our 10x ICME Solution features an integration with our
partner iPoint-systems GmbH’s unique software that allows you to identify and avoid
restricted substances early in the design.
6. Effect of Processing – The injection molding process for polymers itself influences
the performance of the final manufactured part, but simulation tools typically leave out
critical details such as fiber distributions and fiber orientations. Our 10x ICME Solution
addresses this by providing mechanical simulations that incorporate critical details of the
FEA simulation, giving greater accuracy of the performance of the part being simulated.
7. Accurate Material Modeling – A manufactured part composed of a reinforced

material for instance must be validated to perform well during its service life. Finite
Element Analysis (FEA) is used in the 10x ICME Solution during its validation process
with accurate material models.
8. Additive Manufacturing (AM) – Typically, modern simulation tools are unable to

simulate the AM process well for polymers and composites, and analysts resort to
physical prototypes to develop AM procedures and components. Our 10x ICME
Solution makes possible the simulation of the AM process for polymers, thus
minimizing costly physical prototypes. It offers a unique combination of material
engineering, process simulation and structural analysis solutions.
9. Material Intelligence (AI) – A significant amount of real and virtual test data is
generated during engineering simulation. The challenge to design engineers is to
make decisions based on the large volumes of data generated. We have uniquely
implemented Artificial Intelligence (AI) methods embedded in our 10x ICME Solution to
improve your material selection process.
10. Digital Continuity & Material Twin – Material development processes usually
involve multiple steps, some of which have simulation equivalents that have
high predictability as a result. Our 10x ICME Solution includes an extensive suite
of simulation equivalents for material development, material characterization,
manufacturing simulation and final part performance. Ultimately a digital twin of the
material manufacturing process is obtained and improves predictability.

AU TONOMOUS
Multi-Resolution
Traffic Simulation
for Connected Car Applications
using VIRES VTD
By AUDI AG: Andreas Kern

Technical University of Munich: Manuel Schiller
Institute of Transportation Systems: Daniel Krajzewicz
VIRES, part of Hexagon: Marius Dupuis

V
ehicular ad hoc networks available computing resources. This
(VANETs) have attracted a lot article presents an approach to solve this
of research attention over the trade-off by coupling multiple resolutions
last few years because they of traffic simulation to get highly accurate
have the potential to improve simulation results where they are needed,
traffic safety, efficiency and driver comfort. and simultaneously achieve an efficient
In fact, several ADAS (Advanced Driver simulation of large-scale scenarios of the
Assistance Systems) applications, such surrounding environment.
as cooperative driving and subsequently
automated driving, can only be achieved
through wireless communication between The Developed Multi-
the vehicles on the road. Resolution Traffic Simulation
Since those systems often exhibit safety- Microscopic Traffic Simulator: SUMO
critical features, rigorous testing and We chose to use Simulation of Urban
validation must be completed before their MObility (SUMO) as the traffic simulator
mass adoption. Although real road tests responsible for the simulation of the
using physical prototype vehicles offer the low-resolution area (LRA). SUMO is
highest degree of realism, the large amount a microscopic, space-continuous,
of resources needed to perform large-scale and time-discrete simulator. While it is
and extensive testing of vehicular networks employed in a wide range of research
renders their use impossible. Simulations domains, its most notable use is shown
are essential to validate the performance in a high number of research papers
of such solutions in large-scale virtual regarding VANET simulations. SUMO is
environments. Furthermore, simulation- well known for its high execution speed,
based evaluation techniques are invaluable as well as for its extensibility. SUMO is
for testing those complex systems in a wide ideally suited to simulate a high number
variety of dangerous and critical scenarios of vehicles residing in the LRA due to
without putting humans at risk. its efficiency, which is partly achieved
through its simplified driver model (which
In the automotive industry, the use of determines the path a vehicle will take).
simulation (Figure 1) is well established
in the development process of traditional Nanoscopic Traffic and Vehicle
driver assistance and active safety systems, Simulator: VIRES Virtual Test Drive
which primarily focus on the simulation of We employ the nanoscopic traffic and
individual vehicles with a very high level of vehicle simulator VIRES Virtual Test Drive
detail. When investigating and evaluating (VTD) for the simulation of the high-
the performance of ADAS based on resolution vehicles. VTD was developed
vehicular communication, this isolated
view of a single vehicle alone or a small
number of vehicles in the simulation is not
sufficient anymore. Potentially, every vehicle
equipped with wireless communication
technology could be coupled in a feedback
loop with the other road users participating
in the vehicular network, and therefore,
the number of influencers that need to be
considered is drastically increased.
These considerations lead to a trade-

off between accuracy in terms of the
simulation details for each vehicle and
scalability in terms of the number of
vehicles that can be simulated with the Figure 1: ADAS simulation

for the automotive industry as a virtual test been reached for SUMO and the condition
environment used for the development TVTD ≥ TVTD + SSUMO is therefore fulfilled, the
of ADAS and Autonomous Vehicles. Its state of the high-resolution vehicles is sent
focus lies on the interactive high-realism to SUMO through a gateway. This triggers
simulation of driver behaviour, vehicle the simulation of the next timestep in the
dynamics, and sensors. VTD is highly low-resolution model, and as a result, the
modular, so any standard component may positions of the low-resolution vehicles
be exchanged by a custom and potentially are passed back. These vehicles are now
more detailed implementation. Its standard classified, and, if applicable, the change
driver model is based on the intelligent of resolution is performed for individual
driver model; however, an external driver vehicles. When an exchange of a vehicle
model may be applied if necessary. The between the simulators happens, the
same concept applies to the vehicle previously mentioned inherent difference
dynamics simulation, where the standard in the underlying road network may cause
single-track model can be substituted by problems if a vehicle cannot be mapped
an arbitrarily complex vehicle dynamics based on its position in a specific lane due
model adapted for specific vehicles. Each to differences in accuracy. This is especially
simulated vehicle can be equipped with true for complex intersections which are
arbitrary simulated sensors, for example a modelled quite differently.
RADAR sensor, which is shown in Figure 2.
After all the resolution changes have been
Offline Pre-processing successfully completed, the simulation is
Figure 2: 3D visualization of a simulated
RADAR sensor in VIRES VTD To Enable Coupling unblocked again and the next timestep can
The two simulators rely on different data be simulated. This synchronization is very
formats representing the modelled road important to ensure reproducible simulation
network. In order to be able to run a results across multiple simulation runs.
co-simulation of both simulators, the
underlying data basis must match. VTD
uses the OpenDRIVE format to specify Dynamic Spatial Partitioning
the road network. This specification of The Simulated Area
models the road geometry as realistically
as possible by using analytical definitions. Our approach aims to couple traffic
SUMO on the other hand approximates the simulation models of different resolutions
road network geometry by line segments. at dynamic regions of interest. Contrary to
There are also differences in the modelling conventional traffic simulation, we are not
of intersections and lane geometries. interested in investigating a large number
To achieve a matching database, we of vehicles from a bird’s eye perspective,
convert the road network in an offline pre- but the focus is rather on a single vehicle
Figure 3: Dynamic partitioning of the processing step from OpenDRIVE to the (or a limited number of vehicles) which are
simulation area file format SUMO supports. used to conduct a test drive in the virtual
environment. This vehicle of interest has the
Online Coupling and Synchronization ADAS system under investigation onboard,
The coupling of the simulators at and is referred to as the EGO car. The
simulation runtime is based on the simulated measurements and sensor values
master-slave principle. Figure 4 shows are fed into the ADAS, and depending on its
this sequence of operations during a type and its use case, the respective ADAS
single simulation step, in which VTD and directly or indirectly influences the vehicle’s
SUMO can operate with different temporal state and behaviour.
resolutions without losing synchronization.
Based on this distance criterion, an area
SVTD is the length of a time step for the high- of interest is defined that centres around
resolution area (HRA), whereas SSUMO is the the EGO car, and in which the defined
length of a time step for the low-resolution simulative high-fidelity requirements
area (LRA). Typically, the nanoscopic must be fulfilled. Since the EGO car is
simulation is run at a higher frequency driving continuously through the virtual
than the microscopic one. TVTD and TSUMO environment, this area of interest is likewise
respectively denote the local simulation being moved along. We therefore partition
time in each simulator. At the beginning of the global area of the simulation dynamically
Figure 4: Comparison of simulation resolution each simulation step, a new timestep is into a high-resolution area (HRA) and a
switching simulated in VTD. If the next timestep has low-resolution area (LRA). Figure 3 shows

