Spotlight on Ansys 12.0

EXCELLENCE IN ENGINEERING SIMULATION ADVANTAGEVOLUME III ISSUE 1 2009

SPOTLIGHT ON ANSYS 12.0

FRAMEWORK MULTIPHYSICS HIGH-PERFORMANCE PAGE 6 PAGE 11 COMPUTING PAGE 25

EDITORIAL

Engineering Simulation: t pa Needed Now

Time and cost benefits of engineering simulation are this simulation method has the potential to reduce crack well documented. Predicting product performance and growth analysis time by over 90 percent compared determining optimal solutions early in the design phase help with manual methods. The productivity gain will enable to avoid late-stage problems and to eliminate trial-and-error engineers to analyze more designs annually, thus keeping testing cycles that drive up costs and bog down schedules. up with increased demand for turbochargers around the Simulation enables engineers to perform what-if studies world and strengthening the company’s leadership and to compare alternatives, processes that otherwise position in this competitive industry sector. would be impractical. Indeed, bottom-line savings are The prediction method is based on improved fracture one key benefit that prompts most companies to implement mechanics capabilities for calculating J integrals, one of simulation, and are most readily quantified in return-on- the many enhancements in ANSYS 12.0. Previewed in the investment calculations. Spotlight section of this issue, the release is a milestone A second, and potentially greater, benefit is boosting for the software supplier and a huge step forward for the top-line revenue growth. With simulation, companies can CAE industry in terms of advancements in individual develop innovative, winning products that stand apart from physics (structural, fluid, thermal and electromagnetics) others, make the status quo obsolete or create entirely new and integration of this functionality into a unified multi- market opportunities. Brand value can be enhanced physics framework for Simulation Driven Product by tuning product performance to specific performance Development — an approach leading to top-line revenue characteristics. Revenue streams may be expanded by growth and bottom-line savings for many companies. increasing design throughput of new products or tackling Discussion of the business value of simulation is projects that otherwise would not be attempted. particularly relevant in today’s world as manufacturers How specific companies leverage simulation in face the toughest economic climate of a lifetime. Indeed, achieving these benefits depends on their unique products, with their survival at stake, forward-thinking companies engineering challenges and business requirements. The recognize the need to invest in engineering simulation now possibilities are limitless. Case in point is detailed in this more than ever to withstand the current market turbulence issue’s article “Predicting 3-D Fatigue Cracks without a and to strengthen their long-term competitive position, Crystal Ball” from Honeywell Turbo Technologies. Engineers brand value and profitability as conditions improve in the used software from ANSYS to predict thermomechanical coming years. ■ fatigue cracks in turbochargers for internal combustion engines. Predicting crack failures early enables engineers to optimize designs upfront and helps to avoid qualification test failures that lead to additional rounds of tests — which can be very expensive and take weeks to complete. Further, John Krouse, Senior Editor and Industry Analyst

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 1 CONTENTS Table of Contents

SPOTLIGHT ON ANSYS 12.0 4 ANSYS 12.0 Launching a New Era of Smart Engineering Simulation A full generation ahead of other solutions, ANSYS 12.0 takes product design and development to the next level. 6 FRAMEWORK 4 Introducing ANSYS Workbench 2.0 Proven simulation technology is delivered in a truly innovative integration framework. 8 GEOMETRY AND MESHING Taking Shape in 12.0 ANSYS combines depth of simulation with industry experience to provide geometry and meshing tools that realize simulation results faster. 11 MULTIPHYSICS Multiphysics for the Real World 8 In ANSYS 12.0, multiphysics capabilities continue to increase in flexibility, application and ease of use. 14 ELECTROMAGNETICS ANSYS Emag 12.0 Generates Solutions Improved accuracy, speed and platform integration advance the capabilities of low-frequency electromagnetic simulation. 15 FLUIDS A Flood of Fluids Developments A new integrated environment and technology enhancements make fluids simulation faster, more intuitive and more accurate.

11 18 STRUCTURAL MECHANICS Designing with Structure Advancements in structural mechanics allow more efficient and higher-fidelity modeling of complex structural phenomena. 22 EXPLICIT DYNAMICS Explicit Dynamics Goes Mainstream ANSYS 12.0 brings native explicit dynamics to ANSYS Workbench and provides the easiest explicit software for nonlinear dynamics. 23 EIGENSOLVER Introducing the Supernode Eigensolver A new eigensolver in ANSYS 12.0 determines large numbers of natural frequency modes more quickly and efficiently than conventional methods. 15 25 HIGH-PERFORMANCE COMPUTING The Need for Speed From desktop to supercomputer, high-performance computing with ANSYS 12.0 continues to race ahead. 28 FUTURE DIRECTIONS Foundations for the Future The many advanced features of ANSYS 12.0 were designed to solve today’s challenging engineering problems and to deliver a platform for tomorrow’s simulation technology.

2 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

CONTENTS

SIMULATION @ WORK DEPARTMENTS 31 AUTOMOTIVE 47 TIPS AND TRICKS Predicting 3-D Fatigue Cracks without Reusing Legacy Meshes a Crystal Ball ANSYS tools enable users to work with finite element models ANSYS tools quickly predict 3-D thermomechanical fatigue in various formats for performing simulations as well as cracking in turbocharger components. making changes to part geometry. 33 HEALTHCARE 49 ACADEMIC Electromagnetics in Medicine Expanding Stent Knowledge Electromagnetic and thermal simulations find use in medical Simulation provides the medical industry with a closer look applications. at stent procedures. 36 ELECTRONICS Keeping Cool in the Field A communications systems company gains millions of dollars by using thermal simulation to bring tactical radios to market faster.

38 BUILT ENVIRONMENT Designing Against the Wind Simulation helps develop screen enclosures that can better withstand hurricane-force winds.

40 ENVIRONMENT Stabilizing Nuclear Waste Fluid simulation solidifies its role in the radioactive waste 38 treatment process.

42 OPTIMIZATION Topology Optimization and Casting: A Perfect Combination Using topology optimization and structural simulation helps a casting company develop better products faster.

44 MARINE Fighting Fire with Simulation The U.K. Ministry of Defence uses engineering simulation to find alternatives to ozone-depleting substances for 49 fire suppression.

For ANSYS, Inc. sales information, call 1.866.267.9724, or visit www.ansys.com. For address changes, contact [email protected]. To subscribe to ANSYS Advantage, go to www.ansys.com/subscribe.

Executive Editor Editors Ad Sales Manager Circulation Manager About the Cover Fran Hensler Erik Ferguson Helen Renshaw Sharon Everts ANSYS introduces release 12.0, Shane Moeykens the next-generation technology Managing Editor Mark Ravenstahl Editorial Advisor for Simulation Driven Product Chris Reeves Kelly Wall Development. The spotlight Contributors begins on page 4. Senior Editor and Susan Wheeler Designer Industry Analyst Marty Mundy Miller Creative Group John Krouse

ANSYS Advantage is published for ANSYS, Inc. customers, partners and others interested in the field of design and analysis applications. Neither ANSYS, Inc. nor the senior editor nor Miller Creative Group guarantees or warrants accuracy or completeness of the material contained in this publication. ANSYS, ANSYS Workbench, Ansoft Designer, CFX, AUTODYN, FLUENT, GAMBIT, POLYFLOW, Airpak, DesignSpace, FIDAP, Flotran, Iceboard, Icechip, Icemax, Icepak, FloWizard, FLOWLAB, G/Turbo, MixSim, Nexxim, Q3D Extractor, Maxwell, Simplorer, Mechanical, Professional, Structural, DesignModeler, TGrid, AI*Environment, ASAS, AQWA, AutoReaGas, Blademodeler, DesignXplorer, Drop Test, ED, Engineering Knowledge Manager, Emag, Fatigue, Icepro, Icewave, Mesh Morpher, ParaMesh, TAS, TASSTRESS, TASFET, TurboGrid, Vista, VT Accelerator, CADOE, CoolSim, SIwave, Turbo Package Analyzer, RMxprt, PExprt, HFSS, Full-Wave SPICE, Simulation Driven Product Development, Smart Engineering Simulation and any and all ANSYS, Inc. brand, product, service, and feature names, logos and slogans are registered trademarks or trademarks of ANSYS, Inc. or its subsidiaries located in the United States or other countries. ICEM CFD is a trademark licensed by ANSYS, Inc. All other brand, product, wwwservice.ansys.com and feature names or trademarks are the property of their respective owners. ANSYS Advantage • Volume III, Issue 1, 2009 3 © 2009 ANSYS, Inc. All rights reserved.

ANSYS 12.0: Launching a New Era of Smart Engineering Simulation A full generation ahead of other solutions, ANSYS 12.0 takes product design and development to the next level.

By Jim Cashman, President and CEO, ANSYS, Inc.

The current economic climate has completely changed simulation in much the same way that the internet and desktop the way most companies view engineering simulation. publishing have revolutionized the broadband distribution Leveraging the power of virtual prototyping to compress the of information. As a direct consequence of a long-standing product development process and drive down costs is no commitment to simulation, ANSYS is the only company longer a choice — it’s a requirement for survival in an offering advanced simulation technologies that span all key increasingly competitive environment. engineering disciplines — and bringing them together in an In nearly every industry, driving product development integrated and flexible software platform designed specifically through engineering simulation technology has become a to support Simulation Driven Product Development. key strategy to develop more innovative products, reduce Over the years ANSYS has made significant technology development and manufacturing costs, and accelerate time investments, acquisitions and partnership to ensure continuing to market. leadership. We recognize that every technology breakthrough Backed by the unmatched power of ANSYS 12.0 or market accomplishment has only been a stepping stone to software, progressive companies are taking engineering our vision. Reflecting these investments — as well as the simulation a step beyond. They have already realized acquired wisdom of four decades in this industry — ANSYS the enormous strategic benefits of virtual prototyping — and 12.0 represents the fullest expression of our leadership posi- are now seeking more from their investments in simulation. tion. It is the most comprehensive engineering simulation ANSYS 12.0 enables these forward-looking companies solution available today. to maximize the efficiency of their simulation processes, to While the following pages offer a wealth of detail, I’d like to increase the accuracy of their virtual prototypes, and to focus on the high-level benefits that our customers will realize capture and reuse their simulation processes and data. This as they leverage the full depth and breadth of ANSYS 12.0 to next level of performance signals a new era of Smart make product development smarter, better, faster and more Engineering Simulation, in which product innovations can be collaborative than they ever thought possible. realized more rapidly, and more cost effectively, than ever before. Smart Technologies = Smart Simulation There is no company better qualified to launch this new At ANSYS, we have applied our long history of tech- era. ANSYS has led the engineering simulation industry nology leadership to create the world’s smartest solution for for nearly 40 years, revolutionizing the field of engineering engineering simulation — more automated, repeatable,

Some images courtesy FluidDA nv, Forschungszentrum Jülich GmbH, Heat Transfer Research, Inc., Riello SPA and © iStockphoto.com/iLexx.

444 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com persistent and intuitive than existing products. The ground- captured and automated with drag-and-drop ease. ANSYS breaking ANSYS Workbench 2.0 platform is a flexible 12.0 amplifies the capabilities and outputs of every member environment that allows engineers to easily set up, visualize of the engineering staff, enabling them to work smarter, and manage their simulations. ANSYS 12.0 offers to intelligently make design trade-offs and to rapidly unequalled technical breadth that allows customers to converge on the best designs. And, because ANSYS 12.0 is explore a complete range of dynamic behavior, from based on the most advanced technology and physics, frequency response to large overall motion of nonlinear design and engineering teams can commit to manufacturing flexible multibody systems. ANSYS has also leveraged operations with confidence — and without investing time its industry-leading capabilities to create an unequalled and money in exhaustive physical testing. depth of simulation physics, including the newly integrated ANSYS FLUENT solver, advancements in all key simulation Redefining Collaboration physics, and enabling technologies for meshing, geometry Real-world simulation projects often involve a wide and design optimization. ANSYS Engineering Knowledge variety of engineering personnel — and generate large Manager allows engineers to easily archive, search, retrieve volumes of data that must be shared across the enterprise. and report their simulation data via a local machine or a With its broad support of simulation disciplines and native centralized data repository. Not only does ANSYS 12.0 project management system, ANSYS 12.0 allows represent the smartest and best individual technologies, but engineering teams to collaborate more freely, without it brings them together in a customized, scalable solution software barriers or other technology obstacles. Within a that meets the highly specific needs of every engineering single project, several engineers can assess their designs team. Powerful and flexible, ANSYS 12.0 can be configured within individual disciplines, as well as easily coordinate for advanced or professional users, deployed to a single user multiphysics simulations. The single-project environ- or enterprise, and executed on laptops or massively parallel ment reduces redundancies and synchronization errors computer clusters. As customer requirements grow and among different engineering teams. ANSYS Engineering mature, ANSYS 12.0 is engineered to scale up accordingly. Knowledge Manager also provides the tools to manage the workflow of a group of engineers and a myriad of Better Prototypes, Better Products simulation projects. With its unique multiphysics, high-performance At ANSYS, we have always believed that engineering computing and complete system modeling capabilities, simulation is a sound investment — and today, it is emerging ANSYS 12.0 is a complete solution that takes virtual proto- as one of the smartest investments an organization can typing to a new level of accuracy, realism and efficiency. make. We understand the incredible time and cost pressures ANSYS 12.0 captures the response of a completely under which our customers operate today, and ANSYS 12.0 assembled system and assesses how a range of highly is specifically designed to help them meet these challenges. complex, real-world physical phenomena will affect not only In the new era of Smart Engineering Simulation heralded individual components but also their interactions with one by ANSYS 12.0, product development teams can work another. Flaws in product functionality can be recognized faster and more effectively than ever before — with a greater before investments are made in full-blown physical proto- degree of confidence in their finished products. Because it types — and ideas that are validated in the virtual world can provides a tremendous opportunity for engineers to design be fast-tracked to maximize agility and capture emerging higher-quality, more innovative products that are manu- market opportunities. Powered by fast and accurate solvers, factured faster, and at a lower cost, ANSYS 12.0 makes the design optimization with ANSYS 12.0 results in prototypes most compelling case yet for engineering simulation as a with a much higher probability of ultimate market success. powerful competitive strategy. But we are far from finished: ANSYS 12.0 is a milestone, not the destination, as we Product Design at Warp Speed continually work to put our tools in the hands of every ANSYS 12.0 automates many manual and tedious tasks engineer who can benefit from them. As the power of involved in simulation, reducing design and analysis cycles ANSYS 12.0 is unleashed by imaginative engineering teams by days or even weeks. An innovative project management around the world, I look forward to the amazing product system allows custom simulation workflows to be created, innovations that will result. ■

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 555

12.0: FRAMEWORK Introducing ANSYS Workbench 2.0 Proven simulation technology is delivered in a truly innovative integration framework.

ANSYS 12.0 delivers innovative, dramatic simulation technology advances in every major physics discipline, along with improvements in computing speed and enhancements to enabling technologies such as geometry handling, meshing and post-processing. These advancements alone represent a major step ahead on the path forward in Simulation Driven Product Development. But ANSYS has reached even further by delivering all this technology in an innovative simulation framework, ANSYS Workbench 2.0.

