Project

Flow Over FSAE Car

GGokula Krishnan A24 Jul 2020

CFD

Objective:
- To simulate the flow over FSAE car for two different races and derive the aerodynamic parameters for the same. The specifications of the race tracks are as follows:
  1. Race 1:
    1. Racing track conditions - Lot of turns
    2. Average lap speeds - 45kmph
    3. 70% turns at 45 degrees, 20% turns at 80 degrees, 10% turns at 20 degrees
  2. Race 2:
    1. Racing track conditions - Straights
    2. Average lap speed - 75kmph
Theory:
- Formula-SAE cars are built for fuel economy, controlability(manoeuvrablity), optimal weight, etc. Of these factors, the fuel economy and controlabilty depend on the aerodynamic parameters of the vehicle.
- The two major aerodynamic forces, namely, the drag force and the lift(here, its negative counterpart, downforce) play a major role in the efficiency of the vehicle.
- The drag force is the force acting parallel to the ground(horizontal) in the direction of fluid flow. As it acts against the direction of motion of the vehicle, reducing it to a major extent improves fuel economy(as the force required to propel the vehicle forward reduces).
- The negative lift or downforce is the force acting perpendicular to the direction of fluid flow towards the ground. This force helps in maintaining the traction with the ground, which is very important when controlability is taken into concern. So, increasing the downforce would benefit the car with controlability at all the circumstances that the car would need to undergo in the race-track.
- To reach all the above objectives and optimize the car, it is required to conduct aerodynamic analysis. For instance, the contribution of aerodynamic analysis towards the evolution of Formula-1 cars is quite commendable as shown below.
- The left most is the most primitive model, middle one- start of this century, and the right most is the advanced & futuristic model.

Procedure:

The geometry of the FSAE car(.stl file) is downloaded and imported into Converge CFD software.
The diagnostic test is run to check for surface imperfections. As it a complex geometry, there are a few problems relating to Non-manifold edges. It is resolved by deleting these erroneous triangles and patching them perfectly by using the Repair-->Patch feature in Converge. Hence, gometry clean-up is done and the diagnostic test is run once again to ensure that the geometry is error-free.
As quoated before, it is a complex geometry, in order to get a clear picture of aerodynamic contributions of various parts present in the geometry, it is split into useful boundaries. This is done by creating boundary fences around each boundary that is required. Boundary-->Fence-->Between 2 vertices option is used to create the fence around each and every boundary one edge at a time. Then they are flagged accordingly for convenience such as, Rear wing, front wheels, rear wheels, front suspension right, front suspension left, rear suspension right, rear suspension left, front wings right, front wings left, under body, human, and the body as shown below:
Having made the geometry perfectly, the next important step is to create the virtual wind-tunnel. As the wake region along the rear end of the car is the point of concern, the length of the wind-tunnel is generally taken around 10 to 50 times that of the length of the car. Here, 3L in front of the car & 10L until the end of the wind-tunnel is taken as the length. The width of the WT is about 2 times the length of the car, so that the flow is not affected by the side walls. The height of the WT is approximately the length of the car(i.e. 2.5m), again for the same reason. The below shows the wind tunnel for race 1 and race 2, separately:
1. As the track of race 2 involves only straights and no major turns, it is designed first as per the above dimensions. After creating the wind-tunnel, the various boundaries such as inlet, outlet, top-side walls, and bottom-wall are flagged in addition to the existing boundaries.
  - Note: Basically, in CFD, the intersection of triangles on one surface with another is prohibited and is not possible. Here, in ideal case, the wheel must touch the bottom wall, which is not possible in CFD. Moreover, having a small space between the wheel and the bottom-wall would affect the simulation results, which would lead to wrong conclusions. Therefore, in order to solve this problem, the wheels in the car geometry are designed to have a flat open portion in the bottom. The bottom-wall is patched leaving the edges pertaining to this hole using Patch around hole feature of Converge CFD. Thus, this just mimics the car standing on a road. The below figure shows how the bottom wall is patched:
2. Now, for race 1, the wind-tunnel is similar in dimension to the previous, but as the race track contains a lot of turns, the car is turned to an angle based on the weighted average of the given angles.(Note: Here, either the car or the wind-tunnel could be tilted, as in either way the effect on the simulation is going to be the same, the car is tilted)
  - `"Weighted Average"=(sum_(i=1)^nw_i*x_i)/(sum_(i=1)^nw_i), "where "w_i-"weight pertaining to the data", x_i-"data values"`
  - `Here,w_1=70%,x_1=45deg,w_2=20%, x_2=80deg,w_3=10%,x_3=20deg`
  - `=>"Weighted Average"=(0.7xx45+0.2xx80+0.1xx20)/(0.7+0.2+0.1)=(31.5+16+2)/1=bb"49.5deg"`
  - Thus, the car is turned thru' 49.5deg about the center of the car, using the Transform-->Rotate option of Converge CFD. To be particular, it is rotated about the z-axis as the reference axis, with an angle of 49.5deg w.r.t. the x-axis. Due to the rotation of the car, a few surface imperfections occur, especially in the bottom-wall of the wind tunnel(near the holes of the wheel). This is resolved by deleting the bottom-wall and recreating it following the same procedure as before. The below figure shows the wind tunnel of the race-1:

The inputs for the simulation are given by means of case-setup:

The fluid/material medium is selected as air, and its default properties are specified.
The Simulation Parameters are specified as follows:
1. The transient solver is selected under Run Parameters as it is a time-dependant simulation and other default parameters are validated.
2. The time for Simulation time parameter is calculated using the Flow thru' time as follows:
  - `"Flow through time"="Length of the pipe/wind-tunnel"/"Velocity of fluid(air)"`
  - `Here, "Length of the pipe/wind-tunnel"=3L+10L"(as stated earlier)"=3xx2.5+10xx2.5=bb"32.5m",` `"Velocity of fluid"_(Race1)=45kmph=45xx5/18=bb"12.5m/s", "Velocity of fluid"_(Race2)=75kmph=75xx5/18=bb"20.833m/s"`
  - Now, `"Flow thru' time"_(Race1)=32.5/12.5=2.6s " and " "Flow thru' time"_(Race2)=32.5/20.833=1.56s`
  - The time for simulation is generally taken as 2 to 4 times the Flow thru' time. Here, 2 times the value is taken as it is quite big. So, time for simulation=`2xx"Flow thru' time"=2xx2.6=bb"5.2s"`. For simplicity, the same value is used in both the race simulations, as this value is higher than the other.
3. The default values are set under the Solver parameters.
The initial conditions are specified, such as the velocity along x-axis as 12.5m/s and 20.833m/s for Race-1 and Race-2 respectively, and the pressure is initialised to atmospheric pressure.

The boundary conditions are given as follows:

S.No.	Boundary	Boundary Condition
1.	Rear wing	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
2.	Front wheels	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
3.	Rear wheels	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
4.	Front suspension right	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
5.	Front suspension left	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
6.	Rear suspension right	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
7.	Rear suspension left	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
8.	Front wing left	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
9.	Front wing right	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
10.	Under body	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
11.	Human	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
12.	Body	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.
13.	Inlet	Inflow boundary- `Velocity_(Race1)=12.5m"/"s` `Velocity_(Race2)=20.833m"/"s` \| Pressure-Neumann's BC (Race 1&2) \| Air is included as the working fluid.
14.	Top-side walls	Symmetric boundary \| As the boundary layer formation is not required near this boundary, it is employed.
15.	Outlet	*Outflow boundary- Pressure=101325Pa (Race 1&2) \| Velocity-Neumann's BC \| Air* is included as the working fluid.
16.	Bottom-wall	Wall boundary- Law of the wall \| As the boundary layer formation is required near this boundary, it is used.

