AirShaper

Parts in the NASA interview series:
- Part 1 - New aerodynamic concepts: https://youtu.be/m4rysu8LlLM
- Part 2 - Propeller design: https://youtu.be/UDZBqDQGU8o
- Part 3 - Battery and component cooling: https://youtu.be/ieB541felas
- Part 4 - Electric flight experience: https://youtu.be/pGn07UTVKBg

In this first out of 4 videos on our interview with NASA, we discuss the basic layout of this novel electric aircraft.

The X-57 is the latest in the famous series of X-planes developed by NASA - their experimental aircraft like the X15.

Sean Clarke and Nick Borer give their take on the design of this aircraft. We discuss the benefits if distributed electric propulsion, especially in terms of new aerodynamic concepts that are being made possible.

-----------------------------------------------------------------------------------------------------------

The AirShaper videos cover the basics of aerodynamics (aerodynamic drag, drag & lift coefficients, boundary layer theory, flow separation, reynolds number...), simulation aspects (computational fluid dynamics, CFD meshing, ...) and aerodynamic testing (wind tunnel testing, flow visualization, ...).

We then use those basics to explain the aerodynamics of (race) cars (aerodynamic efficiency of electric vehicles, aerodynamic drag, downforce, aero maps, formula one aerodynamics, ...), drones and airplanes (propellers, airfoils, electric aviation, eVTOLS, ...), motorcycles (wind buffeting, motogp aerodynamics, ...) and more!

For more information, visit www.airshaper.com

Electric Aviation - NASA interview on Maxwell X57 - Part 1: New Aerodynamic Concepts

To learn more about CFD simulations: https://youtu.be/dyiREvdc4Gg

Rotating Wall Boundary Condition
The simplest form is the rotating wall boundary condition. Without rotation, the velocity of the airflow is always zero at the surface because of the no slip condition. The rotating wall boundary condition simply enforces a tangential velocity onto the surface of the rotating object. This velocity is equal to the distance from the center multiplied by the rotational velocity. The further you move away from the center, the higher the velocity.
This technique is very simple and doesn’t significantly change the cost of computation, as no extra equations are introduced and the flow can be solved in a steady state manner. It’s very suitable on surfaces that are tangential to the local direction of rotation.
At AirShaper, we chose this technique to take the rotating wheels of cars into account, as it’s a very robust solution. When you click a tire, the axis of rotation will be detected automatically, the components inside will be selected as well and the rotational velocity is automatically linked to the driving speed of the vehicle

MRF – Multiple Reference Frame
For more complex geometries with a lot of non-tangential surfaces, the more advanced MRF technique can be used. It stands for multiple reference frames. In essence, you create a separate region around the rotating object. Inside this region, for each point, you calculate the relative velocity, which is a combination of the absolute velocity and the rotational velocity.
The Navier-Stokes flow equations are then built on top of this relative velocity, which results in extra terms that take the Coriolis and centrifugal forces into account. So, you end up with a set of equations for the stationary region and another set for the rotating region.
This method is a bit more expensive in terms of computational cost, but still allows for steady-state calculations and it’s able to accurately capture the “instantaneous” flow pattern around a propeller for example.
At AirShaper, we chose to use this technique when adding rotating propellers to a simulation. When selecting a propeller, the central axis will automatically be detected. The direction of rotation can be flipped and the RPM can be set by the user. And it is possible to add multiple propellers to a single simulation 

AMI – Arbitrary Mesh Interface
This instantaneous flow, however, can depend heavily on the relative position of the rotating part versus the static part, like a propeller versus a strut. So, for very advanced simulations, there’s also the AMI technique, which stands for arbitrary mesh interface. In this technique, the mesh around the rotating object is actually cut and then made to rotate. This can only be done using transient simulations, as the rotation of the mesh is now linked to the actual time step. At the sliding mesh interface, the cells of the stationary and rotating mesh exchange information on the flow.

This technique is usually the most accurate, but also very demanding in terms of computational effort. It can also be challenging in terms of stability, as the mesh interface needs to be defined very well. This type of simulation usually requires a manual approach.