The ANSYS Workbench environ- The toolbox, at left, contains systems that form a project’s building blocks. In this single-physics example, the user drags ment is the glue that binds the the system (from left) into the project schematic (at right), then sets up and solves the system, working from the top simulation process; this has not down through the cells in the system. As shown, the Fluid Flow system (at right) is complete through mesh generation, as shown by green check marks. changed with version 2.0. In the original ANSYS Workbench, the user interacted as static structural analysis, the user schematic. Connections are created with the analysis as a whole using the locates the appropriate analysis automatically and data is transferred platform’s project page: launching the system in the toolbox and, using drag- behind the scenes, delivering drag-and- various applications and tracking the and-drop, introduces it into the project drop multiphysics with unprecedented resulting files employed in the process schematic. That individual system con- ease of use. of creating an analysis. Tight integration sists of multiple cells, each of which The ANSYS Workbench environ- between the component applications represents a particular phase or step ment tracks dependencies among the yielded unprecedented ease of use for in the analysis. Working through the various types of data in the project. If setup and solution of even complex system from the top down, the user something changes in an upstream multiphysics simulations. completes the analysis, starting with a cell, the project schematic shows that In ANSYS 12.0, while the core parametric connection to the original downstream cells need to be updated applications may seem familiar, they CAD geometry and continuing through to reflect these changes. A project- are bound together via the innovative to post-processing of the analysis level update mechanism allows these project page that introduces the result. As each step is completed, changes to be propagated through all concept of the project schematic. progress is shown clearly at the project dependent cells and downstream This expands on the project page level. (A green check mark in a cell indi- systems in batch mode, dramatically concept. Rather than offer a simple cates that an analysis step has been reducing the effort required to repeat list of files, the project schematic completed.) variations on a previously completed presents a comprehensive view of Passing files and data from one analysis. the entire analysis project in flow- application to the next is managed Parameters are managed at the chart form in which explicit data entirely by the framework, and data project level, where it is possible to relationships are readily apparent. and state dependencies are directly change CAD and geometry parameters, Building and interacting with these represented. More-complex analyses material properties and boundary flowcharts is straightforward. A toolbox can be constructed by joining multiple condition values. Multiple parametric contains a selection of systems that systems. The user simply drags a new cases can be defined in advance form the building blocks of the project. system from the toolbox and drops and managed as a set of design To perform a typical simulation, such it onto the existing system in the points, summarized in tabular form

666 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com 12.0: FRAMEWORK

on the ANSYS Workbench project page. Design Exploration systems can be Managing Simulation Data connected to these same project-level With the ever-increasing use of simulation, keeping track of the parameters to drive automated design expanding volume of simulation data becomes more and more difficult. investigations, such as Design of Experi- The need to be able to quickly locate information for reuse is paramount to ments, goal-driven optimization or Design increasing productivity and reducing development costs. for Six Sigma. ANSYS EKM Desktop is a new tool, integrated in the ANSYS In addition to serving as a framework Workbench environment, that facilitates managing simulation data from for the integration of existing applications, multiple projects. ANSYS EKM Desktop is a single-user configuration the ANSYS Workbench 2.0 platform also of EKM that allows users to add files from any project to a local virtual serves as an application development repository. Simulation properties and other metadata are automatically framework and will ultimately provide extracted (or created) from files when added, and users can tag files with project-wide scripting, reporting, a user unique identifiers at any time. These attributes can all be used to search interface (UI) toolkit and standard data and retrieve files based on keywords or complex search criteria. Reports interfaces. These capabilities will emerge can be easily generated to allow efficient side-by-side comparison of the over this and subsequent releases. At attributes of related analyses. Search queries and reports can be saved for ANSYS 12.0, Engineering Data and later re-use. Files that are retrieved can be directly launched in their associ- ANSYS DesignXplorer are no longer ated simulation application from within the ANSYS EKM Desktop tool. independent applications: They have been re-engineered using the UI toolkit and integrated within the ANSYS Workbench project window. Beyond managing individual simulation projects, ANSYS Workbench interfaces with the ANSYS Engineering Knowledge Manager (EKM) product for simulation process and data management. At ANSYS 12.0, ANSYS Workbench includes the single-user configuration of ANSYS EKM, called ANSYS EKM Desktop. (See sidebar.) ANSYS Workbench 2.0 represents a sizable step forward in engineering simulation. Within this innovative software framework, analysts can leverage a complete range of proven simulation technology, including common tools for More-complex analyses involving multiple physics can be built up by connecting systems. Data dependencies are CAD integration, geometry repair and indicated clearly as connections. State icons at the right of each cell indicate whether cells are up to date, require user meshing. A novel project schematic input or need to be updated — for example, whether they are just meshed or fully solved. concept guides users through complex analyses, illustrating explicit data relationships and capturing the process for automating subsequent analyses. Meanwhile, its parametric and persistent modeling environment in conjunction with integral tools for design optimization and statistical studies enable engineers to arrive at the best design faster. Looking beyond ANSYS 12.0, the ANSYS Workbench platform will be further refined: The aim is to deliver a comprehensive set of simulation technology in an open, adaptive software architecture that allows for pervasive customization and the integration of third-party applications. ■

Judd Kaiser, Shantanu Bhide, Scott Gilmore and Todd Two analyses from the schematics shown in the previous figure are shown here in the mechanical simulation application. McDevitt of ANSYS, Inc. contributed to this article. Launched from the schematic, individual applications may be familiar to existing users. www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 7

12.0: GEOMETRY AND MESHING

Taking Shape in 12.0 Automated cleanup and repair of imported geometry: ANSYS combines depth of simulation New tools automatically detect and fix typical problems, such as small edges, sliver faces, holes, seams and faces with industry experience to provide with sharp angles. Geometry models can now be prepared for analysis at a much faster pace. These images show an geometry and meshing tools that realize aircraft model before (top) and after (bottom) cleanup. simulation results faster.

Engineering simulation software CFX tools into the ANSYS meshing for IGES, STEP, ACIS®, Parasolid®, users have been known to spend up to platform — which provides the foundation CATIA® V4 and CATIA V5. At ANSYS 90 percent of their simulation-related for unifying and leveraging meshing tech- 12.0, geometry interfaces have been time working on pre-processing tasks. nologies, making them interoperable and enhanced to import more information By targeting developments in capabilities available in multiple applications. Taking from CAD systems, including new data to increase ease of use, simplifying advantage of the enhanced ANSYS types such as line bodies for modeling pre-processing tasks, and increasing the Workbench 2.0 framework, the company beams, additional attributes such as capabilities of pre-processing tools, provides further significant improve- colors and coordinate systems, and ANSYS has systematically delivered ments for ANSYS 12.0 geometry and improved support for named selections exciting advances to increase the meshing applications. created within the CAD systems. efficiency of simulation. For pre-processing larger models, ANSYS has combined rich CAD Connections release 12.0 includes support for 64-bit geometry and meshing techniques ANSYS continues to deliver a leading operating systems, and smart and with its depth of knowledge and CAD-neutral CAE integration environ- selective updates of CAD parts. The experience, and the end result is ment, providing direct, associative and newly introduced ability to selectively products capable of harnessing bi-directional interfaces with all major update CAD components allows users integrated geometry and meshing CAD systems, including Unigraphics®, to update individual parts instead of an solutions that share core libraries Autodesk® Inventor®, Pro/ENGINEER®, entire assembly, thus making geometry with other applications. At releases CATIA® V5, PTC CoCreate® Modeling, updates much faster and more targeted. 10.0 and 11.0, ANSYS introduced SolidEdge®, SolidWorks®, and Autodesk® robust, new meshing capabilities Mechanical Desktop®. Software from from ANSYS ICEM CFD and ANSYS ANSYS also supports file-based readers

“ANSYS 12.0 will set the stage for major improvements in our design processes. Two of Cummins’ core tools, ANSYS FLUENT and ANSYS Mechanical, are coming together in the ANSYS Workbench environment. I am also very pleased to see that geometry import continues to improve, and we have several more meshing options.”

Improved surface extension: Users can select and extend — Bob Tickel multiple groups of surfaces in a single step, a procedure Director of Structural and Dynamic Analysis that greatly simplifies the process of closing gaps between parts after mid-surface extraction. The images show a Cummins, Inc. sample model before and after surface extension.

888 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

12.0: GEOMETRY AND MESHING

Geometry Handling in Improved attribute support is ANSYS DesignModeler available with ANSYS DesignModeler Geometry modeling in the ANSYS 12.0. This includes options to create Workbench environment is greatly attributes within ANSYS DesignModeler improved to provide increased as well as to import additional attributes automation, greater flexibility and from external CAD, including named improved ease of use for the task of selections, coordinate systems and preparing geometry for analysis. The work points. feature-based, parametric ANSYS DesignModeler tool, which can be used ANSYS Meshing Platform Thin solid sweep method: Using the thin solid sweep mesh method, complicated sheet metal parts can be easily hex to create parametric geometry from A primary focus for ANSYS 12.0 has meshed without the need for midsurfacing or welding. The scratch or to prepare an existing CAD been to provide an automated meshing mesh can be generated to conform to the shared interface to increase the accuracy and speed of the solution. geometry for analysis, now includes solution that is best in class for fluid automated options for simplification, dynamics. With the addition of cap- cleanup, repair and defeaturing. abilities from GAMBIT and TGrid Merge, Connect and Project meshing applications, major improve- features have been added for improved ments have been made in the automatic surface modeling in ANSYS 12.0. Face generation of CFD-appropriate tetra- and Edge merge operations can be hedral meshes with minimal user input. used to easily simplify models by Advanced size functions (similar to those eliminating unnecessary features and found in GAMBIT), prism/tet meshing boundaries, leading to improved mesh (from TGrid) and other ANSYS meshing and solution quality. The Connect technologies combine to provide operation can be applied to ensure improved smoothness, quality, speed, Patch conformal tet method with advanced size functions: proper connectivity in models with gaps curvature and proximity feature With minimal input, ANSYS size function–based triangulation and inflation technology can handle advanced CFD meshing and overlaps. capturing, and boundary layer capturing. challenges, such as this benchmark aircraft model. Automated cleanup and repair In the area of hex meshing, the tra- capabilities have been improved in the ditional sweep and thin sweep methods 12.0 release. New tools automatically have seen evolutionary improvements. In the area of hybrid meshing, the detect and fix typical problems, such as A new method called MultiZone has MultiZone method allows for compli- small edges, sliver faces, holes, seams been integrated into the ANSYS cated regions to be meshed with a and faces with sharp angles. Geometry meshing platform. By combining hybrid mesh (tet, hex-core, hex-domi- models can now be prepared for analy- existing ANSYS ICEM CFD Hexa nant), further improving the flexibility and sis at a much faster pace. As always, technology with improvements in automation of this meshing approach. analysis settings remain persistent after automation, MultiZone allows the user For more control in key areas of concern, performing these operations and are to automatically create hex meshes for the Sweep and Patch Conforming updated automatically in response to many complex geometries without methods can be employed with changes in geometry. requiring geometry decomposition. conformal inflation layers throughout. Shell modeling has been enhanced Though many of these enhance- in several ways, including improved ments were driven by fluid dynamics surface extensions. The ability to select needs, they also benefit users of other and extend groups of surfaces greatly types of simulation. For example, users simplifies the process of closing gaps performing structural analyses will benefit between parts after mid-surface extrac- from the improved automation and mesh tion. The result is easier modeling of quality. Additional meshing enhance- welds, for example. ments for structural analyses include: Analysis-specific tools within the • Physics-based meshing ANSYS DesignModeler product now improvements include an automated option to extract • Rigid body meshing for contact flow volumes for fluid dynamics analy- • Automated meshing of gaskets ses. In addition, several new features, including user-defined offsets, user- • Improved handling of beams defined cross sections and better MultiZone mesh method: Using the new MultiZone mesh • Thin solid meshing improvements orientation controls, are available for method, a user can mesh complicated models with a pure • Support for multiple elements hex mesh without the need for geometry decomposition. improved beam modeling for structural This brake rotor example can be meshed with a pure hex through the thickness analyses. mesh in a single operation.

wwwwww.ansys.com.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 999 12.0: GEOMETRY AND MESHING

technology within the ANSYS meshing platform and continued development to enhance the ANSYS ICEM CFD product for interactive meshing customers. Because the ANSYS ICEM CFD integration involves the sharing of core libraries, improvements made for the ANSYS meshing platform also enhance the ANSYS ICEM CFD meshing product (and vice versa). New developments in the ANSYS TurboGrid software MultiZone meshing is an example are used to create high-quality meshes for bladed Hybrid mesh: Using a combination of sweep and tetra- components with minimal user input. hedral mesh methods, a user can quickly control the mesh of a crossover technology that has Geometry courtesy PCA Engineers. in regions of interest to improve the accuracy of the received special attention in both solution without the need for a pure hex mesh (and the time required to generate it). ANSYS meshing and the stand-alone ANSYS ICEM CFD meshing product. • Generation of conformal meshes This hybrid meshing method combines Enhancements to in multi-body parts the strengths of various meshers, such Turbomachinery Tools • Enhanced and new mesh as ANSYS ICEM CFD Hexa and TGrid, With release 12.0, a number of controls in a semi-automatic blocking frame- enhancements have been incorpo- rated into ANSYS BladeModeler, • Pinch features to help in work. Within the ANSYS Workbench the design tool tailored to bladed defeaturing models environment, multizone automation provides multi-source, multi-target geometries for rotating machinery. • Improved smoothing and multi-direction sweep capabilities Within the BladeGen component, • Improved flexibility in size reminiscent of the GAMBIT Cooper tool. the integrated tools for determining controls and mesh refinement In the stand-alone ANSYS ICEM CFD initial blade shape and size (which • Arbitrary mesh matching to product, this is an excellent way to were developed in conjunction with improve node linking and mesh for external aerodynamics in a partner PCA Engineers Limited) solver accuracy semi-automated way that provides have been expanded to cover cen- rapid hybrid meshing with a high degree trifugal compressors and axial fans These improvements, though driven of control and quality. in addition to radial turbines and by structural analysis needs, provide Improvements for ANSYS ICEM centrifugal pumps. The other com- benefits to the entire spectrum of CFD 12.0 include process and interface ponent of ANSYS BladeModeler, ANSYS users. streamlining, new hexa features, BFCart BladeEditor, includes new blade mesher enhancements, mesh editing geometry modeling capabilities to ANSYS ICEM CFD advancements, output format updates create and modify one or more For ANSYS 12.0, ANSYS ICEM CFD and more. ■ bladed components. As an add-in to meshing development focused on two ANSYS DesignModeler, ANSYS Ben Klinkhammer, Shyam Kishor, Erling Eklund, primary tasks: improved implement- Simon Pereira and Scott Gilmore of ANSYS, Inc. BladeModeler provides access to ation of ANSYS ICEM CFD meshing contributed to this article. ANSYS DesignModeler’s extensive functionality to create nonstandard geometry components and features. ANSYS TurboGrid software includes a number of evolutionary improvements in release 12.0, and introduces a completely new meshing technology. This tool fully automates a series of topology and smoothing steps to largely eliminate the need to manually adjust mesh controls, yet still generates high-quality Named selection manager: This new feature allows a user ANSYS ICEM CFD: MultiZone meshing that combines to create and save named selections within CAD systems the strength of various meshing tools, automatically fluid dynamics meshes for bladed and then to use them within ANSYS applications. This generated this hybrid grid for a tidal turbine. turbomachinery components. example uses the named selection manager within Pro/ENGINEER.

101010 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

12.0: MULTIPHYSICS

Multiphysics for the Real World In ANSYS 12.0, multiphysics capabilities continue to increase in flexibility, application and ease of use.

Continuing to build on the foundation of prior releases, ANSYS 12.0 expands the company’s industry-leading comprehensive multiphysics solutions. New features and enhancements are available for solving both direct and sequentially coupled multiphysics problems, and the ANSYS Workbench framework makes performing multiphysics simulations even faster than before.

ANSYS Workbench Integration The integration of the broad array of ANSYS solver technologies has taken a considerable step forward with release 12.0. The ANSYS Workbench environment has been redesigned for an efficient multiphysics workflow by inte- grating the solver technology into one unified simulation environment. This platform now includes drag-and-drop The electric potential for the transformer busbar shown here was analyzed within the ANSYS Workbench environment and required the use of temperature-dependent multiphysics, which allows the user to easily set up and material properties. Courtesy WEG Electrical Equipment. visualize multiphysics analysis, significantly reducing the time necessary to obtain solutions to complex multiphysics temperature-dependent material properties and advanced problems. thermoelectric effects, including Peltier and Seebeck effects. Another new enhancement to the ANSYS Workbench The applications for this new technology include Joule framework is the support for steady-state electric conduc- heating of integrated circuits and electronic traces, tion. There is a new analysis system that exposes 3-D solid busbars, and thermoelectric coolers and generators. electric conduction elements (SOLID231 and SOLID232) in the ANSYS Workbench platform. All the benefits of this Solver Performance popular environment — leveraging CAD data, meshing ANSYS 12.0 extends the distributed sparse solver to complex geometry and design optimization features — are support unsymmetric and complex matrices for both shared now available for electric conduction analysis. and distributed memory parallel environments. This new Also new in ANSYS Workbench at version 12.0 is sup- solver technology dramatically reduces the time needed to port for direct coupled-field analysis. Relevant elements perform certain direct coupled solutions including Peltier and (SOLID226 and SOLID227) are now natively supported Seebeck effects as well as thermoelasticity. Thermo- in the ANSYS Workbench platform for thermal–electric elasticity, including thermoelastic damping, is an important coupling. There also is a new analysis system for thermal– loss mechanism for many MEMS devices, such as block electric coupling that supports Joule heating problems with resonators and silicon ring gyroscopes.

Elements A new family of direct coupled- field elements is available in ANSYS 12.0; these new elements enable the modeling of fluid flow through a porous media. This exciting new

The project schematic shows the multiphysics workflow for a coupled electric conduction, heat transfer and capability, comprising coupled subsequent thermal stress analysis. pore–pressure mechanical solids,

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 111111 12.0 MULTIPHYSICS

enables multiphysics modeling of new classes of civil and biomed- ical engineering problems that rely on fluid pore pressures. The elements allow users to model fluid pore pressures in soils (for simulating building foundations) and biometric materials (for modeling bone in order to develop prosthetic implants).

Fluid Structure Interaction One of the major enhancements for fluid structure interaction (FSI) is a new immersed solid FSI solution. This technique is based on a mesh superposition method in Sequence of images showing simulation of the motion of a screw pump solved using immersed which the fluid and the solid are meshed independently solid fluid structure interaction from one another. The solution enables engineers to model fluid structure interaction of immersed rigid solids with 4 imposed motion. Rotating, translating and explicit motion of rigid–solid objects can be defined, and the CFD solver 3 accounts for the imposed motion of the solid object in the fluid. This solution technique provides rapid FSI simulations, since there is no need to morph or remesh the fluid mesh 2 based on the solid motion. The model preparation for the Scale of Solution Speed new immersed solid technique is also very straightforward: 1 The entire setup for the FSI solution can be performed 1 2 3 4 5 6 7 8 9 10 11 12 entirely within ANSYS CFX software. This technology is especially applicable to fluid structure interaction problems Number of Processors with large imposed rigid-body motions, such as closing Solution scaling of a thermoelectric cooler model with 500,000 degrees of freedom enables a speedup of four valves, gear pumps and screw compressors. The method is times for 12 processors. also useful for rapid first-pass FSI simulations.

