See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/37421723

Experimental and analytical study of two-phase pressure drops during

evaporation in horizontal tubes

Article · January 2005

DOI: 10.5075/epfl-thesis-3337 · Source: OAI

CITATIONS READS

47 15,192

1 author:

Jesús Moreno Quibén

Wolverine Tube, Inc.
22 PUBLICATIONS   818 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Jesús Moreno Quibén on 22 May 2014.

The user has requested enhancement of the downloaded file.

EXPERIMENTAL AND ANALYTICAL STUDY OF
TWO-PHASE PRESSURE DROPS DURING
EVAPORATION IN HORIZONTAL TUBES

THÈSE NO 3337 (2005)

PRÉSENTÉE À LA FACULTÉ SCIENCES ET TECHNIQUES DE L'INGÉNIEUR

Institut des sciences de l'énergie

SECTION DE GÉNIE MÉCANIQUE

ÉCOLE POLYTECHNIQUE FÉDÉRALE DE LAUSANNE

POUR L'OBTENTION DU GRADE DE DOCTEUR ÈS SCIENCES

PAR

Jesus MORENO QUIBEN

DEA - Dynamique des Fluides et des Transferts, Université Pierre et Marie Curie, Paris, France
et de nationalité espagnole

acceptée sur proposition du jury:

Prof. J. Thome, directeur de thèse

Prof. J. Corberan Salvador, rapporteur
Prof. D. Favrat, rapporteur
Prof. P. von Rohr, rapporteur

Lausanne, EPFL
2005
2
Acknowledgements

This study has been carried out at the Laboratory of Heat and Mass Transfer (LTCM), Swiss
Federal Institute of Technology Lausanne (EPFL), under the direction of Prof. John R. Thome.
The project has been supported financially by the Fond National Swiss (FNS) contract number
21-57210.99 and by the Air-Conditioning and Refrigeration Technology Institute (ARTI) contract
number 605-20040, which are gratefully acknowledged.
I would like to thank Prof. John R. Thome for his advice and for giving me the opportunity and
motivation to undertake this work. He was all the time available for fruitful research discussions
and his guidance and encouragement during this period were a benefical contribution to this study.
Finally, I appreciate his total confidence and the responsibility granted me throughout this period.
I thank Prof. Daniel Favrat of the Laboratory of Industrial Energy Systems, Swiss Federal Institute
of Technology Lausanne (EPFL), Prof. Philipp Rudolf von Rohr of the Laboratory for Transport
Processes and Reactions, Swiss Federal Institute of Technology Zürich (ETH) and Prof. José Miguel
Corberán of the Institute for Energy Engineering, Universidad Politécnica de Valencia (UPV) for
being the examiners of this thesis.
Special thanks to my colleagues of the LTCM for their cooperation and friendship. Among them, my
special gratitude to Dr. Leszek Wojtan for his constant helpfulness and productive collaboration.
I also acknowledge our technicians Laurent Chevalley and Alfred Thomas who have skillfully par-
ticipated to the construction and maintenance of the experimental facility.
Finally, many thanks to Valeria and Soledad for being comprehensive in the time involved in this
work.

3
4
Abstract

Two-phase flow of gases and liquids or vapors and liquids in pipes, channels, equipment, etc. is
frequently encountered in industry and has been studied intensively for many years. The reliable
prediction of pressure drop in two-phase flow is thereby an important aim. Because of the com-
plexity of these types of flow, empirical or semiempirical relationships are only of limited reliability
and pressure drops predicted using leading methods may differ by up to 100%. In order to improve
prediction methods, this work presents an experimental and analytical investigation of two-phase
pressure drops during evaporation in horizontal tubes. The goal of the experimental part was
to obtain accurate two-phase pressure drop values over a wide range of experimental conditions.
The existing LTCM intube refrigerant test loop has been modified and adapted to the new test
conditions and measurement methods. Two new test sections have been also implemented into
the modified test rig. The new test section consists of two zones: diabatic and adiabatic. This
configuration allows tests to be run that obtain experimental two-phase pressure drop values un-
der diabatic and adiabatic conditions simultaneously. The experimental campaign acquired 2543
experimental two-phase pressure drop values. Based on a comprehensive state-of-the-art review
and comparison with two-phase frictional pressure drop prediction methods, it is proven that none
of these methods were able to accurately, reliably predict the present experimental values. In the
second part of this work, an analytical study was undertaken in order to develop a new two-phase
frictional prediction method. It has been shown in the literature that the so called ”phenomenolog-
ical approach” tends to provide more accurate and realistic predictions as the interfacial structure
between the phases is taken into account. Based on that, a phenomenological flow pattern approach
was chosen in the present study. The recent Wojtan-Ursenbacher-Thome [155] map was chosen to
provide the corresponding interfacial structure. The new model treats each flow regime (i.e. inter-
facial structure) separately and then ensures a smooth transition in between, being in agreement
with the experimental observations. Another important feature of the proposed model is that it
matches the correct limits at x = 0 (single-phase liquid flow) and x = 1 (single-phase gas flow).
Based on a statistically comparison, it is concluded that the new two-phase frictional pressure drop
model based on flow pattern map successfully predicts the new experimental data. The present
work completes the fourth basic step in LTCM’s flow pattern based work on two-phase flow and
heat transfer inside horizontal round tubes: (i) generalized flow pattern map, (ii) flow boiling heat
transfer model, (iii) convective condensation model and (iv) two-phase frictional pressure drop
model.

5
6
Résumé

Les écoulements biphasiques liquide/gaz ou liquide/vapeur en tubes, canaux ou dans différentes

géométries sont un problème fréquemment rencontré dans les applications industrielles et ont été
largement étudiés ces dernières années. De par leur importance pratique, la prédiction des pertes de
charges des écoulements biphasiques doit être précise. La complexité de ces types d’écoulements fait
que les relations empiriques ou semi-empiriques usuelles sont peu précises et leurs prédictions peu-
vent différer parfois de 100%. Ce travail présente une investigation expérimentale et analytique des
pertes de charges biphasiques durant l’évaporation en tubes horizontaux en vue d’en améliorer les
méthodes de prédiction. La campagne expérimentale a permis d’obtenir une base de données élargie
et fiable de pertes de charges biphasiques. Une boucle de test existante au LTCM pour l’étude des
réfrigérants dans des tubes a été modifiée et adaptée aux nouvelles conditions de test et méthodes
de mesures. Deux nouvelles sections de tests ont été implantées dans la boucle modifiée. Elles
sont composées de deux zones : l’une adiabatique et l’autre non adiabatique. Cette configuration
permet d’étudier simultanément les pertes de charges biphasiques en zone adiabatique et en zone
non adiabatique. La campagne expérimentale a permis d’obtenir 2543 valeurs de pertes de charges
biphasiques. Une étude bibliographique approfondie et une comparaison avec différentes méthodes
de prédiction de pertes de charges biphasiques ont montré qu’aucune de ces méthodes ne permettait
une prédiction précise et fiable des ces résultats expérimentaux. Dans la seconde partie de ce tra-
vail, une étude analytique a été réalisée afin de développer une nouvelle méthode de prédiction des
pertes de charges biphasiques par frottement. Une approche phénoménologique a été adoptée dans
cette étude car il a été démontré dans la littérature qu’elle permet des prédictions plus réalistes
et plus précises en prenant en compte la structure de l’interface entre les phases. Cette structure
de l’interface entre les phases a été obtenue en se basant sur la carte d’écoulement de Wojtan-
Ursenbacher-Thome [155]. Ainsi le nouveau modèle traite chaque type d’écoulement séparément
et assure également des transitions correctes, en accord avec les observations expérimentales. Une
autre innovation importante de ce nouveau modèle est qu’il prend correctement en compte les deux
limites a x = 0 (écoulement monophasique liquide) et x = 1 (écoulement monophasique gazeux).
Une étude statistique a permis de conclure que ce nouveau modèle basé sur les cartes d’écoulements
prédit avec succès les résultats expérimentaux. Ainsi cette étude complète la démarche en 4 étapes
du LTCM concernant les écoulements biphasiques et les transferts de chaleur internes dans des
tubes horizontaux circulaires: (i) carte d’écoulement non adiabatique généralisée, (ii) modèle de
transfert de chaleur en ébullition, (iii) modèle de transfert de chaleur en condensation et (iv) modèle
de pertes de charges biphasiques.

7
8
Contents

Acknowledgements 3

Abstract 5

Résumé 7

1 Introduction 19

2 Fundamental Definitions in Two-Phase Flow 21

2.1 Two-phase flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2 Vapor quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3 Void fraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.4 Velocities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.5 Definition of non-dimensional numbers . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.6 Basic equations of two-phase flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.6.1 Conservation of mass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.6.2 Conservation of momentum . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.6.3 Conservation of energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3 Overview of Flow Pattern Maps 35

3.1 Flow patterns in horizontal flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Flow pattern maps in horizontal flow . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.2.1 Baker flow pattern map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.2.2 Mandhane et al. flow pattern map . . . . . . . . . . . . . . . . . . . . . . . . 38
3.2.3 Taitel and Dukler flow pattern map . . . . . . . . . . . . . . . . . . . . . . . 39
3.2.4 Hashizume flow pattern map . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.2.5 Kattan-Thome-Favrat flow pattern map . . . . . . . . . . . . . . . . . . . . . 42

9
10 CONTENTS

3.2.6 Thome and El Hajal flow pattern map . . . . . . . . . . . . . . . . . . . . . . 46

3.2.7 Wojtan-Ursenbacher-Thome flow pattern map . . . . . . . . . . . . . . . . . . 48
3.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

4 Two-Phase Pressure Drop Models 53

4.1 Two-phase pressure drops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.2 Empirical methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.2.1 Lockhart-Martinelli [90] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.2.2 Bankoff [9] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.2.3 Cicchitti et al. [34] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.2.4 Thom [129] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.2.5 Pierre [107] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.2.6 Baroczy [11] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.2.7 Chawla [27] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.2.8 Chisholm [31] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.2.9 Friedel [46] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.2.10 Grönnerud [53] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.2.11 Müller-Steinhagen and Heck [97] . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.3 Analytical methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.4 Phenomenological methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.4.1 Bandel [8] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.4.2 Beattie and Whalley [14] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.4.3 Hashizume et al. [61] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.4.4 Olujić [104] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.4.5 ARS Model [58] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

