Numerical Study on Aerodynamic Drag Reduction on a Rear Wing of a Formula Student Car

(1)

SA118X DEGREE PROJECT IN TECHNOLOGY, FIRST CYCLE, 15 CREDITS

STOCKHOLM, SWEDEN 2021

Numerical Study on Aerodynamic Drag Reduction on a Rear Wing of a Formula Student Car

MAHIM AHSAN

KTH ROYAL INSTITUTE OF TECHNOLOGY SCHOOL OF ENGINEERING SCIENCES

(2)

Abstract

The importance of aerodynamics in racing has increased substantially over the past decades.

Racing vehicles utilize different aerodynamic devices, to redirect the airflow around the vehicle in a beneficial way. The increase in downforce using an aerodynamic wing package increases drag force. The drag force limits the top speed and the maximum acceleration of the vehicle and is an unwanted effect of increased downforce. To combat this, top-performance vehicles utilize a drag reduction system (DRS). The DRS is an active system of the vehicle, which allows adjustment of the angle of attack (AoA) of wings on the vehicle while driving on track when deemed beneficial. This thesis will investigate the possibility of adding a DRS to the rear wing of a Formula Student vehicle. All tests were done by simulations, no physical tests were done in this thesis.

The investigation of a DRS was done in three stages. The first step was to create a 3-dimensional CAD-model that complied with the Formula Student Germany rules. The design of the rear wing was based on the current rear wing on the KTH Formula Student vehicle. The current rear wing was slightly redesigned for this project to facilitate the investigation of a DRS. The second step was to create a mesh model. A mesh is a representation of a geometry, it defines the physical shape of an object, in this case the rear wing. The more detailed a mesh is, with a higher number of cells, the more accurate the results will be from the simulations. However, a finer mesh with a high number of cells increases the simulation time significantly. Therefore, it is important to find a good balance between the number of cells and the time it takes to make a simulation. After a mesh was generated, physics models for the simulation were determined.

The final step was to make simulations based on the physics models chosen. For this project, Siemens NX was used to design a three-dimensional model of the wing and Siemens STAR- CCM+ was used to make the simulations and meshes.

The resulting reduction in drag was calculated to 78 per cent between the closed configuration and the open configuration with the least drag. However, there is a large amount of separation between the airflow and the rear wing on all configurations simulated except at the AoA -10^°. When the airflow separates downforce decreases and drag increases and is not desirable.

The drag on the rear wing can be decreased by utilizing a DRS as observed from the results.

The drag reduced by the DRS at 15^° and -10^° AoA and a DRS would certainly be beneficial.

Further investigation of computational fluid dynamics and aerodynamics is recommended to complement the result. There are uncertainties in the results due to inadequacy in the mesh and design of the rear wing. By addressing these uncertainties better results can be achieved.

(3)

Sammanfattning

Betydelsen av aerodynamik i racing har ökat markant de senaste decennierna. Racingbilar använder olika aerodynamiska enheter i syfte att omdirigera luftflödet runt fordonet på ett fördelaktigt sätt. Ökningen av lyftkraft med hjälp av ett aerodynamiskt vingpaket resulterar i en ökning av luftmotstånd. Luftmotståndet som uppstår begränsar topphastigheten och maximal acceleration av fordonet och är en oönskad effekt av ökad lyftkraft. För att motverka detta använder racingbilar ett drag reduction system (DRS). DRS är ett aktivt system i fordonet som gör det möjligt att justera attackvinkeln (AoA) på fordonet under körning på banor, när det anses fördelaktigt. Denna rapport kommer att undersöka möjligheten till att använda en DRS till den bakre vingen av ett Formula Student fordon. Alla tester utfördes genom simuleringar, inga fysiska tester utfördes i denna rapport.

Undersökningen av en DRS utfördes i tre steg. Det första steget var att skapa en tredimensionell CAD-modell som uppfyllde reglerna för Formula Student Tyskland. Designen av bakvingen baserades på den nuvarande bakvingen på KTH Formula Student fordonet. Den nuvarande bakvingen ändrades för detta projekt för att underlätta utredningen av en DRS. Det andra steget var att skapa en mesh. En mesh representerar en geometri, det definierar den fysiska formen på ett föremål, i detta fall den bakre vingen. Ju mer detaljerat meshen är, med fler antal celler, desto bättre blir resultaten från simuleringarna. En finare mesh med många celler ökar dock simuleringstiden avsevärt. Därför är det viktigt att hitta en bra balans mellan antalet celler och den tid det tar att utföra en simulering. Efter att en mesh genererats bestämdes fysikmodeller för simuleringen. Det sista steget var att göra simuleringar baserat på de valda modellerna. För detta projekt användes Siemens NX för att designa en tredimensionell modell av vingen och Siemens STAR-CCM+ användes för att generera en mesh och utföra simuleringar.

