Contributed by Dr. Ilhan Bayraktar
Old Dominion University

Norfolk, VA

November 2001

External truck aerodynamics animation. Models such as this will lead to modifications that increase fuel efficiency by minimizing drag force. Reducing drag force will also lessen environmental pollution and improve stability and vehicle control. Download AVI Animation (3349 KB)

Dr. Ilhan Bayraktar is a Ph.D. Candidate at Old Dominion University’s Aerospace Engineering Department. His primary research interest is numerical simulation of complex, large-scale flow problems involving heat transfer and structural interactions in participating media (in this case flow-structure interaction). The applications that he is currently working on include compressible and incompressible flows, detonation in aerodynamics and heat transfer problems.

Outlined results are from Ilhan’s dissertation, which has been conducted at Old Dominion University and Langley Full Scale Wind Tunnel. Old Dominion University has assembled a research group on Ground Vehicle Aerodynamics. This particular research project, directed by Dr. Oktay Baysal, focuses on analyzing heavy ground vehicle aerodynamics and understanding complex wake flow behind vehicle bodies. (Further research areas include under-body and under-the-hood aerodynamics, internal aerodynamics and heat transfer problems.)

The animation above shows pressure contours on a truck surface. Maximum pressure occurs in the front region of the model (red regions). The side of the trailer has relatively low pressure and no separation exists (there is no wake or reattachment in the flow). The figure below shows a circulation region on the back of the truck – a very important region for aerodynamic drag. Although the highest pressure takes place on the front surfaces, most of the drag occurs at the back due to the separation of the flow.

A circulation region on the back of the truck can be easily seen in this image. Most of the drag force takes place due to the separation of the flow at the back.

There are two types of drag forces on bluff bodies; friction drag and pressure drag. Computational studies show that about 80% of total drag is from pressure drag, and the rest is from friction. The maximum pressure difference is observed at the back surface of the truck, where complex flow phenomena, such as separation, reattachment and vortices are found.

Maximum pressure contours can be seen on the front of the truck.

The computational part of this work was conducted on Sun 10000 Supercomputer using up to 32 processors. Full-scale domain was created with 12.5 million mesh elements. Approximately 10 GBs of memory was allocated by implicit Reynolds averaged Navier-Stokes Solver and roughly one week runtime was spent for converged result.

Visualization is a key component in transforming the raw data into something useful for engineering and scientific analysis. Using Tecplot, Ilhan can quickly explore the experimental and computational data to better understand the aerodynamics of the flow and generate plots to communicate the results. In the images above, 3-D stream tubes help identify separation and recirculation regions.

A truck inside the Langley Full Scale Wind Tunnel

Langley Full Scale Wind Tunnel (LFST), the largest university-operated wind tunnel in the world, plays a very important role in the research group. LFST tests full-scale commercial vehicles like trucks, cars and aircraft. Since the research group works on full-scale aerodynamics of ground vehicles, they need a wind tunnel to test full-scale ground vehicles. A purely computational study is not complete without experimental support or vice versa. Because there are just a few places, which can test full-scale ground vehicle models, in the world – LFST is the perfect place for comparison and validation studies for such studies.

This study will be a benchmark case resulting in a database containing both experimental and computational results. These results will lead to modification devices that minimize drag force and increase fuel efficiency. Reducing drag force will also lessen environmental pollution and improve stability and vehicle control.

Pressure contours on various back surface configurations. (Bluff bodies represent simplified ground vehicles.)

Contributed by Jonathan Binns
BMW Oracle Racing

The graphics of refined grids show the results of refining the grid for a boundary element code. The top image shows the dynamic pressure distribution for the older grid, the bottom shows the dynamic pressure for a refined grid.

The Researchers

The design team members of BMW Oracle Racing share a common, single goal: to design the International Americas Cup Class (IACC) yacht that will win the 32nd America’s Cup in 2007. The history of yacht design involves the use of leading-edge technology in all fields; much of which has traditionally been outsourced. BMW Oracle Racing has, however, taken a very different approach by doing as much of the research as possible in-house. One area which has remained almost entirely in-house is Computational Fluid Dynamics (CFD).

With viscous, free-surface and lifting surface effects all playing critical roles in sailing yacht design, CFD has always pushed existing technologies to the limit. The use of CFD in the design of sailing yachts is complicated by the existence of a large number of requirements placed upon the design.

For example, to win the America’s Cup a yacht is required to sail efficiently at all speeds up to 18 knots and at all heel angles up to 35 degrees. The yachts also must sail into the wind, requiring lift-to-drag ratios in excess of 5.0 for the hull, yet be able to sail away from the wind with near zero lift requirements.

Photographer: Gilles Martin-Raget

Photographer: Gilles Martin-Raget

To win, all of these conditions must be met more efficiently than their opponents. To analyze all these conditions for all possible design variations using the most complex analysis procedures available within the time frame of an IACC design cycle is a difficult engineering feat.