a schematic view of the dynamic spatial
partitioning. There, the HRA is defined
as the area of a circle which is centred
around the EGO vehicle. Red vehicles
are within that circle and are therefore
simulated in high resolution by the involved
nanoscopic simulator, whereas the green
vehicles are outside of the circle and are VTD is highly modular, so any standard
consequently simulated in low resolution component may be exchanged by a
by a microscopic simulation. All vehicles
exist in the microscopic simulation, but in
custom and potentially more detailed
the nanoscopic simulation contains only implementation.
the high-resolution vehicles, and their
movements are applied to their proxy
counterparts in the microscopic simulator.
Due to the dynamic nature of road traffic,

the EGO car, the high-resolution vehicles
as well as the low-resolution vehicles
are permitted to move continuously. The
classification of the assigned resolution
mode is therefore performed after each
time step of the simulation. Vehicles the oncoming traffic flow. This artificial
for which the classification has led to a road was first modelled in the OpenDRIVE
change in resolution are transferred to format and was then converted to the
the appropriate simulator. This change of SUMO road network format.
resolution is possible in both directions at
every time step. However, since the HRA is We performed two series of experiments.
defined to be centred around the EGO car, In the first series, the nanoscopic traffic
it is always simulated in high resolution. simulator VTD was applied to the whole
simulated area. In the second series,
we used the described multi-resolution Figure 5: Hysteresis control of the simulation
In order to prevent vehicles which are close
resolution
to the boundary between HRA and LRA concept to partition the simulation area
from oscillating very frequently between between VTD and SUMO. We chose
the two resolution areas, a hysteresis a timestep of SVTD 20 ms for the high-
controller as depicted in Figure 5 is applied resolution area in VTD and a timestep of
in the classification process. As shown in SSUMO 1 s for the low-resolution area in the duration is around 12 ms, which is less
Figure 3, the two thresholds Rin and Rout SUMO. The hysteresis thresholds which than the timestep length of 20 ms and
are defined. A vehicle is transferred into define the dynamic area of interest were therefore yet fulfils the real-time constraint.
the high-resolution simulation only if its set to Rin 500 m and Rout 550 m. At around 150 vehicles, the duration
distance to the EGO car falls below the is beyond these 20 ms and real-time
value of Rin. The exchange back to the Performance Evaluation simulation is not possible anymore. With
low-resolution simulation is carried not out We measured the duration it takes to increasing vehicle count, the duration for
until the threshold Rout is exceeded. perform each simulation step over the each timestep also considerably increases
simulation period of 1,800 s, while the and reaches 180 ms at the end of the
number of vehicles was constantly being simulation period. This results in a factor
Simulation & Evaluation increased. Each series consists of five 15 computation time increase compared
separate simulation runs to account for to the amount of at the beginning of the
Scenario and Simulation Setup fluctuations in the measured execution simulation. The overall simulation took over
A synthetic scenario was created times. To illustrate the trends of the 120 min to complete, which is four times
for testing the coupling concept and measurements more clearly, the moving more than the simulated time.
evaluating its performance. It consists of average is also displayed in the following
a single straight road running west to east figures. Figure 7 shows the performance
with a length of 50 km and two lanes, one development of the multi-resolution
for each direction. Each lane is configured Figure 6 shows the performance simulation in the same simulation scenario
to have a constant inlet of 1,000 vehicles development of the nanoscopic simulation over the same simulation period. While the
per hour heading either east or west. The while increasing the simulated vehicle count total vehicle count is increased the same
EGO car is located near the start of the over the simulation period. The duration way as in the pure nanoscopic simulation,
road. It drives from west to east and is of each simulation step is almost constant the separately plotted nanoscopic vehicle
followed by a traffic flow, heading towards up to a count of 70 vehicles. Until then, count illustrates the number of cars

Figure 6: Simulation performance—nanoscopic simulation only Figure 7: Simulation performance: multi-resolution simulation
which are within the high-resolution area. on a dynamically-determined area of Reference

It shows that reducing the nanoscopic interest. The presented methodology
“Multi-resolution Traffic Simulation for Large-
model’s area of interest fulfils the aim of partitions the simulation area into a
Scale High-Fidelity Evaluation of VANET
reducing the overall simulation time. After variable, highly detailed region of interest Applications”, Manuel Schiller, Marius Dupuis,
a local maximum of 11 nanoscopic cars is represented by a nanoscopic model, Daniel Krajzewicz, Andreas Kern and Alois
reached, this count decreases slowly since with VIRES Virtual Test Drive (VTD), and Knoll, © 3rd SUMO Conference 2015 Berlin,
Germany
slower vehicles are left behind the faster the surrounding area simulated at low
moving EGO car. At around simulation resolution by a microscopic model. The
time 1,350 s, the two traffic flows from evaluation shows a dramatic reduction of
each end of the road meet in the middle computation time in comparison with a
of the road, which then increases the pure nanoscopic simulation of the same
nanoscopic vehicle count. However, simulation dimensions, which even makes
because the extent of the HRA is limited, real-time simulation possible. This divide-
the nanoscopic vehicle count does not and-conquer strategy enables accurate,
exceed a certain limit, which for the given realistic, and large-scale testing and
configuration is at around 27 vehicles. The validation of real implementations of driver
duration for the timesteps stays on average assistance systems based on vehicular
constant around 12 ms, so the overhead networks in a virtual environment. As
resulting from the coupling of the two the next step, we are investigating
simulators is negligible. The execution time the application of the multi-resolution
of the microscopic simulator is also shown simulation methodology for the other
to be negligible due to its less detailed, yet domains relevant for the simulation of
much more efficient, simulation model. The vehicular networks, namely network
overall simulation took less than 18 min simulation and application emulation,
to complete, so the simulation was faster to model the whole system across all
than real time by factor 1.66 and the real- domains efficiently at high fidelity.
time constraint was fulfilled throughout the
whole simulation period.
Conclusion
In this article, we proposed a concept

