Search

Tip vortex cavitation (TVC) is an important cavitation phenomenon in marine propeller. The formation and evolution of tip vortex cavitation are hot topics consistently both in engineering application and mechanism research. In this paper some recent studies on tip vortex cavitation inception and the noise of tip vortex cavitation evolution are presented. The effects of both flow field and water qualities on tip vortex cavitation inception are considered by experiments and numerical simulations. The results show that besides the average minimum pressure in the vortex core the turbulence fluctuation and water qualities including air content and nuclei distribution have great influence on tip vortex cavitation inception. Based on the idea of first nucleus cavitating in tip vortex core new prediction formula for tip vortex cavitation inception is proposed. The synchronous technique of high speed video observation and noise measurement are adopted to study the development of tip vortex cavitation. S-type total noise characteristics are obtained when cavitation number from low to high. Vortex singing is found in the case where the tip vortex cavitation just before leaves the tip region. The excited mechanism of vortex singing is proposed by analyzing the wave propagation on the interface of vortex cavity.

In this paper, recent measurements of tip vortex flow with and without cavitation carried out in Cavitation Mechanism Tunnel of China Ship Scientific Research Center (CSSRC) are presented. The elliptic hydrofoil with section NACA 662-415 was adopted as test model. High-speed video (HSV) camera was used to visualize the trajectory of tip vortex core and the form of tip vortex cavitation (TVC) in different cavitation situations. Laser Doppler velocimetry (LDV) was employed to measure the tip vortex flow field in some typical sections along the vortex trajectory with the case of cavitation free. Stereo particle image velocimetry (SPIV) system was used to measure the velocity and vorticity distributions with and without cavitation. Series measurement results such as velocity and vorticity distributions, the trajectory of tip vortex core, the vortex core radius, cavity size and cavitation inception number were obtained. The results demonstrated that the minimum pressure coefficient in the vortex core obtained by flow field measurement was quite coincident with the tip vortex cavitation inception number obtained under the condition of high incoming velocity and low air content. And TVC would decrease the vortex strength comparing with the case without cavitation.

The numerical simulation of the axial flow impeller blood pump NIVADIII is carried out by using a CFD multiphase flow model. The hydrodynamic performance of the pump and the flow field in the pump are analyzed, and the shear stress distribution is obtained. A hemolytic prediction model based on the shear stress is built based on the calculation results, and it can be used for quantitative predictions of the hemolytic behavior of a blood pump. Hemolysis tests in vitro were performed 6 times with fresh bovine blood. At each time, the flow of the pump NIVADIII is 5 L/min and the outflow tract pressure is 100 mmHg. According to the tests, the plasma free hemoglobin (FHB) content and the hematocrit (HCT) are measured after 0 s, 0.5 s, 1 s, 1.5 s, …4 s. At the end of each experiment Normal Index of Hemolysis (NIH) of NIVADIII is calculated. The average of NIH is 0.0055 g/100L, almost identical with that obtained from the hemolytic prediction model. This method can be applied in the selection stage of a blood pump.

This article presents an approach which employs a commercial Reynolds-Averaged Navier-Stokes (RANS) solver to predict the steady wake field and loading distributions for a rim driven thruster. Four different cases of propeller blades are chosen to be calculated with the presented method. The propeller blade radial circulation and chordwise circulation density distributions are analyzed. The maximum radial circulation is found at the blade tip, which is different from conventional shaft driven propeller. The numerical results indicate that there is no tip leakage vortex in rim driven propulors. But there exist the tip joint vortex and the root region vortex. Bollard characteristics are calculated by taking rim surface effect into account. From the predicted results the second case in this paper is selected as the final one to perform hydrodynamic experiment. The calculation results with empirical rim surface corrections are compared with the measurement. It shows that the developed numerical method can well predict hydrodynamic performances of the rim driven thruster.

