Flow Pattern Transition Research Articles

Experimental study on heat transfer and flow pattern transition for R1233zd/FC72 mixtures across tube bundles

An experimental investigation was conducted to study the heat transfer and two-phase flow patterns of pure R1233zd and non-azeotropic R1233zd/FC72 mixtures (0.75/0.25, 0.5/0.5, and 0.25/0.75 by mass) across tube bundles. The test section of tube bundles was transparent to observe the flow pattern. The flow boiling heat transfer was measured. The experiments were carried out with vapor qualities of 0.1–0.9, mass flow rates of 3–5 kg/h and heat fluxes of 3000–7000 W/m2. Falling film flow, bubbly flow, bubbly-shear flow, and shear flow were observed. The flow pattern transition maps were presented for various conditions. Increasing the mass fraction of the volatile component R1233zd causes the flow pattern transitions occurring at lower vapor qualities. The heat transfer coefficient of mixtures is lower than that of pure R1233zd. With increasing vapor quality, the heat transfer coefficient increases for both pure R1233zd and the mixture with a R1233zd mass fraction of 75 %. However, for mixtures with R1233zd mass fractions of 50 % and 25 %, the heat transfer coefficient initially increases and then decreases with increasing vapor quality. For vapor qualities higher than 0.5, increasing the volatile component, R1233zd, significantly enhances heat transfer. The heat transfer coefficient increases with increasing mass flow rate and heat flux. The experimental data were compared with correlations in the literature with poor agreement. An improved heat transfer correlation for non-azeotropic mixtures flowing across tube bundles was proposed, which effectively predicted the experimental data.

Two-phase flow characterization for metal concentric tubes with capacitance measurement

The two-phase flow characteristics for the concentric tube would be different to a plain tube. But the existing research focused on the two-phase flow in circular tubes, without concentric structures. The purpose of this paper was to reveal the two-phase flow characteristics for R134a in both horizontal and vertical upward flow between two concentric tubes by capacitance sensors. The capacitive signals corresponding to different vapor qualities were measured with the mass fluxes from 18.37 kg m−2 s−1 to 64.31 kg m−2 s−1. The experimental data showed that the value of the time-averaged capacitive signal was mainly determined by the vapor quality. The measured capacitance was independent of the mass flux and electrode width, but greatly influenced by the flow direction. The result in horizontal flow was flatter when the fluid was closer to the single-phase and steeper at moderate vapor quality (from 0.1 to 0.9), while it was close to a straight line in vertical upward flow. Moreover, the capacitive signals had oscillation characteristics. By the probability density function, it was found that the amplitude features and oscillation levels were related to the flow patterns. The amplitude degree reflected the stability of the two-phase flow. The oscillation levels could be explained by the flow pattern transition. The corresponding relationship between flow patterns and oscillation characteristics was discovered, which could provide more information for the identification of flow patterns.

Cross-dimensional isothermal model for the transient restart of weakly compressible laminar/turbulent flow of time-dependent non-Newtonian fluid in a long pipeline

Based on reasonable simplifications of three-dimensional governing equations, a cross-dimensional isothermal model for the transient flow restart of time-dependent non-Newtonian fluid in a long pipeline is established, in which conservation equations are in one-dimensional (1D) form and two key variables (average velocity and wall shear stress) in the 1D conservation equations are related to some two-dimensional (2D) equations. This model considers fluid compressibility, complex non-Newtonian rheological behavior, and flow pattern transition. A prediction-correction method and a trial calculation method are combined to solve the cross-dimensional model. The presented cross-dimensional simulation is validated from different perspectives using a theoretical formula, a mature empirical formula, and numerical results from 2D and 1.5-dimensional (1.5D) simulations. The presented cross-dimensional simulation is considered to have strong universality for predicting flow processes in long pipelines. Moreover, the complex flow restart in a long waxy crude oil pipeline is numerically simulated. The results indicate that the flow restart can be divided into three stages according to differences in various influence factors and flow features. These three stages include the laminar/turbulent flow in the yield front movement, the laminar/turbulent flow in the fluid structure breakdown and flow pattern transition, and the turbulent flow in the pure fluid structure breakdown. Analysis of flow processes in these three stages reveals the complex characteristics of the flow restart in a long pipeline.

