Rapid Flow Modeling with FLUENT for CATIA V5 – Part 2

Christoph Hiemcke (Aerospace CFD Engineer, Fluent Inc.) and Tushar Sambharam (Applications Engineer, Fluent India Pvt Ltd.)

In the last issue you were introduced to FLUENT for CATIA V5, which is a flow solver that is fully embedded inside CATIA V5. It is implemented as a workbench and the designer never leaves the familiar CATIA graphical user interface (GUI). Also, the resulting analysis is fully associative to the original part or assembly, so that changes in the part are immediately reflected in both the mesh and the analysis. Last month, you also saw a number of flow solutions produced using FLUENT for CATIA V5 for different industry areas.

In this article, we will focus on the technical collaboration between Dassault Systèmes and Fluent that has led to some CATIA FEM Surface & Solid (FMS and FMD) mesher enhancements. These enhancements benefit the CFD process and are accessible from FMS and FMD via the FLUENT for CATIA V5 workbench. We will demonstrate the enhancements using an example from the aerospace industry, a transport airplane.

Figure 1: FLUENT for CATIA V5 screenshot: Contours (fringes) of static pressure on the skin of a transport airplane

One of the "grand challenges" for any mesher is to produce a mesh suitable for external aerodynamics, i.e., the prediction of the lifting force, drag force, and pitching moment acting on an airplane. Such a mesh must have enough resolution to resolve small gaps and edges and regions of high surface curvature, but it must also be coarse enough to avoid a huge cell count.

Our first step in planning the mesh for the transport airplane is to anticipate the flow conditions. For this transport airplane our goal is to predict the lift and moment at cruise conditions. The physics of such a flow indicates that there will be hardly any separation, so we need not place prism cells (extruded triangles) next to the skin in order to resolve the boundary layer. Also, since there will not be any separation except very near the trailing edge, we can keep the mesh over most of the wing quite coarse.

However, we must be careful to have enough resolution at the leading edge to avoid large slope discontinuities there, and we should resolve the wing tip sufficiently to adequately capture the tip vortex.

The shape of the leading edge is often defined by means of the leading edge radius. In other words, we can imagine that a typical rib has a semicircle at its nose. One rule of thumb is to make sure that between 15 and 20 cells make up that semicircle.

For the wing tip, the minimalist's rule of thumb is to make sure that at least three cells span the thickness of the center section of the wing tip profile.

As we look at the image of the transport airplane, we observe that our comments about the leading edges apply not just to the wing, but also to the vertical and horizontal tail, to the engine inlet lip, and to the engine nacelle pylon. Similarly, the rule of thumb for the wing tip also applies to the tips of the vertical and horizontal tails.

Just for fun, here is an image of a mesh for the transport airplane that was generated without being very careful (see Figure 2). We can see that all parts of the airplane are under resolved, especially the leading edges and tips. Also compare how unacceptable spotty the pressure distribution is compared to the one shown in the first figure.

Figure 2: Unacceptable mesh for the transport airplane

One way to refine the mesh at the leading edges and wing tips is to begin with some global mesh parameters, and to then make some local mesh modifications. But such local modifications in CATIA FMS are tedious. It would be nicer to have a global way of producing local refinements automatically!

Based on our experience with meshing such geometries, Fluent Inc. developed size functions for its meshers to control the mesh globally. The idea here is to make the mesher sensitive to gaps and local surface curvatures. The size functions also make sure that any transitions from fine to coarse mesh occur in a smooth manner and are controllable via a growth rate.

One approach for detecting curvature is to first cover the faces with a first triangular mesh, in order to discretize it. For each triangular face we compute the face centroid and the face normal. We then measure the angle between its normal and the normals of its neighbors, and that gives us an idea of how well we are resolving the local curvature. We also compute the distances between the face centroids to detect proximity, i.e., sections of the wall that are very near one another.

Through the CAAV5 Partnership with Fluent, Dassault Systèmes implemented their own version of the proximity size function and added a user-specified growth rate to govern the smooth transition from the fine mesh created by curvature size functions or local mesh specifications to the coarser mesh away from those areas.

Figure 3 shows an idealized wing. Here we simply selected some common global mesh parameters: we can see that the leading edge is under resolved. Also note the side walls of the wind tunnel test section.

Figure 3: Coarse mesh from the standard mesher

Next we use the curvature size function that is part of the standard FMS/FMD workbench. Note in Figure 4 that we have used the 2D Standard mesher and that we have specified a relative sag of 0.02. That value allows us to resolve the leading edge as we had planned. However, we note that the mesh grows too quickly as we move aft from the leading edge - we need to control that transition using the growth rate.

