## Introduction

Aerodynamics is a very complex field. Air is invisible and to see what is going on, you need to somehow add something to the flow (like smoke) or the surface of the model (tufts) or measure a derived quantity (like pressure) at different points to re-construct what is happening in terms of velocity from your experience with similar cases.

While this has worked well in the past and fantastic things have been achieved by talented engineers with years of experience, today’s race for performance, efficiency and energy saving can only be won if detailed and precise 3D datasets are available. To obtain this, there are basically two, sometimes competing approaches available: CFD (computational fluid dynamics) and wind tunnel flow measurements. Each in his own way, both approaches are similarly complex.

CFD requires to solve the complex Navier-Stokes equations that describe the flow around an object. Which in theory is fairly straight-forward but solving them for larger setups quickly becomes extremely challenging. The discretization (meshing) of the model geometry (often including ventilation openings, rotating wheels and small details like bolts and wires), the large number of computational cells typically used (50 Mio cells and more) and the effect of model surface structure that determines the point of turbulent transition and flow detachment need to be tackled, to name just a few. This results in simulations that must be run overnight on high-performance clusters, produce datasets in the order of several tens Gb per configuration and require experienced engineers to setup, run and analyze.

The picture is the same if you turn to highly advanced optical flow measurement techniques like PIV (particle image velocimetry) or traversing systems capable of mapping the flow around the model in a wind tunnel. It helps that wind tunnel tests are anyways required for validation or WLTP emissions calculations. And you can be reasonably confident that the measurement data represents reality fairly well, at least for the setup in the wind tunnel. However, an expensive setup including powerful lasers and cameras, or traversing equipment is needed. Plus, again, a set of well-trained engineers that can handle the calibration of the optics or the programming of the traversing system. Not to mention the finer points like laser safety, seeding, several hours of measurement time and days taken up by post-processing.

Bottom line for both approaches is: The achievable quality and accuracy of the data is impressive. But for an initial assessment of a design during the concept phase, a quick comparison of configurations and most optimization tasks they are both simply too expensive and time consuming. To accelerate aerodynamic development, we must provide a simplified, yet effective toolset.

## Speeding up insight

In terms of simulation, such options include cloud platforms like AirShaper that automates the process of CFD simulations. Instead of spending days on pre-processing, solving & post-processing, the process is streamlined to provide results often within hours. The compromise is that many of the small, expert software settings are hidden, in return for a more generalized / standardized approach & simulation output.

<YouTubeEmbed
 videoId="O_DNDxCtI_I"
 title="How to Runa Simulation on AirShaper"
/>

Along the same idea but for experimental wind tunnel work streamwise gmbh developed ProCap, a scan-and-paint approach that allows to analyze the flow in real-time while at the same time providing a fully 3D volumetric dataset. Rather than providing ultra-precise results in terms of spacial resolution and accuracy, this approach offers a much more flexible, time efficient way to obtain quantitative data in the form of a fully 3D flow field, perfect also to validate CFD results.

<YouTubeEmbed videoId="TFdnU6Gnz_I" title="ProCap in Wind Tunnels" />

## Case study

We want to introduce the two new concepts in some detail here but also clearly and openly illustrate the benefits and drawbacks of both techniques. To this end, both AirShaper simulations and ProCap measurements were applied to the Drivaer model, a reference car geometry widely used in the automotive industry. As both techniques provide full 3D data, the comparison of the results is straight forward using a CFD post-processor. The setup was a 1:4 scale model with fixed wheels at 25 m/s.

### simulation

In terms of simulation, one of the main choices to make is how to approach the boundary layer. This is the layer of air sticking to the surface and it features high velocity gradients normal to the wall. One way to take this into account is to use prism layers, which are very "flat" cells used to increase the resolution normal to the wall, to capture the velocity gradients and flow separation locations. But this is expensive in terms of mesh count and it is also more challenging to automate, especially on complex geometries. Therefore, wall functions are often used: these are "precalculated" velocity profiles that are then applied to the relatively large cells close to the boundary layer. It makes the whole process of meshing and solving a lot more robust, but can create slight differences in terms of boundary layer behaviour. The benefits of AirShaper, though, include the possibility to work with non-watertight models (no need to spend weeks closing small gaps & holes) and have results within hours of uploading the 3D model.

### measurement

The accuracy of a measurement is typically limited by how accurately you can re-create reality in the wind tunnel. Error sources includes the presence of walls, boundary layers on the floor, mounting points of the model, etc. When using a probe, the finite size of the probe will also affect the flow and especially close to a surface (below one probe tip diameter or 4 mm for ProCap), errors become significant. When working with ProCap, an additional source of error (quite similar to mesh resolution in CFD) must be taken into account. As the spatial resolution is a direct consequence of the number of measurement points per volume, a too short measurement time can result in poor resolution of velocity gradients. Typical measurement times for the volumetric data shown in this comparison are 30 minutes total, with only 30 minutes to factor in for setting up at the beginning. To do a comparison of different configurations one would focus on a single plane and thus be able to cut measurement further. Compared to using a traversing system, ProCap typically achieves a speed-up of 50 to 100.

<ImageWithCaption
 src="/assets/images/accelerating-erodynamic-design-picture-1.jpg"
 alt="Figure 1: Overview of the two datasets. AirShaper (CFD) left, ProCap (measurement) right."
 width="800"
 height="542"
 positionCenter={true}
/>

### results

Overall flow pattern (figure 1): the overall flow & pressure patterns provided by AirShaper and ProCAP measurements are very similar. They both provide an excellent overview and understanding of the flow topology and lead to the same conclusions in terms of wake size & structure, high & low pressure zones around the car etc. at a fraction of time and money. In some areas differences are visible as they would be in any comparison of numerical and experimental data, some of which are pointed out here (Figure 2).

<ImageWithCaption
 src="/assets/images/accelerating-erodynamic-design-picture-2.jpg"
 alt="Figure 2: Highlighting areas with differences between the two datasets."
 width="800"
 height="344"
 positionCenter={true}
/>

Boundary layer (Figure 2, A): In the simulation data, the thickness of the boundary layer is determined by the cell size in this area. For ProCap, the resolution close to the surface is limited by the physical dimensions of the probe (4 mm) which explains that the velocity does not reach zero at the wall.

Underfloor aerodynamics (Figure 2, B): the velocity of the air underneath the car is higher in the AirShaper simulations compared to the ProCAP results. One explanation for this is the impact of scaled model testing: the ProCAP results are based on physical wind tunnel results, where surface roughness on both car & ground can influence results. Also, depending on the upstream length of the wind tunnel test section, the boundary build-up can be quite significant (reducing the air speed below the car), as the floor is static. The AirShaper simulation also has a static floor (although a moving floor is also possible), but of limited length. Another argument is that the wall functions under-predict the boundary layer thickness, especially because a perfectly smooth surface is assumed. And in terms of ProCAP measurements, the relatively coarse spatial resolution of the capturing in that area (due to the confined space and difficulties in the optical tracking with cameras placed above the model) leads to interpolation effects, reducing local gradients and peak velocities.

Shear layers (Figure 2, C): the "thickness" of the transition layer between the wake behind the car (slower moving air being dragged along) and the free stream air (air flowing on top of the wake) is smaller in simulation compared to the ProCAP measurements. One reason for this is that the AIrShaper simulation uses a turbulence model to resolve velocity variations, and perhaps the applied model underpredicts the thickness of this shear layer. In terms of ProCAP measurements, the finite dimension of the probe and resolution of the grid creates smoothing effect, which can overpredict the shear layer thickness.

Wake regions (Figure 2, D): Differences of the wake behind the wheel illustrate the sensitivity to surface roughness and small details, both difficult to reproduce correctly in CFD geometry and the wind tunnel model.

## Which tool to use?

Both the AirShaper and ProCap offer an alternative to the high-end high-cost approach in aerodynamic development able to provide 80% of the information for 20% of the budget. Can they provide valuable data and help to understand and improve the design? Absolutely. Is there a tradeoff between speed and accuracy? To some extent, yes. But the goal is not to replace the high-end stuff but rather to offer an option where otherwise no data at all would be available.

