Turbo Charger Simulation

Objective:

To perform the simulation of a turbocharger turbine ,find its contours and output parameters.

Theory:

A turbocharger, colloquially known as a turbo, is a turbine-driven,forced induction device that increases an IC engine\'s efficiency and power output by forcing extra compressed air into the combustion chamber.This improvement over a naturally asperated \'s power output is due to the fact that the compressor can force more air—and proportionately more fuel—into the combustion chamber than atmospheric pressure (and for that matter, ram air effects ) alone.

Turbochargers were originally known as turbosuperchargers when all forced induction devices were classified as superchargers. Today, the term \"supercharger\" is typically applied only to mechanically driven forced induction devices. The key difference between a turbocharger and a conventional supercharger is that a supercharger is mechanically driven by the engine, often through a belt connected to the crankshaft, whereas a turbocharger is powered by a turbine driven by the engine\'s exhaust gas . Compared with a mechanically driven supercharger, turbochargers tend to be more efficient, but less responsive. Twincharger refers to an engine with both a supercharger and a turbocharger.

Manufacturers commonly use turbochargers in truck, car, train, aircraft, and construction-equipment engines. They are most often used with Otto cycle and Diesel cycle internal combustion engiine.

Geometry:

Provided geometry in stl file was imported in converge .

Impeller:

Case setup:

Gas simulation with air as participating species

(Note: Our main aim here was to understand the software and how does fluid structure intraction happens in converge .So instead of including actual exhaust species we set air as the participating species )

Dynamic timestep was used with time parameter set as :

Turbocharger turbines usually rotate in a range of 100000rpm

so time taken per revolution =0.0006s

we make an assumption here that for the time of 100 cycles we\'ll get converged results

so End time was set as 0.06s

Boundary conditions:

Pressure outflow : 100000pa; Temperature :300k

Inflow pressure: 250000pa ; Temperature : 800k

Casing set to Staitonary wall ; Temperature :300k

Impeller set to FSI wall ; Temperature :450k

Region defined :

Turbulence model used :

RNG K-epsilon model used as recomended for such types of flow simualtions and we did\'nt have computation capacity to make the mesh finer so that our y+ is suitable for k-omega .

Setting the FSI wall parameter:

Mass ,Rotating about axis and center of rotation was provided as follows

Meshing:

Base grid : 0.0035 m

Fixed Embedding at the boundary of rotating wheel(impeller) scale of 1 and no of layers set to 1

Result plots:

Simulation:

We observe from the animation that the turbine starts rotating after certain time ,this is because of its inertia,that it has to overcome first .Detailed explaination is given ahed in turbo lag section.

Cellcounts:

As the Rotating wheel has some mass ,inertia to it ,it takes some time to rotate and overcome its state of rest.And as wall is rotating we see less fluctuation in cellcount in starting and more fluctuations when the impeller starts attaining its usual speed.

Massflow rate :

Inlet:

Massflow rate at inlet convered at around 0.065kg/s

Outler:

One way we trust our results is we shound get similar massflow rates at inlet & outlet .

Reason :Softwares usually use Linear upwind diffrencing or Advanced discretisation scheme to get the gradient value accurately . This methods take into account massflow rate as an input and output at each and every cell generated so if our solution is correct we should get similar massflow rates .

Velocity plot:

We observe Inlet velocity is around 450m/s and that at outlet is 100m/s .We observe diffrence in velocity as kinetic energy of air is used to produce rotating work.

Pressure plot:

Inlet pressure was 250000pa and pressure at outlet was found to be 104525pa,diffrence is because this pressure energy is used to produce work.

Rotation captured :

We could only capture 4.5 rotations because at the start it takes some time for the impeller to change its state of rest and overcome its inertia.

Torque generated at impeller:

We observe that our impeller will be producing around 1N-m of torque .Using gear ratios this can be further supplied to the pump so that it delivers compressed more dense air to the prime mover.

Temperature of impeller:

Impeller temperature : 550k

Casing avg temperature : 580k

Fluid expands in impeller blades ,expansion decreases temperature of fluid ,so this is the reason we observe pressure diffrence between impeller & casing .

Rotational velocity:

Here we observe that the rotational velocity has not converged ,It went upto 900rad/s (8600RPM).

to reach upto 100000 RPM we expect to keep the simulation running till 1.5 s(which right now consdering the computation capacity is not feisable )

This is common problem with turbocharged engines ,lag occurs in turbocharger which is also known as turbolag ,This is also one of the reasons why race cars driver prefer engine revving at the start of the race .We can obtain a stable rpm of turbine if we increase the End time of the simulation .

What is turbo lag?

Turbo lag is the time it takes for a turbocharger to “light up” or produce positive manifold pressure drastically changing the power output of a motor.

Let’s assume you would like even more of an explanation. Remember that turbochargers are driven off of exhaust gasses. The gasses spin a turbine wheel that is connected to a compressor wheel that shoves compressed air into the engine providing it more air, and therefor the ability to make more power. It takes time for these turbine and compressor wheels to build up the speed required to push positive pressure air into the motor. This “dead time” between you putting your foot down on the gas peddle and the turbo providing full torque are what we call “lag”. And depending on the rpm engine is turbo lag varies .

Prevention:

• Reduce mechanical losses
• Use a smaller & light weight turbine .

Flow pattern:

We observe the flow pattern ,when the flow strikes the blade velocity of fluid increases ,reason being that the blades are curved in such a way that their area decreases .As we want to maximize the work done ,this energy is utilised to produce work.

At the outlet we want the pressure and velocity minimum considering backflow or flow stagnation does not take place.Which based on our result is good as we have an outlet pressure of 1.04bar which is greater than atm.

1. In the first animation we observe that the blades are unevenly loaded .The pressure at the attacking side of impeller is high around 2.0bar and that behind the blade is negative .This will have an effect of recirculation of flow in this areas were the velocity vector becomes negative and opposes the rotation of impeller .
2. We observe from the 2nd and the 4th contour animation of pressure and velocity that where the velocity increases the pressure in that area near the throat decreases and attains a negative value .
3. Also the loading on the blades decreases as you move away from the inlet side .

Conclusion:

• The simulation of turbocharger turbine was completed successfully .
• Flow pattern was analysed
• Turbo lag can be calculated based on this results
• Loading effects can be observed and based on this results FEA analysis can be done .