Covering the design space with the greatest efficiency requires investment of design resources into different CFD methods. In order to accomplish this BMW Oracle Racing has invested heavily in two CFD methods: finite volume prediction methods and boundary element methods. Calculation speed and physical detail variations in the methods mean that comparing the results of both bring greater insight than one alone.

Taking the Right Track to Better Hull Design

Jonathan Binns, PhD, a CFD and experimental operator with the BMW Oracle design team, says that the design team is currently performing CFD analyses using three finite volume codes and six boundary element codes. Each code carries its own advantages in terms of throughput, physical simulation detail and pre- and post-processing requirements. However, ultimately all of the results must be related to each other and then to the design of the yacht.

“This is where Tecplot has become important to our research effort in providing a universal format with a high degree of programmability,” says Jonathan. “Tecplot has provided the visualization link required for many situations.”

The graphics of refined grids in Figures 1 to 3 show the results of refining the grid for a boundary element code. The top image shows the dynamic pressure distribution for the older grid, the bottom shows the dynamic pressure for a refined grid. “These images provide information to the CFD researchers concerning the improvements made and visual proof to the designers that the results will provide a better answer,” says Jonathan.

The resulting plot shows the predicted variation in dynamic pressure coefficient for an IACC yacht sailing to windward. From the results Jonathan and the BMW Oracle design team learned that the refined grid and the newer program were producing a much smoother, more realistic pressure distribution. “Spikes in the pressure distribution have been dramatically reduced and so the results can be used with much greater confidence for subtle design geometry changes in these regions,” says Jonathan.

The graphics of the finite volume results (right image of Figure 4) against the boundary element results (left image of Figure 4) provide much greater understanding of the effects of the differences between the two CFD methods. “Armed with this information our CFD operators can better provide our designers with the information they need in the time frame they need it.”

The plot was created using data generated by a boundary element time domain flow solver and an unsteady finite volume solver. To create the plot, Jonathan exported the final ten steps of a converged solution from the flow solver to an ASCII data file, and then imported those steps into Tecplot software using a customized add-on he wrote to average the final ten steps. The layouts were then linked to ensure that they could be examined from an identical angle and location.

Besides pressure distributions the design team also uses Tecplot 360 to visualize shear stress distributions for CFD data and for plotting large sets of XY data from experiments. “Tecplot also gives us the ability to easily plot and print multiple data sets of 4 million data points with overlays and graphics,” adds Jonathan.

The results of these visualizations are used to explore the largest possible design space and to gain a greater insight into flow topology. “In this way,” says Jonathan, “small incremental advancements can be made to produce greater lift for lower drag whilst sailing. Experimental data is used hand-in-hand with CFD results both as validation and as design insight.”

Visualizing their data with Tecplot not only provides the design team with a way to view physical representations of their CFD analyses, but also provides valuable design information on the effects of design variations.

Jonathan believes that Tecplot’s three greatest strengths are its Macro and ADK programming; its ability to handle very large data sets; and its common data format. Without Tecplot, Jonathan says that customized comparisons would not be possible, which might leave his dream of being a part of an America’s Cup winning yacht design team high and dry.

Contributed by Bruce Rosen
Bay Simulations, Inc.

The America’s Cup is the world’s oldest sporting trophy. Big, technology-laden racing yachts representing yacht clubs from countries around the world compete for the right to hold and defend this ornate piece of silver.

Since 1992, the “IACC rule” has governed the size and other general characteristics of the roughly 70-foot yachts. Now a proposed 90-foot rule would give the naval architects more opportunities to explore a new design space, developing hull and appendage shapes intended to give their teams an edge over the competition.

“Gridgen and Tecplot give us the confidence that we will be able to effectively communicate and collaborate with America’s Cup teams and yacht designers all over the world.” – Bruce Rosen, South Bay Simulations, Inc.

With recent syndicate budgets running in excess of $100 million, teams invest significant effort in testing their hydrodynamic designs. Every angle is investigated with a wide range of tools from simple empirical relationships to very expensive scale model towing tank tests.

Designers also use an arsenal of advanced software to perform extremely sophisticated numerical calculations aimed at solving the fully 3D and viscous Navier-Stokes equations that govern the fluid dynamic flow fields under the water and in the air.
SPLASH Software Helps Designers Understand Flow Dynamics

One popular tool for predicting yacht performance is the SPLASH free-surface flow software, developed by Bruce Rosen, president of South Bay Simulations, Inc. SPLASH is a panel code, a computerized method for solving the 3D inviscid fluid flow equations governing the underwater portion of the flow.

SPLASH was originally developed to support skipper Dennis Conner’s successful 1987 campaign to win back the America’s Cup from Australia, who had just ended the New York Yacht Club’s 150-year monopoly of the trophy. Since then, SPLASH has undergone continuous refinement, and has been used by about one-half of the many competing syndicates.

In SPLASH, fluid source and/or doublet singularities of unknown strength are distributed across each panel of the model. Boundary conditions are applied at panel control points, yielding a set of linear equations to solve for the strengths of the sources and doublets. This makes it possible to compute the corresponding flow field and performance-determining drag, side forces, and rolling and yawing moments that act on the yacht as it moves through the water.