Coupling Electromagnetics

By joining forces with Ansoft, ANSYS can deliver simulation environment started almost immediately after greater multiphysics capabilities — specifically electro- the acquisition. While the combined development team is magnetics — to the ANSYS suite. The plan to integrate this working toward a seamlessly integrated bidirectional electromagnetics technology within the existing ANSYS solution, several electromagnetic-centric case studies already have demonstrated the ability to couple electromagnetic, thermal Start and structural tools within the adaptive architecture of the ANSYS Workbench environment. Create and solve the electromagnetic Import the geometry into application using HFSS ANSYS Mechanical and create the For example, a high-power elec- corresponding ANSYS thermal model tronic connector used in a radar application to connect a transmitter to

Export geometry and thermal link file an antenna must be engineered from from HFSS to ANSYS Mechanical Import surface and/or volumetric electromagnetic, thermal and structural losses using the imported load option (beta) in ANSYS Workbench perspectives to ensure success. The simulation was performed by coupling Ansoft’s HFSS software with the

ANSYS Workbench runs HFSS in batch Solve the ANSYS thermal model and ANSYS Workbench environment, using End to perform the load interpolation post-process the thermal results advanced thermal and structural capabilities. Engineers used HFSS to ensure Case study procedure of one-way coupling between Ansoft (blue) and ANSYS (yellow) software that the device was transmitting in the

12 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

12.0: MULTIPHYSICS

temperatures or surface forces between ANSYS FLUENT and ANSYS mechanical products based on ANSYS CFX- Post. The most appropriate applications include those that require one-way transfer of fluid pressures or temperatures from CFD to a mechanical analysis, such as automotive exhaust manifolds, heat sinks for electronics cooling and turbomachinery.

Multi-Field Solver The multi-field solver (used for performing implicit sequential coupling) contains a number of new enhancements at release 12.0. The first is a new solution option that The results of an RF MEMS switch solved by coupling the electrostatic, fluid and controls writing a multiframe restart file. This capability mechanical behavior of the switch in one analysis using FLUID136 to represent squeeze allows a user to restart an analysis from any multi-field time film effects. Image courtesy EPCOS NL and Philips Applied Technologies. step, which allows for better control over the availability of a restart file with less hard drive usage. Another enhancement Another new capability for fluid structure interaction in is more-flexible results file controls. This capability reduces ANSYS 12.0, FLUID136 now solves the nonlinear Reynolds the results file sizes for the multi-field solver, and it allows for squeeze film equations for nonlinear transient FSI applica- synchronizing the fluid and mechanical results in an FSI tions involving thin fluid films. Since the nonlinear fluidic and solution. The final improvement is new convergence con- structural responses are coupled at the finite element level, trols for the multi-field solution to provide more flexible the solution is very fast and robust for thin fluid film applica- solution controls for nonlinear convergence of the multi-field tions. Any squeeze film application can benefit from this solver. The applications for these enhancements are any technology, including thin film fluid damping often found in multiphysics application using sequential coupling including RF MEMS switches. fluid structure interaction. ■ Version 12.0 offers another exciting new FSI capability: the ability to perform one-way fluid structure interaction using ANSYS FLUENT software as the CFD solver. Stephen Scampoli of ANSYS, Inc. and Ansoft LLC technical specialists This capability enables one-way load transfer for surface contributed to this article.

proper path, by calculating the high-frequency electromagnetic fields, power loss density distribution and S-parameters. In such high-power applications, it is critical to determine the temperature distribution to ensure the device stays below temperatures that cause material failure, such as melting. The power loss density results from the HFSS simulation were used

as the source for the thermal Deformation of the high-power electronic connector can be predicted by combining simulation performed within Ansoft HFSS and ANSYS Mechanical software. ANSYS Mechanical software, which simulated the tempera- solenoid. The power loss was used as an input for a thermal ture distribution of the device. simulation performed with ANSYS Mechanical software to In another case, a valve- determine the temperature profile of the device. Subse- actuating solenoid application quently, the application predicted how the device deformed used a coupled ANSYS and due to the rise in temperature. Such coupling delivers a Ansoft simulation to analyze powerful analysis framework needed to solve these complex, temperature distribution. interrelated physics problems. Thus, engineers can Maxwell software was used to address electro-thermal-stress problems associated with calculate the power loss from optimizing state-of-the-art radio frequency (RF) and electro- the low-frequency electromechanical components including antennas, actuators, Eddy current and conduction loss calculated by Ansoft’s Maxwell software magnetic fields within the power converters and printed circuit boards (PCBs).

ANSYS Advantage • Volume III, Issue 1, 2009 13 12.0: ELECTROMAGNETICS

SOLID236 ANSYS Emag 12.0 3-D 20-node brick Generates Solutions

Improved accuracy, speed and platform integration advance SOLID237 3-D 10-node the capabilities of low-frequency electromagnetic simulation. tetrahedron

As the combined development teams from Ansoft and and SOLID237 elements support both distributed and ANSYS set out to integrate the world-class Ansoft electronic shared-memory parallel processing for low-frequency design products into the ANSYS portfolio, ANSYS electromagnetic solutions. As a result of faster simulation customers can benefit immediately from improved and speeds, users can solve much larger and more complex extended electromagnetics capabilities in release 12.0. low-frequency electromagnetic models.

Elements ANSYS Workbench Integration A new family of 3-D solid elements for low-frequency Release 12.0 offers several ANSYS Workbench electromagnetic simulation is included in the 12.0 release of enhancements for electromagnetic simulation. A new ANSYS Emag software. Solid elements (SOLID236 and capability facilitates multiple load step analysis for magneto- SOLID237) are available for modeling magnetostatic, quasi- statics. This allows users to compute the magnetostatic static time harmonic, and quasi-static time-transient response to time-dependent loading, specifying voltage and magnetic fields. These two elements are formulated using current loads with time-dependent an edge-based magnetic vector potential formulation, tabular data. The results are which allows for improved accuracy for low-frequency more flexibility for magneto- electromagnetic simulation. The elements also provide a static problems with true volt degree of freedom — as opposed to a time- time-dependent loads integrated electric potential — enabling circuit coupling along with transient with discrete circuit elements and simplifying pre- simulation for elec- and post-processing for electromagnetic simulation. tromagnetics, with SOLID236 and SOLID237 also include much faster the addition of a gauging than prior releases, which significantly reduces simple command overall solution times. Users can apply this new element snippet, within the technology to most low-frequency electromagnetic ANSYS Workbench applications, such as electric motors, solenoids, environment. electromagnets and generators. The integrated plat- Nonlinear transient rotational test form also includes an rig solved in the ANSYS Workbench Solvers option for a meshed environment using SOLID236, SOLID237 and the new stranded conductor option At release 12.0, the distributed sparse solver includes representation of a (TEAM24 benchmark) support for low-frequency electromagnetics. SOLID236 stranded conductor. The current density for the new stranded conductor DANSYS for Low-Frequency Electromagnetics supports tabular loading for the new multi-step mag- 5 netostatic analysis. This capability allows for a more accurate representation of current, improves overall 4 simulation accuracy and leverages existing CAD data for 3 coil geometry. This new ANSYS Workbench technology can be applied to any electromagnetic application 2 Solutions Speedup subject to time-dependent loading, including electric machines, solenoids and generators. ■ 1 12 3 4 5 6 7 8 Number of Processors Solution scaling of a SOLID237 model with 550,000 degrees of freedom Stephen Scampoli of ANSYS, Inc. contributied to this article.

141414 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com 12.0: FLUIDS

A Flood of Fluids Developments A new integrated environment and technology enhancements make fluids simulation faster, more intuitive and more accurate.

With release 12.0, ANSYS con- access to bidirectional CAD tinues to deliver on its commitment to connections, powerful geometry develop the world’s most advanced modeling and advanced mesh genera- fluid dynamics technology and make tion. (See the article Taking Shape in it easier and more efficient to use. 12.0.) Users can examine analysis Fuel injector model with close-up of vapor volume Through its use, engineers can results in full detail using CFD-Post, fraction contours at the injector surface develop the most competitive prod- also available within the ANSYS ucts and manufacturing processes Workbench environment. possible. In addition to delivering growth and needs of high-performance numerous new advancements in Multiphysics computing. (See the article The Need physics, numerics and performance, In some cases, fluid simulations for Speed.) ANSYS has combined the function- must consider physics beyond basic The perennial goal of improving ality of both ANSYS CFX and ANSYS fluid flow. Both ANSYS CFX and accuracy without sacrificing robustness FLUENT into the ANSYS Workbench ANSYS FLUENT technologies provide motivated numerous developments, platform. Customers can use this many multiphysics simulation options including new discretization options integrated environment to leverage and approaches, including coupling such as the bounded second-order simulation technology, including to ANSYS Mechanical software to option in ANSYS FLUENT and the superior CAD connectivity, geometry analyze fluid structure interaction iteratively-bounded high-resolution creation and repair, and advanced (FSI) within the ANSYS Workbench discretization scheme in ANSYS CFX. meshing, all engineered to improve environment. Being able to consistently use higher- simulation efficiency and compress Another new capability is the order discretization schemes means the overall design and analysis cycle. immersed solid technique in ANSYS that users will see further increases in CFX 12.0 that allows users to include the accuracy of flow simulations without Integration into ANSYS Workbench the effects of large solid motion penalties in robustness. ANSYS 12.0 introduces the full in their analyses. (See the article integration of its fluids products into Multiphysics for the Real World.) User Interface ANSYS Workbench together with the Ease of use has been enhanced in capability to manage simulation General Solver Improvements various ways. Most noticeably, the workflows within the environment. This ANSYS continues to make ANSYS FLUENT user interface has allows users — whether they employ progress on basic core solver speed, a taken a significant step forward by ANSYS CFX or ANSYS FLUENT soft- benefit to all users for all types of appli- adopting a single-window interface ware (or both) — to create, connect cations, steady or transient. A suite of paradigm, consistent with other and re-use systems; perform auto- cases that span the range of industrial applications integrated in ANSYS mated parametric analyses; and applications has consistently shown Workbench. A new navigation pane seamlessly manage simulations increases in solver speed of 10 to 20 and icon bar and new task pages and using multiple physics all within percent, or even more, for both ANSYS tools for graphics window manage- one environment. CFX and ANSYS FLUENT software. ment all reflect a more modern and The integration of the core CFD Beyond core solver efficiency, improve- intuitive interface while providing products into the ANSYS Workbench ments to various aspects of parallel access to the previous version’s menu environment also provides users with efficiency address the continued bar and text user interface. wwwwww.ansys.com.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 151515 12.0: FLUIDS

For ANSYS CFX software, a host of improvements have been added to the graphical user interface (GUI). There is a completely new capability that allows users to customize GUI appearance, including the option to create additional input panels. These custom panels provide the ability to encapsulate best practices and common processes by giving the user control over GUI layout and required input. Evolution of the free surface of oil in a reciprocating compressor. The blue area is the gas/oil rotating domain inside the shaft, and the gray surface at the bottom shows the oil level of the reservoir. As the shaft rotates, oil is pumped Specific Focus Areas up due to body forces. Image courtesy Embraco. Internal Combustion Engines Internal combustion (IC) engines Multiphase bladed geometry design and mesh gen- are a primary target application for Multiphase flow modeling con- eration, continue to evolve and improve. the development of numerous tinues to receive a great deal of (See the Geometry and Meshing article features. While this development is development attention, in terms of for more details.) driven by the specific needs of IC numerics and robustness improve- An exciting new development for turbo- engine simulations, it benefits many ments as well as extended modeling machinery analysts is the introduction of other applications and users: capabilities. ANSYS FLUENT software the through-flow code ANSYS Vista™ TF. • New options and flexibility for extends the single-phase coupling Developed together with partner PCA handling variations in physics technology, introduced previously for Engineers Limited, Vista TF complements complexity required at different the pressure-based solver, to include full 3-D fluid dynamics analysis to provide phases of analyses Eulerian multiphase simulations. This basic performance predictions on one or • Further-integrated options enhancement provides more robust more bladed components in a matter of and controls for remeshing, convergence, especially for steady- seconds, allowing users to quickly and including an IC-specific option state flows. ANSYS CFX users will find easily screen initial designs. for setting up an entire engine that improvements to the option to simulation include solution of the volume fraction And More … equations as part of the coupled set of These enhancements represent just • Extensions and improvements equations make it more broadly usable the tip of the iceberg in new and to discrete particle-tracking in applications with separate velocity improved models and capabilities within capabilities fields for each phase. Other modeling core fluids products from ANSYS. Some • Numerous enhancements to enhancements include the implemen- other new developments include: combustion models and their tation of a wall boiling model and • Turbulence modeling extensions usability additional non-drag forces in ANSYS and improvements CFX as well as more robust cavitation ■ Reynolds-averaged Navier– and immiscible fluid models in ANSYS Stokes (RANS) models FLUENT. ■ Laminar–turbulent transition

Turbomachinery ■ Large eddy simulation (LES) The significant proportion of cus- ■ Detached eddy simulation (DES) tomers using products from ANSYS for the design and optimization of rotating ■ Scale-adaptive simulation (SAS) machinery ensured that this field • Ability to use real gas properties received a substantial development with the pressure-based solver in focus. This latest release contains a ANSYS FLUENT and, therefore, variety of enhancements to core solver include these in reaction modeling Internal combustion engine simulation is one of the technology that couple rotating and focus applications for ANSYS 12.0. This snapshot from a • Faster, more accurate chemistry transient simulation of the complete engine cycle shows stationary components more robustly, the flow just after the intake valves open and the direct more accurately and more efficiently. across the board injection of fuel. New flow feature extraction options in CFD-Post are used to highlight vortex structures with ANSYS BladeModeler and ANSYS • Dramatic speedups in view factor velocity vectors. Image courtesy BMW Group. TurboGrid, specialized products for calculations in ANSYS FLUENT

16 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

12.0: FLUIDS

“ANSYS CFX 12.0 showed a 30 percent solver speedup in comparison with the previous release. This significant improvement allows us to examine more design variations in the same time, enabling further design optimization and considerably reducing the total development time. This helps Embraco bring our products to the market more quickly.”

— Celso Kenzo Takemori Product and Process Technology Management Embraco

• Inclusion of convective terms in CFX-Post application, CFD-Post pro- Conclusion solids to model conjugate heat vides a complete range of graphical This is only a sampling of what the transfer in moving solids in post-processing options to allow users fluid dynamics development teams ANSYS CFX to visualize and assess the flow predic- have produced for ANSYS 12.0. The • Ability to model thin surfaces in tions they have made and to create combined depth and breadth of ANSYS CFX insightful 2-D and 3-D images and CFD knowledge and experience is animations. The application includes delivering benefits to all users as • Much more in areas such as powerful tools for quantitative analysis, technologies are combined and devel- particle tracking, fuel cells, such as a complete range of options for opment teams drive simulation acoustics, material properties calculating weighted averages and technology to new levels of achieve- and population balance methods automatic report-generation capabil- ment. With release 12.0, ANSYS ities. All steps can be scripted, allowing continues its commitment to provide CFD-Post for fully automated post-processing. leading-edge CFD technology. ■ An exciting introduction is the Among the specific enhancements in common post-processing application release 12.0 are the ability to open and This article was written through contributions from Chris Wolfe and John Stokes of ANSYS, Inc. CFD-Post. The result of combining compare multiple cases in the same technologies from both ANSYS CFD-Post session and the addition of FLUENT and ANSYS CFX tools and tools to locate vortex cores in the building upon the well-established predicted flow field.

In work sponsored by BMT Seatech, partially-filled tanks on marine vessels are being simulated by researchers at the University of Southampton to predict structural loads and changes in vessel behavior due to the sloshing of the fluid. CFD-Post can be used to compare multiple designs directly, both by examining them side by side and by looking at the calculated difference between results. Geometry courtesy CADFEM GmbH.

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 171717 12.0: STRUCTURAL MECHANICS

Warping and ovalization of pipe structures with the new pipe elements

Designing with Structure Advancements in structural mechanics allow more efficient and higher-fidelity modeling of complex structural phenomena.

