Engineering topics

Mehta, R. D.

Publications and source records attributed to Mehta, R. D..

At least 19 records

NASA NTRS1998

Low-Speed Flow Studies Using the Pressure Sensitive Paint Technique

Optical pressure measurements have been made on a NACA 0012 airfoil coated with Pressure Sensitive Paint (PSP) at very low flow speeds (less than 50 m/s). Angle of attack was limited to 5 deg. for most measurements. Effects of temperature gradients and mis-registration errors on PSP response have been established and minimized. By reducing measurement error caused by these effects. PSP sensitivity has been enhanced. Acceptable aerodynamic data at flow speeds down to 20 m/s have been obtained and valid pressure paint response was observed down to 10 m/s. Measurement errors (in terms of pressure and pressure coefficient) using PSP with pressure taps as a reference are provided for the range of flow speeds from 50 m/s to 10 m/s.

Brown, O. C.↗

NASA NTRS1997

PSP Testing at NASA Ames Research Center

Pressure sensitive paints (PSPs) are now used routinely for measuring surface pressures on wind tunnel models at transonic and supersonic Mach numbers. The method utilizes a surface coating containing fluorescent or phosphorescent materials, the brightness of which varies with the local air pressure on the surface. The present paper will summarize PSP activities (in progress and planned) at the NASA Ames Research Center. One of the main accomplishments at NASA Ames has been the development of a PSP measurement system that is production testing capable. This system has been integrated successfully into the large-scale wind tunnel facilities at Ames. There are several problems related to PSP testing which are unique to large-scale wind tunnel testing. The hardware is often difficult to set-up and must operate under harsh conditions (e.g. high pressures and low temperatures). The data acquisition and reduction times need to be kept to a minimum so that the overall wind tunnel productivity is not compromised. The pressure sensitive paints needs to be very robust; the paints must readily adhere to different surfaces with varying geometries and remain functional for long running times. The paint must have well understood, and preferably minimal, temperature sensitivity since fine control of the tunnel temperature is not easily achievable in the larger wind tunnels. In an effort to improve the overall accuracy of the PSP technique, we are currently evaluating some referenced pressure sensitive paints which contain a pressure- independent luminophor in addition to the one which is affected by the surface pressure. The two luminophors are chosen so that their emission wavelengths are somewhat different. Then by taking two 'wind-on' images with either two cameras (with different filters) or one camera with a rotating filter system, the need for 'wind-off' images can be eliminated. The ratio of the two wind-on images accounts for nonuniform lighting and model motion problems (physical movement of the model under aerodynamic loading) are not a concern any more. The paints and data acquisition and reduction software for this type of a PSI system are currently under development and evaluation. Additional information is contained within the original extended abstract.

Bell, J. H.↗

NASA NTRS1993

Effects of small changes in initial conditions on mixing layer three-dimensionality

The effects of relatively small changes in the initial conditions of the development of the 3D structure of a plane mixing layer originating from laminar boundary layers were experimentally studied. It is found that, while the exact shapes and positions of the streamwise vortex structures are not the same for the two initial conditions, their overall distribution, reorganization, and decay are very similar. The results likely that while some of the specific details of the streamwise vortex structure may be facility-dependent, a relatively strong structure which produces significant three-dimensionality should form in all mixing layers originating from laminar boundary layers. After some initial readjustments, the structure will appear in the form of counterrotating pairs of streamwise vortices which in the mean, grow with the mixing layer and decay in strength. The results also serve to warn that small changes in initial conditions may significantly affect the Reynolds stress distributions in the near field.