Panel codes are often not considered to be true Computational Fluid Dynamics (CFD), due in large part to the greatly simplifying inviscid flow assumption. But, for a yacht moving through the water, the additional physical and numerical aspects due to the free-surface waves at the air/water interface can be just as complicated, and difficult to capture, as other types of phenomena normally pursued using viscous CFD codes.

 Unraveling the Complexities of Yacht Design

As seen in Figures 1 and 2, the wave shape, the waterline contour, and the extent of the wetted portion of the hull are not known in advance of the flow simulations and must be determined as part of the flow calculations. This requires special hydrostatic and hydrodynamic boundary conditions on the free-surface panels.

Figure 1: Hull bottom: The distribution of grid angles along the waterline is highly unusual and difficult to generate using standard methods. Other enhanced capabilities include adjusting the usual first spacing at the boundary, as the mesh is being generated, to maintain a desired first spacing normal to the boundary.

These boundary conditions are nonlinear, and they incorporate additional radiation conditions to ensure that waves propagate only downstream and not upstream. As a result, the free-surface panel boundary conditions are considerably more complicated than the standard ones used on solid surface panels such as for the hull and appendages.

The dynamic sink and trim (pitch angle) of the model “underway” are also not known in advance. A series of alternating SPLASH flow calculations and model re-flotations and re-panelizations are conducted, ultimately converging to final solutions. For yachts, the entire process can be automated within ACCPAN software, a companion product to SPLASH.

Figure 2: Model details allow for capturing free-surface waves at nonzero heel angles while including the effects of an immersed transom and the lifting surfaces (keel and rudder) and their trailing vortex sheet wakes.

ACCPAN constructs the SPLASH panel models using structured surface meshes distributed over the various yacht components, and over a portion of the free surface in the near vicinity of the yacht. ACCPAN automatically generates the panelizations based on designer-supplied yacht surface geometry databases, and other model and test condition inputs. This generation uses the elliptic grid algorithms available in Gridgen, CFD meshing software from Pointwise, Inc.

When setting up the simulations, the main objective is to distribute a sufficient number of panels over the yacht and free surface to adequately resolve all significant aspects of the flow. This is difficult to accomplish, yet quality of the panelization is critical.

“These concepts and approaches are very consistent with one another and, taken together, help to compensate for most if not all of the other accompanying complexities and difficulties.” – Bruce Rosen, South Bay Simulations, Inc.

The modeling of the lifting appendages and their trailing vortex sheets adds significantly to the overall complexity. It is difficult to achieve adequate panel density on the hull and free surface adjacent to the appendages. Even more important, at high heel angles, the overall quality of the hull and free-surface panelizations suffer greatly due to the attachment of the keel and rudder-trailing vortex wakes to the hull centerline, regardless of heel angle. An attendant bunching up of panels occurs to windward, and a stretching out of panels occurs to leeward.

Increasing the panel density alone does not always yield higher quality results. Computer memory and CPU execution time increase exponentially with the number of panels. While a single SPLASH panel model flow solution requires only about one CPU-minute to complete, an average of about eight such solutions must be repeated to capture all the nonlinear effects, and this must be carried out at each of 100 to 200 distinct test points.

To add to the complexity, designers’ tools must contend with highly asymmetric mesh boundaries, a waterline contour, and wetted surfaces that feature regions of very high curvature. These result somewhat randomly from the intersection between the yacht geometry and the yacht’s own steady wave system.

Gridgen Helps Designers Converge on Better Designs

Gridgen’s elliptic meshing algorithms provide the high-quality panelizations required to make it all work. ACCPAN also incorporates methods to externally specify parameters used within Gridgen to control elliptic mesh generation. “Millions of distinct ACCPAN panelizations and SPLASH flow calculations have been produced to date, in support of numerous America’s Cup design programs,” says Rosen.

The final SPLASH flow solution for this panel model (Figure 3) shows the distribution of pressure over the yacht surfaces and on the free surface where the wave elevation is proportional to the pressure coefficient.

Figure 3: View of hull bottom. The final SPLASH flow solution for this panel model shows the distribution of pressure over the yacht surfaces and on the free surface (where the wave elevation is proportional to the pressure coefficient).

Rosen feels there are certain advantages in terms of reliability and accuracy using a low-order panel code, with structured meshes, strictly planar wake models, and panel pressure integration for forces and moments. “These concepts and approaches are very consistent with one another and, taken together, help to compensate for most if not all of the other accompanying complexities and difficulties,” says Rosen.

Results of numerical flow simulation test results are often evaluated by comparison with experimental towing tank data for a subscale model (Figures 4 and 5). Hydrodynamic characteristics of interest include the drag at non-lifting conditions (i.e., with yaw, tab and rudder set to zero), and the effective span (a single measure of side-force generating efficiency). These are functions of both the yacht’s forward speed and its heel angle. The figures illustrate typical absolute and relative (boat-to-boat) agreement between SPLASH calculations and experiment.

