Search

This paper presents an experimental study of two-phase R134a flow in the inlet header of brazed plate heat exchangers and its effect on flow distribution among the channels. Visualization of the two-phase flow is accomplished through a 3-D printed transparent window on the inlet header of the heat exchangers. The two-phase flow distribution among channels is quantified based on infrared images of the heat exchanger sidewalls. At the test conditions, the observed flow regimes in the inlet header are periodic and two or three stages are identified in one cycle: top corner vapor flow, vapor jet flow, and (conditional) liquid blockage flow. Among them, the top corner vapor flow affects the distribution most, in which the vapor refrigerant is mainly present at the top corner of the header and branches out through the first several channels, leaving the liquid refrigerant to occupy the rest flow area of the inlet header and present a single-phase like distribution profile. When the inlet vapor quality increases, the distribution of the liquid refrigerant is improved since the vapor refrigerant can reach more downstream channels to help balance the total pressure drop. With the mass flux increases, the maldistribution caused by the header induced pressure drop is more significant, which compromises the benefit brought by the higher vapor momentum. When the number of plates is increased, the liquid refrigerant distribution is worse due to an increased pressure drop in the headers.

This paper presents a method to quantify two-phase refrigerant flow distribution in brazed plate heat exchangers (BPHEs) from infrared images. In this method, the two-phase length of the refrigerant channels is first estimated from the IR images of the BPHE sidewall. Then, the refrigerant distribution is adjusted in a BPHE evaporator model, to have the predicted two-phase length of the channels match the identified results from the IR images. The overall performance of the BPHE is also predicted in the quantification process. The proposed quantification method is validated by the experimental data: heat exchanger capacities, pressure profiles, and surface temperature profiles of the sidewall. This non-intrusive method can effectively quantify two-phase refrigerant distribution in BPHEs with the premise that all liquid refrigerant has evaporated in the heat exchangers.

This paper presents new capacitive sensors built and calibrated to measure cross-sectional void fraction between tubes in vertical headers of microchannel heat exchangers (MCHEs). Eleven individual sensors are assembled into one test header to measure local capacitance independently in real-time. A calibration procedure based on visualization, pressure drop, signal patterns, and mass measurement (quick-closing valve technique, QCV) is proposed. After the calibration, most void fraction data predicted by all eleven sensors fall into ±10% deviations of the experimental results by QCV. The sensors are ready to be utilized to measure local void fraction along vertical inlet or intermediate headers of MCHEs in future studies.

The separation circuitry has been proven in the past to improve the performance of microchannel condensers. In the vertical second header of the condenser, liquid separates from vapor mainly due to gravity. However, separation is usually not perfect, expressed through the separation efficiency. This study presents the phase separation result in the second header calculated by the Euler-Euler method of Computational Fluid Dynamics (CFD). Simulations are conducted for two-phase refrigerant R134a flow in the second header with 21 microchannel tubes in the 1st pass. The inlet mass flux to the second header (through the microchannels of the 1st pass) in the simulation is 166 kg m − 2 s − 1, 207 kg m − 2 s − 1, and 311 kg m − 2 s − 1. The inlet quality is 0.13 to 0.21. The results agree well with the experimental results with flow visualization and the results of a simpler 1-D numerical model. Results show that the liquid separation efficiency decreases as the vapor separation efficiency increases, following a linear trend in the experimental range. The void fraction result shows liquid mainly flows in the half of the header without microchannel tube intrusions. The velocity profile in the header is presented and reverse flow is identified on the exit planes of the inlet section connecting to the 2nd-upper pass and the 2nd-lower pass. The pressure profile in the header is also revealed and it indicates that the 1-D pressure assumption may still apply to two-phase flow in a header.

This paper presents an experimental and numerical investigation of the transient refrigerant and oil distribution in a residential heat pump water heater (HPWH) system. In the experiments, R134a is paired with POE 22 oil. The quick closing valve technique (QCVT) is used to trap the refrigerant and oil in each section of the system. The trapped refrigerant mass is measured after recovered by liquid nitrogen. The retained oil mass in each section, except for the compressor, is determined by the mix and sample technique (MST). Five experiments are conducted to cover a full warm-up of five hours. The experimental data shows that most of the refrigerant is in the heat exchangers. During the water warm-up, the inventory of the refrigerant generally decreases in the evaporator, while in the condenser, it decreases in the first three hours and later increases. The measurements also indicate that most of the oil stays in the compressor and the escaped oil is mainly retained in the heat exchangers. Besides, a model is developed to predict the refrigerant and oil retention in the heat exchangers by applying the mixture vapor quality to calculate the void fraction. The developed model predicts the refrigerant inventory in the condenser with an average deviation of 7.4%, and in the evaporator, 9.1%. The average deviations in the predictions of the oil retention in the two heat exchangers are 15.8% and 10.2%.

