Wilson C. Chin

Engine power simulation for transonic flow-through nacelles

Conclusions Heterogeneous nucleation is seen to begin at partial pressures of about 10 times the wall partial pressure. Homogeneous nucleation is seen to begin at partial pressures on the order of 10 times the wall partial pressure. Initial gas temperature has only a weak effect on the onset of snow as long as the wall temperatures are substantially below this inlet temperature. Snow may be avoided by. staying below the onset curves, lowering the inlet partial pressures of the water vapor, or increasing the wall temperatures. These calculations and curves apply to any uniform temperature water cold trap. These curves can even be used to predict the humidity required for the thin natural convection boundary layer flowing down the outside of a glass of ice water to be visible due to small droplet formation (fog) in this boundary layer.

Effect of dissipation and dispersion on slowly varying wavetrains

A kinematic model for disturbance wave motions slowly modulated in space and time is developed, which describes the effects of amplitude and frequency dispersion, modal dependence, flow inhomogeneities, dissipation, and high-order wave dispersion and diffusion. The results, based on general variational principles, apply to waves in arbitrary continuous media, but are discussed primarily in the fluid-dynamic context. The more complete theory presented here reduces to several known wave models in various limits and reconciles, within the framework of a broader unifying approach, a number of differences found among existing, although more specialized, formulations. These latter models are surveyed and, in particular, we emphasize how their domains of applicability may be limited by different kinds of dominant high-order dispersive or diffusive wave mechanisms. **;

Supersonic Wave Drag for Nonplanar Singularity Distributions

Introduction T paper derives a planar counterpart to von Karmans * well-known drag formula for axisymrnetric lineal source distributions but which includes the effect of doublets. It provides an explicit evaluation of the authors formal results (for general nonplanar singularity distributions) in the planar case. Current wave drag prediction methods in supersonic aerodynamics are based on the classical analyses of von K&rman and Hayes. An exact result due to Hayes applies only to the source problem and is summarized easily. For sources of density/(Q), where Q is the source coordinate, define an equivalent density/such that

Stability of Inviscid Shear Flow over Flexible Membranes

A minimum of four terms in the solution form was necessary to predict accurately the divergence velocities as a function of the length-to-radius ratio ?/a. Divergence velocities varied widely before approaching a limit as the number of terms in the assumed solution was increased. This was particularly true for ?/a> 10 and for higher-order modes (/?>3). A study of the effect of the thickness-to-radius ratio on critical divergence velocities showed that higher-order circumferential modes are associated with shorter and/or thinner shells. The lowest critical divergence velocity, for long shells (£/tf>40), is associated with the beam-type mode (n=\). Critical divergence velocity as a function of lamina orientation, for a given circumferential mode, is shown in Fig. 3 for £/#=10 and hi a = 0.01. The present analysis was repeated for the isotropic case and compared with the results given in Ref. 7. Good agreement was noted. Conclusions Natural frequencies of fiber-reinfor ced shells decrease with increasing fluid velocity, as do isotropic shells, until static divergence occurs. Natural frequency is a function of the length-to-radius ratio, the radius-to-thickness ratio, the circumferential mode number, and the orientation of the lamina. Beam-type theory was found to be suitable for long shells, but in any other case shell theory must be used. Qualitatively, the behavior of anisotropic shells appears to be much the same as that of isotropic shells. However, it is reasonable to expect that certain lamina stacking sequences will result in considerably different numerical values of divergence velocity as a function of lamina orientation.

Some singular aspects of three-dimensional transonic flow

a series of elastic concentration- factor values from 1.05 to 10 and for values of the Ramberg- Osgood .exponent from 5 to 200. The data for Figs. 1 and 2 were obtained from the tabular output of the program. Conclusions 1) Design relations have been developed for plastic discontinuity stresses using Neubers equation and the Ramberg-Osgood analytic approximation of stress-strain properties. 2) The limiting value of the discontinuity stress in the developed relations is the Ramberg-Osgood end-value. The latter is the maximum value of stress for which the Ramberg- Osgood equation provides a satisfactory fit of experimental stress-strain data. The end-point of the equation has been identified using a non-dimensional strain parameter. The end- stress is obtained by computation. 3) The relative reduction of the stress concentration factor associated with plastic discontinuity stresses is independent of the elastic concentration factor. The reduction is mainly dependent on the maximum discontinuity stress and the Ramberg-Osgood exponent value. 4) Limiting values of the stress-concentration reduction factor occur when the discontinuity stress is at the Ramberg- Osgood end-value and the reference stress is elastic: a) If the end-value is the secant-yield stress, the plastic concentration factor is 83.7% of the elastic concentration factor for all values of the Ramberg-Osgood exponent and for all elastic values of the reference stress; or b) If the end value is greater than the secant-yield stress, the reduction value is a function of both the exponent and the end-stress. However, for given values of the two parameters, the limiting value of the reduction factor is constant for all elastic values of the reference stress.

Similar solutions for unsteady transonic flow

the entrainment into thermals occurs primarily near the rear stagnation point. The path from the rim to the rear stagnation point of a multiburst thermal is long, and it seems likely that the entrainment would be less efficient, i.e., that a. would be smaller. It is hoped that the entrainment constant can be determined experimentally. It should not be difficult to do experiments similar to those that have been done before, but with many small sources rather than one source. The consequences of the change in cloud geometry have been worked out using the closed-form solution of Ref. 1; but the same change in geometry could be made in a more general analysis and the same general conclusions would be expected.