Den resulterande reduceringen av luftmotstånd beräknades till 78 procent mellan den stängda konfigurationen och den öppna konfigurationen med minst luftmotstånd. Det sker dock en stor separation mellan luftflödet och den bakre vingen på alla simulerade konfigurationer utom vid attackvinkel -10^°. När luftflödet separeras reduceras lyftkraften och luftmotståndet ökar och är inte önskvärt.

Luftmotståndet på den bakre vingen kan reduceras genom att använda en DRS, vilket resultaten påvisar. Reduktionen av luftmotståndet vid 15^° and -10^° AoA är signifikant och en DRS skulle vara fördelaktig. Ytterligare undersökning av CFD och aerodynamik rekommenderas för att komplettera resultaten. Det råder osäkerhet i resultaten på grund av brist på meshens kvalitet och design på bakvingen. Genom att ta itu med dessa osäkerheter kan bättre resultat uppnås.

(4)

Acknowledgements

I would like to thank my supervisor Malte Rothhämel for sharing his guidance and experience and valuable discussions throughout the course of this project. I would also like to thank the head of KTH Formula Student, Tamas Vass, for handing me this project. I would like to thank Emelie Trigell for presenting the basics of STAR-CCM+. Lastly, I would like to thank Guillaume Defromont for sharing his invaluable expertise in NX and STAR-CCM+.

(5)

List of Figures

FIGURE 1:BASIC ILLUSTRATION OF THE DIFFERENCE BETWEEN A RACE CAR AIRFOIL AND AN

AEROPLANE AIRFOIL ... -3-

FIGURE 2:DIFFERENT PARTS OF THE REAR WING ... -3-

FIGURE 3:EXISTING DRS SOLUTIONS:(A) PUSH-UP TYPE,(B) POD-ROCKER TYPE AND (C) POD-PULL TYPE [6] ... -4-

FIGURE 4:THE CURRENT REAR WING AND THE NEW DESIGN ... -7-

FIGURE 5:DESIGN CHANGE OF THE SECOND WING FLAP ... -7-

FIGURE 6:PRISM LAYERS ... -8-

FIGURE 7:MESH OF THE CLOSED CONFIGURATION ... -8-

FIGURE 8:SIMPLIFIED ILLUSTRATION OF THE WORK PROCESS ... -9-

FIGURE 9:SEPARATION OF THE AIRFLOW, REAR SIDE, CLOSED CONFIGURATION AOA-44^° ... -10-

FIGURE 10:SEPARATION OF THE AIRFLOW, FRONT SIDE, OPEN CONFIGURATION AOA-10^° ... -10-

FIGURE 11:SEPARATION OF THE AIRFLOW, FRONT SIDE, OPEN CONFIGURATION AOA15° ... -11-

FIGURE 12:ANGLE DIFFERENCE BETWEEN THE MIDDLE CORD AND THE OUTER CORD OF THE SECOND WING FLAP ... -12-

FIGURE 13:VORTICES GENERATED BY THE REAR WING AT AOA-10° ... -13-

(7)

List of Tables

TABLE 1:CD AND CL FOR DIFFERENT AOA AND THE REDUCTION IN DRAG ... -10-

(8)

Notations

Cd Coefficient of Drag [*]

Cl Coefficient of Lift [*]

Abbreviations

AoA Angle of Attack

CAD Computer-aided Design

CFD Computational Fluid Dynamics DRS Drag Reduction System

FS Formula Student

FSG Formula Student Germany

KTH Kungliga Tekniska Högskolan (The Royal Institute of Technology)

(9)

- 1 -

1 Introduction

The importance of aerodynamics in racing has increased substantially over the past decades.

The main purpose is to increase the normal load on the tires for increased grip without adding any mass to the car. These aerodynamic advantages have improved considerably with the progress of technology, requiring rules and regulations to scale down these advantages to reasonable levels.