By means of numerical simulation, the evolvement of tip vortex and vortical trailing wake of a propeller were studied in details. From upstream of leading edge up to downstream of trailing edge, the tip vortices region extends rapidly and becomes sizable, which reason is that the transport of vortices between trailing wake and tip vortex takes place. The trailing wake region or the tip vortex region, each consists of sub-regions. The direction of the vortices in one of the sub-regions is opposite to that of the other sub-region. Each sub-region is structured with vortex layers of varying strength. The trailing wake possesses significant thickness and the tip vortex region occupies sizable space. The present analysis also show that within certain distance downstream from tailing edge, although the size and the configuration of the vortices region varies with going backwards, the total vortices flux through each station almost remains constant.

Using Singhal’s full cavitation model and a modified k−ω RNG turbulence model, the unsteady cavitating flow and the dynamic shedding of cloud cavitation from the 3D elliptical foil were numerically investigated. For validating the simulation result, the corresponding experiments were carried out in cavitation tunnel at CSSRC. The modification of turbulence model made to the turbulence model addresses the influence of compressibility effect on turbulence by artificially modifying the turbulent viscosity and takes effect only in the compressible mixture flow region. The main features of cloud cavitating dynamics shedding agree well with the experimental observations.

Based on established numerical methods and hydrodynamic performance testing facilities, CFD technique are applied to improve the design of the implantable axial flow blood pump and the flow analysis. (1) Applying brushless machine magnet steel, reducing its thickness while increasing the length, the flow channel's cross-section is increased, with no space connection between the large and the small rotators, and with a cone transition segment from the bearing to the principal axis, the flow is made smoother. The rotating speed is lowered by 1000 rpm – 1200 rpm under the same flow rate and pressure head, and thus the hemolysis can be avoided. (2) Different outlet stator guiding vanes are selected for the same blood pump for analyses of hydrodynamic performances and flow fields. An excellent design not only can regulate the rotating flow field into an axial one, reduce the circulation loss and improve the pump efficiency, but also can avoid backflow, vortex and secondary flow at the pump outlet, and thus the thrombus can be prevented. (3) The calf live tests show that some residual clots exist at the inner wall of the outlet connection bends, which are analyzed and explained by CFD techniques and the corresponding improvements are proposed. All results are verified by hydrodynamic performance tests and PIV flow field tests, and consistent conclusions are obtained.

A process for numerical analysis of radial circulation distribution of propeller blade is proposed and presented. It is based on the results of numerical simulation of the velocity field around propeller blades and in the wake. The well-known traditional method using tangential velocity data in the wake and applying Stockes’s theorem was also examined in the investigation. The results from two approaches are compared with each other. It is found that if the traditional way is utilized, in many cases an unexpected “hump” appears in the circulation distribution at certain outer radius. The authors calculated the circulations directly around blade sections, and it is referred as direct method. The unexpected hump of the circulation distribution disappears in the results of direct method. This article also discusses the reasons of the appearance of the unexpected hump in traditional approach. The direct method is proposed to have a potential in analyzing or verifying the radial road distribution for designed propeller and the numerical analysis instead of experimental validation for circulation distribution can be as a tool in the propeller design process.

This article presents the design of a new implantable axial-flow blood pump. The special feature of the flow channel inside the blood pump is that the blood is driven by a big-small tandem impeller installed in the inner hole of the cylinder magnet of a brushless direct current motor. The inner hole makes the main flow channel possible, while the gap between the inner end of the stator and the outer end of the cylinder magnet gives the shape of the tributary flow channel. There is no motor magnet inside the main flow channel, therefore, more blood can pass through it. The gap of the tributary flow channel is very small, but the blood flow in it is not blocked. Thus, the efficiency is increased and the volume and weight of blood pump can be reduced greatly. The outer diameter, length and weight of the manufactured implantable axial-flow blood pump are 29.6 mm, 76 mm and 158 g, respectively. The impeller spins at the speed of 9000 rpm and can generate a pressure head of 100 mmHg and a flow rate of 8 L/min. In an animal experiment, the blood pump has been successfully applied as a Ventricular Assist Device (VAD) in the chest of a small cow. Besides, a mathematical model is established to simulate the flow inside an axial-flow blood pump of implantable VAD. The numerical studies on the performance of the implantable axial-flow blood pump are carried out by combining this mathematical model and the Fluent software. The numerical results agree well with those of experiments, with the maximum error less than 10%.