Objective. Combustion. Non interfering diagnostics for the study of thermofluidynamic processes in ICE

Transcription

1 Objective Non interfering diagnostics for the study of thermofluidynamic processes in ICE B. M. Vaglieco Istituto Motori CNR Napoli, ITALY Study of chemical and physical phenomena in internal combustion engine by non intrusive techniques at high spatial (< micron) and temporal resolution (nanosecond) Combustion CONVENTIONAL MEASUREMENTS Compression Ignition Spark Ignition Homogeneous Charge Compression Ignition Fuel injector Hot Flame Spark plug Low temperature combustion Engine parameters quartz piezoelectric pressure transducer In-cylinder pressure current and voltage sensors spark and injection soot NO Flame front NO Exhaust emissions UV-IR analyzer and opacimeter HC, O 2, CO, CO 2, NO X particulate mass concentration Diesel Engine Gasoline Engine HCCI Engine Determination of Start of Combustion Piezoelectric Pressure transducer The variation of slope of pressure curve The heat release indicates : in the absolute minimum the overcoming of the first exothermic combustion reactions with respect to endothermic due to the evaporation of fuel injected

2 Pressure measurements.2 ST=3 CAD BTDC ST=3 CAD BTDC Interaction light-matter.8 Pressure [bar] ST=3 CAD BTDC knock pressure [bar] Qualitative and quantitative characterization on transient phenomena in optically accessible combustion system (d) GAS (d) -.2 Non intrusive techniques Non interference on phenomena Crank Angle Degree Crank Angle Degree High spatial and temporal resolution Capability to follow stationary phenomena and to measure in situ EXCITATION Occurs when an electron in an atom is given energy causing it to jump to a higher orbit. This can happen through collisions or photon absorption (the photon absorption must exactly match the energy jump). Radio waves are produced by electrons moving up and down an antenna Visible light is produced by electrons changing energy states in an atom ELECTROMAGNIC SPECTRUM The excited atom usually de-excites in about millionth of a second. The subsequent emitted radiation has an energy that matches that of the orbital change in the atom. This emitted radiation gives the characteristic colors of the element involved EM Waves Radio Waves Microwaves Infrared Visible Ultraviolet X-rays Gamma rays Sources Vibrating charges Molecular rotations Molecular vibrations Atomic vibrations Atomic vibrations Atomic vibrations Nuclear vibrations Emission Spectra Continuous Emission Spectrum Emission Spectra of Hydrogen Slit Slit Discrete Emission Spectrum Film White Light Source Prism Low Density Glowing Hydrogen Gas Prism Photographic Film Photographic Film 2

3 Absorption Spectra of Hydrogen Discrete Emission Spectrum ABSORPTION SPECTRA Frequencies of light that represent the correct energy jumps in the atom will be absorbed. When the atom de-excites, it emits the same kinds of frequencies it absorbed. However, this emission is in all directions. White Light Source Slit Hydrogen Gas Prism Discrete Absorption Spectrum Photographic Film Film INCANDESCENCE Electron transitions occur not only in the parent atom but in adjacent atoms as well FLUORESCENCE Some materials that are excited by UV emit visible. These materials are referred to as fluorescent materials. PHOSPHORESCENCE Electrons get "stuck" in excited states in the atoms and de-excitation occurs at different times for different atoms. A continuous glow occurs for some time. Bioluminescence MAGIC BOX MAGIC EYE UV-visible natural emission pros non-intrusive method simultaneous multispecies detection (spectroscopy) differentiation at same wavelength (chemiluminescence) cons line-of-sight method low signal/noise h CFD and optical measurement are suitable tools for explanation of complicated combustion mechanism. UV mirror Quartz window hν hν h If the other sub-models (spray,etc) are effective, recent CFD with detailed kinetics model is though to be robust or adaptable to the various combustion and emission description. 3

4 section view Transparent diesel engine Commercial cylinder head Common Rail Injector Conditioned intake Air Exhaust gas Transparent engine Lateral and Top Cross-Sections Toroidal bowl optical access diameter = 34mm; Lateral window diameter =4mm upper edge of bowl FRONT VIEW edge of quartz window in bottom piston injector Non lubricated condition (Bronze-Teflon Ring) Coolant temperature control system exhaust valve replaced by quartz window OUT IN pressure transducer upper edge of bowl TOP VIEW Optical Setup Injector Pump DIGITAL IMAGING of fuel spray and combustion phase Acquisition through piston crown window CCD CCD CCD UV-Mirror PC Common Rail Digital imaging Encoder CCD camera 6x48 pixels - Minimum exposure time μs Unit Control Electric Engine Combustion Pressure [bar] 7 6 Pre+M PSOC Pre PSOC M crank angle [degree] Pre+Main Injection 8 6 Combustion Pressure [bar] 7 6 Inj. Pressure 6 bar M PSOC crank angle [degree] Main Injection 8 6 Experiments vs Modelling Pre Main Autoignition R) H + O 2 = O + R2) O + H 2 = H + R3) H 2 + = H 2 O + H H 2 as fuel Heywood, BTDC 2.5 ATDC 6.5 BTDC 3.5 ATDC 4

