Cross-Sectional Imaging of the Liquid Film in Horizontal Two-Phase Annular Flow

Volume 3 ◽  
2004 ◽  
Author(s):  
Daniel J. Rodri´guez ◽  
Timothy A. Shedd

Planar laser induced fluorescence (PLIF) was applied to horizontal air/water two-phase annular flow in order to clearly image the liquid film and interfacial wave behavior at the top, side and bottom of the tube. The visualization section was fabricated from FEP, which has nearly the same refractive index as water at room temperature. This index-matched test section was used to allow imaging of the water to within approximately 10 microns of the 15.1 mm I.D. tube wall. A small amount of dye was added to the water with a peak excitation wavelength near that of a pulsed Nd:YAG laser (532 nm). The laser system generated an approximately 5 ns pulsed light sheet at 30 Hz. Images of the liquid film were captured using a digital video camera with a macro lens for a resolution of about 8.2 microns/pixel. Cross-sectional data at 68 annular flow conditions were obtained. The observations of the liquid film between waves indicated that the film thickness was relatively insensitive to both gas and liquid flow in the annular regime, confirming film thickness measurements reported elsewhere. In addition, the structure of waves changes significantly from wavy-annular, where peaked or cresting waves dominate, to fully annular, where the waves are much more turbulent and unstructured. The wave height decreases with increased gas flow and is relatively insensitive to increased liquid flow in the annular regime. The entrainment of gas in the liquid by the waves is very apparent from these images. Although the precise entrainment mechanisms are not entirely clear, a viable folding action mechanism is proposed. The visualization results will be discussed in relation to both conceptual and computational annular flow modeling.

Volume 3 ◽  
2004 ◽  
Author(s):  
DuWayne Schubring ◽  
Timothy A. Shedd

In this study, non-intrusive pressure drop, liquid film thickness distribution and wave behavior measurements have been obtained for 140 and 220 two-phase flow conditions in horizontal 8.8 mm I.D and 15.1 mm I.D. tubes, respectively. Horizontal flow regimes ranging from stratified-wavy to annular were studied in long clear test sections. Pressure drop data appeared to show different trends for the wavy, wavy-annular and annular flow regimes, suggesting that a unique model may be required for each. In addition, wave frequency showed clearly different behavior for these regimes, with only minor liquid flow dependence in the wavy and wavy-annular flows and strong liquid flow dependence in annular flow. Interestingly, disturbance wave velocity could be correlated to within 10% by the gas friction velocity in the annular regime and within 20% in the wavy-annular regime, leading to a simple correlation between pressure drop and wave velocity. Base film thickness data (between waves) show that the film is relatively insensitive to gas flow at the side and top of the tube and that the film thickness around the tube becomes nearly independent of liquid flow rate at high gas flows. Empirical correlations of the various data sets are presented with the goal of aiding general horizontal two-phase flow modeling efforts.


Author(s):  
Hiroshi Kanno ◽  
Youngbae Han ◽  
Yusuke Saito ◽  
Naoki Shikazono

Heat transfer in micro scale two-phase flow attracts large attention since it can achieve large heat transfer area per density. At high quality, annular flow becomes one of the major flow regimes in micro two-phase flow. Heat is transferred by evaporation or condensation of the liquid film, which are the dominant mechanisms of micro scale heat transfer. Therefore, liquid film thickness is one of the most important parameters in modeling the phenomena. In macro tubes, large numbers of researches have been conducted to investigate the liquid film thickness. However, in micro tubes, quantitative information for the annular liquid film thickness is still limited. In the present study, annular liquid film thickness is measured using a confocal method, which is used in the previous study [1, 2]. Glass tubes with inner diameters of 0.3, 0.5 and 1.0 mm are used. Degassed water and FC40 are used as working fluids, and the total mass flux is varied from G = 100 to 500 kg/m2s. Liquid film thickness is measured by laser confocal displacement meter (LCDM), and the liquid-gas interface profile is observed by a high-speed camera. Mean liquid film thickness is then plotted against quality for different flow rates and tube diameters. Mean thickness data is compared with the smooth annular film model of Revellin et al. [3]. Annular film model predictions overestimated the experimental values especially at low quality. It is considered that this overestimation is attributed to the disturbances caused by the interface ripples.


