James C. Hermanson

Effects of Coflow on Turbulent Flame Puffs

Pulsed, turbulent jet diffusion flames in an air coflow of variable strength were examined experimentally. In all cases, the flames were fully modulated, that is, the fuel flow was completely shut off between pulses. Isolated puffs of unheated ethylene fuel were injected using a 2-mm-diam-nozzle into a combustor with an air coflow at 1-atm pressure. For short injection times (τ<50 ms), compact, pufflike structures were generated. The mean flame length of these puffs was at least 51% less than that of a steady-state, that is, nonpulsed, flame for the same injection Reynolds number. More elongated flame structures, with a flame length closer to that of steady-state flames, occurred for longer injection times of up to 300 ms. The addition of coflow generally causes an increase in the mean flame length. For short injection times (r < 50 ms), this resulted in an increase in flame length of up to 27% for a coflow strength of U cof /U jet = 0.02. The fractional increase in the flame length due to coflow of pulsed flames with longer injection times, as well as steady flames, was significantly less. The mean flame length for the flame with the coflow duct generally exceeded that of the corresponding free flame, even for the case of zero coflow. The amount of coflow required to achieve a given increase in mean flame length is quantitatively consistent with a scaling argument developed as part of this investigation.

Experimental investigation of convective structure evolution and heat transfer in quasi-steady evaporating liquid films

The stability, convective structure, and heat transfer characteristics of upward-facing, evaporating, thin liquid films were studied experimentally. Dichloromethane, chloroform, methanol, and acetone films with initial thicknesses of 2–5 mm were subjected to constant levels of superheating until film rupture occurred (typically at a thickness of around 50 μm). The films resided on a temperature controlled, polished copper plate incorporated into a closed pressure chamber free of non-condensable gasses. The dynamic film thickness was measured at multiple points using a non-intrusive ultrasound ranging system. Instability wavelength and convective structure information was obtained using double-pass schlieren imaging. The sequence of the convective structures as the film thins due to evaporation is observed to be as follows: (1) large, highly variable cells, (2) concentric rings and spirals, and (3) apparent end of convection. The transition from large, variable cells to concentric rings and spirals occurs ...

Dynamics of Supersonic Droplets of Volatile Liquids

T HE breakup and vaporization of liquid droplets in supersonic flow is an interesting research problem with potentially important implications for supersonic combustion ramjets (scramjets). Noncryogenic, liquid hydrocarbons have substantial benefits as scramjet fuel [1], including higher energy density, lower cost, and ease of handling compared to liquid hydrogen fuel. The complications and time associated with the atomization and vaporization of liquid hydrocarbon scramjet fuel can, in principle, be avoided by injecting the fuel in the vapor phase. In some situations, however, such as a “cold start” in which the fuel is not preheated, hydrocarbon fuel may necessarily be injected while still in liquid phase. In this case, the rates and physical mechanisms associated with the disruption and vaporization of liquid droplets under supersonic flow conditions become critical issues to scramjet combustor performance. One possible technique to increase the dispersion of liquid fuels is to exploit the accelerated vaporization made possible by superheating the liquid. Investigation of the vaporization of superheated droplets and sprays have to date been largely confined to incompressible flows [2,3], with the physics of superheated liquid droplet disruption and vaporization in supersonic flow not yet well established. Studies of droplet disruption in compressible flows that have been conducted [4–6] have generally not considered the effects of superheating. The numerical study of Joseph et al. [7] did suggest a flash vaporization mechanism in the disruption of liquid drops in steady, supersonic flow, arguing that superheating may occur over small regions of the droplet surface due to a local combination of low pressure and frictional heating. The disruption of droplets in high-speed flow has often been studied by the sudden application of aerodynamic loads though the use of shock tubes [4,6,8–10]. Though this technique can produce liquid droplets under locally supersonic conditions, this is accomplished only after the passage of a shock wave through the droplets. Other droplet disruption studies have been conducted at subsonic speeds [11,12], for example, by droplets falling across a high-speed gas jet [13], droplet-bearing jets in cross flow [14], and in drop tubes [15]. In any case, these techniques do not typically result in the droplets achieving a significant degree of superheating. The research presented here investigated the dynamics of droplets consisting of volatile fluid smoothly accelerated to supersonic Mach numbers without passage of shock waves through the droplet. This was accomplished over a range of liquid vapor pressures using a compact, underexpanded supersonic jet configuration. This is an extension of a previous study of superheated droplet disruption with compressible, but subsonic, flow relative to the droplets [16]. One challenging aspect in the study of liquid drops in high-speed airstreams is the measurement of the drop velocity and acceleration [17]. Double-pulsed, planar laser imaging was employed here to determine the velocity of the droplets for various test fluids.