The turbulence model is selected as `"Standard "k-omega" model"` as it is an external flow. This model is precise for capturing the flow characteristics near the wall. This records the viscous sub-layer in an accurate way. Hence, it is selected.
Then the mesh sizes are given under Grid control section, considering the following:
- The mesh size near the wall is determined by means of the Y-plus value considered, for this the website: https://www.pointwise.com/yplus/ is used. The inputs such as free-stream velocity, density, dynamic viscosity, reference length and Y+ value are given and the corresponding wall spacing is obtained as output.
- For Race 1:
  - The velocity for Race-1 is given as 12.5`m"/"s`, reference length as length of the car(2.5m), and the rest of the values are default. For Y+ values of 1(viscous sub-layer) and 30(turbulent layer), the refined mesh size or wall spacing are found to be 0.000029m and 0.00089m respectively. From this the base mesh size is derived as shown below, assuming 5 levels of scaling using fixed embedding:(Note: Two values of Y+ are considered to demonstrate the variation in mesh size, and also that the case of Y+=1 is not possible due to system constraints)
  - `=>0.000029xx2^5=0.0009m(Y+=1),`
  - Also,`0.00089xx2^5=0.028m(Y+=30)`
- For Race 2:
  - The velocity for Race-1 is given as 20.833`m"/"s`, reference length as length of the car(2.5m), and the rest of the values are default. For Y+ values of 1(viscous sub-layer) and 30(turbulent layer), the refined mesh size or wall spacing are found to be 0.0000185m and 0.00055m respectively. Similar to the previous case, the base mesh size is derived as shown below, assuming 5 levels of scaling using fixed embedding:
  - `=>0.0000185xx2^5=0.00059m(Y+=1),`
  - Also,`0.00055xx2^5=0.018m(Y+=30)`
- As the above derived base grid values(for even Y+=30) cannot be used due to system constraints, the maximum possible refinement and the base grid size are determined thru' a series of trials. By the virtue of this, the most appropriate base grid size taking the system constraint into concern is 0.25m. In order to reduce the number of cells, the cells are stretched along x-axis(Stretching- Max.50% of the actual value) as variation along x-axis is to an extent insignificant, so the base grid dimension is 0.32X0.25X0.25m.
- Moreover, to refine the cells near the various boundaries of the car the Fixed Embedding option is used. This helps in capturing the characteristics of the flow in greater detail. Here, two types of Fixed Embedding is used, one box type and another, boundary type. The maximum possible scale value based on the system constraint is 2, both for the boundary and the box. Hence, it is employed. To elucidate, the box type Fixed Embedding is used to capture the wake region, which is 3D in nature, so its centre is placed near the rear end of the car, having a span of over 2L from this point along the length.
Then, the Y+ is included in the Post variable selection under Output/Post Processing tab, so that its variation can be recorded. The time-period for writing text & 3D outputs are given and the wall boundary outputs are enabled to capture the wall related outputs, such as lift, drag,etc. in the Outputs files section.

As all the required inputs are given, the input files are exported to a folder and the converge executable file is run.
The output files are generated as the simulation runs. These output files are converted into required form as soon as the simulation is complete, so as to view using Paraview software.
Then the converted files are post-processed using Paraview software and the simulation results are analysed.

Observations and Inferences:
- Note: As the Race-2 is a general case(involving only straights), and Race-1 is a particular case(involving lots of turns), Race-2 is explained first for simplicity.
- Race-2:
  - Conditions:
    - Racing track conditions - Straights
    - Average lap speed - 75kmph
  - Visualisation of the flow:
  - - Observations:
      - The flow is diverted by the car to upward, bottom and side-ways directions.
      - The density of flow tracers in the locations near the rear-end of the car is very low, and the flow characteristics are very varied in the leeward side.
    - Inference:
      - The front part of the car, the body and the front wing cause the flow to divert smoothly. It is one of the functions of the wing to divert the air upward, bottom & sideways. This is done to cool some critical parts of the system(say, radiator) and reduce the area of the region of low velocity behind the wheels, called the dirty air, as this would cause some problems to performance.
      - There is a formation of partial vacuum in the leeward side of the car, which causes the the varied characteristics of the flow. Moreover, this also leads to the wake formation, which is briefed in detail in the next section.
  - Velocity Contour:
    - Velocity variation over the surface(Front & side):
    - - Observations:
        
        The velocity is less to an extent in front part of the car(i.e., wind-ward side of the car).
        
        The velocity is very high(indicated red) above the car and very low in the leeward side and sides of the body(indicated blue) of the car.
      - Inference:
        
        As the fluid strikes the car, it looses some amount of its kinetic energy which leads to the decrease in velocity of the fluid in wind-ward side of the car.
        
        The fluid incident on the car is diverted to the upper part of it, which adds up to the velocity of the fluid actually flowing above the car, thus an increase in velocity is witnessed above the car.
        
