Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM An Open Cycle Simulation of DI Diesel Engine Flow Field Effect on Spray Processes 2012-01-0696 Published 04/16/2012 Kohei Fukuda, Abbas Ghasemi, Ronald Barron and Ram Balachandar University of Windsor Copyright © 2012 SAE International doi:10.4271/2012-01-0696 ABSTRACT Clean diesel engines are one of the fuel efficient and low emission engines of interest in the automotive industry. The combustion chamber flow field and its effect on fuel spray characteristics plays an important role in improving the efficiency and reducing the pollutant emission in a direct injection diesel engine, in terms of influencing processes of breakup, evaporation mixture formation, ignition, combustion and pollutant formation. Ultra-high injection pressure fuel sprays have benefits in jet atomization, penetration and air entrainment, which promote better fuel-air mixture and combustion. CFD modeling is a valuable tool to acquire detailed information about these important processes. In this research, the characteristics of ultra-high injection pressure diesel fuel sprays are simulated and validated in a quiescent constant volume chamber. A profile function is utilized in order to apply variable velocity and mass flow rate at the nozzle exit. The CFD model is also applied to an open cycle engine model to study the effects of engine flow field features such as swirl and tumble motions on the spray behavior. In particular, the effect of the above mentioned parameters on spray penetration Sauter mean diameter (SMD) and fuel distribution in the chamber are extensively discussed. INTRODUCTION Demands on clean diesel engines are continuously growing in the automotive sector. One of the solutions advanced in diesel engine applications is to improve fuel droplet atomization and air-fuel mixing. The development of high pressure injectors will result in finer atomization and provide for better air-fuel mixture [1]. Recently, the injection pressure of the injection systems has reached to 300 MPa or above in these applications and a growing number of researchers have shown interest in the performance of the ultra-high injectors. Kato et al. [2] and Yokota et al. [3] have reported on experiments in the ultra-high injection system at the early stage of the research. They examined the effects of the injection pressure, ranging from 55 to 250 MPa, and also the variations of nozzle orifice and injection duration. From their studies they concluded that the Sauter mean diameter is correlated with the average injection pressure and the transition of the injection pressure by time. Moreover, a shorter combustion process and reduced soot formation are realized by utilizing ultra-high injection pressure and smaller orifice diameter. Nishida's research group of University of Hiroshima has conducted numerous experiments using various ultra-high injection pressures, micro-hole nozzles, spray wallimpingement, and diesel and alternative diesel fuels [4,5,6,7,8,9,10]. The combination of 300 MPa injection pressure and 0.08 mm nozzle-hole diameter reportedly gave the best performance in terms of turbulent mixture rate and droplet size reduction to decrease the mixture process and lean mixture formation. Lee et al. [11] experimentally and numerically investigated free sprays at ultra-high injection pressure in the range of 150 to 355 MPa. The limitation of Sauter mean diameter after reaching an injection pressure of 300 MPa and the reduced growth rate of the penetration length are reported. Tao and Bergstrand [12] studied the effect of ultra-high injection pressures on engine ignition and combustion using three-dimensional numerical simulations. The advantage of high pressure injection in producing reduced ignition delay, short combustion phase, and fast flame propagation is reported. Additionally, three different rates of injection profiles were examined. Rate falling injection results were found to shorten at the early stage of combustion and expand Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM at the late stage, and rate rising injection performs inversely. On the other hand, rate rising injection estimates a wider flame area at high temperature and reduces NO formation due to faster cooling after combustion. Flame lift-off lengths were observed to be constant at different injection pressures, in contrast to the case of injection in a constant volume chamber. To study the characteristics of ultra-high injection pressure sprays numerically, it is important to understand the effect of spray modeling. Comprehensive reviews of droplet phenomena have been presented by Lin and Reitz [13] and Jiang et al. [14]. The difference between the popular breakup models has been discussed by Djavareshkian and Ghasemi [15] and Hossainpour and Binesh [16]. They reported on implementation of WAVE and KH-RT models, and found better agreement with experimental data using KH-RT. The interaction of the mesh, turbulence model and spray has been studied by Karrholm and Nordin [17] in a constant volume chamber. The effect of spray-in-cylinder-flow interaction is realized in the combustion process [18] and studied by a number of researchers. Choi