-----------------------------------------------------------------------------------------------------------

The AirShaper videos cover the basics of aerodynamics (aerodynamic drag, drag & lift coefficients, boundary layer theory, flow separation, reynolds number...), simulation aspects (computational fluid dynamics, CFD meshing, ...) and aerodynamic testing (wind tunnel testing, flow visualization, ...).

We then use those basics to explain the aerodynamics of (race) cars (aerodynamic efficiency of electric vehicles, aerodynamic drag, downforce, aero maps, formula one aerodynamics, ...), drones and airplanes (propellers, airfoils, electric aviation, eVTOLS, ...), motorcycles (wind buffeting, motogp aerodynamics, ...) and more!

For more information, visit www.airshaper.com

How to model Rotating Elements in CFD (computational fluid dynamics) using MRF, AMI and more

Check out Fluid Mechanics 101:
https://www.youtube.com/channel/UCcqQi9LT0ETkRoUu8eYaEkg

On Aidan's channel, you can also find the uncut, full length video of this interview:
https://youtu.be/Ntl15C2KXCQ
----------------------------------------------------------------------------------------

For this video, we interviewed Aidan Wimshurst, who did his PhD at Oxford University (United Kingdom) on Tital Power Generation.

Concept
Tidal currents originate because of the tides: as water levels change, water flows from one area to another. When the water needs to fit through a narrow channel or move around certain coastlines, flow speeds can be quite high. These streams are very predictable, which makes it much easier to integrate the energy they produce into the grid.

The general idea is to take a typical 3-bladed wind turbine and put it under water. But there are a number of key differences to wind turbines that make tidal turbines unique.

Wing tip vortex
On an airplane, the air at the high pressure location below the wing wants to move to the low pressure location at the top of the wing via the side (the wing tip). This creates a curling motion called wing tip vortex.

On a turbine, this vortex also exists but it leaves a spiral trace instead of a linear one. And this increased vorticity & turbulence can impact turbines further downstream. It also represents efficiency losses for the turbine itself, which is why some (wind turbine) blades are equipped with winglets.

Cavitation
If pressure is reduced below the vapor pressure, you will start getting bubbles. When these implode, they can cause serious damage to the blades, structure, and so on. The risk of cavitation is reduced when the hydrostatic pressure is high, close to the sea bed. But close to the surface, this hydrostatic pressure is much lower, increasing the risk of cavitation. So the risk of cavitation varies along a single revolution.

Farm layouts
For a given plot of land, the goal is to maximize energy and thus put as many turbines as possible in place. But if you put them too close to each other, the turbulent wake of one turbine will lower the efficiency and lifetime expectancy of other turbines, so there is an optimum to be found.

With tidal energy, the orientation of the flow is always the same, only changing direction between tides. So these are typically placed in an array. The flow around a turbine is accelerated, meaning that a turbine placed in this accelerated flow will yield more power, ánd the original turbine will also perform better. This is called constructive interference. This will be analyzed in the MeyGen project.

Size trends
Compared to wind turbines, the constraints are different for tidal energy. The water depth is typically between 20 and 50 meter, so that limits the rotor size. Also, the velocity shear (difference between the velocity at the bottom and free water surface) poses strong bending loads on the rotor.

Suitable sites
Tidal energy will never be able to provide the full energy demand of a country, but it can provide a valuable contribution. And it's also very useful for remote sites, which are typically islands, as they have difficult access to other sources of energy.


-----------------------------------------------------------------------------------------------------------

The AirShaper videos cover the basics of aerodynamics (aerodynamic drag, drag & lift coefficients, boundary layer theory, flow separation, reynolds number...), simulation aspects (computational fluid dynamics, CFD meshing, ...) and aerodynamic testing (wind tunnel testing, flow visualization, ...).

We then use those basics to explain the aerodynamics of (race) cars (aerodynamic efficiency of electric vehicles, aerodynamic drag, downforce, aero maps, formula one aerodynamics, ...), drones and airplanes (propellers, airfoils, electric aviation, eVTOLS, ...), motorcycles (wind buffeting, motogp aerodynamics, ...) and more!