Challenging in this comparison, but also in most real-life cases, is the lack of ground truth. As so often, there remains a grey zone in terms of which method accounts for what difference. Of course, when validating & developing such tools, it's important to benchmark results versus "well known & documented" cases (like this Tesla Model Y). This helps to build a "general understanding" of the limitations of each tool. But for specific, non-standardized cases, an amount of uncertainty around result accuracy will always remain.

While AirShaper strongly accelerates the digital phase, with a fully automated CFD workflow able to analyze & compare the full flow pattern of various design concepts in just hours, ProCap provides substantial advantages during physical wind tunnel testing, correctly capturing fine details & flow separation on the fly in the wind tunnel.

Which tool? Both please :)

Interesting links:

 

<ExternalLink href="https://www.streamwise.ch/">Streamwise</ExternalLink>

 

<Link href="https://app.airshaper.com/projects/new?accuracy=regular">
 Run Your Own Simulation
</Link>

Accelerating the aerodynamic design process

The <ExternalLink href="https://catesbytunnel.com/">Catesby Tunnel</ExternalLink> is a multi-million pound project that is set to become a new global benchmark for full scale vehicle testing. The project has been in development since 2013 and this week it will be open for business for the very first time. To find out more, we spoke to Jon Paton, Group Leader at <ExternalLink href="https://www.totalsimulation.co.uk/computational-fluid-dynamics/catesby-tunnel/">TotalSim</ExternalLink> who are leading the engineering behind the project.

---

## What is the Catesby Tunnel?

The Catesby Tunnel was originally constructed as a Victorian railway tunnel on the Great Central railway main line in the UK. It was built in 1897 and was in service for 68 years until it eventually closed in 1966. It is a perfectly straight 2.7km long tunnel, which took approximately 30 million bricks to build.

<ImageWithCaption
 src="/assets/images/roadcar-catesby-tunnel.jpeg"
 alt="A roadcar testing in the refurbished Catesby Tunnel. CREDIT: www.catesbytunnel.com"
 width="810"
 height="540"
 positionCenter={true}
/>

In 2013, the Aero Research Partners Ltd initiated the design phase to convert the disused railway tunnel into a state-of-the-art vehicle testing facility. A science and technology park will also be built adjacent to the tunnel, offering CFD, 3D scanning and other services. The Aero Research Partners Ltd is made up of TotalSim, a private investor and an automotive OEM.

---

## The Catesby Tunnel in numbers

The tunnel is 2.7km (1.7 miles) in length and 8.2m wide which is unusually wide for a tunnel in the Victorian era. In fact, it's width is more than double that of conventional tunnels at that time, pushing Victorian construction to the limit. This results in a cross-sectional area of 40m².

Another impressive statistic is the gradient of the track bed, which is a constant 1:176. So for each 176m travelled, the height only changes by 1m, which is particularly useful for repeatable testing. The Victorian engineers couldn’t have done much more to make it perfect for an aerodynamic test facility.

<ImageWithCaption
 src="/assets/images/1949-catesby-tunnel.jpeg"
 alt="The Catesby Tunnel in 1949. It's large width makes it ideal for an aerodynamic test facility. CREDIT: www.warwickshirerailways.com"
 width="810"
 height="565"
 positionCenter={true}
/>

---

## How does it compare to road testing?

Real world testing such as on a test track or proving ground is an important element in vehicle development programmes. From an aerodynamics point of view, it offers many advantages over testing in a wind tunnel:

- The engine and other components are running at true operating temperatures
- The asphalt and therefore tyre behaviour is more realistic as opposed to a partial or full moving floor in a wind tunnel
- There are no flow-disturbing connection elements holding the car in place, as would be the case in a wind tunnel
- There are minimal <Link href="/videos/how-wind-tunnels-work-blockage-factor-wall-effects-scaled-model-similarity-number-moving-floor/qqYiz1eo-F4">blockage effects</Link> which can affect the quality of the airflow

However, the downside of testing in the real world is the lack of repeatability due to changing weather conditions, which is particularly challenging in the UK:

- The magnitude and orientation of the wind can change, drastically impacting aerodynamic drag
- The temperature, humidity and density of the air can change, again impacting drag and also engine performance
- Real roads are often not perfectly flat and can feature bumps

The Catesby Tunnel, by contrast, suffers from very few of these problems. It sits below thousands of tons of earth, providing it with an enormous thermal inertia. This greatly reduces variations in temperature, density and humidity, which strongly increases the repeatability of vehicle tests.

<ImageWithCaption
 src="/assets/images/catesby-tunnel-road-resurfacing.png"
 alt="Relaying the road surface of the Catesby Tunnel. CREDIT: www.semlep.com"
 width="810"
 height="405"
 positionCenter={true}
/>

"The length of the tunnel also means that there are no wind effects so the flow distribution is much more uniform than in a conventional wind tunnel,' explains Jon Paton, Group Leader at TotalSim. 'This means there is no need to implement flow conditioning techniques such as <Link href="/videos/how-wind-tunnels-work-contraction-test-section-diffuser-fan-turning-vanes-and-settling-chamber/L1AYo9Mk1EI">meshes and grids</Link> to reduce perturbations in flow velocity.'

By improving the repeatability of tests, engineers are then able to isolate and compare smaller differences in measured forces, knowing that any changes are real and not just measurement uncertainties. This allows for more detailed design iterations where every marginal gain can add up to a substantial performance improvement.

---

## How does it compare to wind tunnel testing?

Wind tunnels are also designed to overcome some of the problems of real world testing. They aim to eliminate the same repeatability problems and add some additional benefits. Forces can be monitored via a <Link href="/videos/how-wind-tunnels-work-measurement-techniques-piv-pressure-taps-force-balance-pressure-probe/yA9FdqJdiKk">force and moment balance</Link> and many different set ups of the vehicle can be analysed within a short time frame, such as aero mapping.

- <Link href="/videos/how-aero-maps-can-improve-race-car-performance/u-GyTRblNmc">
 How aero maps can improve car performance
 </Link>

However, nothing comes for free, so there are also a few downsides to using wind tunnels. 'This is where the unusually large width of the Catesby Tunnel can add real value to a test programme,' highlights Paton. 'The cross-sectional area of 40m² is considerably larger than a typical car. Take the Tesla Model 3 for example, which has a frontal surface area of around 2.3m², this means the blockage ratio is less than 6%.'

'The large width of the Catesby Tunnel also helps to minimise wall interactions which would normally affect the quality of the airflow. The Tesla Model 3 is around 1.85m wide compared to the 6m wide roadway in the tunnel.'

<YouTubeEmbed
 videoId="_Z6_QYlbS8c"
 title="The transformation of Catesby Tunnel"
/>

Another benefit to testing at Catesby is that the vehicles are actually being driven. As a vehicle moves through air, it drags along a certain amount of air – an effect called “added mass”. This will be captured more accurately in the Catesby Tunnel then it would be in a conventional wind tunnel.

Although, wind tunnels are also used in the early stages of a vehicle's design life; testing models with no chassis components. Whereas at the Catesby Tunnel, cars have to be fully drivable and so have to have a functional drivetrain. To get round this issue, customers can use an electric driving platform and mount items that they want to test, such as 3D printed bodywork designs, to this platform.

---

## What types of tests can be done?

Alongside full scale aerodynamic performance testing, the Catesby Tunnel can also be used for vehicle performance evaluation. Low and high speed, acceleration and braking can all be tested along with the thermal performance of the powertrain, radiators and brakes. Aero-acoustics such as cabin noise, engine sound and exhaust systems can also be tested, removing the need for manufacturers to visit specific acoustic testing facilities.

<ImageWithCaption
 src="/assets/images/catesby-tunnel-cyclist.jpg"
 alt="The Catesby Tunnel can also be a useful facility for optimise the aerodynamics of athletes. CREDIT: www.catesbytunnel.com"
 width="810"
 height="387"
 positionCenter={true}
/>

- <Link href="/videos/designing-an-aero-bike/DqlDpUFcScc">
 How to design an aero bike to win the Ironman
 </Link>

Catesby can also be used for emissions testing, with Paton hinting that the tunnel may also support manufacturers towards WLTP certification. WLTP (World Harmonized Light Vehicles Test Procedure) is a global standard which determines fuel consumption, range and emission figures in a more representative way. The controlled environmental conditions offered by the Catesby Tunnel could, over time, also make it suitable to conduct the necessary WLTP tests.

