Experiments

Costes and Couderc [1] used two different cylindrical tanks, that geometrically

similar, with flat bottom and four vertical baffles to study turbulent flow induced by a Rushton turbine in a stirred tank by laser Doppler anemometer. Mean velocity and velocity fluctuations were measured at different impeller rotational speed. The discharge flow number and the circulation flow number were also calculated. They find that, in non-dimensional form, all the profiles are approximately independent of the system size and of the rotational speed of the impeller. The three components of velocity intensity are found to be the same order in the bulk region of the tank, and therefore, turbulence may be considered as isotropic and homogeneous in the bulk region. The pumping coefficient was measured as 0.73. The value of total circulation number is 3.4 in a plane 45° from the baffle.

Wu and Petterson [2] used one dimensional laser Doppler anemometer to profile mean and rms velocity of all the three component on vertical surface of impeller swept volume with periodicity that correspond to impeller blade frequency and twice that frequency removed. Radial profiles of pumping capacity in impeller stream have been presented. The pumping capacity increases with radius and maximum is around 2.5 times the discharge pumping capacity. Turbulent length scale has been reported for impeller stream. Energy balance around the impeller region and profiles of local energy dissipation rates have been reported. 30% of input energy dissipated in impeller region, about 30% dissipated in the impeller stream, and the rest, about 40%

was dissipated in the bulk of the tank. The mean velocities, pumping capacities, and turbulence intensity in the impeller stream were approximately proportional to the impeller speed, in other words, the profiles on non-dimension are independent of rotational speed of impeller.

Ranade and Joshi [3] want to establish a complete understanding of the flow generated by a disc turbine in a fully baffled cylindrical vessel, as a reference case.

They measured the mean flow and turbulence intensities for bulk region using a laser Doppler anemometer. An energy balance around the impeller has been performed and the effect of vessel diameter on flow characteristics has also been studied. The comparison with published data of previous papers was also presented. The hydraulic efficiency of the disc turbine was found to be about 65% in the 500 mm i.d. vessel and 61% in the 300 mm i.d. vessel. And around 35%-40% of the input energy is dissipated in the impeller region. Various vertical plane located at different angles from the baffles has been measured. These profiles demonstrate the existence of vortices behind the baffle. The value of primary pumping number obtained in the present work is 0.74.

Yianneskis et al. [4] presented flow data for the trailing vortices and flow between two blades of impeller. Data presented in various r-z and r-θ planes near the impeller were used to characterize the trailing vortices behind the blades. The measurements quantify the mean flow and the kinetic energy of turbulence in the vortices and show that ensemble-average measurements of the flow over 360 of revolution can overestimate the turbulence fluctuations by up to 400%. The vortices structure is present in most cycles shown and each recording follows a similar pattern: the vortices appear to be generated constantly behind the blades, but have a short duration and are erratically broken-up occasionally. They also report the power number is 4.8 for the present germetry.

°

Mishra and Joshi [5] compared the flow generated by three different diameters of disc turbine with each other, with the flow generated by a standard disc turbine located on difference clearances and with flow generated by other four designs of radial flow impellers, namely modified disc turbine, curved blade turbine, straight blade turbine and Brumagin impeller. The effect of impeller geometry on the turbulent flow was

measured using a laser Doppler anemometer. The comparison is presented in terms of mean velocities, turbulent kinetic energy, flow number, hydraulic efficiency and pumping effectiveness. The flow number of the disc turbine was found to be maximal at an impeller clearance of H/3, and for an impeller diameter of T/3. The hydraulic efficiency of the disc turbine is maximal for an impeller clearance of H/2, and for an impeller diameter of T/3. The straight blade turbine produces an unsteady flow with intermittent changes in direction of flow.

Mishra and Joshi [6] investigated three dimensional turbulent flow generated by multiple impellers using two kind of impellers that are a standard Rushton disc turbine( DT) and a pitched blade downflow turbine(PTD) and two impeller combinations: (1)DT-DT and (2)DT-PTD. Three dimensional mean and rms velocities generated by a double system have been reported over a wide range of impeller clearance and impeller location from the vessel bottom. The flow number and power number have also been measured for all the configurations. Energy balance had been established and the values of hydraulic efficiency and pumping effectiveness have been reported. The hydraulic efficiency and the pumping effectiveness for the DT-PTD impeller was found to be higher than the DT-DT impeller or a single disc turbine. They supposed that it is likely that PTD-PTD combination is superior even to the DT-PTD combination for a flow controlled operation.

Lee and Yianneslis [7] investigated the structure of the flow in a vessel stirred by a 100mm diameter Rushton turbine by laser Doppler anemometry. The flow field and turbulence structure produced in the vicinity of a Rushton impeller has been described in detail. The time and length scales of turbulence were determined and used to estimate the dissipation rate of turbulence energy. The lever of turbulence energy and dissipation are high near the turbine and decrease rapidly with increasing distance

from the turbine blade. The degree of anisotropy of the turbulence in the vicinity of the impeller blades decreases with distance from the blade tip, as indicated by comparison of the three normal stress components.

Ranade et al. [8] studied the trailing vortices behind the blades of a standard Rushton turbine using PIV technique. Angle resolved and angle averaged flow fields near the impeller blades were measured and the structure of trailing vortices was studied in detail. They also do the computational simulation of flow field agitated by Rushton turbine with standard k-ε and RNG k-ε turbulent models. Predicted results were compared with the angle resolved PIV measurements. The trailing vortices were found to retain their coherent structure up to about 30° behind the leading blade.

Predicted flow field show the presence of trailing vortices but under predicted the strength and the kinetic energy.

2. Dual impeller

Rutherford et al. [9] investigated the flows generated in vessels stirred by two Rushton turbines using flow visualization, power consumption, mixing time, and

ensemble-averaged and angle-resolved LDA measurement techniques. They use two vessels of diameter T=100mm and T=294mm with impeller diameter D=T/3. The flow depended strongly on the clearance of the lower impeller above the base of the vessel(C1), the separation between the impellers(C2), and the submergence(C3) of the upper impeller below the top of the liquid height. When these distances were varied, three stable and four unstable flow patterns were observed. The three kind of stable flow patterns are the parallel flow, merging flow, and diverging flow with the clearance (C1,C2,C3) equal to (0.25T, 0.5T, 0.25T), (T/3, T/3, T/3), and (0.15T, 0.5T, 0.35T) respectively as shown in Fig1.3. The total power number of the parallel flow pattern was 10, of the diverging flow pattern was 9.5, of the merging flow was 8.4 at 360°

1°

the impeller rotational speed equal to 250rpm.

Markopoulos et al. [10] presented that the dependence of power consumption on baffle length, L, in vessels agitated by a dual Rushton turbine system was studied within the turbulent regime, and also in relation to the impeller spacing ΔH(Fig. 1.4).

The two Rushton turbines act independently at ΔH >1.65d. As the baffle length decrease, an increased mutual interaction between the two impellers can be observed for a large regime of ΔH/d values. Power number is not affected by ΔH for the unbaffled agitated systems studied.