The ability to drive the engineering design process in this requires local remeshing during the simulation structural applications has taken a significant step forward process. The 2-D rezoning introduced with release 11.0 with the improvements in release 12.0. New features and extends further in ANSYS 12.0, increasing the flexibility of tools, many integrated into the ANSYS Workbench platform, the remeshing process: The user can now define transition help reduce overall solution time. Specific improvements regions within the refined zones and use meshes created focus on elements, materials and contact and solver in external meshing tools. performance, along with linear, rigid and flexible dynamics. Materials Elements Accounting for proper cyclic softening or hardening or The most notable new element in release 12.0 is the damage of materials is a key factor for elastomer applica- four-noded tetrahedron for modeling complex geometries in tions and, more generally speaking, any structure whose hyperelastic or forming applications. The element provides material variation depends on the strain rate. Release 12.0 a convenient way to automate the meshing of complex introduces several additions to the wide choice of materi- structures, avoiding the need for pure hexahedral meshes. als already available. Other feature improvements include: This reduces the time it takes to develop a case from geom- • Rate-dependent Chaboche plasticity, which can etry through solution, while maintaining the accuracy of the benefit turbine and engine design solution. See the table below for a summary of new and • Bergström–Boyce model to enhance elastomer enhanced elements. modeling capabilities When simulating a nonlinear process, large deformation can introduce too much distortion of the elements. Resolving • New damage model based on the Ogden–Roxburgh formulation

Element New Improved Capability Applications Four-noded tetrahedron X Provides a convenient way to automate Modeling complex geometries for forming or meshing of complex structures, avoiding hyperelastic applications need for pure hexahedral meshes

General axisymmetric element X Supports contact Compatible with 3-D non-axisymmetric loading and can use arbitrary axis of rotation

Various pipe model elements X Increased accuracy To provide refined behavior of structures in case of ovalization, warping or similar deformations of cross section for thin or moderately thick pipes and nonlinear material behavior support

Shell: linear, quadratic, axisymmetric X Improved shell thickness updating scheme Provides greater accuracy in the behavior of shell models and improved convergence as well as a faster solution for nonlinear problems

Beam X Supports cubic shape function Provides additional accuracy to coarse meshes and greater support of complex load patterns

Reinforcement elements X Allows modeling of discrete fibers with a Stresses in reinforcements can be analyzed variety of nonlinear material behavior separately from host elements

Summary of new and enhanced element features in ANSYS 12.0 structural analysis products

1818 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com 12.0: STRUCTURAL MECHANICS

contact search algorithms, contact trimming logic and smart over-constraint elimination for multipoint constraint (MPC) contact.

Solver Performance Solver performance has improved in many different areas. ANSYS 12.0 introduces a new modal solver, called SNODE, that increases the speed of computation for problems with a large number of modes — in the realm of several hundred — on large structures that typically have over a million degrees of freedom. This solver is well suited for automotive or aerospace applications and for large beams and shell assemblies. Beyond its ability to compute a larger number of modes in a reduced amount of time, Crack tip analysis SNODE also significantly reduces the amount of I/O of turbine blade Courtesy PADT required to compute the solution. (See the Supernode Eigensolver article.) Many enhancements have been made to the distributed solver to improve the scalability of the solution. (See • Anand’s viscoplasticity model, useful for metal the article on High Performance Computing.) More solver forming applications such as solder joints techniques are supported, including: • Improvements in the calculation of J-integrals to • Partial solve capability that computes only a portion account for mixed-mode stress intensity factors, of the solution which benefit improvements in fracture mechanics • Prestressed analysis • Initial strain and initial plastic stress import • Models that employ the use of unsymmetric capabilities that allow for state transfer from matrices, which are useful for scenarios that involve a 2-D model to a 3-D model high-friction coefficients, for example

Contact These new features can be combined for applications As assemblies have become a de facto standard in such as brake squeal, which might combine the partial simulation, the need for advanced contact features has solve and unsymmetric matrix capabilities. grown accordingly. ANSYS 12.0 developments include a number of additional contact modeling features as well as 80,000 significant improvements in solving contact problems. While Coulomb’s law for friction is widely used, there are 60,000 circumstances in which more elaborate modeling is Block Lanczos required, such as wear modeling or pipelines resting on sea Supernode 40,000 beds. Release 12.0 supports a friction coefficient definition ime (seconds) T that depends upon the contact state itself and accounts for CPU complex frictional behavior. Specifically, the user is able to 20,000 define the dependency of the friction on contact parameters, such as sliding distance or contact pressure. 0 A typical contact application involves seals that are sub- 100 1,000 4,000 8,000 Number of Modes ject to fluid pressure. Release 12.0 provides support of fluid Performance of new modal solver pressure penetration, to model scenarios in which pressure rises higher than the contact pressure around the seal. Linear Dynamics Pressures in such cases can be applied only on the free Some of these element, material, contact and solver faces of the structure and evolve with the contact state. improvements benefit the field of linear dynamics as well. Contact simulation is usually a time-consuming They are complemented by enhancements specific to this process. The latest release introduces contact modeling simulation area, especially for mode superposition analysis. improvements that significantly reduce computation time For harmonic or transient loadings, the mode superposition and results file size. These enhancements include new methods exhibit better performance, especially during the

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 191919

12.0: STRUCTURAL MECHANICS

so-called expansion pass that computes results at each solution speed by a factor of three or four — the greater the frequency or time step on the full model. For very large number of sectors, the better the performance. structures, the total computation effort can be reduced Rotating machinery applications profit from an extended by up to 75 percent. The mode combination for spectral set of capabilities for rotordynamics analysis. These include analysis benefits from similar advancements. Instability the extension of the gyroscopic effect to shell and predictions, such as the case of brake squeal, can be 2-D elements and inclusion of rotating damping that takes computed faster due to several enhancements to the hysteretic behavior into account. damped eigensolver. Random vibration and spectral analysis users gain new The introduction of ANSYS Variational Technology tools as well as a greater flexibility in modeling structures, provides faster mode computation for cyclic symmetric including support of spectrum analysis in the ANSYS structures, such as those found in many turbine Workbench platform. New tools include the United States applications. Using this technique can typically improve Nuclear Regulatory Commission–compliant computation of missing masses and support of rigid modes, along with the ability to use residual vectors to account for higher energy modes. The global number of spectra applied simultaneously to the structure has been increased up to 50 as has the number of modes used in a combination — now up to 10,000. When analyzing design variations, comparing data from different simulation cases, or correlating simulation and test data, comparison between modal content of the models is required. The modal assurance criterion (MAC) in release 12.0 provides a convenient tool to compare the results of two modal analyses. Typical use cases for the Instability analysis for brake squeal Modal analysis of a cyclic–symmetric geometry criteria include tuning of misaligned turbine blades or Courtesy PADT, Inc. validation of new component designs, each with respect to their vibration behavior.

New Element Reduces Meshing Time ZF Boge Elastmetall GmbH develops, manufactures and time-consuming meshing that was previously required supplies vibration control components and parts for the when using hexahedral elements. Boge’s work proved that automotive industry. These components include plastic by employing this new element, users can determine parts, energy-absorbing elements for vehicle safety, and the stresses and strains for a durability calculation in a rubber–metal components such as chassis suspension reasonable time. mounts, control arm bushes (also known as bushings) and engine mounts. The German company uses simulation to reduce Deformation of an automotive development time and costs. When developing models for suspension components with hyperelastic material properties, company mount engineers require an element type that can be freely meshed; can accommodate extreme deformation, stable contact and short computing time; and can provide reliable results. By using the new SOLID285 four-noded tetrahedron element available in ANSYS 12.0, ZF Boge Elastmetall engineers considerably reduced meshing time. Close correlation between the simulation and physical measurement allowed them to determine the spring rate of strongly deformed structures without the complex and

202020 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com 12.0: STRUCTURAL MECHANICS

ANSYS Workbench Integration • Ability to associate contact to the top or bottom The integration of the structural applications within of shell face the ANSYS Workbench platform provides additional Post-processing capabilities have drastically improved productivity to users, including: with release 12.0. The user can now plot any structural sim- • New meshing techniques to improve mesh quality ulation data stored in the results files. Mathematical • Support of additional elements, such as gasket operations involving elementary results can be introduced elements as well as quadratic shells and beams to create additional user-defined criteria. Complex mode that include offset definitions shapes, plotting on linear paths, stress linearization (which depends upon path plotting), and the ability to display • Boundary condition definitions that provide a unaveraged results at element nodes complement the list of spatial dependency for loads the features that increase productivity at ANSYS 12.0. ■ • Coupling conditions Pierre Thieffry and Siddharth Shah of ANSYS, Inc. contributed to • Remote points this article.

Multibody Dynamics At release 12.0, a number of improvements in the • Ability to export forces and moments at any time general area of multibody dynamics enable the rapid design within a transient simulation and analysis of complete mechanical systems undergoing large overall motion. ANSYS Rigid Dynamics software has a For durability studies, exported loads can be used in a new Runge–Kutta 5 integrator, the preferred solution for long static structural analysis as an efficient first-pass failure transient simulations. A new bushing joint, a “stops and analysis. Although it won’t provide the complete picture locks” option for most other joint types, and the ability obtained from comprehensive flexible dynamics simulation, to specify preload for springs give new flexibility when a static structural simulation is typically much less compu- simulating complex multiple-part assemblies and tationally expensive. Flexible dynamics simulations benefit component interactions. at release 12.0 from robust component modal synthesis, or For complex assemblies, conducting an initial simulation CMS. This method uses an internal substructuring with the ANSYS Rigid Dynamics product is the key to approach and requires that the CMS parts of an assembly achieving robust flexible dynamics results. Creating over- are constructed with linear materials. The procedure simpli- constrained assemblies is an inconvenient reality; release fies a problem by accounting only for a few degrees of 12.0 adds a redundancy analysis and repair tool to identify freedom, which results in solution times that are often a overconstrained assemblies, points out which joints or fraction of those found using the standard full computation degrees of freedom are redundant, and allows selective method. Time-to-solution reductions of several hundred unconstraining to create a properly constrained mechanism. percent are not uncommon. A number of improvements to data and process handling increase ease of use for multibody simulations: • Enhanced load data fitting (no longer requires curve fitting) • Ability to read in complex load input, such as simulated or measured multi-channel road surface or seismic data, and apply as load data to parts or joints • Ability to use remote solution manager (RSM) to offload the solving effort to a server or other capable CPU (benefits long- duration and multi-channel input transient simulations)

Multibody dynamics capabilities were used to simulate this leaf spring suspension.

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 212121 12.0: EXPLICIT DYNAMICS

Explicit Dynamics Goes Mainstream ANSYS 12.0 brings native explicit dynamics to ANSYS Workbench and provides the easiest explicit software for nonlinear dynamics.

ANSYS has expended significant they already know most of what is • Associatively link to a parametric effort in the area of explicit dynamics for needed to use ANSYS Explicit STR. CAD model or import a geometry release 12.0 — including the addition of The ANSYS Explicit STR tool is well • Create a smooth explicit mesh a new product that will make this tech- suited to solving: using the new explicit preference nology accessible to users independent • Drop tests (electronics and option and/or patch-independent of their simulation experience. In addi- consumer goods) mesh method within the ANSYS tion, enhancements to both the ANSYS • Low- to high-speed solid-to-solid meshing platform; automatically AUTODYN and ANSYS LS-DYNA impacts (a wide range of applica- create part-to-part contact by using products provide considerable benefits tions from sporting goods to the new body interactions tool to their users. aerospace) Newly introduced in ANSYS 12.0, • Fine-tune contact specifications if ANSYS Explicit STR software is the first • Highly nonlinear plastic desired by utilizing breakable or explicit dynamics product with a native buckling events (for ultimate eroding contact options ANSYS Workbench interface. It is based limit state design) • Load and/or support an assembly on the Lagrangian portion of the ANSYS • Complete material failure and/or parts as usual AUTODYN product. The technology will applications (defense and • Assign material properties from the appeal to those who want to model homeland security) comprehensive material library transient dynamic events such as drop tests, as well as quasi-static events • Breakable contact, such as • Solve interactively either in the involving rapidly changing contact adhesives or spot welds background or via remote solution conditions, sophisticated material (electronics and automotive) manager (RSM) failure/damage and/or severe displace- The real benefit of ANSYS Explicit • View progress of solution in real ments and rotations of structures. In STR software is the work flow afforded time using concurrent post- addition, it will appeal to users who can by operating in the ANSYS Workbench processing capability, new to benefit from the productivity provided by environment. While many different ANSYS Workbench at 12.0 other applications integrated within the simulation processes are possible, here • Explore alternative design ideas ANSYS Workbench environment. is an example of the typical steps a user via parametric changes to the CAD Those who have previous experience might take: model and easily perform re-solves, using ANSYS Workbench will find that just like other ANSYS Workbench based applications • Use the ANSYS Design Exploration capability to automate the parametric model space exploration In addition, users of the full version of ANSYS AUTODYN (structural- plus fluids- capable) have access to the ANSYS Explicit STR interface; consequently, they will be able to transfer implicit solutions from the ANSYS Workbench environment for doing implicit–explicit solutions, such as bird strike analysis of a pre-stressed fan blade. ANSYS LS-DYNA software users will be able to use the pre-processing portion of ANSYS Explicit STR and output a .K file for solving and post-processing outside of ANSYS Workbench. ■

Wim J. Slagter of ANSYS, Inc. is available to ANSYS Explicit STR is the first explicit dynamics product with a native ANSYS Workbench interface. answer your questions about explicit dynamics.

222222 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com 12.0: EIGENSOLVER

Introducing the Supernode Eigensolver A new eigensolver in ANSYS 12.0 determines large numbers of natural frequency modes more quickly and efficiently than conventional methods.

By Jeff Beisheim, Senior Development Engineer, ANSYS, Inc.

In a wide range of applications, parts are subject to cyclic mechanical loading, and engineers must use an tejay eigensolver to determine the structure’s natural frequencies — also known as eigen modes. With some modes, large vibration amplitudes can interfere with product performance and cause damage, such as fatigue cracking. In most © istockphoto.com/Ma cases, only the first few modes with the largest deformations are of particular interest, though determining even dozens of modes can be common. In the CAE industry, the block Lanczos eigensolver is typically used more than any other for these types of calcu- The ANSYS supernode eigensolver is well suited for applications such as seismic lations. This proven algorithm has been used in many finite analysis of power plant cooling towers, skyscrapers and other structures in which element software packages, including ANSYS Mechanical hundreds of modes must be extracted to determine the response of the structures to multiple short-duration transient shock/impact loadings. technology. It brings together the efficiency and accuracy of the Lanczos algorithm and the robustness of a sparse direct For such cases, the ANSYS release 12.0 includes a new equation solver. The software works in a sequential fashion supernode eigensolver. Instead of computing each mode by computing one mode (or a block of modes) at a time until individually and working with mode shapes in the global all desired modes have been computed. model space, the supernode algorithm uses a mathematical Although the method is considered efficient in solving approach based on substructuring to simultaneously deter- for each of these eigen modes, the amount of time and mine all modes within a given frequency range and to computer resources (both memory and I/O) required adds manage data in a reduced model space. up when many dozens of eigen modes must be found. By utilizing fewer resources than block Lanczos, this Elapsed solution times of several hours — or days — are supernode eigensolver becomes an ideal choice when typical in applications that involve thousands of modes. solving on a desktop computer, which can have limited Generally, determining large numbers of modes is required memory and relatively slow I/O performance. When com- in capturing system response for studies such as transient bined with current eigensolver technology already available or harmonic analyses using the mode superposition in mechanical software from ANSYS, virtually all modal method. analyses can be efficiently solved.

25,000

Linux 64-bit server: 3.4 GHz Comparing Eigensolvers 20,000 single-core Xeon®, Red Hat® 4, A sample comparison shows that the super- 64 GB RAM, 200 GB disk

node eigensolver offers no significant performance 15,000 ime (seconds)

advantage over block Lanczos for a low number of T Block Lanczos modes. In fact, supernode is slower when 50 10,000 or fewer modes are requested. However, when Supernode more than 200 modes are requested, the

Solver Elapsed 5,000 supernode eigensolver is significantly faster

than block Lanczos — with efficiency increasing 0 considerably as the number rises. 10 50 100 200 500 1,000 2,000 Number of Requested Modes Performance of block Lanczos and supernode eigensolvers at 1 million DOF www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 232323 12.0: EIGENSOLVER

Using Supernode Eigensolver factor (RangeFact) field on the SNOPTION command. The supernode eigensolver can be selected in the Higher values of RangeFact lead to more accurate solu- ANSYS Mechanical traditional interface using the SNODE tions at the cost of extra computations that somewhat slow label with the MODOPT command or via the Analysis Options down eigensolver performance. dialog box. ANSYS Workbench users can choose this When computing the final mode shapes, the super- eigensolver by adding a command snippet that includes the node eigensolver often does the bulk of I/O transfer to and MODOPT,SNODE command. from disk, and the amount of I/O transfer is often The MODOPT command allows users to specify the num- significantly less than a similar run using block Lanczos. To ber of natural frequencies and what range those frequencies maximize supernode solver efficiency, I/O can be further lie within. With other eigensolvers, the number of requested minimized using the block size (BlockSize) field on the modes primarily affects solver performance, while the fre- SNOPTION command. Larger values of block size will quency range is, essentially, optional. Asking for more reduce the amount of I/O transfer by holding more data in modes increases solution time, while the frequency range memory during the eigenvalue/eigenvector output phase, generally decides which computed modes are computed. which generally speeds up the overall solution time. The supernode eigensolver behaves completely oppo- However, this is recommended only if there is enough site: It computes all modes within the specified frequency physical memory to do so. range regardless of how many modes are requested. Therefore, for maximum efficiency, users should input Application Guidelines a range that covers only the spectrum of frequencies The following general guidelines can be used in deter- between the first and last mode of interest. The number mining when to use the supernode eigensolver, which is of modes requested on the MODOPT command then typically most efficient when the following three conditions decides how many of the computed frequencies are are met: provided by the software. • The model would be a good candidate for using the Today, with the prevalence of multi-core processors, the sparse solver in a similar static or full transient first release of this new eigensolver will support shared- analysis (that is, dominated with beam/shell elements memory parallelism. For users who want full control of the or having thin structure). solver, a new SNOPTION command allows control over several important parameters that affect accuracy and • The number of requested modes is greater than 200. efficiency. • The beginning frequency input on the MODOPT command is zero (or near zero). Controlling Parameters The supernode eigensolver does not compute exact eigenvalues. Typically, this is not an issue, since the lowest For models that have dominantly solid elements or modes in the system (often used to compute the dominant bulky geometry, the supernode eigensolver can be more resonant frequencies) are computed very accurately — gen- efficient than other eigensolvers, but it may require higher erally within less than 1 percent compared to using block numbers of modes to consider it the best choice. Also, Lanczos. Accuracy drifts somewhat with higher modes, other factors such as computing hardware can affect the however, in which computed values may be off by as much decision. For example, on machines with slow I/O perform- as a few percent compared with Lanczos. In these cases, ance, the supernode eigensolver may be the better choice, the accuracy of the solver may be tightened using the range even when solving for less than 200 modes. ■

Examining Real-World Performance A heavy-equipment cab model with over 7 million equations was used to demonstrate the power of the supernode eigensolver. This model was solved using a single core on a machine with the Windows® 64-bit operating system with 32 gigabytes of RAM. Time spent computing 300 modes with block Lanczos was about 31.8 hours. The solution time dropped to 15.7 hours (a two-times speedup) using the supernode eigensolver. The model illustrates real-world performance for a bulkier model with only 300 modes requested. For modal analyses in which hundreds or thousands of modes are requested, users often see a speedup of 10 times or more with the supernode eigensolver compared with block Lanczos. In Total displacement for the tenth-lowest natural one recent project, a major industrial equipment manufacturer frequency is plotted for a heavy-equipment cab model represented by more than 7 million equations. reduced analysis run time from 1.5 hours to just 10 minutes by Model courtesy PTC. switching from block Lanczos to supernode eigensolver.