Figure 4: Drag predictions at non-lifting conditions vs. tank data.

There are many other tasks where use of Gridgen is critical, says Rosen. These include:

• Interrogation of designer-supplied yacht surface geometry databases
• Modification and repair of database surfaces; visual inspection and troubleshooting of ACCPAN-generated models
• Research and development of new panelization strategies
• Customized panel model generation for non-standard yacht geometries or for non-yachting applications

Gridgen software also offers a complete suite of volume meshing tools for structured, unstructured, and hybrid meshes so it can be used when simulation needs call for other types of CFD simulation, such as full Navier-Stokes.

America’s Cup teams typically undertake numerical model tests for a very large number of similar but slightly different yacht designs. In these tests, the Gridgen scripting language, Glyph, can be used to automate as much of the database preprocessing as possible.

Tecplot Helps Designer Communicate with Racing Teams

Performing flow calculations is critical to the improvement of racing yacht designs. However, using software such as SPLASH and Gridgen can only get designers so far. Once simulation runs are completed, the designers must be able to view the results and communicate the significance thereof to others.

To view the results of their flow simulations, the analysts at South Bay Simulations use Tecplot scientific visualization software from Tecplot, Inc. “We work on Linux with command-line codes and shell scripts, which aren’t very graphical,” says Rosen, adding that his company uses Tecplot for all their presentation needs (including all the figures shown here).

Summing it all up, Rosen says that “Gridgen and Tecplot give us the confidence that we will be able to effectively communicate and collaborate with America’s Cup teams and yacht designers all over the world.”

February 2011

Tecplot software provides critical CFD visualization component in cutting-edge technology

At Toyota Motorsport in Cologne, Germany, engineers use wind tunnels with sophisticated built-in PIV systems to help design faster high-performance cars.

The blink of an eye may mean nothing to most people, but in the high-pressure, high-stakes world of auto racing, it can be the difference between winning and losing. Glance at the results of almost any Formula One race and you’ll see that the top three cars usually finish within a few tenths of a second of each other — barring major mechanical failure or driver error, of course. Getting the best performance possible out of these complex speed machines requires intense scrutiny and religious maintenance by experienced engineering teams who are relentlessly focused on finding ways to improve air flow, increase downforce, and reduce drag.

Motorsports engineers like Frank Michaux, a CFD/PIV researcher at Toyota Motorsport in Cologne, Germany, use a number of advanced analysis techniques to help them get peak performance out of all kinds of vehicles — from NASCAR, Sports Car, Formula One (F1) and other races to finely built model lines for highway use. Michaux’s team maintains and operates two state-of-the-art automotive wind tunnels, each equipped with a built-in particle image velocimetry (PIV) system.

During the 2009 Formula One season, engineers wanted to study the wake behind the front wheels of a vehicle – a critical part of the flow that can affect the performance of the entire vehicle.

To be useful, the complex data derived from these state-of-the art systems must be quickly and accurately interpreted and displayed. Toyota Motorsport sustains its winning edge by correlating its PIV results with computational fluid dynamics (CFD) simulations, both accurately measured and rendered by Tecplot software.

During the 2009 Formula One season, engineers wanted to study the wake behind the front wheels of a vehicle – a critical part of the flow that can affect the performance of the entire vehicle.

During a particle image velocimetry (PIV) test, Toyota Motorsport engineers position a camera at a 90-degree angle to the plane of the flow field. The wind tunnel is filled with the fog, and the part that needs to be visualized is illuminated with a high-powered laser, creating a 2D plane.

Inside the Wind Tunnels

Toyota Motorsport is a renowned center of excellence for automotive design and development. A service-focused company, it offers use of its wind tunnels to outside engineers and researchers as well as to those within the Toyota organization. Both wind tunnels are state-of-the-art facilities built less than ten years ago. One is equipped for full-scale testing, with a rolling road on which vehicles can reach simulated speeds of 70 meters per second. The second is used for model testing at a scale of approximately 60 percent. Both feature permanent PIV systems, a rare and valuable research tool for design engineers.

PIV is an optical method of flow visualization that has been used for a couple of decades to obtain instantaneous velocity measurements and related properties in fluids. But it wasn’t until very recently that PIV became a practical design tool for engineers, largely due to the increasing power and decreasing cost of both computers and digital cameras.

Development teams from many of the world’s largest and best-known auto makers travel to Cologne to take advantage of the advanced technology and expertise offered by the Toyota wind tunnel team.

“Many of today’s motorsports cars are based on existing, commercially-available cars,” said Frank Michaux. “If researchers can identify a way to reduce drag on a motorsports car, it’s reasonable to assume that this information also may apply to future versions of a normal road car.”

Optimization in auto racing is like everything else. It is a continuous process and, according to Michaux, most engineers sport a bit of a perfectionist streak. As a result, he says, they are never totally satisfied with what they’ve accomplished. This is a good thing, though, because each part or mechanical adjustment in a car—no matter how tiny—affects the flow, force or drag of the entire machine. Without that perfectionism, the small details could be easily overlooked. Since optimization is continuous and evolutionary, the speed with which engineers can compile and analyze data, and then apply it to a current project, becomes vital. A delay of even a few days can mean the difference between successful integration and failure.