Formula Student is an engineering competition held annually in different locations. Student teams from around the world design, manufacture, and test a formula-style racing car. The car must be based on a series of rules, to ensure safety and promote thinking outside the box for clever solutions. During the competition, the cars are judged on different criteria. One of the criteria will test the endurance of the vehicle around a track. To achieve good lap times the aerodynamics is important.

1.1 Background

Racing vehicles utilize different aerodynamic devices, to redirect the airflow around the vehicle in a beneficial way. These devices consist of an assortment of wing profiles in different shapes and placements all around the vehicle, to create a negative lift force on the vehicle (more commonly known as downforce). An increase in downforce results in a larger normal force between the tyres and the road, therefore increasing traction, which results in faster lap times around a track. The aerodynamic devices which contribute most to increased downforce on a KTH Formula Student vehicle are the undertray, the front wing, and the rear wing.

1.2 Problem

The increase in downforce using an aerodynamic wing package increases drag force. The drag force limits the top speed and the maximum acceleration of the vehicle and is an unwanted effect of increased downforce. To combat this, top-performance vehicles utilize a drag reduction system (DRS). The DRS is an active system of the vehicle, which allows adjustment of the angle of attack (AoA) of wings on the vehicle while driving on track when deemed beneficial. When the car is moving in a straight-line a DRS would be beneficial since downforce is not as necessary. Therefore, a DRS could be utilized to reduce drag when the car is moving in a straight-line and increase top speed and maximum acceleration. But when entering a corner, maximum downforce is vital to attain maximum grip which allows the car to break in a shorter distance and travel faster around the corner. KTH FS would like to investigate the possibility of adding a DRS to the rear wing of the vehicle.

1.3 Purpose

The purpose of this bachelor thesis is to investigate if it is possible to reduce the drag force on the rear wing and if so, get an impression of how much the drag can be decreased. If the decrease in drag is of significance a DRS would be beneficial. The increase in performance will help KTH FS achieve better results in competitions. Furthermore, the purpose is also to

(10)

- 2 -

fulfil all learning objectives set by the examiner to receive a bachelor’s degree at The Royal Institute of Technology.

1.4 Delimitations

The project will only investigate the possibility of implementing a DRS on the rear wing and no other parts of the vehicle. The reason being KTH FS only wants to investigate the rear wing. Only a suggestion of a DRS will be presented and not a finished physical rear wing. The cost of the suggested DRS should not exceed 5000 SEK and should ideally be under 4000 SEK.

The overall weight for the DRS should not exceed two kilograms and should ideally be under one kilogram. The DRS must be able to reliably operate (open and close) at the maximum vehicle speed of 150 kilometres per hour. Furthermore, the DRS must comply with all regulations of the Formula Student Class, Formula Student Germany (FSG 2020) [1]. All tests will be done by simulations, no physical tests will be done. The suggested rear wing design with a DRS will be based on the existing static rear wing on the current KTH FS vehicle, a new rear wing will not be designed from the ground up. Since the current rear wings design has been optimized the simulations for this study will only optimize the operation of the DRS and at what AoA the coefficient of drag (Cd) is the least. The type of actuator will be suggested but how it is operated will not be discussed in this thesis. How the DRS will be controlled is not a part of this thesis. The reason for this is other subgroups of KTH FS need to be involved in case electronics and hydraulics is used to operate the DRS.

2 Literature Study

A literature study was conducted before the start of the project to get an insight into past and modern solutions for DRS but also to learn how a rear wing function. There are several aspects to take into consideration when designing a DRS, such as actuation types, materials, and computational tools, to name a few. The research found in this area has either been performed physically or with simulations. Since no tests will be done physically, i.e., in a wind tunnel, the research done in simulations has been evaluated more thoroughly. Furthermore, to design the rear wing and make simulations, computer tools will be needed. Siemens NX [2] will be used for designing the rear wing since it is used by KTH FS. There are different tools for making simulation which will be discussed in chapter 2.2.