Plesniak, M. W.↗

NASA NTRS1992

Sensitivity of the structure of untripped mixing layers to small changes in initial conditions

An experimental study was conducted concerning the influence of small changes in initial conditions on the near- and far-field evolution of the three-dimensional structure of a plan mixing layer. A two-stream mixing layer with a velocity ratio of 0.6 was generated with the initial boundary layers on the splitter plate laminar and was nominally two-dimensional. The initial conditions were changed slightly by interchanging the high- and low-speed sides of the wind tunnel, while maintaining the same velocities, and hence velocity ratio. This resulted in small changes in the initial boundary layer properties, and the perturbations present in the boundary layers were interchanged between the high- and low-speed sides for the two cases. The results indicate that, even with this relatively minor change in initial conditions, the near-field regions of the two cases differ significantly. The peak Reynolds stress levels in the near-field differ by up to 100 percent, and this is attributed to a difference in the location of the initial spanwise vortex roll-up. In addition, the positions and shapes of the individual streamwise vortical structures differ for the two cases, although the overall structures differ for the two cases, although the overall qualitative description of these structures is comparable. The subsequent reorganization and decay of the streamwise vortical structures is very similar for the two cases. As a result, in the far field, both mixing layers achieve similar structure, yielding comparable growth rates, Reynolds stress, distribution, and spectral content.

Plesniak, M. W.↗

NASA NTRS1992

Structure of a swirling turbulent mixing layer

A single-stream swirling turbulent mixing layer was produced in which the angular momentum instability acts over most of the layer for most of its development length. The instability is associated with a large increase in the level of the conventional Reynolds stresses over those in an unswirled mixing layer. This paper presents measurements of the three-dimensionality of the mean flow at one axial location well downstream of the origin. The output of a fixed hot-wire probe was ensemble-averaged to produce an approximation to the mean flow field that would be seen by a rotating observer. The main goals of this study were to establish the presence or absence of Taylor-Goertler vortices found in unstable laminar boundary layers. The variations in the mean velocities and Reynolds stresses are generally consistent with a bodily 'wrinkling' of the mixing layer, without a significant variation in the layer thickness. The wrinkling is most probably a result of circumferential variations in the initial boundary layer, which in turn originate at the swirl generator.

Wood, D. H.↗

NASA NTRS1992

Three-dimensional structure of a curved wake

The effects of streamwise curvature on the 3D structure of a plane wake are studied. Preliminary results indicate that the wake is initially grossly 3D with large spanwise distortions in the mean velocity and Reynolds stress contours. Local extrema distributed in a quasi-periodic manner are seen. The spanwise variations are caused by the presence of relatively strong quadrupoles of mean streamwise vorticity. Further downstream the mean vorticity decays at a rate lower on the unstable side of the wake so that only the unstable side row of vortices persist to the far-field region. As a result, the spanwise distortions and local extrema seem to persist with a relatively slow decay. The effects of curvature are also apparent in the Reynolds stress results. In particular, the shear stress is significantly affected by curvature right from the start such that significantly higher magnitudes are observed on the unstable side. The curved wake clearly develops a very 3D asymmetric structure which shows no signs of reaching an equilibrium state.

Weygandt, J. H.↗

NASA NTRS1991

Some effects of swirl on turbulent mixing layer development

Effects of swirl on the development of an axisymmetric turbulent mixing layer have been studied. The development of the zero swirl case is found to be consistent with that of a single-stream mixing layer developing from a tripped boundary layer. In particular, the consistency of the measured stress levels shows the accuracy of the measurement system. Swirl increases all the Reynolds stress levels with the increased scaling with swirl number. The shear stresses are found to be affected significantly more than the normal stresses. The stress increases are partly explained by considering extra production terms in the transport equations that are activated by the presense of the swirl. The thickness and growth rate for the lower swirl rate are found to be comparable to the zero swirl case, while those for the higher swirl rate are increased significantly, in line with the increased Reynolds stresses.

Mehta, R. D.↗

NASA NTRS1991

Effects of initial spanwise perturbations and three-dimensionality in a plane mixing layer

An experimental study concerning the influence of small changaes in initial conditions on the near- and far-field evolution of the th ree-dimensional structure in a plane mixing layer has been conducted. A two-stream mixing layer with a velocity ratio of 0.6 was generated with the initial boundary layers laminar and nominally two-dimensional. The initial conditions were changed slightly by interchanging the high- and low-speed sides of the wind tunnel, while maintaining the same velocity ratio. The results indicate that, even with this minor change in initial conditions, the near-field regions of the two cases differ significantly, with the peak Reynolds stress levels varying by up to 100 percent. In addition, the distribution of the individual streamwise vortical structures differs in intensity and position for the two cases, although the overall qualitative description is comparable.

