Session DP - Experimental Techniques II.
ORAL session, Sunday afternoon, November 18
Manchester 2, Marriott Hotel and Marina San Diego

[DP.001] Comparison of Defocusing DPIV with Stereo DPIV *

DPIV and Stereo DPIV velocity measurement techniques have been used and therefore established for a number of years. While, PIV has provided two-component velocity measurements within a two-dimensional sheet, stereo DPIV has upgraded the performance of DPIV by providing three-component velocity measurements within a two-dimensional sheet. The advent of the newly developed DDPIV (Defocusing Digital Particle Image Velocimetry) velocity measurement technique provides three-component velocity measurements within a volume, and therefore allows a unique opportunity to study three-dimensional time-evolving flows. Since the DDPIV method is new, no comparisons have been made to evaluate its performance with respect to previously established techniques, such as stereo DPIV. In this regard, we present simultaneous data taken by both stereo DPIV as well as DDPIV. Through their comparison, we evaluate quantities such as mean values, and standard deviations, in order to determine the performance of the DDPIV technique.

*Supported by ONR contract No. N00014-97-1-0303

[DP.002] A Novel Method for Instantaneous, Quantitative Measurement of Molecular Mixing in Gaseous Flows

Various techniques have been developed in the past for obtaining a resolution-free estimate of the extent of molecular mixing in non-reacting gas flows. These techniques often use two tracers, one of which has a large fluorescence quenching rate in the presense of a quencher, and the other does not. Other two-tracer approaches based on sensitized phosphorescence have also been developed. In these methods, two detectors and one or two laser sources are needed. We describe a novel method where a single tracer, laser and dual-frame detector are used to obtain quantitative measurements of the extent of molecular mixing. This method takes advantage of the effective oxygen quenching of phosphorescence of tracers such as acetone and biacetyl that are often used in laser induced fluorscence studies. In this work the details of the technique are described along with its application in an excited nitrogen jet discharging into air. The instantaneous planar distributions of molecularly mixed fluid and mixing efficiency are presented.

[DP.003] Particle Proximity Sensors for Suspension Flows

A nonintrusive technique to indicate when noncolloidal particles suspended in a liquid are within a few microns of a solid surface---and therefore can interact with the surface---would be valuable in studying particle deposition and coating processes. We are developing such ``particle proximity sensors'' based upon calcium ion-specific fluorescent indicators, chemical compounds whose fluorescence is triggered by the presence of Ca^++. By coating a solid surface with zeolites that absorb Ca^++, particles dyed with these indicators suspended in a calcium ion-containing solution will stop fluorescing when they enter the Ca^++-depleted mass boundary layer next to this surface. Since these fluorescent indicators are developed for aqueous chemistries, an aqueous model suspension system was developed for non-colloidal PMMA particles (diameters 10--200 \mum) suspended in a liquid mixture of water, glycerine, and ammonium thiocyanate with a viscosity about five times that of water. This refractive-index matched, neutrally-buoyant suspension was characterized for a range of temperatures, volume fractions and chemical compositions. The intensity of fluorescence emitted from Ca^++ indicator-dyed PMMA particles near a zeolite-coated glass surface was measured as a function of distance normal to the surface at various zeolite thicknesses and flow speeds to determine the diffusion characteristics of this zeolite-Ca^++ system.

[DP.004] Particle tracking velocimetry applied to an opposed-jet mixer configuration

A new version of our particle tracking velocimeter (PTV) has been developed that improves the spatial resolution by a factor of four and the time resolution by at a factor of two. Two synchronized video cameras are used with two PC computers (1 gbyte ram). The video pictures can be obtained stereoscopically and then are used for particle tracking in a three-dimensional flow field. Nearly 4000 images can be taken at the full 60 Hz rate which is equivalent to a 100 ft roll of movie film. Smaller size images can be taken at higher speeds (up to 200 Hz). Application to a highly turbulent, opposed-jet system is used to illustrate the long-time averaged nature of the flow and short-term dynamic motions that can be resolved. The inlet jet pipes Reynolds numbers are about 2000. This is considerably higher than the 200 or so level that results in fully turbulent conditions in the chamber.