2021 ◽  
Author(s):  
Antai Liu ◽  
Haifeng Gu ◽  
Fuqiang Zhu ◽  
Changqi Yan

Abstract As a key physical parameter in annular flow, liquid film thickness is crucial to study the behavior characteristics about gas-liquid interface under annular flow conditions. In this study, the narrow rectangular channel is taken as the research object, and air-water were used as the media to conduct annular flow experiments under atmospheric pressure. The cross-sectional area of the narrow rectangular channel is 70mm × 2mm. The PCB liquid film sensor can realize multi-point measurement of liquid film thickness. A total of 10 × 16 measuring points are arranged in rows and columns on the surface of the channel, with a spatial resolution of 4.4mm × 4.4mm and a measurement speed of 1000 frames per second. The results show the fluctuation of liquid film is dominated by the ripple wave at low superficial liquid velocity. The frequency distribution of film thickness becomes sharper because of the increase of gas flow, i.e. the interfacial surface becomes smoother. The liquid film will become thinner with the increase of gas flow, but the effect is reduced when the gas flow reaches a certain value. The liquid film will thicken and the number of disturbance waves will increase as the increase of the liquid flow.


2012 ◽  
Vol 44 (1) ◽  
pp. 71-78 ◽  
Author(s):  
Bo-An Lee ◽  
Byong-Jo Yun ◽  
Kyung-Youn Kim ◽  
Sin Kim

2019 ◽  
Vol 141 (6) ◽  
Author(s):  
Mingyang Liu ◽  
Haixiao Liu

A mechanistic model of film movements is developed based on the treatments on the annular flow field. The initial conditions at the inlet are determined by adopting a validated film thickness correlation of fully developed upward annular flow in vertical pipes. The overall pressure gradient is assumed to be uniform all along the axial distance within the elbow and the static pressure is also uniform on every cross section. The axial velocities of the liquid film and the core region are both uniform on the cross-sectional plane. The droplets are assumed to travel in straight lines normal to the inlet plane until colliding on and absorbed by the liquid film surface. The liquid film motion is divided into the axial and radial directions. Energy conservation law and Newton's second law are, respectively, used in the two directions. The film motion calculation is executed by using a discrete method with an explicit solution. The average film thickness and the circumferential thickness distribution on an arbitrary cross section can be obtained for the given flow conditions. The mechanistic model is verified by comparing the predicted circumferential distribution of film thickness with three series of experimental data from the literature. Parametric studies are also conducted to investigate the parameter effects and the range of application. The present work proves that the variation and distribution of film thickness within the elbows can be efficiently described by the mechanistic model.


Author(s):  
Ashwin A. Gadgil ◽  
Robert E. Randall

Annular flow is a flow regime of two-phase gas-liquid flow dominated by high gas flowrate moving through the center of the pipe (gas core). In this paper we have developed and studied an innovative phenomenological model which combines the Zuber & Findlay’s Drift Flux Model’s weighted mean value approach [1] with the 1-D flow approximation equations. The flow is described in terms of a distribution parameter and an averaged local velocity difference between the phases across the pipe cross-section. The average void fraction is calculated as a function of the ratio of weighted mean gas velocity to the weighted mean liquid velocity (Slip ratio) and the drift flux velocity. The void fraction thus estimated is then applied to the 1-D continuity, momentum and energy equations. The equations are solved simultaneously to obtain the pressure gradient. Lastly, we obtain the liquid film thickness using the triangular hydrodynamic relationship between the liquid flow rate, pressure gradient and the liquid film thickness. The thickness of layer obtained, is then used to verify the original estimate of the void fraction. An iterative procedure is used to match the original estimate to the final value. The results from this study were validated against PipeSIM© software and two field measurements conducted on a wet-gas field in Brazil. As opposed to conventional drift flux models which are based on four simultaneous equations, this model relies on three, thereby significantly reducing the computational resources necessary and is more accurate as we account for variable velocities and void fractions across the pipe cross-section.


Author(s):  
G. F. Hewitt

Many of the important features of annular two-phase flow are governed principally by the existence and behaviour of waves on the interface between the phases. In particular, the annular flow system is characterized by rapid wave growth leading to extremely large waves known as ‘disturbance waves’. By studying these waves, much can be learnt about annular flow. This paper summarizes previously published Harwell work in which a number of parameters of the waves have been established using a variety of experimental techniques, including (1) high-speed ciné studies using both normal and axial viewing; (2) tracking of the waves using flush-mounted conductance probes; (3) continuous film thickness measurements using a fluorescent technique; and (4) velocity measurement using an optical-mechanical stroboscopic device. Extensive data have been obtained on wave frequency, amplitude, velocity, and break-up. If a long enough channel is used, the wave frequency and velocity become constant. It is likely that wave growth continues but that break-up of the waves occurs giving rise to liquid droplet entrainment which is balanced by droplet redeposition, thus giving a quasi-stable situation. Although the frequency becomes constant, the waves are by no means periodic. The fluorescence measurements show that the waves have an amplitude about five times greater than the mean film thickness and that the wave has an axial size of about the same order as the tube diameter. The large amplitude of the waves is confirmed by the axial view photography which also demonstrates large and transient amplitude variations around the inner periphery of the wave.


Sign in / Sign up

Export Citation Format

Share Document