The two end plates in plate heat exchangers transfer heat through contact with the corrugated surface of the adjacent thermal plate. This phenomenon effectively increases the active heat transfer area and significantly affects the heat transfer when the heat exchanger has a very small number of plates. That is especially important when such a heat exchanger is used for the determination of the heat transfer coefficient in the development of heat transfer correlations. This paper presents a theoretical analysis of the effect of the end plate on heat transfer in brazed plate heat exchangers. The governing equations are derived through the energy balance analysis for a small element of the end plate. Numerical and analytical models are developed to solve the governing equations. The predictions by the two models agree well with each other. The heat transfer data for single-phase water in the BPHEs with different numbers of plates are used to validate the developed models. Both experiments and theoretical analyses show that the influence of the end plate is more significant in a heat exchanger with a small number of plates and a lower convective heat transfer coefficient. An iteration algorithm is developed to compensate for the end-plate effect in the development of heat transfer correlations.

Maldistribution of fluid among parallel channels is one of the main issues in applications of plate heat exchangers. This paper presents an experimental and numerical investigation of the single-phase flow distribution in the brazed plate heat exchangers, but the results can be applied to plate-and-frame and plate-and-shell designs. In the experiments, the pressure profile in the heat exchanger is measured by the probes inserted into the headers. The flow distribution is determined by the measured pressure drop across the channels and the developed in-channel friction factor correlation. The experimental results indicate that in a U-type brazed plate heat exchanger, the channel flow rate first increases for the first several channels near the heat exchanger entrance due to the sudden expansion of flow in the inlet header. For the rest channels, the flow rate decreases with the distance away from the entrance/exit of the heat exchanger. Such a distribution profile is associated with the axial momentum transfer in the inlet header. The influence of the total flow rate on the distribution profile is trivial, but the maldistribution is more severe with an increased number of plates. Two distribution models presented are developed based on the principle of equal total pressure drop for all flow paths. Two models calculate the pressure profile in the headers by 3-D CFD modeling and 1-D mass and momentum conservation equation. The experimental results validate the models.

This paper presents an evaluation of the performance of a refrigerant distributor with visualization of the two-phase flow. The distributor is a typical conical design w/o the orifice for about 4 kW R134a system. Three factors influencing two-phase flow distribution are varied: mass flow rate, distributor inlet quality, and orientation. We looked at the uniformity of mass flow rates, evaporator inlet qualities, and capacities of each circuit. The authors built a transparent distributor following the exact geometry of an original distributor to visualize two-phase flow regimes exiting the expansion device and entering the distributor. We related flow regimes with distributor performance. A high-speed camera is used to capture two-phase flow regimes. After acquiring results for a baseline situation, the authors applied two approaches in an attempt to improve flow distribution: The first is manual adjustment of the resistance of each circuit individually to achieve uniform mass flow rates, and the second is the homogenization of the flow regime before division by adding an orifice in the distributor before separation to four channels. The average deviations of capacity for each branch are 9% for the baseline case, 3.3% for manual adjusting, and only 1.6% for the homogenization approach, stressing the importance of that simple strategy.

This paper presents the identification of slug/stratified-wavy, stratified-wavy and annular regimes for horizontal flow (in the range of mass fluxes 40 – 150 kg m−2 s−1), and slug, churn, and annular flow regimes for vertical upward flow (in the range 65 – 115 kg m−2 s−1) for R134a flow through 7 mm ID tube. Flow regimes are characterized based on time plot of normalized capacitive signals, kernel density estimation (KDE), power spectral density (PSD), and visualization results from a high-speed camera. Sensors with different axial lengths (D, 2D/3, and D/2) are also tested to study the adequacy of shorter sensors for the characterization of flow regimes. Results show that all three sensors have a similar capability of characterizing flow regimes, justifying the use of the shorter sensors in many applications with limited space.

The low GWP (Global Warming Potential) refrigerant R1234ze(E) is a potential alternative of R134a. This study puts the emphasis on the comparison of the flow boiling pressure drop and heat transfer of R134a and R1234ze(E) in a dimpled flat duct. For the air-conditioning applications, the experiments are conducted at mass flux from 100 to 200 kg m−2 s−1, saturation temperature from 5 to 15 °C, heat flux from 2.5 to 10 kW m−2, and vapor quality from 0.1 to 0.95. The results show that the frictional pressure gradient of R1234ze(E) is 1.10 to 1.27 times that of R134a. The characteristics of the flow boiling heat transfer coefficient of R1234ze(E) are similar to R134a. The increasing flow boiling heat transfer coefficient with vapor quality in the tested range implies the convective evaporation dominates the heat transfer. It is also observed that the flow boiling heat transfer coefficient visibly increases with a rise in mass flux and decreases with the saturation temperature increment due to the increase in vapor to liquid density ratio. The enhancement effect of heat flux on the flow boiling heat transfer shows up at the highest value of 10 kW m−2 only. The heat transfer coefficient ratio of R1234ze(E) to R134a increases from 0.89 to 1.20 with an increase in vapor quality. The comparison of frictional pressure gradient and heat transfer characteristics between R1234ze(E) and R134a has been discussed from the viewpoint of fluid properties.