---

## How will accurate measurements be achieved?

Similar to real world testing, measurements are currently limited to on-vehicle or in-vehicle test equipment. For force measurements, Paton and his team are currently developing force plates which can be integrated into the ground, measuring the force as the vehicle drives over them.

'We may also include a system that will measure pressure differences on the wall of the tunnel to capture the pressure reflections of a vehicle passing by,' says Paton. 'By modelling the same vehicle passing through CFD [Computational Fluid Dynamics] simulations, we can then determine a correlation factor between pressure measurements and vehicle drag. This will allow us to estimate the drag on the car.'

There is also a plan to install an ultra-wide band indoor GPS system to measure the position and speed of the vehicle inside the tunnel with a precision of 0.1 km/h. This will allow for more accurate coast-down testing to measure vehicle drag and rolling resistance.

- <Link href="/videos/drag-measurement-coastdown-testing/Gip3iG2A2zE">
 How to use coastdown testing to measure drag
 </Link>

---

## How will testing be kept confidential?

Confidentiality is always a huge consideration for OEMs and other users of facilities like this. Even at some of the most secure outdoor testing facilities, spyshots from unathorised photographers can be leaked, but this is impossible at Catesby.

"The tunnel is effectively a closed-off room that customers can hire and seal, so the Catesby team ensures full confidentiality," says Paton. "Vehicles are unloaded behind closed doors and then enter the tunnel directly. With 99% of measurements expected to be on-vehicle, we anticipate very few confidential setup and component changes to happen outside of the tunnel itself."

<ImageWithCaption
 src="/assets/images/catesby-tunnel-facility.jpeg"
 alt="A science and technology park will be built adjacent to the tunnel, offering CFD, 3D scanning and other services. CREDIT: www.catesbytunnel.com"
 width="810"
 height="458"
 positionCenter={true}
/>

Overall, the Catesby Tunnel is an all-in-one cost effective solution which allows customers to test many aspects at the same time and on the same vehicle. "We have developed the Catesby Tunnel to offer a number of unique advantages in terms of repeatability, real-world conditions and cost-effectiveness," concludes Paton. "This makes it ideal for range and downforce optimization as well as achieving those all important marginal gains. By blending CFD simulations with real world testing, customers can conduct advanced vehicle development programmes and ultimately improve the performance of their vehicles."

---

##### Interesting links:

 

<ExternalLink href="https://catesbytunnel.com/">Catesby Tunnel</ExternalLink>

 

<ExternalLink href="https://www.wltpfacts.eu/">WLTP</ExternalLink>

 

<Link href="/videos/how-to-run-an-aerodynamic-simulation/O_DNDxCtI_I">
 How to use coastdown testing to measure drag
</Link>

What the Catesby Tunnel brings to vehicle testing

## A brief history

Aerodynamics and motorcycle racing have had a patchy relationship over the last seventy years. Whilst Formula 1, and subsequently most forms of four-wheel racing, went all-in during the 1970s, it has remained on the periphery for bikes until comparatively recently.

That’s not to say that aerodynamics hasn't been important or thought about, more that it was limited in its application. This vintage advert from 1965 below shows how aerodynamic road bike fairings were used for performance and rider comfort on Royal Enfields.

<ImageWithCaption
 src="/assets/images/royal-enfield-1965-advert.jpeg"
 alt="A 1965 advert promotes 'aerodynamically designed fairings' for Royal Enfields. CREDIT: www.cybermotorcycle.com"
 width="800"
 height="637"
 positionCenter={true}
/>

In 1950s racing, the dustbin fairing was used to reduce drag for performance until banned on safety grounds. Doubts were raised about crosswind stability, especially at events like the Isle of Man TT, where weather conditions can significantly affect the race. Even today, fairings are often drilled in particularly windy conditions for this reason.

<ImageWithCaption
 src="/assets/images/1950s-old-dustbin-fairing.jpeg"
 alt="In the 1950s, dustbin fairings were used to minimise drag but were banned due to crosswind stability. CREDIT: Onebauer media"
 width="600"
 height="351"
 positionCenter={true}
/>

Racing bikes continued to develop fairings and work on rider positioning to reduce drag within the regulations, but this was more or less the scope of aerodynamic development for many years. That has changed recently, as competition and technology have driven teams to look for performance gains into areas other than just drag reduction.

## Drag and slipstreaming

To start with, let's discuss drag as it has been an ever-present consideration in motorcycle performance, both in racing and on the road. Increasing top speed almost always guarantees performance, which is why reducing drag is so important.

MotoGP engine development is one of the key battlefields in the quest for race wins and millions of dollars are spent extracting horsepower. That investment is of no use if power gains are wasted in aerodynamic losses. Compared to braking or cornering, it’s also generally slightly easier to extract lap time from a top speed increase. Throttle open, hold on!

Unlike a car, particularly closed cockpit racers, bikes have multiple exposed elements. These include the front wheels and forks, fairing edges, cooling air exiting from the radiator, the rider, exhausts and swing arm. The resulting airflow is highly turbulent, with multiple separations and reversed flow that add to the pressure drag force that has to be overcome by engine power.

<ImageWithCaption
 src="/assets/images/motorcycle-drag.png"
 alt="The red surfaces show the wakes generated by a motorbike"
 width="750"
 height="394"
 positionCenter={true}
/>

This means that, whilst considerably smaller in frontal area than a car, a bike’s drag coefficient is much higher. The net result is a similar drag level to a car, despite a bike's smaller size. With generally less engine power available, this equates to a top speed performance deficit for bikes, but also means that more performance can be extracted by reducing drag.

<ImageWithCaption
 src="/assets/images/car-vs-motorbike-drag.png"
 alt="Despite a smaller frontal area, bikes can have almost double the drag coefficient of a modern car"
 width="700"
 height="432"
 positionCenter={true}
/>

The other, perhaps less obvious, factor driving drag reduction is the desire to reduce the size of the wake a bike produces. This “hole in the air” gives an advantage to following bikes, giving them a useful speed increase to help overtaking. This is of particular interest in Moto2 and Moto3 where more common bike parts mean that small differences can have a large effect. Brian Gillen from MV Agusta discusses this with AirShaper in this video.

<YouTubeEmbed
 videoId="BUkaejkHRn8"
 title="Race Talk with MV Agusta - Top Aerodynamic Priorities for MotoGP bikes"
/>

## The importance of the rider

The other defining feature of a motorcycle is the exposed rider and how much they move around. From a drag point of view, optimising rider position whilst on the straight is key to managing drag and the resulting wakes.

Teams analyse body, arm and leg positioning as well as the hump on the back of riders’ leathers, the evolution of which is discussed in <ExternalLink href="https://demonerosso.dainese.com/the-aerodynamic-hump-and-its-history">this article by Dainese</ExternalLink>. Small positional changes to foot pegs and even the angle of a rider’s feet can make a big difference.

<YouTubeEmbed
 videoId="qrH_QBOrqbw"
 title="Find out how rider posture can increase top speed by 10mph"
/>

Rider comfort, probably better described as rider stability, can’t be ignored either. In the unsteady flow regimes that exist on a motorcycle, it is desirable to control the way in which flow detaches from the fairing and the rider’s helmet. If this flow is poorly controlled, it can result in unwanted forces on the rider’s body and head that will affect human performance, particularly over a race distance.

The rider can also be used as an aerodynamic aid. Moving out from behind the fairing changes the frontal area and drag coefficient of the bike dramatically. From a simplistic point of view, this increases the drag force and is why it has become a common sight to see riders hanging a leg out under braking.

Looking in more detail, the rider rarely moves in a way that is symmetrical, instead hanging off the bike to one side or the other. This means that the centre of pressure of drag on the bike is no longer along the centreline, instead it is to one side; generating a yaw moment. Harnessing this can be used to help the bike turn-in and corner.

## Keeping the wheels on the ground

Moving away from drag in a top speed sense, brings us to one of the dominant features of motorcycle performance: the wheelie. For a long time, controlling this has been the domain of mechanical design, set-up and electronics. But why is it such a problem?