Flow pattern diagram of compressible non-equilibrium gas flow around a circular cylinder

An investigation into the non-equilibrium gas flow around a circular cylinder within the Knudsen number (Kn) range of 0.001–1 and the free-stream Mach number (Ma) range of 0.01–6 is presented using the unstructured grid unified gas kinetic scheme. The primary objective is to examine the impact of Kn and Ma on flow patterns. The flow pattern diagram illustrating seven flow patterns in the Ma-Kn space is provided, including the transition boundary between bow shock-wave with laminar flow (BS+L) and bow shock-wave with vortex flow (BS+V). The relationships between Re-Kn and Ma-Re both follow the power function: y=eβxα, where α and β are constants. The study also provides a more precise critical curve of vortex shedding in subsonic inflow, the boundary of tailing shock-wave, and the boundary of vortex shedding in a transonic inflow. The flow pattern diagram indicates that the variation of flow separation with Kn is non-monotonic across the entire Ma range but is monotonic at Ma>1. In the subsonic inflow, the critical Re of flow separation (Rec) increases with Ma, while Rec initially increases and then decreases with Kn. The critical Ma at the turning point is about 0.72. In supersonic inflow, the critical Re associated with the onset of flow separation either increases or decreases with the increase in Ma or Kn. The critical Re of vortex shedding is non-monotonic with Kn. The critical Re of the trailing shock-wave decreases with both Kn and Ma. In the transonic inflow, the critical Re and critical Ma of vortex shedding decrease with Kn. As rarefaction increases, the type of flow patterns decreases. The flow pattern diagram provides a visual representation of the impact of rarefaction and compressibility effects on flow pattern transitions and assists in determining the applicable range of the drag coefficient model.

Numerical simulation of dense-phase pneumatic conveying in vertical pipe for gasifier

Abstract The stable transportation of pulverized coal in the vertical pipe is significant for the operation of the gasifier. There are few studies on the flow characteristics and flow pattern transition of particles in vertical pipes with small diameters. This paper has modeled and analyzed the flow characteristic of powder in dense-phase pneumatic conveying through 25 mm vertical pipe using CFD. Firstly, the grid independence is verified to determine the optimal mesh size. Then, the influences of different solid loading ratios (SLRs) and conveying velocities on particle flow characteristics, flow stability, and flow pattern transition are investigated. The results show that the flow pattern in the vertical pipe changes from annular flow to uniform flow at high SLR and low conveying velocity. Moreover, the evolution regulation of resistance characteristics under different conveying velocities is further revealed. Considering the conveying stability and economic benefit, the most suitable conveying velocity is 6 m/s.

Numerical simulation of boiling behavior in vertical microchannels

High heat flux electronic devices put forward new requirements for heat dissipation, and boiling heat transfer technology is widely used because of its higher heat dissipation capacity. In this study, the volume of fluid method was employed, along with the incorporation of the Lee phase-change mass transfer model, to investigate two-phase flow and heat transfer in vertical upward rectangular microchannels. The heat flux was varied within the range of 10–40 kW/m2, while the mass flux was varied within the range of 200–600 kg/m2 s. With the increase in heat flux, bubble flow, slug flow, churn flow, and annular flow were found successively. A phase diagram was established to predict the flow pattern transition during the boiling process. When the flow pattern changes to the churn and the annular flow, the active nucleation site density increases obviously with the Boiling number (Bo). A new correlation was proposed for two-phase flow boiling heat transfer, suitable for vertical upward channels in microscale fluids. The friction factor obtained using the Darcy friction factor equation agrees well with the simulation results at a high-pressure drop. The instability in microchannels increases with the increase in heat flux, particularly in annular flow, resulting in more severe wall temperature fluctuations.

Critical criterion for droplet breakup in a contractive microchannel

Predicting the critical transition condition for different behaviors of droplet flowing through a micro contraction is a classic academic issue, and there is currently a lack of a simple and universally applicable prediction formula. This article aims to construct critical transition conditions for different droplet behaviors through their characteristic time. In present work, the droplet behavior, including deformation and breakup, are observed in a locally contractive microchannel by a high-speed camera. By tracing the dynamic evolution of droplet interface, it is found that the essential differences between droplet deformation and breakup is whether the minimum neck width will be less than the critical value for the appearance of the irreversible collapse. Based on this, two characteristic times are proposed to describe the deformation and breakup processes, which well explain the trend of flow pattern transition lines. Two mathematical models are established for the two characteristic times, and the critical conditions for the transformation from deformation to breakup is derived from these models. The predicted results are in good agreement with the experimental results, indicating the applicability of the proposed method in this paper.

