Transient state simulation of flow over a throttle body

• Objective: To perform transient state simulation of flow over throttle body in Converge Studio software and study the flow pattern related to the same.

• Procedure:

1. The throttle body geometry is downloaded and imported into Converge Studio software.
2. The geometry preparation is done in order to facilitate simulation like, deletion of unwanted surfaces which don't have any effect on the simulation, patching a few surfaces, etc.
3. Boundary flagging is done logically as per simulation requirements, like inlet, outlet, body and throttle body (Note: The throttle body is considered as a surface as the flow over it is the point of concern).
4. The input parameters for simulation such as, fluid medium, pressure inlet and outlet, mesh grid size, etc., are set-up and the input files generated (Note: The 'Fixed Embedding' feature is used to refine the mesh near the throttle body as the flow over it is the major concern).
5. As it is a transient simulation the time for the simulation and the instant at which the throttle body should rotate & to what angle is specified while giving the input conditions. (Note: The time for the simulation is calculated using the following formula: `"Flow through Time" = "Length of the pipe" / "Velocity of air inlet"``, where flow through time is the time for particle to flow once through the entire elbow)
6. The converge executable file is run to generate the output files.
7. Post-processing is done by running the post-convert executable file.
8. Paraview software is used to view the simulations of various parameters such as velocity, pressure, etc.
9. In addition to the above, the 'steam tracer' tool of Paraview is used to find how the flow propagates over the throttle.
10. Animations of the same is recorded.
11. In Converge Studio, the plots of parameters like pressure, velocity, mass flow rate, etc. are made.

Note: Calculation of Time for simulation

w.k.t., `"Flow through Time" = "Length of the pipe" / "Velocity of air inlet"`

`"Here, Length of the pipe"~~0.21m`

`"Also, the Velocity (from the steady state Simulation)=" 190.608"m/s"`

`=>"Flow through Time="0.21/190.608=0.0011~~0.001s`

Generally for simulation, 4 to 5 times the flow through time is considered, but here the challenge is to simulate the flow for 0.008s with the following specifications:

• From 0 to 2 ms the valve should rotate by 25 degrees.
• From 2 to 4ms it should stay at 25 degrees.
• After which by 8ms the valve should get back to it’s original position

• Observations:

Mesh size: dx = 2e-3m, dy = 2e-3m, dz = 2e-3m

*Pressure Glyphs:(to show the direction of flow)

=> Inference:

• The direction of the flow is from top-right (inlet) of the pipe to the bottom (outlet). So, in the following contours the same convention is followed.

*Flow Tracer:(to show how the throttle body rotates)

=> Inference:

• The above shows how the throttle body rotates in the stated rate and the corresponding effects on the flow of the fluid.

*Pressure Contours:(as the simulation involves many dynamic changes, all the contours are presented in video format)

=> Inference:(As the 'Surface with edges' option was covering the contour, it was disabled)

• Pressure at the inlet is 1.5 bar and that of the outlet is 1 bar.
• At first the pressure is uniform throughout the elbow as the throttle is stationary. The pressure starts to change drastically as the throttle rotates.
• As the throttle starts to rotate the pressure starts to develop in the windward side, as the flow propagates further, the pressure becomes very much higher(indicated red).
• This is because the particles strike the windward side which converts the kinetic energy of the particles to pressure energy.
• Consequently, the pressure in the leeward side is very much lower(indicated blue). This is because the possibility of particles reaching this side is less as the flow is obstructed by the rotation of the throttle.
• As the throttle returns to its postion, the pressure becomes uniform slowly.
• Also the pressure drops in the junction because the amount of air passing thro' the junction is low, which is similar to the case of intermediate pressure in convergent-divergent nozzle.
• Moreover, the pressure plot is so smooth as the meshing is done with cells of smaller size.