Oil separation is commonly needed in air conditioning and refrigeration systems to reduce the oil circulation rate and keep the oil inside the compressor. For compactness, the oil separation structure integrated into the compressor is more and more popular than traditional external oil separators. However, a quantitative guideline for the design of oil separation structure is missing for irregular geometry and realistic flow condition at the compressor discharge. This paper presents impinging separation mechanism, one of the basic mechanisms of droplet separation. The separating structures are studied by flow visualization and experimental measurement under realistic compressor discharge conditions. The video of oil mist flowing through the baffles and plates is captured by a high-speed camera and analyzed quantitatively. The effect of the oil flow condition and the effect of separator geometry design are investigated based on the experimental data. With the help of flow visualization, the flow details are analyzed, and a semi-empirical model is proposed to predict the separation efficiency and pressure drop of a certain wave-plates structure. The results and conclusions from this study give useful guidelines about how to reduce the oil circulation ratio by designing oil separators.

This article presents a novel approach to visualize the flow boiling in round tubes under “diabatic conditions”. In the visualization test section, a 3D printed round tube is placed inside a glass ...

This paper presents a new capacitive sensor built and used to measure void fraction of both horizontal and vertical upward flow in the low mass flux range in circular tubes with an inner diameter of 7 mm. Three sensors with different axial lengths (D, 2D/3, and D/2) are built and evaluated to examine the possibility of utilizing shorter sensors in applications with space limitations. Results show all three sensors have the capability to measure void fraction. Due to the nonlinear relation between capacitive signals and void fraction, a calibration procedure based on mass measurement (quick-closing valve technique, QCV) is proposed. After the calibration procedure, most void fraction data measured by the sensor falls into the ±15% deviations of the experimental results by QCV for horizontal flow and ±10% for vertical upward flow. Sensors with the same design can be utilized directly to measure void fraction for the similar test conditions in the future studies.

This paper presents an improved flow pattern map for flow boiling in horizontal micro-fin tubes considering the effect of helix angle. The proposed map is modified from the one developed for horizontal smooth tubes by Wojtan et al. (2005) and takes the early transition to annular flow due to micro-fin geometry into account. Visualization of R410A flow boiling is carried out through the transparent smooth and micro-fin tubes with 0°, 10° and 18° helix angle and plotted on the proposed map for comparison. The result shows that the flow pattern map for micro-fin tube with 0° helix angle is generally similar to that for the smooth tube. For the tube with 10° and 18° helix angle, the annular flow occurs at lower vapor quality and lower mass flux. The proposed map shows good agreements when compared with the visualization and reflects the downward shift (to lower mass flux and vapor quality conditions) of the SW-WA transition with the increase of the helix angle. In order to further validate the proposed model, the flow pattern transitions calculated from the equations are also compared with the experimental data of the micro-fin tube in literature. The results show that the modified map increase the prediction accuracy of the flow patterns in the micro-fin tube.

The effect of airflow profile and condensation pressure on the performance of air-cooled condensers is investigated experimentally. Two large, flattened-tube air-cooled steam condensers are studied...

Oil separation is commonly needed in air conditioning or refrigeration systems to reduce the oil circulation rate and keep the oil inside the compressor. For compactness, the oil separation structure integrated into the compressor is more and more popular than traditional external oil separator. However, a quantitative guideline for the design of oil separation structure is missing for irregular geometry and realistic flow condition at the compressor discharge. This paper presents coalescence, one of the basic mechanisms of droplet separation, studied by flow visualization and analytical models. The misty oil flow through separator is visualized by a high-speed camera and analyzed quantitatively. Oil droplet size distribution is estimated by video processing. Important flow details are revealed, including oil droplet collision, oil droplet coalescence, oil film breakup, and re-entrainment. Separation efficiency is estimated by the ratio between drained and incoming oil mass flow rate. Pressure drop is also measured to evaluate the cost brought by separation structures. An analytical model for coalescing oil separator is developed based on the mass flow balance through multiple layers of coalescing separators. Development of the model starts with the experimental and analytical analysis of first one wire, then a serious of wires to come to a wire pad. The results of the model are verified by experimental measurements in a full vapor compression system with R134a and PAG oil. The results of the model show good agreement with experiment and conclusion provides guidelines for oil separator design and operation.

This paper presents a novel visualization approach to understand the effect of axial micro-fins on two-phase flow behavior inside horizontal round tubes. In order to reflect the real inner geometry in the commercial metal tubes (smooth and micro-finned tubes), clear resin tubes were made by an SLA 3D printer to reproduce the geometry. The 3D printed tube for visualization was installed right after the heat transfer test section. In the heat transfer test section, R410A flow boiling experiments were conducted at 10 °C saturation temperature with heat flux of 15 kW/m2. The results indicated the axial micro-fins do not provide additional force to lift up the liquid in the round tube and the annular flow pattern does not occur earlier (at lower vapor quality or mass flux conditions) than the smooth tube. For low mass flux and vapor quality, the liquid-phase refrigerant is easily trapped in the grooves at the top or side of the axial micro-finned tube. The main liquid level in the axial micro-finned tube is, therefore, lower than that in the smooth tube under the same conditions. This liquid level change is quantified by tracing the liquid-vapor interfaces in the captured videos with change point analysis.

