Falling Films on Arrays of Horizontal Tubes with R-134a, Part I: Boiling Heat Transfer Results for Four Types of Tubes

A new falling film heat transfer test facility has been built for the measurement of local heat transfer coefficients on a vertical array of horizontal tubes, including flow visualization capabilities, for use with refrigerants. Presently, the facility has been used for evaporation tests on four types of tubes at three tube pitches and three nominal heat flux levels for R-134a at 5°C. A new method for determining local heat transfer coefficients using hot water heating has been applied, and test results for a wide range of liquid film Reynolds numbers have been measured for arrays made of plain, Turbo-BII HP, Gewa-B, and High-Flux tubes. The results show that there is a transition to partial dryout as the film Reynolds number is reduced, marked by a sharp falloff in heat transfer. Above this transition, the heat transfer coefficients are nearly insensitive to the film Reynolds number, apparently because vigorous nucleate boiling is always seen in the liquid film. The corresponding nucleate pool boiling data for the four types of tubes were also measured for direct comparison purposes. Overall, about 15,000 local heat transfer data points were obtained in this study as a function of heat flux, film Reynolds number, tube spacing, and type.

Two-phase operational maps, pressure drop, and heat transfer for flow boiling of R236fa in a micro-pin fin evaporator

Navid Borhani, Chiara Falsetti, Hamideh Jafarpoorchekab, Mirco Magnini, John Richard Thome

Cooling the new generation of 3D high power electronic chips is one of the leading challenges in microelectronics, as it is a key to achieve high computational performance at lower cooling system power consumption, thus reducing the operating cost. Two-phase flow boiling in two micro-pin fin heat sinks is studied here for this cooling process. Each micro-evaporator has a heated area of 1 cm(2) and contains 66 rows of cylindrical micro-pin fins with an in-line configuration and diameter, height and pitch of respectively 50 mu m,100 gm and 91.7 mu m. The fluid tested is refrigerant R236fa. Channel entrances with and without inlet restrictions are tested in order to evaluate their relative effect on the stability of the flow. The inlet restrictions consist of an extra row of micro-pin fins with a larger diameter of 100 mu m placed at the inlet of the heated area. The present study investigates operational stable and unstable flow regimes, pressure drop and heat transfer performance for the two micro-evaporators (one with and one without inlet restrictions) by coupling high speed visualization of the micro-pin fins flow area with fine resolution infrared temperature measurements of the micro-evaporator base, which yields a 2D map of local heat transfer coefficients of the whole heated area. Working conditions tested have mass flux varying from 500 kg m(-2) S-1 to 2500 kg m(-2) S-1, heat flux ranging from 20 W cm(-2) to 48 W cm(-2), and a constant outlet saturation temperature of 30.5 degrees C. In agreement with previous studies for multi-microchannel evaporators, it is here observed that inlet restrictions extend the map of stable operational regimes, in particular toward lower values of the mass flux, without any appreciable increase of the pressure drop. Unlike evaporation through parallel microchannels, the heat transfer coefficient trends and magnitudes vary dramatically with the flow conditions (mass flux and heat flux), thus suggesting that micro-pin fins have a strong impact on the two-phase flow pattern development, in particular delaying the transition to annular flow. (C) 2016 Elsevier Ltd. All rights reserved.

Pergamon-Elsevier Science Ltd2017

Contribution to the heat transfer analysis of natural and substitute refrigerants evaporated in a horizontal tube

