**4. KIVA-3V simulations of work of the test engine 2SZ-FE**

The carried out simulations were focused on the determination and comparison of the differences in the combustion process in the cylinders of the engine working with the portand dual-injection of fuel in conditions similar to occurring at experimental research.

In order to determination of the phenomena occurring in the cylinder, computer simulations were performed in programme KIVA-3V. Used for three-dimensional modelling of processes in internal combustion engines KIVA-3V program takes into account the physical and chemical phenomena that occur during forming of the mixture and its combustion [17,18]. The pro‐ gramme takes into account movement of fuel droplets and their atomization in the air using a stochastic model of the injection.

KIVA-3V has the ability to simulate the engine operation using different fuels. In the described work a hydrocarbon with the chemical formula C8H17 was used as the fuel. One can see similarities to octane (C8H18), however, this substance have more comparable proportions of carbon and hydrogen in the molecule to the petrol than octane. Therefore, it can be regarded as a special kind of single-component petrol. The C8H17 fuel is oxidized according to the reaction (7).

$$4\text{C}\_8\text{H}\_{17} + 49\text{O}\_2 \rightarrow 32\text{CO}\_2 + 34\text{H}\_2\text{O} \tag{7}$$

equal thickness falls to 81% of piston stroke starting from the bottom dead centre. The remaining 14 layers around the top dead centre was concentrated to obtain more advantageous terms of the simulation of combustion process that takes place there (combustion chamber). The cylinder mesh has transverse-sectional dimensions respectively 38 x 34. It gives together

Combustion Process in the Spark-Ignition Engine with Dual-Injection System

http://dx.doi.org/10.5772/54160

73

Used in the research an engine model was developed based on available technical data of 2SZ-FE engine. Dimensions required to generate a grid, especially cylinder head and valves lift

In both of the simulation, with indirect fuel injection and dual-injection of fuel in both simulations conditions, such as occurring during research which results were presented in Figure 14, were maintained. In the case of simulation of engine with dual-injection of fuel the whole amount of fuel was divided between indirect injection systems and direct, so that the fraction of direct injection xDI was 0.62. At this fraction the engine gained the best value of the total efficiency. List of crucial assumptions and sub-models used in the simulations were

**Parameter/Sub-model MPI DI+MPI**

were obtained by direct measurement of elements of the modified engine.

presented in Table 4, respectively for indirect- and dual-injection of fuel.

Intake valve opening/closing 4° CA before TDC / 46° CA after BDC

Mass of fuel injected into the intake channel 0.01610 g/cycle 0.01061 g/cycle Mass of fuel injected into the cylinder - 0.00600 g/cycle Whole mass of fuel 0.01610 g/cycle 0.01661 g/cycle

Start of injection into the cylinder - 281°CA before TDC

Composition of the mixture stoichiometric Intake manifold absolute pressure 0.079 MPa Engine rotational speed 2000 RPM

Start of injection into the intake channel 360° CA before TDC

Ignition angle 14° CA before TDC Total time of a spark discharge 1.33 ms / 16° CA Ambient absolute pressure 0.097 MPa Backpressure in the exhaust channel 0.110 MPa Temperature of the cylinder sleeve (constant) 450 K Temperature of the cylinder head (constant) 500 K Temperature of the piston crown (constant) 530 K Model of fuel injection Reitz

around 45000 cells in the whole cylinder volume.

**4.1. Initial and boundary conditions for simulations**

Fuel oxidation described by chemical equation (1) is a basic chemical reaction that occurs during the simulation in the program KIVA-3V. Other processes important for the simulation take place according to the formulas (8) to (10).

$$\text{N}\_2\text{+O} \rightarrow \text{N} + \text{NO} \tag{8}$$

$$\text{N} + \text{O}\_2 \rightarrow \text{O} + \text{NO} \tag{9}$$

$$\text{N} + \text{OH} \rightarrow \text{H} + \text{NO} \tag{10}$$

A set of reactions (2) - (4) describes a so-called thermal mechanism of formation of nitric oxide, which occurs at high temperatures, e.g. in the conditions occurring in the combustion chamber of the engine. From the name of Russian scientist Yakov Borisovich Zeldovich, who described this mechanism, in the literature it is often referred to as the extended Zeldovich mechanism.

Preparations for simulations have included a generation of a mesh of one of the engine's cylinder and modifying the source code of KIVA-3V in order enable the simulation of work with both of the fuel injectors at the same time, what in the basic version of the program is not possible. The computational mesh was built based on the results of previous positively verified solutions in that matter. The grid consists of a cylinder 35 of horizontal layers. 21 layers of equal thickness falls to 81% of piston stroke starting from the bottom dead centre. The remaining 14 layers around the top dead centre was concentrated to obtain more advantageous terms of the simulation of combustion process that takes place there (combustion chamber). The cylinder mesh has transverse-sectional dimensions respectively 38 x 34. It gives together around 45000 cells in the whole cylinder volume.