Open Standards Essential for Self Driving? Download our
for coupling traffic simulators of different
Free Whitepaper: www.mscsoftware.com/openstandards
simulation resolutions to achieve a multi-
resolution traffic simulation which focuses

This divide-and-conquer
strategy enables accurate,
realistic, and large-scale
testing and validation of real
implementations of driver
assistance systems based on
vehicular networks in a
virtual environment.

M U LTIB ODY DYN A MI C S
Catching Structural Fatigue of

Concrete Mixer Truck
at the Design Stage
with Modal Synthesis
By SANY Automobile: Xia Xuewen, Lei Xinjun

MSC Software: Wang Chengkai
S
ANY concrete mixer trucks carry heavy loads with complex operating
conditions, and as such new models are usually subjected to rigorous
road tests with a specified mileage for at least two months before mass
production. The intensified test field includes twisted road, a fish scale pit
road and washboard road. When subjected to such demanding proving
grounds, the vehicles would suffer significant damage to the frame structure that typically
occurred under fatigue-torsion conditions.
To reduce the time it took to test vehicles and find the design flaws earlier in the test
process, it was necessary to conduct the fatigue simulation analysis of certain structural
parts on the mixer truck. In this project, we used MSC software tools to build a multibody
dynamics model for the four-axis mixer truck so we could perform the virtual proving
ground analysis. We also implemented physical test validation of the stresses and
vibrations to further calculate the fatigue results, to provide an early warning of any under-
designed components.

MSC Adams Virtual Proving Ground MSC Nastran Flexible Parts
Validate Loads
SANY Vehicle Vibration Test Vehicle Dynamics Simulation Fatigue Analysis
Figure 2
Full Vehicle Multibody Dynamics Modelling
Structural Components Modal Stress Recovery Theory

The large structural parts of the mixer truck include the main frame, the sub-frame,
the front, the back and the escalator. If a conventional finite element method is used to
simulate the deformation of the structural parts, it drastically increases the degrees of
freedom for the model and thus the computational cost. Since the structural response is
controlled by the low-order modes, it is not necessary to solve the high-order dynamic
equations of the whole structure for a few low-order modes. Therefore, the modal
Figure 1: Physical road testing for the mixer truck
synthesis method is used to simulate the deformation of the structural components, and
one of the most representative methods, the Craig-Bampton method, was used.
In this application, MSC Nastran is used to analyse the free modes of the main frame,
sub-frame, front, back and escalator to generate an MNF file. This is imported into the MSC
Adams multibody dynamics software to create a flexible frame, as shown in Figure 3.
Full Vehicle Modelling

MSC Adams dynamic simulation analysis software was used to build the entire vehicle model.
Adams is the most widely used mechanical system dynamics simulation tool, with applications
across industries like automotive, aerospace, aerospace, railway, medical and machinery.
Adams enables engineers to more effectively evaluate various system dynamic characteristics,
improve product performance and reduce expensive and time-consuming physical testing.
Adams provides a dynamic model library for buses and trucks. The truck model library has
a sophisticated three-bridge template. This article describes how SANY built a complete
vehicle dynamics model for the concrete mixer step by step by starting from this template.
Figure 3: Flexible truck frame model
1. Input the parameters: This included the vehicle parameters, steering system, front
suspension, front leaf spring, rear leaf spring, frame, cab, mixing drum, powertrain and
tyre parameters. In general, we extracted the hard points, component mass inertia,
bushing characteristics and other parameters, etc.
2. Built a frame model for the mixer truck: We created a four-bridge mixer frame
model based on the three-bridge template that included the addition of the second
bridge and the mixer drum subsystem.
3. Subsystem modelling: We modified the existing template subsystem based on input
parameters, established a mixer truck steering system, front suspension system, rear
suspension system, frame, cab, powertrain system, braking system and tyre system;
new front leaf spring system, rear leaf spring model and mixing drum system.
4. Full vehicle modelling: Finally, we replaced the subsystems in the previous mixer
Figure 4: Rigid-flexible coupled full vehicle truck model to generate the full vehicle model, as shown in Figure 4.
model

Figure 5: MSC Adams Virtual Proving Ground
Virtual Proving Ground Modelling

We used the Adams 3D shell road model to
establish three special types of virtual roads
and the full-circle proving ground. These
special virtual roads were used to verify the
dynamic model and to create the virtual
proving ground that is used to extract the
loads exerted by the roads (Figure 5).
Multibody Dynamics
Analysis
Model Tuning
There are 3 pairs of leaf springs in the
four-bridge mixer truck. Each piece is
modelled by discrete beam elements.
The degrees are nearly 800 degrees of
freedom for the rigid-flexible coupled full
vehicle model. With such a complicated Figure 6: Stress and Vibration Test Results
model, the model tuning needs to be
completed on both the suspension level,
and the full vehicle level.
Virtual Pavement Simulation

Analysis
The most important task of full vehicle
tuning is to perform sensitivity analysis. We
performed the sensitivity study based on
the key parameters that have an impact
on the stress results. These parameters
include the twisted road ramp, tyre
pressure, bushing stiffness and so on.
Figure 7: Correlation between simulation and testing results
After analysis, the importance for those
different parameters are ranked as follows:
reducer bushing stiffness > eccentricity
> ramp > tyre pressure. The amplitude There were 24 symmetrically-arranged According to the order of importance for
increases as the damping ratio decreases. stress monitoring points. Eight of those various parameters in the sensitivity study,
were dedicated to the principal stress; we incorporated the tyre stiffness, vehicle
16 were related to the Von Misses stress. drift & pull and ramp slope parameters with
Stress Vibration Overall there are 13 measurement points for the precise tyre pressures, and we focused
Experiment Validation vibration, focusing on the accelerations in on tuning the bushing stiffness. Adams’s
the vertical direction, as shown in Figure 6. modal stress recovery feature was used
Stress Vibration Experiment to compare the simulated stress results
In the testing facility, the vehicle stress and Simulation-test Correlation with testing results. We also compared the
vibration tests were carried out in strict For the twisted road, we correlated the full results on vertical acceleration between the
accordance with the test procedures. vehicle simulation results on the stress value. simulation and testing (Figure 7).

