Week 4: Project - Transient simulation of flow over a throttle body

Transient state simulation of flow over a throttle body

As soon as the geometry is imported in converge, diagnosis check is run in order to check for errors if any present. Here the elbow geometry consists of some thickness which we actually do not require for our simulation. So,

Step1: Automatic boundary creation: Create automatic boundary fence regions, then delete the thickness of the elbow, so that only a single surface remains.

Step-2 Repair the inlet and outlet passages: Repair has a ‘patch’ option which when applied to the open edge at inlet and outlet will patch the edge.

Step-3: Again, run the diagnosis check by clicking on find, if there are no errors, click on ‘clean’ also. Also tick the box for ‘check all the normal that are pointing to the interior ‘. After doing this, display the normal once to check if the fluid flow is inwards. in the pipe.

Setp-4 Create boundaries: Once the errors are removed, create boundaries by going in boundary option and choosing flag and ‘add ‘for inlet, outlet, elbow and throttle and assign the respective areas in the boundaries. We can also hide /unhide selected boundaries by ticking/unticking the boundary check box in the ‘view options’ column on right side.

Step-5 Case-setup: Once all of this is done, then case -setup should be done. It consists of many important steps. It sets the particular parameters for the particular simulation.

Case set-up

Click the geometry bounding box in ‘options’ to find the dimensions of the geometry. Converge assumes the file to be in metres. If the geometry is made initially in mm, then it needs to be scaled in converge. Transform consists of scale option where we have to scale the geometry to 0.001.

First: Select the time-based solver. Air as the fluid flowing. Solver type should be transient state solver.

Second: For transient state simulation, the run time is in time. So,

End time Calculation: Length of pipe (can be taken as 0.2 approximately)/velocity of flow(106m/s from the line plot curve of velocity that we got for steady state simulation) i.e. approximately 0.002 seconds for the flow to occur once, to run the simulation 10 times like this , we can say that the total end time is 10 x 0.002 i.e. 0.02 i.e.20 milli seconds.

Initial time step- 1e-9

Minimum time step- 1

Time interval of writing 3d outputs for restarting -0.001(so that we get output at regular small intervals)

Time interval for writing text output: 1e-6

Maximum CFL limit: always lesser than 1 i.e. here 0.5 can be taken.

Third: Region and installation – add the region and drag all the boundaries created (from view options) into the region column. Now go to the ‘boundary’ set-up and choose the following:

Inlet: Type- inflow

Pressure-1.5 bar

Temperture-300K

Fluid-air

Outlet- type- Outflow

Pressure- 1bar

Temperature- 300k

Fluid-air

Throttle- type- wall

Rotating type, moving boundary

Now, we draw three arcs on the top face of the cylindrical portion, so that we can define a centre and arc normal of the face. Go to ‘direction’ option in ‘measure’ and choose arc normal option. Choose 3 vertices on the face of cylinder and click apply. Now in message log, we get arc centre and arc normal. We could have also selected triangle normal instead of the arc normal option, that would have also given similar result.

Here we need rotation origin: This is the arc centre

Rotation about: This is the arc normal

Rotation rate: We want angle to go till 25 degree in 2 milli seconds, then remain 25 degree till 4 seconds and then come back to original by 8 milli seconds.

Time angle

0.0 0.0

0.002 25.0

0.004 25.0

0.008 0.0

Elbow- Type- wall

Base grid:

Mesh grid size of 2mm was chosen optimum here and fixed embedding for cells created near the throttle is done, so that cells near throttle become smaller in size, here they are scaled to level of 2 i.e. basically 2-time reduction by a factor of 2. There is very less area near the throttle because almost all the area in that region is occupied by the throttle and hence, a smaller number of cells are accommodated there, so in order to refine the mesh, those cells are decreased in size. This increases the cell count overall. Mesh defining is the most important part of this case setup.

Fourth: For physical modelling, the rng k-Ɛ model is selected. No changes are made in output files and post variable selection except selecting the species oxygen and nitrogen.

Fifth: After case-set-up, Export the files to a folder and run the simulation in Cygwin.

Sixth: After the simulation is run, results are post processed in converge and then opened in Para view.

Seventh: Open Para view, inside the multiblock inspector adjust the transparency of the elbow portion so that the throttle becomes visible.

CONCEPT/EXPLANATION OF THE FLOW:

While the fluid (air here) moves through the inlet with a slightly higher pressure and the throttle is moving now i.e. the position of throttle will change at different times. Here throttle acts as an obstruction to the flow. The flow as soon as it hits the throttle wall, strikes it and basically divides its path into an upper flow and lower flow, i.e. above and below the throttle (which is a disc like element placed in the pipe generally to regulate and control the flow). This helps creates a separated area like region in the mid region where fluid strikes, thus creating a region of low velocity there. Velocity is higher in the portion where the fluid passes through the tiny passage between the throttle wall and the pipe wall. This is a region of low pressure i.e. As the flow reaches the throttle valve, it faces decrease in the area of flow passage between the pipe wall and the throttle wall. Now the throttle is tilting also with time, so the velocity will be higher in the region where due to the tit , less area for fluid passage is there. So, velocity of flow increases because the area decreased. Due to this increase in velocity, as per Bernoulli theorem, the net energy should be conserved. So, to compensate kinetic energy, the pressure of the fluid decreases slightly. When the flow leaves the throttle, (the flow is still separated and there is still a stagnation region and velocity drops in this region also) the pressure again increases to some extent because the area of pipe again increased which leads to decrease in velocity. The velocity is somewhat higher on the mid-surface of the cylinder in the throttle face because that cylinder portion created a region of lesser area passage for the fluid flow than it surroundings.