5 Description of Experimental Test Facility 75

5.1 General description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2 Refrigerant circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.3 Hot water circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.4 Test sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.5 Experimental procedure and data acquisition . . . . . . . . . . . . . . . . . . . . . . 80
CONTENTS 11

5.6 Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.6.1 Pressure drop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.6.2 Refrigerant temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.6.3 Heat Flux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.6.4 Vapor quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.7 Accuracy of measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.7.1 Pressure drop measurement accuracy . . . . . . . . . . . . . . . . . . . . . . . 82
5.7.2 Heat flux measurement accuracy . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.7.3 Vapor quality measurement accuracy . . . . . . . . . . . . . . . . . . . . . . . 83
5.8 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

6 Experimental Results 85
6.1 Reduction of experimental data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
6.2 Two-phase pressure drop measurements . . . . . . . . . . . . . . . . . . . . . . . . . 89
6.2.1 Some comparisons for different experimental parameters . . . . . . . . . . . . 89
6.2.2 Results for R134a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
6.2.3 Result for R22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
6.2.4 Result for R410A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

7 New Two-Phase Frictional Pressure Drop Model Based on Flow Pattern Map 109
7.1 Comparison to existing methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
7.2 New two-phase pressure drop model . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
7.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
7.2.2 Segregation of experimental data . . . . . . . . . . . . . . . . . . . . . . . . . 116
7.2.3 Model development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
7.2.4 Comparisons of experimental to predicted values . . . . . . . . . . . . . . . . 125
7.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

8 Conclusions 137

A Fluid Physical Properties 139

B Liquid-Vapor Interfaces 141

12 CONTENTS

Bibliography 142

Nomenclature 155

Curriculum Vitae 159

List of Figures

2.1 Cross-sectional void fraction (
). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

2.2 Idealized model for multiphase-flow in an inclined channel. . . . . . . . . . . . . . . . 28

3.1 Two-phase flow patterns in horizontal flow. . . . . . . . . . . . . . . . . . . . . . . . 36

3.2 Flow patterns during evaporation in a horizontal tube ([35]). . . . . . . . . . . . . . 37
3.3 Baker (1954) flow pattern map for horizontal flow in a tube. . . . . . . . . . . . . . . 38
3.4 Mandhane (1974) flow pattern map for horizontal flow in a tube. . . . . . . . . . . . 39
3.5 Taitel and Dukler (1976) flow pattern map for horizontal flow in a tube. . . . . . . . 40
3.6 Hashizume (1983) flow pattern map for horizontal flow in a tube. . . . . . . . . . . . 42
3.7 Kattan (1998) flow pattern map (solid lines) evaluated for refrigerant R-410A at
Tsat = 5o C in 13.84 mm internal diameter tube. . . . . . . . . . . . . . . . . . . . . . 43
3.8 Cross-sectional and peripheral fractions in a circular tube. . . . . . . . . . . . . . . . 45
3.9 Stratified angle in two-phase flow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.10 Thome-El Hajal (2002) flow pattern map or refrigerant R-22 at Tsat = 5o C and
qf lux = 7.5 kW/m2 in 13.84 mm internal diameter tube. . . . . . . . . . . . . . . . . 48
3.11 Flow pattern map for R-22 at Tsat = 5o C in a 13.84 mm internal diameter tube at
G = 100 kg/m2 s and q = 2.1kW/m2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

4.1 Lockhart-Martinelli correlation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

4.2 Multipliers r2, r3, and r4 for boiling flow of water and steam (Thom 1964). . . . . . 58
4.3 Two-phase frictional pressure drop correlation (Baroczy 1965). . . . . . . . . . . . . 60
4.4 Mass velocity correction vs property index (Baroczy 1965). . . . . . . . . . . . . . . 61
4.5 Linear interpolation for transient flow (Hashizume 1985). . . . . . . . . . . . . . . . 71
4.6 Schematic representation of gas-liquid flow with a small liquid holdup
L in straight
smooth tubes (ARS 1989). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

5.1 Overall view of the modified test facility including the new two-zone test section. . . 76

13
14 LIST OF FIGURES

5.2 Schematic of the refrigerant circuit. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

5.3 Schematic of the hot water circuit ([75]). . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.4 New two-zone test section. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

6.1 Total, frictional and momentum pressure drops vs. vapor quality at different exper-
imental conditions (d = 13.8 mm). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
6.2 Total, frictional and momentum pressure drops vs. vapor quality at different exper-
imental conditions (d = 8 mm). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
6.3 Adiabatic vs. diabatic test section results for R410A. . . . . . . . . . . . . . . . . . . 89
6.4 Frictional pressure gradients vs. vapor quality for R22 and R410A. . . . . . . . . . . 90
6.5 Comparison of the frictional pressure gradients for R134a,R22 and R410A at two set
of experimental conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
6.6 Frictional pressure gradients vs. vapor quality for different mass velocities. . . . . . . 91
6.7 Frictional pressure gradients vs. vapor quality at different heat fluxes. . . . . . . . . 91
6.8 Frictional pressure gradients vs. vapor quality for R134a at different experimental
conditions (d = 13.8 mm). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
6.9 Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 13.8 mm) – (I). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
6.10 Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 13.8 mm) – (II). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.11 Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 13.8 mm) – (III). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
6.12 Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 13.8 mm) – (IV). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.13 Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 8 mm) – (I). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6.14 Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 8 mm) – (II). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.15 Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 13.8 mm) – (I). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.16 Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 13.8 mm) – (II). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
6.17 Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 13.8 mm) – (III). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.18 Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 13.8 mm) – (IV). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.19 Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 8 mm) – (I). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
LIST OF FIGURES 15

6.20 Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 8 mm) – (II). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

6.21 Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 8 mm) – (III). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

6.22 Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 8 mm) – (IV). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

7.1 Frictional pressure gradients vs. three prediction methods for R22 at different ex-
perimental conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

7.2 Frictional pressure gradients vs. three prediction methods for R410A at different
experimental conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

7.3 Comparisons of experimental to predicted values using Friedel correlation for the
entire database: a) 67.33% are predicted within ±30%, b) 51.81% are predicted
within ±20%. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

7.4 Comparisons of experimental to predicted values using Grönnerud correlation for

the entire database: a) 46.15% are predicted within ±30%, b) 40.45% are predicted
within ±20%. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

7.5 Comparisons of experimental to predicted values using Müller-Steinhagen and Heck

correlation for the entire database: a) 75.79% are predicted within ±30%, b) 49.64%
are predicted within ±20%. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

7.6 Flow pattern maps for R-22, Tsat = 5o C, D = 13.84 mm at G = 300 kg/m2 s and
heat fluxes: a) 7.5 kW/m2 , b) 17.5 kW/m2 , c) 37.5 kW/m2 , d) 57.5 kW/m2 . . . . . 117

7.7 Flow pattern maps for R-410A, Tsat = 5o C, D = 8.00 mm at G = 400 kg/m2 s and
initial heat fluxes: a) 37.5 kW/m2 , b) 57.5 kW/m2 . . . . . . . . . . . . . . . . . . . . 117

7.8 Flow pattern maps for R-410A, Tsat = 5o C, D = 13.84 mm at G = 300 kg/m2 s and
initial heat fluxes: a) 7.5 kW/m2 , b) 17.5 kW/m2 , c) 37.5 kW/m2 , d) 57.5 kW/m2 . 118

7.9 Simplified annular flow configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

7.10 Simplified stratified flow configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . 121

7.11 Dryout zone during evaporation in horizontal tube. . . . . . . . . . . . . . . . . . . . 123

7.12 Comparisons of experimental to predicted values for the entire database: a) 82.30%
are predicted within ±30%, b) 64.71% are predicted within ±20%. . . . . . . . . . . 126

7.13 Comparisons of experimental to predicted values for R22 and both tested diameters:
a) 82.67% are predicted within ±30%, b) 60.26% are predicted within ±20%. . . . . 127

7.14 Comparisons of experimental to predicted values for R410A and both tested diame-
ters: a) 83.02% are predicted within ±30%, b) 69.46% are predicted within ±20%. . 127

7.15 Comparisons of experimental to predicted values for d = 8 mm and both tested

fluids: a) 82.06% are predicted within ±30%, b) 67.64% are predicted within ±20%. 128
16 LIST OF FIGURES

7.16 Comparisons of experimental to predicted values for d = 13 mm and both tested

diameters: a) 82.24% are predicted within ±30%, b) 62.86% are predicted within
±20%. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
7.17 Comparisons of experimental to predicted values for the entire database segregated
by flow regime: a) annular, b) Slug+Intermittent, c) SW, d)Slug+SW. . . . . . . . . 129
7.18 Experimental and predicted values vs. vapor quality for R134a at different experi-
mental conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
7.19 Experimental and predicted values vs. vapor quality for R22 at different experimen-
tal conditions and d = 13.8 mm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
7.20 Experimental and predicted values vs. vapor quality for R22 at different experimen-
tal conditions and d = 8 mm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
7.21 Experimental and predicted values vs. vapor quality for R410A at different experi-
mental conditions and d = 13.8 mm. . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
7.22 Experimental and predicted values vs. vapor quality for R410A at different experi-
mental conditions and d = 8 mm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

B.1 Perturbed vapor-liquid interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

List of Tables

4.1 Values of C to fit the empirical curves of Lockhart and Martinelli. . . . . . . . . . . 56

5.1 Experimental conditions for in-tube evaporation tests. . . . . . . . . . . . . . . . . . 75

5.2 Main properties and geometrical dimensions of the implemented test sections. . . . . 79

7.1 Segregated experimental values by flow regime using Wojtan et al. flow pattern
map, where S= stratified flow, SW= stratified-wavy flow, I= intermittent flow, A=
annular flow, M= mist flow and D represents the transition zone between annular
and mist flow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
7.2 Compilation of percentages of the database within an specific range for the new
proposed model and three existing methods. . . . . . . . . . . . . . . . . . . . . . . . 125

A.1 Physical properties for fluids used during the experimental campaign at three differ-
ent saturation temperatures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