Ultrasonic measurement of condensate film thickness

The current work describes a modified time-of-flight ultrasound signal processing technique applied to the study of a distal liquid layer with a free surface. The technique simulates multiple reflections analytically and determines the film thickness by comparison to the measured pulse echo signal. The technique is applied with 20 MHz transducers to an n-pentane film condensing on a copper plate. The technique proved capable of measuring liquid thickness from approximately 8 microm, 16 the acoustic wavelength in pentane, to greater than 5 mm. Near the lower thickness limit, echoes from the liquid/vapor interface overlap each other and the significantly larger echoes from the metal/liquid interface.

Breakup and vaporization of droplets under locally supersonic conditions

The disruption and vaporization of simulated fuel droplets in an accelerating supersonic flow was examined experimentally in a draw-down supersonic wind tunnel. The droplets achieved supersonic velocities relative to the surrounding air to give relative Mach numbers of up to 1.8 and Weber numbers of up to 300. Mono-disperse, 100 μm-diameter fluid droplets were generated using a droplet-on-demand generator upstream of the tunnel entrance. Direct close-up single- and multiple-exposure imaging was used to examine the features of droplet breakup and to determine the droplet velocities. Laser-induced fluorescence (LIF) imaging of the disrupting droplets was performed using acetone fluorescence to determine the dispersion of the expelled vapor. Three test liquids were employed: 2-propanol and tetraethylene glycol dimethyl ether as non-volatile fluids and a 50/50 hexanol-pentane mixture (Hex-Pen 50/50). The vapor pressure of the Hex-Pen 50/50 was sufficiently high to cause the droplet fluid to potentially become...

Disruption of Volatile and Nonvolatile Droplets Under Locally Supersonic Conditions

The disruption of simulated fuel droplets in supersonic flow is examined experimentally in a drawdown supersonic wind tunnel. The droplets are accelerated in the supersonic flow, achieving supersonic velocities relative to the surrounding air with relative Mach numbers as high as 1.8 and Weber numbers as high as 300. Monodisperse 100m-diam fluid droplets are generated using a droplet-on-demand generator upstream of the tunnel entrance. The droplets are imaged by direct close-up singleandmultiple-exposure imaging. Three test liquidswere employed: 2-propanol and tetraethylene glycol dimethyl ether as nonvolatile fluids, and a more volatile 50=50 hexanol-pentane mixture. The decreased static pressure in the supersonic stream had the potential to give rise to superheating of the droplet fluid, as in some cases, the static pressure became significantly lower than the vapor pressure of the droplet liquid. Droplet lifetimes for the hexanol/pentane mixture appear to be shorter due to accelerated vaporization consistent with superheating, although little impact is observed on the droplet velocity and relative Mach number. Droplet-disruption patterns for these supersonic flow conditions can be classified into four different flow regions by considering the changes in the Weber number with downstream distance as the droplets accelerate. The drag coefficients associatedwith the droplet disruption under locally supersonic conditions are generally higher than those expected for solid spheres, largely due to the cross-sectional area change associated with droplet deformation/ breakup.