242424 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

12.0: HIGH-PERFORMANCE COMPUTING

The Need for Speed From desktop to supercomputer, high-performance computing with ANSYS 12.0 continues to race ahead.

Tuning software from ANSYS on the latest high- performance computing technologies for optimal performance has been — and will continue to be — a major focus area within the software development organization at ANSYS. This effort has yielded significant performance gains and new functionality in ANSYS 12.0, with important implications for more productive use of simulation by customers. High-performance computing, or HPC, refers to the use of high-speed processors (CPUs) and related technologies to solve computationally intensive problems. In recent years, HPC has become much more widely available and affordable, primarily due to the use of multiple low-cost processors that work in parallel on the computational task. Today, clusters of affordable compute servers make large-scale parallel processing a very viable strategy for ANSYS customers. In fact, the new multi-core processors have turned even desktop workstations into high-performance platforms for single-job execution. This wider availability of HPC systems is enabling important trends in engineering simulation. Simu- lation models are getting larger — using more computer memory and requiring more computational time — as engineers include greater geometric detail and more-realistic treatment of physical phenomena (Figure 1). These higher-fidelity models are critical for Figure 1. Simulations as large as 1 billion cells are now supported at release 12.0. This 1 billion-scale racing yacht simulation was conducted on a cluster of 208 HP simulation to reduce the need for expensive physical testing. ProLiant™ server blades. (For more information, visit www.ansys.com/one-billion.) HPC systems make higher-fidelity simulations practical by Image courtesy Ignazio Maria Viola. yielding results within the engineering project’s required time frame. A second important trend is toward more simulations — enabling engineers to consider multiple design ideas, conduct parametric studies and even perform automated HPC on Workstations? design optimization. HPC systems provide the throughput While purists might argue whether workstations can required for completing multiple simulations simultaneously, be considered high-performance computing platforms, thus allowing design decisions to be made early in the project. the performance possibilities for ANSYS 12.0 running on Software from ANSYS takes advantage of workstations are noteworthy. With the latest quad-core multi-processor and/or multi-core systems by employing processor technology, an eight-core workstation running domain decomposition, which divides the simulation model Windows® can deliver a speedup of five to six times for into multiple pieces or sub-domains. Each sub-domain is then users of mechanical products from ANSYS (Figure 2) computed on a separate processor (or core), and the multiple and over seven times for users of its fluid dynamics processors work in parallel to speed up the computation. In the products (Figure 4). This means that parallel processing ideal case, speedup is linear, meaning that the simulation turn- now provides tremendous ROI for both large engineering around time can be reduced in proportion to the number of groups and individual workstation users, enabling faster processors used. Parallel processing also allows larger turnaround, higher-fidelity models and parametric problems to be tackled, since the processing power and modeling. With release 12.0 and 2009 computing memory requirements can be distributed across the cluster of platforms, parallel processing improves productivity processors. Whether performed on a multi-core desktop work- for all simulation types, from workstation to cluster, for station, desk-side cluster or scaled-out HPC system, parallel mechanical or fluids simulations. www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 252525 12.0: HIGH-PERFORMANCE COMPUTING

8.83 processing provides excellent return on investment by improving 9.00 8.00 the productivity of the engineers who perform simulation.

8.00 ANSYS 12.0 provides many important advances in areas related to parallel processing and HPC, delivering scalability 7.00 5.74 5.46 from desktop systems to supercomputers. For users of the 5.53 5.83 6.00 ANSYS Mechanical product line, release 12.0 introduces 5.00 4.55 expanded functionality in the Distributed ANSYS (DANSYS) 4.00 solvers, including support for all multi-field simulations, pre- stress effects and analyses involving cyclic symmetry. In 3.00 addition, DANSYS now supports both symmetric and non- Core Solver Speedup 2.00 symmetric matrices as well as all electromagnetic analyses. 1.00 Mechanical simulations benefit from significantly improved 0.00 8 scaling on the latest multi-core processors. Simulations in the 4 size range of 2 million to 3 million degrees of freedom (DOF) Ideal 2 BMD-2 BMD-3 1 now show good scaling on eight cores (Figure 2). Based on BMD-4 BMD-5 benchmark problem performance, customers can expect to BMD-6 BMD-7 Number of Cores get answers back five to six times faster on eight cores. Even Figure 2. Speedup of Distributed ANSYS Mechanical 12.0 software using the more impressive is the scale-out behavior shown in Figure 3, 11.0 SP1 benchmark problems. Simulations running eight-way parallel show typical with a 10 million DOF simulation showing solver speedup of speedup of between five and six times. Data was collected on a Cray CX-1 Personal 68 times on 128 cores. Supercomputer using two quad-core Intel Xeon Processor E5472 running Microsoft® Windows HPC Server 2008. With turnaround times measured in tens of seconds, parametric studies and automated design optimization are now well within the grasp of ANSYS customers who perform 80 mechanical simulations. These benchmarks are noteworthy, 70 60 in part, as they show execution with all cores on the cluster 50 fully utilized, indicating that the latest quad-core processors 40 have sufficient memory bandwidth to support parallel 30 processing for memory-hungry mechanical simulations. 20 Software tuning has contributed to improved scaling as well, Core Solver Speedup 10 including improved domain decomposition, load balancing 0 1 2 4 8 32 64 128 and distributed matrix generation. To help customers maxi- Number of Cores mize their ANSYS solver performance, the online help system Figure 3. Scaling of a 10M DOF simulation using the ANSYS Mechanical 12.0 iterative now includes a performance guide that provides a compre- PCG solver on a cluster of Intel Xeon 5500 Processor series. All cores on these quad- hensive summary of factors that impact the performance of core processors are fully utilized for the benchmark. mechanical simulations on current hardware systems. Explicit simulations using ANSYS AUTODYN technology take great advantage of HPC systems at release 12.0. Full 8.00 64-bit support is now available, allowing much larger simu- 7.58 8.00 7.16 7.18 7.55 lations to be considered from pre-processing to solution and 7.00 6.23 6.79 post-processing. 6.47 For users of fluid dynamics software from ANSYS, release 6.00 12.0 builds on the strong foundation of excellent scaling in 5.00 both the ANSYS FLUENT and ANSYS CFX solvers. These 4.00 fluids simulation codes run massively parallel, with sustained

3.00 scaling at hundreds or even thousands of cores. The release

Core Solver Speedup incorporates tuning for the latest multi-core processors, 2.00 including enhanced cache re-utilization, optimal mapping and 1.00 binding of processes to cores (for better memory locality and 0.00 system utilization), and leveraging the latest compiler opti- 8 mizations. The resulting ANSYS FLUENT and ANSYS CFX Ideal 4 FL5L2 ® ® FL5L3 2 performance on the newly released Intel Xeon 5500 1 Sedan_4M Processor series is shown in Figure 4, with outstanding ANSYS FLUENT Truck_14MIndy Car LeMans Turbine Number of Cores speedup of over seven times for many benchmark cases. In ANSYS CFX addition, the new release delivers significant performance improvements at large core counts, the result of general Figure 4. Scalability of ANSYS FLUENT and ANSYS CFX benchmark problems on the Intel Xeon 5500 Processor series quad-core platform. Simulations running eight-way solver enhancements and optimized communications over parallel show typical speedup of over seven times. the latest high-speed interconnects. Figure 5 demonstrates

262626 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com 12.0: HIGH-PERFORMANCE COMPUTING

scaling achieved by ANSYS CFX software on a cluster of 120.00 quad-core AMD processors. Nearly ideal linear scaling to 1,024 cores — and very good efficiency up to 2,048 cores — 100.00 has been demonstrated with ANSYS FLUENT (Figure 6). Both fluids codes provide improvements to mesh partitioning that 80.00 Transonic Airfoil Benchmark enhance scalability. ANSYS FLUENT software now provides Model — 10M Nodes dynamic load balancing based on mesh- and solution- 60.00 derived criteria. This enables optimal scalability for Speedup simulations involving multiphysics, such as particle-laden 40.00 Ideal flows. The ANSYS CFX code delivers improved partitioning Infiniband for moving and/or rotating meshes, yielding important 20.00 Gigabit reductions in memory use and improved performance for 0.00 turbomachinery and related applications. Finally, ANSYS 0 16 32 48 64 80 96 FLUENT users will benefit from several usability improve- Number of Cores ments, including built-in tools for checking system network Figure 5. Scalability of ANSYS CFX 12.0 on a 10M node transonic airfoil benchmark bandwidth, latency and resource utilization — all helping to example. Data was collected on a cluster of AMD Opteron™ 2218 processors, showing identify potential scaling bottlenecks on the cluster. the benefit of a high-speed interconnect. Beyond solver speedup, the ANSYS 12.0 focus on HPC addresses issues related to file input and output (I/O). Both ANSYS FLUENT and ANSYS CFX software have updated I/O 2,048 algorithms to speed up writing of results files on clusters, enhancing the practicality of periodic solution snapshots 1,536 when checkpointing or running time-dependent simulations. Truck Benchmark Model — 1,581 ANSYS FLUENT includes improvements in the standard file I/O 111M Cells as well as new support for fully parallel I/O based on parallel file 1,024 991 systems. Order of magnitude improvements in I/O throughput Solver Speedup have been demonstrated on large test cases (Figure 7), virtually 512 Speedup 521 eliminating I/O as a potential bottleneck for large-scale simula- Ideal tions. ANSYS CFX improves I/O performance via data 265 128 0 compression during the process of gathering from the cluster 0 512 1,024 1,536 2,048 nodes, therefore reducing file write times. Proper I/O configura- Number of Cores tion is also an important aspect of cluster performance for the Figure 6. Scaling of ANSYS FLUENT 12.0 software is nearly ideal up to 1,024 processors ANSYS Mechanical product line. and 78 percent of ideal at 2,048 processors. Data courtesy SGI, based on the SGI Altix® Recognizing that cluster deployment and management ICE 8200EX using quad-core Intel Xeon Processor E5472 with Infiniband®. are key concerns, ANSYS 12.0 includes a focus on compatibility with the overall HPC ecosystem. ANSYS products are registered and tested as part of the Intel Cluster Ready program, confirming that these products conform to standards of 400 PanFS – FLUENT 12 compatibility that contribute to successful deployment PanFS — FLUENT 12.0 NFS –— FLUENT6.3 FLUENT6.3 345 (www.ansys.com/intelclusterready). In addition to supporting 300 334 enterprise Linux® distributions from Red Hat® and Novell, 273 200

ANSYS 12.0 products are supported on clusters based on Write (MBytes/sec) 31X Microsoft Windows HPC Server 2008. ANSYS has also ta 39X 20X worked with hardware OEMs, including HP®, SGI®, IBM®, Dell®, 100 Cray® and others, to define reference configurations that are te for Da optimally designed to run simulation software from ANSYS I/O Ra 0 64 128 256 (www.ansys.com/reference-configs). Number of Cores As computing technology continues to evolve, ANSYS is Figure 7. Parallel I/O in ANSYS FLUENT 12.0 using the Panasas© file system, compared working with HPC leaders to ensure support for the break- to serial I/O in the previous release using NFS. Parallel treatment of I/O provides through capability that will make simulation more productive. important speedup for time-varying simulations on large clusters. Looking forward, important emerging technologies include many-core processors, general purpose graphical processing units (GP-GPUs) and fault tolerance at large scale. ■

Contributions to this article were made by Barbara Hutchings, Ray Browell and Prasad Alavilli of ANSYS, Inc.

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 27 12.0: FUTURE DIRECTIONS

Foundations for the Future The many advanced features of ANSYS 12.0 were designed to solve today’s challenging engineering problems and to deliver a platform for tomorrow’s simulation technology.

As this special spotlight in ANSYS Advantage attests, Release 12.0 is a notable milestone in the company’s release 12.0 delivers a compelling advancement in what the nearly 40-year history of innovating engineering simulation, CAE industry has, until now, only envisioned — a full range and it sets the stage for a new era of Smart Engineering of best-in-class simulation capabilities assembled into a Simulation — an era in which ANSYS customers will gain flexible multiphysics simulation environment specifically more from their investment in simulation by increasing the designed to increase engineering insight, productivity and efficiency of their processes, increasing the accuracy of innovation. Whether the need is structural analysis, fluid their virtual prototypes, and capturing and reusing their flow, thermal, electromagnetics, geometry preparation or simulation processes and data. However, the advance- meshing, ANSYS customers can rely on release 12.0 for the ments of ANSYS 12.0 notwithstanding, the journey is far depth and breadth of simulation capabilities to overcome from complete. To address the simulation challenges on the their engineering challenges. horizon, ANSYS will continue to reinvest in research and Staying true to our commitment to develop the most development and to explore new technologies. In particular, advanced simulation technologies, release 12.0 has further there are a few areas that we consider vital in the pursuit of expanded the depth of individual physics and more Simulation Driven Product Development — areas in which intimately coupled them to form an engineering simulation ANSYS has laid strong foundations and remains committed capability second to none. A multitude of new material to build upon as we look beyond release 12.0. models, physics and algorithms enable simulating real-world operating conditions and coupled physical Physics First phenomena, while new solver technology and parallel ANSYS customers rely heavily on simulation before processing improvements have dramatically reduced run making commitments to product designs or manufacturing times and made complete system simulations more processes. High-fidelity engineering simulation is absolutely computationally affordable. paramount when upstream engineering decisions can Shouldering the array of technology in release 12.0 determine the overall success of a product and, in some is our next-generation simulation platform, ANSYS cases, the company’s financial success. At ANSYS, we Workbench 2.0. Seamlessly spanning all stages of believe our customers should never have to compromise by engineering simulation, ANSYS Workbench 2.0 has been making broad-based engineering assumptions due to engineered to manage the complexities of today’s simu- limitations in their analysis software. That is why we have lations and to accelerate innovation. taken a comprehensive multiphysics approach to simulation, and it starts with a foundation of individual physics. Looking beyond release 12.0, ANSYS will continue to invest and demonstrate leadership in all the key physics. And as

28 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com 12.0: FUTURE DIRECTIONS

we develop tomorrow’s advanced capabilities, we will con- applications — data management, parameterization, scripting tinue to allow them to be combined in ways that free and graphics, among others. Release 12.0 relies heavily on engineers from making the assumptions associated with the framework’s data management and parameterization single-physics simulations. Within the ANSYS Workbench services to integrate existing applications into the ANSYS simulation paradigm, we will enable engineers to routinely Workbench environment, where they have become highly consider the effects of fully coupled physical phenomena. interoperable. Over subsequent releases, these applications will leverage the framework’s graphical toolkit to establish a High-Performance Computing consistent user interface and further blend the various As one might expect, high-performance computing applications integrated into the platform. At the onset of (HPC) is a strategic enabling technology for ANSYS. The developing ANSYS Workbench 2.0, we identified scripting appearance of quad-core machines on the desktop and the and journaling as fundamental requirements of the new increased availability of compute clusters have ushered in architecture. As such, a top-level scripting engine has been a new era of parallel and distributed computing for our thoughtfully designed and lays the groundwork for future customers. ANSYS has kept pace with the exponential ANSYS Workbench customization and batch processing. increase in computational horsepower with prolific develop- Looking beyond release 12.0, all these services will be ment in the areas of parallel and distributed computing and further refined and will fuel rapid add-in development and numerical methods. The result is improved scalability and a further expansion of capabilities. Over time, ANSYS dramatically reduced run times for large-scale fluid flow, customers and partners will leverage the framework’s structural and electromagnetic simulations. open architecture, enlisting its services to create tailored Solving large-scale problems with meshes exceeding applications, and will elevate ANSYS Workbench as 1 billion cells has been the latest stretch goal for fluid an application development platform for the engineering flow simulation. Recently, HPC and software from ANSYS simulation community. were combined to investigate the aerodynamics of a racing yacht using 1 billion computational cells. Breaking Simulation Process and Data Management this barrier demonstrates our conviction for high- ANSYS Workbench 2.0 is an environment in which a performance scientific computing. As computational single analyst creates and executes one or more steps of an resources increase and engineering simulations become engineering simulation workflow. ANSYS Engineering larger and more complex, we will continue to ensure that Knowledge Manager (EKM) extends ANSYS Workbench by our solvers scale appropriately. Moreover, our forward providing the tools to manage the work of a group of deployment of HPC technology is not limited to solvers. analysts and myriad simulation workflows. This includes The complexity of today’s models and massive amounts system-level services to manage and foster collaboration of results data require more-scalable solutions for on thousands of models, terabytes of results, hundreds of preparing models and interpreting results as well. defined processes and huge investments in simulation. Looking forward, ANSYS believes that managing data ANSYS Workbench Framework and processes will become integral with engineering simu- The ANSYS Workbench 2.0 platform is a powerful multi- lation. Ten years ago, simulation comprised three discrete domain simulation environment that harnesses the core and sequential phases: pre-processing, solving and post- physics from ANSYS; enables their interoperability; and processing. With the evolution of ANSYS Workbench, we provides common tools for interfacing with CAD, repairing now look at engineering simulation as a continuous geometry, creating meshes and post-processing results. workflow intertwining these steps. In the same way, Instrumental to the successful integration of this unparal- process and data management will become intertwined leled breadth of technology is a “well-architected,” open and extendable software framework. The ANSYS Workbench framework is designed to provide common services for engineering simulation