So, the pressure decreases around the throttle and then again increases to atmospheric pressure. It does not reduce lower than atmospheric pressure while it flows through the throttle. Velocity increases as the flow passes the throttle and again decreases as the flow has passed towards the outlet.

More detailed explanation has been provided below the velocity and pressure stream line contours.

Mesh image:

Observations:

Output line plots for velocity, pressure, mass flow rate output, cell count rank

Velocity:

Pressure:

Mass flow rate input:

Mass flow rate output:

The mass inlet and outlet become constant after a certain number of cycles. The mass flow rate increases for the output and then converges and become constant. The mass flow rate for input decreases and then converges after certain cycles. The change in the rate is so steep because it takes place suddenly and then goes on converging.

The pressure and velocity output goes on converging after almost 0.01 seconds. The simulation is run for20milli seconds.

Cell count: It goes on changing here because as throttle moves, at different times the cells needed become different and hence cell count differs. When the throttle stops moving the cell count became constant (after 0.01seconds) .

Post-processed image:

Here, the throttle is clearly visible. The elbow is shaded and inlet and outlet is also visible here.

Refined mesh image with scaled mesh size near throttle:

Pressure contour at the end of cycle:

Considering the movement of the throttle, till 0-2 milli seconds, the throttle tilts from zero to 25 degrees and hence the area for flow above the throttle is lesser than earlier and the area of flow below the throttle is higher than earlier here. So, from Bernoulli theorem high pressure gets created in the lower region and low pressure gets created in the upper portion. As the angle decreases from 4-8 ms, the pressure balances and lower pressure above and higher pressure below reach a state of equal mid value pressure.

(when the throttle is steady i.e. at one position after 8 ms)

Initially the pressure is high at inlet side, after that as the flow arrives the throttle the pressure decreases as due to valve plate the velocity has increased. Pressure decreases to some value(still greater than atmospheric outlet pressure value) and decreases even more at the cylinder portion mid surface region (because there the area for flow is least technically and velocity is highest) .As the flow leaves the throttle and reaches the outlet it decreases as now, the pressure becomes same as the atmospheric pressure. At the side where the fluid flows between the pipe wall and the throttle , the area being the smallest makes the velocity higher in that region , thus increasing the pressure in that area.

Pressure at 2 ms:

At 4 ms:

Pressure at 6ms:

Pressure at 8ms:

Pressure stream lines at 2 ms:

Pressure stream lines at 4 ms:

Pressure stream lines at 8 ms:

The flow as we can see separates into 2 regions at the incoming to throttle and also as the flow leaves the throttle. The pressure when the throttle tilts from 0- 2 ms, from zero to 25 degree is less where the area is less and then goes on increasing a little as area increases for both above and below regions of the throttle as the area changes.

Velocity contour at the end of the cycle:

Velocity at 2 ms:

Velocity at 4 ms:

Velocity at 8 ms:

These images actually project the flow that is for the cell just above the throttle because there is no velocity at the surface of the throttle actually.

Detailed image of the throttle with stream line:

50-100 points are taken in stream tracer properties whose paths /streamlines are traced here and the color change is w.r.t velocity.

Velocity at 2 ms:

Velocity at 4 ms:

Velocity at 6 ms:

Velocity at 8 ms:

Observations for velocity: As the fluid flows through the inlet and strikes the throttle, it separates into two regions as the throttle acts as a obstruction to it. As the throttle tilts from 0-2ms, from zero to 25 degrees, the flow that passes above and below the throttle and due to the throttle which is at an angle now, there is a region where fluid has a very less region to flow and then the area for fluid flow expands in the upper region of throttle and similarly for the lower region of throttle, the fluid initially gets more area and then experiences a decrease in area due to the tilt of throttle , the velocity is high in that region both above and below the throttle where there is less area. The flow velocity decreases as the area for flow increases. This can be seen in the image above at 2ms. When the throttle comes back to zero from 5- 8 ms, as the throttle comes back from 25 to zero degrees and becomes straight, again the passage for fluid flow becomes more and thus the velocity in those regions decrease now. If we see the image for 6 ms, the streams are less red near the throttle because the area is increasing for fluid flow now. Also we can see that the flow goes on separating as it has passed the throttle and moves to the outlet.

Glpyh image:

Velocity animation video link:

Pressure animation video link:

Conclusion: Thus, the transient flow of fluid over a throttle has been simulated. The results agree with the theoretical knowledge and hence are found to be correct. Actually, the velocity of particles would be zero technically which touch the surface of the throttle and hence velocity contour is supposed to be nil at the throttle but this here which converge shows is the velocity contour for the nearest cells on the pipe region above this throttle and in a way that flow is projected. Stream tracer lines are also shown which describe how the streamlines move from inlet to outlet here and the color change describes the velocity changes. Throttle is basically used to control and regulate the flow in the passage for e.g. In engines inlets to control the amount of air going in or the amount of fuel going in. Here I showed how the flow changes as the throttle moves with time.