The simultaneous measurement of two-component velocities and static pressure is tried in a plane jet by using the combined probe of X-type hot-wire and static pressure probe. First, we attempted to improve the frequency response of the static pressure probe by installing a damper inside the probe tube. Second, an optimum arrangement of the pressure probe and the X-type hot-wire is examined. The calibration results for the pressure probe show that the frequency response is improved by the damper. The distributions of cross-correlation between velocities and static pressure in the plane jet are investigated.

View full abstract

For an axisymmetric turbulent jet diffusion field with scalar(heat) transfer, the experimental investigations of passive and active controls are performed with vortex generators that have angle of attack of 30° with respect to the flow. Passive controls with 1, 2, and 6 vortex generators being projected and active controls with vortex generatos being projected in phase, in turn, and in alternate shifts are conducted. Effects of vortex generatos and their motion on turbulent jet are examined.

View full abstract

Turbulence statistics and scalar diffusion field ill all axisymmetric turbulent water jet are experimentally investigated. A commercial dye (Sc ≃ 3,800) is used as a diffusion matter. Instantenous two-component velocities and instantenous concentration are simultaneously measured using a combined probe of an X-type hot-film probe and a fiber-optic concentration sensor. The results of preparatory concentration measurements show that the radial profiles of the mean concentration agree well with Gaussian distribution. The probability density functions of fluctuating concentration are skewed negatively. The power spectra of fluctuating concentration show the -5/3 scaling region. These results are consistent with the past data of scalar field in the turbulent jet. The joint statistics of velocity and concentration will be shown at the conference.

View full abstract

Contributing the establishment of technical basis for advanced codes and standards for the management of flow accelerated corrosion (FAC)-induced pipe thinning. 3D PIV measurements on orifice downstream flow including the global flow patterns and the influence of forced swirl was performed. There is some correlation between the large-scale motion and the velocity gradient close to the wall. This result shows that the large scale motion has some effect on the pipe thinning mechanism.

View full abstract

The purpose of the study is experimentally investigating effects of the flow behavior and the flow resistance that the acceleration and the inlet shape exerted. The experiments were performed under ranges from rest to Reynolds number Re=4×10^4, with two kinds of acceleration and two kinds of inlet shape. The flow behavior and wall shear stress are measured by the three hot wire type probe and surface probe in test section. The unsteady flow maintains the laminar over the critical Reynolds number in steady state (Re=2300), and then the time of transition to turbulent delaying from the inlet to downstream. This is caused by flow rate increases without disturbance in the accelerating flow, turbulence generated at the inlet convection to downstream by the mainstream. Therefore, its transition time in inlet region is decided by time when turbulence is generated near an inlet.

View full abstract

Experiments were performed to investigate the effect of a line of roughness elements on flat-plate boundary layer transition. Each roughness element was a cylinder 2 mm in both diameter and height. Eleven elements formed a row in the spanwise direction. Wedge-shaped turbulent regions ("turbulence wedges") developed downstream from the respective roughness elements. Further downstream, two adjacent wedges merged together and then a two-dimensional turbulent boundary layer was formed. Mean and fluctuating velocities were measured by hot-wire anemometers. From the velocities an intermittency factor and turbulent dissipation, etc. were obtained. The manner how the three-dimensional turbulence wedges change into two-dimensional boundary layer were investigated. At first, the fluctuating velocity on the generator of roughness element and center between roughness elements overshot Spalart distribution. And further downstream, both distribution of fluctuating velocity undershot it.

View full abstract

Measurements were made of the response of a turbulent pipe flow to a short length (about eight pipe radius) of surface roughness. A ring-type groove has a rectangular square cross-section of 3mm which is a height of 0.075 times the pipe radius, and the pitch to height ratio is equal to 2. The measurements include the wall static pressure and mean velocity. At the measurement station, 40 pipe radius from the roughness, mean velocity field had relaxed to the universal smooth-wall structure. Present flow field disturbed by a short length roughness was weakly affected as compared with the case of a ring type manipulator.