5 Emission intensity [a.u.] Spectra at Autoignition of Pre + Main 3 BTDC= SOCpre BTDC=.6 ASOCpre Autoignition phase of pre injection occurs during the first times of main injection.5 BTDC=.5 ASOCpre Brightness versus color curve for different temperatures f T Black body Relative Energy Wavelength (nm) (measured in Kelvins) Imaging, Temperature and Soot concentration rpm - P inj 6 bar Imaging, Temperature and Soot concentration rpm - P inj 6 bar Temperature Visible Flame Soot Temperature Visible Flame Soot Temperature scale [K] 8 3 btdc Current [Ampere] KL factor scale 5 44 Soot mass concentration [mg/m 3 ] Temperature scale [K] 8 2 btdc Current [Ampere] KL factor scale 5 44 Soot mass concentration [mg/m 3 ] - 6 Crank angke [degree] - 6 Crank angke [degree] Imaging, Temperature and Soot concentration rpm - P inj 6 bar Imaging, Temperature and Soot concentration rpm - P inj 6 bar Temperature Visible Flame Soot Temperature Visible Flame Soot Temperature scale [K] 8 tdc Current [Ampere] KL factor scale 5 44 Soot mass concentration [mg/m 3 ] Temperature scale [K] 8 6 atdc Current [Ampere] KL factor scale 5 44 Soot mass concentration [mg/m 3 ] - 6 Crank angke [degree] - 6 Crank angke [degree] 5

6 New concepts for diesel combustion Injection strategies Injections per cycle Late injection MK Concept Nissan Motors HiMIcs HINO Motors UNIBUS - PCI 4 2 PREDIC/MUL DIC 3 (with 3 injectors) HCCI Denbratt 8: Varied Varied Compression ratio 8: 6.5: 6: 2: to 2: 7: to.5: Displacement [cc] Fuel wall impingement Early injection very limited yes yes yes yes EGR rate [%] high low high high high 5 Combustion Pressure [bar] 7 6 Pinj = 6 bar CR 8 6 Combustion Pressure [bar] 6 Pinj = 7 bar HCCI 8 6 Load range limited large large very limited large Test speed [rpm] Knock at high load P. Pinchon at al., IFP Thiesel 4 no no high high no NOx very low low very low very low very low Smoke very low low very low improved very low crank angle [degree] crank angle [degree] Engine speed: rpm - Fuel amount: 8 mm 3 /stroke Pinj [bar] Tin [ C] Pin (abs) [bar] Fuel BMEP Pilot [Kg/h] [bar] Pilot Pre Pre Main Main Post Post After After CR Pre+Main+Post \ \ \ \ HCCI Heat release rate for HCCI HCCI injection phase 68 btdc 58 btdc 48 btdc 38 btdc 28 btdc Rate Of Heat Release [kj/kg/cad] 8 Drive current [Ampere] Time A [ms] Ignition delay LTR HTR Temperature [K].5 A. A.5 A 2. A I II III IV V Cylinder Pressure [bar] P inj = 7 bar I II III IV V btdc 56 btdc 46 btdc 36 btdc 26 btdc A = After Start Of Injection Chemiluminescence Measurements Chemiluminescence Measurements RR [kj/kg ] 8 EOI SOC RR [kj/kg ] 8 EOI SOC 6 atdc HCO HO is the most important radical for driver the ignition process. It is producing by the decomposition of H 2 O 2 at K (ref. --77) GAYDON # atdc HO HCO