        Some amount of fluid is diverted to the sides of the car(away from the car) by means of the front wheels and the body, which leads to less or no fluid flow in the region between the front and rear wheels(along the body of the car), which ultimately causes lower velocity in the sides of the car(indicated blue in the side view of the car).
        
        The low velocity in the leeward side is accorded to the flow separation taking place at that location due to the geometry of the car. Due to this, a region of partial vacuum is formed in the leeward side of the car. This leads to flow reversal or recirculation in the rear end of the car, ultimately giving rise to vortex formation in the leeward side. These vortices tend to move along the wind direction and this entire region of low velocity in along the leeward side of the car is known as the wake region. This contributes a lot towards the drag of the car.
  - Velocity Glyph vector:
    - - Observations:
        
        The flow separation near the rear-end is evident(as the intensity of arrows in the rear-end along the car is low).
        
        The flow reversal or recirculation near the rear-end of the car is found(as a few arrows point in the direction opposite to that of flow).
      - Inference:
        
        The phenomenon of flow separation and flow reversal/recirculation are ascertained from the corresponding observations.
        
        These are accorded to the geometry of the car.
  - Pressure Contour:
    - Pressure variation over the surface(Front & Side):
    - - Observations:
        
        The pressure is quite high in the wind-ward side(the front part of the wheels, body, and the human) of the car(indicated red).
        
        The pressure is considerably lower in the upper and leeward side of the car(indicated blue).
      - Inference:
        
        As the air strikes the car, the kinetic energy of the car is converted into pressure energy, thus, there is an increase in pressure in the front part of the car which is incident to the free-stream. This also adds up to the drag acting on the car.
        
        The low pressure in upper part of the car is attributed to the high velocity in this region.
        
        Varied from the above, the low pressure in the leeward side of the car is due to the wake region, which is formed due to the flow separation, partial vacuum and recirculation as briefed previously. This pressure is also called as suction pressure, which contributes to the major amount of drag.
  - Mesh Generation and Total Cell Count:
    - Total Cell Count Vs time:
    - - Observations:
        
        The mesh is finer near the car, and quite coarse in farther regions.
        
        The total cell count is around 155,000
      - Inference:
        
        As the fixed embedding feature with 2 scales is used, the mesh is finer near the car.
        
        However, it is coarse when compared to the actual practice, and the cell count is considerably meagre, which is the cause of the less accurate simulation so obtained(accorded to the system constrains).
  - Y-Plus vs Time:
    - - Observations:
        
        The Y-Plus is far higher than the permissible limits(i.e., most of it is in the range of 1000s).
      - Inference:
        
        This is accorded to the coarse meshing done. So, the simulation results are less accurate to an extent, as discussed earlier.
  - Comparison of Drag force:
    - Here, the drag force for each part is extracted from the simulation and average of the last 1s is taken to determine the appropriate value for the same.
    - Link for Excel sheet containing the drag force data: https://drive.google.com/file/d/1IpeJQ4YsX7hd-5ah30t26Y32cPVCcyEb/view?usp=sharing
    - - (Note: The first plot shows the drag force value for each part separately, the second represents the same, but, as the centre of action of body and under body are same, the net force is considered for it, for simplicity. Moreover, it is done to get a brief view of forces acting on various other parts. The same procedure is followed for all the upcoming plots)
      - Observations:
        
        The maximum amount of drag is produced by the front wing.
        
        The entities like front suspension, wheels contribute to some amount of drag.
        
        The net drag, when body and underbody are considered, is found to support the motion of the car(i.e., acts in the direction of motion).
      - Inferences/Recommendations:
        
        The front wing, which is supposed to produce less drag, is a major contributor for the same. This can be accorded to ineffective positioning and selection of aerofoil for the wing. This can be resolved by proper design of the front wing taking into concern the above parameters.
        
        As the pressure acting on the entities like front suspension, wheels are considerable, these produce some amount of drag on the car.
        
        The net drag force value, taking body and underbody into concern, is negative, which is quite different from the usual.
  - Comparison of Down-force:
    - Here, the downforce for each part is extracted from the simulation and average of the last 1s is taken to determine the appropriate value for the same.
    - Link for Excel sheet containing the downforce data: https://drive.google.com/file/d/1HxITZheQC6_oSdoLI-adnjM_1acB5vrx/view?usp=sharing
    - - Observations:
        
        The maximum amount of downforce(negative lift) is produced by the Human.
        