View full abstract

A spot, generated by an artificial disturbance introduced into a laminar boundary layer, had a highly reproducibility in its velocity perturbations during the initial stage of its downstream development. Thus, we called it a "laminar spot" and investigated its downstream development in detail. The purpose of this study was to investigate the generation of an irregularity in the velocity perturbation. The results show that the irregularity occurred locally in the low-speed region away from the wall, while it spread wide in the vicinity of the wall. The irregularity of the phase shifts was predominant in both the low- and high-speed region, and that of the amplitude arose between their regions. Then, a new 'valley' occurred in the velocity perturbation, and the turbulence grew rapidly in the tail part of the spot.

View full abstract

Laminar-Turbulent transition of a boundary layer forced by a jet in the inlet region of a circular pipe was experimentally investigated. The jet was inserted radially from a small hole in the inlet region into the pipe flow periodically. Isolated turbulence patches were formed from the jets, and then they propagated downstream. Axial velocity component was measured by a hot-wire anemometer. Mean and fluctuating velocities and intermittency function were ensemble-averaged. The contour lines of the intermittency were drawn. The jet reached the central region of the pipe, then damped and disappeared as goes downstream. The jet induced a turbulent patch near the wall. The turbulent patch grew in the axial, radial and peripheral directions. The patch penetrated the boundary layer, and grew in the axial direction because of the difference of propagation velocity between leading and trailing edges.

View full abstract

The turbulent stripe, which was found by previous DNS (direct numerical simulation) studies on a turbulent channel flow at transitional Reynolds numbers, was investigated numerically using large-scale DNSs and experimentally by flow visualizations. The bulk Reynolds number at the transition from turbulent to intermittent flow was varied from 2800 to 1580, and the intermittency factor was measured. During the determination of the intermittency factor, the threshold value of the detector function based on the wall-normal turbulent intensity was verified by flow visualizations. Conditional averaging revealed that, in the quasi-laminar region, the Reynolds shear stress was almost negligible but the streamwise and spanwise velocity fluctuations were still significant.

View full abstract

We present the results obtained from direct numerical simulations of square-duct flow in a long streamwise computational domain at a range of Reynolds numbers 750≤Re_b≤1500, Re_b being based on mean bulk velocity and duct half width. Localized turbulence travelling downstream, equiliblium 'puff', is identified at the Reynolds number Re_b=850-1050. The mean secondary flow in 'puff' region is found to exhibit eight-vortex patterns observed in the secondary motion of Prandtl's second kind in turbulent square duct flow.

View full abstract

The streamwise interface between laminar and turbulent regions is experimentally investigated. Piezo ceramic actuators were used to introduce a turbulent region into the flat-plate boundary layer. A flow visualization experiment was carried out to investigate the flow structure at the laminar/turbulent interface. The spanwise correlations of velocity fluctuations inside the calmed region were obtained by applying the conditional sampling technique with an ensemble-averaged method. The results showed that the spanwise correlations are influenced by the frequency bandwidths of the filtered velocity fluctuations.

View full abstract

Spatial direct numerical simulations are used to study the development of turbulent structures and the resultant sound emission mechanism in a compressible boundary layer, for a high-speed train model conducted in a moving model facility, where the free stream velocity is 500Km/h (corresponding Mach number is 0.41). Disturbances of compressible isotropic turbulence are superimposed on the laminar profile at the inlet boundary layer in the computational box. The Reynolds number at the inlet based on the displacement thickness is 1640, which meets the measurement of the boundary layer undergoing transition. Simulation results show that the inlet disturbances affect the transition and the low-frequency velocity/pressure fluctuations. Low frequency waves are generated in the turbulent boundary layer and radiated toward an open area.