This paper presents heat transfer coefficient and pressure gradient of R1234ze(E) in a tube of 0.643 mm. In addition, visualization sections are added for evaluation of flow patterns. Both heat transfer coefficient and pressure gradient are presented against real saturation pressure, while flow pattern captures at exit of data points are presented in the same plot ( Fig. 8 ). Experiments are conducted on a 24-port microchannel tube with a hydraulic diameter of 0.643 mm. The single-phase friction factor shows fRe = 64 in laminar and underestimation of Churchill prediction in transition (1700

Flow regimes and void fraction in horizontal and vertical round smooth tubes ID 7 mm with R134a in the adiabatic conditions (saturation temperature at 33 °C) and low mass flux (40–150 kg/m2 s for horizontal tubes and 65–115 kg/m2 s for vertical tubes) are presented in this paper. Flow regimes are captured by a high-speed camera while void fraction is measured by quick-closing valve method. The horizontal flow patterns are compared to Wojtan et al. (2005a) flow-regime map and some modifications based on visualization results are proposed. Void fraction results for both horizontal and vertical flows are compared to some widely used correlations. Influences of tube orientation and mass flux on void fraction are discussed. When the vapor quality keeps constant, void fraction of horizontal tubes is larger than that of vertical tubes. Higher mass flux also results in larger void fraction compared that of lower mass flux.

This paper presents the discussion of regimes of two-phase flow of R1234ze(E) in a 24-port microchannel tube with average hydraulic diameter of 0.643 mm, obtained by visualization. The experiment is conducted in the same facility in the previous work for R32 (Li and Hrnjak, 2019). The conditions cover mass flux from 50 to 225 kg m−2 s−1. The inlet saturation temperature is 30 °C. The two-phase flow is generated by adding heat in several steps to refrigerant initially subcooled at the entrance of test line. As vapor quality increases at a fixed mass flux, the flow is firstly plug/slug, then transitional, and finally is annular flow regime at high quality. When mass flux is 50 kg m−2 s−1, no annular flow observed in the tube. The vapor quality of boundary between two flow patterns decreases as mass flux increases. The annular flow starts at x = 0.85 (G = 100 kg m−2 s−1) and x = 0.63 (225 kg m−2 s−1). The transitional flow starts at x = 0.8 (G = 50 kg m−2 s−1) and x = 0.3 (225 kg m−2 s−1). Comparing to R32 and R134a, flow patterns of R1234ze(E) transition at lower quality due to the lower vapor density and thus higher vapor velocity. Three flow pattern maps in literature are compared to the results and they have limited agreement with our observations. The plug/slug flow behaves as homogeneous. The velocity at interface between liquid slug and vapor plug is close to the homogeneous velocity. The vapor plug length fraction also agrees to the homogeneous void fraction. The length of vapor plug increases dramatically as vapor quality increases at fixed mass flux. The length of liquid slug first increases and then unchanged as quality increases. The video supports that there are liquid droplets formed from the liquid slug and liquid ring collision.

Two-phase flow during condensation in smooth horizontal round tubes of R245fa, R1233zd(E), R1234ze(E), R134a, R32 from superheated vapor is visualized and presented in this paper. Flow regimes under different mass fluxes, heat fluxes, saturation pressures, specific enthalpies and tube sizes (1, 4, 6 mm) are identified. The paper describes to the flow regime transitions according to the visualizations. The driving force behind the annular-stratified flow transition is identified to be the force balance between shear, gravity and surface tension. The mechanism that dictates the annular-intermittent flow transition is the comparison between wave-height and the tube size. The slip ratio, which generates the Kelvin-Helmholtz instability, is considered to be the reason of transition from stratified-wavy to the fully-stratified flow. The more complicated scenarios where characteristics of different flow regimes coexist are detailed and methods for simplification are provided. The results are also compared to two different flow regime maps. The flow regime map that does not consider the non-equilibrium effects does not provide information beyond bulk quality 1 and 0. Additionally, it does not capture the annular entrance during condensation either. The flow regime map with non-equilibrium taken into account addresses issues above while having its own defects. For instance, it is highly empirical and some transition lines do not properly reflect experimental observations. A more mechanistic flow regime map is recommended.

A flow regime map for condensation from superheated vapor is proposed in this paper. The flow regime map takes the non-equilibrium effects into account by following the development of liquid film from the real onset of condensation to the end. It can be applied to both conventional tubes and microchannels. The transition mechanism between annular and stratified-wavy flow is determined to be the force balance between the shear force, gravity and surface tension. The transition mechanism from annular to intermittent flow is found to be the comparison between wave heights and tube diameter. The transition mechanism for stratified-wavy and fully-stratified is kept the same as elaborated by Xiao and Hrnjak (2017). The connections between the dimensionless number We, Fr and Bo and the transition criteria are analyzed. The flow regime map is validated by the experimental data in Xiao and Hrnjak (2017) and some current visualizations, which includes diabatic visualizations of R134a, R1234ze(E), R32, R245fa and R1233zd(E) condensing at 30 and 50 °C in 1, 4 and 6 mm tubes.