        The wheels contribute to considerable amount of downforce, whereas the wings(both front & rear) produce only negligible amount of downforce!
        
        The combined effect of body and under body is found to produce lift denoted by positive value.
      - Inferences/Recommendations:
        
        The downforce acting on the Human is attributed to the component of force due to pressure acting on it, along the vertical direction. For a similar reason, there is downforce acting on the wheels. The same is shown in the below image.
        
        The purpose of installing wings is to impart the major amount of downforce on the car. But there is negligible influence on downforce by the wings. This is because of the inappropriate positioning and selection of airfoil for the wings. Moreover, it could be improved by providing flaps(gurney) and varying the angle of attack to increase the downforce. The below shows the variation of pressure over the front and rear wing of the car:(to depict the low pressure on these elements, which causes the above observation)
        
        The body and under body producing positive lift is quite different from the usual observation. The below shows the variation of pressure on the body and under body of the car, which clearly shows a lower pressure on both the entities, which could have led the above result.
        
        (Note:The results are approximate, accorded to the coarse meshing due to the system constraints)
- Race-1:
  - Conditions:
    - Racing track conditions - Lot of turns
    - Average lap speeds - 45kmph
    - 70% turns at 45 degrees, 20% turns at 80 degrees, 10% turns at 20 degrees
  - Realisation of flow:
  - - Observations:
      - The flow is diverted by the left part of the car(viewed from the screen facing the front part of the car) to various directions abruptly.
      - The flow in the farther-side is very varied in characteristics, in 3D fashion.
    - Inference:
      - As the car is rotated thru' an angle 49.5deg, the fluid is incident at the left part of the car, hence it is the wind-ward side. After it strikes the car at various locations, like wheels, wings(though at an angle), body, and human, it is diverted abruptly as the entities like wheels, wings are facing sideways which restrain it from causing a smooth diversion as the previous case.
      - The flow separation and wake formation just past the axis of symmetry of the car causes the varied behavior of the flow in the farther-side. This is dealt in detail in the following sections.
    - (Note: Here, the flow characteristics are unsymmetrical as the car is rotated. So, the aerodynamic parameters vary drastically)
  - Velocity Contour:
    - Velocity variation over the surface(Front, back & top):
    - - Observations:
        
        The velocity is low in the wind-ward side(indicated blue) and quite high in & around the regions pertaining to the axis of symmetry of the car(indicated red).
        
        The velocity is very low in the leeward side of the car(indicated dark blue) and it extends to a vast region along the rear-end of the car.
      - Inference:
        
        The low velocity in the wind-ward side is due to the loss of kinetic energy of the fluid as it strikes the car, which is usual.
        
        As the some part of stricken fluid is diverted to the upper part of the car(i.e., the the region of axis of symmetry), it adds up to the velocity of the fluid already flowing in this region. Hence, the velocity is high in this region.
        
        The flow separation takes place in the right side(viewed in the same fashion as before) along the axis of symmetry of the car. This, as usual, leads to formation of partial vacuum, which inturn gives rise to flow reversal/recirculation. This causes very low velocity in this region. Moreover, as the car is at an angle, this region extends to some distance along the leeward side. Also due to the presence of the rear wing(facing the free-stream at an angle) that the low velocity region is extended to such vast areas.
        
        As the low velocity region, i.e. the wake region, is vast, the drag associated will also increase to a greater extent.
  - Velocity Glyph vector:
    - - Observations:
        
        The flow separation is witnessed in the leeward side(as there are hardly any arrows along the rear-side of the car).
        
        The flow reversal/recirculation is also clear, as there many arrows pointing in the direction opposite to that of the flow, in the leeward side.
        
        The region of flow reversal/ recirculation is quite vast as the region behind the car has scarce amount of arrows.
      - Inference:
        
        The phenomenon of flow separation and recirculation are ascertained by the observations.
        
        The area under flow reversal is high as the flow is diverted abruptly by various elements such as rear wing, side part of the body, as specified earlier.
  - Pressure Contour:
    - Pressure variation over the surface(Front & Back):
    - - Observations:
        
        The pressure is very high in the wind-ward side of the car(indicated red), i.e., the left part of the car(front & rear left tyres, left part of the body, left front wing).
        