17
18 LIST OF TABLES
Chapter 1

Introduction

Liquid-vapor phase-change processes play a very important role in many technological applications.
The virtually isothermal heat transfer associated with boiling and condensation processes makes
their inclusion in power and refrigeration cycles very advantageous from a thermodynamic effi-
ciency viewpoint. In addition boiling and condensation can achieve large heat transfer rates with
small temperatures differences. Processes associated with phase-change phenomena are among
the most complex transport processes encountered in engineering applications. These processes
include all the complexity of single-phase convective transport (non-linearities, transition to turbu-
lence, instabilities) and additional elements resulting from the motion of the vapor-liquid interface,
non-equilibrium effects and interactions between the phases.
Two-phase flow of gases and liquids or vapors and liquids in pipes, channels, equipment, etc. is
frequently encountered in industry and has been studied intensively for many years. The reliable
prediction of pressure drop in two-phase flows is thereby an important aim. Because of the com-
plexity mention above, empirical or semiempirical relationships are only of limited applicability.
Yet, pressure drops predicted using leading methods differ by up to 100% according to Ould Didi,
Kattan and Thome [105] and Moreno Quibén and Thome [93]. Hence, increasingly, attempts are
being made to develop prediction methods which are based on physical models and which can be
correlated with the measured results. The mathematical content is kept as small as possible in
order to keep the application practical.
Kattan, Thome and Favrat [76, 77, 78] proposed a new physically based (based on a simplified
interfacial two-phase structure) flow pattern oriented model to predict heat transfer coefficients
during evaporation in horizontal tubes. The new heat transfer model was a significant step ahead,
improving significantly the predictive accuracy. In addition, in the previously mentioned study by
Ould Didi et al. it is shown that classifying the flow by local flow pattern and then using the best
two-phase pressure drop prediction method for that particular flow pattern results in a significant
improvement in accuracy.
Based on that, the idea is to extend this physically based approach to the development of a
two-phase pressure drop prediction method. Hence, in order to develop this approach an experi-
mental and analytical study was undertaken. The main experimental objectives of this study are
to: (i) accurately and reliably measure two-phase pressure drops over a wide range of experimental
conditions and (ii) be able to obtain experimental two-phase pressure drop values in the different
flow regimes by a consequent design of the test sections. The analytical objective of this study is
then the development of a flow pattern based pressure drop model that respects the two-phase flow

19
20 CHAPTER 1. INTRODUCTION

structure of the various flow regimes as much as possible while maintaining a degree of simplicity.
The present work will complete the fourth basic step in LTCM’s flow pattern based work on
two-phase flow and heat transfer inside horizontal round tubes: (i) generalized flow pattern map,
(ii) flow boiling heat transfer model (iii) convective condensation model and (iv) two-phase frictional
pressure drop model.
The expected result from this project is a much better two-phase pressure drop prediction method
and therefore a more accurate design method that better simulates experimental trends and the
effects of the principal variables on the process.
The manuscript is organized as follows:

• Chapter 1 - Introduction: Background and purpose of the present study.

• Chapter 2 - Fundamentals: Definition of the main variables and basic equations used in
two-phase flow and two-phase pressure drop analysis.

• Chapter 3 - Overview of flow pattern maps: Presentation of the state-of-the-art of existing

flow pattern maps.

• Chapter 4 - Two-phase pressure drop models: State-of-the-art of two-phase pressure drop

models.

• Chapter 5 - Description of experiments: Test facility and data collection methods are de-
scribed.

• Chapter 6 - Experimental results: Results of the two-phase pressure drop measurements are
presented and analyzed.

• Chapter 7 - New two-phase pressure drop flow pattern based model: Presentation of a new
flow pattern based, two-phase frictional pressure drop model.

• Chapter 8 - Conclusions: General conclusions of this present study are summarized.

Chapter 2

Fundamental Definitions in
Two-Phase Flow

This chapter introduces the primary variables used throughout this work and derives some simple
relationships between them for the case of one-dimensional flow. To distinguish between gas and
liquid the subscripts ’L ’ for liquid and ’G ’ for vapor will be used. Basic equations for two-phase
flows are also introduced at the end of the chapter.

2.1 Two-phase flow

Classical thermodynamics tell us that a phase is a macroscopic state of matter which is homogeneous
in chemical composition and physical structure; e.g. a gas, a liquid or solid of a pure component.
Two-phase flow is the simplest case of multiphase flow in which two phases are present for a pure
component.
In internal convective vaporization and condensation processes, the vapor and liquid are in si-
multaneous motion inside the pipe. The resulting two-phase flow is generally more complicated
physically than single-flow. In addition to the usual inertia, viscous, and pressure forces present in
single-phase flow, two-phase flows are also affected by interfacial tension forces, the wetting char-
acteristics of the liquid on the tube wall, and the exchange of momentum between the liquid and
vapor phases in the flow.

2.2 Vapor quality

The vapor quality (x) is defined to be the ratio of the vapor mass flow rate (ṀG [kg s−1 ]) divided
by the total mass flow rate (ṀG + ṀL ):

ṀG
x= (2.1)
ṀG + ṀL

When phase change does not take place in the channel, one needs to measure the mass flow rate
of each phase, and the quality is then determined for the entire channel. In case there is a phase

21
22 CHAPTER 2. FUNDAMENTAL DEFINITIONS IN TWO-PHASE FLOW

change in the channel, e.g. if the channel is heated and boiling takes place, then the quality will
increase (inverse for condensation) downstream with the flow.
Since often there is not thermal equilibrium between the phases, one cannot calculate the quality
merely by knowing the inlet quality and the heat flux from the wall. Unfortunately, it is very
difficult to measure or calculate with precision the quality of the liquid-vapor mixture flowing in
a channel where a change of phase takes place. A fictitious quality, the so called thermodynamic
equilibrium quality can be calculated by assuming that both phases are saturated, i.e., that their
temperatures are equal to the saturation temperature corresponding to their common pressure.
The so-called thermodynamic equilibrium quality can be calculated as:

h(z) − hL
x= (2.2)
hLG

where hL [J kg−1 ] is the enthalpy of the saturated liquid, hLG [J kg−1 ] is the latent heat of vapor-
ization, and h(z) [J kg−1 ] is the enthalpy at a cross section z, which can be calculated from:

z
1
h(z) = hinlet + q
(z)dz (2.3)
Ṁ 0

where hinlet is the enthalpy of the fluid at the inlet and q
(z) [W m−1 ] is the heat input per unit
length of channel.

2.3 Void fraction

In two-phase flow, void fraction is one of the most important parameters to be defined. It defines
the cross-sectional area occupied by each phase. As it determines mean velocities of the liquid and
the vapor, it represents a fundamental parameter in the calculation of pressure drop, flow pattern
transitions and heat transfer coefficients.
L iq u id
V a p o r

Figure 2.1: Cross-sectional void fraction (
).

The void fraction of the vapor is defined as:

AG

= (2.4)
AG + AL
where AG is the sum of areas occupied by voids and AL is the sum of areas occupied by the liquid.
The total cross-sectional area of the channel is called A.
2.4. VELOCITIES 23

2.4 Velocities

In two-phase flow there are a number of velocities that can be defined. Also, in general, the phases
will not have the same velocity and there will be a relative velocity between them.

True average velocities

The true average velocities (also called actual velocities) of the phases uG and uL are the velocities
by which the phases actually travel. The cross sectional average true velocities are determined by
the volumetric flow rates Q̇G and Q̇L [m3 s−1 ] of the vapor and liquid divided by the cross-sectional
areas occupied by the respective phases:

Q̇G Q̇G
uG = = (2.5)
AG
A

Q̇L Q̇L
uL = = (2.6)
AL (1 −
)A

From the equation of continuity it is possible to define liquid and vapor true mean velocities referred
to their own cross sectional areas and their own mass flow rates as follows:

x Ṁ G x
uG = = (2.7)

ρG A ρG

1 − x Ṁ G 1−x
uL = = (2.8)
1 −
ρL A ρL 1 −

Superficial velocities

The superficial velocities (also called volumetric fluxes) of the phases jG and jL are defined as the
volumetric flow rate of the phase through the total cross-sectional area of the two-phase flow. It
might also be expressed as the phase velocity if it would flow alone in the entire cross section. Thus:

Q̇G G
jG = = x =
uG (2.9)
A ρG

Q̇L G
jL = = (1 − x) = (1 −
) uL (2.10)
A ρL

The total superficial velocity is defined as:

j = jG + jL (2.11)
24 CHAPTER 2. FUNDAMENTAL DEFINITIONS IN TWO-PHASE FLOW

Drift velocities

The drift velocities of the phases VGj and VLj are defined as the true average velocity of the phase
in relation to the total superficial velocity, namely:

VGj = uG − j (2.12)

VLj = uL − j (2.13)

The drift fluxes of the phases jGj and jLj are defined as follows:

jGj =
VGj =
(uG − j) (2.14)

jLj =
VLj = (1 −
)(uL − j) (2.15)

It follows, from equations (2.14), (2.15) and (2.11) that:

jGj + jLj = 0 (2.16)

Mass velocity

The mass velocity (G) is defined to be the ratio of the mass flow rate (Ṁ ) divided by the cross-
sectional area of the flow channel:

Ṁ
G= (2.17)
A

Considering the continuity law, the mass velocity is the expression of the mean flow velocity mul-
tiplied by the mean density. The mass velocity has units of [kg/m2 s].

2.5 Definition of non-dimensional numbers

The principal non-dimensional numbers used in the present study are defined below. Different
definitions of main non-dimensional numbers, particularly for the Reynolds and Froude number,
can be found in the literature. In order to be coherent in this work, the corresponding used here
definitions are introduced.