[DP.005] Experimental Evaluation of a Micro Optical Shear Stress Sensor

Recent progress in the development of a micro optical shear stress sensor is described. The sensor is based on the original concept developed by Naqwi and Reynolds who measured the velocity gradient within the first hundred microns above the wall using a diverging fringe pattern originating at the wall. The use of integrated optics and micro fabrication techniques has reduced the original large optical setup to a sensor 15 mm in diameter and 15 mm long, which is easily embeddable into a model. The performance of the sensor is tested in a laminar boundary layer with a weak favorable pressure gradient. Excellent agreement is obtained between the wall velocity gradient measured by the shear stress sensor and that calculated from a boundary layer velocity survey using a miniature LDV. The expected accuracy of the present sensor configuration is discussed for both laminar and turbulent boundary layers.

[DP.006] Liquid Crystal Flow Visualization of Supersonic Flow Separation Ahead of Cylindrical Protuberances

An experimental flow visualization technique using shear-sensitive liquid crystals was used to capture the footprints of the flow separation ahead of a circular cylinder mounted on a flat surface. The experiments were conducted at the University of Kansas supersonic wind tunnel at a test Mach number of 2.0. In a single cylinder configuration, a circular cylinder with a diameter of 10.0 mm and a height of 10.0 mm was mounted perpendicular to a flat surface and the test section flow. In a second test, a similar cylinder was mounted to the flat surface with a sweep angle of 40 degrees. A thin coating of unsealed shear-sensitive liquid crystal was employed to visualize the characteristics of the junction flows in supersonic regime adjacent to the flat surface and ahead of the cylinder. The results confirmed the ability of the liquid crystal in identifying the footprints of the flow separation, shocks and vortical structures ahead of the cylindrical protuberances in supersonic flow.

[DP.007] Direct Velocity Gradient Measurements Using a Novel Laser Probe

An evaluation of a novel experimental method for the measurement of the velocity gradient and vorticity in transparent fluids was performed. The non-intrusive technique is based on the collection and direct heterodyning of the coherent light scattered from individual seeds particles in two adjacent locations. Beat frequency analyzed by conventional LDA processor is proportional to the difference of the same velocity components of the particles in those two points. Otugen et al. in 1998 [1] applied this method for direct measurements of the velocity differences in radial direction in the laminar flow between two concentric cylinders. In the present, we carry out a more comprehensive test of the technique with a wider range of flow parameters. Systematic experiments are carried out in the near field of a mixing layer for a range of velocity ratios, U1/U2 in order to critically evaluate the technique and to determine optimal optical configurations as well as particle size and loading. The time resolved velocity gradient measurements obtained using the present method are compared to those obtained independently but simultaneously using a time-of-flight method. Both data sets are also compared to LDV results obtained earlier. The results and the conclusions of this evaluation will serve for future development and applications of this technique.

[1] Otugen et al, Meas. Sci. Technol., 9, (1998), 267-274

[DP.008] A single-element, thermal, flow-velocity sensor with wide dynamic range

Thermal flow sensors with a wide dynamic range approaching 1:1000 are presently not available in spite of the large demand for such sensors in practical fluid flow measurements. During the last meeting (paper JG4, Bul.\ Am.\ Phys.\ Soc.\ 45, no.\ 9, p.\ 141, 2000), we described such a probe consisting of a minute wire heated using sinusoidal alternating current and two sensing wires acting as resistance thermometers and set parallel to, and at a small distance on either side of, the pulsed wire. Herein we detail the development of a single wire heated using square waves of electrical current. The elimination of the sensing wires reduces the complexity as well as the cost of the sensor and improves its spatial resolution. Unlike time-of-flight sensors, however, the present single-element sensor is sensitive to the physical properties and temperature of the ambient fluid. The present device is suited for measuring slowly-varying unidirectional flows over a very wide dynamic range. For a given current amplitude and frequency, the nominal output of the single sensor is the increase in wire temperature (or resistance) between times just before the leading edge of the current pulse and just after the trailing edge of the pulse. In practice, an integral of the resistance over the pulse duration is computed and averaged over several pulses. This output is a function of the wire’s time constant or thermal inertia and thus of the flow speed as well as the heat convected from the heated wire to the flow. We exploit the fact that the time constant decreases as the flow speed increases while the rate of heat transfer increases. At very low flow speeds, the response is determined almost entirely by the time constant whereas at high speeds the device acts almost like a constant-current hot-wire anemometer. At low speeds, therefore, the wire thermal inertia augments the output signal of the basic hot wire increasing its speed range and sensitivity above that of a conventional hot-wire flowmeter. We demonstrate that a wire of diameter 12.5 microns excited using a square wave of 17-msec duration and 30-Hz frequency is usable in the speed range of 0.01 to 25 m/s, yielding a dynamic range of 1:2500, almost two orders of magnitude broader range than that for traditional time-of-flight pulsed-wire anemometers.