“At Toyota Motorsport, we need to deliver data quickly. If you see that you are not capturing the flow correctly, then you need to adjust your CFD methodology until you get it right,” Michaux continued. “The sooner and faster you can do that, the better.”

Using Non-Intrusive Methods to Visualize and Quantify Air Flow

Before the advent of PIV, engineers would create simple vehicle models and then study the flow on the surface of the model. This was inefficient because they could only see the flow at the surface. They didn’t have the means to produce a 2D plane, for example, or accurately observe and record the critical wake of the wheels.

To compensate for these shortcomings, engineers employed several ad hoc techniques, like putting a smoke probe in the wind tunnel in order to better “see” the air flow. But in an industry where seconds count, these methods didn’t accomplish what engineers really wanted: to fully capture accurate flow data for the area of interest, and to do so very quickly.

“By looking at the smoke, we could visualize the flow, but we were unable to quantify it. You could only make assumptions about speeds based on the visual aspects of the flow,” explained Michaux. “The whole problem in this industry is how to visualize air flow without introducing something new into that flow that could potentially compromise the results. With the PIV method, you can really attach numbers to air flow.”

PIV allows for the visualization of the flow field almost exactly as it appears in the wind tunnel, without influencing the very flow field that engineers are seeking to measure. Toyota’s PIV system involves filling the tunnel with a fog or mist with essentially the same density as air. When the air flows through the tunnel, the small particles that make up the fog simply float, making this PIV method as non-intrusive as current technology will allow.

For a PIV test, engineers position a camera at a 90-degree angle to the plane of the flow field they want to study. The tunnel is then filled with the fog, and all tunnel lights are turned off. Next, engineers illuminate the part that needs to be visualized with a high-powered laser, creating a 2D plane. Simultaneously, a series of two-set photos are taken at extremely rapid intervals—generally 10 to 20 microseconds. Equipped with this sort of ultra-slow-motion digital imaging, engineers can easily measure the rate and direction of the flow.

A laser is used to illuminate the flow field around the car’s front wheel to take particle image velocimetry (PIV) measurements at Toyota Motorsport.

A laser is used to illuminate the flow field around the car’s front wheel to take particle image velocimetry (PIV) measurements at Toyota Motorsport.

Integrating PIV with Tecplot for Complete Visualization and Understanding

During the 2009 Formula One season, engineers wanted to study the wake behind the front wheels of a vehicle. This is a critical part of the flow; if it isn’t perfectly calibrated, the performance of the entire vehicle could be severely compromised. Upon being presented with the problem, the Toyota Motorsport team realized they needed to look at options for adding or modifying various front-end parts, for example, adding an under-nose turning vane or modifying the front wing to create or influence an outwash. The goal is to push the flow out from under the nose of the car, forcing the wake of the car as far “outboard” as possible.

After gathering the raw data from the PIV measurements of the “separation point” on the front tires, engineers post-processed the data using Tecplot software, which allowed them to see and measure the exact position of separation. Each of Toyota’s PIV measurements consist of 300 datasets, with each dataset containing two images taken 10–20 microseconds apart. The end result of the PIV process is a complete 2D field of vectors. Engineers subsequently plotted the velocity magnitude, or vorticity, with vectors based on the average of all 300 datasets. The corresponding CFD result was then also imported into Tecplot software. Engineers compared the PIV and CFD data sets to determine whether their CFD methods were within tolerances. Whenever necessary, the engineers tweaked the CFD process to get it closer to the wind tunnel results.

In the case of the separation point on the front tires, initial tests showed that the separation point was late and too far back from the tires. The engineers altered the CFD methodology based on these observations, imported the new results into their Tecplot software, and compared it with the PIV results to evaluate their progress. The process was repeated until they arrived at a design that placed the separation point at an optimal position on the tires. This method of CFD / PIV analysis helps engineers derive simulations of real world conditions that are as accurate as possible in a surprisingly short timeframe.

Toyota Motorsport engineers compare the PIV results with their CFD simulations to make sure that the wind tunnel results and CFD methodology correlate.

“The choice of Tecplot software for our automotive testing was fairly obvious,” observed Michaux. “Almost everyone in the automotive research industry uses it, making it easy to share information with other researchers or engineers. And if you need to use PIV software from a different provider, the output can still be read in Tecplot software. It’s simple to switch back and forth without completely changing your post-processing chain.”

With speed and accuracy the name of the game, Toyota Motorsport, along with Tecplot software, can provide a number of diverse automotive clients the data gathering, interpretation and visualization they need to lap the competition.

“We have the ability to accommodate a wide range of CFD and PIV requirements, and have a team of knowledgeable engineers available to assist with acquiring and visualizing the results,” said Michaux. “Tecplot software plays a significant role in helping us achieve that level of automotive science and service.”