An experimental study of steam condensation in a power-plant air-cooled condenser is presented. This is the third of a four-part group of papers. The first two parts (Kang et al., 2017; Davies et al., 2017), published in the same journal present the facility, pressure drop, void fraction, and flow regime results, while this study presents heat transfer results and analysis. A follow-up paper will investigate the effect of varying inclination angle. The condenser test section is half of a flattened steel tube with brazed aluminum fins. The full size of a condenser tube is 10.72 m × 214 mm × 18 mm. The condenser tube is cut in half lengthwise and covered with a polycarbonate window to perform visualization simultaneously with the heat transfer measurements. All tests are performed with condensing pressure slightly above atmospheric. Stratified flow is found for all test conditions and all locations along the condenser, with both filmwise and dropwise condensation along the condenser wall. Steam-side heat transfer coefficient is found to depend on wall-steam temperature difference, and not quality or Reynolds number for vapor. As a result, steam-side heat transfer coefficient does not decrease along the condenser length, as is common for smaller condenser tubes with higher mass flux. This phenomenon disagrees with the predictions of many of the published correlations. Overall condenser heat transfer coefficient is found to decrease along the condenser length, due to an increase in the thickness of the stratified condensate layer.

An experimental study of convective steam condensation inside a large, inclined, flattened tube used in air-cooled condensers for power plants is presented. This is the fourth of a four-part group of papers. The first three parts (Kang et al., 2017; Davies et al., 2017, 2018) published in the same journal present the facility, pressure drop, void fraction, flow regime and heat transfer results, while this study presents the effects of inclination on heat transfer and flow regimes. The condenser is a flattened steel tube with brazed aluminum fins. The full tube has dimensions 10.72 m × 214 mm × 18 mm. The condenser tube is cut in half lengthwise and covered with a polycarbonate window to perform visualization simultaneously with the heat transfer measurements. The steam is condensed at atmospheric pressure, and cooled by air at a uniform velocity profile. The angle of inclination is varied from horizontal (0°) to 75° downward. Condenser performance is also predicted with a model. The majority of the condenser is found to be in the stratified flow regime for all inclinations tested, with only a short annular section at the inlet of the condenser. The tubes inclined greater than 60° are also found to have stratified-wavy flow near the condenser outlet. Overall condenser U is found to increase with increasing downward inclination angle of the condenser, with a maximum increase of approximately 4% at 75° inclination. This improvement is found to be the result of improved drainage and increased void fraction near the condenser outlet. Mean steam-side heat transfer coefficient 1 (HTC) is found to remain constant along the tube, and for the entire condenser, with changes in tube inclination angle. Commonly-used inclined condensation HTC correlations are found to underpredict the magnitude of the experimentally-determined steam-side HTC.

This paper presents the heat transfer coefficient, pressure gradient, and flow pattern of R1234yf in a microchannel tube. Both heat transfer coefficient and pressure gradient are presented against real saturation pressure, while flow pattern captures at the exit of data points are presented in the same plot. The experiment was conducted on a 24-port microchannel tube with an average hydraulic diameter of 0.643 mm. The experiment covers mass flux from 100 to 200 kg m−2s−1, heat flux from 0 to 6 kW m−2, vapor quality from 0 to 1, and inlet saturation temperature from 10 to 30 °C. Comparing the correlations to the HTC measurements at very low quality (about 0.1), Gorenflo, D., and Kenning, D. (2010, Pool Boiling, in: VDI Heat Atlas, 2nd ed, Springer, pp. 757–788) agree with the results. As vapor quality increases, pressure gradient increases. The adiabatic pressure gradient is a strong function of mass flux and saturation pressure (temperature). Flow patterns of R1234yf are also affected by mass flux and saturation pressure. The heat transfer coefficient is a strong function of mass flux and heat flux. The saturation temperature has a smaller effect on HTC in the condition range (10 – 30 °C). Under the test range, the accelerating pressure drop is insignificant compared to friction. Comparing to the results, Mishima, K., and Hibiki, T. (1996, “Some Characteristics of Air-Water Two-Phase Flow in Small Diameter Vertical Tubes,” Int. J. Multiph. Flow, 22(4), pp. 703–712) and Muller-Steinhagen, H., and Heck, K. (1986, “A Simple Friction Pressure Drop Correlation for Two-Phase Flow in Pipes,” Accessed March 1, 2018)., 20, pp. 297–308.) have small mean absolute error (MAE) to predict local pressure gradient. For the heat transfer coefficient, Sun, L., and Mishima, K. (2009, “An Evaluation of Prediction Methods for Saturated Flow Boiling Heat Transfer in Mini-Channels,” Int. J. Heat Mass Transf, 52(23–24), pp. 5323–5329) and Gungor, K. E., and Winterton, R. H. S. (1986, “A General Correlation for Flow Boiling in Tubes and Annuli,” Int. J. Heat Mass Transf, 29(3), pp. 351–358) have an MAE less than 30%.