For more information, visit www.airshaper.com

Tidal Stream Turbines - Renewable Energy from the Tides

- MV Agusta MotoGP - part 1: https://youtu.be/BUkaejkHRn8
- MV Agusta MotoGP - part 2: https://youtu.be/fGNdMaVft_w
- Motorbike wheelies: https://youtu.be/M2Z_GuLHmHs
- AMA superbikes: https://youtu.be/qrH_QBOrqbw
- Damon motorcycles - electric motorbikes: https://youtu.be/asYdvFfzHTs

For this video, we interviewed Derek Dorresteyn, CTO at Damon Motorcycles (https://www.damon.com).

Hyperdrive platform
The case of the battery is the main stressed component of the motorcycle, which means they got rid of the traditional motorcycle frame. It has reduced mass, the stiffness can be tuned and it provided more room.
Higher top speed, faster acceleration and far less rotating mass compared to an ICE motorcycle. It’s a really simple motor-gear reduction and this reduction in rotating mass significantly reduces the apparent mass, which otherwise adds forces that resist the vehicle leaning over and changing direction. Without the added forces, the bike feels much lighter in corner transitions.

Cooling
The electric drive train is far more efficient: 85% efficiency for the electric drivetrain versus 30% efficiency for the internal combustion engine. But the threshold temperature of a battery pack is much lower than that of a combustion engine, which again increases the required size of the radiator. Both more or less cancel each other out, meaning you end up with more or less the same radiator size.

Range
There is a lot of variability in the shape and size of the people that will be on the motorcycle, but also the way they interact with the bike can be different. For the optimization work in terms of drag, they have used a standard rider in terms of height and mass combined with the SAE J2982 drive cycle.

Aerodynamics
If they made a vehicle with the lowest possible drag coefficient, it likely would not be very commercial. So they have this tension between current design trends, consumer taste and the desire to improve aerodynamic performance.
The aerodynamics are extremely important: if they wouldn’t care at all, the range would just be half of the claimed 200 miles.

-----------------------------------------------------------------------------------------------------------

The AirShaper videos cover the basics of aerodynamics (aerodynamic drag, drag & lift coefficients, boundary layer theory, flow separation, reynolds number...), simulation aspects (computational fluid dynamics, CFD meshing, ...) and aerodynamic testing (wind tunnel testing, flow visualization, ...).

We then use those basics to explain the aerodynamics of (race) cars (aerodynamic efficiency of electric vehicles, aerodynamic drag, downforce, aero maps, formula one aerodynamics, ...), drones and airplanes (propellers, airfoils, electric aviation, eVTOLS, ...), motorcycles (wind buffeting, motogp aerodynamics, ...) and more!

For more information, visit www.airshaper.com

How to Design Electric Motorbikes - Interview with Damon Motorcycles

For more information about:
- F150 coastdown testing: https://youtu.be/Gip3iG2A2zE

In this video, we’re going to improve the aerodynamics of America’s favorite pickup truck, the Ford F150. As a starting point, we used a highly detailed 3D scan of the latest version of the 2021 Ford F150, here in its hybrid version, provided by our partner A2MAC1. We uploaded this 3D scan directly to AirShaper, set the speed to 30 meters per second, added rotating wheels and launched a Regular Simulation.
A few hours later, we could immediately spot a number of problem areas in terms of aerodynamics:
- Right after the cabin, the flow separates and turns into a turbulent wake. Some of the flow tumbles down into the cargo bed and then hits the tail gate on the way out, again creating a wake.
- The huge mirrors are without a doubt very practical and safe, but they also generate a lot of drag.
- The underbody features many exposed components. That’s very practical for easy access and quick repairs, but it strongly obstructs the flow underneath the car, which creates resistance.
- The wheels look nice, but with their sharp edges and large openings, they create what is called ventilation drag – it’s the energy lost on mixing the air, which translates into extra drag. Also, they create dirty air which affects the aerodynamics of the car further downstream.
- The ride height of the car is quite high – great for off-roading, but usually not so good for the aerodynamics of a car, as more air is now travelling through the channel between the underfloor and the ground. And with the underfloor being so exposed, this can also create extra drag.
The verdict: a drag coefficient of 0.463. Not very impressive compared to for example the 0.21 of the Lucid Air.