Olivier Zürcher

The substitution of CFC refrigerants in refrigeration systems, heat pumps and organic Rankine cycles, requests a good knowledge of the heat transfer and pressure drop properties of substitute fluids. A contribution to this international effort is proposed with the study of two hydrofluorocarbon refrigerants (HFC-134a and the zeotropic mixture HFC-407C) and the study of the natural refrigerant ammonia. The HFCs have been experimentally tested on the first test rig developed in the Laboratory of Industrial Energy Systems (LENI) which was substantially modified to cope with ammonia, both in terms of safety requirements and operational conditions. The experimental test section is composed of two concentric tubes, with in tube evaporation of the refrigerant inside the inner tube of counter-current Annular water heating. A new database of the local heat transfer coefficients for the refrigerants HFC-134a, HFC-407C and ammonia together with pressure drop measurements has been collected and used to define and calibrate a new and more general heat transfer model. The two HFC refrigerants have been tested on the test rig developed by Kattan [36]; due to the chemical and the thermophysical properties of ammonia, a new test rig has been designed for the ammonia tests and a new calculation procedure based on the mean temperature measurement on the tube wall has been used. The experimental database covers evaporation tests on plain and on microfin tubes, and with nominal oil concentrations varying from 0 to 5 [%] (by wt.). Detailed modeling is concentrated on the tests on plain tube without oil. Each of the three refrigerants has been evaporated at 4[°C]. The other experimental conditions for the HFCs refrigerants were an inside tube diameter of 10.92[mm], a heat flux range of 2 – 5 [kW/m2] and three mass velocities of G = 100, 200, 300 [kg/(m2s)] visualized as Stratified-Wavy flows (G = 100[kg/(m2s)]), and mainly Annular for the higher mass velocities (G = 200, 300[kg/(m2s)]). The other global conditions for ammonia were an internal tube diameter of 14.00 [mm], a heat flux range of 5 – 70[kW/m2] and eleven mass velocities of G = 10,20,30,40,45,50,55,60,80,120,140[kg/(m2s)], corresponding to Stratified, Stratified-Wavy, Intermittent and Annular flow patterns; complementary tests, varying the mass velocity at constant vapor qualities (x = 20,50,80[%]), have been made in particular to better characterize boundaries between flow patterns. The flow patterns are visualized through glass sections at both ends of the 3.1 [m] long test tube. An extensive review of void fraction models with sensitivity analysis on the actual available flow pattern map has been made. From the 17 correlations reviewed, the models of Taitel-Dukler [71] and Rouhani [79] have been used. Other improvements to the threshold criteria of the Stratified-Wavy to Annular transition, lead to an accurate diabatic map to predict the flow patterns and their transition for very different families of fluids like HFC's and ammonia. A new approach in the prediction of two-phase flow heat transfer has been proposed through the study of each flow pattern separately, according to a new criterion defining the onset of nucleate boiling as a function of the critical convective heat transfer coefficient representative of the location where nucleate boiling might occur. A function based on a pseudo Biot number allows, from the mean heat flux around the tube periphery, to base the model on two different mean heat fluxes applied respectively to the parts of the perimeter in contact with the liquid and the vapor in stratified types of flow. Considering pure convective heat transfer, or mixed convective and nucleate heat transfer, this division allows the use of a common criterion to be applied to each flow pattern. The two-phase flow heat transfer coefficient has finally been obtained as a weighted mean function of the vapor and the liquid heat transfer coefficient with respect to their contact surface with the heated tube. Based on the database of refrigerants HFC-134a and ammonia, the standard deviation of the new heat transfer model is of σ = 27.9[%]. Even if the database showed that the flow conditions were close to, or in the turbulent to laminar flow transition, and even if the major part of the experimental points were purposely obtained close to the various flow pattern transitions, the new model showed very good agreement with the experimental database. Due to the precision of the new flow pattern map and the effectiveness of the onset on nucleate boiling criterion, this new heat transfer model accurately predicts the heat transfer conditions during evaporation. Finally, a new method to model separated flows is proposed, based on partial hydraulic diameters and a mean interface velocity.

EPFL2000

Experimental and analytical investigation of void fraction and heat transfer during evaporation in horizontal tubes

Leszek Wojtan

A new optical void fraction measurement system has been coupled to the existing LTCM flow boiling test facility to obtain dynamic and time-averaged void fractions, simultaneously with measurements of the local heat transfer coefficients. A series of evaporation tests have been run for R-22 and R-410A in 13.84 mm and 8.00 mm ID tubes. Using our newly developed image processing system for processing laser illuminated cross-sectional views of the flow, about 310 000 images have been analysed in this study to provide the same number of dynamic void fraction measurements. From these images, 238 and 87 time-averaged void fraction values have been obtained for the 13.60 mm and 8.00 mm diameter glass tubes, respectively. The same number of time-averaged dry angles has been obtained. The measured void fractions have been compared to the principal prediction models, showing good agreement to the Steiner version of Rouhani-Axelsson drift flux model. Based on analysis of the void fraction evolution as a function of time, several modifications of the Kattan et al. flow pattern map [30] (1998a) have been made to improve the heat transfer prediction model in stratified-wavy flow and to extend its application to vapor qualities below 0.15. Over 1250 new flow boiling heat transfer points have been acquired at mass velocities from 70 to 700 kg/m2s and heat fluxes from 2.0 to 57.5 kW/m2. Based on the 368 heat transfer points obtained in dryout and mist flow conditions, new boundaries for the transitions (from/to) annular/dryout and dryout/mist flow have been defined and integrated into the flow pattern map of Kattan et al. A new heat transfer model for dryout and mist flow conditions has also been proposed, extending the flow pattern oriented model of Kattan et al. to these flow regimes.