Used in the research an engine model was developed based on available technical data of 2SZ-FE engine. Dimensions required to generate a grid, especially cylinder head and valves lift were obtained by direct measurement of elements of the modified engine.

## **4.1. Initial and boundary conditions for simulations**

**4. KIVA-3V simulations of work of the test engine 2SZ-FE**

a stochastic model of the injection.

72 Advances in Internal Combustion Engines and Fuel Technologies

take place according to the formulas (8) to (10).

reaction (7).

The carried out simulations were focused on the determination and comparison of the differences in the combustion process in the cylinders of the engine working with the port-

In order to determination of the phenomena occurring in the cylinder, computer simulations were performed in programme KIVA-3V. Used for three-dimensional modelling of processes in internal combustion engines KIVA-3V program takes into account the physical and chemical phenomena that occur during forming of the mixture and its combustion [17,18]. The pro‐ gramme takes into account movement of fuel droplets and their atomization in the air using

KIVA-3V has the ability to simulate the engine operation using different fuels. In the described work a hydrocarbon with the chemical formula C8H17 was used as the fuel. One can see similarities to octane (C8H18), however, this substance have more comparable proportions of carbon and hydrogen in the molecule to the petrol than octane. Therefore, it can be regarded as a special kind of single-component petrol. The C8H17 fuel is oxidized according to the

Fuel oxidation described by chemical equation (1) is a basic chemical reaction that occurs during the simulation in the program KIVA-3V. Other processes important for the simulation

A set of reactions (2) - (4) describes a so-called thermal mechanism of formation of nitric oxide, which occurs at high temperatures, e.g. in the conditions occurring in the combustion chamber of the engine. From the name of Russian scientist Yakov Borisovich Zeldovich, who described this mechanism, in the literature it is often referred to as the extended Zeldovich mechanism. Preparations for simulations have included a generation of a mesh of one of the engine's cylinder and modifying the source code of KIVA-3V in order enable the simulation of work with both of the fuel injectors at the same time, what in the basic version of the program is not possible. The computational mesh was built based on the results of previous positively verified solutions in that matter. The grid consists of a cylinder 35 of horizontal layers. 21 layers of

8 17 2 2 2 4C H + 49 O 32 CO + 34 H O ® (7)

N + O N + NO <sup>2</sup> ® (8)

N + O O + NO <sup>2</sup> ® (9)

N + OH H + NO ® (10)

and dual-injection of fuel in conditions similar to occurring at experimental research.

In both of the simulation, with indirect fuel injection and dual-injection of fuel in both simulations conditions, such as occurring during research which results were presented in Figure 14, were maintained. In the case of simulation of engine with dual-injection of fuel the whole amount of fuel was divided between indirect injection systems and direct, so that the fraction of direct injection xDI was 0.62. At this fraction the engine gained the best value of the total efficiency. List of crucial assumptions and sub-models used in the simulations were presented in Table 4, respectively for indirect- and dual-injection of fuel.



**Table 4.** List of crucial assumptions and sub-models used in the simulations

#### **4.2. Comparison of selected simulation results for both of the fuel systems**

Figure 21 shows traces of pressure in the cylinder pc as a function of a cylinder volume in the case of indirect fuel injection and during work with the dual-injection system.

represented on the chart by achieving zero by the curve of the green (mass of liquid fuel) and the maximum of the curve of the blue (mass of fuel vapour), which takes place about 120° CA

**Figure 22.** The change of the fuel mass as a function of a crank angle for the engine's operation with dual-injection

Combustion Process in the Spark-Ignition Engine with Dual-Injection System

http://dx.doi.org/10.5772/54160

75

The angular momentum of the charge Ktot is a indicator of the intensity of swirl and tumble turbulence in the cylinder, which affect the intensity of the evaporation of the fuel, its propa‐ gation within the cylinder volume, and consequently the speed of the flame spread. Traces of

**Figure 23.** The total angular momentum of charge Ktot in a function of a crank angle for both of the considered fuel

One can see the impact of fuel stream injected directly into the cylinder on charge. In the case of dual-injection of fuel the angular momentum in the intake and compression process achieves values greater than is the case of fuel injection only into intake channel. Intensification of the

systems

before TDC, while the ignition timing in the simulation was assumed at 14° CA.

system and with the port fuel injection

the total angular momentum of the cylinder charge were shown in Figure 23.

**Figure 21.** The traces of pressure in the cylinder as a function of a cylinder volume for both of the fuel systems: MPI and DI + MPI

**•** One can see the difference in the value of the peak pressure and a slightly larger area under the curve of pressure in the cylinder of engine working with dual- injection system.