Figure 8: Fatigue simulation results correlate well with physical test results
The values of the first five maximum stress prerequisites: a finite element model, Based on the modal stress recovery
points were all greater than 200 MPa, and Adams loads and material information. theory, the fatigue life of the frame is
the maximum error of simulation and test The finite element model is provided by calculated using the MSC durability
amplitude was 13%. The average stress the BDF file from MSC Nastran. The road simulation workflow. The simulated stress
value was greater than 100 MPa, with an loads calculated by Adams are included vibration had an 83% correlation with the
83% correlation level with the test results. in a DAC file. The material information physical testing and the simulation and
The acceleration results also correlated includes the main sub-frame SN curve and test results have a trend consistency of
well between simulation and testing, and the escalator SN curve. The calculated life over 90%. The proving ground durability
the trend simulation error is only 9%. cycle of the frame is shown in Figure 8. results also verified the accuracy and
effectiveness of the rigid-flexible coupled
Through fatigue calculation, the minimum simulation methodology and the fatigue life
Fatigue Simulation Analysis life cycle of the frame was shown to be calculation. The entire durability workflow
29,609 km, and the minimum life cycle of can be used to predict the life cycle of
Virtual Proving Ground the sub-frame was 22,010 km, both of newly developed vehicle models, which
Simulation Analysis which are far greater than the 10,000 km has drastically reduced overall testing
Adams has a code-driven function that design requirement. Through the intensified campaign duration, development costs
specifies the steering, throttle, brake, gear physical tests in the real proving ground, and the need for physical prototyping.
and clutch signals with pre-programmed we can prove that there hasn’t been any
events. The test bench controls the vehicle fatigue cracking failure that occurred in the
to drive around the virtual proving ground frame structures, as the simulation results Reference
through a pre-built program. Subsequently, had predicted.
“The Component Fatigue Analysis of Concrete
the Adams Durability module can directly
Mixer Truck Based on Virtual Road”, Xia
output load files to the MSC Fatigue tool Xuewen, Wang Chengkai, Lei Xinjun, ©
for fatigue calculation. Conclusion Automobile Applied Technology. Article ID: 1671-
7988(2018)23-216-04
Fatigue Calculation Results In this project, the four-bridge mixer truck

The MSC Fatigue calculation has three is taken as the main research object.

CFD

A
nalog Way was initially
founded in France in 1989.
Since then, I am proud to
say we’ve established a
worldwide reputation for
quality, reliability and performance. For 30
years, we have pioneered the design and
engineering of numerous award-winning
image processing solutions.
At Analog Way, we have delivered some

incredible experiences to our customers.
From panoramic LED canvas in sports
books to large live corporate events
and concerts, from product launches to
installations in prestigious locations. As
an example of our capabilities, we’ve
delivered equipment to drive the highest-
resolution LED screen in Times Square.
For this project, we partnered with SNA
Displays to provide a full 8K resolution
processing solution for a massive wrap-
around display canvas, one of the largest
continuous exterior displays in the world
representing more than 17,000 square
feet of LED display technology. For the
record this is equivalent to almost 4
Designing
basketball courts.
Premium Innovative
Solutions for 30 Years
By Philippe Vitali, Marketing & Comms

Director, Analog Way

Our only real way
forward to drive this
sort of innovation was
Cradle CFD simulation
As a result of efforts like these, Analog to power increasingly rich, interactive

Way is seen as one of the undisputed experiences. We know they demand
leaders in video presentation experiences. industrial grade reliability. They also need
This is an important element of who we a high degree of customization built into
are, and the capabilities we want to be some very flexible, modular, systems.
able to deliver to our customers.
In order to deliver what we considered the
It feels great to be able to talk about next generation of video processing, we
these capabilities and how we strive estimated we’re going to need a system
to offer premium innovative solutions capable of dealing with 24 x 4K inputs, as
Figure 1: Cradle CFD Solution
to our customers. It doesn’t come well as evolving requirements around 8K.
accidently and is the result of a clearly This meant we needed deliver 3 x times the
choregraphed, and very intentional, effort. video processing performance over and the design variations our customers are
above what we’d achieved in Times Square. going to be able to select. The modular
So, how do you follow up an example like design of the LivePremier™ series allows
Times Square? How do you maintain, or These are significant improvements in customers to easily swap in I/O cards to
extend your high-end positioning? This core performance. In practical reality accommodate a variety of connectivity
is the real question. It’s one thing to want these improvements needed to be arrangements to match their source and
to be among the leaders on your market. delivered with the same, or tougher, display requirements. In fact, per chassis,
How do you actually do that though? physical constraints. In fact, we needed the LivePremier™ series features up to
to deliver this performance in one of the 24 inputs and 20 outputs configurable as
We knew we needed to maintain our most compact systems in the world. This single screens, edge-blended widescreens
extreme performance edge. However, increase in power meant a huge power or scaled auxiliary outputs, 2 dedicated
to really make a difference, we had to footprint in a relatively small confined multiviewer outputs and up to 24x 4K or
start with some basics. We had to turn space, often in some tough conditions. 48x HD freely assignable layers.
our desire to be a leader into something
tangible that meant something to our Consider an outside display in Las Vegas. Clearly, there’s a lot to consider… and
customers. We know our customers need During the day it is not uncommon to we hadn’t even considered our own
see temperatures hitting 40 C, dropping deadlines and design processes. For
during the evening, and incorporating all example, how would we iterate on our
the dust and dirt you might expect with a designs in a timely manner so we could
desert environment. In these environments, reach our performance goals. We feel we
the effective dissipation of heat, and have an extremely talented engineering
the management of severe temperature team, that is arguably the best in our
variation is key to maintaining effective industry. However, it quickly became
performance our customers can rely on. clear that traditional processes, relying on
We therefore needed to absolutely, 110%, experience, educated guess work and
ensure our systems are able to deliver 24/7 physical testing was going to struggle as
in these environments. we tried to ensure we reach our goals.
Already a tough engineering challenge, we Our only real way forward to drive this sort
then needed to ensure we can commit to of innovation was Cradle CFD Solution
our performance considerations for all of simulation. We’d be able to iterate on our

designs, and simulate our performance
in a fraction of the time we’d done in the
past, and gain more insight along the way.
For this we chose MSC Cradle scStream.

Until MSC Cradle, frankly, we’d not
considered Computational Fluid Dynamics
(CFD) too much. However, our needs here
meant we needed something pragmatic
that could actively help our team reach
our goals. We were not looking for a fluid
dynamics science project. We simply
wanted to drive our design forward.
We needed something that was able to

deal with our fairly large, configurable,
assemblies and help us gain the insight
we needed, quickly, accurately, and
effectively. We needed to get a clear view
on the thermal conditions in our compact
footprint. In particular we needed to
understand the parameters required to
maximize air cooling system performance,
and clearly ascertain maximum
temperatures, and their fluctuation.
The initial model, with some support from

MSC, only took 2-3 days to setup, and
we had some of the detailed results we
needed within a week. From a physical
perspective we’re now confident we can
effectively and reliably dissipate some
2KW across 133 In2 (approx. 800cm2).
To provide some context the amount
of energy we’re generating is similar to
a household electric fire. It’s therefore
important we get this right.
So far, we’ve been pleased with our

decision. So much so, we also used MSC
Cradle CFD to produce an 8K video to
highlight our core differentiative features
- front panel configuration, swappable
cards, redundant power supplies, and of
course our industry leading cooling. We
even featured it’s results on our website!
However, it’s not enough to make us happy.