2.1 How a rear wing function

The main purpose of a rear wing on a race car is to generate negative lift (downforce), the opposite to aeroplanes where the wings generate positive lift. Figure 1 illustrates an example of how an airfoil can be shaped and the main difference between a race car airfoil and an aeroplane airfoil. The two airfoils in Figure 1 are mirror images of each other, the purpose of the figure is only to illustrate the differences. Positive lift is generated when the pressure above the airfoil is lowered. The cause of lift is the introduction of a shape into an airflow, which curves the streamlines and introduces changes in pressure, lower pressure on the upper surface and higher pressure on the lower surface [3]. It is the curvature of the airfoil that creates lift and not the difference in distance between the upper side and the lower side [3]. In theory, by flipping an aeroplane airfoil vertically it can be utilized as a rear wing of a race car to generate downforce.

(11)

- 3 -

Figure 1: Basic illustration of the difference between a race car airfoil and an aeroplane airfoil

By combining different shapes and sizes of airfoils the downforce on the rear wing can be optimized. Figure 2 illustrates the rear wing design utilized for this thesis and identifies different parts of the rear wing. The purpose of endplates is to manage the airflow around the wing elements and decrease drag. By controlling how the air comes off the wing and how it is directed the drag can be decreased. The curvature on the first wing element is to avoid the odd vortices of air (“dirty air”) coming off the driver’s head which is positioned in front of the rear wing.

Figure 2: Different parts of the rear wing

2.2 Actuators

Actuators are mechanical devices that convert energy into motion. The motion includes lifting, blocking, ejecting, and clamping. The motion can be linear (one or both directions), and circular. Three different actuator types are mainly used in the industry, hydraulic, pneumatic, and electric actuators [4]. All three have their pros and cons. All three actuators convert some form of stored energy into motion. The main difference between them is the power they can

Airflow Direction

Negative lift (downforce)

Positive lift

Racecar airfoil

Aeroplane airfoil

Endplates

Second wing element

First wing element Angle of Attack

(AoA)

(12)

- 4 -

handle and their ability to convert energy to physical work varies. A hydraulic actuator is superior to pneumatic and electric actuators handling power and has a longer life cycle [4]. The biggest concern with a hydraulic actuator is leakage, but with proper maintenance, the risk of leakage reduces significantly. If the current KTH FS vehicle utilizes a hydraulic system, the same one can be used for operating a DRS thus saving weight.

Deployment of rear wing flap

Three types of actuators that are/were most common on DRSs for race cars, push-up, pod- rocker, and pod-pull [5]. Figure 3 displays the three different types of actuators in the closed configuration and the dashed grey line shows the open configuration. In the push-up solution, the packaging sits under the rear wing element. This solution is unaffected by the removal of the rear wing. The push-up solution was disadvantageous since the mechanism interfered with the underside of the wings surface, disturbing the flow of the air.

The pod-rocker solution moved the actuator to the top of the rear wing element. If hydraulics is used, it will require the hydraulic connectors to be disconnected if the rear wing needs to be removed, which is never ideal to do. With this solution, the wing will open fractions faster than the push-up solution and it removes the interference with the underside of the wing.

The pod-pull is the most common solution today. The mechanism has been simplified and the pod has been reduced in size. Since the pod-pull does not have the same mechanical advantage as the pod-rocker, this system presumably requires more hydraulic effort for the same speed of opening.

Figure 3: Existing DRS solutions: (a) push-up type, (b) pod-rocker type and (c) pod-pull type [6]

2.3 Simulation Tools

There are numerous simulation tools to choose from. Siemens’ and Ansys’ licenses are available through KTH FS and KTH, therefore these two will be considered for this project.

Siemens and Ansys both offer computational fluid dynamics (CFD) software, both commonly used. The purpose of simulation software is to predict how product designs will behave in real- world environments.

(13)

- 5 -

STAR-CCM+ [7] is Siemens’ CFD solver and Fluent [8] is one of Ansys’ CFD solver and is used to predict fluid flow and other related phenomena. Each software requires CAD-models to mesh. A mesh is a representation of a geometry, it defines the physical shape of an object, in this case the rear wing. A mesh consists of vertices, edges, and faces. Creating a high-quality mesh is important to obtain accurate results of the simulations. Generally, a smaller mesh size provides a more accurate solution. The trade-off is with higher quality meshes the computational cost will increase significantly and thus solve times are extended. A high-quality mesh is an optimal balance between the mesh size and the computational cost needed.

Since NX is the design tool used in this project, which is also provided by Siemens, using STAR-CCM+ might be an advantage since both software is from the same company.