Figure 4: Effect of using the size function of the standard mesher

The growth rate can be controlled by using the 2D Optimized mesher, as shown in Figure 5. We have kept the same value for the relative sag but we also observe the two new controls at the bottom of the panel. The default growth rate of 1.1 keeps changes in the size of adjacent cells below 10%. We used the minimum face number to dictate that any gaps or ledges (in this case the trailing edge) should be resolved using at least one layer of cells. We note on the right of Figure 5 that the new mesh transitions smoothly from the leading and trailing edges.

Figure 5: Optimized meshing via size function

Having demonstrated the principles involved in the size function using a simple wing, we end by showing an array of images (see Figure 6) from the size function based mesh for the transport airplane. Note the smooth transitions and that the fine mesh is only used where it is really needed, which gives both accurate results and fast convergence. Also note that the high-quality pressure field contours shown in Figure 1 were calculated using this optimized mesh.

Overall mesh	Wing tip
Horizontal tail	Engine nacelle
Figure 6: Transport airplane with optimized mesh based on size functions

For more information on FLUENT for CATIA V5, please visit: www.fluentforcatia.com. Also visit www.fluent.com for more general information. Christoph Hiemcke can be reached at cbh@fluent.com.

Royal Canadian Mint Promotes Innovation with PLM Solutions from IBM and Dassault Systèmes

Esteemed international producer of circulation and commemorative coins reduces costs and time to market for new designs

In advance of the XXIV Mint Director's Conference (Paris, May 1-7), IBM and Dassault Systèmes announced that the prestigious Royal Canadian Mint (RCM) has streamlined the design and production of its high-quality coins and medals with their product lifecycle management (PLM) solutions.

Recognized as one of the largest and most versatile mints in the world, the RCM is leveraging PLM to foster innovation and business efficiency for the best benefits of its customers. It has made gains throughout the entire product development process by using CATIA V5 for 3D product design and SMARTEAM for product data and lifecycle management.

"Our PLM solution automates and integrates the entire quotation, new product development and engineering process," said RCM Chief Information Officer Neil Hallam. "Simplifying and bringing better control to the engineering change process, and allowing users to more quickly and easily access existing data for reuse, has driven down engineering and manufacturing costs. It furthers our goal of becoming a lean enterprise."

CATIA V5 enables RCM engineers to create 3D virtual definitions of a coin, its engineering specifications such as weight and dimensions, as well as the associated tooling and packaging. Using SMARTEAM, the RCM incorporates into its business processes information such as the metal's formula and properties, engineering specifications, tooling and production data. By consolidating its intellectual property in a single repository accessible by everyone in the production and marketing cycles, the RCM has now increased customer responsiveness, developing and marketing innovative coins more efficiently.

"Like many organizations, prior to its PLM initiative, the Royal Canadian Mint had vast knowledge of its industry and processes, but few mechanisms to use that expertise to its competitive advantage," said Joel Lemke, president, Dassault Systèmes Americas Corp. "The innovation the RCM achieves with IBM and Dassault Systèmes' PLM solutions demonstrates how a forward-looking company can do better work in less time by bringing its institutional knowledge into every facet of design and production."

About the Royal Canadian Mint
The RCM, an ISO 9001-2000 certified company, is the Crown corporation responsible for the minting and distribution of Canada's circulation coins. In operation since 1908, the RCM is recognized as one of the largest and most versatile mints in the world, offering a wide range of specialized, high quality coinage products and related services on an international scale. For more information on the RCM, its products and services, visit www.mint.ca.

DELMIA Academic Partner Program Helps Future Engineers

In an effort to give students real-world experience in engineering and prepare them for future careers, DELMIA is reaching out to the educational and business communities with the DELMIA Academic Partnership Program (DAPP). This academic program provides both career awareness and career preparation by offering educational institutions - from elementary schools to universities - academic partnerships that make highly sophisticated digital software available through special licensing and purchase arrangements.

DAPP offers many benefits by providing powerful technology that enhance curricula and allow students the opportunity to learn about advances in the digital manufacturing arena. DELMIA facilitates this educational journey by bringing together a set of digital manufacturing solutions targeted for the industry sectors where continuing optimization is a determinant factor. These solutions not only teach tomorrow's engineers, but help today's manufacturers bring their products to market more quickly, while reducing production costs and encouraging innovation.

Through DAPP, academic partners can individually determine how they intend to implement DELMIA digital manufacturing solutions into their education program. Plans can include one or more of these approaches:

• Integration of simulation in education to help teach design and manufacturing concepts within a current program
• Creation of new courses and project-based activities to prepare students for a workplace that will find a pervasive use of simulation as a design and management tool
• Creating a link with industry to utilize DELMIA solutions to demonstrate how cost savings and reduced time-to-market can be achieved by the practical application of simulation software tools

In addition, we can demonstrate the Integration of Product (CATIA) "What to Build", DELMIA (Process) "How to Build it" and Resource (CATIA) "Build with what." Plus, show how our customers -- such as Boeing, Toyota and DaimlerChrysler-- are proving-out, testing and building their products completely in a digital environment, prior to the physical build. When the design changes, the process and resource changes.

