SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
A numerical model of turbulent boundary layer flows over compliant walls was investigated. The model is based on Burton's observation that outer flow structures in turbulent boundary layers produce large scale pressure fluctuations near the wall. The results of calculations indicate that certain small wavelength wall motions can have a significant effect upon the stability of turbulent boundary layers.
The traditional procedure for estimating the performance of slotted walls for airfoil wind tunnels is reviewed, and a modification which improves the accuracy of this procedure is described. Unlike the traditional procedure, the modified procedure indicates that the design of airfoil wind-tunnel walls which induce minimal blockage and streamline-curvature effects is feasible. The design and testing of such a slotted wall is described. It is shown experimentally that the presence of a model can affect the plenum pressure and thus make the use of the plenum pressure as a calibration reference questionable. Finally, an ONERA experiment which shows the effect of the sidewall boundary layer on the measured model normal force is discussed.
In three recent papers, some results of an experimental investigation of a freely expanding coflowing jet as well as a three dimensional wall jet have been presented. A flat plate as well as a curved wall surface intended to model a wing-flap combination in a high lift V/STOL configuration have been investigated. In these papers, the ratio of the jet exit plane velocity to the free stream velocity, was 5.1. This paper explores the effects of increasing the velocity ratio. The quantities measured include the width of the mixing region, the mean velocity field, turbulent intensities and time scales. In addition, wall and static pressure-velocity correlations and coherences are presented. The velocity measurements are made using a laser Doppler velocimeter with a phase-locked loop processor. The fluctuating pressures are monitored using condenser-type microphones.
A method for assessing the wall interference in transonic two dimensional wind tunnel test was developed and implemented in a computer program. The method involves three successive solutions of the transonic small disturbance potential equation to define the wind tunnel flow, the perturbation attriburable to the model, and the equivalent free air flow around the model. Input includes pressure distributions on the model and along the top and bottom tunnel walls which are used as boundary conditions for the wind tunnel flow. The wall induced perturbation fields is determined as the difference between the perturbation in the tunnel flow solution and the perturbation attributable to the model. The methodology used in the program is described and detailed descriptions of the computer program input and output are presented. Input and output for a sample case are given.
This design eliminates the uncertainties in data from conventional transonic test sections. Sidewalls are rigid, and the flexible floor and ceiling are positioned by motorized jacks controlled by on-line computer to minimize run times. The tunnel-computer combination is self-streamlining without reference to the model. Data is taken from the model only when the walls are good streamlines, and is corrected for the small, known but inevitable residual interferences. Two-dimensional validation testing in the Mach range up to about 0.85 where the walls are just supercritical shows good agreement with reference data using a height:chord ratio of 1.5. Techniques are under development to extend Mach number above 1. This work has demonstrated the feasibility of almost eliminating wall interferences, improving flow quality, and reducing power requirements or increasing Reynolds number. Extensions to three-dimensional testing are outlined.
Measurements were made in a two dimensional incompressible wall jet submerged under a thick upstream boundary layer with a zero pressure gradient and an adverse pressure gradient. The measurements included mean velocity and Reynolds stresses profiles, skin friction, and turbulence spectra. The measurements were confined to practical ratios (less than 2) of the jet velocity to the free stream velocity. The wall jet used in the experiments had an asymmetric velocity profile with a relatively higher concentration of momentum away from the wall. An asymmetric jet velocity profile has distinct advantages over a uniform jet velocity profile, especially in the control of separation. Predictions were made using Irwin's (1974) method for blown boundary layers. The predictions clearly show the difference in flow development between an asymmetric jet velocity profile and a uniform jet velocity profile.
Natural instability related to negative wall deformation was studied. It was shown that natural instabilities which propagate in a laminar boundary layer of a flat plate are in agreement with stability theory. It is found that if a wall has a deformation, a second frequency does exist, which is not predicted and is twice the first frequency. This second instable frequency only appears if there is a negative velocity gradient. The phenomenon is located very closely to the wall and drops off rapidly when moved away from it.
The variables affecting the design of a composite turbine shroud, consisting of a metal base, an interlayer of porous metal, and an outer layer of yttria-stabilized zirconia, are analyzed. Results show that significant reductions in the cooling-air to gas-flow ratio are indicated for the composite shrouds compared to an all-metal shroud that was only impingement air cooled. The good insulating properties of the ceramic reduced the temperatures of the porous metal and support wall significantly. For a given porous metal density and coolant- to gas-flow ratio, decreasing the thickness of the porous metal and increasing ceramic thickness resulted in lower support wall temperatures. To maintain given allowable inter-layer temperatures and coolantto gas-flow ratios, porous-metal density or thermal conductivity must increase as the ratio of the thickness of the ceramic-to-porous metal decreases. It is concluded that a 1.78 mm thickness of porous material with a density of 0.2 and a 1.78 mm thickness of ceramic appears to be a good composite wall configuration for the assumed conditions
The acoustic field of a rectangular wall jet was surveyed for Reynolds numbers between 2000 and 9200. The study includes measurements of amplitude directivity, acoustic power spectra, and flow visualization. The effects of jet speed, wall length and trailing edge angle were examined. The flow visualization survey suggests an important role of discrete vortex convection in wall tone generation.
