Mark P. B. Musculus

On the Correlation between NOx Emissions and the Diesel Premixed Burn

It is generally accepted that exhaust NO x emissions of diesel engines increase with the degree of premixed burning. Although several mechanisms proposed in the literature are likely responsible for some aspects of the correlation, taken together, they cannot explain all observations of this correlation. In this study, thermodynamic analyses and optical/imaging diagnostics were employed in an optically-accessible, heavy-duty Dl diesel engine to examine the in-cylinder mechanisms by which fuel/air premixing affects engine-out NO x emissions. Exhaust NO and NO x emissions were measured and correlated with observations of soot luminosity and jet penetration as the intake-temperature and injection timing were varied The engine was operated at low-load conditions, for which the premixed burn was a significant fraction of the total heat released. As injection timing was retarded to the misfire limit, dramatic reductions in soot luminosity accompanied increased exhaust NO x emissions, even as the calculated adiabatic flame temperatures decreased. Since thermal NO formation increases with temperature, the reduction of the cooling effect of soot radiative heat transfer may increase the actual flame temperature, and thus NO formation, as the jets become less sooty. Compression heating of the reactant mixture by the large pressure rise during premixed combustion was found to be insufficient to explain exhaust NO x trends. Compression of post-flame burned-gases, however, could be responsible for some aspects of the correlation. Though the products of the premixed burn are normally too rich to form significant thermal NO, under very long ignition-delay conditions, portions of the mixture at ignition may be lean enough for significant thermal NO formation.

Effects of the In-Cylinder Environment on Diffusion Flame Lift-Off in a DI Diesel Engine

The diffusion flame lift-off length of isolated, free diesel jets in quiescent atmospheres is known to have a strong influence on soot formation by affecting fuel/air mixing prior to combustion. In realistic engine environments, the proximity and temperature of in-cylinder surfaces, in-cylinder gas flows (swirl), and interactions between adjacent jets may affect the behavior of the flame lift-off, and thereby affect soot formation. To better understand the influence of these factors on the lift-off length and on soot formation, optical imaging diagnostics were employed to measure the flame lift-off length in an optically-accessible heavy-duty Direct Injection (DI) diesel engine. A two-camera OH chemiluminescence diagnostic was developed and employed to measure the flame lift-off length for a range of injector nozzle geometries and engine operating conditions. A two-camera OH Planar Laser-Induced Fluorescence (OH-PLIF) diagnostic for side-on illumination of the jet,was also developed and utilized to aid in the interpretation of the OH chemiluminescence data. The diesel flame lift-off lengths in the optical engine were shorter than those observed previously for single, isolated jets in a quiescent atmosphere. Also, high cycle-to-cycle variation was observed in the lift-off length data from the engine, yielding considerable uncertainty in most of the reported trends. However, on average, the lift-off length was 7% longer on the windward side of the jet relative to the weak in-cylinder swirl flow. The proximity of the firedeck to the jets (14° down-angle) did not cause preferential vertical asymmetry in the flame shape, on average. A reduction in the interjet spacing angle from 90° (i.e., 4-holes) to 45° (8-holes) was correlated with a 35% reduction in the lift-off length, along with decreased sensitivity to ambient gas temperature and density. The mechanism by which the proximity of adjacent jets affects the lift-off length was not firmly identified, but fluid mechanical/thermal coupling between jets and/or internal injector fuel flow characteristics may be important. In realistic Dl diesel engine environments, high cycle-to-cycle variability, decreased temperature dependence, and shortening of the lift-off length may hinder soot reduction methods that rely on manipulation of the mean flame lift-off length.

2-Color Thermometry Experiments and High-Speed Imaging of Multi-Mode Diesel Engine Combustion

Although in-cylinder optical diagnostics have provided significant understanding of conventional diesel combustion, most alternative combustion strategies have not yet been explored to the same extent. In an effort to build the knowledge base for alternative low-temperature combustion strategies, this paper presents a comparison of three alternative low-temperature combustion strategies to two high-temperature conventional diesel combustion conditions. The baseline conditions, representative of conventional high-temperature diesel combustion, have either a short or a long ignition delay. The other three conditions are representative of some alternative combustion strategies, employing significant charge-gas dilution along with either early or late fuel injection, or a combination of both (double-injection). These operating conditions are investigated for soot volume fraction, soot temperatures, calculated adiabatic flame temperatures, and soot radiation heat loss through 2-color soot thermometry experiments. The spatial location of in-cylinder soot is imaged using a high-speed CMOS camera, and exhaust-gas NO x is also measured. The soot thermometry and high-speed soot luminosity imaging show that the low-temperature operating conditions have lower in-cylinder soot than the high-temperature conditions. Also, soot is formed upstream in the jet for high-temperature operating conditions, but for low-temperature operating conditions, the soot is formed farther downstream, closer to the bowl edge. For all conditions, the onset of in-cylinder soot occurs after the premixed bum, during the mixing-controlled combustion phase. As the amount of soot decreases, the radiation heat loss also decreases drastically. For conventional diesel diffusion combustion operating condition, radiation from soot is about 1.1 percent of the total fuel energy, but for low-temperature combustion operating conditions, the soot radiative heat loss is almost negligible (≈ 0.01 percent). The condition with high soot radiation had peak soot temperatures as much as 300 K lower than the peak adiabatic flame temperatures near 2700 K, and exhaust NO x emissions were near 600 ppm. For the low-temperature conditions, the peak soot temperatures were only about 200 K lower than the peak adiabatic temperatures near 2200 K, and the exhaust NO x concentrations were less than 10 ppm.