et al. [19] found that the flow pattern around the jet is similar at different injection pressures, but strong flow recirculation is observed at higher injection pressure. Spray characteristics in crossflow was studied by Desantes et al. [20], McCracken and Abraham [21], and Park et al. [22] to observe the effect on particle size and mixing process. Correlation of penetration and dispersion of gas jet and sprays were examined by Iyer and Abraham [23]. The effects of gas density and vaporization on penetration, injection condition and dispersion of spray are discussed by Naber and Siebers [24], Kennaird et al. [25], and Post et al. [26]. Jagus et al. [27] assessed injection and mixing using LES turbulence modeling. Not many numerical studies have been conducted with ultrahigh injection pressures. The objective of this research is to take advantage of numerical simulation to investigate the effect of high pressure injection on atomization and fuel mixing with in-cylinder flow. As a first step, the numerical setup is optimized and validation of the spray model is achieved in the constant volume vessel with different grid sizes. The experimental setup and data for the validation are acquired from Wang et al [7]. In the next stage, an engine model with vertical ports is meshed and the flow structure is verified. Finally, the spray models are introduced into the engine simulation with three different injection pressures and two inlet pressure cases. The spray characteristics are investigated and the interaction of the in-cylinder flow with the sprays is studied. NUMERICAL METHODOLOGY MESHING An engine model to be simulated in ANSYS FLUENT 13.0 [28] is meshed with a hybrid topology (half-model shown in Fig. 1). The flow domain is divided into four major zones: chamber, ports, piston layer valve layer. The zones adjacent to reciprocating boundaries such as the piston and valves are meshed with quadrilateral cells (structured mesh). Tetrahedral cells (unstructured mesh) are used in the chamber zone because the valves move into this zone and its cells deform and must subsequently be remeshed. Interfaces must be created between the chamber and valve layer zones to transfer nodal values from one side to the other. Figure 1. Engine geometry with vertical ports, illustrating the mesh at BDC SPRAY AND BREAKUP MODELS In CFD, spray mechanisms are represented by mathematical models. Two approaches, Euler-Lagrange and Euler-Euler, are used in multiphase flows. In both of these approaches, the fluid phase is regarded as a continuum and modeled by the Navier-Stokes equations. For the Euler-Lagrange approach, the Lagrangian discrete phase model is introduced in general CFD codes to calculate the disperse phase by tracking particles, droplets, or parcels [29]. The trajectories of particles in a turbulent flow field are predicted by the turbulent dispersion models. To reduce the computational time of the particle collision calculation, the O'Rourke algorithm is employed [30]. The outcomes of collisions are also determined by this algorithm i.e., whether the droplets coalesce or reflect apart [26]. Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM The hydrodynamic model, in which the diffusion of droplet only controls its vaporization, or the kinetic model, which is concerned with the molecules' detachment from the surface of droplet, is utilized in this study [31]. The disintegration of existing droplets is modeled to numerically simulate different kinds of breakup modes. WAVE, or Kelvin-Helmholtz (KH) instability models [32], and Kelvin-Helmholtz-RayleighTaylor (KH-RT) instability model [33], known as a hybrid model, are favored in high speed high Weber number (We >100) fuel-injection models. On the other hand, the Taylor Analogy Breakup (TAB) is more commonly used in low speed and low Weber number flows. KH-RT incorporates the effects of aerodynamic breakup and instabilities of droplet acceleration; thus, it is capable of handling both TAB and WAVE models. Recently, there have been many hybrid models developed by combining different breakup models to estimate the spray characteristics accurately over a variety of Weber number [22, 34]. In this study, due to the high Weber number condition, WAVE and KH-RT model are most suitable. In the next subsections, both of these models are discussed in details to estimate the coefficients of these models for good modeling. WAVE Model Reitz [32] developed a model called WAVE based on droplet breakup due to the relative velocity between the gaseous and liquid phases. The model is formulated from the KelvinHelmholtz instability's wavelength and growth rate to determine the size of the droplets. The model is limited by Weber number. We, which must be larger than 100 so that Kelvin-Helmholtz instability is dominant in droplet breakup. The maximum wave growth rate (also the most unstable surface wave), Ω, and corresponding wavelength, Λ, are defined as: (1) (2) where , , , , and . Z is the Ohnesorge number, T is the Taylor number and r0 is the radius of the undisturbed jet. We and Re are the Weber number and the Reynolds number and subscripts l and g represent liquid and gas phase, respectively. The