STUDYING THE AIRFLOW OVER A CAR USING AN AHMED BODY

STUDYING THE AIRFLOW OVER AN AHMED BODY

OBJECTIVE: We are going to see the significance of Ahmed body , what it is and why it is used for. Then we are going to determine the drag over the ahmed body and will going to run grid dependency test for the mesh refinement and result refinement.

What is Ahmed body

• The most biggest problem in the present scenario is burning of fossil fuels . One mode of burning is from vehicles. manufacturers are introducing more fuel efficient cars to the market. One main contributor to fuel burn is the car's aerodynamics drag. Complexly shaped, cars are very challenging to model and it's difficult to quantify the aerodynamic drag computationally. The Ahmed body is a benchmark model widely used in the automotive industry for validating simulation tools. It is widely used for validating the CFD codes , like if i have the CFD codes and want to check wheather my codes are giving correct results then i will have to run the codes for Ahmed body and check wheather my codes are giving the correct drag coefficient values and if yes then my codes are ready to use for further application. The Ahmed body is simple enough to model, while maintaing car-like geometry features.

• Ahmed body was first created by S.R.Ahmed in his research in 1984 since then, it has become a benchmark for aerodynamics simulation tools. The simple geometrical shape has a length of 1.044 meters, height of 0.288 meters and width of 0.389 meters. It also has 0.5 meter cylindrical legs attached to the bottom of the body and the rear surface has a slant that ha s40 degrees from top plane.

side view (length of 1.044 meters)

isometric view

front view(width of 0.389 meters)

Why Study the Drag Coefficient in Cars

The drag coefficient quantifies the resistance of an object in a fluid environment. It is not an absolute constant for a body's shape because it varies with speed and direction of the flow, Object shape and size and the density and viscosity of the fluid. The lower the drag coefficient of an object , The less aerodynamics or hydrodynamic drag occurs. In terms of a car, the lower the drag coefficient, the more efficient the car is. As well as affecting the top speed of a vehicle, the drag coefficient also affects the handelling. Car with low drag is better but decreasing the drag drastically can reduce the downforce and lead to loss in road traction and a higher and a higher chance of car accidents.

DRAG CALCULATION USING ANSYS AND REFINEMENT OF THE RESULT USING GRID DEPENDENCY TEST

for any computational process we require following steps to follow-

1. Modeling of the vehicle
2. Meshing
3. Pre-processing
4. Setting-up the physics and solving
5. Post-processing

modeling of the vehicle

we have modeled the geometry using solidworks and imported in spaceclaim for the geometry cleanup(pre-processing).

Meshing

We took a baseline mesh first with number of elements and number of nodes are 39681 and 8064 respectively just because we wanted to see how changes take place with varying parametrs.

Pre-processing

The imported geometry is further divided into two equal parts as it was symmetric baout that plane. Now an enclosure was formed as it is region where air flow take place and that field is called flow-field. A flow-field is the region where all physical properties of the flow can be calculated using eularian approch.

Setting-up the physics and solving

We have selected a Density-based solver and k-epsilon turbulent model . And it is solved using steady state solver.

Now the flow is turbulent because we are determining the drag due to the flow which has free-stream velocity 25m/s. Density of the air is 1.225 kg/m^3, viscosity of the air is 1.7894e-05. The flow is wheather laminar or turbulent is desided on the basis of dimensionless number called reynolds number.

Reynolds number(Re)= [(density)*(velocity of flow)* (characteristic length of the geometry)]/kinematic viscosity

Re comes out to be 17,86,772.1024 which is in turbulent zone.

turbulent model solver

property of fluid which is flowing in the enclosure

solving

Now the system initiated the solution using hybrid initialization for 1000 itteration.

residual plot

velocity contour

vector representation of velocity

Now that flow is something imposible as there are unexpected region where the velocity get increased and to get the correct flow field we have to increase the number of cell and nodes . So we have to procced for the grid dependency test where all the boundary conditions will remain the same but the number of elements will increase so that we get more number of points where we get the soutions and desired solutions.

pressure contour

The results get converged within 1000 itteration at 713 itteration . Drag-coefficient value around the Ahmed body is 19.631 as we can see in the figure below.

