Planar Doppler Velocimetry Technique Research Articles

Development of Megahertz-Rate Planar Doppler Velocimetry for High Speed Flows

A pulse burst laser and either one or two high-speed charge-coupled-device cameras were used to perform onecomponent time-resolved planar-Doppler-velocimetry (PDV) measurements in a rectangular Mach 2.0 jet. The measurements were carried out on a streamwise plane passing through the jet centerline and covering approximately 6‐12 jet heights downstream of the jet exit. The pulse burst laser operated at 0.532-µ mw avelength and produced 28 pulses at 250 kHz with approximately 9 mJ/pulse energy. Velocity image sequences consisting of 28 frames showed dynamics of the velocity field over a time span of 108 µs (approximately 4.5 convective timescales). A typical sequence of images is presented, which demonstrates the process of entrainment of low-speed fluid into the high-speed region of the jet. Mean and standard deviation statistics of the velocity calculations produced expected trends and showed good agreement between the single- and two-camera experiments. An error analysis revealed speckle as the predominant source of noise, as in a conventional PDV technique. At a transmission ratio of 0.5, the estimated total error is 13 m/s for the single-camera experiment and 15 m/s for the two-camera experiment. I. Introduction P LANAR Doppler velocimetry is a powerful optical diagnostic technique that can be used to measure all three components of instantaneous velocity over a two-dimensional plane within a flowfield with high spatial resolution. This is accomplished by using an atomic or molecular vapor filter to measure the frequency shift of light as it is scattered by particles contained in the flowfield. The �

Three-component phase-averaged velocitymeasurements of an optically perturbed supersonicjet using multi-component planar Doppler velocimetry

In the planar Doppler velocimetry (PDV) technique, a molecular/atomicfilter is employed to convert frequency shifts in scattered light to intensityvariations. The potential for instantaneous three-component velocitymeasurements over an entire laser sheet makes PDV attractive for use in largewind tunnels. The development and integration of hardware and software fordeployment of a multi-component PDV system is reported. Hardware issuesaddressed include observation of the long-term stability of a starved iodinecell. In addition, the accuracy of the dot-locating scheme, essential in the PDVdata-reduction process, is investigated using numerically generated images.Finally, operation of a two-component system is demonstrated with velocitymeasurements in a supersonic jet with a large-scale perturbation. Here, energyfrom a Nd:YAG laser is focused into the shear layer near the lip of the nozzleto create a small thermal spot. The PDV technique is then employed to study theevolution of the large-scale disturbance. Two orientations of the two-componentvelocity measurement system are used to produce phase-averaged three-componentmean velocity measurements of the perturbed jet 170 and 220 µs after theintroduction of the disturbance.

Two-Color Planar Doppler Velocimetry

Anovel two-colorapproachto planarDopplervelocimetry (PDV)isdemonstratedin a supersonice ow. Thetechnique was implemented using a frequency-doubled Nd:YAG laser and a Nd:YAG-pumped dye laser (¸ =618 nm) for illumination and a color charge-coupled device (CCD) camera for detection. In testing two-color PDV, we obtain good agreement between experimental and theoretical velocities (difference of » 11%) in a compressible freejet e ow. The two-color approach enjoys an inherent advantage over single-color systems in that both the e ltered and reference e ow images are captured with a single camera. The need to split the scattering (and direct to two cameras ) is eliminated, and the e elds of view of the e ltered and reference images are naturally aligned. Thus, the dife culty of experimental setup, which is an impediment to the wide application of PDV, is signie cantly reduced. Furthermore, the needed equipment, at least for one velocity component, was essentially that employed for particle-imaging velocimetry (PIV). However, as a tradeoff one must characterize the irradiance distributions of the two laser sheets, as well as carefully overlap the two sheets within the probe region, and record the beam energies for thetwo lasers on a shot-by-shot basis. Furthermore, asseed particlesizeincreasesbeyond theRayleigh limit(where dRayleigh <100 nm for visible wavelengths ) the ratio of red and green scattering signals will depend on particle size. Nonetheless, the ease with which the two-color PDV technique can be applied, and its complementary nature with equipment needed for PIV, makes it attractive. The strengths and weaknesses of our particular approach are discussed, namely CCD color bleed effects, as are potential alternatives.

Two-Component Planar Doppler Velocimetry in High Speed Flows

A planar Doppler velocimetry technique was developed and used to measure two components of instantaneous velocity in an ideally expanded, Mach 2, freejet. The technique utilizes a molecular e lter as a frequency discriminator and uses two cameras to resolve two components of velocity on a plane that is illuminated by a laser sheet. Careful formation of the laser sheet, proper calibration of the splitter/recombiner imaging system, and precise image registration were found to be critical steps in the technique. The velocity measurements obtained with this techniquewerecompared with referencelaser Dopplervelocimetry measurementstaken in thesame e owe eld. The mean velocity results were in very good agreement with the reference measurements, and the turbulence results captured the correct trend but were higher than thereference results. A detailed error analysis describes the error sources inherent in the planar Doppler velocimetry technique. For the current two-component planar Doppler velocimetry system, theestimated uncertainty forthe xandy componentsofvelocity were §5and §3:5%ofthejet centerline velocity, respectively. The errors in the x component of velocity were within the estimated uncertainty. The errors in the y component of velocity exceeded the estimated uncertainty. Possible causes for this difference are discussed.