So, the guys at A2MAC1 went to work, drew inspiration from the Tesla Cybertruck and came up with a number of improvements – let’s check them out and see how much they each reduce drag!
- First, the rear cover: by closing the cargo bed, the flow across and around the cabin now stays largely attached to the rear cover. The air is prevented from hitting the tailgate and now nicely slides downward to help reduce the wake behind the car. This solution alone lowers the aerodynamic drag by a staggering 14.82%! Optimizing this global shape of the car helps to improve what is called the pressure recovery: at the front of the car, the air hits the nose and creates a high pressure, pushing the car backwards. Once it has crossed the mid-point of the car, the flow starts to contract again. If you support the airflow in the right way, this helps to create a forward push, compensating for some of the drag at the front. The base shape of the Tesla Cybertruck offers a lot of potential in this regard, although the sharp edges might need to be tuned a little.
- Secondly, the F150 was given a smooth underfloor. This is something we also saw when we analyzed the Porsche Taycan and which applies to many electric vehicles: because of the electric drivetrain, it’s possible to create an almost entirely flat underfloor. This facilitates a much smoother and less turbulent airflow which strongly lowers the drag. In this case, a very crude underfloor design already lowered drag by 3.4%!
- Then, the aerodynamic wheel covers of the Tesla model 3 were pulled from A2MAC1’s database and, after some stretching and scaling, applied to the Ford’s wheels. Without any optimization, these cut drag by 2.23%.
- Another tweak meant taking off the side view mirrors altogether. Yes, in reality you’d have to put camera’s instead, if regulations allow it at all. But just to show the potential, we ran a simulation and noted a 5.5% drop in aerodynamic drag!
- Last but not least, the truck was lowered by 2 inches – this is a trick some other cars already apply to lower drag at cruising speeds. In this case, it lowered drag by another 2%. 

When adding all of these modifications at the same time, we observed a total reduction in drag of 25.05% - that is a massive difference! If we assume that aerodynamics are responsible for roughly half of the energy or fuel consumption of a car, then this means a 12.5 reduction in fuel consumption, or roughly 12.5% more range!


-----------------------------------------------------------------------------------------------------------

The AirShaper videos cover the basics of aerodynamics (aerodynamic drag, drag & lift coefficients, boundary layer theory, flow separation, reynolds number...), simulation aspects (computational fluid dynamics, CFD meshing, ...) and aerodynamic testing (wind tunnel testing, flow visualization, ...).

We then use those basics to explain the aerodynamics of (race) cars (aerodynamic efficiency of electric vehicles, aerodynamic drag, downforce, aero maps, formula one aerodynamics, ...), drones and airplanes (propellers, airfoils, electric aviation, eVTOLS, ...), motorcycles (wind buffeting, motogp aerodynamics, ...) and more!

For more information, visit www.airshaper.com

Ford F150 Aerodynamics - How to add more than 10% range!

For more information, visit https://www.airshaper.com or email info@airshaper.com
----------------------------------------------------------------------------------------

In this video we’ll explain flow separation (or boundary layer separation).
In our previous video on boundary layers, we saw that the air close to the surface will stick to it, forming a boundary layer. But in some cases, it can become too difficult for the air to follow the curvature of the surface, causing it to detach. Let’s have a look at why this can happen.

When you look at air flowing across a flat plate, as in our previous video, the free stream velocity remains constant. Because of Bernoulli’s equation, which states that pressure changes with velocity, this means the pressure gradient in the X-direction is more or less to zero. Also, the pressure is imposed onto the boundary layer and so doesn’t really change much in the direction perpendicular to the surface either. That means the pressure gradient in both the x and y direction are close to zero.

But when the air flows around a curved surface, it speeds up and slows down. Take the flow around a cylinder for example: at the front, the pressure is the highest as the air comes to a complete standstill – this is the stagnation point. As the air then curves around the cylinder, it speeds up and this creates a drop in pressure. As the pressure goes down with increasing x, this is a negative pressure gradient, which is good: the air is being pushed downstream, overcoming the friction in the boundary layer, which gradually builds up.

