Figure 14. The predicted torques acting on the shaft and on the vessel. (a) N = 90 rpm; (b) N = 120 rpm.
As the momentum transfer occurs between the blade and the fluid, leading to the vortex motion, thus, the torques, like other transient simulation variables, oscillate continuously with time, as shown in Figure 14. It is seen that the amplitude of the torque oscillation acting on the vessel is slightly larger than that on the shaft. The power consumption is then calculated by taking the average values of the torque.
The measured and predicted power consumption and the corresponding power number, NP , is listed in Table 3. Here, two cases of N = 90 rpm and N = 120 rpm are considered. We can see that only a slight difference (less than 0.5%) exists between the calculated NP values by the torques acting on the shaft and on the wall. In addition, the CFD predictions demonstrate satisfactory agreement with the experimental results, with a mean error of 2.30%. It is noted that the experimental Nprises by 11.99 to 12.94 when N increases from 90 rpm to 120 rpm, although the flow is in a turbulent regime30. Meanwhile, the averaged value of NP in the case ofH /T =1.5 can reach 12.5. Whereas for the Maxblend impeller, the NP value is just around 2 in a turbulent regime in the case of H /T =1.131. This indicates that the proposed MBC agitator can load the energy into the fluid effectively, which is why efficient mass exchange in both axial and radial directions have been achieved with improved TKE performance.
Table 3. The predicted and experimental torques and the corresponding power number during an averaged time period of 50s.