It all comes down to the relative positions of the centre of gravity and centre of aerodynamic pressure of a motorcycle compared to a car. This is illustrated in the image below where Lewis Hamilton and Valentino Rossi swapped machinery in 2019.

<ImageWithCaption
 src="/assets/images/hamilton-and-rossi-swap.jpeg"
 alt="Lewis Hamilton and Valentino Rossi swapped racecar for race bike in 2019. CREDIT: www.mercedesamgf1.com"
 width="750"
 height="558"
 positionCenter={true}
/>

The centre of gravity for the MotoGP bike is much higher than the Formula 1 car. This is coupled with a very short wheelbase (around 1.5m for the bike compared to 3.5m for the car) and a low overall weight (around 210kg for the bike and rider compared to 752kg for car and driver). This makes the vertical loads going through the bike’s tyres very sensitive to acceleration and drag.

<ImageWithCaption
 src="/assets/images/motorcycle_balance_with_downforce.png"
 alt="Illustration of the forces acting on a motorcycle with downforce"
 width="700"
 height="468"
 positionCenter={true}
/>

A little maths shows the result of these differences. Ignoring aerodynamic drag and downforce, the load transfer off the front axle onto the rear due to acceleration is given by the formula below:

<ImageWithCaption
 src="/assets/images/load-transfer-formula-4.png"
 alt="Load transfer formula"
 width="600"
 height="287"
 positionCenter={true}
/>

If we assume the weight is split 50/50 between front and rear axle, we can work out the starting load on each axle for the bike (1030N) and car (3687N). Estimating some values for car and bike dimensions gives the table below. As you can see, at only 0.5g acceleration, (where g is the acceleration due to gravity and is equal 9.807 m/s²) the bike transfers from front to rear over 50% of the static axle load available (549N out of 1030N), whilst the car transfers less than 10% (316N out of 3687N).

<ImageWithCaption
 src="/assets/images/transfer-load-table.png"
 alt="Transfer load and axle load table"
 width="650"
 height="103"
 positionCenter={true}
/>

Add in the effect of drag and the load transfer increases further. It takes very little to reduce the front axle load on the bike to near zero, the wheelie. As the front axle loses load, the steering becomes less effective and the front cornering grip reduces considerably, as discussed in this video.

<YouTubeEmbed videoId="M2Z_GuLHmHs" title="Motorbike wheelies" />

The same problems affect an F1 car to some extent, but really are not an issue at high speeds. The loads on the axles are so high from the additional aerodynamic load that the reduction is barely noticed. This is starting to be addressed on MotoGP bikes in the same way.

Arguably, Ducati have been at the forefront of much of this as they have looked to take on the dominance of Honda and Marc Marquez. For the most part, this has been in the form of front fairing winglets which attempt to add load to the front axle to combat load transfer. The rise of these winglets over the last few seasons has been a hot topic of conversation and a controversial one at that.

<ImageWithCaption
 src="/assets/images/ducati-front-winglets-2.jpg"
 alt="Ducati's front winglets have been a controversial topic over the last few easons of MotoGP. CREDIT: www.missionwinnow.com"
 width="600"
 height="376"
 positionCenter={true}
/>

The complexity of the movement of a bike means that these types of development are not always just gains. As a bike leans into a corner, the aerodynamic surfaces become angled. As the lean increases, the proportion of downforce that is acting normal to the track decreases. Instead, the downforce-producing elements are generating a side force pulling the bike out of the corner. This may lead to a requirement for the rider to hang off the bike more in corners, whilst keeping the chassis more vertical.

<ImageWithCaption
 src="/assets/images/downforce_tilt.png"
 alt="Downforce acts in line with the lean angle of the bike"
 width="700"
 height="457"
 positionCenter={true}
/>

## Cooling it all down

The final area of aerodynamic development has actually been around for a long time. The dustbin fairings mentioned above had various scoops added to them to allow the brakes to cool, supply air to the engine intakes and to keep the rider comfortable.

There is performance to be gained from engine cooling, reduced weight of components and controlling the drag by optimising the air flow into and around the radiator. The same goes for the ram air effect into the engine intake - more oxygen means more engine power.

<ImageWithCaption
 src="/assets/images/engine-intake-motorbike.png"
 alt="Effectively cooling the engine can lead to signficant performance gains"
 width="700"
 height="365"
 positionCenter={true}
/>

Manufacturers are always looking at the next big development. Aprilia recently published a patent which looked at reducing the effect of blockage on the flow to the radiator from the front wheel as it was turned. In doing this, the airflow to the radiators could be improved, particularly at low speeds when there was less airflow available in the first place. This may allow smaller or lighter coolers.

## The future

Unless the <ExternalLink href="https://www.fim-moto.com/en/sports/view/fim-world-championship-grand-prix-4327">FIM regulations</ExternalLink> step in to make changes, the role of aerodynamics is set to increase over the seasons ahead. Competition between manufacturers will always drive technology and the detail with which everything is scrutinised. As CFD and virtual prototyping capabilities continue to advance, development will follow. Much of that will come from improved software capabilities and processing, as discussed in this interview with MV Agusta.

<YouTubeEmbed
 videoId="fGNdMaVft_w"
 title="Race Talk with MV Agusta - How to Improve Moto GP Aerodynamics"
/>

Winglets are just the start, wheel fairings have appeared on MotoGP bikes and various ducting solutions are being developed. As bikes lean, parts of the lower fairing and wheels get close to the ground, which perhaps opens the door to harnessing ground effect for performance purposes. As usual, these are shrouded in secrecy in a way that only motorsport can, but there is plenty of speculation as to their purpose.

<ExternalLink href="https://www.motorsportmagazine.com/articles/motorcycles/motogp/is-ducati-using-ground-effect-for-more-grip-in-motogp">
 - Is Ducati using ground effect to gain grip in MotoGP?
</ExternalLink>

It will also drive development in road bikes. Fashion and engineering mean that future (and current) road bikes will start to feature the items that are racing today.

Interesting links:

 

<ExternalLink href="https://demonerosso.dainese.com/the-aerodynamic-hump-and-its-history">
 The aerodynamic hump and its history
</ExternalLink>

 

<ExternalLink href="https://www.mercedesamgf1.com/en/news/2019/11/lewis-hamilton-valentino-rossi-swap-machinery/">
 Lewis Hamilton and Valentino Rossi swap machinery
</ExternalLink>

 

<ExternalLink href="https://www.bennetts.co.uk/bikesocial/news-and-views/news/2020/june/aprilias-aero-trickery">
 Revolutionary Aprilia MotoGP aero could influence future RSV4
</ExternalLink>

 

<ExternalLink href="https://www.motorsportmagazine.com/articles/motorcycles/motogp/is-ducati-using-ground-effect-for-more-grip-in-motogp">
 Is Ducati using ground effect for more grip in MotoGP?
</ExternalLink>

 

<ExternalLink href="https://www.ducati.com/ww/en/bikes/panigale/superleggera-v4/aerodynamics">
 Interview with Edoardo Lenoci, Ducati Aerodynamic Development Manager.
</ExternalLink>

 

<ExternalLink href="https://app.airshaper.com/projects/new?accuracy=regular">
 Run Your Own Simulation
</ExternalLink>

The secrets behind MotoGP aerodynamics

If you’ve been keeping your eye on the Tokyo 2020 Olympics, you would have undoubtedly heard the commentators refer to aerodynamics at some point. Whether it’s cycling, golf or running, when you’re trying to be the best in the world you need to exploit every performance gain you can - and often the biggest gain to be found is in aerodynamics.

## Cycling

Cycling has been an Olympic sport since the 1896 Athens games. For Tokyo 2020 there are a total of 22 different cycling events across five disciplines: track cycling, road cycling, mountain bike, BMX Racing and BMX Freestyle. But whatever the category, riders want to extract the maximum performance out of every effort they put in. Over 80% of the energy a cyclist puts into the pedals is used to overcome drag, so improving aerodynamic efficiency is the best way to ride faster for longer.

