Partial cavitation dynamics in an axisymmetric converging-diverging nozzle are investigated experimentally. Shadowgraphy is used to visualize and analyze different cavitation regimes. These regimes are generated by changing the global static pressure and flow velocity independently. Cloud cavitation is the most interesting and complex regime, because the shedding of vapor clouds is caused by two different mechanisms: the re-entrant jet mechanism and the bubbly shock mechanism. The dynamics are investigated using a position-time diagram. Using such a diagram we show that for cavitation number σ > 0.95 the cavity shedding is caused by the re-entrant jet mechanism, and for σ < 0.75 the mechanism responsible for periodic cavity shedding is the bubbly shock mechanism. Both mechanisms are observed in the transition region, 0.75 < σ < 0.95. The shedding frequencies, expressed as Strouhal numbers, collapse on a single curve when plotted against the cavitation number, except for the transition region. The re-entrant jet mechanism is a pressure gradient driven phenomenon, which is caused by a temporary stagnation point at the cavity front. This leads to stick-slip behavior of the cavity. In the bubbly shock regime, a shock wave is induced by a collapse of the previously shedded vapor bubbles downstream of the venturi, which triggers the initiation of the detachment of the growing cavity. The propagation velocity of the shock wave is quantified both in the liquid and the mixture phase by means of the position-time diagram.

In this paper, the cavitating flow around a bluff body is studied both experimentally and numerically. The bluff body has a finite length with semi-circular cross section and is mounted on a surface in the throat of a converging-diverging channel. This set-up creates various 3D flow structures around the body, from cavitation inception to super cavities, at high Reynolds numbers ( Re = 5 . 6 × 10 4 − 2 . 2 × 10 5 ) and low cavitation numbers ( σ = 0 . 56 − 1 . 69 ). Earlier studies have shown this flow to be erosive and the erosion pattern varies by changing the flow rate and w/o the cylinder; hence, this study is an attempt to understand different features of the cavitating flow due to the cylinder effect. In the experiments, high-speed imaging is used. Two of the test cases are investigated in more detail through numerical simulations using a homogeneous mixture model. Non-cavitating simulations have also been performed to study the effect of cavitation on the flow field. Based on the observed results, vortex shedding can have different patterns in cavitating flows. While at higher cavitation numbers the vortices are shed in a cyclic pattern, at very low cavitation numbers large fixed cavities are formed in the wake area. For mid-range cavitation numbers a transitional regime is seen in the shedding process. In addition, the vapour structures have a small effect on the flow behaviour for high cavitation numbers, while at lower cavitation numbers they have significant influence on the exerted forces on the bluff body as well as vortical structures and shedding mechanisms. Besides, at very low cavitation numbers, a reverse flow is observed that moves upstream and causes the detachment of the whole cavity from the cylinder. Such a disturbance is not seen in non-cavitating flows.

Cavitation is a complicated multiphase phenomenon, where the production of vapor cavities leads to an opaque flow. Exploring the internal structures of the cavitating flows is one of the most significant challenges in this field of study. While it is not possible to visualize the interior of the cavity with visible light, we use X-ray computed tomography to obtain the time-averaged void fraction distribution in an axisymmetric converging-diverging nozzle ('venturi'). This technique is based on the amount of energy absorbed by the material, which in turn depends on its density and thickness. Using this technique, two different partial cavitation mechanisms are examined: the re-entrant jet mechanism and the bubbly shock mechanism. 3D reconstruction of the X-ray images is used (i) to differentiate between vapor and liquid phase, (ii) to obtain radial geometric features of the flow, and (iii) to quantify the local void fraction. The void fraction downstream of the venturi in the bubbly shock mechanism is found to be more than twice compared to the re-entrant jet mechanism. The results show the presence of intense cavitation at the walls of the venturi. Moreover, the vapor phase mixes with the liquid phase downstream of the venturi, resulting in cloud-like cavitation.

Cavitating flow dynamics are investigated in an axisymmetric converging-diverging Venturi nozzle. Computational Fluid Dynamics (CFD) results are compared with those from previous experiments. New analysis performed on the quantitative results from both datasets reveals a coherent trend and shows that the simulations and experiments agree well. The CFD results have confirmed the interpretation of the high-speed images of the Venturi flow, which indicated that there are two vapor shedding mechanisms that exist under different running conditions: re-entrant jet and condensation shock. Moreover, they provide further details of the flow mechanisms that cannot be extracted from the experiments. For the first time with this cavitating Venturi nozzle, the re-entrant jet shedding mechanism is reliably achieved in CFD simulations. The condensation shock shedding mechanism is also confirmed, and details of the process are presented. These CFD results compare well with the experimental shadowgraphs, space-time plots, and time-averaged reconstructed computed tomography slices of vapor fraction.

This study focuses on the measurement accuracy of Magnetic Resonance Velocimetry (MRV) in high-speed turbulent flows. One of the most prominent errors in MRV is the displacement error, which describes the misregistration of spatial coordinates and velocity components in moving fluids. Displacement errors are particularly critical for experiments with high flow velocity and high spatial resolution. The degree of displacement error also depends on the sequence structure of the MRV technique. In this study, two MRV sequence types are examined regarding their measurement capabilities in highspeed turbulent flows: a conventional MRV sequence based on the popular "4D FLOW" technique, and a newly developed sequence, named "SYNC SPI". Compared to conventional MRV, SYNC SPI is designed for high measurement accuracy, and not for imaging speed, which limits its application to statistically stationary flows. Both sequence types are evaluated in a flow experiment with a converging-diverging nozzle. Time-averaged results are presented for velocities up to 12 m/s at the throat. Supported by Particle Imaging Velocimetry, it is shown that SYNC SPI is capable of acquiring accurate velocity data in these highly turbulent flows. In contrast, the data from the conventional MRV sequence exhibits substantial displacement errors with a maximum displacement of 21 mm. The long acquisition time is the main disadvantage of the SYNC SPI sequence. Therefore, it is examined if undersampling and non-linear reconstruction, known as Compressed Sensing, can be utilized to make data acquisition more efficient. In the presented measurements, Compressed Sensing is successfully applied to shorten the acquisition time by up to 70% with almost no reduction in measurement accuracy.