After comparing the PIV results with the CFD results, the CFD methodology is optimized and is then used to run further CFD simulations.

 

Tecplot Chorus helps Swift Engineering Inc. change mushroom-shaped wakes on Formula race cars to make passing easier and races more exciting

The race does not always go to the swift, especially when it’s Formula racing, where the wake of a race car can be turbulent enough to prevent others from passing it, no matter how fast they’re going.

“That can make it pretty boring to watch, as well,” said John Winkler, Ph.D. and Chief Aerodynamicist at Swift Engineering Inc. “Less passing means fewer contests for the lead. Cars line up behind one other and just go around the track until they near the end. Many believe this has taken away from the overall excitement of the sport.”

“One idea was to increase the number of times the lead can change by making it easier for cars to pass each other. The answer to that was to do a little mushroom-busting.” — John Winkler, Ph.D., Swift Engineering

Now it’s Swift that’s going to the race, so to speak, with some innovative new designs to help make it more exciting to watch. This is not the first time the company has delivered innovation. Swift Engineering’s design and composite manufacturing innovations have yielded championship winning race cars for nearly thirty years. From its early Formula Ford design in 1983 to the present Formula Nippon carbon fiber race car, which is capable of speeds approaching 200 MPH and 4 G cornering, Swift is known for innovation and quality.

Busting a few mushrooms

“It began when we started talking about what we could contribute, as designers, that could make racing more interesting to watch,” said Winkler. “One idea was to increase the number of times the lead can change by making it easier for cars to pass each other. The answer to that was to do a little mushroom-busting.”

Winkler is referring to the shape of the wake left behind by the rear wing of the race car, which looks like a mushroom when visualized and viewed from the rear. By reshaping the wake and moving it up and away from the racetrack surface, Winkler and his team are trying to remove or reduce this very turbulent obstacle to cars attempting to pass.

They are using Tecplot Chorus to help. Below is a screen shot of Tecplot Chorus.

Process overview

Working with Swift’s Chief Scientist Mark Page and Aerodynamicist Andy Luo, Winkler began by developing a simplified perturbation study on the rear wing of the Swift 017.n to get a better understanding of how changes in the wing’s position and rotation affect aerodynamic performance as well as the shape of the wake downstream.

“Although we were examining ways of changing the shape of the wake downstream, it was equally important to make sure we did not sacrifice performance,” said Winkler. “Increasing the enjoyment and excitement of watching a race is one thing, but it would defeat the purpose if the car is slower, can’t corner as well, or is generally less responsive.”

They then used the qualitative flow visualization results to explain trends in the quantitative aerodynamic parameters such as downforce and drag. Finally, they used the knowledge gained to tailor a design that could promote passing without sacrificing performance.

Grid and solver

Winkler and Luo created a hybrid mesh with a completely structured far-field and an unstructured grid around the car. Using the Metacomp CFD++ flow solver, with the car running at 150 miles per hour in a standard atmosphere at 59 degrees F, the initial run took almost nine hours on an in-house HPC cluster. The team then interpolated the original run as the initial condition for the subsequent runs, cutting the computing time to about 3.5 hours per run.

The results were imported into Tecplot Chorus for further analysis, where Luo and Winkler created surrogate plots of the overall vehicle downforce and rear wing downforce. The plots revealed some peculiar and unexpected results, especially in the cases with rotation.

“With only small changes in the wing’s orientation and position, we saw significant changes in the performance results,” said Luo. “That was one thing we were scratching our heads over. If the wings were moving only slightly, how was there such a big difference in the numbers?”

Luo and Winkler then used Tecplot Chorus to look at and compare surface pressures on top of and beneath the wing.

“As we went into the qualitative assessment of different runs, the downforce numbers were telling us that something was different, something drastic was happening,” said Luo. “But we didn’t know if it was valid. It was only by stepping through these comparisons that we could confirm what we saw was real, not an error.”

The final step was to generate a series of symmetry plane cuts showing a cross section of the wing and the flow around it. This helped explain the physics and confirmed that the downforce numbers were correct. One of the most interesting findings the process yielded was that optimal performance also generated the highest, least obstructive wake.

The current study was only a preliminary investigation to help the engineers better understand the sensitivity of the rear downforce and wake shape in relation to wing position and incidence. It involved a small number of perturbations and a small number of design variables. The next steps will be to expand the study to include more design variables, perform a DOE to isolate design variables with high sensitivities, and drive CFD++ through an outer optimization routine.

Tecplot Chorus delivers speed, efficiency, organization

Winkler and Luo both agree that Tecplot Chorus facilitated the comparison of multiple visualized flow fields in an efficient manner. The results from these comparisons helped to explain the peculiar downforce and drag trends, while the cut-planes in the wake downstream highlighted the wake movement as a function of the perturbed parameters.

“From a designer’s standpoint, it is important to understand why a certain concept or design is better than another,” said Winkler. “Inspecting the flowfield is a key component of the aerodynamic design process, because it allows the designer to not only validate the solution but also to gain a better understanding as to how certain design variables affect the flowfield.”