Figure 22 presents the change of the fuel mass as a function of crank angle for both of the considered injection systems.

In the case of fuel injection only to the intake channel in the considered period of time in the cylinder exist only the fuel vapours. When the dual-injection system is used, the fuel injected directly into the cylinder evaporates completely before the moment of ignition. This fact is Combustion Process in the Spark-Ignition Engine with Dual-Injection System http://dx.doi.org/10.5772/54160 75

**Parameter/Sub-model MPI DI+MPI**

Model of droplet breakup Taylor Analogy of Breakup

Model of heat transfer Improved Law-of-the-Wall

**4.2. Comparison of selected simulation results for both of the fuel systems**

case of indirect fuel injection and during work with the dual-injection system.

Figure 21 shows traces of pressure in the cylinder pc as a function of a cylinder volume in the

**Figure 21.** The traces of pressure in the cylinder as a function of a cylinder volume for both of the fuel systems: MPI

**•** One can see the difference in the value of the peak pressure and a slightly larger area under the curve of pressure in the cylinder of engine working with dual- injection system.

Figure 22 presents the change of the fuel mass as a function of crank angle for both of the

In the case of fuel injection only to the intake channel in the considered period of time in the cylinder exist only the fuel vapours. When the dual-injection system is used, the fuel injected directly into the cylinder evaporates completely before the moment of ignition. This fact is

Model of combustion Mixing-Controlled Turbulent Combustion NO formation extended Zeldovich mechanism (thermal)

Model of droplet evaporation Spalding

74 Advances in Internal Combustion Engines and Fuel Technologies

Number of considered chemical species 12

and DI + MPI

considered injection systems.

**Table 4.** List of crucial assumptions and sub-models used in the simulations

Model of wall impingement Naber and Reitz Model of turbulence standard k-ε

**Figure 22.** The change of the fuel mass as a function of a crank angle for the engine's operation with dual-injection system and with the port fuel injection

represented on the chart by achieving zero by the curve of the green (mass of liquid fuel) and the maximum of the curve of the blue (mass of fuel vapour), which takes place about 120° CA before TDC, while the ignition timing in the simulation was assumed at 14° CA.

The angular momentum of the charge Ktot is a indicator of the intensity of swirl and tumble turbulence in the cylinder, which affect the intensity of the evaporation of the fuel, its propa‐ gation within the cylinder volume, and consequently the speed of the flame spread. Traces of the total angular momentum of the cylinder charge were shown in Figure 23.

**Figure 23.** The total angular momentum of charge Ktot in a function of a crank angle for both of the considered fuel systems

One can see the impact of fuel stream injected directly into the cylinder on charge. In the case of dual-injection of fuel the angular momentum in the intake and compression process achieves values greater than is the case of fuel injection only into intake channel. Intensification of the cylinder charge turbulence has undoubtedly an important influence on the improvement of the combustion process, and thus, to increase the torque of the engine.

Figure 25 shows the distribution of the mass fraction of fuel in the longitudinal section of the

Combustion Process in the Spark-Ignition Engine with Dual-Injection System

http://dx.doi.org/10.5772/54160

77

The stream of fuel injected directly into the engine cylinder is clearly visible in Figure 25 b.

The distributions of mass fraction of hydroxyl radicals OH in the longitudinal cross-section of the cylinder at crank angle 5º before TDC obtained by simulations carried out for both fuel

**Figure 26.** The distribution of mass fraction of hydroxyl radicals OH in the longitudinal cross-section of the cylinder at crank angle 5º before TDC obtained by simulation with port fuel injection (a) and with dual-injection of fuel (b)

**•** On the basis of the analysis of Figure 26 it can be concluded that the combustion develops in the initial stage significantly faster when the mixture is formed by two injectors per cylinder.

**•** The temperature distribution in the cylinder at the crank angle 24° after TDC is presented

**Figure 27.** The temperature distribution in the cylinder at the crank angle 24° after TDC for port fuel injection (a) and

cylinder in the intake stroke for the each considered fuel system.

systems were shown in Figure 26.

in Figure 27 for both considered fuel systems.

dual-injection of fuel (b)

On the Figure 24 mass fraction of hydrocarbons HC, carbon monoxide CO and nitrogen oxide NO in the cylinder were shown as a function of crank angle for indirect injection and for the dual-injection of fuel.

**Figure 24.** The mass fraction of HC, CO i NO in the cylinder in a function of crank angle for both systems of fuel supply

On the basis of analysis of the graphs contained in Figure 24 it can be concluded that there are some differences in the formation of carbon monoxide CO, hydrocarbons HC and nitrogen monoxide NO depending on the concerned injection system. After end of the combustion in the cylinderofthe engineworkingwithindirectfuelinjectionthere is slightlymoreCOandNOthan in the case when the amount of fuel divided between the two injection systems. When the fuel is injected by two injectors a fraction of unburned hydrocarbons is higher than with the indirect injection. The difference amounts to about 80 ppm, so it is not a significant disadvantage.