Reynolds number

The Reynolds number represents the ratio of the inertial forces to the viscous forces. For the
particular case of forced convection inside a tubular channel, the liquid Reynolds number for a
single-phase in a channel can be expressed in the following form:
2.5. DEFINITION OF NON-DIMENSIONAL NUMBERS 25

ρL uL Dh
ReL = (2.18)
µL

where Dh is the hydraulic diameter defined as the ratio of the cross-sectional A to the wetted
perimeter PL and is calculated as follows:

4A
Dh = (2.19)
PL
In the particular case of circular tubes, Dh = D.
Considering one-dimensional flow and using the definition of the true mean velocity from equation
(2.6), the liquid Reynolds number in a two-phase flow can be expressed as:

GDh 1 − x
ReL = (2.20)
µL 1 −

where Dh for the liquid phase is expressed in the following form:

4AL 4(1 −
)A

Dh = = (2.21)
PL PL

Substituting equation (2.21) into (2.20) the liquid Reynolds number is defined as:

4G(1 − x)A
ReL = (2.22)
µL PL

The same approach will be used for the vapor Reynolds number which is defined as:

GDh x
ReG = (2.23)
µG

where the Dh for the vapor phase is expressed as:

4AG 4A

Dh = = (2.24)
PL PL

Finally, substituting equation (2.24) into (2.23), the vapor Reynolds number is expressed as:

4GxA
ReG = (2.25)
µG PL

Kattan et al. ([76],[77],[78]) and Wojtan [153] assumed that the hydraulic diameter Dh for the vapor
phase is equal to the tube diameter D like in single-phase flow. As the void fraction increased very
rapidly with increasing the vapor quality, this assumption seems to be reasonable. Assuming this
assumption, the vapor Reynolds number, equation (2.23), can be expressed as:

GD x
ReG = (2.26)
µG

26 CHAPTER 2. FUNDAMENTAL DEFINITIONS IN TWO-PHASE FLOW

Nusselt number

The Nusselt number expresses the ratio of convective to conduction temperature gradient. In
internal forced convection, the reference length is the tube diameter:

hD
Nu = (2.27)
λ

where h is the heat transfer coefficient, D is the tube diameter, λ is the thermal conductivity.

Liquid Froude number

The Froude number represents the ratio of the inertia forces to the gravitational forces. The general
expression is:

u2
Fr = (2.28)
gL

Wojtan [153], based on existing flow boiling correlations that use a different Froude number defi-
nition, uses the following form of the liquid Froude number:

G2
F rL = (2.29)
ρL gD

In order to be coherent with the mentioned previous work, the same expression will be used through-
out this work.

Liquid Weber number

The liquid Weber number expresses the ratio of inertia to surface tension forces. As for the Nusselt
number, the reference length is the tube diameter.
It is expressed for liquid phase as:

ρL u2L D
W eL = (2.30)
σ

Prandtl number

The Prandtl number is the ratio between the momentum diffusivity and the thermal diffusivity.
It is expressed for a liquid as:

cpL µL
P rL = (2.31)
λL

The same expression for a vapor is:

2.6. BASIC EQUATIONS OF TWO-PHASE FLOW 27

cpG µG
P rG = (2.32)
λG

Martinelli parameter

The Martinelli parameter [90] is defined as the ratio between the theoretical pressure gradients
which would occur if either fluid were flowing alone in the pipe with the original flow rate of each
phase. The Martinelli parameter Xtt is calculated as:

(∆pF /∆L)L0
Xtt2 = (2.33)
(∆pF /∆L)G0

Xtt2 is void fraction independent and is a measure of the degree to which the two-phase mixture
is closer to being a liquid, i.e. Xtt2 >> 1, or to being a gas, i.e. Xtt2 << 1. The subscript tt is
sometimes used and signifies that both phases are turbulent.
Modelling the pressure drop of each phase with its superficial velocity and friction factors in the
classical form:

fL = CL Re−n
L (2.34)

fG = CG Re−m
G (2.35)

and assuming the same friction model for both phases (both turbulent or both laminar) which
means that m = n and CL = CG , equation (2.33) reduces to:

 (−n+2)/2  n/2  1/2

1−x µL ρG
Xtt = (2.36)
x µG ρL

Based on equation (2.34) Lockhart and Martinelli [90] and then Taitel and Dukler [126] used
n = m = 0.2 and CL = CG = 0.046 for a smooth pipe. Steiner [137], Kattan [75] and Zürcher [162]
used n = m = 0.25 and CL = CG = 0.3146 for a smooth pipe. In this work the same expression as
Steiner, Kattan and Zürcher will be used:

 0.875  0.125  0.5

1−x µL ρG
Xtt = (2.37)
x µG ρL

2.6 Basic equations of two-phase flow

Developments of the full governing equations for three-dimensional, time-varying two-phase flow
can be found in Ishii [67], Bouré [20] and Delhaye [37]. The form of the governing equations
can be simplified by invoking time and/or space averaging. The averaging processes make the
equations more tractable but, at the same time, useful information about the flow is lost at each
simplifying step. In the present context, for sake of simplicity, the flow is considered to be steady
28 CHAPTER 2. FUNDAMENTAL DEFINITIONS IN TWO-PHASE FLOW

a k

P k w

...
'2 ',
s '1 ',
s e
d z p h a

e 'k '
G , p h a s
k

d F k

q f k w

Figure 2.2: Idealized model for multiphase-flow in an inclined channel.

and one-dimensional in the sense that all dependent variables are idealized as being constant over
any section of the tube, varying only in the axial direction.
To facilitate development of a one-dimensional analysis of multi-phase flow, we will consider the
system shown in Fig. (2.2). A stratified flow is chosen to allow the equations to be derived for
the general case where each phase is in contact with the channel wall as well as having a common
interface. It is further assumed that the pressure across any phase normal to the channel is uniform.
Mean values of velocity and density of each phase are assumed to exist across any phase normal to
the flow.

2.6.1 Conservation of mass

The equation expressing the conservation of mass in the absence of any removal or addition of fluid
through the channel walls for phase k is:

∂ ∂
(A
k ρk ) + (A
k ρk uk ) = Γk (2.38)
∂t ∂z

where
k is the void fraction of phase k, ρk is the density of phase k, uk is the true average velocity
of phase k and Γk is the mass transfer (mass flow rate Ṁ per unit length) to phase k from the
various interphase mass transfer, where:


Γk = 0 (2.39)
k

For the particular case of steady state two-phase gas(G )/liquid(L ) flow in a constant area channel
this reduces to two expressions:
2.6. BASIC EQUATIONS OF TWO-PHASE FLOW 29

d
(AG ρG uG ) = ΓG
dz
d
(AL ρL uL ) = ΓL (2.40)
dz

and

dṀG dṀL
ΓG = −ΓL = =− (2.41)
dz dz

2.6.2 Conservation of momentum

The rate of creation of momentum of phase k plus the rate of inflow of momentum is balanced
against the sum of the forces acting on that phase plus the momentum generation due to mass
transfer,as follows:

 
∂ ∂
(Ṁk δz) + Ṁk uk + δz (Ṁk uk ) − Ṁk uk =
∂t ∂z
1
     
∂ ∂
A
k p − A
k p + δz (A
k p) + p −δz (A
k ) −
∂z ∂z
2

n
A
k ρk δz gsinθ − τkw Pkw δz + τknz Pkn δz + uk Γ k (2.42)
3 4 6
1
5

A brief discussion of the terms in this equation is given below:

• the terms in box 1 represent the rate of creation of momentum plus the rate of inflow of
momentum within the control element.

• the terms in box 2 represent the pressure forces in the control element.

• the term in box 3 represents the gravitational forces.

• the term in box 4 represents the wall shear force (dFk ), where τkw is the wall shear stress
between the phase k and the channel wall and Pkw is the contact perimeter between the wall
and phase k.

• the term in box 5 represents the sum of the interfacial shear forces (S1 , S2 , . . . ), where where
τknz is the z component of the interfacial shear stress between phase k and phase n and Pkn
is the contact perimeter between phase k and phase n.

• the term in box 6 represents the rate of generation of momentum of phase k due to mass
transfer assuming that that the mass transferred across the interface is accelerated to the
mean velocity of the receiving phase.
30 CHAPTER 2. FUNDAMENTAL DEFINITIONS IN TWO-PHASE FLOW

Rearranging terms, equation (2.42) can be written as:

∂p  n
−A
k − τkw Pkw δz + τknz Pkn δz − A
k ρk δzgsinθ + uk Γk =
∂z 1
∂ ∂
(Ṁk δz) + δz (Ṁk uk ) (2.43)
∂t ∂z

Thus for a steady-state two-phase gas(G )/liquid(L ) flow in a constant area channel, we have for the
gas phase

−AG dp − τGW PGw dz + τGL PGL dz − AG ρG dz g sinθ + uG ΓG = ṀG d uG (2.44)

and for the liquid phase

−AL dp − τLw PLw dz + τLG PLG dz − AL ρL dz g sinθ + uL ΓL = ṀL d uL (2.45)

Adding equations (2.44) and (2.45) and using the conservation of momentum across the interface

τGL PGL dz + uG ΓG = τLG PLG dz + uL ΓG (2.46)

we obtain the following equation which represents the basic differential equation for this simplified
approach

−Adp + τGw PGw dz − τLw PLw dz − g sinθ[AL ρL + AG ρG ] = d(ṀL uL + ṀG uG ) (2.47)

The net frictional force acting on each phase may be expressed in terms of the ones occupied by
each phase

 
dp
(dFG + S) = −τGw PGw dz − τGL PGL dz = −AG F dz (2.48)
dz
  G
dp
(dFL − S) = −τLw PLw dz + τGL PGL dz = −AL F dz (2.49)
dz L

Adding equations (2.48) and (2.49), we obtain the expression for the total net frictional force
 
dp
(dFG + dFL ) = −τGw PGw dz − τLw PLw dz = −A F dz (2.50)
dz
 
dp
where the term F represents the part of the overall pressure gradient required to overcome
dz
friction.
Substitution of equation (2.50) into equation (2.47) and rearrangement yields
       
dp dp dp dp
= F + a + z (2.51)
dz dz dz dz
2.6. BASIC EQUATIONS OF TWO-PHASE FLOW 31
 
dp
where the term a reflects the change in kinetic energy of the flow and can be expressed, using
dz
relationships introduced in this chapter, as:

   2 
dp 1 d

2 d x (1 − x)2
− a = ṀG uG + ṀL uL = G + (2.52)
dz A dz dz
ρG (1 −
) ρL
 
dp
and the term z reflects the change in static head and can be expressed as:
dz
   
dp AG AL
− z = g sinθ ρG + ρL = g sinθ [
ρG + (1 −
)ρL ] (2.53)
dz A A

The above derivation introduces the use of the momentum equation to relate the total pressure
gradient in terms of its three separate components: friction, acceleration and static head. It should
be explained at this point that the frictional component has been derived in terms of the total wall
shear force (dFG + dFL ).