radius of a newly formed droplet from a parent droplet during the breakup process is assumed proportional to the wavelength ΛKH. (3) The constant B0 is set equal to the experimentally determined value of 0.61. Additionally, the rate of change of the parent droplet is defined by (4) where the breakup time, τKH, is given by (5) The breakup time constant B1 is an adjustable variable which is recommended to be in the range of 1.73 to 60 [32, 35, 36]. Larger value of B1, produces fewer breakups and more penetration. An estimation of the B1, factor has been proposed by Liu et al. [37]: (6) However, Liu et al. [37] conclude that Eq. (6) does not determine the valus of this constant qualitatively and recommend that it should be used only for reference value as the first guess. Kelvin-Helmholtz-Rayleigh-Taylor (KH-RT) Model The KH-RT instability model is a combination of KelvinHelmholtz (KH) instability and Ravleigh-Tavlor (RT) instability models. Both Kelvin-Helmholtz and RavleighTaylor models decide droplet breakup by detecting the fastest growing surface wave on the droplets. The source of the Kelvin-Helmholtz wave is induced by aerodynamic forces between gas and liquid phases, whereas the Rayleigh-Taylor wave is the result of acceleration of shed drops ejected into free-stream conditions. Hwang et al. [33] showed in their experiments the sequential breakup process in the catastrophic breakup regime. A droplet first gets flatten by the aerodynamic force on it and breaks up due to the deceleration of the sheet droplet by means of Rayleigh-Taylor instability model. Further breakups proceeded by the smaller wavelength of Kelvin-Helmholtz wave found at the edge of the fragments above. In high Weber number cases with high droplet acceleration, Rayleigh-Taylor instability grows faster and dominates the breakup of droplets. For the numerical model, both instability models are utilized simultaneously and breakups are determined by the fastest growth rate of Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM waves. In the Rayleigh-Taylor model, the fastest growing wave's growth rate and its corresponding wavelength are given by Moreover, Senecal [40] analytically determined the relationship of the Levich constant, CL, and the breakup time constant of the WAVE model, B1. Considering the breakup length of the WAVE model to be LKH = U ·τKH and assuming that the viscosity is zero, Eq. (5) reduces to (7) (13) (8) and therefore where a is the droplet acceleration, CRT is the RayleighTaylor breakup constant and KRT is the wave number given by (14) Comparing Eq. (11) and (14), the correlation of the coefficients is given by (9) Breakup time in the Rayleigh-Taylor model is defined as (10) where Cτ is the breakup time constant and the liquid core length, LRT, is obtained from Levich theory [38]: (11) The radius of the new droplets is calculated as half of the wavelength obtained above . Based on their experiments, Hiroyasu and Arai [39] proposed the correlation of density and diameter to the Levich constant, CL, as (12) (15) As this relation shows, the Levich constant is also adjustable and ranges from 5 to 20. In addition, Patterson and Reitz [35] investigated the effects of Rayleigh-Taylor breakup constant, CRT, in the range 1.0 ≤ CRT ≤ 5.33. GRID INDEPENDENCE Initially, grid independency tests were conducted to optimize the grid size in the engine simulation. The simulations were performed on a 60 mm (D) × 60 mm (W) × 80 mm (H) fixed chamber with similar conditions as Wang's experiments [7]. Five different sets of grids, 70 × 70 × 105 (0.5 million), 87 × 87 × 130 (1.0 million), 93 × 93 × 140 (1.2 million), 100 × 100 × 150 (1.5 million), and 106 × 106 × 160 (1.8 million) nodes, were generated. Figure 2 shows the results for a 300 MPa spray injection pressure with the five different meshes. It is seen that 0.5 million cells over-estimates the penetration significantly. Moreover, the domain with 1.0 million cells slightly over-predicts the penetration in comparison with the rest of the grids. From this information, we concluded that, in the case of ultra-high injection spray model, the domain should be meshed with 93 × 93 × 140 cells or finer to avoid any effects of grid size on penetration length. For our subsequent calculations, the optimal grid size is chosen to be 93 × 93 × 140, with the largest grid size set at 0.65 mm, which is around 60 mm divided by 93, in the chamber of the engine model. Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM X and negative Y directions. For convenience, the injectors are referred to as INJ-0, INJ-90, INJ-180, and INJ-270, respectively. Nozzle geometry and injection conditions are summarized in Table 3. Table 1. Engine operation conditions and initial setups Figure 2. Spray tip penetration results for the grid independency tests Table 2. Diesel fuel properties COMPUTATIONAL DOMAIN SETUP The geometry utilized in this study is obtained from the KIVA3V manual [41]. The engine is sized 82.55 mm in bore and 92.075 mm in stroke and a high compression ratio of 17.2:1 is reached by the design. Its connecting rod length is 174 mm and it is operated at 1500 rpm. Other engine operating conditions are listed in Table 1. The engine model is set at top dead center (TDC) position initially by the requirement of FLUENT. The geometry is appropriately meshed with a hybrid mesh and sized by the optimum size obtained from the previous section. The maximum number of cells for the engine geometry is 166,500 cells at TDC and 570,000 cells when the piston surface reaches the bottom dead center (BDC). All wall boundaries, including the moving boundaries such as piston and valves, are kept constant at 360 K during the simulation. A pressure inlet is used at the inlet boundary (one in positive Y direction) with two inlet pressures (1.0 and 1.5 atm) and fixed temperature at 318 K. The flow direction at the inlet is set parallel to the intake runner walls. The outlet boundary (in negative Y direction) is set as a pressure-outlet where the gauge pressure is zero. The internal interfaces between zones, as discussed above, do not contribute physically in the calculation; however, when the valves are fully closed, the interfaces are treated as walls and the boundary conditions are determined from the adjacent wall boundaries. The diesel fuel which is utilized in this simulation is n-decane (C10H22) with properties taken according to Wang's experimental data (Table 2) [7], Four injection points are set at the center of the cylinder head, offset by 0.5 mm from the cylinder axis. The nozzle holes are 90 degrees apart from each other and face towards positive X, positive Y, negative Table 3. Nozzle configurations and spray injection parameters RESULTS AND DISCUSSION Due to the simplified geometry of the engine used in this study, no validation with experimental data is available. Therefore, the quality of the model in this study is compared with other numerical simulation which uses the same geometry. Jonnalagedda et al. [42] studied an HCCI model with this simplified vertical port engine geometry and investigated the grid dependency with RANS and LES models in KIVA3V. Figure 3 shows the difference of the volume averaged in-cylinder pressure between the turbulence models [43]; standard k-ε in our study and Reynolds Averaged Navier-Stokes (RANS) and large eddy simulation (LES) models by Jonnalagedda [42]. The pressure difference at the peak between the standard k-ε and LES model is 8.34 % while the difference between the RANS and LES models is −5.73 %. The flow structure of the models has also been examined. Figure 4 is the stream traces on the cut plane through the middle of the ports (YZ plane). These traces are captured at 200 CA where the large turbulent kinetic energy Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM is expected at the time when the intake valve is closed. Unlike the LES model, the standard k-ε model is not able to depict small eddies in the section; however, it shows the largest eddies at the bottom right corner, the middle right, and the left edge of the intake valve pocket. Flow at the time of injection is also investigated for the validation. Figure 5 captures the stream traces of the natural aspirated and turbocharged cases in the section of a horizontal (XY) plane 5.7 mm offset from the surface of the cylinder head to piston surface. In the natural aspirated case, vortices are observed near the intake valve, which are also captured in the small flat piston engine [44], In summary, the in-cylinder flow calculated by the standard k-ε turbulence model does not duplicate the complex localized details of the flow predicted by the LES model but is able to reconstruct the larger eddies. In addition comparison of the flow characteristics before TDC shows the similar vortex positions. Since our current interest is to study features of the general flow structure, the standard k-ε model seems to be adequate for this purpose. In Fig. 5, it is observed that the sprays emanating from INJ-0 and INJ-180 are exposed to the opposing flow whereas the spray of INJ- 90 will be injected into the downwind and INJ-270 will face a much more complex flow compared to the others. INJ-0 and INJ-180 are sprayed to the upwind flow and fuel injected from INJ-90 and INJ-270 will experience catastrophic flow with vortices near the wall in the turbocharged case. The cores of the vortices that are captured near the intake port shift to the exhaust port side due to stronger flow by turbocharging. The effects of these incylinder flows are further discussed in the spray characteristics section. Figure 5. Stream traces superimposed by velocity magnitude contour of naturally aspirated (left) and turbocharged (right) case at the start of injection (−25.5 ATDC) on XV plane 5.7 mm offset from the cylinder head surface Figure 3. Mean cylinder pressure variation for standard k- ε in present work, and RANS and LES models [42] Figure 4. Stream traces on YZ cut plane at 200 CA with standard k-ε in present work (left) and LES model [42] (right) Once the fuel is injected into the chamber, strong flows of the sprays become dominant and build symmetric flow structure about the axes (Fig. 6). The magnitude of the fuel spray velocity changes linearly by increasing the injection pressure and the backflow is also increased relatively; hence, our simulations predict higher relative