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 29 12.0: FUTURE DIRECTIONS

As mechanical and electrical engineering worlds converge, the combination of ANSYS and Ansoft technologies will allow engineers to analyze the behavior of combined systems. © iStockphoto.com/weicheltfilm, © iStockphoto.com/meduedik with simulation, expanding its role and aligning it with ANSYS has extended its range of simulation technology business processes such as product lifecycle and supply by incorporating Ansoft’s world-class product portfolio. chain management. Standardizing on ANSYS Workbench for Simulation Driven Product Development means establishing a common Electromechanical System Simulation platform on which to further develop both mechanical and The ANSYS acquisition of Ansoft anticipates a trend in electronic components and analyze the behavior of the the realm of engineering and design: The mechanical, elec- combined systems. Driving innovation with mechatronics trical and software engineering worlds will rapidly converge. will require a comprehensive electromechanical simulation Several years ago, the synchronization of these worlds environment developed by a leader in both mechanical and was coined “mechatronics,” and, today, the combined electronic simulation software. disciplines are responsible for engineering the electromechanical systems found in everything from washing The Future Begins Now machines to airplanes. A simple examination of the auto- With its advancements in individual physics, high- motive industry reveals that the more recent and exciting performance computing, multidomain simulation, meshing, advancements have relied on mechatronics. So, at a time and key enabling technologies such as simulation workflow when greeting cards and tennis shoes contain micro- and data management, release 12.0 clearly delivers on the processors and sensors, mechatronics is not just for ANSYS vision for Simulation Driven Product Development. high-end cars and appliances; rather it is the key to But even though we have come a long way with the advent unleashing innovation in every industry. of ANSYS 12.0, there is still an exciting journey ahead. For many years, electrical and mechanical engineering Standing on the strong foundation of all that ANSYS has teams have increasingly relied on simulation to accelerate learned and developed in almost 40 years of leadership in innovation, but each camp has adopted simulation tools engineering simulation, we see many new opportunities on that were not fully capable of addressing the needs of the the horizon that will extend the reach of how customers use other — until now. As the separation between the electronic our technology. The ANSYS vision and strategy continue to and mechanical worlds becomes increasingly blurred, set our bearings, and we continue to invest in pioneering new frontiers of the industry. And most important is that we remain committed to enabling customers to use simulation to develop innovative products that perform better, cost less and are brought to market faster. ■

This article was written through contributions from Todd McDevitt of ANSYS, Inc.

303030 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

AUTOMOTIVE

Predicting 3-D Fatigue Cracks without a Crystal Ball ANSYS tools quickly predict 3-D thermomechanical fatigue cracking in turbocharger components.

By Shailendra Bist, Senior Engineer, and Ragupathy Kannusamy, Principal Engineer, Structures and Fatigue Group, Honeywell Turbo Technologies, California, U.S.A.

Turbochargers increase the power environments and boost the fuel efficiency of internal with multi-axial combustion engines, but engineering loading. As a result, teams find they pose unique design many part designs challenges. For example, because the are based on modifying turbine is driven by the engine’s own previous geometries, trial- hot exhaust gases, components must and-error testing cycles and, in withstand widely varying thermal many cases, “crystal ball” best-guess Honeywell Turbo Technologies produces nearly 9 million turbochargers annually for stresses as temperatures cycle predictions based partly on conjecture the automotive industry. Because turbochargers between 120 and 1,050 degrees Celsius and simplified assumptions. undergo wide thermal swings, they are subject for engine speed variations relating to Honeywell Turbo Technologies to thermomechanical fatigue cracking. idle, acceleration and braking. overcomes these limitations by using In particular, components such as ANSYS Mechanical software together the cast-iron housing that directs hot with the ANSYS Parametric Design fracture mechanics parameter that gases into the turbine are subject to Language (APDL) scripting tool to calculates energy release rate and thermomechanical fatigue cracking — calculate the probability of a crack intensity of deformation at the crack a problem that often is not discovered initiating as well as its most likely front for linear and nonlinear material until parts fail in qualification tests. To growth rate, length and 3-D path. behaviors. The J-integral approach replicate four to five years of severe Predicting crack fractures in this man- generally works best with hexahedral thermal shock loading — far greater ner at the early stages of component meshes for the highest possible than parts would experience in normal development enables engineers to accuracy. But representing the entire operation — engineers perform rounds optimize designs upfront and help structure with a hex mesh is a of tests that each can be very expen- avoid qualification test failures. tremendous drain on computational sive and take weeks to complete. Conversely, the analysis gives engi- resources. So in this case, Honeywell Several of these rounds generally must neers information on the presence of Turbo engineers used two separate be performed before arriving at a work- small benign cracks that do not lead to meshing techniques: hexahedral able design that passes scrutiny. Many loss of component functionality (for elements for representing the instanta- stress intensity factor formulas are example, gas leakage or turbine wheel neous crack front (a cylindrical volume available in handbooks for predicting rub) and can, therefore, be ignored. around the crack front called the crack fatigue crack growth with simplified 2-D For this application, J-integral tube) and tetrahedral elements for the geometries; typically, though, these analysis capabilities in ANSYS 12.0 remaining part volume. formulas are not applicable for provide a robust solution to predict Connectivity between the two complex part geometries under elastic– crack behavior at high temperatures. different mesh patterns is assured with plastic conditions in high-temperature The J-integral is a path-independent ANSYS transition elements. The size of

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 31

AUTOMOTIVE

X5 X4 X3 Crack front with virtual crack X2 extension X directions 1 X6

X7 X8 X9

Hexahedral elements represent the expected path of 3-D crack The 3-D crack growth direction determining the propagation propagation (called the crack tube), and less-complex tetrahedral path is based on a virtual extension direction angle in which elements are used for the remaining volume of the part. maximum energy is released. the 3-D crack tube depends on the meshed, solved and post-processed. capabilities of the APDL scripting tool. volume of the crack path’s plastic zone The cycle continues until a target Along with techniques such as and is based on the number of rings criterion is reached. All processes submodeling and load blocks for more of elements and the number of are integrated and controlled using efficient solution processing, such contours to be used in calculating in-house APDL scripts. By leveraging automation radically increases the J-integral values using the ANSYS improved fracture mechanics capabil- speed of performing these iterative CINT command. The number of ities in ANSYS 12.0 for calculating calculations. element rings and contours should J-integrals, the method provides a Honeywell Turbo analyzed a test be high enough to maintain path new approach to model and simulate case using this method to predict independence and accuracy of energy arbitrary 3-D crack growth and to growth behavior of paths in a cruciform release rate. compute mixed mode stress intensity specimen under uniaxial and biaxial In this way, ANSYS software factors along the crack front within the loading. The uniaxial load case shows calculates J-integral values at each simulation software. prominent crack turning while the increment of crack propagation along This method requires calculations biaxial case shows near planar growth. several user-defined virtual crack to be performed iteratively for thou- The results obtained validate the extension directions. The crack feature sands of crack-growth cycles — a approach. The team completed further is updated in a third-party CAD code at prohibitively labor-intensive and time- runs to validate crack growth rates each increment, then imported into consuming task if performed manually that show promising results. ANSYS Mechanical software where it is but one well-suited to the automation Using this automated ANSYS fatigue crack prediction process has the potential to increase engineering productivity significantly, with crack growth analysis time reduced by more than 90 percent compared to manual Crack path methods. This speedup has significant from test result value, since Honeywell Turbo engineers must analyze as many as 400 designs annually, and demands will likely increase in the coming years as turbochargers are implemented on a growing number of vehicle models around the world. In this way, technology from ANSYS is playing a critical role in enabling the turbocharger company to strengthen its leadership Crack path position in this competitive industry from test sector. ■

Crack path directions in cruciform specimens under uniaxial loading (top) and biaxial loading (bottom)

32 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com HEALTHCARE © iStockphoto.com/ kwanisik Electromagnetics in Medicine Electromagnetic and thermal simulations find use in medical applications.

By Martin Vogel, Senior Member of the Technical Staff, Ansoft LLC

Electromagnetic fields are used more and more in Model of the open MRI system, which combines an advanced medical applications such as magnetic reso- MRI model generated by Philips Healthcare with the ANSYS human body model nance imaging (MRI), implants and hyperthermia treatment. As the state of the art advances, devices are becoming more complex and simulation more indispensable in the Frequency-dependent electromagnetic material parameters product design phase. With simulation, a designer can are also included in the model. study device functionality and address safety concerns The RF coil design requires optimization for appropriate without exposing a patient to harm or otherwise. image quality: The coils need to resonate at 42.6 MHz for a In the design of an open MRI system, for example, the 1 tesla system and produce a rotating magnetic field that is details of the radio-frequency (RF) coils, a human body strong and smooth in the region of interest but minimizes model, and the large volume of the entire examination room undesired field components. If the field varies strongly, must all be included in an electromagnetic simulation model some parts of the image will appear to be overexposed, to determine the resulting field accurately. The finite element while other areas will remain too dark, both of which are method found in HFSS (High-Frequency Structure Simulator) detrimental for contrast. Once the specifications related to software, an electromagnetic field simulation tool new to the image quality are satisfied, the designer needs to make sure ANSYS portfolio, is well suited for this purpose as it uses that specific absorption rate (SAR) safety regulations are small mesh elements where refinement is needed and larger met. SAR is a measure of how much RF power is absorbed mesh elements elsewhere. The human body model available by, and thus creates heat in, the body. When limits through ANSYS comprises 300 objects that, detailed down are exceeded in any part of the body, the patient can to the millimeter, represent organs, bones and muscles. experience discomfort and tissue damage.

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 333333 HEALTHCARE

Simulation results from the open MRI case indicate hot spots under the armpits, a result that agrees with practical experience. Analysis also indicates resonant hot spots on the legs, even though they are not directly under the coils in the model. Given the frequency and material parameters of the body, the expected wavelength in the body is a little less than 1 meter, and resonances such as these are indeed Sample of specific absorption rate that results on the body when using the open MRI system, as simulated using the possible. The SAR is not quite symmetric; this is expected, HFSS electromagnetic field simulation tool as the excitations are not symmetric either. Entire scan protocols can be simulated in the software by moving the body automatically through the scanner. Another medical application in which human comfort is important is the design of wireless implants. Implants that require directly wired power supplies can be uncomfortable for the patient. But wireless power supplies that use low-frequency coupling require a bulky transmitter, reducing patient freedom. Wireless solutions that use higher frequencies can potentially provide both comfort and freedom. One design challenge is to transmit maximum power to the implant while also satisfying radiation and SAR regulations. Simulations of wireless implants provide details that The electric field (magnitude) that results when using a receiver otherwise are not easily obtained for several transmitter and implanted in epidural space in conjunction with a wireless receiver locations. One important finding is that, in order to transmitter placed behind the back; the image shows a horizontal cross section of the torso and arms of a person, get accurate results, interior body components such as standing, using a wireless implant. organs, bones and fat tissue must be included in the simulation model. If not, the results can easily be off by more than a factor two. One final medical simulation example models an RF phased-array applicator for hyperthermia cancer treat- ments. In hyperthermia, a tumor is heated with RF power and held at an elevated temperature for some time, such as 15 minutes to 60 minutes. This weakens the tumor, which helps to make other therapies more effective. The challenge is to concentrate the hot spot in the tumor while minimally affecting healthy tissue. The applicator consists of several dipole antennas printed on the surface of a cylindrical plastic shell that mounts around the patient’s leg, the location of the tumor Model of a hyperthermia applicator and leg with tumor; in the image, some applicator and water cooling system components for this case. The chosen frequency for the device, have been removed for clarity. The green object is the tumor. 138 MHz, is a compromise between hot spot size and Applicator design and tumor geometry provided by Duke University. penetration depth. A higher frequency can provide a smaller XY Plot 2 hot spot, but it would be harder to penetrate deep into 6 17W 15W the tissue. Water cooling prevents skin heating during the 5 20W procedure and is accounted for in the simulation model. 15W 10W 4 A realistic tumor object, created using MRI data for this Simulation 20W 3 patient, is inserted into the leg of the human body model. 30W By using the electromagnetic simulation capabilities in 2 Measurement

emperatire Increase [deg C] HFSS software, the applicator and its settings are optimized T 1 to focus the hot spot in the tumor. Next, the power-loss

0 information for every mesh element in the model is 0 5 10 15 20 25 Time [min] transferred automatically to the thermal simulation Comparison of simulated and measured temperatures tool, ePhysics. The ePhysics product then computes in the tumor for a hyperthermia treatment case Measured results provided by Duke University. temperature distribution as a function of time, taking

34 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

HEALTHCARE

into account thermal material properties as well as water With these simulations, modeling software progresses cooling, blood perfusion, air convection and thermal radiation. beyond device design into treatment planning. Finding the Blood perfusion refers to blood flow through capillary proper operating conditions through simulation relieves the vessels in muscles and organs. This flow removes excess patient from invasive experimental procedures. To efficiently heat and must be included in hyperthermia simulations. To optimize conditions for a variety of patients in a hospital include all the details of the capillary blood vessels would be environment, engineers must improve methods to translate too complicated; therefore, a simpler model is used. It is MRI scan data into personalized human body models that assumed that a certain amount of blood enters a volume of are ready for simulation. tissue at a specified rate; it is also assumed that blood Electromagnetic and thermal simulations are well assumes the tissue’s temperature and leaves the volume, understood and used regularly for the design of medical taking a corresponding amount of heat with it. Perfusion for equipment and procedures. The next breakthrough is several tissue types can be found in literature [1] and is expected when personalized human body models can be quantified in the simulation model as a temperature- generated efficiently and doctors use simulation for dependent negative heat source. Overall, the simulation treatment planning. ■ results proved to be very sensitive to blood perfusion. The input power to the applicator is varied over time for The author wishes to acknowledge Philips Healthcare in the Netherlands for its work on MRI and Duke University in the United States for its work both simulation and experiment. The outer layer of the on hyperthermia. tumor is assumed to have a higher perfusion rate than the core, as is consistent with literature. Deviations between References simulation results and experimental data in the early stages [1] Erdmann, B; Lang, J; and Seebass, M. “Optimization of Temperature Distributions for Regional Hyperthermia Based on a Nonlinear Heat are likely due to the fact that initial thermal conditions in the Transfer Model.” Ann. N. Y. Acad. Sci., Vol. 858, September 11, 1998, simulation did not exactly match those in the experiment. pp. 36–46.

wwwwww.ansys.com.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 35 ELECTRONICS

Fan Tray Keeping Cool

CAD model of the in the Field radio chassis A communications systems company gains millions of dollars by using thermal simulation to bring tactical radios to market faster.

By Patrick Weber, Mechanical Engineer, Datron World Communications, Inc., California, U.S.A.

The communications systems Datron mechanical engineers face need to limit — for safety reasons — designed and built by Datron World the challenge of providing cooling external temperature of the heat sink Communications, Inc. present major management within a completely to 15 degrees C above ambient. thermal design challenges. The com- sealed radio cabinet in up to 60-degree Historically, thermal management pany’s radios travel with today’s war Celsius (C) ambient temperatures. design was based on engineering fighters around the world in helicopters Communication systems are designed experience and instinct. In order to and Humvees® as well as on foot. The with heat sinks external to the cabinet understand the cause of any thermal devices are designed to survive in a that use forced-air conventional problems, engineers had to test a wide wide variety of environments, ranging cooling. Components with the highest range of prospective solutions and from a sandstorm in the desert to a levels of power dissipation are mounted corresponding prototypes. The cost mountain blizzard. These systems internally near those fins. Radios con- of developing, building and testing dissipate substantial amounts of heat tain printed circuit boards (PCBs) for prototypes was high. But the resulting yet must be sealed to the outside the power supply, radio frequency (RF) delays in bringing each new product environment to prevent damage to filter, CPU and audio functions. These to market were even more costly. internal components — for example, PCBs generate substantial amounts of Datron engineers have improved the if the radio falls into a creek, it still heat. In addition to keeping junction thermal design process by using must work — and to prevent electro- temperatures of board components thermal simulation. magnetic interference. within specifications, Datron engineers The company now practices Simu- lation Driven Product Development and begins the thermal modeling early in the design process. Radios typically generate 125 watts output and dissipate approximately 220 watts inside a 15-inch wide by 15-inch deep by 5.5-inch high box. Initial models are developed based on very limited information, such as the size of the chassis, the RF output power and the expected efficiency of the radio. Engineers select primitive objects, such as cubes, as building blocks and parametrically assign dimensions and material properties. Surface properties are assigned to the outside surface of the enclosure to represent the olive paint that is typically used on the final product. In the early design stages, the Original radio design with ferrite core filters shows hot spots.