Furthermore, the student licence for Ansys Fluent limits the number of nodes that one can have in a mesh, which is not favourable. Another CFD solver from Ansys is Discovery [9].

Discovery is Ansys’ next-generation design software. Discovery is the first simulation-driven design tool to combine instant physics simulation [9]. The result is a near-real-time simulation system for both fluid and static structural problems. Discovery has discarded the need for a mesh, which is required both in Fluent and STAR-CCM+.

2.4 Summary

For this project, a hydraulic actuator is favourable because of its superiority to pneumatic and electric actuators handling power. The pod-pull solution will be suggested for deploying the DRS because of its mechanical simplification and advantages in size compared to the other solutions. STAR-CCM+ and NX are the software used by KTH FS and will therefore be used for this thesis. Discovery is relatively new and has not been tested to the same extent as Fluent or STAR-CCM+ and given the short time frame for this thesis, it would be a risk to use such a new tool.

3 Sustainability

One of KTH FS’s core values is sustainability, which pervades every aspect of development.

Every year the team strives to improve the design of the car, which means components of the car will acquire a new design. This results in components being wasted due to only being used for one year. Utilizing a part for a single year is not sustainable and therefore material selection is exceedingly critical. Several components of the vehicle are made of carbon fibre due to its physical properties. However, carbon fibre is not biodegradable and thus cannot be recycled.

95 per cent of carbon fibre ends up in landfills [10] and since it is not biodegradable it will stay there until a recycling solution, or an alternative has been discovered. The current rear wing is made of carbon fibre and next year’s rear wing with a DRS will also be made of carbon fibre.

Only the first wing element from the current rear wing will be able to be reused on the next year’s rear wing since the other elements have been redesigned. Ycom together with Bcomp Ltd. has developed a sustainable alternative to carbon fibre [11]. The natural fibre developed by the companies is still 40 per cent heavier than its carbon fibre counterpart but enables a 50 per cent reduction of carbon dioxide on the composite site [11]. It is still not on par with carbon fibre but if the natural fibre can reduce its weight, it will be a good alternative to carbon fibre.

(14)

- 6 -

However, the use of carbon fibre instead of metals on vehicles such as automobiles or aircraft significantly reduces the weight of the vehicle. This results in improved fuel economy (also for electrically powered cars) which also results in a reduction of emissions. Less drag also contributes to better fuel economy and less electricity needed to power the vehicle. KTH FS’

vehicle is electrically powered and a lighter car with less drag will reduce the electricity needed considerably. If less electricity is needed a smaller battery will achieve the same range as a heavier car with a larger battery and worse drag. A smaller battery would further reduce the weight of the car, improving the range even more, and have a smaller impact on the environment since constructing a battery produces a lot of emissions.

One of the main motivations of racing is to transfer the technology developed to regular road cars. The development of race technology is expensive and not sharing these technologies would be a huge waste of resources. Regular car companies cannot afford the level of development in racing, therefore utilizing material such as carbon fibre can be justified.

Developing cheaper ways to produce carbon fibre and normalizing it for road cars could be beneficial since using carbon fibre on-road cars would reduce their weight and consequently reduce fuel consumption. The research done in this thesis will not be of much help for regular cars since they do not require a rear wing, the addition of a rear wing would increase the drag.

4 Method

The investigation of a DRS was done in three stages. The first step was to create a 3-dimensional CAD-model that complied with the FSG rules. The rules complied with is found in Appendix A. The second step was to create a high-quality mesh model, which included an enclosure that defined the simulation field. Lastly, simulations were executed, which provided the coefficient of lift and drag on the rear wing. For this project, Siemens NX was used to design a three- dimensional model of the wing and Siemens STAR-CCM+ was used to make the simulations and meshes.

4.1 CAD-Model

The design of the rear wing is based on the current rear wing on the KTH FS vehicle. The existing rear wing, displayed to the left in Figure 4, consists of three wing elements.

Implementing a DRS on both the second and third wing element would help decrease drag and retain the amount of downforce when the DRS is deactivated. The gain would not be great since the third element only contributes to 16 per cent of the overall rear wing drag [12]. Creating a DRS for the two upper wing elements was considered too complicated. Therefore, the rear wing was redesigned to consist of only two wing elements, displayed to the right in Figure 4.