Three-dimensional flow field measurements were made near a transonic slotted wall. Field velocity vectors and static pressure distributions have been obtained. The boundary layer displacement thickness was found to vary in the transverse plane with its maximum at the slot center line but decreased with increasing suction rate through the slot. The boundary layer characteristics were sensitive to the mass transfer through the slot. The projection on the flow field velocity vectors on the transverse plane reveals a vortex-like flow formation. The center of this secondary flow was located nearly at the edge of the wall shear layer and decreased in strength with applied suction. The secondary vortex motion may be attributed to the mean flow skewing, inhomogeneous transverse plane boundary layer and the wall turbulence anisotropy.
A procedure for the evaluation of wall interference corrections for three-dimensional models is presented. The Mach number and angle-of-attack corrections require the numerical solution of the potential equation about a simplified representation of the experimental model. Pressure measurements are required near the wind tunnel walls. The correction procedure also requires knowledge of the free-stream Mach number, the model angle of attack, and the lift force experienced by the model. The procedure provides an estimate of the accuracy of the correction. For slender configurations at Mach numbers close to one, the Equivalence Rule formulation is adopted to calculate the wall interference effects. Preliminary results are presented for both general and slender-body configurations.
Walls of historic buildings charted by neutron radiography. Neutron source and Gamma-Ray Detector alined with each other yield map of composition of wall. Points spaced for minimal overlap based on mean free path of gamma rays emitted from wall materials. Map indicates nature and extent of changes in building materials so proper treatment is applied.
Progress in implementing and refining two near-wall turbulence models in which the near-wall region is divided into either two or three zones is outlined. These models were successfully applied to the computation of recirculating flows. The research was further extended to obtaining experimental results of two different recirculating flow conditions in order to check the validity of the present models. Two different experimental apparatuses were set up: axisymmetric turbulent impinging jets on a flat plate, and turbulent flows in a circular pipe with a abrupt pipe expansion. It is shown that generally better results are obtained by using the present near-wall models, and among the models the three-zone model is superior to the two-zone model.
Aerodynamic tests in wind tunnels are jeopardized by the lateral limitations of the throat. This influence expands with increasing size of the model in proportion to the cross-section of the throat. Wall interference of this type can be avoided by giving the wall the form of a stream surface that would be identical to the one observed during free flight. To solve this problem, flexible walls that can adapt to every contour of surface flow are needed.
In wind tunnel tests, the problems arise of determining the wall pressure distribution, calculating the wall contour, and controlling adjustment of the walls. This report shows how these problems have been solved for the high speed wind tunnel of the Technical University of Berlin.
Conventional methods for the calculation of wall interference corrections are based on boundary conditions which require a knowledge of ventilated wall porosity parameters, and which are unsuitable for deformed walls. The method described uses a simple exponential decay of pressure distribution beyond the most upstream and downstream limits in order to evaluate Mach number and incidence corrections given by the method proposed by Papelier et al. (1978). It is found that, while the upstream contribution to incidence correction is significant, the upstream and downstream contributions to Mach number correction are negligible.
The effects of wind-tunnel walls on the flow over a swept wing were greatly reduced by wall contouring. Significant reductions in spanwise pressure gradients were achieved by shaping all of the walls to conform to the streamlines over the model in free air. Surface pressure and oil-flow data were used to evaluate the effects of Mach and Reynolds numbers on the design. Comparisons of these data with inviscid calculations indicate that free-air flow is established at a Mach number of 0.74 and at Reynolds numbers above 4.7 million.
To make computer codes for two-dimensional compressible flows more robust and economical, wall functions for these flows, under adiabatic conditions, have been developed and tested. These wall functions have been applied to three two-equation models of turbulence. The tests consist of comparisons of calculated and experimental results for transonic and supersonic flow over a flat plate and for two-dimensional and axisymmetrical transonic shock-wave/boundary-layer interaction flows with and without separation. The calculations are performed with an implicit algorithm that solves the Reynolds-averaged Navier-Stokes equations. It is shown that results obtained agree very well with the data for the complex compressible flows tested, provided criteria for use of the wall functions are followed. The expected savings in cost of the computations and improved robustness of the code were achieved.