2.6.3 Conservation of energy

The rate of increase of total energy for phase k (internal plus kinetic energy) within the control
element plus the rate at which total energy is convected into the control element is balanced against
the rate at which heat is added to phase k plus the rate at which work is done on phase k plus the
rate at which energy energy is transferred across the interface to the control element. The equation
expressing the differential energy balance is:

          
∂ u2k u2k u2k ∂ u2k

k ρk ek + Aδz + Ṁk ek + δz − Ṁk ek + − δz Ṁk ek + =
∂t 2 2 2 ∂z 2
1
   
n
Ṁk p Ṁk p ∂ Ṁk p
φkw Pkw δz + φkn Pkn δz + φ̇k A
k δz + − + δz −
ρk ρk ∂z ρk
1
2 3

n  
∂
k δz p u2k
Ṁk g sinθδz − pAδz + Γk + uk τkn pkn δz + Γk δz ek + (2.54)
∂t 4 ρk 2
1 6
5

A brief discussion of terms in this equation is given below:

• the terms in box 1 represent the rate of increase of total energy plus the rate at which energy
enters within the control element in the absence of the addition or subtraction of mass through
the channel walls.

• the terms in box 2 represent the rate at which heat enters phase k within the control volume,
namely: the heat flow via the channel wall over the perimeter Pkw , the heat flow via the
various interfaces with the other n phases and the internal heat generation for phase k within
the control element itself.

• the terms in box 3 represent the work done by pressure forces.

32 CHAPTER 2. FUNDAMENTAL DEFINITIONS IN TWO-PHASE FLOW

• the terms in box 4 represent the work done by body forces.

• the terms in box 5 represent the work done by shear and pressure forces at the interface with
other phases.

• the term in box 6 represents the rate at which energy is added to phase k by virtue of mass
transfer across the interface.

Rearranging terms, equation (2.54) may be written as:

   
∂ u2k ∂ u2k
A
k ρk ek + + Ṁk ik + =
∂t 2 ∂z 2
n
−Ṁk g sin θ + φwk Pwk + φkn Pkn + φ̇k A
k −
1
  
u2
n
∂
k
pA + Γk ik + k + uk τkn Pkn (2.55)
∂t 2 1

where ik represents the enthalpy of phase k per unit of mass

p
ik = uk + (2.56)
ρk

For the particular case of steady-state, two-phase gas(G )/liquid(L ) flow in a constant area channel
with no internal heat generation (φ̇k ), equation (2.55) reduces to the following expressions for the
vapor and liquid phases:

  
u2
d ṀG iG + G + ṀG g sin θ δz =
2
 
u2
φwG PwG δz + φGL PGL δz + uG τGL PGL δz + ΓG δz iG + G (2.57)
2

  
u2
d ṀL iL + L + ṀL g sin θ δz =
2
 
u2
φwL PwL δz + φLG PLG δz + uL τLG PLG δz + ΓL δz iL + L (2.58)
2

Adding equations (2.57) and (2.58) and taking into account the conservation of energy across the
interface, the results gives:

   
u2 u2
ΓG iG + G + φGL PGL + uG τGL PGL = ΓL iL + L + φLG PLG + uL τLG PLG (2.59)
2 2

We can now obtain the following differential equation for this simplified approach:
2.6. BASIC EQUATIONS OF TWO-PHASE FLOW 33

 
d   d ṀG u2G ṀL u2L
ṀG iG + ṀL iL + + + (ṀG + ṀL ) g sin θ = QwL (2.60)
dz dz 2 2

where the term QwL (= φwL PwL + φwG PwG ) represents the heat transferred to the fluid across the
channel length per unit of wall.
Using relationships introduced in this chapter, equation (2.60) may be written
   
dp x (1 − x) dE QwL
− + = − +
dz ρG ρL dz (ṀG + ṀL )
    3 
d x (1 − x) G2 d x (1 − x)3
+ p + + + + g sin θ (2.61)
dz ρG ρL 2 dz ρ2G
2 ρ2L (1 −
)2

where E = x eG + (1 − x) eL is the flow-weighted mixture internal energy per unit of mass.

The above equation shows that the total pressure gradient can be expressed in terms of a frictional
dissipation term (first bracketed term), an accelerational term (second bracketed term) and a static
head term (final term).
 
dE QwL
It should pointed out that the frictional dissipation term − includes the dissipa-
dz ṀG + ṀL
tion of mechanical energy not only within the fluid due to friction at the channel walls but also at
the interface due to the relative motion of the phases.
34 CHAPTER 2. FUNDAMENTAL DEFINITIONS IN TWO-PHASE FLOW
Chapter 3

Overview of Flow Pattern Maps

For two-phase flow, the respective distribution of the liquid and vapor phases is an important
factor of their description. Their respective distribution take on some commonly observed flow
structures, which are defined as two-phase flow patterns or flow regimes, and they present
particular identifying characteristics. In fact, pressure drops and heat transfer coefficients are
closely related to the local flow structure of the fluid, and thus two-phase flow pattern prediction
is an important aspect of modelling evaporation and condensation. Hence, in order to obtain local
flow pattern based models, a reliable flow pattern map to identify what type of flow pattern
exists at the local flow conditions is needed. Analogous to predicting the transition from laminar
to turbulent flow in single-phase flows, two-phase flow pattern maps are used for predicting the
transition from one type of two-phase flow pattern to another.
In this chapter, first the geometric characteristics of flow patterns inside horizontal tubes will be
described. Second, several widely quoted, older flow pattern maps for horizontal flows will be
presented. Finally, a recent flow pattern map and corresponding flow regime transition equations
specifically for diabatic flows and in particular for evaporation in horizontal tubes will be presented.

3.1 Flow patterns in horizontal flow

Flow patterns for co-current flow of gas and liquid in a horizontal tube are strongly influenced by
gravity that acts to stratify the liquid to the bottom of the tube and the gas to the top. The liquid
and gas phases distribute themselves into several recognizable flow structures. These are referred
to as flow patterns or flow regimes and they are shown in Fig. (3.1) and can be described as follows:

• Bubbly flow. The gas bubbles are dispersed in the liquid with a high concentration of
bubbles in the upper half of the tube due to their buoyancy. When shear forces are dominant,
the bubbles tend to disperse uniformly in the tube. For horizontal flows, this regime only
occurs at high mass flow rates.

• Stratified flow. At low liquid and gas velocities, complete separation of the two phases
occurs. The gas goes to the top and the liquid to the bottom of the tube. Both phases are
separated by an undisturbed flat interface. Therefore, the liquid and gas are fully stratified
in this regime.

35
36 CHAPTER 3. OVERVIEW OF FLOW PATTERN MAPS

Liquid Vapor

Flow

Bubbly flow Stratified flow

Stratified-wavy flow Plug flow

Slug flow Annular flow

Intermittent flow Mist flow

Figure 3.1: Two-phase flow patterns in horizontal flow.

• Stratified-wavy flow. Increasing the gas velocity in a stratified flow, waves are formed on
the interface and travel in the direction of the flow. The amplitude of the waves is notable
and depends on the relative velocity of the two phases; however, their crests do not reach the
top of the tube. The waves climb up the sides of the tube, leaving behind thin films of liquid
on the wall.

• Intermittent flow. Further increasing the gas velocity, these interfacial waves become large
enough to wash the top of the tube. This regime is characterized by large amplitude waves
intermittently washing the top of the tube with smaller amplitude waves in between. Large
amplitude waves often contain entrained bubbles. The top wall is nearly continuously wetted
by the large amplitude waves and the thin liquid films left behind. Intermittent flow is also a
composite of the plug and slug flow regimes. These subcategories are characterized as follows:

– Plug flow. This regimen has liquid plugs that are separated by elongated bubbles. The
diameters of the elongated bubbles are much smaller than the tube such that the liquid
phase is continuous along the bottom of the tube below the elongated bubbles.
– Slug flow. At higher gas velocities, the diameters of elongated bubbles become similar
in size to the channel height. The liquid slugs separating such elongated bubbles can
also be described as large amplitude waves.

• Annular flow. At even larger gas flow rates, the liquid forms a continuous film around
the perimeter of the tube. The interface between the liquid annulus and the vapor core is
disturbed by small amplitude waves and droplets may be dispersed in the gas core. At high
gas fractions, the top of the tube with its thinner film becomes dry first, so that the annular
film covers only part of the tube perimeter and thus this is then classified as stratified-wavy
flow.
3.2. FLOW PATTERN MAPS IN HORIZONTAL FLOW 37

• Mist flow. At very high velocities most of the liquid is entrained as spray by the gas. The
spray appears to be produced by the high velocity gas ripping the annular liquid film off from
the wall.

3.2 Flow pattern maps in horizontal flow

The analysis of single-phase flow is made easier if one can establish that the flow is either laminar or
turbulent and whether any separation or secondary flow effect occurs. This information is equally
important in the study of gas-liquid flow. However, perhaps of greater importance in the latter
case is the topology or geometry of the flow, i.e. the corresponding flow patterns or flow regimes.
Fig. (3.2) shows a schematic representation of a horizontal tubular channel heated by a uniform
low heat flux and fed with liquid just below the saturation temperature.

Figure 3.2: Flow patterns during evaporation in a horizontal tube ([35]).

To predict the local flow pattern in a tube, a flow pattern map is used. These are an attempt,
on a two-dimensional graph, to separate the space into areas corresponding to the various flow
regimes. It should be pointed out that the flow pattern is also influence by a number of secondary
variables but it is not possible to represent their influence using only a two-dimensional plot. One
should be aware that transition curves on flow pattern maps should be considered as transition
zones analogous to that between laminar and turbulent flows.

3.2.1 Baker flow pattern map

The first to recognize the importance of the flow pattern as a starting point for the calculation of
pressure drop, void fraction, and heat and mass transfer was Baker [7] in 1954. He published the
earliest flow pattern map for horizontal flow, presented in Fig. (3.3).
To utilize this map, first the mass velocities of the liquid GL and vapor GG must be determined.
Then the gas-phase parameter λ and the liquid-phase parameter ψ are calculated as follows:

 1/2
ρG ρL
λ= (3.1)
ρair ρwater
38 CHAPTER 3. OVERVIEW OF FLOW PATTERN MAPS

Figure 3.3: Baker (1954) flow pattern map for horizontal flow in a tube.