(15)

- 7 -

Figure 4: The current rear wing and the new design

The endplates were reduced in size but kept a similar shape to the previous endplate. The surface of the second wing element was enlarged to compensate for the lost downforce from the discarded wing element when the DRS is deactivated. The difference between the previous and current second wing elements is displayed in Figure 5. The design of the rear wing complies with the FSG rules.

Figure 5: Design change of the second wing flap

4.2 Mesh

Before a mesh was generated, an enclosure was formed where a fluid region was defined.

The dimensioning of the enclosure was performed according to the rule of thumb that each side of the enclosure should be 20 times larger than the width, height, and depth of the rear wing [12]. Each side of the enclosure was assigned a region with different boundary conditions.

Since the rear wing is axially symmetrical, it was enough to make simulations on half the rear wing. Generating a mesh for only half of the rear wing reduced the number of cells of the mesh significantly, which consequently reduced the simulation time. Furthermore, a surface wrapper was required before generating a mesh [12]. A surface wrapper can be utilized when a CAD- model is of poor quality. Creating a fine CAD-model is difficult and a surface wrapper is often necessary. The resulting surface quality from the surface wrapper is not excellent and is therefore used together with the surface remesher, providing a higher quality starting surface.

Current

New

Current New

(16)

- 8 -

The mesh operation is automated. The user must specify what meshers are to be used, the base size for the cells, and other parameters depending on what meshers have been chosen and custom controls for specific parts if desired [12],[13]. Prism layers were necessary to improve the accuracy of the flow solution. The prism layers allowed the simulation solver to resolve near-wall flow accurately, which was critical to determining forces and flow separation. The prism layers are displayed in Figure 6. After these parameters were determined a mesh could be generated. Every time the AoA was changed a new mesh was required. In Figure 7 below, the resulting mesh of the closed configuration is displayed.

Figure 6: Prism Layers

Figure 7: Mesh of the closed configuration

The darker areas in Figure 7 are finer cells. Finer cells were required on the edges and curves of the wing to get a smooth surface which improves the accuracy of the simulations.

(17)

- 9 -

Mesh Independence study

A mesh independence study was suggested [12] and was therefore conducted to determine that the solutions would fully converge and were not dependent on the mesh resolution. The study was done on the first simulation, which was the closed configuration. The base cell size for the first mesh was initiated with a larger size and was then uniformly refined while observing how the calculated Cd changed with decreasing number of cells. The base size with the minimum number of cells that returned acceptable results was then used for the remaining simulations. A mesh independence study should ideally provide results accurate enough while minimizing the computation time. The number of cells for each mesh was around five million which was reasonable, and feasible given the computational power available (six cores with 16 GB RAM).

4.3 Simulation

After a mesh was generated, physics models for the simulation were determined. The fluid chosen was air. The velocity of the air was set to 42 meters per second. If the airflow is below 100 meters per second, the fluid can be regarded as incompressible and the density of the air constant [14]. The airflow was chosen to be turbulent. Menter’s SST k-omega turbulence model was used in this project. SST k-omega is commonly used for CFD and is a two-equation model which provides general descriptions of turbulence. The model is well suited for simulating flow near-wall regions [15]. Lastly, reports and plots for CD, Cl, and scalar scenes displaying separation and velocity behaviour were created. In Figure 8 below, a simplified illustration of the work process is shown, from the beginning to the end of one simulation. Each simulation was iterated 1000 times to make sure that the solution had converged. After a simulation was done the same process was repeated but with a different AoA for the rear wing.

Each simulation took between two to three hours. A more detailed description of the work process and different parameters chosen in STAR-CCM+ can be found in Appendix B.

Figure 8: Simplified illustration of the work process

5 Results

The result from the simulations is presented in this chapter. Nine simulations were done in total and the relevant results will be presented. Statistics of the drag and lift coefficients are shown in Table 1. Zero-degree AoA is when the angle is parallel to the ground.