### Shin tape <a name="olympicsshintape"></a>

One of the most controversial talking points from the Tokyo Velodrome has been, quite bizarrely, kinesiology tape. Lining up for their qualifying round in the men’s team pursuit, each of the Danish riders had a strip of kinesiology tape on the front of their shins. Each strip was in exactly the same position and extended upwards from within the socks to just below the knee.

<ImageWithCaption
 src="/assets/images/danish-team-pursuit-shin-tape.png"
 alt="Shin tape on the Danish team pursuit riders at Tokyo 2020. CREDIT: Getty Images and www.cyclingweekly.com"
 width="750"
 height="512"
 positionCenter={true}
/>

A rider’s shins can be an area of significant aerodynamic gain. Increasing the surface roughness in this location can initiate a turbulent boundary layer which mixes with surrounding air and retains kinetic energy. This energy allows the airflow to remain attached for longer, reducing the wake behind the legs and the associated pressure drag.

<ImageWithCaption
 src="/assets/images/surface-friction-cyclist.png"
 alt="The red circles highlight transitions from high to low surface friction to identify areas of flow separation"
 width="700"
 height="461"
 positionCenter={true}
/>

In the past, teams have used socks and overshoes to try and exploit this aerodynamic benefit. Overshoes that extend up to the knee can reduce drag by approximately 1%. Although on the 1st of January 2019, the UCI introduced a sock height rule to ensure that socks were only used for clothing and safety and not for gaining an aerodynamic advantage.

It was highly unlikely that each Danish rider was experiencing the same injury on both shins and therefore required kinesiology tape. Of course, this was quickly spotted and protested by other teams and the tape was consequently banned with the Danish team issued a warning.

### Helmet design

One of the biggest differences you will spot between indoor and outdoor cycling is the range of helmet designs. Indoor cyclists will often opt for a streamlined tear-drop shaped helmet with an extended tail. By having a seamless transition from the helmet to the back, the airflow can remain attached, avoiding separation and reducing drag. Well-designed tear drop helmets can achieve lower drag coefficients than more conventional spherical shaped helmets, but only if the rider maintains the optimum head position.

However, riding this type of helmet in an outdoor event could have a disastrous effect. Side winds change the angle of attack of the rider and consequently increase their effective frontal area, particularly when riding with covered wheels. This not only increases drag but also introduces side forces which can destabilise the bike. Wearing a long helmet will exacerbate this affect, increasing the frontal area and the resulting lateral forces.

<ImageWithCaption
 src="/assets/images/track-cycling-teardrop-aero-helmet.png"
 alt="With no side wind indoors, the bike, helmet and clothing can be optimised for one position in track cycling.CREDIT: AFP and www.telegraph.co.uk"
 width="700"
 height="439"
 positionCenter={true}
/>

From an aerodynamic point of view, indoor cycling is an aerodynamicists playground as the lack of side wind means that the bike, rider and helmet can all be optimised for only one position. This is why track cyclists will have much more aerodynamic helmets, skinsuits and bikes.

### Rider posture

Whether it’s track or road cycling, the rider’s posture has a huge influence on their aerodynamic efficiency. Riders are larger than their bikes and can account for up to 80% of the drag. Moving the torso closer to the bike will reduce frontal area and drag, but if too low, this can affect the rider’s breathing and power output. While the centre of gravity of the rider and the bike should be aligned, with the rider exerting a constant vertical force throughout the four phases of the pedal stroke.

To accurately analyse a rider’s posture and body movements, motion analysis software now plays a key role in helping athletes race faster. Reflective markers are placed on a rider’s articulations and infrared lights along with high speed cameras track the rider’s movements. This can then be used to optimise the rider’s posture, seat height and pedal frequency.

<ImageWithCaption
 src="/assets/images/athlete-motion-analysis-software.jpg"
 alt="Motion analysis software is being used to optimise athelte's posture and performance in many sports. CREDIT: www.motionanalysis.com"
 width="700"
 height="558"
 positionCenter={true}
/>

## Athletics

### Clothing

Each Olympic games brings with it new innovations in aerodynamic clothing. The baggy shorts and T-shirts of the past are now replaced with tight fitting synthetic speedsuits that incorporate several different technologies.

One fascinating technology that is now used in kits, therapeutic tapes as well as leg and arm sleeves is the Nike AeroBlades. Exploiting a similar philosophy to golf ball dimples, miniature nodes are located on specific areas of the fabric and channel air around the athlete in the most effective way; reducing drag.

- <Link href="/videos/golf-ball-dimples-aerodynamics-how-do-they-work-and-are-they-relevant-for-your-design/KzXcglAhxps">
 Why do golf balls have dimples?
 </Link>

<ImageWithCaption
 src="/assets/images/nike-aeroblades.jpeg"
 alt="Nike AeroBlades are miniature nodes on fabric which help to reduce drag. CREDIT: www.nike.news.com"
 width="600"
 height="429"
 positionCenter={true}
/>

The precise location of these nodes is defined by wind tunnel testing, where mannequins of the same size and shape of Nike’s athletes are used to identify areas of the highest wind resistance. This data then guides the placement of the AeroBlades for that specific athlete to achieve maximum performance. This, along with other technology in Nike’s speedsuits can achieve gains of up to 0.023 seconds over 100m.

<ImageWithCaption
 src="/assets/images/aeroblades-tape-on-athlete.jpeg"
 alt="Nike AeroBlades is incoporated into AeroSwift tape as worn here. CREDIT: www.nike.news.com"
 width="700"
 height="500"
 positionCenter={true}
/>

## The stadium

It’s not just the athletes and equipment that are designed to be aerodynamically efficient, but the stadiums are too. A 2m/s tail wind can drop the 100m sprint time by around a tenth of a second, while a head wind can increase that time by around 0.1s. So within the official limits stated by the World Athletics (+- 2m/s - to be double-checked), times can vary within a range of 0.2 tenths due to aerodynamics alone.

In reality though, the behaviour of the airflow within a stadium is more complex as usually the wind speed along the track is not constant. To investigate this, we used a public 3D model of the Tokyo Olympics stadium and ran CFD simulations to understand how wind outside the stadium translates into complex flow patterns inside the stadium. As you can see from the image below, the result can be a variety of large and small vortices along the race track, illustrated by the white line.

<ImageWithCaption
 src="/assets/images/tokyo-stadium-vortices.png"
 alt="Streamline trace for the flow along the middle lane of the race track as shown by the white line"
 width="750"
 height="388"
 positionCenter={true}
/>

At the midpoint of the track, where the official measurement is taken, we observed a 'virtual wind speed measurement' of around 0.14m/s. At the same time, we saw that at the beginning of the track there was a tail wind of 0.4m/s, which was 280% more than the measured velocity. While near the end, athletes would experience a head wind of 0.1m/s as shown by the graph below. However, as Bert Blocken argues in the link below, there may be some inaccuracies within the measurements of this study.

<ExternalLink href="https://www.linkedin.com/pulse/why-elaine-thompson-herah-only-fastest-woman-alive-ever-bert-blocken/?trackingId=mucmVDR36K8W4jm1ZW1Jaw%3D%3D">
 Why Elaine Thompson-Herah is the fastest woman ever on the planet.
</ExternalLink>

<ImageWithCaption
 src="/assets/images/tokyo-stadium-flow-patterns.png"
 alt="Top: Birds eye view of the complex flow patterns within the Tokyo Olympic Stadium. Bottom: Graph showing the longitudinal flow velocity along the middle lane of the race track."
 width="750"
 height="376"
 positionCenter={true}
/>

Another aspect is the comfort of the spectators, both inside and outside the stadium. The complex flow structures inside the stadium can make some areas relatively windy, while others can be nicely shielded and more comfortable. Outside the stadium, the building and surroundings can strongly impact flow patterns, making the difference between sitting comfortably on a bench or being blown over when cycling past.

<ImageWithCaption
 src="/assets/images/cp-tokyo-stadium.png"
 alt="External aerodynamics of the Tokyo Olympic Stadium"
 width="650"
 height="380"
 positionCenter={true}
/>

So next time you’re watching the Olympics, see how many aerodynamic innovations you can spot on the athlete's themselves, on their equipment and in their surroundings.