This paper presents heat transfer coefficient and pressure drop of R32 measured in the same facility as introduced in Li and Hrnjak (2017). Experiments are conducted on a 24-port microchannel tube with a hydraulic diameter of 0.643 mm. The pressure drop of R32 is lower than R134a, and the heat transfer coefficient is higher at the same condition. The pressure drop increases with quality and mass flux but decreases with reduction of saturation pressure. R32 heat transfer coefficient increases when heat or mass flux rises. Heat transfer coefficient of R32 first increases when quality increases due to convective effects and then drops at moderate quality due to the absence of nucleate boiling and dry-out. A comparison of adiabatic pressure drop and heat transfer coefficient to correlations has been made. Based on the Blasius equation and homogeneous density, the viscosity model from Cicchitti et al. (1960) has the best fit to predict two-phase pressure drop. Kim and Mudawar (2012) has the smallest MAE to predict pressure drop. For heat transfer coefficient, Sun and Mishima (2009) is the best fit. Further research will be on expanding the database and understanding hydraulic and boiling behavior.

A mechanistic pressure drop model is proposed in this paper for condensation in horizontal smooth round tubes in order to account for the non-equilibrium effects. The model makes use of a flow regime map and void fraction correlation as well as a mechanistic heat transfer model that are all developed for condensation of superheated vapor in a vapor-compression system. The model provide seamless transition between single-phase and two-phase regions including the superheated, condensing-superheated, two-phase, condensing-subcooled and subcooled regions. Diabatic flow visualizations are used to analyze the effects on pressure drop from the formation of waves. An enhancement factor to represent the frequency and magnitude of the waves is established using Kelvin-Helmholtz and Rayleigh-Taylor instability. The two-phase pressure drop is modeled based on the single-phase pressure drop correlations, the flow regimes, void fractions as well as the enhancement factor. Data obtained from R134a, R32, R1234ze(E), R1233zd(E) and R245fa with mass fluxes from 100 kg m−2 s−1 to 400 kg m−2 s−1 and heat fluxes from 5 kW m−2 to 15 kW m−2 inside two different tubes of 4 and 6 mm are used to validate the model.

Pressure drop during condensation from superheated vapor is reported in this paper. Three different refrigerants including R134a, R32 and R1233zd(E) are evaluated in two horizontal round copper tubes with 6.1 and 4.0 mm inner diameters. The test conditions cover a range of mass fluxes from 100 kg m−2 s−1 to 400 kg m−2 s−1 and heat fluxes from 5 kW m−2 to 15 kW m−2. The condensing pressures are selected to be those that correspond to 30 °C. The measurements are all conducted with diabatic flows which start in the superheated (SH) region to observe how the onset of condensation affects the pressure drop. The visualizations of condensation are compared with the measured pressure drop, suggesting the pressure drop behaviors in different regions during the condensation process are different due to different flow regimes. The onsets of condensation all happens in the superheated region before specific enthalpy drops to that of bulk quality 1. The pressure drop depends on the mass flux, specific enthalpy and tube diameter. The result also shows that several thermal properties of the refrigerants such as the density ratio and viscosity have an effect on the pressure drop, which can be explained by the generation of waves and the differences in vapor and liquid velocity. A peak in the pressure drop curve is consistently present between the onset and end of condensation around bulk quality 1, which could be a result of the competition between increasing liquid-wave strength and decreasing flow velocity as condensation proceeds after the onset of liquid film. Since none of the current existing correlations are sufficient in accounting for the different mechanisms in different stages of the condensation process, a pressure drop model that is specifically developed for condensation of superheated vapor by taking non-equilibrium into consideration is in need.

An experimental study of convective steam condensation inside a large, inclined, flattened-tube air-cooled condenser for power plants is presented. This is the second of a four-part group of papers. The first part presents pressure drop and visualization results, while this study presents the experimental method along with heat transfer results. Follow-up papers present further heat transfer results and the effect of inclination. The condenser in this study is steel with brazed aluminum fins. The condenser measures 10.72 m in length, with a cross section of 214 mm × 16 mm. The condenser tube was cut in half lengthwise and covered with a polycarbonate viewing window in order to provide visualization access simultaneously with the heat transfer measurements. Inlet steam mass flux ranged from 6.2 to 9.5 kg m−2 s−1, and condenser capacity varied from 25 to 31 kW. The angle of inclination was varied from horizontal to 75° downward. The experiments were performed with a uniform fin-face velocity of crossflowing air at 2.2 m/s. Condenser capacity was found to increase linearly with increasing downward inclination angle of the condenser, at a rate of 0.041% per degree of inclination below horizontal. This improvement was found to be the result of improved drainage and increased void fraction near the condenser outlet.

The amount and frequency of reverse flow in microchannel evaporators largely depends on the thermophysical properties of the working fluid. In this study, four refrigerants (R134a, R1234yf, R245fa and R32) are examined in the same air conditioning system operated under flash gas bypass mode with venting of reversed vapor (to provide better control of the flow). The flow rate and frequency of reversed vapor are measured for each refrigerant. It has been found that fluids with lower heat of vaporization and higher specific volume difference between vapor and liquid phase) tend to generate more reversed vapor flow. Experimental results validate qualitatively and in most cases quantitatively the mechanistic model presented in Li and Hrnjak (2017a). The validated model is used to demonstrate the effects of thermophysical properties on flow reversal for nine widely used refrigerants.

