Benjamin W. Knox

Combustion Recession after End of Injection in Diesel Sprays

This work contributes to the understanding of physical mechanisms that control flashback, or more appropriately combustion recession, in diesel-like sprays. Combustion recession is the process whereby a lifted flame retreats back towards the injector after end-of-injection under conditions that favor autoignition. The motivation for this study is that failure of combustion recession can result in unburned hydrocarbon emissions. A large dataset, comprising many fuels, injection pressures, ambient temperatures, ambient oxygen concentrations, ambient densities, and nozzle diameters is used to explore experimental trends for the behavior of combustion recession. Then, a reduced-order model, capable of modeling non-reacting and reacting conditions, is used to help interpret the experimental trends. Finally, the reduced-order model is used to predict how a controlled ramp-down rate-ofinjection can enhance the likelihood of combustion recession for conditions that would not normally exhibit combustion recession. In general, fuel, ambient conditions, and the spray rate-of-injection transient during the end-of-injection determine the success or failure of combustion recession. The likelihood of combustion recession increases for higher ambient temperatures and oxygen concentrations as well as for higher reactivity fuels. In the transition between high and low ambient temperature (or oxygen concentration), the behavior of combustion recession changes from spatially sequential ignition to separated, or isolated, ignition sites that eventually merge. In contradistinction to typical diesel ignition delay trends where the autoignition times are longer for increasing injection pressure, the time required for combustion recession increases with injection pressure.

A Comparison of Fluidic and Physical Obstacles for Deflagration-to-Detonation Transition

Abstract : A fluidic obstacle has been proposed as an alternative to conventional deflagration-to-detonation transition (DDT) enhancement devices for use in a Pulsed Detonation Engine (PDE). Experimental results have been obtained utilizing unsteady reacting and steady non-reacting flow to gain insight on the relative performance of a fluidic obstacle. Using stoichiometric premixed hydrogen-air, transition to detonation has been achieved using solely a fluidic obstacle with comparable DDT distances to that of a physical orifice plate. Flame acceleration is achieved due to the intense turbulent mixing characteristics inherent of a high-velocity jet and the blockage created by the virtual obstacle. Turbulence intensity (T.I.) measurements, taken downstream of both obstacles with hot-film anemometry during non-reacting steady flow, show a conservative trend that a fluidic obstacle produces approximately a 240% increase in turbulence intensity compared to that of a physical obstacle. Ignition times were reduced approximately 45%, attributable to the increase in upstream T.I. levels relative to the fluidic obstacle during the fill portion of the PDEs cycle. Transition to detonation was obtained for injection compositions of both premixed stoichiometric hydrogen-air and pure air.

Reduced-order numerical model for transient reacting diesel sprays with detailed kinetics

A reduced-order model of transient diesel spray combustion is presented that utilizes simplified fluid mechanics and detailed chemical kinetics, premised on the similarity between dense turbulent gaseous jets and diesel sprays at engine conditions. The presented model offers a new capability for detailed chemistry predictions in transient diesel sprays since the use of large chemical mechanisms is prohibitively expensive in more detailed modeling approaches such as multidimensional computational fluid dynamics. The numerical model is validated against Engine Combustion Network spray-H experimental data. Predictions of vapor penetration, axial mixture fraction distribution, ignition delay, axial location of cool-flame reaction, and end-of-injection combustion recession show excellent agreement with experimental measurements. The model is applied to study modern diesel injection strategies that involve significant transient mixing and combustion behavior, including fuel injection rate shaping and close-coupled split-injection strategies. In general, the model is shown to enable a detailed examination of modern diesel injection strategies and the expected impact of these strategies on emissions. A slow ramp down of fueling rate at the end of injection is found to limit over-mixing in the near field of the injector, enabling recession of second-stage ignition toward the injector after end of injection. This is advantageous for consumption of unburned hydrocarbons and improved combustion efficiency. Compared to slow ramp-down injection strategies, close-coupled split injections are less effective for unburned hydrocarbon reduction due to a strong end-of-injection entrainment wave that accompanies both injections, causing rapid over-leaning and no recession of second-stage ignition.