It’s our customers we need to impress. Our
LivePremier™ solution line won two best of
show awards at the 2019 Infocomm show
in Orlando, and this is only the beginning!
Find out More About MSC Software CFD Solutions:

www.mscsoftware.com/cfd

AD D I TIV E MANUFA C TU R I N G
Flexibility
Through Additive Manufacturing:
How Simulation Supports
3D Prototyping
Simufact and its technology partner toolcraft shows

in a best practice case how additive manufacturing helps
to save time and money in the production of prototypes.
M
BFZ toolcraft GmbH require the component to have a long service process suitable for series production.
from Georgensgmünd life and high wear resistance so that it can However, before a new blade geometry
in Middle Franconia withstand mechanical and thermal loads. can be used with the required properties,
has optimized together many tests are required for which
with its software partner prototypes or small batches of blades are
Simufact Engineering from Hamburg the From Prototype to Series / required. In exceptional cases - depending
additive production of a turbine wheel Take a View On Manufacturing on the number of parts required - the
from ABB Turbo Systems AG. Typically, and Its Challenges in Serial turbine blades required for testing can
these components can be found in drive Manufacturing also be produced by casting in very small
units of heavy machines and vehicles, series. In general, these processes are
such as diesel locomotives, off-highway Filigree blade geometries are typically very time-consuming and cost-intensive
trucks or dump trucks. Depending on the produced by casting processes as and therefore not much more than two
application, manufacturers an economical and robust production prototypes are available to develop the

CHALLENGE: Transfer prototyping SOLUTION: Generate variant diversity
into serial manufacturing. Using the with the help of additive manufacturing.
example of a filigree blade geometry we This technique helps you save time
consider the challenges of traditional and money. Reach the first-time-right
manufacturing. approach through simulation.
USED PRODUCTS: Simufact Additive USER: MBFZ toolcraft GmbH
final product for use in series turbines. Due to component geometry and thermal
At this point, additive manufacturing has stress, high stresses occur during the
become a key technology that saves time building process. This is due to the special
and money. Furthermore, the technology features of the geometry, which on the one
offers a maximum flexibility, one of the hand has a solid core with a lot of material
most important requirements in the field and volume, while on the other hand the
of prototyping. With the help of this blades are very filigree. As a result, there
innovative manufacturing process, a are large cross-sectional changes in the
variety of turbine blades can be produced component, which favour the residual
in a very short time, which ultimately stresses during the manufacturing
leads to a better product. This is where process. These in turn result in a high
MBFZ toolcraft‘s high manufacturing susceptibility to distortion.
competence throughout the entire value-
added chain in turbine blade production MBFZ toolcraft solves this problem with
proves its worth. Within the framework of a careful simulation-based as-is analysis
the cooperation between MBFZ toolcraft in which critical areas are identified. From
and ABB Turbo Systems AG, the products this, the necessary measures can then
can be designed and implemented as 3D be derived to counteract the distortion
printing right from the start. problem. This includes the development
of suitable support structures that
generally minimize distortion and thus
Simulation Provides Reliable ensure a safe construction process. But
Information on Distortion and the ideal alignment of the components
Stresses in the Component to be printed on the base plate can also
be very helpful in individual cases. The
For MBFZ toolcraft, the greatest challenge last step is an automated compensation
in manufacturing prototypes is maintaining of the remaining distortion based on a
the required tolerances and dimensional quantitative distortion analysis, with which
Image 1: Simulation helps to reduce component
accuracy. The decisive factor here is the the remaining distortion is determined. The distortion and thus to keep tight tolerances.
component distortion caused by the AM results obtained in this way can be used
process. In order to keep the distortions as to derive the print preparation. Thanks to
low as possible, MBFZ toolcraft relies on the simulation, MBFZ toolcraft achieves a
Simufact Additive. By using the user-friendly low-distortion component structure and
and process-oriented simulation solution, can thus remain to its “ first-time-right“
MBFZ toolcraft makes it possible to approach - to fulfil all requirements on
significantly minimize distortions by means the component with the first print. The
of suitable process parameters and to use of additive manufacturing enables
compensate where they cannot be avoided. MBFZ toolcraft to react flexibly and quickly
In this way, MBFZ toolcraft can meet all to customer requests, such as design
required tolerances, thus eliminating the changes, and thus to significantly reduce
need for time-consuming reworking. project lead times. The virtual engineering
offered by the powerful simulation solution
enables significantly tighter processes in
Problems and Challenges in the process development of 3D printing
the Building Process projects. This approach can be realized
through the reliable software Simufact
Image 2: From design to simulation to the
A closer look at the building process clearly Additive. finished component – less distortions thanks
reveals the challenges and problems. to Simufact Additive.

S T RUCT U R E S
MSC Nastran and AVL EXCITE™ Integration Delivers
Better, Faster
and More Efficient
Powertrain Simulations
By Bernhard Loibnegger, AVL
Paul Herster, AVL Excite Product Manager,
Bhoomi Gadhia, Product Marketing Manager, MSC Software
O
riginally developed for NASA engineers, MSC Nastran is the most
trusted FEA solver for design and analysis of aerospace and automotive
structures. Industries have commonly used the proven solutions of MSC
Nastran for the past 50 years. To build upon the same practice, MSC
Nastran provides streamlined workflows for powertrain simulations,
improved verification options for contact models and enhanced HPC methods for
simulation of large-scale problems.
Through a direct coupling between MSC Nastran and the AVL EXCITE, engineers can perform
the analysis of dynamics and acoustics of powertrain systems faster and more accurate.
AVL EXCITE is a multi-body dynamic simulation software that provides rigid and flexible
solutions for the analysis of powertrain systems. The robust solutions of AVL EXCITE
family are equipped with advanced technology to calculate dynamics, strength, vibration,
and acoustics of conventional and electrified powertrains. Engineers can create accurate
models that account for all physical properties of materials and lubricants to predict
contact behavior, friction, and wear in slider bearings, pistons, piston rings and gears.
With the new direct coupling, AVL EXCITE can now be used with MSC Nastran to increase
the accuracy of Multi-Body simulations by providing a better representation of the flexible
components. Similarly, the FEA engineers can benefit from this integration to examine each
component with realistic boundary conditions and dynamic loads. Direct export of AVL
EXCITE input (.EXB) files from MSC Nastran makes data transfer much easier and faster,
eliminating the need for creating and translating extensive intermediate data.