 2 1/3
σ
 µ  ρ
water L water
ψ= (3.2)
σ µwater ρL

where ρG , ρL , µL and σ are the properties of the fluid and ρwater , ρair , µwater and σwater are
the reference properties of air and water at standard atmospheric pressure and room temperature.
This map was developed based on air-water data. Note that λ and ψ are standard dimensionless
parameters that should take into account the variation in the properties of the fluid.

3.2.2 Mandhane et al. flow pattern map

In the interest of simplicity, Mandhane et al. [91] in 1974 proposed a basic flow pattern map based on
air-water data, and then attempt to apply physical property corrections. While this was certainly
not a new approach, previous workers did not have access to the amount of data that were available
for this study. The proposed flow pattern map is shown in Fig. (3.4). It should be noted that the
transition between adjacent flow patterns do not occur suddenly but over a range of flow rates.
Thus, in this figure, transitions between flow patterns are shown as broad bands instead of lines.
The transition boundaries indicated were located on the basis of a log VSL vs. log VSG plot of the
1178 observations for the air-water system, where VSL (jL ) and VSG (jG ) are respectively the liquid
and vapor superficial velocities. One of the mayor improvements of this approach was that they
showed that the effect of tube diameter was adequately taken into account by using the superficial
velocities, VSL and VSG , as the coordinate axes.
3.2. FLOW PATTERN MAPS IN HORIZONTAL FLOW 39

Figure 3.4: Mandhane (1974) flow pattern map for horizontal flow in a tube.

As pointed out, the Mandhane et al. flow pattern map was developed using air-water data. In order
to take into account the variation in the properties of the fluid, they proposed two dimensionless
parameters. These factor are applied to the flow pattern boundaries rather than to the axes of the
map following the procedure well described in their paper. The proposed map, and the physical
property correction procedure were then compared with a huge data bank.

3.2.3 Taitel and Dukler flow pattern map

Taitel and Dukler [125] proposed in 1976 a flow pattern map for horizontal flow in tubes. This map
is shown in Fig. (3.5) and is based on their analytical analysis of the flow transition mechanisms
together with empirical selection of several parameters. The proposed map has a better scientific
basis than many of the previous attempts and thus extrapolates better than the others maps. The
map uses the Martinelli parameter X, the gas Froude number F rG and the parameters T and K
and is composed of three graphs.
The Martinelli parameter is:

 1/2
(dp/dz)L
X= (3.3)
(dp/dz)G
40 CHAPTER 3. OVERVIEW OF FLOW PATTERN MAPS

Figure 3.5: Taitel and Dukler (1976) flow pattern map for horizontal flow in a tube.

The gas-phase Froude number is:

GG
F rG = (3.4)
[ρG (ρL − ρG )Dg]1/2

Their parameter T is:

 1/2
|(dp/dz)L |
T = (3.5)
g(ρL − ρG )

Their parameter K is:

1/2
K = F rG ReL (3.6)
3.2. FLOW PATTERN MAPS IN HORIZONTAL FLOW 41

where the liquid-phase and vapor-phase Reynolds numbers are:

GL D
ReL = (3.7)
µL

GG D
ReG = (3.8)
µG

The pressure gradient of the flow for phase k (where k is either L or G) is:

2fk G2k
(dp/dz)k = − (3.9)
ρk D

For Rek ≤ 2000, the laminar flow friction factor is used:

16
fk = (3.10)
Rek

For Rek > 2000, the turbulent flow friction factor equation is used:

0.079
fk = 1/4
(3.11)
Rek

To implement the map, one first determines the Martinelli parameter X and F rG . Using these
two parameters on the top graph, if their coordinates fall in the annular flow regime, then the flow
pattern is annular. If the coordinates of X and F rG fall in the lower left zone of the top of the
graph, then K is calculated. Using K and X in the middle graph, the flow regime is identified
as either stratified-wavy or fully stratified. If the coordinates of X and F rG fall in the right zone
on the top graph, then T is calculated. Using T and X in the bottom graph, the flow regime is
identified as either bubbly flow or intermittent (plug or slug) flow.

3.2.4 Hashizume flow pattern map

Hashizume [59] in 1983 performed flow pattern observation experiments for refrigerant two-phase
flow in a horizontal tube. He showed that the flow pattern boundaries of refrigerant two-phase flows
differ considerably from those of Baker map, which was based on air-water data. He concluded that
the flow patterns of refrigerant two-phase flows can be presented on a revised Baker map, where
the property correction factor on surface tension was modified. The revised Baker map is shown
in Fig. (3.6).
The modified property correction factor proposed is:

 2 1/3
σ
1/4  µ   ρ

water L water
ψ = (3.12)
σ µwater ρL

Because the correction term (σwater /σ) for the air-water system is nearly unity, the modified Baker
map with Eq. (3.12) is practically identical to the original Baker map for air-water system.
42 CHAPTER 3. OVERVIEW OF FLOW PATTERN MAPS

Figure 3.6: Hashizume (1983) flow pattern map for horizontal flow in a tube.

In fact, Hashizume was one of the first to propose a flow pattern map based on observations for
refrigerants instead of air-water as in most of previous work.

3.2.5 Kattan-Thome-Favrat flow pattern map

These above flow patterns maps were all developed for adiabatic two-phase flows but are often
extrapolated for use with the diabatic process of evaporation. As with any extrapolation, this may
or may not produce reliable results. Important factors influencing the flow during evaporation,
which may have an effect on transition between flow regimes, are nucleate boiling, evaporation of
liquid films and the accelerational of the flow due to the phase change. It is desirable to define for
this type of flow a flow pattern map that includes the influences of heat flux and dryout on the flow
pattern transition boundaries and one one which is also easier to implement than the frequently
used log − log format. As a first step in this direction Kattan,Thome and Favrat [76, 77, 78]
proposed a modification of the Steiner [122] map, which is itself a modified Taitel-Dukler map,
and included a method for predicting the onset of dryout at the top of the tube in evaporating
annular flows. This flow pattern map will be presented here and the corresponding flow regime
transition boundaries are depicted in Fig. (3.7) (bubbly flow occurs at very high velocities and is not
shown). This map provides the transition boundaries (calculated from their underlying transitions
equations) on a linear-linear graph with mass velocity plotted versus gas or vapor fraction for the
particular fluid and flow channel, which is much easier to use than the log − log format of other
maps.
For the sake of simplicity, the transition between stratified and stratified-wavy flow will be des-
ignated as ”S-SW”, between stratified-wavy and intermittent/annular as ”SW-I/A”, between
3.2. FLOW PATTERN MAPS IN HORIZONTAL FLOW 43

intermittent and annular as ”I-A”, between annular and mist flow ”A-M” and between intermit-
tent and bubbly flow as ”I-B”.

Flow pattern map − R−410A, Tsat=5 oC, D=13.84mm

700

M
600

500
q=7.5kW/m2

I A
Mass Velocity [kg/m2s]

400
q=17.5kW/m2

300 q=37.5kW/m2
Steiner Kattan

200

SW
100

S
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Vapor quality [−]

Figure 3.7: Kattan (1998) flow pattern map (solid lines) evaluated for refrigerant R-410A at
Tsat = 5o C in 13.84 mm internal diameter tube.

The transition ”S-SW” is given by the expression:

 1/3
(226.3)2 ALD A2GD ρG (ρL − ρG )µL g
Gstrat = (3.13)
x2 (1 − x)π 3

The transition ”S-I/A” is:

   −F2 (q) 0.5

16A3GD gDρL ρG π2 We
Gwavy = (1 − x)−F1 (q) +1 + 50 (3.14)
x2 π 2 (1 − (2hLD − 1)2 )0.5 25h2LD Fr L

The high vapor quality portion of this curve depends on the ratio of the Froude number (F rL )
to the Weber number (W eL ), where F rL is the ratio of the inertia to the surface tension forces
while W eL is the ratio of inertia to gravity forces. The transition ”A-M” is given by the following
expression:

   0.5
7680A2GD gDρL ρG Fr
Gmist = (3.15)
x2 π 2 ξP h We L
44 CHAPTER 3. OVERVIEW OF FLOW PATTERN MAPS

Evaluating the above expression for the minimum mass velocity of the mist transition gives the
value of xmin , and when x > xmin :
Gmist = Gmin (3.16)

The transition threshold into bubbly flow (not shown in Fig. (3.7)) is given by:

 1/1.75
256AGD A2LD D1.25 ρL (ρL − ρG )g
Gbubbly = (3.17)
0.3164(1 − x)1.75 π 2 PiD µ0.25
L

In the above equations, the ratio of W e to F r is

 
We gD2 ρL
= (3.18)
Fr L σ

and the friction factor is

  −2
π
ξP h = 1.138 + 2log (3.19)
1.5ALD

The non-dimensional empirical exponents F1 (q) and F2 (q) in the ”S-SW” boundary equation
include the effect of heat flux on the onset of dryout of the annular film, i.e. the transition from
annular flow into annular flow with partial dryout around the upper perimeter of thr tubr, the
latter is classified as stratified-wavy flow by the map. They are given as:

 2  
q q
F1 (q) = 646.0 + 64.8 (3.20)
qcrit qcrit
 
q
F2 (q) = 18.8 + 1.023 (3.21)
qcrit

The Kutateladze [85] correlation for the heat flux at the departure from nucleate boiling, qcrit , is
used to normalize the local heat flux:

1/2
qcrit = 0.131ρG hLG [g(ρL − ρG )σ]1/4 (3.22)

The transition ”I-A” is assumed to occur at a fixed value of the Martinelli parameter, Xtt equal to
0.34. Solving for x in Eq. (2.37), the threshold line intermittent-to-annular flow transition at xIA
is:

  −1/1.75  −1/7  −1

ρG µL
xIA = 0.2914 +1 (3.23)
ρL µG

Fig. (3.8) defines the geometrical dimensions of the flow where PL is the wetted perimeter of the
tube, PG is the dry perimeter in contact with only vapor, hL is the height of the completely stratified
liquid layer, and Pi is the length of the phase interface. Similarly AL and AG are the corresponding
3.2. FLOW PATTERN MAPS IN HORIZONTAL FLOW 45

Vapor
PV
AV
qdry

D
qwet

AL Pi

hL
Liquid
PL
Figure 3.8: Cross-sectional and peripheral fractions in a circular tube.