Experimental study of liquid/dense-gas pipe flow

The pre-salt reservoirs play a significant role in the current Brazilian ultradeep water offshore scenario. Their importance is expected to increase soon. The fluids are transported through horizontal, inclined, and vertical wells of several kilometers, and due to the environmental conditions at the reservoir, dense-gas/liquid flow occurs in the pipelines. However, the literature lacks experimental data for gas-liquid flow with densities of the same order of magnitude, which affects the mechanistic model's accuracy for flow patterns, pressure gradients, and holdup predictions. To improve the understanding of that specific two-phase flow, an experimental apparatus was built, and an experimental study was carried out using sulfur hexafluoride (100 kg/m3) and mineral oil (850 kg/m3) at medium pressures. A density ratio lower than 10 was obtained for steady-state horizontal flow in a 50.8-mm-i.d. stainless-steel pipe, and flow patterns, pressure gradient, and holdup were measured. The experimental results show the effect of the low-density ratio on the stratified to non-stratified flow-pattern transition and the occurrence of the dual-continuous flow pattern at high mixture superficial velocity. Moreover, the pressure gradient and holdup data were compared with the results of a commercial multiphase flow simulator, showing good agreement.

Flow in a Taylor–Couette Reactor with Ribbed Rotors

This paper investigates the flow structure and flow pattern transition within a conical ribbed Taylor–Couette reactor (TCR), which is 4 mm in gap width and 200 mm in height, via particle image velocimetry (PIV) and numerical simulation methods. The effect of various parameters on the vortex structure and on flow transition, including the structural parameters of the ribs (rib spacing and rib width) and the operating parameters (Taylor number and axial Reynolds number), were investigated. Without axial flow, the ribbed TCR can control the flow structure while maintaining the symmetry of the flow field. Under certain conditions, a Taylor vortex pair can form between the ribs, with the down vortex rotating clockwise and the up vortex rotating counterclockwise. The axial dimension of the Taylor vortex can be controlled by adjusting the rib spacing, which can be summarized into four different conditions according to the size of the rib spacing. With axial flow, the axial Reynolds number greatly impacts the Taylor vortex structure within the ribbed TCR, and as the axial Reynolds number increases, the up vortex appears to be compressed and the down vortex appears to be stretched. The double vortex flow pattern between the ribs is eventually transformed into a single vortex. The critical axial Reynolds number for flow pattern transition is defined and correlated with the Taylor number and rib spacing. The results show that the critical axial Reynolds number is positively proportional to the Taylor number and is inversely proportional to rib spacing. The empirical correlation equation developed in this study shows strong predictive power and is validated using the experimental results. Overall, this study provides a comprehensive understanding of the flow structure and pattern transition within a ribbed TCR and lays the foundation for the further optimization of TCR design.

Intercolumn two-phase flow patterns across falling film tube bundles

Two-phase flow patterns are critical in falling film heat devices. Hydrodynamic characteristics of turbulent falling films across a horizontal triangular tube bundle were studied under a range of film Reynolds numbers and contact angles. A flow pattern map based on the k-means clustering approach was proposed after the intercolumn two-phase flow patterns were grouped. The results demonstrated that that the intercolumn liquid-vapor flow patterns in a horizontal tube bundle could be effectively and reasonably grouped using the k-means clustering approach according to the results of the void fraction and interface area. There are four intercolumn liquid-vapor flow patterns that were identified: bubbles flow, annular-slug-column flow, half-annular flow, and multi-bridges flow. The annular-slug-column flow pattern and the multi-bridges flow pattern are next to one another in the flow pattern map and are capable of transitioning one into the other. It is only through the two transitional flow patterns that the half-annular-column flow pattern and bubbles flow pattern can turn to one another. In the bubbles' flow pattern and half-annular-column flow pattern, respectively, the highest and least averaged heat transfer coefficients are attained. Simply raising the film Reynolds number or lowering the contact angle will not produce the best heat transfer results.