velocity and air entraimnent at the edges of sprays, as observed in the experiments of Choi et al [19]. The effects of the breakup models, turbocharging, and injection pressures on spray tip penetration are also examined. Figure 7 presents the spray tip penetrations for different injection pressures and densities in the engine model, using the experiment data by Wang et al [7] and the simulation in constant volume chamber as a reference. The difference between simulations and experiments in the constant volume case indicates that the spray models are properly set. The reduction of the penetration from the constant volume case to the engine model confirms the effect of the flow inside the cylinder. It is also noticed in Fig. 7 that the KH-RT model cases penetrate slightly less than the WAVE model cases, as expected [16]. Since the KH-RT model generally breaks droplets up more quickly into smaller droplets, evaporation is induced and penetration is reduced. Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM The small droplet size and the effective vaporization generated by the RT mechanism of KH-RT agree with the result of Ricart et al. [45] and Xin et al. [46]. The effect of turbocharging is clearly seen in the slower penetration as shown in Fig. 7. As stated in the experimental works [7, 24], high air density induced by turbocharging reduces the spray penetration significantly. Figure 6. Velocity vector field of INJ-0 superimposed by velocity magnitude contour at 2.5 CA after start of injection (−23.0 ATDC) on XY plane 5.7 mm offset from the cylinder head surface Figure 7. Spray tip penetration variations of INJ-0 at different pressures (100, 200 and 300 MPa) and conditions associated with the comparison between the experiment and simulation in the constant volume vessel Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM Figure 8 presents the variation of each injector's spray penetration in the turbo and non-turbocharging cases. In the naturally aspirated case, as discussed previously in Fig. 5, INJ-0 and INJ-180 have upwind flow towards the injection points and hence their penetrations are similar. On the other hand, the penetration of INJ-270 is greatly reduced from the other sprays due to complexity of the flow and the existence of a vortex near the wall. In the case of turbocharging, the difference is not large compared to the naturally aspirated case; however, the penetrations of INJ-90 and INJ-270 are reduced since both of them are exposed to complicated flow. The effect of the vortices on vaporization is consistent with the discussion by Jagus [27]. Also note that the spray penetration at 300 MPa injection pressure is lower than 100 and 200 MPa in the experiment and the computation. Fast atomization induced by higher injection pressure results in droplet vaporization and the penetration is reduced. Figure 8. Deviation of spray tip penetration among injectors at 300 MPa injection pressure with naturally aspirated (top) and turbocharged (bottom) cases The results for Sauter mean diameter (Fig. 9) display transition of droplet size during the injection. Within 1.0 CA after the start of injection each model shows rapid change of droplet size. The maximum difference of SMD between WAVE and KH-RT model is found to be 19.5% in the 200 MPa turbocharging case. Additionally, the difference between naturally aspirated and turbocharged case is not significant since the effect of the air density on SMD is factored by the power 0.06 as defined by the modified SMD correlation of Ejim et al. [47], (16) where SMD is given in µm, kinematic viscosity is in m2/s, surface tension is in N/m, ρl and pg (density of the fuel and air) are in kg/m3, and ΔP (the pressure difference between injection and ambient pressure) is in bar. Within 1.0 CA after the start of injection, droplet sizes of all models converge to similar diameter and remain constant, as discussed by Nishida et al. [5]. The higher the injection pressure, the higher the velocity, as discussed in the previous section. Only the 300 MPa injection pressure case is atomized the droplets break into smaller size than SMDs obtained from the correlation (red-dashed line in Fig. 9). Since the correlation of SMD is developed within 0.5 ms [47], the larger SMDs in lower injection pressures are due to the short time period of the breakup process. Still, the correlation between high spray velocity and small SMD is observed in the results and match the study of Post and Abraham [26]. The SMD limitation after 300 MPa injection pressure reported by Lee et al. [11] is not reproduced in this study; however, the reduction of the rate of change in SMD is observed where the rate of reduction from 100 MPa to 200 MPa is 48.9% whereas the change rate from 200 MPa to 300 MPa is 41.8%. The dispersion of sprays is analyzed by the spray cone angles in different conditions. The correlation of the cone angle with the injection pressure is observed in Fig. 10. The results of larger cone angle at higher injection pressure matches the discussion of Park et al. [48] that small droplets are found downstream of the cross-flow. In this study the small droplets on the edge of the spray are entrained by the opposing flow, captured by the flow, and enhance the size of the cone angle. The effect of turbocharging is also shown for the 300 MPa case (Fig. 10) which predicts wider cone angle, as found by Naber and Siebers [24] whereas no change or even narrower cone angles are captured in the 100 and 200 MPa cases. It is assumed that the sprays are not yet fully developed at 2.5 CA after the start of injection for those cases. The change of cone angle in time shows oscillation during the injection but it is likely to maintain the constant value right after the start of injection. The constant but oscillating spray cone angle during the injection is also found at different location of spray in Naber and Siebers' work [24]. Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM Figure 10. Variations of spray cone angle in different cases and injectors at 2.5 CA after start of injection (−23.0 ATDC) and transition of spray cone angle of 300 MPa injection pressure until sprays impinge to the walls Figure 9. Sauter mean diameter; INJ-0 at different injection pressures (100, 200, and 300 MPa) Finally, the fuel mass fractions for each case are examined. Figure 11 shows the sections of XY, YZ, and ZX plane at two different times and is placed right, top, and left, respectively, in each group of contours. Larger portion of the fuel mass contour in the 300 MPa cases at 2.5 CA after start of injection indicates fast injection process due to high injection pressure. Fast penetration of spray at high injection pressure was also observed by Tao and Bergstrand [12]. The completion of injection at an early stage results in extra time for fuel-air mixing and contributes to a better mixture formation at combustion as reported by Yokota [3]. The uniform color contour of the high injection pressure cases at TDC indicates an even and lean mixture of fuel in the chamber. Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM strong swirl motion but generates a complicated flow at the time of injection. From this study, in the ultra-high injection pressure cases, the following conclusions regarding spray and flow characteristics can be made, and are in agreement with the results reported by other researchers. • KH-RT hybrid breakup model estimates smaller droplets and larger dispersion than the WAVE model. However, the spray characteristics calculated by the WAVE model are not significantly different from KH-RT since the breakup constant is found by using the correlation to RT constants. • Four sprays injected in the chamber are exposed to different flow structures and found to be in good agreement with other works on spray characteristics such as spray tip penetration Sauter mean diameter and spray cone angle. The ultra-high injection pressure breaks droplets to smaller size; thus, reduced spray penetration and higher dispersion rate is predicted. Also, the droplet vaporization is accelerated by the existence of vortices in the direction of the sprays. • Turbocharging is found to reduce spray tip penetration and to dissipate the angle of spray. On the other hand, the effect of turbocharging on SMD is not predicted as indicated in the correlation stated by Ejim [47]. • The ultra-high injection pressure promotes a faster and effective mixture process and allows extending the time to form the mixture of air-fuel. • The rate of SMD change is reduced at higher injection pressure, as reported by Lee et al [11]. However, the SMD limitation which Lee et al. has observed is not detected in this study. REFERENCES 1. Heywood, J.B., “Internal Combustion Engine Fundamentals,” McGraw-Hill, ISBN-0-07-028637-X: 512-554, 1988. 2. Kato, T., Tsujimura, K., Shintani, M., Minami, T. et al., “Spray Characteristics and Combustion Improvement of D.I. Diesel Engine with High Pressure Fuel Injection,” SAE Technical Paper 890265, 1989, doi: 10.4271/890265. Figure 11. Contours of fuel mass fraction at two different times CONCLUSIONS The effects of ultra-high injection pressure sprays are studied in a simplified engine geometry. The engine model is properly set up and different conditions are examined by means of injection pressures, spray models and inlet pressures. The simple engine geometry does not induce 3. Yokota, H., Kamimoto, T., Kosaka, H., and Tsujimura, K., “Fast Burning and Reduced Soot Formation via Ultra-High Pressure Diesel Fuel Injection,” SAE Technical Paper 910225, 1991, doi: 10.4271/910225. 4. Kuti, O.A., Xiangang, W.G., Zhang, W., Nishida, K., and Huang, Z.H., “Characteristics of the Ignition and Combustion of Biodiesel Fuel Spray Injected by a Common-Rail Injection System for a Direct-Injection Diesel Engine,” Proceeding of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering 224:1581-1596, 2010, doi: 10.1243/09544070JAUTO 1503. 5. NISHIDA, K., ZHANG, W., and MANABE, T., “Effects of Micro-Hole and Ultra-High Injection Pressure on Mixture Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM Properties of D.I. Diesel Spray,” SAE Technical Paper 2007-01-1890, 2007, doi:10.4271/2007-01-1890. 6. Wang, X.G., Kuti, O. A., Zhang, W., Nishida, K., and Huang, Z.H., “Effect of Injection Pressure on Flame and Soot Characteristics of the Biodiesel Fuel Spray,” Combustion Science and Technology 182:1369-1390, 2010, doi: 10.1080/00102201003789139. 