Interesting links:

 

<Link href="/videos/sports-aerodynamics-1-winning-the-ironman-triathlon/DqlDpUFcScc">
 Cycling: From 3D scan to aerodynamic analysis
</Link>

 

<ExternalLink href="https://app.airshaper.com/projects/new?accuracy=regular">
 Run Your Own Simulation
</ExternalLink>

Aerodynamics at the Olympics

## Introduction

In modern cycling, there is a continuous search for marginal gains. Bicycle manufacturers, cyclists and teams are constantly trying to find new ways to take their performance to the next level. Team DSM is one of the most performance (and science) driven teams in the UCI World Tour. We had the chance to talk with Team DSM’s Aerodynamics Expert, Harm Ubbens.

<ImageWithCaption
 src="/assets/images/harm-ubbens-profile.jpg"
 alt="Harm Ubbens - Aerodynamics Expert at Team DSM"
 width="200"
 height="200"
 positionCenter={true}
/>

Harm studied an MSc in Aerospace Engineering at the Technical University of Delft. During his degree his interest shifted from airplane to sports aerodynamics and he completed his thesis on bobsleigh aerodynamics. He also helped design the wind tunnel at BikeValley in Belgium during his internship and later worked there as aerodynamicist and wind tunnel operator. At the beginning of 2020 he joined Team DSM as their Aerodynamics Expert.

## A history of cycling aerodynamics

In the past, there have been a number of pioneers in cycling aerodynamics. For example, Greg LeMond in the finishing time trial of the Tour de France in 1989. He was the first cyclist to use an aerobar extension on his steer as well as wear a profiled time trial helmet. At that time, other riders wore just a cap, or nothing at all. This allowed LeMond to ride in a more streamlined position which gave him the competitive edge to win the Tour de France. He won by an 8 second time difference over the 87 hours of cycling.

<ImageWithCaption
 src="/assets/images/greg-lemond.jpg"
 alt="Greg LeMond in aerodynamic position during the final stage of the 1989 Tour de FRance - image credit: capovelo.Com"
 width="800"
 height="524"
 positionCenter={true}
/>

Graeme Obree was another pioneer. Working towards his world hour record attempt in 1993, he built a bike using parts of his washing machine. He combined this with extremely compact or stretched positions on his bike, and broke the world hour record twice.

<ImageWithCaption
 src="/assets/images/graeme-obree1.jpg"
 alt="Graeme Obree's tuck position - image credit: cyclingweekly.com"
 width="800"
 height="532"
 positionCenter={true}
/>

<ImageWithCaption
 src="/assets/images/graeme-obree2.jpg"
 alt="Graeme Obree's superman position - image credit: roadcyclinguk.com"
 width="800"
 height="469"
 positionCenter={true}
/>

## The importance of aerodynamics in cycling

In general, cycling can be seen as overcoming resistance. On the flats, about 90% of the total resistance a cyclist feels, is due to aerodynamic drag. If you can minimize this, the cyclist can go faster with the same power output; gaining time over their competitors.

When climbing very steep gradients, the relative importance of aerodynamics decreases as the importance of gravity increases. Luckily, there are many ways to reduce aerodynamic drag and improve performance through special clothing and custom components such as aero helmets and aero wheels.

Changing gravity, by contrast, is quite difficult. So aerodynamics is definitely one of the most important factors that can be influenced to increase performance. It is, however, important to always optimize it in combination with other aspects like ergonomics. The most aerodynamic position usually leads to a reduced power output and as such, does not make the cyclist faster.

- <Link href="/videos/designing-an-aero-bike/DqlDpUFcScc">
 Find out how CFD can help optimise rider position
 </Link>

## Aerodynamic cycling positions

In general, every pro rider is into aero aerodynamics, but you could divide the peloton into two groups. The first are those specialized in climbing. The second group are the more powerful riders. If you belong to the second group, you cannot ignore the significance of aerodynamics.

There are some examples of riders willing to do everything they can to go faster. World hour record holder Victor Campenaerts is a good example of a rider who is passionate about aerodynamics. He is prepared to do everything to improve his aerodynamics, including getting into some incredible positions.

## Bike versus rider

The bike is an important part of a cyclist’s aerodynamics, but not as important as riding position. That's why amateur riders will get more value for money by optimising their position through a bike fitting than buying an aero bike.

## Aerodynamic innovation within the UCI rules

Today, it’s still possible to improve the aerodynamics through continuous innovation in terms of production methods. The biggest evolution over the last five years is probably the integration of braking and shifting cables. These cables are round and tend to initiate flow separation, reducing the flow quality along the body and frame. Integrating these cables provides a significant reduction in aerodynamic drag.

Otherwise it has become difficult to improve the aerodynamics of a bike within the UCI rules. Typically, an aerodynamic bike is heavier due to the special shapes which has a negative effect when climbing. That is why manufacturers are now aiming to make aero bikes as light as possible. Examples of bikes that are both aerodynamic and lightweight are the Scott Addict RC and the S-Works Tarmac SL7.

<ImageWithCaption
 src="/assets/images/specialized-tarmac-sl7.jpg"
 alt="Specialized Tarmac SL7 - image credit: specialized.com"
 width="800"
 height="514"
 positionCenter={true}
/>

<ImageWithCaption
 src="/assets/images/scott-addict-rc.jpg"
 alt="Scott Addict RC - image credit: scott-sports.com"
 width="800"
 height="501"
 positionCenter={true}
/>

## Future innovations in cycling aerodynamics

When manufacturers switched from steel frames to aluminium ones, a lot of research was carried out on the shape of the tubes. This was again analysed when aluminium frames were replaced with carbon ones. The next innovation was the integration of brake cables which was then followed by minimizing weight.

The current approach often optimizes the bike, wheels, helmet etc individually without considering the interaction of the cyclist. Adapting their riding position to the equipment can lead to unpleasant ergonomics for riders. Therefore, in the future, we will likely see a more holistic approach where both the rider and bike are optimized simultaneously.

## Components of aerodynamic drag

The aerodynamic drag on a cyclist can be split into skin friction drag and pressure drag. Skin friction is the result of the air sliding across the surface. It is literally the friction between the air and the skin or <Link href="/videos/aerodynamic-clothing/A0Gx-lBzSFA">aerodynamic clothing</Link> of a cyclist.

Pressure drag is the difference between the high pressure at the front of the cyclist and the low pressure at the rear. This delta is mainly caused by flow separation and vortices behind the cyclist, which reduce pressure recovery. These forces hold the cyclist back at higher speeds.

When riding a bike on a levelled road, around 90% of the resistance is the result of aerodynamic drag. Because a cyclist is a bluff body (a non-streamlined object), about 85% of this drag is pressure drag and 15% is skin friction drag. Although 15% is not negligible, pressure drag is the most important factor. That is why the focus is on optimizing the shape of components and athlete position, rather than minimizing skin friction.

## Aerodynamic testing at Team DSM

Our main techniques are track testing and wind tunnel testing. Track testing involves the cyclist tweaking their position while doing laps on a track. Obviously, it’s important to keep the power output constant and to ride the exact same line every lap. This is a clear drawback of this technique as not every cyclist can perform this test accurately.

Wind tunnel testing on the other hand allows you to improve small details in a controlled environment. The biggest downside of the tunnel, however, is that you’re working on a fixed position most of the time.

Keep in mind that the tests described above are done in ideal circumstances. There is no influence of any natural wind or rain. For this reason, outdoor testing is a popular tool to analyse the effect of aerodynamic devices. This provides more accurate results than the wind tunnel or track testing.

Our partner, TU Delft, developed an outdoor testing technology called 'The Ring Of Fire'. This is where a cyclist rides through a 'ring' of helium bubbles. As the flow pattern around the rider moves the bubbles, the bubbles are tracked by PIV (Particle Image Velocimetry).

<YouTubeEmbed
 videoId="w5WnYEHTgus"
 imgSize="default"
 title="The Ring of Fire - TU Delft"
/>

Performing PIV measurements in open air allows us to gain more information about the aerodynamics of the individual cyclist. It also helps to learn which team tactics are best to implement in TT (Time Trial) and TTT (Team Time Trial) events.

Each type of testing has its own upsides and downsides. The kind of test we do depends on what data we want to get out of it. So in conclusion, we can state we use a combination of different techniques throughout our search for optimal aerodynamics.