The maximum velocity is reached somewhere around the midpoint of the cylinder, after which the air starts to slow down again. This means the pressure gradient is now reversed: as the air travels across the surface, the pressure goes up. This is what is called an “adverse” pressure gradient and the air is no longer driven but obstructed by the pressure difference: it has to flow into regions with higher pressure.

If the pressure gradient is positive or zero, as in the case of the flat plate, then the velocity gradient perpendicular to the wall is positive: the further you move away from the wall, the higher the velocity. But when the air travels against an adverse pressure gradient, the velocity profile is pushed back and this reduces the velocity gradient at the wall.

At some point, the velocity gradient becomes zero, which is the point where the flow separates or detaches. Beyond this point, the velocity profile is zero both at the wall, because of the no-slip condition, and at the inflection point, which is where the velocity crosses from negative into positive. In the negative velocity region, between the surface and the inflection point, the air flows in a direction opposite to the main flow  – this is called a recirculation. The separated flow will now surf on top of this recirculation flow and the boundary layer and wake will continue to grow, which can have a negative effect on drag.

In reality, shapes can be much more complex than cylinders and flows can detach and reattach in various locations. Even on a smooth wing, the flow can detach and reattach. So, this is very much a 3-dimensional challenge and getting it right requires a careful analysis and optimization of local and global pressure gradients.

Also, keep in mind that flow separation can happen in both laminar and turbulent boundary layers. Turbulent boundary layers, however, carry more momentum and will typically stay attached to the surface further downstream compared to a laminar one – just check our video on golf ball dimples or vortex generators to learn more.

Flow Separation - Boundary layer separation explained

For more information, visit https://www.airshaper.com or email info@airshaper.com
----------------------------------------------------------------------------------------
In this video, we explain what a boundary layer is.

Let’s look at two extremes first:
No-slip condition: no matter how smooth the surface is, the flow will always stick to it, having a flow velocity of zero on the surface of the object.
Free stream velocity: the velocity of the undisturbed air, far away from the object

To understand what happens in between these two extremes, let’s look at air flowing across a flat plate. As the undisturbed air meets the leading edge of the plate, it will locally stick to it because of the no-slip condition. As the air moves or slides across the plate, this layer or air sticking to it will grow thicker. This region, where the air moves slower than the free stream velocity, is called the boundary layer and it is mainly determined by the viscous forces. Outside of the boundary layer, the viscous forces are much less important and sometimes even ignored in fluid modeling

You can have a laminar boundary layer, with layers of air moving parallel to the surface, or a turbulent boundary layer, filled with vortices featuring velocity variations in all directions.
If the incoming airflow is laminar itself and if there are no disturbances that can trip the flow, the boundary layer will start off as laminar. As the air continues to move along the plate, so does the distance traveled from the leading edge. If we use this distance to calculate the Reynolds number, it’s clear that the Reynolds number will continue to rise as we move further down the plate. 

When the Reynolds number crosses a critical value, which can be different for each application or geometry, the laminar boundary layer will start to transition into a turbulent one. The velocity profile of both types is very different and can have a large impact on the total friction & pressure force on the object.

In applications like airplanes, where shapes are smooth and streamlined, the goal is to maintain a laminar boundary layer as long as possible: a turbulent boundary layer increases the friction drag and typically is thicker than a laminar one, which increases the effective thickness and thus the pressure drag of the wing.

In applications with less streamlined shapes, like sports athletes or golf balls, it can be beneficial to trip the boundary layer and make it turbulent: turbulent boundary layers carry more momentum and will push the separation point further downstream – that’s the reason golf balls have dimples and athletes wear special suits or even shin tape, reducing the wake they leave behind.

What is a Boundary Layer - Laminar and Turbulent boundary layers explained

The boundary layer is a thin region of slow-moving air next to the surface of an aerodynamic body and can be the main source of friction drag.

What is a Boundary layer?