(18)

- 10 -

Table 1: Cd and Cl for different AoA and the reduction in drag

-44^°

(closed) -10^° -5^° -3° 0^° 3^° 5^° 10^° 15^°

Cd 0.808 0.365 0.305 0.285 0.262 0.237 0.224 0.195 0.173

Cl -2.356 -1.634 -1.478 -1.408 -1.318 -1.226 -1.176 -1.061 -0.955 Drag

Redu ction

54% 62% 65% 68% 71% 72% 76% 78%

The resulting reduction in drag is also shown in Table 1 for all the open configurations. Figures 9, 10, and 11 display the wall shear stress on the lefthand side and the velocity streamline contours of the airflow on the righthand side. The non-red areas shown in Figure 9, 10, and 11 is where the separation between the air and the rear wing occurs. The contours displayed on the right-hand side are cut through in the symmetry plane of the wing and the red circles indicate where the separation occurs.

Figure 9: Separation of the airflow, rear side, closed configuration AoA -44^°

Figure 10: Separation of the airflow, front side, open configuration AoA -10^°

(19)

- 11 -

Figure 11: Separation of the airflow, front side, open configuration AoA 15°

Additional results for the closed configuration and the open configurations with AoA -10° and 15° can be found in Appendix C, D and E, respectively.

6 Discussion

The resulting reduction in drag on the rear wing was calculated to 78 per cent between the closed configuration and the configuration with the lowest drag. However, there is a large amount of separation between the airflow and the rear wing on all configurations simulated except at AoA -10^°. Separation essentially means that the rear wing is not working properly and is, therefore, less effective. When the airflow separates downforce decreases and drag increases and is not desirable. The reduction of 78 per cent in drag is presumably inaccurate due to the separation, and the number is quite high

.

In Figure 9 the airflow separates on the backside of the second wing element reducing the downforce and increasing the drag. This is due to the high AoA and reducing the AoA will likely eliminate the separation for the closed configuration. For the open configurations, the issue is the angle difference between the middle cord and the outer cord of the second wing element shown in Figure 11. Since the angle is higher towards the middle separation begins from the middle and as the AoA is increased the separation increases as well, extending outwards. Again, the separation will result in reduced downforce and an increase in drag, which is not the goal for a DRS. Due to the lack of time, further simulations with these changes has not been completed.

In the one configuration where the airflow did not separate, shown in Figure 10, the drag was reduced by 55 per cent. Reducing the drag by 55 per cent is acceptable and this configuration can be suggested for use until the separation issue has been corrected and new simulations have been completed. Adjusting the angle difference displayed in Figure 12 from 14° to 5°, which is the same angle difference for the cord of the first wing element, the separation should be eliminated, and the drag should be reduced further.

In Figure 13, vortices created by the wingtips are shown. The wingtip vortices produce drag due to slowing the air down behind the wing, therefore slowing the car down. To combat this, louvres can be placed on the front of the endplates which allows high-pressure air inside the endplates to mix with the low-pressure air outside the endplates, reducing the pressure difference. Furthermore, the vortices generated by the wing elements and the vortices generated

(20)

- 12 -

behind the wing elements, the vortices can be neutralized and reduced even more since it would allow more of the air to mix.

The mesh generated gave good enough results since the results converged. Refining the mesh further would mean a significant increase in simulation time which was not feasible with the timeframe given and the computer power available. When the cell number was increased from 5 million to 15 million, the simulation time increased from three hours to 10 hours. Performing nine simulations that would take ten hours each was not an option.

The development for a DRS was done independently from other projects in KTH FS. Results gathered in this paper are only from simulations on the rear wing alone and further simulations should be done with the rest of the vehicle to include aerodynamic effects on the rear wing from other parts of the vehicle. Furthermore, since electronics and hydraulics might be involved, this project will need to be discussed and cooperated with other subgroups in KTH FS to be sure that everything works well with other components of the vehicle and is rule compliant.

Further investigation of CFD and STAR-CCM+ is recommended to complement the result.

The amount of knowledge required for acquiring good results was underestimated. A more extensive literature study should have been conducted, notably about STAR-CCM+ and how the simulation tool could be better utilized.

Figure 12: Angle difference between the middle cord and the outer cord of the second wing flap

(21)

- 13 -

Figure 13: Vortices generated by the rear wing at AoA -10°

7 Conclusion and Future work

The drag on the rear wing can be decreased by utilizing a drag reduction system (DRS) as observed from the results. The drag reduced by the DRS at 15^° and -10^° angle of attack is of significance and a DRS would certainly be beneficial. Further investigation of computational fluid dynamics and aerodynamics is recommended to complement the result. There are uncertainties in the results due to inadequacy in the mesh and design of the rear wing. By addressing these uncertainties better results can be achieved. Furthermore, this thesis only investigated the rear wing, independently from the rest of the KTH FS vehicle.