        The pressure is very low along the axis of symmetry of the car(indicated dark blue) and considerably low in leeward side of the car(indicated light blue).
      - Inference:
        
        As the air strikes the wind-ward side of the car, the kinetic energy is converted into pressure energy, hence, the pressure is very high here.
        
        The low pressure along the axis of symmetry of the car is accorded to the high velocity in these regions.
        
        The low pressure in the leeward side is attributed to the flow separation taking place at there. As the partial vacuum forms and flow reversal takes place in the leeward side, it leads to low pressure in this region.
  - Mesh Generation and Total Cell Count:
    - Total Cell Count Vs time:
    - - Observations:
        
        Similar to the previous one, the mesh is finer near the car, and quite coarse in farther regions.
        
        The total cell count is around 2,20,000.
      - Inference:
        
        Again, the fixed embedding feature with 2 scales is used, so, the mesh is finer near the car.
        
        However, it is coarse when compared to actual practice, and also that the total cell count is quite optimal, which is the cause of the less accurate simulation so obtained(accorded to the system constrains).
  - Y-Plus vs Time:
    - - Observations:
        
        Similar to the previous case, Y-Plus is higher than the permissible limits(i.e., it is in and around 1000s). But it is lower than the previous.
      - Inference:
        
        The higher value is again accorded to coarse meshing done. It is quite lower than the previous because the velocity is low in this case comparatively.
  - Comparison of Drag force:
    - Here, the drag force for each part is extracted from the simulation and average of the last 1s is taken to determine the appropriate value for the same.
    - Link for Excel sheet containing the drag force data: https://drive.google.com/file/d/1JIv0Xx1kgkGw5m6myY-lo8r3YXdPo7Ny/view?usp=sharing
    - - Observations:
        
        The left front wing contributes to the major amount of the drag on the car.
        
        Also the wheels and right front suspension contribute to a small amount of drag.
        
        The right front wing, rear suspension, combinatory force of body & under body are found to push the vehicle in the direction of motion by producing -ve force.
      - Inference/ Recommendations:
        
        As the pressure force acting on the left front wing is high(as shown in the figure below), the drag force imparted by it on the car is quite higher. Similarly, a considerable amount of net pressure acts on the wheels, which produces some amount of drag(as shown in the figure below). Unlike the previous, the right front suspension experiences drag as it belongs to the wake region(low pressure region).
        
        The pressure acting on the left part of the right front wing is lower than that in the right part, so the horizontal component of net-force acts in the -ve direction. The same is shown in the below figure.
        
        Moreover, the drag, for this case, could be minimized by making changes in the design, considering the turning positions too. For example, the wings could be designed such that the they are aerodynamic even while turning.
  - Comparison of Downforce:
    - Here, the downforce for each part is extracted from the simulation and average of the last 1s is taken to determine the appropriate value for the same.
    - Link for Excel sheet containing the downforce data: https://drive.google.com/file/d/1Uk4L8xrsqo4mcGCkc2JsEJSkf40S8s-L/view?usp=sharing
    - - Observations:
        
        The maximum amount of downforce is produced by the Human.
        
        Considerable amount of downforce is imparted by the wheels.
        
        The combinatory effect of the Body and the Under body is found to offer lift.
      - Inference/ Recommendations:
        
        The downforce produced by Human is accorded to the component of resultant force acting on it due to pressure as shown below.
        
        Similarly, as the car is turned, it is the component of pressure force that causes the downforce on the wheels(as depicted in the previous image relating to pressure on wheels).
        
        The Body and Under body together offer lift, which is different from the usual scenario. The pressure is higher in the wind-ward side, lower in the leeward side when body is taken into concern, and it is lower for under-body(shown in the below images). Thus, as it is a combination of the various pressure forces, the component of the resultant could have caused the lift prevalent.
Result:
- The simulation of flow over FSAE car for two different track conditions has been done and the various aerodynamic parameters derived.

Flow Over FSAE Car

Ship work like this
in your own course.

Talk to a career counsellor

Ship work like thisin your own course.

Talk to a career counsellor

Ship work like this
in your own course.