To learn more about:
- Boundary layers: https://youtu.be/TwOxa9rAOfE
- Flow separation: https://youtu.be/HfICmKv9SzE
- Golf ball aerodynamics: https://youtu.be/KzXcglAhxps

In this video, we'll explain the aerodynamics behind curved free kicks, corners, and so on! Want to bend it like Beckham? Then stay tuned!

When you kick the ball off center, you'll give it a spin around the vertical axis. At one side, the surface is spinning against the airflow, so the relative velocity between the surface and the air is increased. This slows down the air and also moves the separation point upstream.

At the other side, the surface is spinning in the same direction as the airflow, reducing the relative velocity. This reduces friction and boundary layer buildup and pushing the separation point further downstream.

First way to explain the force - Bernoulli:
the faster moving air will feature lower pressure, creating a suction effect bigger than on the other side. This generates a net lateral force.

Second way to explain the force - momentum:
the wake behind the soccer ball is pulled towards the side of the forward spinning side. Because of Newton's law of action and reaction, this generates an equal but opposite effect on the ball.

The effect described in this video is called the magnus effect.


-----------------------------------------------------------------------------------------------------------

The AirShaper videos cover the basics of aerodynamics (aerodynamic drag, drag & lift coefficients, boundary layer theory, flow separation, reynolds number...), simulation aspects (computational fluid dynamics, CFD meshing, ...) and aerodynamic testing (wind tunnel testing, flow visualization, ...).

We then use those basics to explain the aerodynamics of (race) cars (aerodynamic efficiency of electric vehicles, aerodynamic drag, downforce, aero maps, formula one aerodynamics, ...), drones and airplanes (propellers, airfoils, electric aviation, eVTOLS, ...), motorcycles (wind buffeting, motogp aerodynamics, ...) and more!

For more information, visit www.airshaper.com

Soccer aerodynamics - How to curve a ball

Part 1: https://youtu.be/K2jgsgihZ_w
Tuft testing on a Vokswagen Beetle: https://youtu.be/CGmqdXKgBiQ

In this second video on the Porsche Taycan Aerodynamics, we’ll take the car to the street to compare tuft movement to simulation data.
Special thanks to the Porsche Centre in Mechelen, Belgium, for letting us play with this fantastic car 😊
https://www.porsche.com/belgium-mechelen

In our last video, we had equipped the Porsche Taycan with a number of tuft

Rear window & spoiler

On the rear window and spoiler, none of the tufts move. This is because the flow in that area stays nicely attached & laminar. There is some slow-down of the flow due to the “negative” curvature of the window, but this doesn’t lead to separation.

Rear wheels
The rear wheels create a turbulent zone which extends downwind (causing a lot of movement of the tufts on the rear bumper) and also upward (creating movement above the wheel arch as well). The latter extends all the way to the rear spoiler.

A-pillar and side window
As the air makes its way from the front wind shield to the side window, it curves around the A-pillar. In doing so, a vortex is created which locally pushes the air upward, just behind the A-pillar. This upward motion extends further downwind at the top of the side window as well.

Front wheels
The front wheels also create a turbulent zone which splashes out upward. Further downstream, this zone is countered by the air coming off the front hood, which creates a zone of attached air curving downward on the flanks.

Headlamps
The headlamps are located in a “pocket” in the bodywork. As the air jumps out of this pocket, it locally separates from the bodywork and leaves behind a local bubble of separated, turbulent flow. This can be seen in both simulation & tuft movement.


-----------------------------------------------------------------------------------------------------------

The AirShaper videos cover the basics of aerodynamics (aerodynamic drag, drag & lift coefficients, boundary layer theory, flow separation, reynolds number...), simulation aspects (computational fluid dynamics, CFD meshing, ...) and aerodynamic testing (wind tunnel testing, flow visualization, ...).

We then use those basics to explain the aerodynamics of (race) cars (aerodynamic efficiency of electric vehicles, aerodynamic drag, downforce, aero maps, formula one aerodynamics, ...), drones and airplanes (propellers, airfoils, electric aviation, eVTOLS, ...), motorcycles (wind buffeting, motogp aerodynamics, ...) and more!

For more information, visit www.airshaper.com

Our Latest Videos Gain aerodynamic insight by watching our videos