Evolution, challenges and path forward for low temperature combustion engines

Nomenclature a = constant; Fig. 1 C , C 0 = constants k = roughness element height, ft N k = average roughness element height, ft L = vehicle length, ft M = Mach number n = exponent; Fig. 1 N R = Poll’s transition parameter; Eq. (1) Reke = roughnessReynolds number based on height k and edge conditions Rekk = roughnessReynolds number based on height k and conditions at k Reμ = Reynolds number based on height μ and edge conditions U = velocity component parallel to test surface or velocity component perpendicularto attachment line, ft /s V = velocity component parallel to attachment line, ft/s X = generalized disturbanceparameter or axial coordinate along windward centerline x = coordinate perpendicular to attachment line, ft Y = generalized transition parameter ® = angle of attack, deg ± = smooth-wall laminar boundary-layer thickness, ft = Poll’s length scale [Eq. (2)], ft μ = smooth-wall laminar boundary-layermomentum thickness, ft 1 = viscosity, lbm/ft ¢ s o = kinematic viscosity, ft2/s 1⁄2 = density, lbm/ft

Review and synthesis of roughness-dominated transition correlations for reentry applications

The starting characteristics of a small-scale rectangular inlet with a thick ingested boundary layer were investigated at nominal Mach 3 conditions. Parameters investigated included Reynolds number, cowl length, and cowl height. Measurements of the maximum and restart contraction ratios were made. Depending on the test configuration, the unstarts were classified into two broad categories as either "hard" or "soft." The hard unstarts appear to occur when the flow at the inlet throat chokes. The soft unstarts occur as large-scale separation develops within the inlet. The ability of the classical Kantrowitz limit to predict the restart contraction ratio was assessed, and it was shown to be applicable for the hard unstart/ restart configurations. The role of fluid injection upstream of the unstarted inlet was also assessed. The use of this injection may ultimately lead to improving the starting characteristics of inlets. Nomenclature A Area hc Cowl height Lc Cowl length rh Mass flow M Mach number M Mass-averaged Mach number P Pressure Pt Freestream total pressure x, y Cartesian coordinates y Ratio of specific heats 6 Boundary layer thickness 6* Boundary layer displacement thickness 0 Boundary layer momentum thickness 9C Cowl angle p Density * Principal staff engineer, senior member AIAA ** Senior staff engineer, member AIAA ^ Associate staff engineer, member AIAA Copyright © American Institute of Aeronautics and Astronautics, Inc., 1996. All rights reserved. Subscripts 0 Freestream 2 Entrance to cowl 4 Inlet throat cl Wall conditions at cowl lip inj Injectant Introduction Airbreathing engines that operate at supersonic and hypersonic speeds require inlets to capture and compress air for processing by the remainder of the engine. The goal in the design of any inlet is to define a minimum weight geometry that provides an efficient compression process, generates minimum drag, produces nearly uniform flow entering the compressor or combustor, and provides these characteristics over a wide range of flight and engine operating conditions. For efficient operation and moderate induced drag, most inlets use a combination of external and internal compression. The introduction of internal contraction in an inlet adds complexity in the design and analysis process in that the starting of the inlet must be ensured. For efficient operation, supersonic and hypersonic inlets must operate in a started mode. The process of inlet starting and unstarting is well understood at a conceptual level, although significant details remain to be resolved. Some variation exists in the very definition of a started inlet. One convention states that a started inlet is one with supersonic flow in the inlet throat, but it is well known that some unstarted inlets can have complex internal flowfields with a significant fraction of supersonic flow in the inlet throat. In the present work, the term "started" is used to denote operation under conditions where flow phenomena in the internal portions of the inlet do not alter the air capture characteristics of the inlet. (Reduction in the captured mass flow through the use of bleed holes or bypass channels is not considered in assessing whether an inlet is started.) An inlet can be unstarted by either over-contracting to the point where the flow chokes at the inlet throat or by raising the back pressure beyond the level that can be sustained by the inlet. Currently, a significant uncertainty exists regarding the conditions under which an inlet will unstart or restart. Part of this uncertainty is due to the large variety of 1 American Institute of Aeronautics and Astronautics geometries that have been considered in designing engines. The design of an inlet is strongly affected by vehicle considerations, and a variety of two-dimensional planar, axisymmetric, and three-dimensional inlet designs have been investigated. The diversity found in inlet designs can be seen in the sample inlets shown in Fig. I.'" Preliminary estimates of the internal contraction that will self-start can be obtained from the Kantrowitz limit. This limit is determined by assuming a normal shock wave at the beginning of the internal contraction and calculating the one-dimensional, isentropic internal area ratio that will produce sonic flow at the inlet throat. For a perfect gas, the Kantrowitz limit can be calculated as follows: Oswatisch inlet HRE-type inlet A2] /KANTROWITZ M, l)MJ

Starting characteristics of supersonic inlets

Effect of fuel injection parameters on combustion stability and emissions of a mineral diesel fueled partially premixed charge compression ignition (PCCI) engine

Aspects of Hypersonic Boundary Layer Transition Control

We present boundary-layer transition data on swept cylinders in hypersonic flow that were obtained in the hypersonic Ludwieg-tube wind tunnel of DLR. Experiments were conducted at M∞ = 5.0 and 6.9 for three swept cylinders with and without end plates and sweep angles of 30, 45, and 60 deg, respectively. To determine the state of the boundary layers, the liquid-crystal technique was applied. In addition, the response of attachment-line transition to surface roughness heights due to trip wires for sweep angles of 45 and 60 deg at M∞ = 5.0 was investigated. Freestream Reynolds numbers based on cylinder diameter for attachment-line transition were about 0.2-0.3 x 10 6 for the end plate disturbances and 0.9-1.2 x 10 6 for no contaminations. Trends of attachment-line transition Reynolds numbers with surface roughness heights are roughly similar to subsonic ones with respect to critical roughness heights. The behavior of the attachment-line transition around the critical roughness heights, however, depended strongly on spanwise Mach numbers. With flow visualizations, off-attachment-line transition could be observed, as well as fine streak pattern in laminar flow region upstream of the transition front, confirming streamwise vortices induced by crossflow instability at hypersonic speeds.

Boundary-layer transition on swept cylinders at hypersonic speeds

Effects of Compression Ratio on Characteristics of PCCI Diesel Combustion with a Hollow Cone Spray

Porous Bleed Model for Boundary Condition of CFD Analysis

Numerical simulations of the unsteady supersonic flow through two-dimensional external compression intakes were performed. The simulations with Euler equations reached to a persistently oscillatory state in subcritical range below some value of the mass flow rate. The frequency of the simulated flow increased with the decrease in the mass flow rate. The intake flows with different inlet Mach numbers, ramp angles and subsonic diffuser area ratios were simulated to examine the effect of those parameters on both of the criteria and frequency of the shock oscillation. The simulations with NavierStokes equations using a low Reynolds number kepsilon turbulence model were also performed, the results of which were compared to the experimental data recently obtained in a supersonic wind tunnel. The shock oscillation frequency predicted by the numerical simulation agrees well with the experimentally obtained frequency, both of which increased with the decrease in the mass flow rate.

Akira Murakami

Papers

Boundary-layer transition on swept cylinders at hypersonic speeds

Effects of Compression Ratio on Characteristics of PCCI Diesel Combustion with a Hollow Cone Spray

Porous Bleed Model for Boundary Condition of CFD Analysis

Numerical Analysis on Shock Oscillation of Two-Dimensional External Compression Intakes

Swirl Measurements and Modeling in Direct Injection Diesel Engines