3636 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com ELECTRONICS

internal components are approximated The Natural Convection Challenge by a single component that dissipates One of the biggest challenges Datron engineers face is simulating natural the total amount of heat in the radio. convection. This is inherently difficult and expensive to simulate because the As the design progresses, more buoyancy forces are constantly changing. The Datron team developed a detailed information on the PCBs typical natural convection problem and compared the ability of all the becomes available. Mechanical engi- leading thermal simulation tools to solve it. Several of the software packages neers model the different PCBs and took 24 hours or more, while ANSYS Icepak software solved the problem in components within the chassis and only 20 minutes. Datron engineers liked the nonconformal meshing tools evaluate the thermal performance. in the ANSYS Icepak product that make it possible to separately mesh — ANSYS Icepak macros are used to usually with a finer mesh than the rest of the model — critical areas within the quickly generate models of standard system, such as high-dissipation components. Such a process increases the packages. Other macros are used to accuracy in the critical areas without unnecessarily increasing computational generate heat fins from parameters time requirements. including the number of fins, fin width and fin spacing. The design team limits to optimize thermal management possible to substitute other suitable the model to approximately 1 million and acoustics. components with lower thermal resist- cells by meshing smaller boxes around Using this approach, Datron engi- ances. If this problem had not been hot spots at higher densities. neers improved the performance of the discovered until after the detailed In a recent project, early models software prototype until it met thermal design process, it would have required showed that junction temperatures requirements within the required margin a considerable amount of time and exceeded the typical maximum of 125 of safety. At that point, they ordered work to correct. In addition, with this to 150 degrees C. The original design the first thermal hardware prototype. change, engineers discovered that specified ferrite core filters that are Testing showed that the thermal proto- they could decrease the number of fins relatively light but have a very low type closely matched the simulation required, which provided more room thermal conductance. Simulation using predictions and also met all of the on the rear panel of the enclosure and the ANSYS Icepak tool showed that thermal design specifications. As a made it possible to reduce the overall the devices heated up the surrounding result, no additional hardware proto- size and weight of the radio. air to the point of overheating neigh- types needed to be built, and the radio For Datron, simulation makes it boring devices. Based on this insight, was brought to market substantially possible to validate and optimize engineers replaced the ferrite filters earlier than if the company’s original designs much earlier in the develop- with aircoil filters that have a higher build and test method had been used. ment process, saving large amounts of thermal conductance. This design In other recent thermal design time and money. Engineering simu- change was the key to significantly projects at Datron, ANSYS Icepak lation has substantially reduced the reducing junction temperatures of high simulations showed that several power time required to bring new, improved power-dissipation components. Once transistors exceeded the junction communications technology to the a working design was obtained, the temperature specification. By knowing marketplace, and this can translate engineers used parametric modeling this early in the design process, it was into millions of dollars in revenue. ■

New design with aircoil filters shows that temperatures are reduced to acceptable levels. ANSYS Icepak model shows the speed of the air from the fans along with temperature (The filter temperatures in degrees C have gone from the 200s to the 90s.) contours on the chassis. Blue indicates cooler temperature.

Designing Against the Wind Simulation helps develop screen enclosures that can better withstand hurricane-force winds.

By Steve Sincere, President, Optimization Analysis Associates, Inc., Florida, U.S.A. Photo courtesy Richard Graulich/The Palm Beach Post.

One of the most popular residential structures performed analytical studies of existing screen enclosure in Florida is the screen enclosure (or screen room), designs using FBC wind loads. The company found that consisting of an extruded aluminum frame covered with the simplified methods failed to accurately calculate forces screen. These structures are primarily intended to keep and moments. Thus, the complex interactions among debris and insects out of swimming pools and to structural members were not adequately accounted for in increase living space to include an outdoor environment. the designs. Even so, they must be designed to resist hurricane-force Finite element analysis (FEA) provides the most winds ranging from 100 mph inland to 150 mph in coastal accurate method of determining such loads and inter- areas, depending on building code requirements. actions. Most engineers in the screen enclosure industry Recent hurricanes have revealed shortcomings in these do not have a background in FEA, however, and those with designs. Most are developed by contractors or enclosure such expertise often forgo these studies due to time and fabricators based on oversimplified analytical assumptions. cost constraints. The answer is an automated FEA-based Components typically are sized without regard to the screen enclosure design tool — one that is fast, is accurate Aluminum Design Manual (ADM), Specifications and and requires no FEA skills. Guidelines for Aluminum Structures as specified by A perfect platform for this task is ANSYS Parametric the Florida Building Code (FBC). Moreover, fasteners and Design Language (APDL) — a scripting language for fastening methods typically are selected for ease of automating common analysis tasks or even building fabrication or accepted convention rather than suitability for models in terms of user-specified input variables. This adap- the high wind loads. tive software architecture enabled Optimization Analysis Using ANSYS Mechanical software, Optimization Associates to create a web-based solution with a graphical Analysis Associates, Inc. — an engineering consulting firm interface through which screen enclosure designs could be specializing in mechanical analysis and design simulation — conveniently specified and automatically evaluated.

3838 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com BUILT ENVIRONMENT

If the user does not have a passing design (or if the design is too conservative), parameters may be revised and another iteration may be performed. Optimization Analysis Associates has written programs for more specialized work as well. A version of the model- building macro allows experienced users of software from ANSYS to create customized structures with nontypical shapes and/or nonstandard bracing configurations. Another macro uses the ADM data to produce color contour plots of interaction ratios, a calculated value of allowable stress ratio not existing in the results file. Locations of failure to meet APDL is used to automatically create, load and solve a full-frame model of a screen the ADM criteria give a quick visual indication of problem enclosure from parameters entered by the user describing the structure. areas. In addition, these allowable stress ratio plots can be animated with a modified version of the animation macro Users are required to enter only minimal input data, ANCNTR.MAC and overlaid on 3-D models showing including basic geometry information of the frame, wind deformed structural geometry. load criteria, a sketch of the plan view (to provide x and y One final specialized macro provides a cost estimate for coordinates for each corner), wall height, roof style, density the construction of the design. This macro interrogates the of structural members (number of columns to be used on a model to determine the length of each extrusion required wall, for instance) and sizes of the structural members. along with the square footage of screen and number of From this input data, three APDL macros then auto- fasteners, brackets, etc. It accesses an external price list file matically perform an analysis, check results against for each item, as well as factors for items such as labor, guidelines and generate layout drawings — all completed in scrap, overhead and profit to determine the total cost. The less than three minutes and requiring no user intervention. final output includes a complete parts list and a breakdown The first APDL macro reads in the data to create, load of all cost components. and solve the full frame model. Beam elements represent The automation of the modeling and simulation- the structural members, which are coupled in the model based evaluation using APDL provides a fast, easy-to-use to simulate hinged or rigid connections as necessary and extremely accurate method of structural frame designs. according to the type of connections used. Shell elements The screen enclosure industry now has the potential to represent the screen in a proprietary method that deter- produce hurricane-resistant structures, to significantly mines the load distribution on structural members. improve design productivity, and to improve cost estimating Solutions are obtained for the eight wind-load cases and profit margins of contractors and fabricators who use prescribed by the FBC. engineering simulation for their designs. ■ A second macro performs all required checks defined by ADM criteria. This complicated process begins by accessing external files containing section properties, material characteristics and other parameters associated with extrusions used in the design. Then a series of nested APDL do-loops performs the ADM calculations for all nodes on every structural member for each load case. The macro enters this data into arrays and sorts through them to determine the limiting members. The limiting members are written to a summary report text file, which is accessed by the web-based interface. The report provides a simple pass/fail output with percent overstress values (or interaction ratios). If the user has a passing design, a third APDL macro produces a layout drawing of the structure. This macro takes advantage of the graphical capabilities of ANSYS Mechanical software in generating annotation for dimensions and labels on screen enclosure 2-D layout drawings. Color contour plots of interaction ratios show locations’ potential wind-force failure in red.

www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 393939 ENVIRONMENT

Stabilizing Nuclear Waste Fluid simulation solidifies its role in the radioactive waste treatment process. Contours of solid particle concentration: During the suction phase, the solids were By Brigette Rosendall, Principal Engineer, Bechtel National, Inc., California, U.S.A. found to become more concentrated along the bottom of the vessel, as shown by the red color in the early suction.

The nuclear site at Hanford, while keeping all mechanical compo- the top to the bottom of the vessel. Washington, houses approximately nents well away from radioactive This model solves a separate set of 60 percent of America’s radioactive materials. Navier–Stokes equations for the fluid waste. Near the Columbia River, the Because there had been little and solid phases. It accounts for the site stores waste in 177 underground previous experience with PJMs in coupling between and within the tanks as a combination of liquid, this mixing environment, it was critical phases using exchange coefficients, sludge and slurry. A vast complex of that the engineering team be able to the most important of which is for the treatment facilities is being constructed accurately predict the ability of the fluid–solid interaction. The results made to convert this waste into a stable units to provide sufficient mixing for it possible to determine whether the glass-like material using a technology each of the different vessels in which mixing criteria were met under given known as vitrification, which involves the wastes will be treated. Within the operating conditions. mixing the waste processed in these waste treatment plant, each of the Each vessel in the plant has a vessels with hot glass formers such mixing vessels has substantially different mixing criterion; however, as rutile (TiO2) or silica. The mixture different geometries and processing most simply require that the solids is then poured into steel canisters requirements. In addition, there is remain in suspension and are mixed and cooled to solidify for permanent considerable variation in the character- well enough for accurate sampling and storage. One of the major challenges istics of the mixture of fluid and transfer to the next step of the vitrifica- in this process is keeping the solids in particles that will be processed in the tion process. Since pulse jet mixing is the waste in suspension during its different tanks due to separation and time in the holding vessels before the concentration of the radioactive com- separation and processing stages. ponents. The mixing performance of the

Avoiding contact of any mechanical PJMs is a function of the geometry of Air Process Air Process Air Process Air Process Air Process Air Process Air Process Air Process components with the slurry being the vessel, number of PJMs per vessel, mixed during holding was crucial and particle size, fluid characteristics, cycle led Bechtel National engineers working time and other variables. It was impor- on the project to select fluidic pulse jet tant to validate the ability of the PJMs mixers (PJMs). The action of the PJMs to keep the particles in suspension in To Vent System is carefully controlled by compressing each tank. air inside them to drive the slurry into To simulate the pulse jet mixing the vessel to create the mixing action. process, Bechtel engineers used the

Only 80 percent of the slurry volume ANSYS FLUENT fluid flow simulation Pulse Jet that is suctioned up into each PJM is package because of the software’s Mixers expelled out of the mixers, which pre- unique capability depth in modeling ABCD vents air from escaping into the vessel. multiphase mixing. The Eulerian At that point, the compressed air is granular multiphase model in ANSYS vented and a vacuum is applied to refill FLUENT software made it possible to the mixers. PJMs thus provide mixing predict the distribution of solids from Pulse jet mixer design

404040 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

ENVIRONMENT

a turbulent process, Bechtel engineers 160 chose ANSYS FLUENT software’s k-epsilon turbulence model based on 140 the results of a preliminary study. In this 120 study, computational fluid dynamics (CFD) specialists compared the results 100 of various turbulence models to 80 CFD experimental data to determine which tion (in) Experiment model was best at predicting the Eleva 60 velocity in scaled hydrodynamics tests. 40 The engineering group controlled 20 time-varying boundary conditions by a user-defined function that prescribed 0 0 0.02 0.04 0.06 0.08 0.1 Volume Fraction

Comparison of fluid flow predictions and experimental results for solid particle volume fractions averaged over tank radius and mixing cycle for a 140-inch-high tank

significant compared to the cyclic varia- different vessel designs and to tions in the concentration. At higher determine whether or not PJMs could elevations, there were more significant provide adequate mixing for each con- differences between the experiment figuration. The use of fluid dynamics in and simulation, with the simulation this application can potentially save a significant amount of time and money that otherwise would be spent on additional physical testing prior to beginning actual waste processing. ■

Mid-suction See also: www.bechtel.waste2glass.com the time-dependent velocities of each www.hanford.gov jet and tracked the solids concentration flowing through the nozzles and at the top of the domain. This eliminated the need to track the free surfaces inside the PJMs and at the fluid–air interfaces inside the mixing vessels, greatly simplifying the models. The Bechtel team could perform only very limited physical testing due to End-suction the high cost of building and testing the vessels and mixers. The company commissioned the construction of predicting more uniform mixing than the a full-scale PJM vessel to perform experiments demonstrated. experimental testing at Battelle Pacific Even though the ANSYS FLUENT Northwest National Laboratory. Fluid results demonstrate slightly better flow predictions of concentration and mixing than the physical experiments, velocity were then compared to the the results were close enough to give measured data. The results showed that Bechtel confidence in the ability of the the ANSYS FLUENT simulations slightly fluid flow model to provide pass–fail At the end of the drive phase, higher concentrations are underpredicted the solid-phase volume judgments in rating the performances predicted at the top boundary of the fluid domain while fraction, except at the higher elevations of the PJMs. Bechtel uses ANSYS concentrations were reduced at the bottom as the solids in the tank. This difference was not FLUENT technology to model the many were pushed away from the jet nozzle exits. www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 414141

OPTIMIZATION

Topology Optimization and Casting: A Perfect Combination Using topology optimization and structural simulation helps a casting company develop better products faster.

By Thorsten Schmidt, Technical Director, Heidenreich & Harbeck AG, Moelln, Germany and Boris Lauber, Application Engineer, FE-DESIGN GmbH, Karlsruhe, Germany

Engineers usually need to ensure both functionality and TOSCA® Structure software from German-based zero defects during component production. This often can FE-DESIGN GmbH. This product interfaces with ANSYS be achieved by simulating production processes and oper- Professional software. ating conditions in the virtual world. Development teams in In the past, the engineering team designed structural the machine tool industry need not only to prove the components with primary consideration to manufacturing mechanical strength of components but also to take into restrictions. But structural analysis of these component account rigidity and cost. designs often revealed weak points, especially for parts with Heidenreich & Harbeck AG in Germany was established a large number of load cases. Engineers then had to in 1927 as a foundry for cast iron components. Today, the perform time-consuming iterations with alternating modifi- company’s range of capabilities has expanded to include cations of CAD design and structural analysis in order to modern machine tools for finishing large, quality castings fulfill customer requirements. that have high accuracy requirements. The company’s Currently, the Heidenreich & Harbeck development in-house development department assists customers’ process starts with modeling the design space, which designers and develops castings of complex machine usually is easy to define. Engineers import the design space structures according to customers’ specifications. geometries into ANSYS Professional software and then The comprehensive software portfolio at Heidenreich & generate meshes. Boundary and loading conditions are Harbeck contains several 3-D CAD tools, process simula- applied. Groups of volume elements that are required for tion software for casting processes and numerical control optimization are defined in ANSYS Professional technology (NC) machining, a sophisticated cost calculation tool based as components. The engineering team exports solver input on 3-D CAD models, and project-planning software. In files from the ANSYS Professional tool and imports them addition, Heidenreich & Harbeck uses ANSYS Professional directly into TOSCA Structure software with the latter’s software for the simulation of mechanical properties. user interface. Using this wizard-based technology, the To provide optimal design proposals to accelerate the optimization setup can be executed with a few mouse clicks development of large castings, the company obtained by re-using group definitions from ANSYS Professional to

Four Guiding Wagons To Be Mounted

Eccentric Load

Model of original design, Design space, as provided Meshed, optimized structure before without optimization by customer with loading including casting restrictions in the definitions defined iterative design process

Topology optimization of support arm for paper unwinder Courtesy Bielematik. 4242 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

OPTIMIZATION

Structural Pre-Processing Optimization Pre-Processing Batch Optimization Process CAD System • FEA model Optimization wizard • Load cases Generation of geometry ANSYS Mechanical for design space • Components ANSYS Mechanical TOSCA Structure

Structural Post-Processing Structural Pre-Processing CAD System • Validate optimization results FEA model of TOSCA Structure • Final evaluation redesigned geometry Redesign using extensive manufacturing knowledge ANSYS Mechanical ANSYS Mechanical

Scheme of topology optimization using TOSCA Structure based on solver from ANSYS

define the design area, frozen areas, evaluation areas for manufactured without the use of cost-intensive cores in the design responses, and areas for the application of manufac- sand mold. An automatic or user-defined parting plane may turing constraints. The optimization procedure is carried out be specified. For the design of stiffening ribs, the casting in a batch process. TOSCA Structure software iteratively constraints may be coupled with a wall thickness constraint. launches the ANSYS Professional solver for the analysis of A customer provided Heidenreich & Harbeck with the the design space model and then launches the optimization design space of a support arm for a large paper roll module that evaluates results and changes material proper- unwinder loaded with an eccentric force. The design with no ties. Users who want to remain in the familiar ANSYS casting restrictions led to a hollow profile without access- product environment may transfer the results produced by ibility for fastening screws. A second optimization with the TOSCA Structure product back to ANSYS Professional casting restrictions resulted in a two-beam structure. The for post-processing using a file containing the material final design combined the benefits of both proposals property values for the finalized optimization. (accessibility for screws along with hollow profile for cable Heidenreich & Harbeck uses an optional module and tube-laying, which the customer added to the specifi- from FE-DESIGN called TOSCA Smooth to convert cations after he became aware of the first design proposal). the optimization results into IGES or STL files containing Due to topology, optimization rigidity was increased by 25 isosurfaces and cutting splines based on the normalized percent, and weight was decreased 34 percent compared material distribution. with the former two-piece design. For the design of castings, consideration of In another project involving a vertical lathe housing, the manufacturing constraints plays a very important role. It is customer delivered two-dimensional sketches with the essential to take into account demolding constraints for expectation of final pattern drawings within only three parts with low-cost restrictions. For a part that is loaded by weeks. Using TOSCA Structure software, the rigidity an eccentric force leading to a torsional loading condition, a requirements were fulfilled with minimal material consump- non-restricted optimization will generate a hollow section tion, and time-consuming design iterations were avoided. that would lead to high torsional rigidity. By applying a This reduced development lead time by approximately demolding constraint in the TOSCA Structure tool, the 50 percent. ■ engineer can obtain a design proposal that is less rigid but has no undercuts and cavities and may, therefore, be Visit www.huhag.de and www.fe-design.de for further information.