MESH-2 ( ELEMENT -7,05,376 and NODE- 3,05,414)

This time we are refining the mesh mesh near the Ahmed body and for that we have created an extra enclosure in spaceclaim of dimensions that is shown in the figure. The motive of making this enclosure is that we can refine the mesh where boundary layers can be formed and wake region is created so we have to ceate mlore points near the Ahmed body.

dimensions of the enclosure

By using share topology mode we have cleaned up the interfacial part in spaceclaim so that mesh should be more structured

Now in the mesh , we have created inflation layer with total thickness of 25mm and total number of layer be 30.

Following mesh commond are used to refine the mesh-

1. multizone with quad dominant elements.
2. body sizing
3. face sizing
4. inflation

default mesh element size is 100mm in the outer region and in the smaller region 50mm.

Solving with the updated mesh with the same boundary conditions

We first check for the Y + wall value and set the reference values as per the problem.

1. Area - projected area of Ahmed body- 0.1102
2. Temperature- 300k
3. Viscosity- 1.7894e-05
4. Density- 1.225 kg/m^3

Y wall + plot Reference values

residual plot

Static pressure distribution over Ahmed body

It is actually used for more clarity of visualising the negative pressure distribution over the Ahmed body.

velocity contour

vector representation of velocity-vector field

pressure distribution

this figure shows the better region where the pressure is negative

Now the drag-coefficient in this case is 0.2505 approximately.

MESH-3 (ELEMENTS-13,00,546 and NODES- 5,22,241)

With no change in enclosure we have just redused the size of elements.

The following operations we did in this mesh refinement-

1. Multizone method (mapped mesh type is HEXA),default element size- 35mm.
2. Body sizing , element size-25mm.
3. Inflation layer, maximum thickness-25mm and total number of layer-30.
4. Face sizing is applied on Ahmed body , element size is 10mm.

The following figure shows the refined mesh and inflation layer applied over the Ahmed body.

Solving with the updated mesh for 1687 itteration-

residual plot

Drag plot

velocity contour

vector representation of the wake region the flow field behind the Ahmed body

vector representation of the wake region the flow field front of the Ahmed body

pressure contour

Static pressure plot for better visualization of negative pressure region

Now the drag-coefficient is 0.2327.

NOW THE QUESTION ARISES THAT WHY THAT NEGATIVE PRESSURE ARISES IN SOME REGIONS IN THE FLOW

As from the above three cases we have seen so far, we can see that there are some region where pressure is negative , areas like near the wall.

There can be many reasons for such negative pressure such as-

1. due to increased velocity , pressure should be reduced according to the Bernoulli theorem.
2. flow seperation takes place and due to that vaccume region or bubbles formed which leads to reduction of pressure .
3. due to flow circulation and wake generation , the pressure inside the region is low as flow seperates.

NOW WHY THE FLOW SEPERATES AND WHAT IS THE SIGNIFICANCE OF POINT OF SEPERATION

Flow seperation is the boundary layer seperation and it occurs whenever the change in velocity of the fluid in either magnitude or direction , is too large for the fluid to adhere to the solid surtface. now the point of seperation as the name suggest, the position at which the boundary layer leaves the surface of a solid body.it may occur with the laminar flow or when the boundary layer is turbulent. Now this concept can be used in designing so as to produce more lift and less drag.

OBSERVATIONS

1. On increasing the number of grids , we get more smooth contour and more refined results, as we have seen in drag-coefficient value . Upon increasing the elements it approaches to 2.3.
2. On increasing the number of grids , the computation time also increases. So we should decide before meshing that what is our requirement.
3. We have to take care of the boundary layer and for that we have to check Y+ wall value .