View full abstract

In order to investigate the sustaining mechanism of the large-scale streamwise vortical structures in a turbulent Couette flow, a vorticity-probe consisted of 4-X type hotwire has been developed. The probe performance was estimated by measuring the fluctuating velocity and vorticity components in turbulent Poiseuille flow and comparing them with DNS and/or LDV measurements. Contributions to Dω^^-_1/Dt from various terms of ω^^-_1 transport equation in Poiseuille flow and Couette flow are estimated using data obtained from vorticity-probe. In Poiseuille flow the total of these terms weaken the streamwise vortex, while the one in Couette flow has no such tendency. These results are consistent with the decaying or sustaining properties of the streamwise vortex in Poiseuille or Couette flow.

View full abstract

The direct numerical simulation (DNSs) with a spectral method are performed to study the effects of spanwise rotation on a turbulent Poiseuille flow at very low-Reynolds numbers. At the small rotation and Reynolds numbers, the region of the zero absolute vorticity, typically observed in the mean velocity profile of a rotating Poiseuille flow, disappears in the channel center, and the mean velocity gradient becomes opposite to that of zero absolute vorticity. When it disappears, very long low-speed streaks coiled by twisted vortical structures, dominate the flow, where low-speed fluids near a wall are excessively pumped up, and trapped into the channel center by the larger-scale twisted vortices away from the wall.

View full abstract

Detailed experiments have been made for the effect of the adverse pressure gradients on the law of the wall in turbulent boundary layers. The wall shear stress was measured by a direct measurement device and scaling law of the mean velocity was discussed with high accuracy experimental data. Consequently it was seen that ignoring the inertia term in the equation of motion was not allowed in the inner layer because of the effects of the adverse pressure gradients. By considering significant contribution of the inertia term, a new velocity scale is defined and a similarity law was yielded for the mean velocity profile, which agrees well to the experimental data.

View full abstract

Spectral direct numerical simulations (DNSs) of zero-pressure-gradient turbulent boundary layer along a flat plate are performed using a fringe method. In the DNSs a particular attention is paid to the possible effect of the fringe method on the statistics of turbulent boundary layer. Comparing the results obtained from DNSs of different parameters, we confirmed that there is a region for which similar Re-dependence of the statistics is obtained. DNS results show that Re-dependence of not only velocity statistics but also pressure statistics is quantitatively consistent with experimental results of much higher Re turbulent boundary layers.

View full abstract

Recently, much attention has been paid to the large scale motions in shear flows (boundary layer, channel flow, pipe flow). To evaluate accurately the turbulence intensities in the inner region and outer region, it is important to measure both the friction velocity and turbulence intensities. This paper discuss this point, and compare the experiments with DNS (Direct Numerical Simulation).

View full abstract

Experimental measurements of the wall pressure and streamwise velocity fluctuations in a flat plate turbulent boundary layer were made using spanwise arrays of 32 microphone and 16 hot-wire sensors. The streamwise streaks are observed in the wall pressure space-time distributions. Appearance of the streaks is inttermittent and there exist carlmed region between active durations with the streaks. Pressure-velocity correlations indicate the size of the near wall streaks and the large scall stractures.

View full abstract

The purpose of this study is to evaluate the frictional resistance with sufficient accuracy and the drag coefficient at high Reynolds numbers. We have measured the resistance of flat plate with using towing tank. Correcting the wave-making resistance, pressure resistance, and drag on turbulence simulator, it is found that the measured frictional resistance is smaller than the Karman-Schoenherr formula. But it agrees with the values suggested by Oil-film interferometry in moderate Reynolds number.

View full abstract

The purpose of this study is to clearly thermal fluctuation in thermal turbulent in cylindrical cell. We have measured the temperature around the side wall with using thermistors. But the center area is unknown, so we measured there about 9 thermistors.