cross-sectional areas. Normalizing with the tube internal diameter, D, six dimensionless variables
are obtained:

hL PL PG Pi AL AG
hLD = ; PLD = ; PGD = ; PiD = ALD = ; AGD = ; (3.24)
D D D D D2 D2

The dimensionless peripheral and cross-sectional area variables that are required for analysis can
be derived from the geometry for a given liquid height hL or for a reference liquid level hLD .
For hLD ≤ 0.5:

 
PLD = 8(hLD )0.5 − 2(hLD (1 − hLD ))0.5 /3 (3.25)

PGD = π − PLD (3.26)

 
ALD = 12(hLD (1 − hLD ))0.5 + 8(hLD )0.5 hLD /15 (3.27)

π
AGD = − ALD (3.28)
4

For hLD > 0.5:

 
PGD = 8(hLD )0.5 − 2(hLD (1 − hLD ))0.5 /3 (3.29)

PLD = π − PGD (3.30)

 
AGD = 12(hLD (1 − hLD ))0.5 + 8(hLD )0.5 hLD /15 (3.31)

π
ALD = − AGD (3.32)
4
46 CHAPTER 3. OVERVIEW OF FLOW PATTERN MAPS

For 0 ≤ hLD ≤ 1:

PiD = 2 (hLD (1 − hLD ))0.5 (3.33)

Since hL is unknown, an iterative method utilizing the following equation is necessary to calculate
the reference liquid level hLD :

 1/4     1/4  
PGD + PiD π2 PGD + PiD PiD π 64A3LD
Xtt2 = + (3.34)
π 64A2GD AGD ALD PLD π 2 PLD

Once the reference liquid level hLD is known, the dimensionless variables are calculated from
Eqs. (3.25) to (3.33) and the transition curves for the flow pattern map are determined with
Eqs. (3.13) to (3.23).
An improved version of this map was proposed by Zürcher et al. [165] (1999), Zürcher [166] (2000)
and Zürcher et al. [163] (2002), respectively. This improved version of Kattan flow pattern map
respects their ammonia flow pattern observations but is in fact quite complex to implement. It
used the Taitel and Dukler [125] (1976) void fraction model for fully stratified flows, the Rouhani-
Axelsson [110] (1970) void fraction model for intermittent and annular flows and interpolates be-
tween these two for stratified-wavy flows. Also, its includes a dissipation function in the ”SW-A/I”
transition curve. The entire set of equations of this improved flow pattern map must be iteratively
solved to find the transition curves, which is physically logical but difficult to implement for general
practice.

3.2.6 Thome and El Hajal flow pattern map

An easier to implement version of the above maps was proposed by Thome and El Hajal [131].
In the previously presented flow pattern maps, the dimensionless variables ALD , AGD , hLD and
PiD were calculated following an iterative procedure using the stratified flow void fraction model
of Taitel-Dukler [125]. On the other hand, the flow pattern based heat transfer model of Kattan et
al. [78] uses the Steiner [122] version of the Rouhani-Axelsson [110] drift flux model for horizontal
tubes to predict the void fraction
:

   −1
x x (1 − x) 1.18(1 − x)[gσ(ρL − ρG )]0.25

= (1 + 0.12(1 − x)) + + (3.35)
ρG ρG ρL G2 ρ0.5
L

This drift flux void fraction model is easy to apply and gives the void fraction as an explicit function
of total mass flux, which the method of Taitel-Dukler does not. Therefore, it makes sense to use the
same void fraction model in both the flow pattern map and the flow boiling heat transfer model,
for which the Rouhani-Axelsson model is a better choice as a general method. This was latter
proven experimentally by making 238 void fraction measurements for R-22 and R-410A in Wojtan
et al. [154].
Then, knowing the cross-sectional area of the tube, the values ALD and AGD are determinable as:

A(1 −
)
ALD = (3.36)
D2
3.2. FLOW PATTERN MAPS IN HORIZONTAL FLOW 47

Figure 3.9: Stratified angle in two-phase flow.

A

AGD = (3.37)
D2

The dimensionless liquid height hLD and the dimensionless length of the liquid interface PiD can
be expressed as a function of stratified angle θstrat :

  
2π − θstrat
hLD = 0.5 1 − cos (3.38)
2

 
2π − θstrat
PiD = sin (3.39)
2

Avoiding any iteration, the stratified angle θstrat can be calculated from an approximate expression,
evaluated in terms of void fraction, proposed by Biberg [16] as follows:

  3π 1/3 
θstrat = 2π − 2 π(1 −
) + 2 [1 − 2(1 −
) + (1 − −
)1/3
1/3 ] (3.40)
1
− 200 (1 −
)
[1 − 2(1 −
)[1 + 4((1 −
)2 +
2 ]

The transition curve ”SW-I/A” is determined using the updated expression of Zürcher et al. [165]
for Gwavy , where Gwavy is in kg/m2 s:

   −F2 (q) 0.5

16A3GD gDρL ρG π2 We
Gwavy = (1 − x)−F1 (q) 1 (3.41)
x2 π 2 (1 − (2hLD − 1)2 )0.5 25h2LD Fr L
„ «
(x2 −0.97)2
− x(1−x)
+50 − 75 e

Similarly, the transition curve from ”S-SW” is determined using the other expression of Zürcher et
al. [165] for Gstrat , where Gstrat is in kg/m2 s:

 1/3
(226.3)2 ALD A2GD ρG (ρL − ρG )µL g
Gstrat = + 20x (3.42)
x2 (1 − x)π 3
48 CHAPTER 3. OVERVIEW OF FLOW PATTERN MAPS

Figure 3.10: Thome-El Hajal (2002) flow pattern map or refrigerant R-22 at Tsat = 5o C and
qf lux = 7.5 kW/m2 in 13.84 mm internal diameter tube.

The non-dimensional empirical exponents F1 (q) and F2 (q) were added to the Gwavy expression to
include the effect of heat flux on the onset of dryout of the annular film, i.e. the transition from
annular flow into annular flow with partial dryout. These parameters are determined as:

 2  
q/2 q/2
F1 (q) = 646.0 + 64.8 (3.43)
qcrit qcrit

 2
q/2
F2 (q) = 18.8 + 1.023 (3.44)
qcrit

where again the Kutateladze correlation Eq. (3.22) for the critical heat flux qcrit was used to
normalize the local heat transfer.
The ”A-M”, ”I-A” and ”I-B” transition boundaries have not been changed and it is recommended
to use them in the same form proposed by Kattan et al. [76].

3.2.7 Wojtan-Ursenbacher-Thome flow pattern map

Based on information obtained from dynamic void fraction measurements and observations of the
cross-sectional locus of the liquid-vapor interface, a more recent version of the Kattan et al. flow
map has been proposed by Wojtan et al. [155], and also includes the effect of heat flux on the
transition to mist flow.
3.2. FLOW PATTERN MAPS IN HORIZONTAL FLOW 49

This new map is illustrated in Fig. (3.11) where S= stratified flow, SW= stratified-wavy flow, I=
intermittent flow, A= annular flow, M= mist flow and D represents the transition zone between
annular and mist flow.

Figure 3.11: Flow pattern map for R-22 at Tsat = 5o C in a 13.84 mm internal diameter tube at
G = 100 kg/m2 s and q = 2.1kW/m2 .

The implementation procedure for the updated version is as follows:

1. First, the geometrical parameters
, ALD , AV D , θstrat , hLD and Pid are calculated from the
following equations:

   −1
x x (1 − x) 1.18(1 − x)[gσ(ρL − ρG )]0.25

= (1 + 0.12(1 − x)) + + (3.45)
ρG ρG ρL G2 ρ0.5
L

A(1 −
)
ALD = (3.46)
D2

A

AGD = (3.47)
D2
  1/3 
θstrat = 2π − 2 π(1 −
) + 3π 2 [1 − 2(1 −
) + (1 −
)1/3 −
1/3 ]
(3.48)
1
− 200 (1 −
)
[1 − 2(1 −
)[1 + 4((1 −
)2 +
2 ]
  
2π − θstrat
hLD = 0.5 1 − cos (3.49)
2
50 CHAPTER 3. OVERVIEW OF FLOW PATTERN MAPS

 
2π − θstrat
PiD = sin (3.50)
2

2. The S-SW transition is calculated from:

 1/3
226.32 ALD A2GD ρG (ρL − ρG )gµL
Gstrat = (3.51)
x2 (1 − x)π 3

where Gstrat = Gstrat (xIA ) at x < xIA .

The flow is stratified whenever G < Gstrat .

3. The SW-I/A boundary is calculated from the following equation:

   −1 0.5
16A3GD gDρL ρG π2 We
Gwavy = +1 + 50 (3.52)
x2 π 2 (1 − (2hLD − 1)2 )0.5 25h2LD Fr L

The stratified-wavy region lies above the stratified region and is then subdivided into three
zones:

• G > Gwavy (xIA ) gives the SLUG zone;

• Gstrat < G < Gwavy (xIA ) and x < xIA give the SLUG/STRATIFIED-WAVY zone;
• x ≥ xIA gives the STRATIFIED-WAVY zone.

4. The I-A transition is calculated from the equation below and is extended down to its inter-
section with Gstrat :

  −1/1.75  1/7  −1

1/0.875 ρG µG
xIA = 0.34 +1 (3.53)
ρL µL

5. The A-D boundary is calculated from:

⎡   0.58  
−0.17 ⎤0.926
1
0.235 ln + 0.52 ρD ·
Gdryout = ⎣
−0.70 ⎦
x σ

−0.37
G−0.25 (3.54)
1 ρG q
gDρG (ρL −ρG ) ρL qcrit

6. The D-M is calculated from:

⎡   0.61  
−0.38 ⎤0.943
1
0.0058 ln + 0.57 ρD ·
Gmist = ⎣
−0.27 ⎦
x Gσ

−0.15
0.09 (3.55)
1 ρG q
gDρG (ρL −ρG ) ρL qcrit

The following conditions are then applied to define the transitions in the high quality range:
If Gstrat ≥ Gdryout , then Gdryout = Gstrat
If Gwavy ≥ Gdryout , then Gdryout = Gwavy
3.3. CONCLUSIONS 51

The maximum values of x to use in Eqs. (3.54) and (3.55) are 0.99
This updated version provides a more accurate prediction of different flow regimes (in particular
the onset and completion of dryout around the tube perimeter) and does not require any iterative
calculations. Therefore, it can be easily used for flow regime identification.