Gas Hydrate Formation and Slurry Flow Characteristics of Gas–Liquid–Solid Multiphase Systems

Hydrate formation, aggregation, and deposition have become major hazards for the safe transportation of oil and gas pipelines. Although there has been a great deal of research into the behavior of particles during hydrate formation and transport, there is less research into the multiphase generation and flow characteristics of hydrate gas and liquids. In this paper, the flow characteristics of hydrate slurry and the flow pattern transition in gas–liquid multiphase system were studied, and the generations of it were compared between a gas–liquid multiphase system and oil–water emulsion system. The results showed that the flow parameters such as pressure, pressure drop, and density fluctuated more drastically in the gas–liquid system than that in the emulsion system, the higher fluctuation meant a higher risk of pipe plugging. At the same time, the influence of pressure and dosage on multiphase flow pattern transformation and boundary change conditions were studied. A gas-slurry flow pattern transformation diagram was drawn based on the defined transformation conditions between flow patterns. The results also indicated that, with the increase of pressure from 5 to 6 MPa, the boundary of slug flow and air mass flow changed to the direction of higher gas and liquid flow velocity. While the boundary of slug flow and wave flow, stratified flow and wave flow changed to the direction of higher gas flow velocity. The boundary of slug flow and air mass flow changed to the direction of lower liquid flow velocity. In contrast, the boundary of slug flow and wave flow, stratified flow and wave flow, slug flow and stratified flow changed to the direction of higher gas flow velocity and lower liquid flow velocity in the 1% inhibitor period compared with that without the inhibitor. This study will provide theoretical references and technological supports for the safety of multiphase transportation pipelines and the safety of deep water production.

Measurement of film fluctuations and aerated characteristics during gas-liquid slug-churn flow transition by wire-mesh sensor

Vertical upward gas-liquid slug flows are frequently encountered in chemical processes and petroleum industries. The measurement of the film fluctuations and the aerated characteristics is of great significance for uncovering the mechanism of slug-churn flow pattern transitions. In this study, a conductance wire-mesh sensor (WMS) measurement system is designed based on a Field Programmable Gate Array (FPGA) to visualize the structures of vertical gas-liquid flows. Liquid film flooding is a significant factor prompting the transition from slug to churn flow. Based on the WMS data, the 3D film structures are derived to indicate film instability during the flow pattern transition. Three types of film fluctuations in stable slug flow, unstable slug flow, and churn flow are presented. Liquid slug aeration is another important factor contributing to the slug-churn flow transition. The spatial distribution and the diameters of the gas bubbles in the liquid slug are detected by the WMS. The coalescence behavior of the bubbles is uncovered. Finally, mechanistic models based on the film flooding and slug aeration are constructed to predict the boundary of the flow pattern transition. The performance of the film flooding model and slug aeration model in predicting the onset of churn flow is evaluated.

Hydrodynamics of gas–liquid flow in a capillary at elevated pressure: Implications for the fixation of CO2 into propylene carbonate in microreaction systems

It is an effective method to capture and utilize carbon dioxide (CO2) by using a microreactor to fix CO2 into propylene carbonate (PC). It is important to investigate the hydrodynamics of gas–liquid flow in order to ensure that the CO2 fixation in microreactor is operated under suitable conditions to obtain efficient mass transfer performance. However, few studies have been concerned with PC-CO2 two-phase flow in microreactors, especially at elevated pressures. Therefore, a high-pressure microreactor observation platform is established in this work to explore the hydrodynamics of PC-CO2 two-phase flow at a pressure range of 0.5–3.0 MPa. Three typical patterns such as bubbly flow, slug flow, and slug-annular flow were observed, and the flow pattern diagram at different pressures was plotted using WeG and WeL. The influence of the pressure on flow pattern transition lines was further discussed. Moreover, the effects of pressure on bubble size, liquid film thickness, bubble frequency, and gas volume fraction were investigated, as well as the corresponding correlation models. This work may provide guidance for the operation of gas–liquid microreactor at elevated pressure, especially for the fixation of CO2 into PC.

Mass transfer enhancement for CO2 chemical absorption in a spiral baffle embedded microchannel

This paper presents an experimental study on CO2 chemical absorption by Monoethanolamine (MEA) solution in microchannel. In particular, a novel spirally distributed baffle structure is proposed to be embedded into the microchannel so as to generate the swirling flow pattern and to enhance the mass transfer. The effect of spiral baffle structure on the two phase flow pattern transition, the CO2 gas void fraction, the pressure drop, and the mass transfer performance are characterized and discussed, under varied gas and liquid flow rates, and MEA concentrations. Results demonstrate that the spiral baffle structure could significantly enhance the gas–liquid mass transfer, with overall liquid side volumetric mass transfer rate (kLa) up to 2.35 s−1 and enhancement number E up to 1.387 at small gas and liquid Reynolds numbers (ReG < 5.5; ReL < 14). New predictive correlations have also been proposed by considering both the structure enhancement factor and the chemical reaction factor.