7. Wang, X.G., Huang, Z.H., Kuti, O.A., Zhang, W., and Nishida, K., “Experimental and Analytical Study on Biodiesel and Diesel Spray Characteristics under Ultra-High Injection Pressure,” International Journal of Heat and Fluid Flow 31:659-666, 2010, doi: 10.1016/j.ijheatfluidflow. 2010.03.006. 8. Wang, X.G., Huang, Z.H., Zhang, W., Kuti, O.A., and Nishida, K., “Effects of Ultra-High Injection Pressure and Micro-Hole Nozzle on Flame Structure and Soot Formation of Impinging Diesel Spray,” Applied Energy 88:1620-1628, 2011, doi: 10.1016/j.apenergy.2010.11.035. 9. Zhang, W., Nishida, K., Gao, J., and Miura, D., “An Experimental Study on Flat-Wall-Impinging Spray of Microhole Nozzles under Ultra-High Injection Pressures,” Proceeding of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering 222:1731-1741, 2008, doi: 10.1243/09544070JAUTO858. 17. Karrholm, F. and Nordin, N., “Numerical Investigation of Mesh/Turbulence/Spray Interaction for Diesel applications,” SAE Technical Paper 2005-01-2115, 2005, doi:10.4271/2005-01-2115. 18. Ghasemi, A., and Djavareshkian, M.H., “Investigation of the Effect of Natural Gas Equivalence Ratio and Piston Bowl Flow Field on Combustion and Pollutant Formation of a DI Dual Fuel Engine.” Journal of Applied Sciences 10:1-11, 2010, doi: 10.3923/jas.2010.1369.1379. 19. Choi, W., and Choi, B.C., “Estimation of the Air Entraimnent Characteristics of a Transient High-Pressure Diesel Spray,” Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering 219:1025-1036, 2005, doi: 10.1243/095440705X34630. 20. Desantes, J.M., Arregle, J., Lopez, J.J., and Garcia, J.M., “Turbulent Gas Jets and Diesel-Like Sprays in a Crossflow: A Study on Axis Deflection and Air Entraimnent,” Fuel 86:2120-2132, 2006, doi: 10.1016/j.fuel.2006.03.025. 21. McCracken, M.E., and Abraham, J., “Characterization of Mixing Enhancement in Swirl-Spray Interactions in Diesel Engines,” Combustion Science and Technology 174(10): 93-124, 2002, doi: 10.1080/00102200290021452. 10. Zhang, W., Nishida, K., and Tian, J.P., “Spray-Spray and Spray-Wall Interactions in Diesel Sprays from Micro-Hole Nozzles under Ultra-High Injection Pressures,” presented at ICLASS 2009, 2009, Vail, CO, July, 2009. 22. Park, S.W., and Lee, C.S., “Investigation of Atomization and Evaporation Characteristics of High-Pressure Injection Diesel Spray Using Kelvin-Helmtholtz Instability/Droplet Deformation and Break-Up Competition Model,” Proceeding of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering 218:767-777, 2004. 11. Lee, S.H., Jeong, D.Y., Lee, J.T., Ryou, H.S., and Hong, K., “Investigation on Spray Characteristics under Ultra-High Injection Pressure Conditions,” International Journal of Automotive Technology 6(2): 125-131, 2005. 23. Iyer, V., and Abraham, J., “Penetration and Dispersion of Transient Gas Jets and Sprays,” Combustion Science and Technology 130: 315-334, 1997, doi: 10.1080/00102209708935747. 12. Tao, F. and Bergstrand, P., “Effect of Ultra-High Injection Pressure on Diesel Ignition and Flame under HighBoost Conditions,” SAE Technical Paper 2008-01-1603, 2008, doi:10.4271/2008-01-1603. 24. Naber, J. and Siebers, D., “Effects of Gas Density and Vaporization on Penetration and Dispersion of Diesel Sprays,” SAE Technical Paper 960034, 1996, doi: 10.4271/960034. 13. Lin, S.P., and Reitz, R.D., “Drop and Spray Formation from a Liquid Jet,” Annual Review of Fluid Mechanics 30:85-105, 1998, doi: 10.1146/annurev.fluid.30.1.85. 25. Kennaird, D., Crua, C., Lacoste, J., Heikal, M. et al., “InCylinder Penetration and Break-Up of Diesel Sprays Using a Common-Rail Injection System,” SAE Technical Paper 2002-01-1626, 2002, doi:10.4271/2002-01-1626. 14. Jiang, X., Siamas, G.A., Jagus, K., and Karayiannis, T.G., “Physical Modelling and Advanced Simulations of Gasliquid Two- phase Jet Flows in Atomization and Sprays.” Progress in Energy and Combustion Science 36:131-167, 2010, doi: 10.1016/j.pecs.2009.09.002. 15. Djavareshkian, M.H., and Ghasemi, A., “Investigation of Jet Break-Up Process in Diesel Engine Spray Modelling,” Journal of Applied Sciences 9(11):2078-2087, 2009. 16. Hossainpour, S., and Binesh, A.R., “Investigation of Fuel Spray Atomization in a DI Heavy-Duty Diesel Engine and Comparison of Various Spray Breakup Models,” Fuel 88:799-805, 2009, doi: 10.1016/j.fuel.2008.10.036. 26. Post, S., and Abraham, J., “Modeling the Outcome of Drop-Drop Collisions in Diesel Sprays,” International Journal of Multiphase Flow 28:997-1019, 2002. 27. Jagus, K., Jiang, X., Dober, G., Greeves, G., Milanovic, N., and Zhao, H., “Assessment of Large-Eddy Simulation Feasibility in Modelling the Unsteady Diesel Fuel Injection and Mixing in a High-Speed Direct-Injection Engine,” Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering 223:1033-1048, 2009, doi: 10.1243/09544070JAUTO1052. Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM 28. “ANSYS FLUENT Theory Guide,” ANSYS, Inc., Canonsburg, PA. 2010. 29. Nijdam, J.J., Guo, B.Y., Fletcher, D.F., and Langrish, T.A.G., “Lagrangian and Eulerian Models for Simulating Turbulent Dispersion and Coalescence of Droplets within a Spray,” Applied Mathematical Modelling 30:1196-1211, 2006, doi: 10.1016/j.apm.2006.02.001. 