Fethi Aloui

Bubbly flow in an axisymmetric sudden expansion: Pressure drop, void fraction, wall shear stress, bubble velocities and sizes

Abstract In this work, a theoretical and experimental study of a bubbly flow in an axisymmetric sudden expansion is presented. A global model for the prediction of the singular pressure drop is derived and an experimental study relative to the dynamic parameters of the bubbly flow downstream of the singularity is given. The singular pressure drop has been obtained as a global formulation that integrates velocity form profiles, turbulence for every phase, and wall shear stress. In the experimental study, data concern pressure drop, wall shear stress, local and global void fraction, average and fluctuating bubble velocities and bubble sizes. These data constitute a quasi-complete experimental data bank with respect to Bel Fdhila and Simonins recent works on the liquid phase of bubbly flow.

Analysis of Inverse Method Applied on Sandwich Probes

Parallel plate disks (PPD) are often used for analyzing the effect of small amplitude oscillations with different frequencies. These devices allow the imposing of a well-known flow kinematics. Mass transfer problems and, particularly, convection-diffusion problems relating wall shear rate to mass transfer can thus be studied. Mass transfer signals can be determined from a sandwich electrodiffusion (ED) sensor frequency response. The experimental database constructed was used to check the inverse method. Indeed, the inverse method (Rehimi et al., 2006, “Inverse Method for Electrochemical Diagnostics of Flows,” Int. J. Heat Mass Transfer, 49, pp. 1242–1254) applied on a sandwich ED sensor was analyzed by comparing its instantaneous numerical wall shear rates to the known local and instantaneous experimental wall shear rate. Oscillatory flows amplitudes, frequencies effects, and flow direction effect have been studied in order to test the robustness of the inverse method to such effects. The little difference between experimental and numerical results is probably caused by the sensitivity of the sandwich sensor to such flow directions or to the neglecting of the insulating gap effect on the inverse method.

On the Stability of Taylor - Couette Flow With Axial Flow

An experimental investigation of Taylor-Couette flows with axial flow is presented. Two techniques are used: Visualization using the Kalliroscope and Electro-diffusion method using electrochemical probes. The fluid is confined between concentric cylinders. It is constituted by an electrochemical solution seeding with 2% of a rheoscopic liquid AQ-1000 (Kalliroscope Corp., U.S.A.). The rheoscopic liquid contains small particles reflecting light in dependence on their orientation imposed by the flow direction. The reflected light intensity of Kalliroscope flakes allows a qualitative study of the flow. While the polarography technique allows the measurement of diffusion limit current intensities delivered by the electrochemical probes. The frequency responses of the probe to the flow allow the determination of the instantaneous and local mass transfer and the instantaneous wall shear rate.Two protocols were adopted to study the effect of an axial flow superposed to Couette-Taylor flows and the history flow effect. The first one consists to impose an azimuthal flow to the inner cylinder. When the regime was established, we superposed the axial flow. This protocol was named “the direct protocol”. While the second protocol consists to impose firstly the axial flow on the gap of the system then the azimuthal flow. We named it “the inverse protocol”. We demonstrated that the Couette-Taylor flow with axial flow is strongly dependent on the flow history (the protocol). For the same Taylor number and axial Reynolds number, the resulting flow is completely different. An axial flow superposed to Couette-Taylor flow can delay the instabilities apparition; generate the displacement of the Taylor vortices in the same direction as the axial flow or in the opposite direction; and modify the instability character of the flow by developing helical vortices or wavy helical vortices.Copyright

Inverse Method Used for the Determination of the Wall Shear Stress in a Sliding Rheometer Using Sandwich Probes

The inverse method, based on a numerical sequential estimation, has been applied for the determination of the wall shear stress of a liquid single phase flow in a sliding rheometer using multi-segment probe. This method requires the inversion of the convection diffusion equation in order to apply it to instantaneous mass transfer measurements. Polarography technique, known as the limiting diffusion current method, has been used. This requires the use of Electro-Diffusion ED probe which allows the determination of the local mass transfer rate for known flow kinematics. In addition, two-segment platinum probe was mounted flush to the inert surface of the upper disk of the sliding rheometer. Hydrodynamic oscillations have been imposed to the torsional flow (type sinusoidal), in order to study the frequency response of the sandwich probe for a fixed polarization voltage. Possible error sources which are likely to affect the interpretation of the results e.g. the directional angle effect, the inertial effect, the diffusion effect and the frequencies of oscillations effect have been studied in order to test the robustness of the inverse method within the presence of such impacts. Furthermore, to demonstrate the possible effect of non-negligible inertia and diffusion, we refer to ED results for both modified Reynolds number defined by [1] and Peclet number ranges as well as for different directional angles. An algorithm has been developed for the numerically filtering of the mass transfer signals, and therefore the wall shear stress signals. It permits to eliminate any possible noise effect due to the imposed vibrations to the torsional flow. The analysis shown that the inverse method is in a good agreement with the ED experimental results for the different cases of study, i.e. for different dimensionless Reynolds numbers, for high and low oscillation frequencies, as well as for different directional angles. The little difference is probably caused by the sensitivity of the double probe to such directional angles or to the neglecting of the insulating gap effect on the inverse method solution as a first step of the study of the inverse method for double probes signals.Copyright

Numerical Investigations of Passive Scalar Transport in Turbulent Taylor-Couette Flows: Code Validation

The highly turbulent flow occurring inside (electro)chemical reactors requires accurate simulation of scalar mixing if CFD methods are to be used with confidence in design. This has motivated the present paper, which describes the implementation of a passive scalar transport equation into a hybrid spectral/finite-element code. For this purpose, direct numerical simulations (DNS) and Large Eddy Simulation (LES) have been performed to study the effects of the gravitational and the centrifugal potentials on the stability of incompressible Taylor-Couette flow. The flow is confined between two concentric cylinders and only the inner cylinder is allowed to rotate while the outer one is at rest. The Navier-Stokes equations and the uncoupled convection–diffusion–reaction (CDR) equation are solved using a code named SFELES which consists on spectral development in one direction combined with a finite element discretisation in the two remaining directions. The performance of the LES code is validated against published DNS data for a channel flow for the velocity and scalar statistics with good agreement between the current LES predictions and DNS data.Copyright