<ImageWithCaption
 src="/assets/images/nikias-arndt.jpg"
 alt="Nikias Arndt in the 2021 UAE Tour TT - image credit: Twitter @ NikiasArndt"
 width="800"
 height="532"
 positionCenter={true}
/>

## Aerodynamics meets biomechanics

When a new cyclist arrives at our team, the first task is to get a bike fitting to avoid injury during training. We then look for the sweet spot between their biomechanics and aerodynamics to achieve the best performance.

Our team aims to have the best specialists of each field work together as one team. Each opinion is equally valuable, without one person taking the lead. This ‘one team, one goal’ philosophy will eventually lead to the best possible performance of our cyclists.

## The science behind cycling aerodynamics

At our team, everything is science-driven. That is why our team searches for the best specialists in every field. As far as I know, Team DSM is one of the few teams which works like this.

Most teams will have someone looking at the aerodynamics, but often that person has a background in biomechanics or sports performance. They know how to make a cyclist go faster, but my aerospace engineering background will hopefully take things to the next level. And I’m very grateful to be in that position!

## Conclusion

Via new production techniques, crazy positions and innovative measurement systems, cycling aerodynamics have come a long way. The continuous search to improve even the smallest aspects will never end. We are very thankful to Team DSM and Harm Ubbens who were willing to share their knowledge with us.

Interesting links:

<ExternalLink href="https://www.team-dsm.com/">Team DSM</ExternalLink>

 

<Link href="/videos/sports-aerodynamics-1-winning-the-ironman-triathlon/DqlDpUFcScc">
 Cycling: From 3D scan to aerodynamic analysis
</Link>

 

<ExternalLink href="https://app.airshaper.com/projects/new?accuracy=regular">
 Run Your Own Simulation
</ExternalLink>

Cycling Aerodynamics - Interview with Team DSM

## Meshing

The behaviour of airflow around an object is extremely complex. Phenomena such as turbulence, flow separation and eddy currents can be present and significantly affect aerodynamic performance. To accurately model the intricacies of this flow, CFD simulations need to run with finer meshes in these areas of interest.

<ImageWithCaption
 src="/assets/images/motorBike_before_refinement_slice_2.png"
 width="700"
 height="350"
 positionCenter={true}
/>

<ImageWithCaption
 src="/assets/images/motorBike_before_refinement_slice_zoomed_2.png"
 alt="A mesh around a motorbike which features a refined mesh close to the surface and a coarser mesh further away"
 width="700"
 height="362"
 positionCenter={true}
/>

In CFD, the volume around an object is divided into miniature 3D blocks called cells. Similar to pixels in a digital photograph, the smaller the cells, the more detail you can capture. However, a higher number of cells requires more processing power. Therefore, to balance this demand for detail with efficient run times, aerodynamicists refine the mesh in areas of complex flow behaviour, using a low resolution for the rest of the mesh.

- <Link href="/videos/computational-fluid-dynamics-explained/dyiREvdc4Gg">
 Learn how CFD simulations work
 </Link>

## Adaptive mesh refinement

However, it is impossible for even the most experienced aerodynamicists to accurately predict where the flow will be the most complicated, after all that’s why they’re using CFD. To help solve this problem, adaptive mesh refinement was invented.

This technique runs an initial simulation with a coarse mesh, determines where higher resolutions are required and then automatically refines the mesh in these areas. This process is repeated until the desired accuracy and run time are satisfied. It is also used to refine/unrefine a mesh dynamically during transient simulations, <ExternalLink href="https://www.youtube.com/watch?v=u-VV3euIsXo&ab_channel=HolzmannCFD">as in this Karman vortex street simulation.</ExternalLink>

At the core of AirShaper is an open source CFD software called OpenFOAM. It utilises adaptive mesh refinement but only refines the cells in the volume around an object and not on the object’s surface itself. AirShaper can work with any given 3D model, <Link href="/videos/save-weeks-run-aerodynamic-simulations-on-open-surface-models/12xA80foGXk">even if the geometry is not completely watertight</Link>, so having the capability to use adaptive mesh refinement for such a wide variety of 3D shapes is essential. Not being able to refine the surface detail of the model could lead to unrepresentative results.

To ensure that adaptive mesh refinement could be used for both the volume around an object and for the object’s surface in AirShaper, we developed our own code. Using the default mesher snappyHexMesh in OpenFOAM, we’re excited to announce that we have now made this part of the code open source.

## Open source repository

The <ExternalLink href="https://github.com/airshaper/adaptive-mesh-refinement">AirShaper adaptive mesh refinement repository</ExternalLink> works in a series of steps, as highlighted in the diagram below.

<ImageWithCaption
 src="/assets/images/adaptive_mesh_refinement_flow_chart.png"
 alt="Flow chart showing the stages of AirShaper's open source adaptive mesh refinement code"
 width="600"
 height="506"
 positionCenter={true}
/>

## Original mesh

The process for a regular snappyHexMesh is as follows:

- **Castellated phase:**

 The blockMesh is refined based on edge, surface and region refinement settings. These are manual refinement settings, which are different from adaptive mesh refinement. Also, any mesh inside an object is removed.

- **Snapping phase:**
 Here, the cut cells are snapped to the surface of the object.

For our adaptive mesh refinement solution, we wanted to keep the castellated mesh and store it separately before it is cut out and snapped. Therefore, the castellated phase and snapping phase are executed separately, with the following modifications:

- **Castellated phase:**

 To avoid the inner mesh being cut away, the patches are set to baffle type. This will leave the entire castellated mesh intact. A copy of this mesh is then stored in the “primal refinement” folder.

- **Snapping phase**
 This picks up the castellated mesh generated in the first step, cuts out the inside cells, and then snaps them to the object, just like it would normally.

This “original mesh” is now ready for a CFD simulation with the results stored in the “primal run” folder.

## Refined mesh

To adaptively refine the mesh based on the flow field, the results from the first CFD simulation are mapped onto the castellated mesh that was stored in the “primal refinement” folder. Then, a refinement parameter called “refVal” is calculated (the repository provides methods for both gradient of pressure and vorticity) using those fields and the castellated mesh. It’s important to do this on the castellated mesh, as refinement in snappyHexMesh cannot be done on a snapped mesh.

<ImageWithCaption
 src="/assets/images/motorBike_after_refinement_slice_curlU_2.png"
 width="700"
 height="311"
 positionCenter={true}
/>

<ImageWithCaption
 src="/assets/images/motorBike_before_refinement_slice_refVal_curlU_2.png"
 alt="Illustration of the refVal field based on flow vorticity. Notice how it nicely captures the wake"
 width="700"
 height="214"
 positionCenter={true}
/>

Once the refVal field has been created, the castellated mesh is refined by running pimpleFoam, which uses the refVal field. The user can play with the threshold values of refVal to select more or less cells for refinement in the dynamicMeshDict file.

Once this castellated mesh has been refined, it is sent back to the “primal run” folder. There, the snapping phase is executed again, but now starting from this refined mesh. In this way, all the cells on the surface will be refined and snapped to the object. Once this snapping phase is complete, a second CFD run is then executed with the newly refined mesh.