*Velocity Contour:

=> Inference:(As the 'Surface with edges' option was covering the contour, it was removed)

• The velocity is low(indicated bluish green) in the windward side as the flow gets obstructed by the movement of the throttle.
• The velocity is a bit higher(indicated greenish yellow) in the leeward side as the fluid particles get guided by the rotation of the throttle and the velocity increases.
• Moreover, in the farther side (near the boundary of the throttle) the velocity is too low, this is because the fluid particles diverging at the incident of throttle, converge on the latter side at a distance from the throttle. (Clearly depicted above in blue color)
• The contour is smooth as the cell size selected is the lower. The variation in velocity can be seen in a far clearer fashion with greater detail.

*Mesh Generation(Surface with edges):

=> Inference:

• Above shows the mesh generation with the 'Surface with edge' option enabled.
• The cells are of very small size.
• The fixed embedding near the throttle body, which refines the mesh to a greater extent for 2 layers and a scale level of 2.
• The video shows how actually the mesh is generated as the throttle rotates and retracts back to its original position. The mesh adapts the boundary of the throttle as it changes.

*Plots:

• Mass Flow rate at inlet and outlet:

=> Observation:

*The mass flow rate is -ve in the inlet and +ve in the outlet.

*The mass flow rate varies drastically in the time period 0.001 to 0.002s, then it follows a mild steep curve.

=> Inference:

*The mass flow rate is so small as the pressure difference between the inlet and outlet is less.

*As the throttle rotates to 25 deg (0s to 0.002s) the mass flow rate increases due to the diversion of fluid by its movement.

*Then, it becomes somewhat steady (0.002s to 0.004s) as the throttle stays in 25 deg steadily.

*The the mass flow steadily increases as the throttle retracts back to its original position, i.e. the throttle opens.

• Velocity at inlet and outlet:

=> Observation:

*The velocity is -ve in the inlet and +ve in the outlet.

*The velocity curve is similar to the mass flow rate.

=> Inference:

*The velocity has a considerable value.

*The velocity increases as the the throttle starts to rotate and follows a mild steep curve.

• Total Pressure at inlet and outlet:

• Static Pressure at inlet and outlet:

=> Observation:

*The total pressure converges to the set-up value of 150000Pa in the inlet and the static pressure converges to 100000Pa in the outlet.

*The total pressure outlet and the static pressure inlet vary drastically.

=> Inference:

*The set-up pressures converges faster than the other variables.

*The total pressure outlet and the static pressure inlet vary as the throttle rotates and then follow a mild steep curve.

• Total number of cells:

=> Observation:

*The total number of cells varies drasticallly between 16000 to 17000 cells.

*The total number of cells increases from 0s to 0.002s as the throttle rotates during this period.

*Then it follows a pattern from 0.002 to 0.004 and is steady.

*Again it slowly reduces to the base value.

=> Inference:

*As the flow is transient the number of cells vary in the above pattern.

*Initially, during the rotation of the throttle the number of cells vary in an unsteady pattern as it is dynamic.

*Then as the throttle stays in 25 degree angle, it is quite stable. Then its again a similar decreasing pattern as that of the rise.

*As the number of cells is very high, the accuracy of the contour and plots is good.

• Animation link for velocity Stream Tracer simulation:

=> Inference:

*The variation in velocity of the air flowing through the elbow is recorded using the Paraview software.

*It shows that as the flow proceeds the particles get obstructed when they are incident to the throttle and then converge at a distance in the other side of the throttle.

*It clearly shows the kinetic energy is converted to pressure enegry as the particles strike the throttle body when it starts rotating, which reduces the velocity at these points (indicated by blue color transition). Also, a lower velocity at the points immediately below the throttle body is witnessed due to the convergence of particle.(indicated by sky-blue color flow lines)

*Moreover on the farther side, the velocity is higher as the particles slide over the throttle as it rotates(indicated greenish-yellow). And the velocity becomes normal as the throttle body comes to its original position(indicated bluish-green color).

• Result:

Thus, the transient state simulation of flow over a throttle body has been done and the inferences derived.