FULL HYDRO case set up (PFI)

AIM

1. To simulate the full hydro case setup of the PFI engine.
2. What is the compression ratio of this engine?
3. Why do we need a wall heat transfer model? Why can't we predict the wall temperature from the CFD simulation?
4. Calculate the combustion efficiency of this engine
5. Use the engine performance calculator, to determine the power and torque for this engine.

INTRODUCTION

In a gasoline engine, fuel is injected in the intake manifold, where it evaporates and mixes with air to form a homogeneous mixture by the time it is compressed in the cylinder. The injection of the fuel in the intake manifold is indicated by the term port fuel injection (PFI).

GEOMETRY

• The geometry was imported in converge and boundary flagging and geometry clean up has been done, to remove any intersection and manifold error and normal.

• Various boundaries are flagged, for simulation. Boundaries which are created are listed below :

• After boundary flagging, the case setup has been done.

CASE SETUP

For application type, a crank-based IC engine is selected. Various parameters associated with the engine is given below :

1. Bore =0.086 m
2. Stroke = 0.09 m
3. Connecting rod length = 0.18 m
4. RPM = 3000

• A transient state density-based solver is selected for the simulation and for algorithm PISO scheme is selected.
• Hydrodynamic simulation mode is selected, for this case.
• Othe simulation parameters considered are :

Start time: -520 deg

End time: 120 deg

Minimum time step: 1e-7 sec

Maximum time step: 0.0001 sec

Materials

• In the IC engine, we are more interested in flow, inside the engine, and reaction inside the combustion chamber.
• For our flow, we have selected IC8H18(Iso-octane), as our fuel as it has the same physical properties as Gasoline fuel. So we have selected parcel species as IC8H18.
• Thermodynamic data of IC8H18 has also been imported to involve more species in reaction mechanisms.

Boundary Conditions

• Boundary conditions for various parts have been defined, for simulation.
• Intake lift and Exhaust have also been provided, for Intake and Exhaust Valve, which will be used to calculate the valve lift.
• Boundary conditions for various parts are shown below:

• For our simulation, we have considered, piston as translating with piston motion provided in converge, and intake valve and exhaust valve are also translation with files provided to then cyclic motion.
• For our simulation we have used, Isoctane(IC8H18), which is similar to gasoline as our primary fuel.
• The inflow boundary condition has a mass fraction of 1, consists of Nitrogen(0.77) and Oxygen(0.23).
• Combustion is stoichiometric, as the outlet consists of CO2, H2O, N2 in a mass fraction of 0.19,0.088, and 0.719.

REACTION :

C8H18+12.5⋅(O2+3.76N2)=8CO2+9H2O+(3.76⋅12.5)N2

Species at Outlet

REGIONS

There are four regions described, which will play the role in the disconnect triangle concept while opening and closing the valve for simulation.

Four regions described are shown below :

• Converge uses events to activate to deactivate the disconnect triangle, which controls when and where flow between the regions is allowed.
• An OPEN event deactivates the disconnected triangle and thus allows flow between the specified regions.
• Close events activate the disconnect triangle and thus prohibit the flow between the disconnect triangle.

We have used two events, one cyclic and one permanent.

Cyclic Event

Cyclic events are used to model the valve opening and closing for inlet and exhaust port for a given referee in every revolution.

Permanent Event

Permanent event is used to model, the port between intake port-1 and intake port -2 as open.

Turbulence modeling

RNG K-Epsilon has been considered for turbulence modeling through the duct. It offers improved accuracy in rotating flows, although there are mixed results in this regard: It has shown improved results for modeling rotating cavities but shown no improvements over the standard model for predicting vortex evolution (both these examples from individual experience). It is favored for indoor air simulations.

Spray Modelling

CONVERGE includes state-of-the-art models for simulating liquid spray phenomena. Spray Modeling in CONVERGE include liquid breakup, collision, and coalescence, vaporization, drag, turbulent dispersion, and drop/wall interaction.

SPRAY SETUP

CONVERGE has two types of liquid injection parameters: nozzles and injectors. An injector is a group of nozzles. In CONVERGE you can set up local coordinate systems, which help you intuitively configure and manipulate the geometry of the nozzles within an injector. In applications such as internal combustion engines, the fuel spray initiates, propagates, and dissipates very quickly and on a very small spatial scale. To dynamically capture the important physics of the injection process, you need a high-density grid around the spray. However, you do not need high resolution far from the spray or before fuel is injected. The automated Adaptive Mesh Refinement (AMR), technology provides the resolution you need exactly when and where you need it: in spray cases, near the injector, and along the path of the spray. The combination of AMR and powerful spray models makes CONVERGE uniquely suited to helping you achieve grid convergent results in a timely manner.

Converge, used the Lagrangian solver to model discrete parcels and the Eulerian solver to model the continuous fluid domain.