Effect of high pressure on severe slugging and multiphase flow pattern transition in a long pipeline-riser system

Severe slugging poses an enormous threat to the safe operation of high-pressure long petroleum pipelines. However, most published studies are conducted in short pipelines under atmospheric pressure. In this work, the gas–liquid two-phase flow experiment is conducted in the pipeline-riser system of 380 m long and 75 mm inner diameter under the separator pressure from atmospheric pressure to high pressure (10 MPa), and flow pattern maps are established. Under high-pressure conditions, liquid fallback and second blowout of severe slugging disappear. In the flow pattern maps, flow pattern types reduce, and severe slugging region shrinks. The boundary between stable flow and severe slugging appears on the left side of the flow pattern map and moves in the direction of higher superficial gas velocity. Increasing separator pressure leads to the increase in the severe slugging period. A correlation equation applicable to predicting the severe slugging period for different pressures (0, 1 and 5 MPa) is proposed, and the average error is less than ± 7.9%. Due to the inhibition of high pressure, the amplitude of the riser differential pressure decreases at most working points and increases only at a few working points where flow pattern transition occurs.

Flow pattern identification for gas-oil two-phase flow based on a virtual capacitance tomography sensor and numerical simulation

Gas-oil two-phase flow is widely encountered in oil exploitation and transportation pipelines. It's complex and transient changes of flow regimes present a great challenge for accurate and real-time measurement. As a non-invasion and real-time measuring method, electrical capacitance tomography (ECT) is suitable for the transient measurement of non-conductive gas-oil flow. However, the highly random and nonlinear nature of multiphase flow make it difficult and limited to investigate the flow parameters based on either static or dynamic measurement. In this research, the whole process of dynamic measurement of ECT applying in gas-oil two-phase flow is thoroughly studied, including simulation calculation, experimental validation and comprehensive data analysis. A simulation approach by coupling the flow and electrostatic field is proposed based on a virtual ECT sensor, in order to monitor the gas-oil two-phase flow characteristics. Based on FLUENT and COMSOL platform, the numerical simulation under six typical flow patterns in a horizontal pipe is carried out. Combining the visualized image generated by ECT measurement and the theory of flow pattern transition, the formation mechanism and structural characteristics of different gas-oil flow patterns are analyzed in detail. Furthermore, this research attempts to analyze the signal fluctuation characteristics caused by flow pattern change, in order to access more in-depth flow information implied in the original capacitance data, via time-series analysis as well as frequency domain analysis based on Flourier Transform. At last, a series of dynamic experiment is conducted to verify the feasibility of the simulation and data analysis approach. The experiment focuses on the flow pattern transition, gas-liquid dynamic characteristics and noise influence in the actual process. It can be concluded from the results of simulation and experiment tests, combining the visualized images and the dynamic characteristics of capacitance signals can make it more effective and intuitive for flow pattern identification, which might be used for the online measurement in real-industry process.

Composite filament with super high effective thermal conductivity

The low thermal conductivity of the clothing filament constrains the efficiency of the thermal management garments. In this paper, a super thermal conductive composite filament (STCCF) was proposed for body thermal management, which was fabricated by hollow Teflon filaments incorporated with hollow copper filaments in the heating and cooling sections. By introducing R134a into the hollow filament, a thermally driven flow was generated due to the pressure imbalance caused by the evaporation and condensation of R134a, which enhanced the heat transport capability significantly. The apparent thermal conductivity of the STCCF increased with the increasing heat power, being up to 5042 W/(m·K) at the heating temperature of 49 °C. Furthermore, bending only had a small effect on the thermal performance of STCCF due to the flow pattern transition in the curved section. Moreover, the heat transfer capacity could be enhanced by a lower temperature difference between the hot and cool sections under constant heat power. Therefore, the proposed STCCF has great potential for human body thermal management, which could contribute to human health and thermal comfort as well as building energy saving.

Thermo-fluid dynamics and synergistic enhancement of heat transfer by interaction between Taylor-Couette flow and heat convection.