View full abstract

The relation between the shape of a vehicle model and surrounding flow has not been clarified. To clarify the relation, two vehicle models were used in experiments. One is easy to entrain side flow to the rear deck. Three-dimensional boundary layer is formed on the rear deck, so the model is called 3D model. Another is difficult to do so. Two-dimensional boundary layer is formed there, so the model is called 2D model. Velocity fluctuation in the model wake was measured with the hot-wire anemometer. The models were pushed up sinusoidally. In the 3D model, the velocity fluctuation in the model wake was small. On the other hand, in the 2D model, the peak detection percentage is larger than another. 2D model has a lot of positive peaks near the rear deck and negative peaks away from it.

View full abstract

Homogeneous scalar fluctution field under constant mean scalar gradiet has been realized in nearly isotropic turbulence generated by a pair of oscillating grid. The second order scalar structure function obtained here indicates the presence of the viscous-convective subrange in high Schimidt number flow.

View full abstract

The objective of this study is to clarify turbulence modulation induced by interaction between a bubble swarm and decay process of oscillating-grid turbulence. First, homogeneous isotropic turbulence was formed by operating oscillating-grid in cylindrical acrylic pipe filled with ion-exchanged and degassed water. Second, we stopped the oscillating-grid in arbitrary time after the homogeneous isotropic turbulence was achieved. A few moments later, the controlled bubble swarm (number of bubbles: 3, equivalent bubble diameter: 3mm, bubble Reynolds number: 859, Weber number: 3.48) was launched into the decaying turbulence. The bubble formation, bubble size and bubble-launch timing are controlled arbitrarily and precisely. The bubble and liquid motion was captured by high-speed video camera and PIV measurement. The liquid-phase ambient turbulence enhanced transition of the bubble motion from zigzag motion to spiral motion. The decay-rate of the turbulence was increased via interaction between the buoyancy flow induced by the bubble swarm and liquid-phase turbulence.

View full abstract

Ray tracing numerical analysis of optical fiber probing, which is one of practical measurement techniques of bubbles and droplets dynamics, has been developed to improve the measurement accuracy of the probe. The analysis is constructed by bounded surfaces of flat planes, cylinders, and ellipsoids and constant refractive index objects in three dimensions. The rays' energy of reflection and refraction are calculated repeatedly on the surfaces considered with polarization. The rays' source is assumed as linear polarized light with random fluctuations simulating the laser source. The numerical results of the probe bursting flat water surface agreed with the experimental results. In addition, the energy intensity of the returned rays formed a sharpen peak just before the sensing edge bursts a gas-liquid surface. The numerical results well agree with the experimental results. The peak positions, however, disagrees. The surface deformation of the bubbles in the experiment is estimated to cause the difference.

View full abstract

The objective of this research is to clarify the dynamic mass transfer process of a single CO_2 bubble to the surrounding liquid and the bubble wake. We newly developed photoelectric probe to measure the point-wise CO_2 concentration quantitatively. It has the advantages of high resolution and quick response. The CO_2 concentration dissolves from a single CO_2 bubble is calculated by measurement results of the electrical conductivity of the solution via photoelectric probe. We measured the CO_2 concentration of the bubble wake quantitatively. As a result, the middle of the bubble wake has high CO_2 concentration, and its value is about 0.4-0.7 mol/l.

View full abstract

Homogeneous quasi-isotropic turbulence fields ranged R_λ=26.1〜406 were generated in a laboratory wind tunnel, covering from small-scale grid turbulence to large-scale turbulence suitable to simulate such as the atmospheric-diffusion fields. Temperature and two velocity components were simultaneously measured behind a thermal plume continuously ejected from a thin pipe of 5.4mm in diameter. The diffusion field developed from short-time diffusion to long time diffusion as it flow downstream. The dependence of turbulent diffusion coefficient K on R_λ was estimated The diffusion coefficient grew larger from K=4.9 to 602cm^2/s as R_λ increased from 26.1 to 406. The growth rate of K, changed as the turbulent Reynolds number exceeded R_λ≒200. The fact corresponds to the shift from the short-time diffusion to the long-time one.