Therefore, it is suggested to perform simulations with the rest of the KTH FS vehicle to attain a better understanding of how the vehicle might affect the rear wing and make changes accordingly. The delimitations concerning the cost and weight of the DRS has not been evaluated in this thesis due to the lack of time and are advocated for future work.

(22)

- 14 -

References

[1] Formula Student Germany. Formula Student Rules 2020 [www]. Retrieved from https://www.formulastudent.de/fileadmin/user_upload/all/2020/rules/FS-

Rules_2020_V1.0.pdf . Published 13 September 2020. Read 4 May 2021.

[2] Siemens. NX [www]. Retrieved from

https://www.plm.automation.siemens.com/global/en/products/nx/ . [No date]. Read 4 May 2021.

[3] University of Cambridge. How wings really work [www]. Retrieved from

https://www.cam.ac.uk/research/news/how-wings-really-work . Published 25 January 2012.

Read 31 May 2021.

[4] York Precision. Hydraulic vs. Pneumatic vs. electric Actuators [www]. Retrieved from https://yorkpmh.com/resources/hydraulic-vs-pneumatic-vs-electric-actuators/ . [No date].

Read 4 May 2021.

[5] Motori Online. DRS activation [www]. Retrieved from

http://forum.motorionline.com/topic/25080-drs-activation/ . Published 31 July 2008. Read 4 May 2021.

[6] Andrea Burzoni, Giulio Reina and Mauro Dimastrogiovanni. An improved active drag reduction system for formula race cars [www]. Retrieved from

https://journals.sagepub.com/doi/10.1177/0954407019862913 . Published 18 July 2019. Read 4 May 2021.

[7] Siemens. STAR-CCM [www]. Retrieved from

https://www.plm.automation.siemens.com/global/en/products/simcenter/STAR-CCM.html . [No date]. Read 4 May 2021.

[8] Ansys. Ansys Fluent [www]. Retrieved from

https://www.ansys.com/products/fluids/ansys-fluent . [No date]. Read 4 May 2021.

[9] Ansys. Ansys Discovery [www]. Retrieved from https://www.ansys.com/products/3d- design/ansys-discovery . [No date]. Read 4 May 2021.

[10] DI He. What’s all the racquet about recycling carbon fibre? [www]. Retrieved from https://cecs.anu.edu.au/news/whats-all-racquet-about-recycling-carbon-fibre . Published 21 October 2019. Read 4 May 2021.

[11] Ycom. Worlds first sustainable alternative to carbon fibre revealed for motorsport crash structure [www]. Retrieved from https://www.ycom.it/natural-fibre-crash-box-ycom-bcomp/ . Published 20 October 2020. Read 4 May 2021.

[12] Defromont, Guillaume; master’s student at KTH Royal Institute of Technology. 2021.

Discussion 29 April.

[13] Trigell, Emelie; PhD student at KTH Royal Institute of Technology. 2021. Discussion 14 April.

[14] CFD Online. Incompressible flow [www]. Retrieved from https://www.cfd- online.com/Wiki/Incompressible_flow . [No date]. Read 28 May 2021.

[15] CFD Online. SST k-omega model [www]. Retrieved from https://www.cfd- online.com/Wiki/SST_k-omega_model . [No date]. Read 4 May 2021.

(23)

i

Appendix

A – Formula Student Germany Rules

(24)

ii

(25)

iii

(26)

i

B – Work process in STAR-CCM+

(27)

ii

Figure 1: Workflow in STAR-CCM+ and the different parameters. The workflow was identical for all configurations

(28)

iii

Figure 2: The enclosure where the fluid region was defined. The red circle indicates the position of the rear wing,

(29)

i

C – Additional results from the closed configuration

Figure 1: Pressure on the wing

Figure 2: Cd plot

Figure 3: Cl plot

(30)

i

D – Additional results from the open configuration, angle of attack -10°

Figure 2: Cd plot

Figure 3: Cl plot

(31)

i

E – Additional results from the open configuration, angle of attack 15°

Figure 2: Cd plot

Figure 3: Cl plot

(32)

Numerical Study on Aerodynamic Drag Reduction on a Rear Wing of a Formula Student Car