This study experimentally and numerically investigated the thermo-fluid dynamics of Taylor-Couette flow with an axial temperature gradient from the chemical engineering perspective. A Taylor-Couette apparatus with a jacket vertically divided into two parts was used in the experiments. Based on the flow visualization and temperature measurement for glycerol aqueous solutions with various concentrations, the flow pattern was classified into six modes: heat convection dominant mode (Case I), heat convection-Taylor vortex flow alternate mode (Case II), Taylor vortex flow dominant mode (Case III), fluctuation maintaining Taylor cell structure mode (Case IV), segregation between Couette flow and Taylor vortex flow mode (Case V) and upward motion mode (Case VI). These flow modes weremapped in terms of the Reynolds and Grashof numbers. Cases II, IV, V and VI are regarded as transition flow patterns between Case I and Case III, depending on the concentration. In addition, numerical simulations showed that in Case II, heat transfer was enhanced when the Taylor-Couette flow was altered by heat convection. Moreover, the average Nusselt number with the alternate flow was higher than that with the stable Taylor vortex flow. Thus, the interaction between heat convection and Taylor-Couette flow is an effective tool to enhance heat transfer. This article is part of the theme issue 'Taylor-Couette and related flows on the centennial of Taylor's seminal Philosophical Transactions paper (Part 2)'.

Experimental study on gas–liquid flow patterns and bubble size in a high-speed rotating impeller of a three-stage centrifugal pump

Head degradation and performance failure will occur when centrifugal pumps work under high gas volume fraction. The gas–liquid two-phase flow patterns and bubble size in a rotating impeller were experimentally studied by high-speed photography technology in a three-stage centrifugal pump. A new impeller was designed with double rings and a transparent shroud, overcoming the speed limitation below 1500 rpm in current studies and reaching the maximum speed of 3500 rpm. Four gas–liquid flow patterns are observed, which are discrete bubble flow, bubble flow, gas pocket flow and stratified flow. The result shows that the transition from bubble flow to gas pocket flow is the critical condition for the severe head degradation. Reducing inlet gas volume fraction, increasing rotational speed and reducing liquid flow rate lower than design value cause the transition from gas pocket flow to bubble flow and the delay of head degradation. By image processing technique, the probability density function (PDF) of bubble diameter in impeller is found to follow a lognormal distribution. Moreover, bubble Sauter diameter dSa has a linear relationship with inlet gas volume fraction λ, rotational speed n−0.75 and liquid flow rate Qw−0.6, respectively. A new correlation is established for predicting bubble Sauter diameter in impeller under high rotational speeds (1500–3500 rpm). The strong relationship between flow pattern transition and head degradation is revealed. Based on the same head performances of three booster stages in pump, the conclusions driven from the visual impeller in the first stage can also be extended to other stages.

Experimental investigation of bubble dynamics and flow patterns during flow boiling in high aspect ratio microchannels with the effect of flow orientation

Multiphase flow and boiling phase-change within microchannels are of paramount importance to thermal management and electronics cooling, among others. The present experimental study investigated the bubble dynamics and flow patterns during flow boiling in a high aspect ratio microchannel at different channel orientations. Hydrofluoroether (HFE−7000) was used as the working fluid within a transparent glass microchannel with a width-to-height aspect ratio of 10 and a hydraulic diameter of 909 μm. A tantalum layer coating was applied to the channel, allowing visual observation to be conducted whilst imposing uniform heating along the channel. Under horizontal and vertical upward flow, the mass fluxes were varied between 14 kg m−2 s−1 and 42 kg m−2 s−1 for Reynolds numbers between 28.4 and 85.2, while the heat fluxes were set from 2.8 kW m−2 to 18.7 kW m−2. The results show that single bubble growth can be divided into three stages, namely: bubble-free growth, partially confined bubble growth, and fully confined growth. The effects of mass flux and heat flux on bubble behaviour and transition points between the defined stages are discussed. Equally important, the channel orientation influences the bubble behaviour in terms of bubble evolution, bubble shape, and bubble nose velocity, with no significant differences observed for the flow patterns when comparing horizontal and vertical upward cases. The comparison of the present work with previous flow pattern map results from the literature shows that earlier flow pattern transition models cannot accurately predict the observed flow patterns in the present configuration. Two types of instabilities are observed, which are associated with the dominant flow pattern during flow boiling and hence a function of the mass flux to heat flux ratio. The first type of instability, high amplitude and long period oscillation, is found during slug flow which is characterised by the presence of bubble rapid expansion along the channel. In addition, a second instability ensues when churn and annular flow create short period oscillations due to the high number of nucleation events, bubble coalescence, and the breakage of liquid slug or liquid film.