[DP.009] A New Technique for Measurement of the Velocity Magnitude and Direction Using Single Hot-Wire Sensors

Thermal anemometry has been used extensively in characterizing turbulent flows for several decades. However, this classical measurement technique is known to suffer from the inability to discern the flow direction. This prohibits the use of hot-wires and -films in the technologically important separated/reattaching flows, where flow reversal occurs. In this work, an innovative approach is used to extract flow direction information from single hot-wire measurements. The new technique, the details of which will be revealed in the talk, may be implemented in full or limited capability modes. In the former, the time-resolved velocity (or wall-shear) signal is captured, whereas in the latter only the mean flow and higher order statistics can be measured. The difference between the two modes will be discussed along with the appropriate sensor parameter range for each. Furthermore, results from applying the technique to measure the velocity in an oscillating pipe flow experiment will be presented.

[DP.010] A probe to simultaneously measure the velocity and temperature fields of strongly heated turbulent flows.

A three-sensor probe capable of simultaneously measuring the velocity and temperature fields of a turbulent hot air flow has been constructed and tested. Two 2.5 micron diameter sensors are in an X-arrangement for velocity component measurements, and the third 1.0 micron diameter sensor is operated at very low overheat for temperature measurements. The sensors are all Platinum-10measurement volume has a diameter of about 1.5 mm. Tests were performed to find the optimal positions and spatial separations of the velocity and temperature sensors in order to minimize the thermal influence on the temperature sensor from the heated velocity sensors for low speed flows. It is found that the contamination of the cold by the hot sensors is not affected by the position as much as by the spatial separation of the sensors. An optimal separation for the fixed overheat ratio chosen has been determined. A combination of Jorgensen's law and a polynomial fit has been found to be the most appropriate to describe the simultaneous effects of velocity and temperature on the output signals. The probe has been tested in a calibration facility capable of inducing flow and temperature fields varying from 0.5-10 m/s and 20-400 deg. C with a velocity component ratio, U/V, of up to 0.36. The measured velocities and temperatures are in excellent agreement with induced values over the whole range of variation as will be shown.

[DP.011] An experimental technique for visualizing Lagrangian coherent structures in aperiodic free-surface flows

We propose a simple experimental technique for visualizing Lagrangian coherent structures (LCS) in turbulent free-surface flows. The technique employs digital photography to record the transport of passive tracers (small paper pieces) introduced manually at the free surface. Coherent eddies are detected by time-averaging the instantaneous light intensity fields over finite-size temporal windows. We demonstrate the potential of the method by applying it to visualize the flow in the vicinity of a surface-piercing rectangular block mounted at one corner of a rectangular open channel. We show that by appropriately choosing the time averaging window, the technique can visualize with clarity the finite-time geometry of LCS and elucidate the complexity of their dynamics over a range of time scales.

[DP.012] Time Resolved Measurements using the PWM-CTA (500KHz?)

The Pulse Width Modulated-CTA (Foss, et al., 1996) driven sensor recovers its equilibrium over heat in one time step. Successful operation of a 16-channel unit at 50KHz has been achieved. Representative results for vorticity measurements in a high R-theta (1.2*10**5) shear layer will be presented. Specifically, the moments of the (omega-z) distribution as well as the (v)-(omega-z) correlation function, as obtained using a Disa 55M (conventional-CTA) and the PWM-CTA, will be reported. The inherent stability of the PWM-CTA circuitry (cf that of a conventional-CTA) as well as the capacity to reduce the cycle time (T) in proportion to the velocity magnitude increase, will - in principle - allow the PWM to be an effective anemometer for M>1. A 500KHz PWM-CTA for high-speed measurements is under development. The results of a collaborative effort with W. Saric (ASU) to make boundary layer stability measurements at M=2.4, will also be reported.

J.F. Foss, D.G. Bohl, and T.J. Hicks (1996) "The Pulse Width Modulated-Constant Temperature Anemometer," Meas. Sci. and Tech., vol. 7, pp. 1388-1395.

Part D of program listing