DELMIA provides advanced digital manufacturing technology for:

• Assembly
• Robotics
• Fastening
• NC Machining
• Human Modeling
• Automation
• Process Planning
• Discrete Event Simulation (QUEST)

Together DAPP and the utilization of DELMIA solutions bring positive results to engineering students and the engineering community overall. Some of the many DAPP achievements include DELMIA's partnership with the Heart of Georgia Technical College to provide statewide PLM software training; the use of DELMIA solutions in the General Motors/Department of Energy Challenge X Project by Mississippi State University; and the partnership with Oakland Schools and DELMIA to form a new academy of digital manufacturing with a curriculum in advanced technology, design and virtual simulation tools.

To find out more about DAPP, contact Roy Smolky by calling 248-205-5180 or e-mailing roy_smolky@delmia.com. You may also visit our website at www.delmia.com.

Case Study: How Collaboration Tools Accelerate Issue Resolution for Electronic Design Automation Users

The Challenge
A provider of high performance EDA tools, with customers and offices located in multiple geographies and with a limited technical support staff, faced the challenge of supporting a growing number of increasingly sophisticated customers across multiple time zones. In addition, the company was fully aware that providing a high level of customer satisfaction was a key factor in maintaining user loyalty and in growing its business.

When customers encountered what they considered to be a problem, the usual process involved the customer contacting the support team by phone or email. The issues generally fell into three categories:

1. Customers would often be "pushing the envelope"; that is, stretching the software's functionality beyond its current capability.
2. Other times, customers were reporting incidents that are due, in fact, to the customer's unfamiliarity with using the provider's tools.
3. Some incidents would identify software defects, which required that data be sent to the support team for further analysis and problem replication.

It is inefficient and time consuming to open support tickets and gather corroborative information when the incident is a result of misuse or miscommunications between the customer and the support agent. Frequently the EDA tool provider requested that the customer transfer confidential designs in order to recreate the reported issue. But because of the confidential nature of the designs, the customer was reluctant to comply. The provider therefore could not reproduce the problem and days or weeks of delay ensued. In some cases, a support person would be flown to the customer site to troubleshoot the issue. In the highly competitive chip design market, every design delay can increase time-to-market and lessen revenue opportunities.

Although the EDA tool provider tried to alleviate the communication problems with web conferencing, the service was prohibitively expensive and the company did not want to force customers into learning a new tool in order to obtain support. In addition, the viewing and refresh response time for the complex designs were so poor that the web conferencing tool introduced further miscommunication, lengthening phone calls and increasing frustration on both sides.

The Solution
The EDA tool provider chose Imera TeamLinks™ because of its integrated online customer support and demonstration capabilities that immediately and securely connect to customer desktops across domains. This is important because many of the customers are highly security conscious and have implemented a strong perimeter defense. To maintain customer confidence with the support organization, the view sharing process had to be seamless. Because all transactions and data remain on the customer's premises, within the control and monitoring of the customer, additional security concerns were eliminated. TeamLinks also facilitates instant collaboration with other team members behind frontline support personnel so that more senior engineers can be immediately called in to support calls and assist with issue resolution.

Using TeamLinks, the provider's support teams can remote drive end-user desktops, with permission, to define and troubleshoot problems without the delays of, or time consumed by, file transfer and asynchronous problem recreation. TeamLinks' refresh rate, designed specifically to handle the demands of data-intensive graphics, provided superior responsiveness when needed for further clarification. And due to TeamLinks presence aware features, technical support engineers can contact each other for additional help or knowledge transfer directly and instantly because they see who is currently online and available. This is especially beneficial when teams are in different time zones. The company is now considering extending this capability to its high touch customers, which will enable them to contact designated support engineers when they come online.

The EDA tool provider has also used TeamLinks for real-time web-based demos of its sophisticated tools. These demonstrations are further enhanced by TeamLinks' ability to allow potential end users to test drive the provider's tools, and obtain an immediate first impression on the software features and usability without the need to issue temporary demo licenses.

Issue Resolution Time (Figure 1)

The Results
The EDA tool provider's support organization reports accelerating resolution time by 15% for calls requiring visualization support. Additionally, support engineers' productivity increased by 20% because they are able to handle more issues in the same timeframe, with no increase in personnel or resources. Issues are resolved faster, and the time normally wasted simply trying to establish communications was removed from the critical path of product issue fixes. By providing immediate access to support, TeamLinks has improved customer satisfaction as well as corporate productivity.

For information about TeamLinks, visit Imera at www.imera.com. To learn more about how TeamLinks can improve your customer support operations visit http://www.imera.com/solutions/onlinesupport.html