The objective of this paper is to present the influence of straight micro-fins on heat transfer coefficient and pressure drop. R410A flow boiling experiments were conducted in both smooth tube and the axial micro-finned tube. Data are taken at 10 °C saturation temperature for the vapor quality from 0.1 to 0.9, mass flux from 100 to 450 kg/s-m2, and heat flux from 10 to 20 kW/m2. The results in the smooth tube were compared with several correlations and used as a baseline. It was observed that straight micro-fin structure played an important role in the evaporative heat transfer. The heat transfer coefficient was significantly enhanced in the axial micro-finned tube, whose enhancement factor ranged from 1.03 to 1.48 with average of 1.34. As the mass flux and vapor quality was low, the straight micro-fins had stronger influence of the heat transfer coefficient since the liquid-phase refrigerant was easily trapped in the grooves of the axial micro-finned tube. The pressure drop penalty factor of the axial micro-finned tube ranged from 0.66 to 1.6, and the average was around 1.23.

The void fraction in headers of microchannel heat exchangers (MCHEs) affects the refrigerant charge, distribution, pressure drop, and heat transfer coefficient of MCHEs. However, due to the complex geometries of headers, experimental measurements of void fraction in headers with refrigerant are still missing in the open literature. A model that predicts local void fraction between microchannel tubes in headers is also widely needed when predicting the refrigerant charge and the performance of MCHEs. This paper first presents experimentally measured void fraction between microchannel tubes in a intermediate (five inlets and five outlets) and inlet headers (ten or five outlets) with R134a by calibrated capacitive sensors. For the intermediate header, the ranges of inlet mass flow rate and quality are 3.2 g s−1 to 5.2 g s−1 and 0.2 to 0.8, respectively. The inlet mass flow rates vary from 3.2 g s−1 to 7.2 g s−1 and qualities range from 0.1 to 0.82 for the inlet headers. As the inlet quality increases, the average void fraction in headers becomes larger. Other local parameters in the headers, such as mass flux, quality, superficial velocity together with the visualization results are also obtained. Void fraction shows a strong relation to the local quality in headers and the refrigerant distribution in microchannel tubes. A drift-flux void fraction model based on the local variables is proposed in this paper. The local void fraction predicted by the model agrees well with the experimentally measured values, with most of the predictions falling into ± 15% deviations. The proposed model has better accuracy when predicting void fraction in headers comparing to some widely used correlations.

Most of the previous research on flash gas bypass (FGB) focused on performance improvement in steady state and demonstrated that compared to direct expansion mode (DX), FGB mode have better performance. However, the control strategy of flash gas bypass system and dynamic behavior during start-ups and transients were not yet clearly defined and investigated. In this paper, a novel control strategy has been proposed for an automobile air conditioning system operating in flash gas bypass mode with R134a as the refrigerant. The proposed control strategy utilized an electronic expansion valve (EV) for the control of subcooling from condenser outlet and a bypass valve (BV) for superheat from compressor inlet. Both start-up and transient system behaviors were studied. The experimental results showed that the proposed cycle control strategy was found to be able to provide reliable control to the system. In addition, proper sizing of bypass valve and flash gas bypass tank have also been studied.

Heat transfer experiments with flow visualization were performed with R134a in a smooth horizontal tube in order to verify that condensation occurs outside the two-phase region. The visualization performed for this work is unusual in that a diabatic visualization section was used, more closely approximating the behavior of the fluid in an actual condenser than typical adiabatic visualization experiments. In addition to the flow visualization, liquid film thickness measurements were taken and used to determine the cumulative latent heat transferred from the refrigerant. The measured film thicknesses and recorded videos confirm the presence of liquid in the bulk superheated region. Condensate appeared first as droplets and rivulets on the tube wall and then formed a thin film. The flow later passed through misty annular and wavy flow regimes as the refrigerant transitioned to the two-phase region.

Boiling instability and periodic flow reversal are commonly seen in microchannel heat exchangers, but have been studied in air conditioning systems only recently. The amount and frequency of reverse flow largely depends on the geometry of the channels. This paper presents results obtained in a R134a automotive air conditioning system operated under flash gas bypass mode with venting of reversed vapor (to provide better control of the flow). Results for multiple microchannel evaporators with different channel diameters and lengths are presented. The mass flow rate and frequency of reversed vapor are measured for each geometry. Meanwhile, a mechanistic model is developed to simulate the bubble dynamics inside of a single microchannel tube. This model is validated against experimental data and employed to explain the effect of channel length and diameter on flow reversal. Both experimental and simulation results show that larger channel diameter and longer channel length result in less flow reversal with a lower frequency.

A new facility has been built for experimental study on microchannel and introduced in this paper. Six heat transfer coefficient and six pressure drop measurements can be conducted simultaneously in one pass from vapor quality of 0 to 1 on the facility. Two water-jacket-design heat exchangers are placed top and bottom on each test section for heating the refrigerant. In each test section, the top and bottom sides are heated and controlled separately to ensure uniform conditions. The environmental heat loss and wall temperature measurements are corrected to increases confidence of results. Port geometry is measured with an optical microscope. Heat transfer coefficient and pressure drop of R134a are measured in a 24-port microchannel tube. The single-phase pressure drop follows the rule of round tube and f Re is 64. The single-phase Nusselt number is more sensitive to Reynolds number, compared to large tubes. As vapor quality increases, two-phase pressure drop increases until quality of 0.9, and then stops increasing or even drops; two-phase heat transfer coefficient increases until quality of 0.5 to 0.6 and then drops. Higher mass flux or lower saturation temperature will increase the pressure drop. Heat flux and mass flux have strong impact on heat transfer coefficient, and the effect of saturation temperature is not clear.