Figure 1: MSC Nastran SOL 103 to AVL Excite to SOL 111 Workflow
The workflow allows engineers to obtain a multiple op2/op4 files from MSC Nastran which are converted by another standalone
full characterization of structural response proprietary program into an EXB file.
by leveraging the nonlinear solution 2. Simple: Only two files are created during step-1 (EXB and external superelement op2)
of MSC Nastran (SOL 400) to include as opposed to multiple op2/op4 files (GEOMUNIT, EIGNUNIT, PSTUNIT, GPSRUNIT,
contact nonlinearities as well as preloaded MKAAUNIT, MFFUNIT, TRANUNIT).
conditions in their powertrain simulations. 3. Fast: For step-1, shared memory parallelization (SMP) would be enabled to make the
Additionally, MSC Nastran can read calculation of computationally intensive higher order mass invariants faster. For step-2,
back output results from AVL EXCITE for EXTSEOUT feature is leveraged to minimize data storage and enable efficient data
further dynamic and acoustic analysis recovery in MSC Nastran.
of sound radiations generated from 4. Convenient: Allows for quick export of EXB file to AVL and quick import of AVL
Internal Combustion Engines and other results back into MSC Nastran.
components. This technology is also
beneficial for vehicle durability simulations, The direct, efficient, fast, and convenient method to exchange data between the two
taking advantage of the award-winning solvers enables significant time savings, resulting in up to 3x performance gain for analysis
embedded fatigue capabilities of MSC of powertrains and engine systems.
Nastran to obtain fatigue life and damage
responses of components.
This functionality will be of benefit to

the automotive industry by providing
the analyst that uses MSC Nastran and
AVL EXCITETM and workflow to perform
various analysis (like dynamic, vibrational,
acoustical) - a direct, convenient and
efficient method to exchange data.
The enhanced MSC Nastran-AVL

EXCITETM Interface with MSC Nastran has
multiple advantages over the currently
used DMAP alters method:
1. Direct: Creation of EXB file from

MSC Nastran rather than exporting Figure 2: MSC Nastran SOL 400 to AVL Excite to SOL 400 Workflow

FLUIDS
Cradle CFD Innovation that

Achieves Excellent
Indoor Air Circulation
By Mr. Masahiro Shigemori, Panasonic
P
anasonic Ecology Systems Ceiling Fans Designed by Malaysia
(hereafter the Company) Fluid-Structural Co-Simulation
has been tackling
Indoor Air Quality (IAQ) Ceiling fans are popular in southeast Asia
improvement for many and middle east countries. The Company
years. As well as dealing with ventilation has provided various types of ceiling fans
and dehumidification, the Company according to design preferences and feature
has pursued achieving quality living requirements in each country (Figure 1).
environments through providing pleasant India
air flow and smell. Their market leading Ceiling fans are given safety measures
product range includes ventilation fans, such as falling prevention. Mr. Shigemori
kitchen hoods, air cleaners, and roof fans, says that ceiling fans can be damaged
and is also available in many countries in or can crash to the floor by blade
Southeast Asia. To design and develop vibration, and in some cases by wind
the equipment needed to facilitate such from air-conditioners, which could add
environments, highly advanced technology wind pressures (external pressures)
and expertise are required. Cradle CFD when hitting the fan. Caution notices are
has been an innovation that brought written on fan manuals, but as fans can
significant improvement to their design be allocated near air-conditioners, the
evaluation process. To help them with Company needs to validate safety by
product development, Cradle CFD has simulations.
been in use and “has played a vital role in
the process,” according to Mr. Masahiro As the analysis involved calculating the
Shigemori, Chief of Ventilation Technology force applied to fan blades and how they
Development Division, Thermo-Fluid could deform blades, the Company has
Development Department, R&D Head been using co-simulations of fluid and
Middle East
Office (Picture). structural analysis. To perform, Software
Cradle’s scFLOW and MSC Software’s
Figure 1: Ceiling fans provided across countries

Figure 2: Transient deformation of ceiling
fan caused by air-conditioner wind
MSC Nastran have been in use. These Future Challenges evaluations. Another goal is to expand the
were initially devised separately, and data areas of co-simulation.”
import was inconvenient. The recently Going forward, Mr. Shigemori comments:
introduced MSC Co-Sim engine has “We have been applying simulations for It is certain that the Company will
enabled better connection and control. qualitative comparison and evaluations, continue to pursue IAQ improvement
but we are hoping to improve accuracy through innovative research and
Values of pressure on blade surfaces and one day apply it for qualitative development.
evaluated by scFLOW are provided
to MSC Nastran, and in reverse, the
degree of deformation evaluated by
MSC Nastran are provided to scFLOW.
From this, where and when deformation
and vibration could take place can be
evaluated. Result comparison shows that
fans rotate at stable rates of deformation
when there is no external force (bottom
of Figure 2), whereas when there is
wind from air-conditioners, the fan
blade vibrates in a complicated manner
(top of Figure 2). “We can observe the
complicated behavior by looking at
simulation results in animation format,”
says Mr. Shigemori.
Applying simulation technology and

evaluating models in the conditions and
environments that products are actually
used, the Company achieved safety
Mr. Masahiro Shigemori, Panasonic
validation of products as well as its
performance improvement.

GENERA L NEW S
AVL and MSC VTD Software to Collaborate on Autonomous Vehicle Projects
MSC Software and AVL List GmbH in Hexagon’s VIRES (Virtual Test Drive) is a “VIRES has worked with AVL in EU
Austria is partnering for research and tool-suite for the creation, configuration, projects such as Enable-S3 for many
development in the field of simulation and presentation and evaluation of virtual years now, as well as together with us at
testing of Autonomous Driving (AD) and environments in the scope of road and MSC on joint projects at European OEMs.
Advanced Driver Assistance Systems rail based ADAS/AD simulations. It As part of this ongoing collaboration,
(ADAS) systems. covers a full range from the generation AVL’s toolchains will interface with VIRES
of 3D content to the simulation of VTD and MSC’s Adams product lines,
As a leading test system provider, AVL complex traffic scenarios and, finally, and our toolchains will interface with AVL’s
is developing a scenario-based open to the simulation of either simplified or Model.CONNECT product. We believe
verification and validation toolchain to physically driven sensors. Due to its open this gives customers access to some of
generate operational scenarios, create the and modular nature, VTD can easily be the best-in-class autonomous simulation
relevant test cases, and execute testing interfaced and integrated into other AD and testing technologies available in the
and homologation of automated driving applications. world.”
functions in combined virtual and real
road environments based on an open and Commenting on this partnership, Paolo For more information:
integrated tool environment. Guglielmini, CEO of MSC Software said: www.mscsoftware.com/avl
MSC Software Selects CADLM to Deliver AI Solutions for CAE Globally
MSC Software announced an extension The ability to compile information Machine Learning techniques via the
to its cooperation with CADLM into a from CAE simulations associated with ODYSSEE suite of software.
strategic market development partnership mathematical models and data mining
for ODYSSEE, its innovative new state-of- allows for very fast analysis of very “We selected CADLM for its game-
the-art Machine Learning-based software complex and highly non-linear simulation changing software platform ODYSSEE
package for real-time parametric design responses such as crash, transient that allows for numerous machine-
and optimization. Paris-based CADLM has dynamic and structural analysis, and learning solutions whose exploitation
been a pioneer in implementing Artificial CFD, but also finance, biomechanics, and is central to real-time CAE design
Intelligence technology in CAE, and MSC defense applications. CADLM develops exploration. Integration of CADLM’s
Software have selected CADLM solutions predictive, real-time design optimization AI technology with MSC’s CAE tools
to add to its product portfolio, globally. models for CAE simulations based on will enhance our mutual solutions and
establish new perspectives, not only in
CAE based structural designs, but also in
digital solutions that create autonomous
connected ecosystems as we move
towards the convergence of the physical
with the digital world, and intelligence
is built-in to all processes,” said Kais
Bouchiba, MSC Sr. Vice President
EMEA & Global A.I. / Machine Learning
Strategist, who welcomes the extension
of the partnership.
For more information:

www.CADLM.com

Shaping the Generative Design Market
MSC Software is positioned to offer material-saving, weight reduction and

its customers a paradigm shift away efficient, cost-effective production.
from general purpose optimization
technologies to an innovative, purpose MSC Apex Generative Design eliminates
built software solution for additive the inefficient manual effort that significantly
manufacturing with the launch of MSC slows today’s workflows, allowing work
Apex Generative Design. steps to be completed in days that would
otherwise take several weeks.
Enabling the efficient production of highly
complex components and providing For more information:
customers with numerous benefits, from www.mscsoftware.com/generativedesign
HxGN LIVE 2019
We are thankful to the thousands of

business leaders from around the world
who came together to experience four
unforgettable, transformative days
at HxGN LIVE 2019! We shared the
latest in CAE, autonomy, and additive
manufacturing at our MSC User’s
Conference, which also took place at
the event.
At this year’s opening keynote, Ola Rollén,

Hexagon President and CEO laid out
Hexagon’s far-reaching vision for putting
data to work in our organizations to reverse
the trend of resource depletion and waste
overwhelming Earth’s systems – and how
higher productivity and quality outcomes
will drive sustainability and growth for our
businesses as well as the planet.
For more information:

www.hxgnlive.com
I AM Tomorrow Challenge
Hexagon has partnered with DSM in the Additive manufacturing is emerging as a

I AM Tomorrow Challenge, an acceleration more sustainable, transparent, and cost-
program for startups and scaleups focused effective way to manufacture products
on data analytics, AI and other advanced by reducing waste and minimizing
software solutions for large-scale 3D transportation. Today, the company actively
printing. Launched in collaboration with collaborates and partners with companies
Hexagon, HighTechXL, and Ultimaker, the across the 3D-printing value chain to unlock
challenge offers select companies a spot on the full potential of additive manufacturing.
a ten-week acceleration program designed The I AM Tomorrow Challenge aims to
to launch or mature their businesses. After stimulate the digital ecosystem that could
the acceleration program, participating shape the future of additive manufacturing
startups may be eligible for funding or a for decades to come.
partnership with DSM.
For more information: :
www.iamtomorrowchallenge.com

PRODUCT REL E A SE S
Adams
• Export geometry from optimization
result directly in a format that can
Adams 2020 | Adams 2020 continues be transferred into common CAD
to extend the ease in which the powerful platforms and used immediately
capabilities of Adams can be leveraged, without manual re-work.
and in less time. This release contains • Read the resulting geometries in
capabilities such as simplifying the Simufact Additive or Digimat AM
creation and publishing of model sub- to significantly reduce the time of
components, increased the flexibility for Design for Additive Manufacturing
exploring model configurations on the (DfAM).
fly, or incorporating custom Adams Car
bushing representations from external
applications via the FMI standard. We
continue to make significant strides in Digimat
performance and in this release there
are dramatic gains for rigid and flex body
contact models. Digimat 2019.1 | Pushing
boundaries of multi-material virtual
testing and part engineering | Digimat’s
MSC Apex MSC Apex | Generative Design latest release provides improvements
addressing its wide range of material
application. Early design values are
MSC Apex Jaguar | The latest MSC Apex Generative Design easier to reach thanks to Digimat-
release of MSC Apex Jaguar, improves MSC Apex Generative Design is the fully FE virtual lab, where microstructure
the interoperability with MSC Nastran automated generative design solution built of both SFRP and CFRP have been
and enables a new workflow interacting on the most intuitive CAE environment in improved. In addition, a new solver
with Adams Car. MSC Nastran based the world, MSC Apex. It exploits all the has been added (FFT solver) to shrink
users can enjoy extensive support of easy-to-use and easy-to-learn features time of virtual data generation. To
edge-ties, coordinate system, geometry of MSC Apex while employing the most reach more reliable virtual allowables of
from mesh and meshing capabilities to innovative generative design engine in CFRP, a new failure package has been
receive productivity gains in more tasks. the background. The software delivers implemented to Digimat-VA based on
While multibody dynamics (MBD) users a new and innovative approach for Prof. Camanho’s work. While in parallel,
can leverage MSC Apex as an interface design optimization which overcomes in order to shrink safety margins,
between Adams and MSC Nastran. the constraints of classical topology Digimat-VA now provides an “effect
From Jaguar, MSC Apex supports optimization techniques and dramatically of defects” flavor allowing to evaluate
load transfer between these two gold decreases the effort required in the design effect of porosity and out-of-plane
standard solvers and enables engineers optimization workflow: waviness. Industry more and more
to post-process models through turns to advanced properties of SFRP,
multiple viewports. MSC Apex Jaguar • Automatically generate multiple for constant load amplitude fatigue,
is a milestone in the road to building a smooth structural design candidates Digimat-RP enriches its fatigue post-
next-generation multiphysics simulation that all satisfy the design criteria processing environment accounting for
platform. Users are encouraged while minimizing the weight. plasticity. To support OEM and Tier ones
to experience its cutting-edge • Import existing geometries or mesh, with SMC material modeling needs, a
technologies to perform productive either in CAD, STL, or MSC Nastran dedicated approach has been deployed
engineering development. BDF format, find optimized design up to structural crash application with
candidates, and perform design Digimat-RP.
validation all inside a single CAE
environment.