While this type of comparison may have been done in another visualization tool using scripts, Tecplot Chorus made it easy to organize, catalog, and visualize all of these solutions.

“One of the best things about Tecplot Chorus is that it lets me spend more time working the problem and less time troubleshooting and writing scripts,” added Luo. “What once took me four or five hours, now takes me only five or ten minutes. And I would much rather invest my time in solving engineering problems for my customers than waste it on tedious tasks like writing and managing scripts.”

Tecplot 360 EX helps Stanford University students optimize design of solar-powered car to compete in the Australian Outback’s World Solar Challenge

The sun shines down upon the immense Australian Outback like nowhere else, drenching the land in solar energy that begs to be tapped. Where better to stage the World Solar Challenge, a design competition established to find the world’s most efficient solar-electric car? Every two years, some of the brightest student minds from leading universities around the world, and even some corporate competitors, gather in Australia’s Outback with solar-powered cars they’ve designed and built from the ground up to vie for the champion’s title. This brutally competitive five-day, cross-continental solar race begins in the Northern Territory city of Darwin and runs more than 3,000 kilometers (1,864 miles) south to Adelaide in South Australia.

“We’re operating with very tight design constraints,” says Guillermo Gomez, team lead for the Stanford Solar Car Project. “We can use no more than six square meters of solar panels, for instance. The solar car has to be a fully-functional vehicle, too, with a suspension, brakes, room for a driver, and so forth. The only two ways we can achieve any real design advantage is through a good power management strategy and an aggressively-shaped aerodynamic design.”

Stanford Solar Car Project finished fourth in the world at the 2013 World Solar Challenge with this solar-powered car. The car has a carbon-fiber body and is powered by energy from the sun that’s collected by six square meters of solar panels and stored in a battery pack as electricity.

CFD Analysis Helps Students Effect Significant Changes in Design

Tecplot 360 EX image showing pressure and airflow around the leading edge of the car’s main airfoil and right fairing.

One key design challenge was optimizing the vehicle’s body for the wide and constant range of cross winds in the Australian Outback. The Stanford team not only had to optimize the body for airflow moving over the vehicle as it moved forward, but also needed to consider how winds might push the side of the body and the wheels from a select range of angles.

Optimizing for frontal drag at 25 meters per second and a range of cross winds, the Stanford team ran their CFD analyses using SU2, the open-source CFD solver code developed in the Aerospace Design Lab at Stanford’s Department of Aeronautics and Astronautics, meshed the car bodies with Pointwise software, and visualized results with Tecplot 360 EX with SZL Technology.

“These three tools integrate so well together that it was easy to set up, run, export, and visualize the data, which was 10 to 15 million cells in size,” says Gomez. “The resulting visuals have been really impressive and very appealing. Of greatest importance, though, the CFD analysis has helped us make significant changes to the body design that should help us compete more effectively.”

There’s still quite a bit of work and testing to be done, and the final design will not be unveiled until the summer of 2015, but Gomez promises that the end result will be a beautiful, aerodynamically-optimized solar-powered car that should give their competitors a run for their money in the Outback.

Close-up showing pressure and airflow around the main airfoil’s leading edge.

Pressure and airflow around the trailing edge of the car’s main body airfoil and the rear of the fairing.

Future Applications

The primary objective of the Stanford Solar Car Project is to provide students with valuable hands-on engineering, technical, and business experience while also serving to help raise community awareness of the power of solar-electric energy. But the technologies being developed today also promise to lead to more efficient electric cars in the future, and if history can tell us anything about the future, it may lead to many uses not yet imagined.

Stanford Solar Car Project  |  Tecplot 360  |  Pointwise  | Standford SU2

Tecplot Chorus Integrates With KARALIT CFD to Save Time With Mesh-free CFD Simulations and Analysis

The Immersed Boundary (IB) methodology, a mesh-free technique for running CFD simulations, has been around since the 1970s. Instead of running Navier-Stokes equations on grids that conform to the shape of an object, as is done with meshing, the IB technique immerses a CAD model of the object into a Cartesian grid and then integrates the equations into that Cartesian grid.

The promise of IB was that it could eliminate the time spent on the set-up and generation of a mesh. It also did not require morphing, which can deteriorate the mesh, or re-meshing the computational domain when the geometry is changed. This technique was hampered, however, by the lack of an effective numerical theory for generating the cells that intersect with the object’s surface and was discarded in favor of mesh generation techniques.

Engine cooling case evaluated using the Immersed Boundary method with KARALIT CFD and Tecplot Chorus. Formula race car provided courtesy of Tatuus Racing.

As geometries became more complex and the size of the data started to increase exponentially, so did the time it took to generate the meshes, and interest in Immersed Boundary methodologies was renewed.

Among the first to commercially and successfully harness the power of this time-saving methodology is KARALIT, a start-up company headquartered in Pula, Italy. KARALIT provides a CFD software that is based on the IB technique. It works seamlessly with Tecplot Chorus, a software tool that integrates CFD post-processing, field and metadata management, and analytics tools into a single environment.

