Development, Implementation, and Validation of a Fuel Impingement Model for Direct Injected Fuels with High Enthalpy of Vaporization (SAE Paper- 2009-01-0306) Craig D. Marriott PE, Matthew A. Wiles PE, GMPT Advanced Engineering Brandon T. Rouse, Michigan Technological University Gamma Technologies GT Power Users Conference Birmingham, MI December 7, 2009
Presentation Outline Introduction - Purpose of the Study Problem statement Background - Engine specifications Fuel specifications Model Construction - Spray geometry Temporal spray formation Thermodynamic representation and assumptions Model Implementation - Layout User interface Virtual sensors and actuators Results and Validation - Volumetric efficiency Heat rejection Engine speed Conclusions 2
Introduction Purpose of the Study General Motors commitment to bio-ethanol usage in Flex- Fuel vehicles (50% in North America by 2012) Reduction of non-renewable fuel consumption Reduction of CO2 emissions from transportation Flex-fuel applications require that engine performance not be compromised on either fuel General Motors Powertrain technology implementation Spark Ignited Direct Injection (SIDI) is key element for future technologies Need to enable Flex-Fuel SIDI engine technology Fuel system dynamic range must be able to handle light load gasoline and high-power ethanol operation 3
Introduction Purpose of the Study Continuation of experimental study published in SAE-2008-01-0319 Detailed investigation of early Start of Injection strategies enabled by smoke reduction of ethanol content Additional injection window available for additional fuel injection 20 E0 (CP Baseline) 20 E85 DI Fuel Pressure, MPa 15 10 5 0.45 0.1 0.1 1.1 0.4 0.15 0.3 1 0.7 0.2 0.35 0.5 0.8 0.5 0.25 0.15 0.8 1 0.9 0.55 0.3 0.4 1.2 0.85 0.4 0.1 0.35 0.95 0.45 0.45 0.15 0.9 1.1 0.7 0.25 0.2 0.25 1.4 1.2 1.0 0.8 0.6 0.4 0.2 Smoke, FSN DI Fuel Pressure, MPa 15 10 5 0.05 0.05 0.05 0.25 0.05 0.35 0.1 0.4 0.15 0.3 0.15 0.2 0.4 0.25 0.1 0.45 0.2 0.1 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 Smoke, FSN 0 0 100 150 200 250 SOI, CAD btdc 300 350 100 150 200 250 SOI, CAD btdc 300 350 4
Introduction Problem Statement Early SOI led to over prediction of Volumetric Efficiency with 1D engine simulation Over prediction appeared to be due misrepresentation of fuel impingement and associated heat transfer mechanisms of high vaporization enthalpy Commercial simulation software standard DI fuel delivery model does not comprehend fuel impingement on cylinder surfaces 110 110 Volumetric Efficiency (%) 100 90 80 70 60 Empirical E85 Analytical E85 Volumetric Efficiency (%) 100 90 80 70 60 Empirical Gasoline Analytical Gasoline 50-400 -300-200 -100 0 Start of Injection (deg atdc) 50-400 -300-200 -100 0 Start of Injection (deg atdc) 5
Background - Engine/Fuel Specifications Manufacturer Engine Configuration Cycle Advertised Displacement GMPT Advanced Engineering Naturally Aspirated, Inline 4- cylinder, Direct Injection, Variable Valve Timing 4-stroke 2.4 liters Valves per cylinder 4 Bore Stroke 88.0 mm 98.0 mm Nominal Compression Ratio 11.9 Intake Valve Peak Lift @ Park Intake Camshaft Duration Intake Cam Timing Range Intake Cam Park Position Exhaust Valve Peak Lift @ Park Exhaust Camshaft Duration Exhaust Camshaft Timing Range Exhaust Cam Park Position Dynamic Injector Flow rate Fuel Pump displacement 131 Crank Deg atdc 267 Crank Deg 50 Crank Deg Full Retard -125 Crank Deg atdc 241 Crank Deg 50 Crank Deg Full Advance 17.5 cc/sec @ 10Mpa 1.1 cc/engine cycle Oil Mineral-based 5W-30 H/C O/C Λ=1. mass air/fuel ratio RON MON LHV Fuel Type [-] [-] [MJ/kg] Mole Weight Specific Gravity Heat of Vapor [kg/kmolcarbon]] [kj/kg] Chevron Philips B-25 1.908 0.000 14.65 104.3 96.5 43.25 13.93 0.736 350 ED85 2.703 0.382 9.85 107.7 89.5 29.62 20.85 0.785 842 6
Model Construction Geometric Representation Variable description: X1 = Axial penetration length X2 = Vertical penetration length X3 = Projected impingement width X4 = Projected impingement sector angle R = Radius of spray cone base α = Spray cone angle to fire deck β = Spray cone dispersion angle S = Piston position from injector tip AI = Projected impingement area AT = Total cone base area Hollow cone is represented by subtracting equivalent geometric calculation of inner cone S x 2 = x 1 A A I T sin 1 = π β ( α ) + cos( α ) tan x 3 = 2 x 2 S cos( α ) R x 4 = arccos R x 3 [ x sin( x ) ( x )] 4 4 cos R = x 1 sin 4 β 2 7
Model Construction Spray Formation Spray length development is a function of time Constant spray dispersion angle (low pressure charge) Fully developed (steady-state) spray penetration is a user input Liquid-Vapor transition is assumed to be linear from 1.0 to user input Spray properties assumed to be axis symmetric SAE-960034 8