CO/UNBURNED HYDROCARBON EMISSIONS OF STRONGLY-PULSED TURBULENT DIFFUSION FLAMES

The CO and unburned hydrocarbon (UHC) emissions of pulsed turbulent diffusion flames were examined by injecting unheated ethylene into a combustor with an air coflow at atmospheric pressure. In all cases the flames were fully modulated (fuel flow fully shut off between injection intervals). Video imaging was performed and time-averaged emissions were measured on the combustor centerline. For short injection times ( ≤ 46 ms), compact, puff-like structures were generated. Longer injection times produced elongated flame structures with flame lengths closer to that of steady flames. The highest emission indices of CO and UHC were found for compact, isolated puffs. The emissions for all flames approached the steady-flame levels for a duty cycle (jet-on fraction) of approximately 0.4. This suggests that there are combinations of injection time and duty cycle for fully modulated, turbulent diffusion flames that produce emissions comparable to the steady flame but with a significantly shorter flame length.

Numerical Simulation of Shock-Enhanced Mixing in Nonuniform Density Turbulent Jets

A numerical study of the interaction of weak normal shock waves with turbulent jets was conducted. The cone guration consisted of a planar jet of air, helium, or carbon dioxide situated on the centerline of a shock tube withanaircoe ow.Theshockstrengthswere Ms =1:2and1:4.Thenumericalmodelwasbasedon atime-dependent, Navier‐Stokes approach and a two-equation q‐! turbulence model. The results indicate that passage of a shock through low-density (helium)jets produces a vortexlike structure not seen for the case of the air or carbon dioxide jets. Helium jets exhibit a decrease in mean jet e uid concentration due to shock interaction of up to approximately 30% at a location 30 jet exit heights downstream of the jet exit for a shock strength of Ms =1:4. The amount of mixing enhancementincreases with increasing shock strength and decreases with increasing downstream distance. In comparison, the air and carbon dioxide jets show a signie cantly smaller degree of mixing enhancement. These results are qualitatively consistent with recent experimental results in axisymmetric, turbulent jets subject to normal shock passage.

CO/NOx Emissions of Strongly Pulsed Jet Diffusion Flames

The CO and NOX emissions of strongly pulsed, turbulent diffusion flames were examined experimentally in a co-flow combustor. Video imaging was performed and time-averaged emissions were measured at the combustor exit and near the visible flame tip. Both the case of a fixed fuel injection velocity during the injection interval and a constant fuelling rate were studied for jet Reynolds numbers ranging from 5,000 to 15,000. For fixed injection velocity, maximum emission indices of CO occurred for compact, isolated flame puffs. CO decreased substantially with decreasing jet-off time as the flame puff interaction increased. The pulsed flames had lower NOX than the steady flames, particularly for the case of isolated flame structures. Similar trends in NO formation were seen for constant fueling rate. The CO emissions were, however, considerably different, largely due to a significant impact of the Reynolds number. Radial emissions profiles suggest an improved fuel/air mixing for shorter jet-off times and longer jet-on times. The correlation of CO/NO emissions with a global flame residence time is discussed.

CO/NO Emissions of Strongly-Pulsed Turbulent Nonpremixed Flames

The CO and NO emissions of strongly-pulsed, turbulent diffusion flames are studied experimentally. In all cases the flames are fully-modulated (fuel completely shut off between pulses). The fuel is ethylene with an air co-flow. The emissions of NO and CO were measured at the combustor exit and on the combustor centerline. For short injection times the CO level substantially exceeds that of the steady (non-pulsed) flame, while the amount of NO is below the steady-flame level. The highest levels of CO and NO occur for short and long injection times, respectively. For sufficiently short jet off-times, as the flame structure interaction increases, levels of both CO and NO tend to reach the steady flame levels. These changes are consistent with an increased interaction between flame structures leading to a decreased air entrainment per amount of injected fuel and a longer flame length. The CO emissions, for a given fueling rate, are strongly dependent on both the injection time and jet off-time for a jet-on fraction less than approximately 50%. The NO levels are generally proportional to the fueling rate. The residence time based on the visual flame dimensions correlates reasonably well with the CO emission indices, reflecting incomplete combustions for low residence time of the flame structures; however, the NO levels do not directly scale with the residence time. Injection conditions exist that produce a more compact flame than the steady flame, and with significantly lower levels of NO but slightly higher CO emissions.