Computational analysis of end-of-injection transients and combustion recession

Mixing and combustion of engine combustion network Spray A after end of injection are modeled using highly resolved multidimensional numerical simulations to explore the physics underlying recent experimental observations of combustion recession. Reacting spray simulations are performed using a traditional Lagrangian–Eulerian coupled formulation for two-phase mixture transport with a Reynolds-averaged Navier–Stokes approach using the open-source computational fluid dynamics code OpenFOAM. Chemical kinetics models for n-dodecane by Cai et al. and Yao et al. are deployed to evaluate the impact of mechanism formulation and low-temperature chemistry on predictions of combustion recession behavior. Simulations with the Cai mechanism show that under standard Spray A conditions, the end-of-injection transient induces second-stage ignition in distinct regions near the nozzle that are initially spatially separated from the lifted diffusion flame, but then rapidly merge with flame. By contrast, the Yao mechanism fails to predict sufficient low-temperature chemistry in mixtures upstream of the diffusion flame during the end-of-injection transient and does not predict combustion recession for the same conditions. The effects of the shape and duration of the end-of-injection transient on the entrainment wave near the nozzle, the likelihood of combustion recession, and the spatiotemporal development of mixing and chemistry in near-nozzle mixtures are also investigated. With a more rapid ramp-down injection profile (ramp-down duration < 400 µs), a weaker combustion recession occurs earlier in time after the start of ramp-down. For extremely fast ramp-down (ramp-down duration = 0), the entrainment flux varies rapidly near the nozzle and over-leaning of the mixture completely suppresses combustion recession. For a slower ramp-down profile with respect to the standard Spray A condition, complete combustion recession back toward the nozzle is observed and combustion recession occurred later in time. Simulations qualitatively agreed with the past experimental and modeling observations of combustion recession with different end-of-injection transients.

Identifying Uncertainties in Diesel Spray Rate-of-Momentum Transients Under Elevated Back Pressure

Rate-of-momentum measurements of transient fuel sprays are valuable for improving current combustion and emission strategies. This data provides boundary conditions for engine computational fluid dynamic (CFD) simulations and provides insight into the transient mixing characteristics of the spray prior to and during combustion. Previous researchers have quantified the rate-of-momentum of transient sprays using the impingement technique, but uncertainties remain in relating the impingement force to the injected fuel momentum at the nozzle exit. Rate-of-momentum measurements are typically performed by directing a spray onto the face of a calibrated transducer in close proximity to the nozzle. The measured impingement force is then used to quantify the rate-of-momentum at the nozzle orifice exit with the aid of a simplified control volume analysis. However, under elevated back pressures, additional terms in the control volume analysis are no longer negligible. Other non-idealities, such as non-orthogonal droplet impingement outcomes and transient mass accumulation in the control volume, can also contribute to errors in the simplified analysis.This paper investigates the impact of non-idealities in impingement-based rate-of-momentum measurements on the quantified fuel injection rate. In specific, we compare the measured rate-of-momentum under back pressure and atmospheric pressure using two different transducers to quantify uncertainties that can arise under back pressure conditions. Uncertainties associated with transient mass accumulation and non-orthogonal spray deflection are also investigated. We found that back pressure affected both the start and end of injection when compared to atmospheric pressure. Under back pressure, there was a lengthened apparent start-of-injection transient, which likely results from a low pressure toroidal vortex occurring at the head of the spray. In addition, there was a longer apparent closing transient, which is likely a result of residual pressure distribution after the end-of-injection. No evidence of transient mass accumulation was observed for the injectors used in this study. Lastly, the transient spray was observed to deflect non-orthogonally from the impact point on the transducer instead of remaining parallel to the transducer face after initial impact. This deflection of the spray leads to uncertainties when quantifying the rate-of-momentum, where the apparent rate-of-momentum could be larger than the actual value.Copyright

Diesel Spray Rate-of-Momentum Measurement Uncertainties and Diagnostic Considerations

Interpretation of combustion and emissions outcomes in diesel engines is often enhanced by accurate knowledge of the transient fuel delivery rate and flow characteristics of the injector nozzle. Important physical characteristics of these flows, including velocity profile and flow separation or cavitation effects, are difficult to measure directly, but can be characterized from a flow-averaged perspective through the measurement of nozzle flow coefficients, namely the discharge, velocity, and area contraction coefficients. Both the transient fuel mass flow rate and the flow-averaged nozzle coefficients can be found by measuring the mass and momentum flux of the fuel stream leaving the nozzle during injection through the application of an impingement technique, where fuel is sprayed onto the face of a transducer calibrated for force measurement in close proximity to the nozzle. While several published experiments have employed the spray impingement method to quantify rate-of-injection, the experimental setup and equipment selections vary widely and may contribute to disagreements in measured rate-of-injection. This paper identifies and provides estimates of measurement uncertainties that can arise when employing different experimental setups using the impingement method. It was observed that the impingement technique was sensitive to the design of the strike cap, specifically the contact area between the cap and transducer diaphragm, in addition to fuel temperature. Conversely, we observed that the impingement technique was relatively insensitive to angular and vertical misalignment, where the uncertainty can be estimated using control volume analysis. Transducer selection, specifically those with low acceleration sensitivity, high resonant frequency, and integrated electronics piezoelectric circuitry substantially reduce the noise in the measurement.Copyright