We have selected the evaporation model to specify how to converge simulate multi-component vaporization and the NTC collision model.

For fuel spray, four nozzles have been defined, with the following parameters :

Nozzle diameter = 250 micro-meter

Circular injection radius = Nozzle radius

Spray cone angle = 10

• Nozzle 0
  • center 0.0823357 0.00100001 0.07019
  • Align Vector -0.732501 0.210489 -0.647408
• Nozzle 1
  • center 0.0823357 -0.00099999 0.07019
  • Align Vector -0.732501 -0.210489 -0.647408
• Nozzle 2
  • center 0.0823357 -0.0004 0.07019
  • Align Vector -0.5 -0.2 -0.647408
• Nozzle 3
  • center 0.0823357 0.0003 0.07019
  • Align Vector -0.5 0.2 -0.647408

• For spray breakup, the KH+RT breakup model is selected.

Combustion Modelling

Converge uses a combustion model for both premixed and non-premixed combustion. We have used the SAGA combustion model for our simulation.

SAGA detailed Chemical Kinetics Solver

SAGA is a characteristic time Combustion model. The combustion of different fuel can be modeled by changing the mechanisms and SAGA also calculate the reaction rate of each elementary reaction while the CFD solver solves the transport equations. Given an accurate mechanism, SAGA can be used for modeling many combustion regimes, it can be used to model either constant volume or constant pressure.

It used adoptive zoning which accelerates the chemistry calculation by grouping together similar computational cells and then invoking the chemistry solver once per group rather than once per cells.

Source/Sink Modeling

• The source/sink volumes are defined by specifying the shape of the source i.e., the sphere in our case. It contains information about the strength of the source and the start and end time of the source.
• As our simulation consists of two energy sources, at the inlet port, which is cyclic in nature.

Spark Ignition Parameters

Source -1

• Start of spark = -15 deg
• Spark duration = 10 deg
• Spark location = -0.003 0 0.0091
• Spark radius = 0.0005m

Source -2

• Start of spark = -15 deg
• Spark duration = -14.5deg
• Spark location = -0.003 0 0.0091
• Spark radius = 0.0005m

Grid Control

The base mesh size for the duct in both directions has been taken 4mm in all direction of the flow, for the base mesh.

Converge used cut cell cartesian grid generation method, and generate mesh during runtime.

Adaptive Mesh Refinement

• It automatically refines the grid-based upon the fluctuating and moving conditions such as temperature and velocity.
• These features may be used to obtain a refined grid inflow region of interest allowing complex phenomena, such as flame propagation or high-velocity flow, without unnecessarily slowing the simulation with a globally refined grid.

Sub-Grid-Scale(SGS) AMR

For a scalar, the sub-grid field is denied as the difference between the actual field and the resolved field,

The dialog box AMR is shown below :

• Intake port-1, intake port-2, and cylinder are selected for AMR modeling.
• And for adaptive mesh refinement, velocity and temperature sub-grid-scale parameters are selected.

Fixed Embedding

• Fixed embedding is used to refine the grid at a specific location in the domain where a finer resolution is critical to the accuracy of the solution.
• Fixed embedding allows the rest of the grid to remain coarser to resolve the complex flow behavior. For example, when simulating sprays you can add an area of fixed embedding by the nozzle to resolve the complex flow behavior.
• An embedding is provided for intake and the exhaust valve angle provides 0.5 mm cells as mesh refinement. An additional embedding is given for our spark plug so when energy is sourced into spark plug the combustion propagates.
• The below image shows the fixed embedding in the shape of a cylinder on the PFI engine:

• For each fixed embedding, we need to specify an embedding scale that indicates how CONVERGE will refine the grid in that location. The scale parameters can be defined by:

`\embed = (\frac{base grid}{2^text(embedscale)})`

• For our flow, we have selected grid scaling of -2, as we have fewer resources to handle a large number of resources.
• After the case file has written, in a specific folder where simulation can be run.

RUNNING SIMULATIONS

• Once the case setup is done and files are exported and saved, then we will use external solver Cygwin to solve the problems using these input files.
• The command used to execute the solving process is shown below :

mpiexec.exe -n 4 "C:\Program Files\Convergent_Science\CONVERGE\3.0.12\bin\converge-intelmpi.exe" restricted

• This command should we run in the same folder where all the case data files are.
• This command is used to solve the problem in parallel with 4 cores.
• When the simulation is completed and converged, the header output files are converted to a suitable format for post-processing.

POST-PROCESSING

1. What is the compression ratio of this engine?

The compression ratio of the engine is defined as the ratio of the total volume of the cylinder to clearance volume.