View full abstract

To understand the flow physics of separating and reattaching flows over porous walls, PIV measurements are carried out for turbulent rib-mounted permeable channel flows. Two kinds of square cylinder ribs: impermeable (solid) and permeable (porous) ribs, are considered. It is found that in the case of the porous rib, the recirculation behind the rib shifts downstream due to the flow going through the porous rib. From the measured turbulence quantities, the flow downstream the porous rib is confirmed to be far less turbulent than that behind the solid rib. By comparison with the solid channel flow LES, the turbulent intensity and the shear stress are found to be lower than those of the LES due to the bypass flow through the porous bottom wall.

View full abstract

We have tested seven fluctuating static pressure probes with different lengths to determine the optimal dimensions for this type of pressure probes. The fundamental performances of the probes, e.g., the frequency response, the effect of angle of attack, and the effect of the probe intrusion, have been investigated. The frequency response depends on the length from the pressure holes to the diaphragm of the microphone. The pressure fluctuation increases with the angle of attack and the response to the angle is changed due to the length from the tip to the pressure hole. The effect of the probe intrusion on the mean velocity is mainly due to the stagnation effect of the connector part. The velocity deficit and the increase of the velocity fluctuation are observed due to the wake of the pressure probe when the flow has an angle of attack to the probe axis.

View full abstract

A constant temperature hot wire anemometer includes a feedback circuit that needs proper care for influence from surrounding electrical noises and its stability. We have developed a novel hot wire anemometer with periodic pulse heating and tested it for a jet flow measurement. The principle of the anemometer is that flow velocity is estimated from rate of thermal decay of a hot wire sensor so that a feedback circuit is unnecessary. Other advantages of this anemometer are less influence of the fluid temperature and small effective volume of the thermal field to determine the measured velocity. The test measurements indicate enough accuracy for a dynamic measurement of flow velocity, compared with a constant temperature hot wire anemometer.

View full abstract

The velocity profile of dilute polymer solutions and a distilled water near the wall surface in a microchannel was clarified using a particle tracking velocimetry technique combined with evanescent wave illumination. Fluorescent particles with the diameter of 20 nm and 100 nm were used as tracer particles. The test polymers are polyethylene-oxide (SE-5 and SE-30) solution at 5 ppm. The results obtained for the velocity profile of a distilled water using fluorescent particles with a diameter of 20 nm was found that the resolution of the velocity profile near the wall surface was drastically improved from the conventional result. On the other hand, the velocity profile of dilute polymer solutions decreased significantly compared with that of distilled water, and the gradient of the velocity profile decreases with the increase of the molecular weight.

View full abstract

Experiments were curried out the measurement of the friction factor of a pipe flow because it is considered that the silk thread of suspensions is a material with small environmental loading. It has been reported that the silk thread suspensions cause the drag reduction phenomena in the turbulent flow range. The size of the test silk thread is 5〜10μm in the diameter of powder, and the length is about 30μm in other type suspensions. It is obtained 5% in the maximum drag reduction ratio of the friction factor.

View full abstract

Ship is an efficient transportation system in terms of ton-mile base energy consumption. Friction between ship's hull and sea water is considered to be the major source of ship's propulsion resistance as sharing 50-80% of the total resistance. Since the ship's hull in contact with sea water is usually coated with paints, development of low friction paints would lead to a significant reduction in the propulsion resistance of vessels and eventually to substantial energy saving. Toms effect is well known as the phenomenon whereby extremely dilute solutions of high-molecular-weight polymers exhibit much lower frictional resistance than the pure solvent. Application of Toms effect to water surrounding the ship may reduce substantially the propulsion drag by addition of effective polymer to ship hull paints. In this study, drag reduction characteristics of different polymer solutions are investigated to select the most appropriate polymer for the developing low friction paint.