7 8 Chemiluminescence Measurements In the whole chamber HCCI vs CR visible combustion atdc 2 atdc 4 atdc 3 BTDC 2 BTDC BTDC BTDC 9 BTDC 8 BTDC 7 BTDC Rate Of H eat Release [kj/kg/cad] 6 HCCI Mean KL factor for HCCI strategy Rate Of Heat Release [kj/kg/cad] 6 CR Mean KL factor for CR strategies RR [kj/kg ] 8 EOI SOC radical is a good marker of LTC combustion process 3 BTDC TDC 4 ATDC ATDC 5 ATDC 7 ATDC 2 ATDC number concentration [part*cm -3 ].x 2.x 8.x 4.x Exhaust particle size distribution ATDC 52 ATDC 66 ATDC.x 9.x 8.x 7.x 6.x 5.x 4.x 3.x 2.x.x diameter [nm] Electrical Low Pressure Impactor Primary and secondary particles at exhaust SCATTERING Experimental method for exhaust characterization OPTICAL TENIQUE PARTICLE SIZE NUMBER DISTRIBUTION EXTINCTION VOLUME FRACTION OPACIMER SMOKE MER EMICAL ANALYSIS N% FSN DRY SOOT PARTICULATE MASS CONCENTRATION Nitrogen Oxides EMILUMINESCENCE ANALYSER SOF OPTICAL TENIQUE ABSORPTION ABSORPTION CROSS SPECTROSCOPY SECTION NUMBER CONCENTRATION SOOT INFRARED ANALYSER Experimental Apparatus EXTINCTION and SCATTERING SPECTRA Spectrometer ICCD BICONVEX LENS out Gas analysers PLANO- CONVEX LENS PLANO- CONVEX LENS DILUTER ELPI Opacimeter in Still plug valve CDPF Pulsed nanosecond light source: laser induced optical breakdown extinction BEAM SPLITTER 64nm Nd:YAG.8x x -6.6x -2 4.x -6 extinction coefficient [cm - ].4x -2.2x -2.x -2 8.x -3 6.x -3 4.x -3 NO rpm - 5bar soot scattering coefficient [cm - sr - ] 3.5x -6 3.x x -6 2.x -6.5x -6.x -6 rpm - 5bar 2.x -3 5.x -7 rpm - 2bar rpm - 2bar.x.x wavelength [nm] wavelength [nm] scattering 7

8 Pressure Drop [mbar] 6 8 UPSTREAM D= nm soot Time [min] Regeneration rpm-2bar Number Concentration [#/cm [#/cm 3 ] 3 ] Number Concentration [#/cm 3 ] 3.x 4 3.x.2x 6 2.x 4 2.x 8.x 5.x 4.x 4.x 5 5 min min 35 min Discr = 8% Discr % soot Discr = 2% graphitic Soot & organic.x.x Diameter [nm] Diameter [nm] [nm] DOWNSTREAM D= 5 nm t = min graphitic-like D= nm t = 5 min soot D = nm soot t = 35 min D 2 = 5 nm % soot % organic Transparent spark ignition engine UV- 45 mirror Normal combustion The combustion process starts at spark timings. The flame front moves across the combustion chamber in likeuniform manner. Abnormal combustion The flame front may be started by hot surface either prior to or after spark ignition Spark knock Can be controlled by the spark advance The knock is identified by intense pressure oscillations that arise around the maximum of pressure. The frequencies of oscillations are typically higher than 5 khz Heywood J. B. - Internal Combustion Engine Fundamentals New York - McGraw-Hill 988. Pressure [bar] knock pressure [bar] CAD CAD CAD ASOS High pass filter 5kHz The knock is identified by intense pressure oscillations that arise around the maximum of pressure. The frequencies of oscillations are typically higher than 5 khz rpm mbar Start of Injection CAD BTDC Duration of injection 96 CAD Spark 8 CAD BTDC Knocking phase rpm mbar Start of Injection CAD BTDC Duration of injection 96 CAD Spark 8 CAD BTDC knock pressure [bar] CAD CAD ASOS Abnormal combustion A combustion process in which a flame front may be started by hot surface either prior to or after spark ignition Hot spots and Surface ignition 29.6 CAD ASOS.4 CAD ASOS 3.6 CAD ASOS 33.6 CAD ASOS The hot spots correspond to very small centres of autoignition due to exothermic reactions. In the same time of hot-spots appearance a flame front (ignition surface) starts from spark plug. This flame is due to the thermal phase of knock (*) 34. CAD ASOS 34.4 CAD ASOS 36. CAD ASOS 36.4 CAD ASOS (*)Maly, R.R.- 25 th Symp.Int. on Combustion. Combustion Institute Ed

9 Knock Hot-spots Spark 8 CAD BTDC Knock Hot-spots Ignition surface Droplets Hot-spots Abnormal combustion Fuel films on cold walls do not fully vaporize during combustion, but instead accumulate over many cycles. [*] As the engine warms, the lighter components of the film vaporize, leaving a film of increasingly heavy composition; Eventually, the wall reaches a temperature where the film fully vaporizes. Intake valves Closed-valve injection Open-valve injection Valve firing Knock Hot-spots 33.6 CAD ASOS [*] Witze, P. O. and Green, R. M. SAE Paper No , 997. Pool fire images of gasoline wall films during a simulated cold start, observed through a window in the piston [*]. [**] C. Arcoumanis et al. Int. J. Engine Research Vol. n.,. Pool fire: soot concentration KL exhaust intake exhaust exhaust intake intake 46 CAD ASOS 7 CAD ASOS 9