Durrell Rittenberg, Tecplot Inc.’s vice president of product management, and Sohail Alizadeh, director of UK and Ireland sales and operations at KARALIT, recently demonstrated the speed at which the design of a Formula Renault Tatuus 2000 race car (provided courtesy of Tatuus Racing) could be assessed using both tools.

In this demonstration Rittenberg and Alizadeh took a little more than an hour to evaluate the performance of the car’s rear wing by changing the geometry and dropping the wing by 10 centimeters, opening the engine bay ventilation flow and external dynamics of the car to observe cooling of the engine, and examining the overall exterior aerodynamics.

“We were able to show that, by combining this mesh-free IB methodology with Tecplot Chorus, you can reduce the time spent running CFD simulations and analysis by an order of magnitude,” says Rittenberg. “In other words, we demonstrated that you can use these two tools to run the CFD, look at and analyze many different solutions, and confidently make design decisions in a few hours instead of weeks.”

KARALIT CFD and the Immersed Boundary Method

With the Immersed Boundary (IB) method, the object’s geometry is immersed directly into a Cartesian mesh.

Alizadeh explains the process for setting up the geometry and running the solution.

“You basically have a complex geometry and you need to run a simulation of it on a computational grid. In KARALIT the grid is a non-isotropic Cartesian unstructured grid,” he says. “The grid needs to sense the complex geometry as it passes through and then the CFD simulation needs to know the boundary conditions, which is where the geometry intersects the grid.”

The grid is categorized into three parts, which don’t appear in traditional CFD:

1. Fluid flow
2. Boundary cells
3. Ghost cells

Ghost cells are reflected as mirror points in the fluid and, during the calculation, the field values are interpolated to the mirror cells. The ghost cells solved during calculation can be set at the correct value so that the resulting boundary condition is forced on to the calculation.

“What that means is that—implicitly and automatically in the background—the entire process of absorbing the geometry and boundary conditions takes place during the simulation and calculation,” says Alizadeh. “It results in a very simple workflow: the CAD geometry is imported into the code (SDL in this case), you set up a few parameters, and run the solution. KARALIT outputs to PLT files, so that it can be easily imported by Tecplot software products.”

KARALIT is set up for parallelization, which means it can do hundreds of nodes, Alizedah notes, adding that the size of the cases that can be run are limited only by the size and memory of the computer.

Analyzing Results Using Tecplot Chorus

Once the simulation was run, the results and the complex geometries could be imported into Tecplot Chorus for fast, accurate analysis and comparison across multiple design candidates. As mentioned earlier, Tecplot Chorus integrates a powerful set of CFD post-processing, field and metadata management, and analytics tools into a single environment, reducing the time it takes to analyze multiple designs in most cases from several weeks to a few hours.

In this demonstration, Rittenberg and Alizadeh were able to quickly determine that that the air pressure on the rear wing of the race car was less in the lowered position than in the original position.

Tecplot 360 shows a side by side comparison in of the fluid dynamics, including pressure fields and streamtraces, of the race car’s base wing in the lower position (left) and upper positon (right).

“The real win here is that we can quickly look at multiple design candidates, evaluate the results, and identify the optimal design,” says Rittenberg. “We are able to skip the busy work. There was no meshing, no macros to write, no loops, and we didn’t have to fuss with the data, among other things. We just looked at the physics and made the design decisions.”

Combining the CFD / IB methodology with Tecplot Chorus works most effectively with complex geometries that have a lot of intricate surfaces, says Rittenberg. Building a project out of even a simple number of solutions can be very convenient when you want to interrogate different views. Tecplot Chorus allows users to do this without generating too many macros and loops, he says, and it allows users to interactively evaluate many key integrated quantities.

This Tecplot Chorus project shows pressures on the rear wing for two cases where the wing geometry has been 1) left in its original position and 2) lowered. This analysis shows that the air pressure on the rear wing in the lowered position is slightly less than the air pressure on the rear wing in the original position.

“You can do so many things with Tecplot Chorus,” Rittenberg adds. “Synchronize styles across multiple solutions, quickly interrogate a set of solutions, look for deltas, and identify what may be driving the physics. You can look at a hundred different solutions at a glance. That, combined with the mesh-free Immersed Boundary method, brings us back to our finding: that you can reduce the time spend running CFD simulations and analysis by an order of magnitude. And that’s a big number.”

Webinar and Video Demonstrations

• For more in-depth information on how users can tap the time-saving of mesh-free CFD simulations with KARALIT and Tecplot Chorus, check out this recorded webinar (YouTube).
• To find out how to  use the Immersed Boundary method and  CFD simulation results with Tecplot Chorus, check out this video demonstration (YouTube).
• To see how  results are evaluated in Tecplot 360, watch this video demonstration (YouTube).
• These videos are available in .mp4 format on our Webinars page.

KARALIT | Tecplot Chorus | Tecplot 360 | Tatuus Racing