Entrainment waves in decelerating transient turbulent jets

A simplified one-dimensional partial differential equation for the integral axial momentum flux during the deceleration phase of single-pulsed transient incompressible jets is derived and solved analytically. The wave speed of the derived first-order nonlinear wave equation shows that the momentum flux transient from the deceleration phase propagates downstream at twice the initial jet penetration rate. Transient-jet velocity data from the existing literature is shown to be consistent with this derivation, and an algebraic analytical solution matches the measured timing and decay of axial velocity after the deceleration transient. The solution also shows that a wave of increased entrainment accompanies the deceleration transient as it travels downstream through the jet. In the long-time limit, the peak entrainment rate at the leading edge of this ‘entrainment wave’ approaches an asymptotic value of three times that of the initial steady jet. The rate of approach to the asymptotic behaviour is controlled by the deceleration rate, which suggests that rate-shaping may be tailored to achieve a desired mixing state at a given time after the end of a single-pulsed jet. In the wake of the entrainment wave, the absolute entrainment rate eventually decays to zero. The local injected fluid concentration also decays, however, so that entrainment rate relative to the local concentration of injected fluid remains higher than in the initial steady jet. An analysis of diesel engine fuel-jets is provided as one example of a transient-jet application in which the considerable increase in the mixing rate after the deceleration phase has important implications.

SIMULTANEOUS OPTICAL DIAGNOSTIC IMAGING OF LOW-TEMPERATURE, DOUBLE-INJECTION COMBUSTION IN A HEAVY-DUTY DI DIESEL ENGINE

Abstract A highly diluted, low-flame-temperature diesel engine combustion strategy with two separate fuel-injections per cycle was investigated using simultaneous optical diagnostics at a low-load operating condition. In-cylinder processes were visualized with a suite of laser/imaging diagnostics. The cool flame first-stage ignition reactions occur along the entire length of the jet for the first combustion event. For both injections, the second-stage ignition reactions occur after the end of injection, primarily in the downstream regions of the jet. OH is found throughout the cross-section of the jet, indicating greater mixing and leaner mixtures than observed for conventional diesel combustion. For the first combustion event, very little soot is formed and it is found in small pockets near the tip of the jet. For the second combustion event, much more soot is formed throughout the downstream jet cross-section. Finally, Sandias conceptual model of diesel combustion has been extended to describe this operating condition.

Effects of exhaust gas recirculation and load on soot in a heavy-duty optical diesel engine with close-coupled post injections for high-efficiency combustion phasing

In-cylinder strategies to reduce soot emissions have demonstrated the potential to lessen the burden on, and likely the size and cost of, exhaust aftertreatment systems for diesel engines. One in-cylinder strategy for soot abatement is the use of close-coupled post injections. These short injections closely following the end of the main injection can alter soot-formation and/or oxidation characteristics enough to significantly reduce engine-out soot. Despite the large body of literature on post injections for soot reduction, a clear consensus has not yet been achieved regarding either the detailed mechanisms that affect the soot reduction, or even the sensitivity of the post-injection efficacy to several important engine operating parameters. We report that post injections reduce soot at a range of close-coupled post-injection durations, intake-oxygen levels, and loads in an optical, heavy-duty diesel research engine. Maximum soot reductions by post injections at the loads and conditions tested range from 40% at 21% intake oxygen (by volume) to 62% at 12.6% intake oxygen. From a more fundamental fluid-mechanical perspective, adding a post injection to a constant main-injection for conditions with low dilution (21% and 18% intake oxygen) decreases soot relative to the original main injection, even though the load is increased by the post injection. High-speed visualization of natural combustion luminosity and laser-induced incandescence of soot suggest that as the post-injection duration increases and the post injection becomes more effective at reducing soot, it interacts more strongly with soot remaining from the main injection.

The influence of large-scale structures on entrainment in a decelerating transient turbulent jet revealed by large eddy simulation

To provide a better understanding of the fluid mechanical mechanisms governing entrainment in decelerating jets, we performed a large eddy simulation (LES) of a transient air jet. The ensemble-averaged LES calculations agree well with the available measurements of centerline velocity, and they reveal a region of increased entrainment that grows as it propagates downstream during deceleration. Within the temporal and spatial domains of the simulation, entrainment during deceleration temporarily increases by roughly a factor of two over that of the quasi-steady jet, and thereafter decays to a level lower than the quasi-steady jet. The LES results also provide large-structure flow details that lend insight into the effects of deceleration on entrainment. The simulations show greater growth and separation of large vortical structures during deceleration. Ambient fluid is engulfed into the gaps between the large-scale structures, causing large-scale indentations in the scalar jet boundary. The changes in the g...