Plesniak, M. W.↗

NASA NTRS1991

Effect of velocity ratio on plane mixing layer development - Influence of the splitter plate wake

The effect of the velocity ratio on the approach of a plane mixing layer to self-similarity was investigated experimentally. In the experiment, plane mixing layers with velocity ratios 0.5, 0.6, 0.7, 0.8, and 0.9 were generated in a mixing-layer wind tunnel consisting of two individually driven legs, in which the two streams were allowed to merge at the sharp edge of a tappered splitter plate. The leg driven by the bigger blower was operated at a free-stream velocity in the test section of 21 m/s, while the flow speed in the other leg was varied to change the velocity ratio. For each velocity ratio, the mean flow- and turbulence measurements were carried out at eight streamwise locations. Results showed that, for velocity ratios between 0.5 and 0.7, self-similarity of the mixing layer was achieved, with the asymptotic states comparable; mixing layers with higher velocity ratios failed to achieve a self-similar state within the measurement domain.

Mehta, R. D.↗

NASA NTRS1990

Verification of three-dimensional laser Doppler velocimeter measurements

An experimental comparison method is proposed for the verification of mean flow and turbulence measurements obtained with a three-dimensional laser Doppler velocimeter system. Such measurements can include large errors caused by problems unique to three-dimensional systems. Direct comparisons of laser and hot cross-wire measurements obtained in two-dimensional flows, as is the common practice, will not bear out all the errors associated with three-dimensional laser systems. It is proposed here that the errors may be adequately quantified by making the direct comparisons in a weakly three-dimensional turbulent shear flow. The weak three-dimensional flow ensures high accuracy of the cross-wire data while still generating sufficiently strong secondary mean flow and Reynolds shear stress so that all the laser measurements may be fully verified. This type of shear flow is easily generated by introducing a weak streamwise vortex into a nominally two-dimensional turbulent shear layer.

Mehta, R. D.↗

NASA NTRS1989

Interaction of an oscillating vortex with a turbulent boundary layer

An oscillating vortex embedded within a turbulent boundary layer was generated experimentally by forcing a periodic lateral translation of a half-delta wing vortex generator. The objective of the experiment was to investigate the possibility that a natural oscillation, or meander, might be responsible for flattened vortex cores observed in previous work, which could also have contaminated previous turbulence measurements. The effect of this forced oscillation was characterized by comparison of measurements of the mean velocities and Reynolds stresses at two streamwise stations, for cases with and without forcing. The Reynolds stresses were affected significantly by the forced oscillation, mainly through contributions from the individual production terms, provided that the vortex was not too diffuse.

Westphal, R. V.↗

NASA NTRS1989

Effect of velocity ratio on plane mixing layer development

An experimental study has been conducted to investigate the effect of velocity ratio on the development of a plane mixing layer to self-similarity. Plane mixing layers with five different velocity ratios (0.5, 0.6, 0.7, 0.8 and 0.9) were generated in a newly designed mixing layer wind tunnel with both initial boundary layers tripped. For each velocity ratio, mean flow and turbulence measurements were obtained at eight streamwise locations with a single cross-wire probe. The results indicate that the splittter plate wake plays a very dominant and, in some cases, a lasting role in the development of the mixing layer. For velocity ratios between 0.5 and 0.7, self-similarity of the mixing layer was observed with the asymptotic states comparable. Mixing layers with the higher velocity ratios failed to achieve a self-similar state within the measurement domain. The development distance decreased with increasing velocity ratio up to 0.7, after which it appeared to increase. Almost all of the observed effects may be attributed to the presence and subsequent behavior of the splitter plate wake.