View full abstract

Effect of blowing polymer solution from a wall upon drag reduction rate is examined by using a discrete element model (spring-dumbbell model). The discrete element model is used to investigate the mechanism of drag reduction by addition of small amount of polymer in turbulent wall-bounded flow of Newtonian fluid by direct numerical simulation. By using the discrete-element model, the spatial concentration of the polymer can be obtained numerically. It is found that the discrete elements are advected from the wall mainly to the buffer layer, leading to drag-reduction effect by decreasing the weighted Reynolds shear stress in the same layer.

View full abstract

Experimental investigation on the turbulent drag reduction with blowing polymer solution from the permeable channel wall was performed. Measurement of the drag reduction has been carried out with changing condition of blowing polymer solution (e.g. blowing rate and concentration). As a result, obvious drag reduction was obtained with blowing polymer solution and the maximum drag reduction rate achieved about 20 %. In addition, it seems that there is close relationship between the drag reduction rate and the mass flux of polymer. Because it is the merit of wall blowing that polymer can exist in the whole near-wall region, we discussed the difference between our results (wall blowing) and the effect of slot injection. Based on these results, the effect of wall blowing polymer solution on the heterogeneous drag reduction was discussed.

View full abstract

Detailed investigation on the drag reduction of water soluble polymer was performed. The drag reducing effect of water soluble polymer solution was measured precisely by a specially designed pipe flow system (1-pass device) and the mean molecular weight in solvent was detected by GPC-MALS. We investigated for 28 kinds of water soluble polymer solution in this study. The polymer solution prepared 10 ppm weight concentration in pure water, and the Reynolds number was ranging from 8000 to 22000 in the 1-pass device. It was found that the polymer with higher molecular weight (more than 1 million) increases obviously the maximum drag reduction. However, when the polymer molecular weight is more than 3 million, the maximum drag reduction tends to a constant value. Moreover, these experimental results have good coincidence with DNS results.

View full abstract

Two-component LDV measurement of the turbulent statistics in the drag reducing flow of polymer homogeneous solution between the concentric cylinders was performed. Measurement has been conducted with changing rotating number of inner cylinder and concentration of the polymer solution. The drag reduction rate was achieved about 37 % in our experiment. As a result of LDV measurement, in the drag reducing flow of polymer solution, the mean azimuthal velocity near the wall of inner cylinder is higher than that of the pure water flow. The rms of azimuthal velocity fluctuation of polymer solution in the near-wall region is larger than that of the pure water, while the rms of radial velocity fluctuation of polymer solution is lower than that of the pure water. The Reynolds shear stress of the polymer solution near the wall of the inner cylinder is lower than that of the pure water.

View full abstract

The normal heat transfer characteristics are maintained in the impinging jet when the counter ion is excessively added to the surfactant solutions. On the other hand, the drag reduction can be attained in a pipe. In order to clarify the phenomena, the impinging jet was visualized. The surfactant used was Ethoquad O/12. The counter ion was salicylic sodium, and was added with an equal and 30 times higher molar concentrations than the surfactant. The two surfactant solutions indicated different boundary layer flows in the wall jet region. The surfactant solution with the molar ratio of 30 showed the same boundary layer thickness as water. On the other hand, the equi-molar counter ion inhibitted to induce the boundary layer flow on the wall at the low Reynolds number.

View full abstract

We studied the effect of textured hydrophobic surfaces on drag reduction in Newtonian laminar flow through a micro channel. Fine fabrication was given to the test wall surfaces so that the groove area ratio may be changed methodically, and their surfaces are coated with PTFE. Drag reduction was estimated by pressure loss measurement in a 0.5×5mm channel. Visualization experiment was carried out to reveal a mechanism of the drag reduction from the form of air-water interface standpoint. A series of experiments showed that the air-water contact area ratio and the air layer thickness influence the drag reduction, and the maximum drag reduction ratio is 15.6%.