Model Construction Fuel Deposition Only the liquid fraction of fuel is considered in an impingement condition Not all of the fuel liquid fraction that impinges upon the piston is of interest Fuel deposition efficiency constant is user input and characterized by empirical measurements 9
Model Construction First Law Analysis M f = injector mass flow rate M v = mass vaporization rate M p = puddle mass formation rate M pv = puddle vaporization rate Q v = charge heat transfer of vaporization Q conv = convective heat transfer of puddle vaporization Q cond = conductive heat transfer of puddle vaporization M & = M & Ratio η * mf puddle injector impinged deposition liquid M & M & puddle injector Ratio impinged = Formation rate of puddle mass = Injector mass flow rate = Projected impingemen t area ratio η deposition = Fuel deposition efficiency mf liquid = Liquid Mass Fraction at impingemen t Q cond = M & puddle h fg = heat of vaporization where 10 h fg
Model Implementation Heat Actuator All retained liquid fuel vaporization energy is assumed to be conducted from the piston Vaporization is assumed to be instantaneous (Mp=Mv, no puddle) Therefore heat transfer rate is equal to vaporization rate Heat is added to the cylinder charge via a heat actuator Q & cond = M & injector mf liwuid h fg Ratio impinged η deposition 11
Results and Validation 2000 RPM WOT Spray cone axis angle 45 deg 110 Axial Penetration 80 mm Spray dispersion angle 80 deg Enthalpy of Vaporization 842 kj/kg Inner spray cone angle 50 deg Deposition Efficiency 0.4 Liquid Fraction at full length 0.2 Volumetric Efficiency (%) 100 90 80 70 60 2000 RPM 1.15 EQR Analytical w Imp 2000 RPM 1.15 EQR Empirical 2000 RPM 1.15 EQR Analytical w/o Imp 4500 12 50 Heat Addition (W) 4000 3500 3000 2500 2000 1500 1000 500 SOI=-210 SOI=-240 SOI=-270 SOI=-300 SOI=-330 SOI=-360 Intake Valve Lift 10 8 6 4 2 Valve Lift (mm) -400-300 -200-100 0 Start of Injection (deg atdc) VE response shows good correlation with empirical data Heat actuator signal shows rapid spray formation and geometric relationship to piston 0 0 180 360 540 720 Crank Angle (deg atdc) 0 Injection ends prior to impingement geometry availability 12
Results and Validation 2000 RPM WOT Cylinder charge heat rejection is reduced by approximately 10% for early SOI conditions Simulated brake thermal efficiency in increased accordingly 22 37 36.8 In-Cylinder Heat Rejection (kw) 20 18 16 with Impingement without Impingement Brake Thermal Efficiency (%) 36.6 36.4 36.2 36 35.8 35.6 35.4 35.2 35 Analytical w/o Imp Analytical w/ Imp 14-400 -300-200 -100 0 Start of Injection (deg atdc) 34.8-400 -300-200 -100 0 Start of Injection (deg atdc) 13
Results and Validation 3000 RPM WOT Simulated volumetric efficiency shows similar trends to empirical data Differences in absolute magnitude are contributed to other model inaccuracies 110 100 Volumetric Efficiency (%) 90 80 70 60 3000 RPM 1.0 EQR Analytical w Imp 3000 RPM 1.0 EQR Empirical 3000 RPM 1.0 EQR Analytical w/o Imp 50-400 -300-200 -100 0 Start of Injection (deg atdc) 14
Results and Validation 4000 RPM WOT Simulated volumetric efficiency shows possible over-prediction of impingement effect Differences in absolute magnitude are attributed to other model inaccuracies 110 110 100 100 Volumetric Efficiency (%) 90 80 70 60 4000 RPM 1.0 EQR Analytical w Imp 4000 RPM 1.0EQR Empirical 4000 RPM 1.0 EQR Analytical w/o Imp Volumetric Efficiency (%) 90 80 70 60 4000 RPM 1.15 EQR Analytical Gas w/o Imp 4000 RPM 1.15 QR Empirical Gas 1.15 EQR 50-400 -300-200 -100 0 Start of Injection (deg atdc) 50-400 -300-200 -100 0 Start of Injection (deg atdc) 15
Results and Validation 4000 RPM WOT High speed operation shows spray formation time is slower relative to piston motion High speed operation shows no impingement develops for the -270 SOI condition Impingement is less significant at high speed due to temporal interactions Injection duration causes impingement throughout geometric availability 3500 3000 12 10 Heat Addition (W) 2500 2000 1500 1000 SOI=-270 SOI=-300 SOI=-330 SOI=-360 Intake Valve Lift 8 6 4 Valve Lift (mm) 500 2 0 0 180 360 540 720 Crank Angle (deg atdc) 0 16
Conclusions Engine simulation of high enthalpy of vaporization fuels must consider impingement to provide accurate results at early SOI conditions Impingement conditions can lead to misrepresentation of heat transfer phenomenon in the cylinder charge A geometric based impingement model was developed to correct for the heat transfer effects of impingement Empirical comparisons show good correlation of impingement trends and magnitude at lower engine speeds (<4000 RPM) suggesting the physics of the problem were correctly represented Geometric impingement model does show reduced magnitude for high speed operation due to temporal interactions Geometric impingement model may over-predict effects of impingement for high speed conditions (>4000 RPM) Air charge motion interactions may have to be represented to improve accuracy for high-speed operation 17
Thank you for your attention 18