Mehta, R. D.↗

NASA NTRS1987

Comparison of experimental and computational techniques for plane mixing layers

In this paper, results from two experimental and two computational investigations of plane turbulent mixing layers are presented and compared. The experimental results include flow visualization data using the smoke laser technique and mean flow and turbulence measurements obtained with hot X-wires and a two-component LDV. Reasonably good agreement is found among these techniques, at least for turbulence quantities up to second order. Reynolds-averaged computations are successful at capturing the complete evolution of the mixing layer, including wake effects in the near field and approach to self-preservation in the far field. The two-dimensional vortex method shows excellent qualitative and quantitative agreement with measured data for the forced mixing layer. For the unforced layer, the results seem to indicate that a three-dimensional computation may be necessary.

Mehta, R. D.↗

NASA NTRS1986

A study of the factors affecting boundary layer two-dimensionality in wind tunnels

The effect of screens, honeycombs, and centrifugal blowers on the two-dimensionality of a boundary layer on the test section floors of low-speed blower tunnels is studied. Surveys of the spanwise variation in surface shear stress in three blower tunnels revealed that the main component responsible for altering the spanwise properties of the test section boundary layer was the last screen, thus confirming previous findings. It was further confirmed that a screen with varying open-area ratio, produced an unstable flow. However, contrary to popular belief, it was also found that for given incoming conditions and a screen free of imperfections, its open-area ratio alone was not enough to describe its performance. The effect of other geometric parameters such as the type of screen, honeycomb, and blower were investigated. In addition, the effect of the order of components in the settling chamber, and of wire Reynolds number were also studied.

Mehta, R. D.↗

NASA NTRS1986

A 3-component laser-Doppler velocimeter data acquisition and reduction system

This report describes a laser Doppler velocimeter capable of measuring all three components of velocity simultaneously in low-speed flows. All the mean velocities, Reynolds stresses, and higher-order products can then be evaluated. The approach followed is to split one of the colors used in a 2-D system, thus creating a third set of beams which is then focused in the flow from an off-axis direction. The third velocity component is computed from the known geometry of the system. In this report, the laser optical hardware and the data acquisition electronics are described in detail. In addition, full operating procedures and listings of the software (written in BASIC and assembly languages) are also included. Some typical measurements obtained with this system in a vortex/mixing layer interaction are presented and compared directly to those obtained with a cross-wire system.

Rodman, L. C.↗

NASA NTRS1985

Double-branched vortex generator

In order to assess the suitability of using a double branched vortex generator in parametric studies involving vortex interactions, an experimental study of the main vortex and secondary flows produced by a double branched vortex generator was conducted in a 20-by-40 cm indraft wind tunnel. Measurements of the cross flow velocities were made with a five hole pressure probe from which vorticity contours and vortex parameters were derived. The results showed that the optimum configuration consisted of chord extensions with the absence of a centerbody.

Cantwell, E. R.↗

NASA NTRS1985

A 3-component laser-Doppler velocimeter data acquisition and reduction system

A laser doppler velocimeter capable of measuring all three components of velocity simultaneously in low-speed flows is described. All the mean velocities, Reynolds stresses, and higher-order products can be evaluated. The approach followed is to split one of the two colors used in a 2-D system, thus creating a third set of beams which is then focused in the flow from an off-axis direction. The third velocity component is computed from the known geometry of the system. The laser optical hardware and the data acquisition electronics are described in detail. In addition, full operating procedures and listings of the software (written in BASIC and ASSEMBLY languages) are also included. Some typical measurements obtained with this system in a vortex/mixing layer interaction are presented and compared directly to those obtained with a cross-wire system.

Rodman, L. C.↗

NASA NTRS1985

An experimental study of plane mixing layer development

Mean flow and turbulence was measured in the near field of two plane mixing layers in air with a maximum velocity of 21 m/sec. The experimental rig enabled mixing layers of velocity ratios 0 and 0.46 to be generated simultaneously. Gases with both tripped and untripped initial boundary layers were studied. It was found that the two stream layer developed to the self preserving state in a distance much shorter than did the single stream layer which followed accepted criteria for the development distance. The asymptotic levels of the turbulence quantities in the two stream layer and the development of the single stream layer showed agreement with existing data. It is suggested that the two stream mixing layer should provide a better test case for the development of turbulence models and calculation methods than the single stream mixing layer.

Mehta, R. D.↗