This paper presents the air-side pressure drop and overall heat transfer coefficient characteristics for serpentine-louvered-fin, microchannel heat exchanger in periodic frosting. It focuses on quantification of the effects of louver angle on heat transfer and pressure drop and on defrost and refrost times. Nine heat exchangers differing in louver angle and fin pitch (i.e. louver angle 15° to 39° and fin pitch 12 to 18 fpi) are studied. The face velocity was 3.5 m s −1 and inlet air relative humidity of 70% and 80%. The major finding of this paper is seen in the proof that it is necessary to perform experiments in several consecutive cycles to be able to understand the situation in operation of the heat exchangers and that geometry that is the best in first frosting test does not stay the best in real operation where multiple consecutive frosting occurs.

This paper presents an experimental study of the inclination effect on pressure drop and flow regime during condensation of steam in a large flattened tube used in air-cooled condensers (ACC) for power plants. Steam with mass flux of about 7 kg m −2 s −1 was condensed inside a 10.7 m long, flattened test tube with inclination angle varied from horizontal up to 70°. The original full-sized steel tube was cut in half along the centerline, and the removed part was replaced by a polycarbonate window to enable simultaneous flow visualization in situ with heat transfer and pressure drop measurements. A uniform velocity profile of 2.03 ± 0.12 m s −1 was imposed on the air side to extract heat from the steam in a cross flow direction. The experimental results showed that increasing the inclination angle led to reductions of pressure drop due to the improvement in the gravity-assisted drainage of condensate inside the test tube. At such low mass fluxes, tube inclination significantly influenced the flow pattern which was observed to be a well-separated stratified flow throughout the tube at all downward inclination angles. The separated flow pattern enabled the direct measurement of void fraction, and the traditional void fraction models using the newly-defined superficial quality successfully predicted the measurements within ±10%. The experimental data were converted to reflect pressure drop in a full tube based on the model that was developed to account for the differences in tube geometry between the full and test tube at the same operating condition. A prediction of pressure drop performance of the same steam condensing system under vacuum condition was also discussed. The negative dependence of total pressure drop on inclination angle also prevailed in both converted results in atmospheric condition and the predicted ones in vacuum condition.

A heat transfer model is proposed in this paper for condensation of annular and stratified flow in horizontal smooth round tubes accounting for the non-equilibrium effects. The model is based on a non-equilibrium flow regime map and void fraction correlation that is developed from diabatic flow visualization and film thickness measurement. The model unifies single-phase and two-phase heat transfer models into one continuous function throughout the superheated, condensing-superheated, two-phase, condensing-subcooled and subcooled regions with seamless transition in between. Film heat transfer coefficient based on the interfacial temperature of the flow is used as a tool for the modeling in the presence of two-phase flow, which is later converted back to heat transfer coefficient based on the bulk temperature of the flow. The two-phase flow model is developed for the annular and stratified flow. The effects of interfacial waviness, liquid entrainment, wall subcooling, gravity, tube diameter as well as non-equilibrium are discussed in separated sections. Data obtained from different refrigerants and working conditions are used for the validation of the model.

This paper presents experimental evaluation of a thermosiphon loop in two systems: conventional and with separation. Thermosiphon with separation is a modification of the conventional system by allowing vapor only to enter condenser, and thus is expected to provide better operation than conventional baseline. The effects of different operating conditions on mass flow rate, height of liquid, pressure drop and temperature difference of two systems in steady state, as well as the startup features of two systems were presented. The system with separation showed improved operation in startup. The conventional system took longer time to stabilize with stronger oscillations of mass flow rate at startup than system with separation. The temperature difference of the system with separation also had a small improvement when heat load was low in steady state. Oversized condenser is identified as the reason why separation did not always realized full potential to improve the performance of the system in steady state. Reducing size of the condenser is identified and the right action and it will be verified in a next step.

This paper introduces the concept of separation of two-phase flow in condensers and discusses its possible application of enhancing the heat transfer performance by capitalizing on the high local heat transfer coefficient of vapor flow. The benefit of vapor–liquid refrigerant separation and the reason why it will improve the condenser performance are explained. Numerical studies are performed on an R-134a microchannel condenser. Model predicts that at the same mass flow rate, the exit temperature is lower by 1.3 K in the separation condenser than in the baseline condenser while the difference of pressure drop remains within 2%. 6.1% more flow rate of condensate is predicted in the separation condenser as another comparison criterion. In addition, the trade-off between high quality and low mass flux for the vapor path downstream of the separation header is investigated by the model and results are presented. Modeling is conducted with pre-assumed separation efficiency in the header. The real value requires further investigation.