Optical Diagnostics of Late-Injection Low-Temperature Combustion in a Heavy-Duty Diesel Engine

A late injection, high exhaust-gas recirculation (EGR)-rate, low-temperature combustion strategy was investigated in a heavy-duty diesel engine using a suite of optical diagnostics: chemiluminescence for visualization of ignition and combustion, laser Mie scattering for liquid fuel imaging, planar laser-induced fluorescence (PLIF) for both OH and vapor-fuel imaging, and laser-induced incandescence (LII) for soot imaging. Fuel is injected at top dead center when the in-cylinder gases are hot and dense. Consequently, the maximum liquid fuel penetration is 27 mm, which is short enough to avoid wall impingement. The cool flame starts 4.5 crank angle degrees (CAD) after the start of injection (ASI), midway between the injector and bowl-rim, and likely helps fuel to vaporize. Within a few CAD, the cool-flame combustion reaches the bowl-rim. A large premixed combustion occurs near 9 CAD ASI, close to the bowl rim. Soot is visible shortly afterwards along the walls, typically between two adjacent jets at the head vortex location. OH PLIF indicates that premixed combustion first occurs within the jet and then spreads along the bowl rim in a thin layer, surrounding soot pockets at the start of the mixing-controlled combustion phase near 17 CAD ASI. During the mixing-controlled phase, soot is not fully oxidized and is still present near the bowl-rim late in the cycle. At the end of combustion near 27 CAD ASI, averaged PLIF images indicate two separate zones. OH PLIF appears near the bowl rim, while broadband PLIF persists late in the cycle near the injector. The most likely source of broadband PLIF is unburned fuel, which indicates that the near-injector region is a potential source of unburned hydrocarbons.

Effect of Load on Close-Coupled Post-Injection Efficacy for Soot Reduction in an Optical Heavy-Duty Diesel Research Engine

The use of close-coupled post injections of fuel is an in-cylinder soot-reduction technique that has much promise for high efficiency, heavy-duty diesel engines. Close-coupled post injections, short injections of fuel that occur soon after the end of the main fuel injection, have been known to reduce engine-out soot at a wide range of engine operating conditions, including variations in injection timing, EGR level, load, boost, and speed. While many studies have investigated the performance of post injections, the details of the mechanism by which soot is reduced remains unclear. In this study, we have measured the efficacy of post injections over a range of load conditions, at constant speed, boost, and rail pressure, in a heavy-duty, optically-accessible research diesel engine. Here, the base load is varied by changing the main-injection duration. Measurements of engine-out soot indicate that not only does the efficacy of a post injection decrease at higher engine loads, but that the range of post-injection durations over which soot reduction is achievable is limited at higher loads. Optical measurements, including natural luminescence of soot and planar laser-induced incandescence of soot, provide information about the spatio-temporal development of in-cylinder soot through the cycle in cases with and without post injections. The optical results indicate that the post injection behaves similarly at different loads, but that its relative efficacy decreases due to the increase in soot resulting from longer main-injection durations.Copyright

Optical Diagnostics of a Late Injection Low-Temperature Combustion in a Heavy Duty Diesel Engine

A late injection, high exhaust-gas recirculation (EGR)-rate, low-temperature combustion strategy was investigated in a heavy-duty diesel engine using a suite of optical diagnostics: chemiluminescence for visualization of ignition and combustion, laser Mie scattering for liquid fuel imaging, planar laser-induced fluorescence (PLIF) for both OH and vapor-fuel imaging, and laser-induced incandescence (LII) for soot imaging. Fuel is injected at top dead center when the in-cylinder gases are hot and dense. Consequently, the maximum liquid fuel penetration is 27 mm, which is short enough to avoid wall impingement. The cool flame starts 4.5 crank angle degrees (CAD) after the start of injection (ASI), midway between the injector and bowl-rim, and likely helps fuel to vaporize. Within a few CAD, the cool-flame combustion reaches the bowl-rim. A large premixed combustion occurs near 9 CAD ASI, close to the bowl rim. Soot is visible shortly afterwards along the walls, typically between two adjacent jets at the head vortex location. OH PLIF indicates that premixed combustion first occurs within the jet and then spreads along the bowl rim in a thin layer, surrounding soot pockets at the start of the mixing-controlled combustion phase near 17 CAD ASI. During the mixing-controlled phase, soot is not fully oxidized and is still present near the bowl-rim late in the cycle. At the end of combustion near 27 CAD ASI, averaged PLIF images indicate two separate zones. OH PLIF appears near the bowl rim, while broadband PLIF persists late in the cycle near the injector. The most likely source of broadband PLIF is unburned fuel, which indicates that the near-injector region is a potential source of unburned hydrocarbons.