View full abstract

We studied the air-liquid interface behavior in the microstructure of the wall surface by the numeric simulation. Level set method was used for an analysis of air-liquid interface. For hydrophobic surface, it was clarified that an air layer keeps its shape in the microstructure at the flow condition. On the other hand, the liquid penetrates in the microstructure when the surface is hydrophilic or velocity is fast even in the hydrophobic surface. In addition, from the analytical result using the proposed model which alternately applied slip and no-slip boundary condition to the wall surface, it was clarified that the drag reduction effect is dependent on not only Re number but also the channel height and the length of the slip boundary condition.

View full abstract

Enegy saving hydrauric equipment of pressure loss,it is important to measure accuturately pressure distribution. Pressure distribution in valve is very effect of inlet flow. With considering pipe fittings on flow in valve, get pressure distribution by Numerical Simulation, inspect by experiment. And assess the pipe coupfittings effect to flow and pressure distribution.

View full abstract

In automobile-exhaust systems, catalytic converters are the main components to produce substantial pressure drops, which induce engine-power loss and fuel-consumption rise. Recently, the authors have reported that a flow deflector placed inside a catalytic-converter diffuser part can reduce drastically the pressure loss in comparison with the no-deflector case, even on pulsating flow as well as steady flow. In the present study, we investigate the velocity-profile improvement by the flow deflector in pulsating flow, using a hot-wire velocimetry.

View full abstract

Velocity profiles of the turbulent boundary layer on hybrid riblets were measured with a hot-wire anemometry in a wind tunnel. Free streamwise velocity was 5.0 m/s. The drag-reducing ability of two types of hybrid riblets (Type1, Type2) was investigated. The drag reduction ratio was estimated from the development of momentum thickness of the turbulent boundary layer. The data on hybrid riblets were compared with those of smooth surface. The maximum drag reduction ratios for Type1 and Type2 were 2 % and 7 %, respectively. It was found that the maximum of streamwise turbulence intensity of hybrid riblet surface was smaller than that of smooth surface.

View full abstract

This paper describes an experimental research on the drag reduction of turbulent surface friction on the compliant wall. The experiment is carried out on the turbulent boundary layer along a hard wall and a compliant wall at Re=3.0×10^5. The compliant wall was made of silicone rubber with 5mm thickness. Its bottom plate is almost fixed end, but partly free end. The boundary layer was measured using I-type hot wire on a solid wall, compliant wall with fixed end and free end. Local skin friction coefficients on the fixed end compliant wall and on the free end compliant wall were smaller than the solid wall by approximately 4% and 2% respectively. So, characteristic frequency of compliant wall is changed free end and fixed end.

View full abstract

Direct numerical simulation (DNS) of a spatially developing turbulent boundary layer is performed with uniform blowing from the wall aiming at friction drag reduction. The amplitude of blowing is set to be 0.001 U_∞ and 0.005 U_∞. It is found that the uniform blowing has a large drag reduction effect. The friction drag is achieved due to a negative contribution of the mean wall-normal velocity.

View full abstract

Two types of wall actuation in channel flow are considered: traveling waves of wall deformation (peristalsis) and traveling waves of blowing and suction. The flow response and its mechanisms are analyzed using numcaical simulations. We show that both actuations induce a flux in the channel in absence of imposed pressure gradient and can thus be characterized as pumping. It is demonstrated that the wall deformation induces forward pumping, whereas the blowing and suction induces backward pumping.

View full abstract

The skin-friction drag reduction effect of the traveling wave-like surface heating/cooling in the channel flow is investigated by using a linear analysis and a direct numerical simulation. The buoyancy force induced by the surface heating/cooling is expected to reduce the Reynolds shear stress in the region near the wall. The linear analysis shows that the downstream traveling wave can reduce the skin-friction drag in laminar flows. Amount of drag reduction is larger at lower Prandtl numbers. For a fully development turbulent channel flow the direct numerical simulation shows that the skin-friction drag is slightly reduced below the uncontrolled level by the high wavenumber of the downstream traveling wave.

View full abstract