To facilitate design engineer work with
additive manufacturing part design, the Marc
printer database of Digimat-AM has been
enriched in order to facilitate access to
process parameters. At the same time Marc 2019 Feature Pack 1
post-processing has been improved to Marc 2019 Feature Pack 1 provides
face printed part challenges such as significant advances in three key areas
investigating skin-core effects related to for the nonlinear community addressing
thermal history. Finally, the Digimat-MX performance, user-friendliness as
database content has been significantly well as use-case specific workflows.
enhanced, providing state-of-the-art “Smart” solver improvements in Marc MSC Nastran
advanced material models from major 2019 Feature Pack 1 result in better
material suppliers for injection molding performance, robustness and flexibility
and additive manufacturing applications. across a wide range of use cases and MSC Nastran 2019
To support material suppliers in material workflows. These improvements include Feature Pack 1 | MSC Nastran
card calibration and broadcast, Digimat- parallelizing the iterative solver to get 2019 Feature Pack 1 focuses on a variety
MX also allows the calibration of thermal significant scaling of up to 4X as well as of enhancements from “truly” coupled
expansion and creep models. using multiple threads and a user defined modes analysis for analyzing launch
time stepping to allow users to easily vehicles with heavy fluids to enhanced
obtain a more efficient solution with GPU support for performing full vehicle
Actran speedup factors up to 5x. NVH analysis up to 50% faster and with
improved accuracy as well. With advances
in electric car manufacturing and quieter
Actran 2020 | Actran’s latest release Furthermore, Marc Mentat now mirrors electric motor engines, higher frequency
provides improvements addressing all some of the MSC Apex interface. In noises and vibrations are becoming more
acoustic applications. The Virtual SEA particular, a dark theme to decrease eye prominent, which makes NVH calculations
analysis, enabling the generation of strain and improved mouse behavior exponentially more expensive. By enabling
reduced energetic vibroacoustic models for more efficient and user-friendly the users to take full advantage of their
for mid- and high-frequency, now offers interaction during and after model GPU resources, MSC Nastran 2019
the possibility to account for acoustic building. Marc 2019 Feature Pack 1 also Feature Pack 1 saves costs as well as
treatment effects in fluid-structure addresses some specific use cases and gives 90% performance improvement
interactions. Tire vibration results from workflow improvements. A few notable for the car manufacturers. Designed to
Abaqus can be imported in order to examples in this release are the addition further improve user productivity, this
study tire noise emissions accounting of the Vegter yield criterion for anistropic release delivers valuable contact analysis
for complete vehicle geometries and plasticity to improve the ease of use enhancements. MSC Nastran 2019 FP1
acoustic treatments. New embedded of material characterization and the now allows you to define contact using
automatic post-processing provides reduced integration Herrmann element a surface definition with segment-to-
automotive engineers access to ISO which improves speed by up to 40% segment method. As there is a need for
results for pass-by noise and acoustic for models involving high levels of metal more lightweight structures, we are leaning
power evaluations. The impact of a plasticity or hyperelastic behavior. towards using more composite materials.
swirling velocity component in acoustic In order to simulate the real-world
propagation inside turbomachinery scenarios accurately in FE models with
ducts can be accounted for by bringing MCS Nastran 2019 FP1, we now support
more accuracy for turbomachinery noise HASHIN Failure Criteria in Linear Solutions.
predictions. Finally, ActranVI ergonomics
has been rethought to improve user
experience.
Request More Information for MSC Software Products:

www.mscsoftware.com/moreinfo

Bridge the Gap Between
Design and Manufacturing
MSC Apex Generative Design is a radically new, fully automated
generative design solution built on the most intuitive CAE
environment in the world, MSC Apex. It exploits all the easy-to-use
and easy-to-learn features of MSC Apex while employing the most
innovative generative design engine in the background.
The software delivers a new and innovative approach for design

optimization which overcomes the constraints of classical topology
optimization techniques and dramatically decreases the effort
required in the design optimization workflow by up to ten-fold.
Learn More or Request a Quote

mscsoftware.com/generativedesign

MSC Software Magazine, Volume X - Winter 2019 by MSC Software

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MSC Software Magazine, Volume X - Winter 2019 by MSC Software

Uploaded by

Copyright:

Available Formats

Sustainability

Makes Good Business Sense

Panasonic GM US Army Samsung

• Actran FEA coupling for high fidelity aeroacoustics

2 | Engineering Reality Magazine

Design & of delivering practical engineering solutions

Engineering commentary on page 14. As Ola pointed

Simulation for opportunity for all industries, as well as a

Sustainability most complex problems facing humanity

Volume X - Winter 2019 | mscsoftware.com | 3

4 | Engineering Reality Magazine

Catching Structural Fatigue of Concrete Mixer Truck errors contained herein.

Volume X - Winter 2019 | mscsoftware.com | 5

6 | Engineering Reality Magazine

Modules are introduced in MSC Nastran as an assembly management tool; it is a

Figure 1: Traditional Model Build Up

Volume X - Winter 2019 | mscsoftware.com | 7

- Dr. Bruns, Volkswagen

modules. Post-processing results are

So, to conclude, MSC Nastran Modules

A module is a standalone Bulk Data section that typically represents a component of

8 | Engineering Reality Magazine

Volume X - Winter 2019 | mscsoftware.com | 9

Will all autonomous cars be I can’t speak to the specifics of timing,

10 | Engineering Reality Magazine

GM’s Approach to Automated Driving

Volume X - Winter 2019 | mscsoftware.com | 11

including design, Cadillac Driver Attention System

Figure 4: GM Super Cruise, before going to

12 | Engineering Reality Magazine

Sustainable Engineering is Good Business Sense:

The concept of sustainability is arguably as old as us humans reusing our nearby

14 | Engineering Reality Magazine

Across the world we emit approximately

Volume X - Winter 2019 | mscsoftware.com | 15

Adams Wind turbine model during a gust event

16 | Engineering Reality Magazine

-Gro Harlem Brandtland

Volume X - Winter 2019 | mscsoftware.com | 17

18 | Engineering Reality Magazine

Volume X - Winter 2019 | mscsoftware.com | 19

Integration with iPoint in MaterialCenter for Compliance Check

20 | Engineering Reality Magazine

Volume X - Winter 2019 | mscsoftware.com | 21

22 | Engineering Reality Magazine

By Tony Ge, Ford Motor Company

The traditional approach to transmission calibration at Ford involved physical testing of

Volume X - Winter 2019 | mscsoftware.com | 23

24 | Engineering Reality Magazine

Figure 2: Simulation Correlation with Test

Volume X - Winter 2019 | mscsoftware.com | 25

26 | Engineering Reality Magazine

By Kambiz Kayvantash, CEO, CADLM

Volume X - Winter 2019 | mscsoftware.com | 27

Real-Time Reduced Order Models

28 | Engineering Reality Magazine

any explorations beyond these bounds

CADLM capabilities can be leveraged

CADLM technology allows simulation users

For More Information: www.cadlm.com

Volume X - Winter 2019 | mscsoftware.com | 29

By Hendrik Schafstall, CEO Simufact

30 | Engineering Reality Magazine

Volume X - Winter 2019 | mscsoftware.com | 31