**Figure 25.** The distribution of the mass fraction of fuel in the longitudinal section of the cylinder in the intake stroke for indirect fuel injection(a) and for dual-injection(b) crank angle – 250º CA before TDC

Figure 25 shows the distribution of the mass fraction of fuel in the longitudinal section of the cylinder in the intake stroke for the each considered fuel system.

cylinder charge turbulence has undoubtedly an important influence on the improvement of

On the Figure 24 mass fraction of hydrocarbons HC, carbon monoxide CO and nitrogen oxide NO in the cylinder were shown as a function of crank angle for indirect injection and for the

**Figure 24.** The mass fraction of HC, CO i NO in the cylinder in a function of crank angle for both systems of fuel supply

On the basis of analysis of the graphs contained in Figure 24 it can be concluded that there are some differences in the formation of carbon monoxide CO, hydrocarbons HC and nitrogen monoxide NO depending on the concerned injection system. After end of the combustion in the cylinderofthe engineworkingwithindirectfuelinjectionthere is slightlymoreCOandNOthan in the case when the amount of fuel divided between the two injection systems. When the fuel is injected by two injectors a fraction of unburned hydrocarbons is higher than with the indirect injection. The difference amounts to about 80 ppm, so it is not a significant disadvantage.

**Figure 25.** The distribution of the mass fraction of fuel in the longitudinal section of the cylinder in the intake stroke

for indirect fuel injection(a) and for dual-injection(b) crank angle – 250º CA before TDC

the combustion process, and thus, to increase the torque of the engine.

76 Advances in Internal Combustion Engines and Fuel Technologies

dual-injection of fuel.

The stream of fuel injected directly into the engine cylinder is clearly visible in Figure 25 b.

The distributions of mass fraction of hydroxyl radicals OH in the longitudinal cross-section of the cylinder at crank angle 5º before TDC obtained by simulations carried out for both fuel systems were shown in Figure 26.

**Figure 26.** The distribution of mass fraction of hydroxyl radicals OH in the longitudinal cross-section of the cylinder at crank angle 5º before TDC obtained by simulation with port fuel injection (a) and with dual-injection of fuel (b)


**Figure 27.** The temperature distribution in the cylinder at the crank angle 24° after TDC for port fuel injection (a) and dual-injection of fuel (b)

**•** It can be seen that at the end of the combustion process slightly higher temperature in the cylinder volume is observed in the case of dual-injection of fuel.

**•** With the dual-injection fuel system in the analyzed operating conditions of the engine a few percent increase in total efficiency was obtained, what in the present state of development of the internal combustion engines is an important value. This fact clearly indicates the

Combustion Process in the Spark-Ignition Engine with Dual-Injection System

http://dx.doi.org/10.5772/54160

79

**•** The analysis of indicator diagrams registered for work with indirect fuel injection and dualinjection of fuel revealed increase in the indicated mean effective pressure and improve

**•** There were no significant changes in the composition of the exhaust gas together with changing the fraction of fuel injected directly into the cylinders. In comparison with the values obtained for indirect fuel injection with the increase in the fraction of fuel injected directly into the cylinder occurs the reduction of nitric oxide concentration with a slight

**•** In view of the total efficiency the optimal value of fraction of fuel injected directly into the

In view of the results of above described tests authors can present topics for further research

**•** Analysis of application of the described fuel system for the formation of stratified lean

**•** Study the impact of the application of dual-injection system on the working parameters of

**•** Assessment of the impact of application of the forming of the mixture according to sprayguided model on working parameters of the engine with dual-injection fuel system

Regarding the concept of Toyota company, it seems there is a future in D-4S injection system. Besides mentioned in introduction 2GR-FSE, after 2005 the D-4S System is used in 4.6 L 1UR-FSE as well as 5.0 L 2UR-FSE and 2UR-GSE V8-engines mounted to various Lexus cars [19]. Since 2012 the FA20 four-cylinder opposed-piston Subaru engine used in Toyota GT86/Scion FS-R car and called 4U-GSE is also equipped with the D-4S dual-injection fuel system.

cylinder grows when increasing the engine load at specified rotational speed,

desirability of conducting research related to the taken issues.

increase in the concentration of carbon monoxide and hydrocarbons.

engine thermal efficiency with dual-injection of fuel.

**6. Future of the dual-injection system**

**Abbreviations and nomenclature**

ε –speed of dissipation of the kinetic energy of turbulence

αthr – opening of the throttle, [%],

the engine burning a quasi-homogeneous lean mixtures,

actions related to the subject:

mixtures,

α–crank angle, [°]