Compression ratio,`\r_c = \frac{\text(Maximum Cylinder Volume)}{\text(Minimum Cylinder Volume)} = \frac{V_d + V_c}{V_c}`

where ,`V_max = V_d +V_c`, maximum cylinder volume, m^3

`\V_min`,minimum cylinder volume,m^3

So for our simulation,

`r_c = \frac{0.000574}{5.732e-5} = 10`

2. Why do we need a wall heat transfer model? Why can't we predict the wall temperature from the CFD simulation?

Modeling of the heat transfer and wall temperature is an important task concerning the accuracy and the predictions of any engine thermodynamic model, although it is many times an overcome task. In order to perform good prediction of engine heat transfer and wall temperature, models are required for accomplishing heat transfer from hot gases to engine parts, heat transfer inside each engine part, and also heat transfer to coolant and lubricating oil.

A spark ignites the fuel and air constituents when compressed inside the cylinder, which emits the flames that spread throughout the cylinder. After a few steps, in our simulation we see the flames vanish completely as soon as it makes content with liner and cylinder head, which makes sense for the head to be transferred to the wall. But as our flow simulation doesn't consider the solid regions for stream ID flow in region setup, we only make use of N-S equation, Scalar transport equations and turbulence, which all govern flow for fluid flows.

These two reasons are sighted for making use of wall heat transfer models to predict temperature ., and our CFD setup doesn't consider solid that's why it will not predict temperature accurately.

Combustion efficiency

Combustion efficiency is a measurement of how well the fuel being burned is being utilized in the combustion process. Complete combustion would result in 100 percent efficiency, but that is not possible, as there are some losses.

For our simulation, the process is calculated as :

Mass of fuel injected = 3e-05 kg

Calorific Value of Iso-octane = 44MJ/kg

Combustion Efficiency, `\eta_c = \frac{text(Total Heat Release)}{text(Total Energy Content of fuel)}`

or, it can be written as ,`\eta_c = \frac{text(Total heat Release)}{text(m x LCV of fuel}`

So, `\eta_c = \frac{1241.18}{3e-05 X 44e06 }`

`\eta_c = \94 %`

Power and torque :

Engine Performance Calculator :

From the given engine performance calculator, we can see that :

Work = 468.646 N-m

Duration = 240.199 deg

From our engine parameters :

Rotation per minutes ,RPM = 3000

Rotation per seconds = 3000/60 = 50

Single rotation of piston = 360 degree

Roration per seconds = 50 * 360 = 18000deg/sec

Time taken for rotation of piston = 240.199 / 18000 = 0.01334 sec

Power = `\frac{text(Work Done)}{text(Time taken)}`

= `\frac{468.646}{0.01334} = \35130.885 W or \35.13 KW`

Torque :

Torque is the twisting or rotating force that the engine exerts on the crankshaft; it is expressed in N-m or Kg-m. It is generally measured at a radius of one meter from the crankshaft's center. Here is the torque-speed curve. As the RPM Of an engine increases, torque increases up to a certain value of RPM.

Torque,`\tau = \frac{Power \text(*) 60}{2 \text(*)pi text(*)N}`

`\tau = \frac{35130.885 \text(*) 60}{2 \text(*)pi text(*)3000} = \111.83 N-m`

5. What is the significance of ca10, ca50, and ca90?

Ca10, Ca50, and Ca90 represent the various stage of the crank angle at the combustion stage.

Ca10, defines when 10 % of combustion is done = 6.83717 deg

Ca50 , defines when 50 % of combustion is done = 18.4623 deg

Ca90, defines when 90 % of combustion is done = 31.701 deg

The physical significance of CA 10 is that's identified as the start of the ignition in an engine and CA 90 represents the end of combustion in the engine.

Plot

Heat Release Rate vs Crank Angle

The given plot shows the Heat release in the engine, as a part of combustion chamber modeling. We can see the heat release is zero almost most of the crack angle and is maximum between 0 to 20 crank angle e that where we have provided heat two heat source in our model with source/sink modeling in Converge.

Emission Parameters vs Crank Angle

From the given plot we can see that the emission of most of the gases is very less other than carbon dioxide, which is more. These exhaust gases are measured in the exhaust pipe. The emission can be decreased by increasing the efficiency which in turn requires better combustion of fuel.

Animation

Mesh Change with Moment of Piston -Cut Out Plane

From the given animation we can see that the change in the mesh when the valve moves with changing the crack angle, is due to Adaptive zoning in this region provided by the Converge.

Fuel Injection Animation

The given animation shows the spray of fuel inside the cylinder at given crank angle and valve timing. We can see as the inlet valve open the spray enter the cylinder after that the inlet valve closes and at exhaust stroke, the exhaust valve is open which is not modeled here.

Conclusion :

• We have successfully performed a Full Hydro simulation on the S18 PFI engine.
• The compression ratio of the engine in 10, and the combustion efficiency is 94 %, Power of the engine is calculated is 35.13 KW and torques are 111.8 N-m.
• Animation of the mesh and injector of fuel is also embedded in the project for a better understanding of the valve opening.
• We can't validate the result because of the lack of experimental data for it.