30. Schmidt, D.P., and Rutland, C. J., “A New Droplet Collision Algorithm,” Journal of Computational Physics 164:62-80, 2000, doi: 10.1006/jcph.2000.6568. 31. Sazhin, S.S., “Advanced Models of Fuel Droplet Heating and Evaporation,” Progress in Energy and Combustion Science 32:162-214, 2006, doi: 10.1016/j.pecs.2005.11.001. 32. Reitz, R.D., “Modeling Atomization Processes in HighPressure Vaporizing Sprays,” Atomization and Spray Technology 3:309-337, 1987. 33. Hwang, S.S., Liu, Z., and Reitz, R.D., “Breakup Mechanisms and Drag Coefficients of High-Speed Vaporizing Liquid Drops,” Atomization and Spray 6:353-376, 1996. 34. Lee, C.S., Kim, H.J., and Park, S.W., “Atomization Characteristics and Prediction Accuracies of Hybrid BreakUp Models for a Gasoline Direct Injection Spray,” Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering 218:1041-1053, 2004. 35. Patterson, M. and Reitz, R., “Modeling the Effects of Fuel Spray Characteristics on Diesel Engine Combustion and Emission,” SAE Technical Paper 980131, 1998, doi: 10.4271/980131. 36. Liu, A., Mather, D., and Reitz, R., “Modeling the Effects of Drop Drag and Breakup on Fuel Sprays,” SAE Technical Paper 930072, 1993, doi: 10.4271/930072. 37. Liu, F.S., Zhou, L., Sun, B.G., Li, Z.J., and Harold, J.S., “Validation and Modification of WAVE Spray Model for Diesel Combustion Simulation,” Fuel 87:3420-3427, 2008, doi: 10.1016/j.fuel.2008.05.001. 42. Jonnalagedda, S., Nguyen, A.T., Zhou, B., and Sobiesiak, A., “Numerical Studies on HCCI Engine Combustion” Proceedings of the 18th Annual Conference of the CFD Society of Canada, CFD 2010, London ON, Canada, May, 2010. 43. Versteeg, H.K., and Malalasekera, W., “An Introduction to Computational Fluid Dynamics,” Pearson Education Limited, Essex, ISBN-978-0-13-127498-3:72-110, 2007. 44. Bianchi, G.M., Cantore, G., Parmeggiani, P., and Michelassi, V., “On Application of Nonlinear k-ε Models for Internal Combustion Engine,” Journal of Engineering for Gas Turbines and Power 124:668-677, 2002, doi: 10.1115/1.1454115. 45. Ricart, L.M., Reitz, R.D., and Dec, J.E., “Comparisons of Diesel Spray Liquid Penetration and Vapor Fuel Distributions with In-Cylinder Optical Measurements,” Journal of Engineering for Gas Turbines and Power 122:588-595, 2000. 46. Xin, J., Ricart, L., and Reitz, R.D., “Computer Modeling of Diesel Spray Atomization and Combustion,” Combustion Science and Technology 137:171-194, 1998, doi: 10.1080/00102209808952050. 47. Ejim, C.E., Fleck, B.A., and Amirfazli, A., “Analytical Study for Atomization of Biodiesels and Their Blends in a Typical Injector: Surface Tension and Viscosity Effects,” Fuel 86:1534-1544, 2007, doi: 10.1016/j.fuel.2006.11.006. 48. Park, S.W., Kim, S., and Lee, C.S., “Breakup and Atomization Characteristics of Mono-Dispersed Diesel Droplets in a Cross- Flow Air Stream” International Journal of Multiphase Flow 32:807-822, 2006, doi: 10.1016/ j.ijmultiphaseflow.2006.02.019. CONTACT INFORMATION Kohei Fukuda fukuda@uwindsor.ca Mechanical, Automotive and Material Engineering University of Windsor 401 Sunset Ave. Windsor, ON, CANADA N9B 3P4 38. Levich, V.G., “Physicochemical Hydrodynamics,” Prentice Hall, New Jersey, ISBN-0-13-674440-0, 1962. ACKNOWLEDGEMENTS 39. Hiroyasu, H. and Arai, M., “Structures of Fuel Sprays in Diesel Engines,” SAE Technical Paper 900475, 1990, doi: 10.4271/900475. This research is funded by Ontario Ministry of Research and Innovation through the Ontario Research Fund under the Green Auto Power Train project. 40. Senecal, P.K., “Development of a Methodology for Internal Combustion Engine Design Using MultiDimensional Modeling with Validation through Experiments,” Ph.D. thesis. Department of Mechanical Engineering, University of Wisconsin-Madison, Madison 2000. 41. Amsden, A.A., “KIVA-3V: A Block-Structured KIVA Program for Engines with Vertical or Canted Valves,” Los Alamos National Laboratory, New Mexico, 1997. DEFINITIONS/ABBREVIATIONS U σ Velocity Surface Tension Gratis copy for Kohei Fukuda Copyright 2012 SAE International E-mailing, copying and internet posting are prohibited Downloaded Thursday, March 15, 2012 10:26:55 AM v ρ ASOI ATDC BDC CA CFD HCCI Kinemetic Viscosity Density After Start of Injection SMD TAB TDC Sauter Mean Diameter Taylor Analogy Breakup Top Dead Center After Top Dead Center Bottom Dead Center Crank Angle Computational Fluid Dynamics Homogeneous Charge Compression Ignition KH-RT Kelvin-Helmholtz Rayleigh-Taylor LES NA RANS Large Eddy Simulation Natural Aspirated Reynolds Averaged Navier-Stokes The Engineering Meetings Board has approved this paper for publication. It has successfully completed SAE's peer review process under the supervision of the session organizer. This process requires a minimum of three (3) reviews by industry experts. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE. ISSN 0148-7191 Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely responsible for the content of the paper. SAE Customer Service: Tel: 877-606-7323 (inside USA and Canada) Tel: 724-776-4970 (outside USA) Fax: 724-776-0790 Email: CustomerService@sae.org SAE Web Address: http://www.sae.org Printed in USA