ANSYS Professional simulation results, which Simulation of the casting process Final component design are evaluated during the optimization process

wwwwww.ansys.com.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 434343

MARINE

Fighting Fire with Simulation The U.K. Ministry of Defence uses engineering simulation to find alternatives to ozone-depleting substances for fire suppression.

By Michael Edwards and Michael Smerdon, U.K. Ministry of Defence, Bristol, U.K. Yehuda Sinai and Chris Staples, ANSYS, Inc.

Fires onboard ships are not uncommon and pose a danger to both crew and equipment. It is vital to develop effective methods to extinguish these fires. At the same time, international agreements such as the Montreal Protocol on Substances that Deplete the Ozone Layer have been signed. These agreements limit the use of firefighting agents such as Halon that, though effective, come with a high environmental price. In order to find an alternative to Halon, the U.K. Ministry of Defence (MOD) completed a comprehensive research program that looked at alternative fire suppression technologies for use on Temperature isosurfaces and droplet trajectories before fire extinction is completed in a ship’s machinery space Royal Navy vessels. The work led to the development of a low-pressure water Because of the complexity of the measurements for two working fluids: mist system, or fine water spray (FWS). application, the simulation involved a water and water with 1 percent by This new FWS system combines large number of software models volume AFFF. The university also salt water from a ship’s high-pressure that included existing capabilities, measured to ascertain whether the salt water (HPSW) system, which typi- existing models that required some additive affected the terminal speed cally operates at a pressure of 7 bar, special functionality extended through of a droplet with a given mass. The together with a 1-percent-concentration FORTRANTM, and some models that SBU measurements were employed in aqueous film-forming foam (AFFF). were implemented entirely through the initial conditions for the particle As part of this program, MOD FORTRAN. The simulation models transport model. validated and used simulation as a tool were validated against data from a To determine how the fire becomes to assess the performance of the FWS large-scale experimental rig. extinguished, the combustion model system, with and without additive, Measurements of the FWS droplet calculates the fuel evaporation rate when fitted onboard a ship. This initial conditions, in air and without fire, from the heat delivered to the fuel by analysis decreased the need for expen- were commissioned at South Bank the fire. The model then predicts where sive fire testing for future assessments University (SBU), London, using high- and how rapidly fuel vapor is burned and design of fire control measures. speed photography. This provided and heat is released exothermically. As The United Kingdom ANSYS office information at a specified, small radial the fire cools after spray initiation and developed a fluid dynamics model distance from the nozzle, for velocity radiation is attenuated by the spray, using ANSYS CFX software, validated (predominantly radial) and mass flow soot, and gaseous products (as well as it blindly against MOD’s full-scale for each of a group of droplet-sized the foam film when that is present), the experiments, and demonstrated its bands, as a function of azimuth heat returned to the pool of liquid fuel application to a real vessel. and elevation. SBU performed is diminished and so is the evaporation

444444 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com

MARINE

rate. If the spray system is appropriately designed, then extinction is achieved when combustion process ceases. Fuel vapor usually vanishes a short while after the fuel evaporation rate falls to zero. The MOD and ANSYS research teams validated the fluids model by comparing it to data from a MOD experimental rig. The rig was large scale with a volume of 1,080 cubic meters. Inside the experimental rig Droplet trajectories and maps of water vapor mole Simulation model of the rig geometry and temperature there were mockups of the large fraction after spray inception isosurfaces before spray inception equipment — diesel generator and gas turbine enclosures typically found within a Royal Navy (RN) machinery 3 and 1.5 square meters, respectively. which had been studied in previous space. The FWS comprised 16 GW The teams validated the simulation research by ANSYS) was an important LoFLowTM K15 nozzles fixed on a against two separate tests: water factor in this regard. Other influences 3-meter grid near the ceiling. Buckets spray for the larger tray and water on the results of the model were identi- at the floor were used to measure spray with additive for the smaller tray. fied: The fuel model used heptane cumulative water delivery. Additional The results of the validation were rather than F-76; the coefficient of instrumentation was added to the generally encouraging, and the pre- restitution was set at zero for water space to enable validation of the dicted extinction times and method of droplets so that when they hit struc- model. Liquid fuel (F-76, which is a extinguishment were reasonably pre- tures they were removed from the common fuel for shipboard diesels, dicted. There were some noticeable model; and positioning of the mockup gas turbines) was provided in one of discrepancies, and there was evidence structures, fuel trays and nozzle posi- two rectangular trays, having areas of that building leakage (the effects of tions represented a worst case.

Engineering Simulation for the Built Environment The technology from ANSYS that can be applied to fire propagation, fire suppression and smoke management for ships, airplanes, trains, cars and trucks is also used for ventilation and thermal modeling in the built environment industry. These comprehensive multiphysics capabilities, which address safety and comfort concerns, are frequently used upfront during the design and construction of buildings. In order to provide information for design improvement, design optimization and energy efficiency in the built environment, predicting conditions such as air velocity, temperature, relative humidity, thermal radiation and contaminants is extremely important. The simulation must also take into account ventilation, heat loss and Courtesy SOLVAY S.A. solar radiation effects on the structure walls, roof, floors, windows and doors, as well as the presence and activity deterioration during catastrophic events. These can of people and equipment in these areas. Simple air flow be analyzed in detail using explicit dynamics and modeling assists engineers and architects in quantifying structural modeling. Solutions from ANSYS allow for and simulating the impact of structural and equipment the analysis of events ranging from explosions that design modifications on the thermal comfort of a space’s encompass blast waves (in the context of homeland occupants. security) to deflagrations in combustible mixtures. Engineering solutions from ANSYS provide a cost- — Thierry Marchal effective and accurate means of designing efficient Industry Marketing Director smoke management and detection systems. The unpar- Materials and Consumer Care, ANSYS, Inc. alleled breadth of solutions across multiple disciplines provides the ability to quantify the behavior of materials subjected to fires or extreme heat and possible structural For more information, visit www.ansys.com/industries/hvac. www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 454545

MARINE

Descriptions of Models Used in the Simulation After completion of the validation, the model was successfully applied to Model Implementation Purpose a real machinery space aboard an RN RANS turbulence modeling (SST) Existing model Determines turbulent transport ship. MOD is proposing the use of Laminar flamelet Extended To include combustion modeling of FWS in its future vessels for fire sup- combustion modeling (Peters) current model heptane fuel and evaporated water pression that was validated by the vapor, with reduced set of species experiment [2] and this work. ■ Soot modeling (Fairweather et al.) Implemented Assesses impact of soot on infrared new model radiation and visibility References Transient Lagrangian particle Existing model Assesses the impact of water spray transport model on fire and fuel, with two-way [1] Sinai, Y., Staples, C., Edwards, M., Smerdon, coupling of mass, momentum, M., “CFD Modelling of Fire Suppression by convective heat and radiant heat Water Mist with CFX Software,” Proc. Interflam 2007, Vol. 1, 2007, pp. 323–333. Multiple droplet size groups Existing model Determines penetration since larger drops are better at penetrating key [2] Hooper, A., Edwards, M., Glockling, J., regions directly, small droplets “Development of Low Pressure Fine Water evaporate quickly and can reach Spray for the Royal Navy: Results of Full key regions by entrainment Scale Tests,” Proc. Halon Options Technical Working Conference, 2004. Coupled fuel evaporation Implemented Calculates fuel burning rate new model Acknowledgments This work was a team effort. The authors wish Subgrid droplet–congestion Implemented Estimates direct removal rate of to thank Dr. J. Glockling of the Fire Protection interactions new model droplets by subgrid congestion Association, Dr. G. Davies and Prof. P. Nolan of Soot scavenging by water droplets Implemented Determines how scavenging affects South Bank University, as well as P. Guilbert, P. new model infrared radiation and visibility; also Stopford, H. Forkel and P. Everitt of ANSYS, Inc. predicts delivery of scavenged for their contributions. substances to boundaries © British Crown Copyright 2009/MOD. Additive effects on water spray Implemented Predicts attenuation of radiant heat Published with the permission of the Controller and fuel evaporation rate new model arriving at pool surface of Her Britannic Majesty’s Stationery Office.

46 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com TIPS AND TRICKS

Reusing Legacy Meshes ANSYS tools enable users to work with finite element models in various formats for performing simulations as well as making changes to part geometry.

By Sébastien Galtier, Software Developer and Pierre Thieffry, Product Manager, ANSYS, Inc.

Figure 1. NASTRAN finite element When designs from past projects must be analyzed, or when a model of a connecting rod modified version of the geometry must be evaluated, the starting point is generally the original CAD model. In some cases, however, only legacy finite element models are available that cannot be imported directly into the user’s current simulation software. These include NASTRAN®, ABAQUS® and ANSYS FLUENT models, for example, as well as many text-based archival versions of ANSYS models. Fortunately, tools in the ANSYS Workbench environment have been developed so users can easily convert these models Figure 2. Geometry created from for use in creating new simulation models of the original segmentation based on curvature design and also in modifying the original shape to meet new design detection requirements. Legacy models such as the mesh for a connecting rod, shown in Figure 1, can be read into ANSYS FE Modeler, located in the Toolbox section of ANSYS Workbench version 12.0. Once imported, the model is handled by the Skin Detection tool in FE Modeler to provide a proper segmentation of the model’s facets. The quality of the segmentation is key to the process — especially when modifying the shape of the model — and the procedure consists of grouping the external faces of finite elements so they Figure 3. Loads and boundary conditions accurately represent faces similar to a geometric model. Edges and applied for analysis with mechanical vertices of the model will then be naturally derived from these faces. simulation software from ANSYS Several methods can be used to identify the faces: detection by angles (between the normal orientations of neighbor elements), detection by curvatures, or employment of facet groups defined by the user. This last method helps in creating specific areas in which loads and boundary conditions can be applied. Figure 2 shows the resulting geometry generated based on curvature detection in FE Modeler from the legacy mesh. A mechanical simulation system from ANSYS then can be linked to FE Modeler to apply loads and boundary conditions, as shown in Figure 3, and the model then can be solved to determine the resulting stresses and deflection (Figure 4). Figure 4. Total deformation results from After such an analysis, the model may need to be the analysis modified because the existing design does not meet current technical requirements. For this purpose, FE Modeler provides capabilities to modify the geometry through a feature called the ANSYS Mesh Morpher. A so-called target configuration is created by duplicating the initial geometry. Then transformations such as offsets, translations or rotations can be applied to the geometric entities. Figure 5 shows how offsets can be used to enlarge or shrink the holes. Once the geometry has been modified, ANSYS Mesh Morpher will Figure 5. Deformed geometry, in which the hole on the left has gotten transform the initial mesh to conform to the target configuration. smaller while the other two have These transformations are parametric, with each geometric feature been enlarged www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 4747 TIPS AND TRICKS

affected by a parameter that is used as a way to control the amount of morphing between the initial and target configurations. Changing the shape of an existing member can be achieved with projection to a new CAD shape. In this case, the faces or edges created by the skin detection process are projected onto an imported CAD model. It is important to note that mesh morphing modifies only the node coordinates, and no remeshing occurs during the process. Once the mesh has been morphed, the model can be used in the mechanical simulation exactly as it was done with the original model. Since the geometry topology remains the same, all loads and boundary conditions Figure 6. Raw result of conversion in Parasolid applied to the initial model are still valid, so the analysis format, with all faces NURBS representations can proceed as before. In this example, changes to the model geometry did not affect the general shape of the model too heavily: No holes were added, for example, and the topology remained the same. The FE Modeler application used in conjunction with ANSYS DesignModeler software provides all necessary tools to allow for such changes. To make more significant changes to the model, the initial geometry must be converted to a Parasolid® model. The result of the conversion is a set of surfaces corresponding to each of the faces obtained from the Skin Detection tool. The surfaces can then be sewn together in FE Modeler to create volume bodies. Figure 6 shows the raw result of this conversion, and Figure 7 illustrates the new design after sewing all faces and modifying Figure 7. New design after sewing all faces together and the geometry with the standard features of ANSYS modifying geometry with ANSYS DesignModeler software DesignModeler. In this way, these tools in the ANSYS Workbench framework allow legacy models to be reused in a process that is not only faster but also less error-prone than manually recreating meshes from scratch. ■

48 ANSYS Advantage • Volume III, Issue 1, 2009 www.ansys.com ACADEMIC

Expanding © iStockphoto.com/cinoby and Stent Knowledge © iStockhpoto.com/fasloof Simulation provides the medical industry with a closer look at stent procedures.

By Matthew R. Hyre, Associate Professor of Mechanical Engineering, James C. Squire, Professor of Electrical Engineering, and Raevon Pulliam, Virginia Military Institute, Virginia, U.S.A.

Heart disease, often caused by partially blocked coro- The team at VMI hypothesizes that restenosis may be nary arteries, is the most common cause of death in the the result of arterial injury incurred during the stenting world. Stenting has become one of the most popular forms procedure itself. During this procedure, the medical team of treatment to open plaque-encrusted atherosclerotic inserts a balloon, sheathed by the stent, into the artery and coronary arteries, with hundreds of thousands of such pro- inflates it. Once the stent expands, the balloon is deflated cedures performed in the United States each year. However, and removed, leaving the stent in place. according to the American Heart Association, about one in The engineering team at VMI identified one possible four stent patients will experience restenosis, a repeated reason for injury: end flare, which is caused by balloon over- narrowing of the stented artery, less than six months after hang at the end of the stent. This exerts increased pressure the procedure. Some patients with restenosis must undergo on the arterial wall and may scrape it during inflation, which a second stenting procedure to alleviate the subsequent could stimulate uncontrolled cell growth in that area. blockage, while for others a full bypass operation is the only The balloon’s mechanical properties vary dramatically solution. A team from the Virginia Military Institute (VMI) is during the expansion process. Though it begins as a highly combining simulation with animation software from Compu- flexible material, the balloon eventually expands in a nonlinear tational Engineering International (CEI) to help identify a fashion as it nears the stent’s final diameter, making the prob- possible cause for restenosis and to find solutions that lem numerically unstable. A factor that is critical to accurately might help reduce the risk of developing it. simulating the problem is how the structure of the balloon, the stent and the artery are meshed. The team used HarpoonTM, from Sharc, Ltd., to generate a complex mesh designed to follow the balloon, stent and artery through the expansion from a 1 millimeter diameter to a 3 millimeter diameter geometry. Once the mesh was established, the data was exported to ANSYS Mechanical software to provide information about stresses and The stent expansion process, with the stent shown in light gray, the balloon in dark gray and the artery colored by arterial stress geometry changes that occur during expansion. The team used EnSight® to turn the simulation data into animations that depict the inflation process. The resulting images allow the medical research team to visualize the process for the entire assembly or to focus on the individual components — options that are impossible during the stenting procedure itself. By using simulation and visualiza- tion tools together, manufacturers may be able to redesign In this image, the stent and balloon are hidden, and the remaining plot depicts only and numerically test stent designs and procedures, arriving the artery after stent inflation. The contours represent arterial stress. The red ring, which occurs at the location of highest stress, aligns with the location at which end at a very clear picture of how each variable affects the flare occurs during stent inflation. overall issue — all without a patient. ■ www.ansys.com ANSYS Advantage • Volume III, Issue 1, 2009 49

ANSYS, Inc. Southpointe 275 Technology Drive Canonsburg, PA U.S.A. 15317

Send address corrections to [email protected].