<ImageWithCaption
 src="/assets/images/motorBike_after_refinement_slice_zoomed_curlU_2.png"
 alt="Example of a mesh refined using the flow vorticity – both volume and surface mesh are refined"
 width="700"
 height="337"
 positionCenter={true}
/>

## What experts say

We’ve had the honour of receiving feedback from a number of leading OpenFOAM experts. Here’s what they had to say about the repository:

- Jozsef Nagy (Founder of <ExternalLink href="http://www.eulerian-solutions.com/solutions.html">eulerion-solutions</ExternalLink> / <ExternalLink href="https://www.youtube.com/channel/UCjdgpuxuAxH9BqheyE82Vvw">OpenFOAM YouTube Channel</ExternalLink> / <ExternalLink href="http://wiki.openfoam.com/">OpenFOAM Wiki</ExternalLink>):
 _"This is a great addition to the community contributions of OpenFOAM. Mesh refinement is an important topic in order to intelligently find the balance between grid resolution and simulation runtime. Try it out!"_

 Jozsef even created a <ExternalLink href="https://wiki.openfoam.com/Adaptive_Mesh_Refinement_by_Air_Shaper">dedicated AirShaper wiki page for this repository </ExternalLink>

- Robin Knowles (Founder of <ExternalLink href="https://www.cfdengine.com/about/">CFD Engine</ExternalLink>):
 _"This is exactly the kind of thing that makes me smile about OpenFOAM - taking the standard "building blocks" & doing something really clever with them.
 The repo's not just useful for those looking to do this kind of refinement, but it's also a practical demonstration of "OpenFOAM thinking" & problem solving.
 Well worth digging into 🙏"_

- Joel Guerrero (CTO of <ExternalLink href="http://www.wolfdynamics.com/">Wolf dynamics</ExternalLink> / Assistant Professor at the <ExternalLink href="https://unige.it/">University of Genova</ExternalLink>):
 _"This workflow is a clever (but not so easy to follow) way to do a-posteriori adaptive mesh refinement exploiting OpenFOAM tools. But the most important feature is that you can refine the backGround mesh using the reference refinement value and then use the new backGround mesh to generate a better body-fitted mesh with snappyHexMesh. This means smaller meshes (because we are not doing unnecessary volume mesh refinement), better boundary layers, and no more missing feature edges, No doubt that I will use this workflow to get better meshes with snappyHexMesh."_

## Conclusion

We’ve created an open-source repository to provide the OpenFOAM community with access to a part of our adaptive mesh refinement code. We are excited to be part of this open-source community and look forward to future contributions from the OpenFoam community.

Interesting links:

<ExternalLink href="https://github.com/airshaper/adaptive-mesh-refinement">
 Github repository
</ExternalLink>

 

<ExternalLink href="https://wiki.openfoam.com/Adaptive_Mesh_Refinement_by_Air_Shaper">
 Wiki page
</ExternalLink>

Open source adaptive mesh refinement

## About the author

Michael Peinsitt is Skill Team Leader RACING at <ExternalLink href="https://www.avl.com/">AVL</ExternalLink>, a company which has been involved in racing for almost 25 years with their in-house developed vehicle dynamics simulation tool AVL VSMTM RACE. Inside VSM, all components of the car are modelled and characterized, such as the suspension, the tyres, the engine, and many more. One area is particularly important towards achieving a good correlation between simulation and reality: aerodynamics. Modern race cars produce quite a lot of downforce, which strongly influences performance. And with downforce significantly varying with ride height, it’s important to map it properly in function of the car’s setup.

## Mapping Aerodynamic Properties

When it comes to aerodynamics, the main challenge is to map downforce & drag in function of changes in input parameters. On top of the ride height, they also vary in function of the chassis slip & roll angle, the steering angle of the front wheels and so on. Grouping all of these correlations (or “sensitivities”) between aero coefficients and car setup results in what we call an <Link href="/videos/race-car-aerodynamics-3-aero-mapping/u-GyTRblNmc">Aero Map </Link> – a dataset of properties that can be used before & during the race weekend to setup & adjust the car in function of the track & the driver’s feedback.

## The Options

One option to obtain these sensitivities is **wind tunnel testing**, but this is typically expensive and requires quite some effort. Additionally, rules in certain race categories prevent teams to use wind tunnels.

The second option is to perform **aero measurements on the racetrack**. Apart from the efforts to equip the car with the required measurement equipment, track testing is always subject to varying external conditions. Changes in for example speed and orientation of the wind, or air temperature, make it difficult to compare results from different runs. Plus, in similar fashion as for the wind tunnel testing, in many race categories the rules ban, or severely restrict track testing.

A third option is to use CFD calculations. To execute a **“traditional” CFD workflow** is also time consuming, as the preparation and the meshing of the CAD model usually requires a certain amount of manual effort, plus the execution of the calculation itself needs considerable computing power to be conducted. Once done, post-processing is needed to convert raw simulation data into meaningful figures & visuals.

To capture the influence of the various input parameters on aero performance, a significant number of runs is needed – which means it takes time until the results are available. With preparation time always being a premium in racing, having to wait for results can seriously reduce the competitiveness of a team.

## Automating the simulation process

We turned to AirShaper to investigate the possibility of reducing or even eliminating many of the existing CFD challenges.

<VideoEmbed
 videos={[
 {
 src: 'https://airshaper.storage.googleapis.com/website/blog/videos/avl-setup.mp4',
 type: 'video/mp4',
 },
 ]}
/>

First of all, the need for “CAD-cleaning” – closing all gaps & holes in the 3D model of the car – can take a significant amount of time. AirShaper, however, runs directly on non-watertight 3D models, although in this particular example we had a watertight model available.

Second of all, we don’t need to worry about meshing, as this is entirely automated: the domain size, the initial mesh, adaptive mesh refinement, … are all included and, perhaps even more important, are repeatable. This means results between different models & setups are comparable.

Thirdly, we don’t need to manually post-process all the results. Frontal area, lift, downforce, aero balance, … are all post-processed automatically into a report. And using the online 3D result viewer, we can spot sources of drag, identify separation locations and so on. And if needed, we can always download the full results to do our own post-processing.

## Benchmarking AirShaper

To verify how realistic the output of these simulations would be, AVL conducted a simulation project with a race car that had already been analyzed using a conventional CFD approach (manual meshing, running, post-processing).

The model used for the simulation was a touring car provided by AVL RACING. The car was set up to drive in a straight line (chassis slip & roll angle as well as the steering angle were set to 0).

The exact same 3D model was provided to AirShaper and within 24 hours, the platform churned out CFD simulation results.

## The results

The first important comparison was a quantitative one, comparing the aerodynamic drag and lift coefficients. We were pleasantly surprised to see just a 5.2% difference in drag and an even smaller 3.6% difference in downforce.

<VideoEmbed
 videos={[
 {
 src: 'https://airshaper.storage.googleapis.com/website/blog/videos/avl-results-screen-capture.mp4',
 type: 'video/mp4',
 },
 ]}
/>

The second comparison was a qualitative one, comparing the surface patterns & flow structures in crucial areas on the car. A first example is the pressure map, which exhibits a near-identical pattern across both simulations.

<ImageWithCaption
 src="/assets/images/avl-airshaper-comparison-pressure-map.jpg"
 alt="Pressure map comparison between AVL and AirShaper"
 width="800"
 height="216"
 positionCenter={true}
/>

A second example is the local flow structure seen on a slice through the car, at 0.6m from the center. The way the flow enters the front wheel well, how it accelerates at the start of the diffuser, the velocity pattern around the rear wing, … all look highly similar.

<ImageWithCaption
 src="/assets/images/avl-airshaper-comparison-velocity-slice.jpg"
 alt="Velocity section comparison between AVL and AirShaper (0.6m from the centre)"
 width="800"
 height="207"
 positionCenter={true}
/>

## Conclusion

For us at AVL RACING, as an engineering & software provider in the racing industry, it is extremely important to first validate new software tools before we offer them to customers. Through this validation project, we have gained insight into the capabilities of AirShaper simulations to help speed up the aerodynamic simulations for our customers.
The good accuracy at a fraction of the cost & throughput time certainly makes AirShaper a viable option for future projects.

Interesting links:

<ExternalLink href="https://www.avl.com/">AVL</ExternalLink>
 

<ExternalLink href="https://airshaper.com/videos/race-car-aerodynamics/GApwSjKYUpQ">
 Race Car Aerodynamics
</ExternalLink>
 

<ExternalLink href="https://app.airshaper.com/projects/new?accuracy=regular">
 Run Your Own Simulation
</ExternalLink>

AirShaper BlogPassion for aerodynamics

Accelerating the aerodynamic design process

Benjamin Rembold

What the Catesby Tunnel brings to vehicle testing

Wouter Remmerie

The secrets behind MotoGP aerodynamics

Geoff Dymott

Aerodynamics at the Olympics

Wouter Remmerie

Cycling Aerodynamics - Interview with Team DSM

Floris Mariën

Open source adaptive mesh refinement

Wouter Remmerie

AVL RACING - Putting AirShaper to the test

Michael Peinsitt

Wind Engineering

Xerxes Chong Xian

Trusted By

Awards and Support

Code contributions by