3.3 Conclusions

Many flow pattern maps are available for predicting adiabatic two-phase flow regimes in horizontal
tubes, but most of them were developed based on air-water data and just few were specifically de-
veloped for refrigerants. In order to overcome this difficulty, some empirical factors were introduced
to extrapolate one of these air-water maps to refrigerants. Another important characteristic is that
most maps were developed for adiabatic conditions and then extrapolated to diabatic conditions.
As it has been pointed out previously, the extrapolation procedure may not always produce reliable
results.
The original Kattan-Thome-Favrat flow pattern map and their respective updates, were developed
specifically for refrigerants under diabatic and adiabatic conditions, overcoming thus the two draw-
backs previously mentioned. Furthermore, the Wojtan-Ursenbacher-Thome version of the original
Kattan-Thome-Favrat flow pattern map includes the influences of heat flux and dryout on the
flow pattern transition boundaries providing a much more accurate prediction of the flow regimes.
Finally this map avoids any iterative calculations and thus can be easily used for flow regime
identification.
52 CHAPTER 3. OVERVIEW OF FLOW PATTERN MAPS
Chapter 4

Two-Phase Pressure Drop Models

Despite numerous theoretical and experimental investigations, no general models are available that
reliably predict two-phase pressure drops. A reason for this is that two-phase flow includes all
the complexities of single-phase like non-linearities, transition to turbulence and instabilities plus
additional two-phase characteristics like motion and deformation of the interface, non-equilibrium
effects and interactions between phases. The prediction of this important design parameter is made
by one of three approaches: empirical correlations, analytical models or phenomenological
models. In this chapter these different approaches are examined and principal advantages and
disadvantages presented. Also in this chapter leading existing methods are presented and critically
reviewed.

4.1 Two-phase pressure drops

The total pressure drop of a fluid is the sum of the variation of potential energy of the fluid, kinetic
energy of the fluid and that due to friction on the channel walls. Thus, the total pressure drop
∆ptotal is the sum of the static pressure drop (elevation head) ∆pstatic , the momentum pressure
drop (acceleration) ∆pmom , and the frictional pressure drop ∆pf rict :

∆ptotal = ∆pstatic + ∆pmom + ∆pf rict (4.1)

For a horizontal tube, there is no change in static head so ∆pstatic = 0. The momentum pressure
drop reflects the change in kinetic energy of the flow and is given by:

    
2 (1 − x)2 x2 (1 − x)2 x2
∆pmom = G + − + (4.2)
ρL (1 −
) ρG
out ρL (1 −
) ρG
in

where G is the total mass velocity of liquid plus vapor and x is the vapor quality.
The momentum pressure drop reflects the change in kinetic energy of the flow and is calculable,
for evaporating flows in horizontal tubes, by input of the inlet and outlet vapor qualities and
void fractions. When measuring two-phase pressure drop for evaporation in horizontal tubes, the
frictional pressure is obtainable by subtracting the momentum pressure drop from the measured
total pressure drop from the measured total pressure drop since the static pressure drop is zero.

53
54 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

4.2 Empirical methods

This is a commonly used approach in modelling two-phase pressure drop. The main reason in
following this approach is that minimum knowledge of the system characteristics is required. Thus,
empirical models are easy to implement and often they provide good accuracy in the range of the
database available for the correlations development. As a consequence, one of principal disadvan-
tages of this approach is that they are limited by the range of their underlying database. Another
important disadvantage of the empirical approach is that no single correlation is able to provide
an acceptable accuracy for general sense.
Below, the most quoted empirical methods are presented and critically reviewed.

4.2.1 Lockhart-Martinelli [90]

This work will be considered first because of its extensive historical use and the continued references
to it in the literature. In this treatment they postulated that two-phase flow could be divided into
four flow regimes:(1) liquid and gas both turbulent (tt), (2) liquid turbulent and gas viscous (tv),
(3) liquid viscous and gas turbulent (vt), (4) liquid and gas both viscous (vv).
The changeover points were selected to be consistent with single-phase flow and to give the best
correlation of the experimental data. There are two basic postulates on which the analysis is
based: (1) The static pressure drop for liquid and gaseous phases must be equal regardless of the
flow pattern; (2) The volume occupied by the liquid plus the gas at any instant (position) must
equal the total volume of the pipe. These postulates imply that the flow pattern does not change
along the tube length. In effect they eliminate those flows which have large pressure fluctuations,
such slug and intermittent flows, and those with radial pressure gradients, such as stratified and
stratified-wavy flows.
In the final correlation, the two-phase frictional pressure drop based on a two-phase multiplier for
the liquid-phase, or the vapor-phase, respectively, is:

∆pf rict = φ2L ∆pL (4.3)

∆pf rict = φ2G ∆pG (4.4)

where the terms ∆pL and ∆pG represent the frictional two-phase pressure drop that would exist if
the the flow as a liquid or gas, respectively, were assumed to flow alone in the entire cross-section
of the channel, and are calculated as:

   
L 2 2 1
∆pL = 4fL G (1 − x) (4.5)
D 2ρL
   
L 2 2 1
∆pG = 4fG G x (4.6)
D 2ρG

The single-phase friction factors of the liquid fL and vapor fG and Reynolds numbers are calculated
using the classical definition with their respective physical properties:
4.2. EMPIRICAL METHODS 55

0.079 G(1 − x)D

fL = where ReL = (4.7)
Re0.25
L µL

0.079 GxD
fG = where ReG = (4.8)
Re0.25
G µG

Then, Martinelli and his co-workers argued that the two-phase friction multipliers φ2L and φ2G could
be correlated uniquely as a function of Martinelli parameter X, which represents the ratio between
the theoretical pressure gradients which would occur if either fluid were flowing alone in the pipe
with the original flow rate of each phase and is calculated with Eq. (2.33), accounting for the flow
regime of each phase. This was verified using their experimental data.

Figure 4.1: Lockhart-Martinelli correlation.

The resulting graphical correlation is shown in Fig. (4.1) where φ (note: not φ2 ) is plotted against
X. All flow regimes were correlated in this manner and the corresponding φ2L and φ2G can be related
to the parameter X by relationships of the form

C 1
φ2L = 1 + + 2 (4.9)
X X

φ2G = 1 + CX + X 2 (4.10)
56 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

The curves in Fig. (4.1) are well represented by Eqs. (4.9) and (4.10) when C takes the following
values:

liquid gas C
turbulent turbulent (tt) 20
viscous turbulent (vt) 12
turbulent viscous (tv) 10
viscous viscous (vv) 5

Table 4.1: Values of C to fit the empirical curves of Lockhart and Martinelli.

To use this method to calculate the two-phase frictional pressure drop, it is only necessary to
calculate the frictional pressure drop for each phase flowing alone in the channel and then use
Fig. (4.1) or alternatively Eqs. (4.9) and (4.10). The correlation was developed for horizontal
two-phase flow of two-component systems at low pressures (close to atmospheric) and its application
to situations outside this range of conditions is not recommended.

4.2.2 Bankoff [9]

Bankoff made an extension of the homogeneous two-phase model by including some two-dimensional
effects. He derived expressions for the axial variation of velocity and void fraction in a tube. The
assumption was that the axial variation could be determined using a power law function. His
two-phase frictional pressure gradient is

   
dp dp 7/4
= φBf (4.11)
dz f ric dz L0

The liquid frictional pressure gradient is

 
dp 2G2
= fL0 (4.12)
dz L0 DρL

and the liquid friction factor fL0 is calculated as follows:

0.079 GD
fL0 = where ReL0 = (4.13)
Re0.25
L0 µL

His two-phase multiplier is

   3   
1 ρG 7 ρL
φBf = 1−γ 1− 1+x −1 (4.14)
1−x ρL ρG

where
4.2. EMPIRICAL METHODS 57

0.71 + 2.35 ρρGL

γ=   ρG
(4.15)
1 + 1−xx ρL

This method was derived using data for steam-water mixtures and is applicable to vapor qualities
from 0 < x < 1.

4.2.3 Cicchitti et al. [34]

Cicchitti et al. developed a simple correlation for steam-water systems. The experimental test
conditions from which this correlation was developed were far from the normal evaporation situation
in an ordinary evaporator. Despite this, their correlation has proven to work reasonably well for
other systems and is therefore included in this work. Their correlation for predicting frictional
pressure drop is as follows:

 
dp 0.092 1.8
= G [µL − x(µL − µG )]0.2 [νL + x(νG − νL )] (4.16)
dz f rict D1.2

where νG and νL are the specific volumes of gas and liquid, respectively.

4.2.4 Thom [129]

Thom proposed a simplified scheme for the calculation of pressure drop during forced circulation of
a two-phase flow of boiling water and steam. He gave curves from which frictional, accelerational,
and gravitational two-phase pressure drops could be estimated from the outlet vapor quality and
the operating pressure. These curves were derived using an extensive set of experimental data for
steam-water pressure drops on heated and unheated horizontal and vertical tubes.
The pressure drop is given by the following expression:

∆ptotal = ∆pstatic + ∆pmom + ∆pf rict =

     
2 1 L 2 1
G r2 + 4fL0 G r3 + ρL gLr4 (4.17)
ρL D 2ρL

where the values r2 , r3 and r4 are determined from curves in Fig. (4.2) and the friction factor fL0
is obtained with Eq. (4.13).

4.2.5 Pierre [107]

Pierre measured both heat transfer and pressure drop for refrigerants evaporating in horizontal
tubes. He derived an equation that would predict both the frictional pressure drop and accelera-
tional pressure drop. From an energy equation Pierre derived the following expression for the total
pressure drop:
58 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

Figure 4.2: Multipliers r2, r3, and r4 for boiling flow of water and steam (Thom 1964).
4.2. EMPIRICAL METHODS 59

 
(xout − xin )D G2 xνG L
∆ptotal = ∆pf rict + ∆pmom = fBP + (4.18)
xL D

The two-phase friction factor for pure refrigerant fBP , valid for the case when Re/Kf > 1, is
expressed as:

fBP = 0.0185(Kf /ReL0 )0.25 (4.19)

where, ReL0 is the all liquid Reynolds number calculated using Eq. (4.13) and Kf is the Pierre
boiling number defined as follows

∆xhLG
Kf = (4.20)
Lg

4.2.6 Baroczy [11]

Baroczy described a systematic correlation for the prediction of two-phase frictional pressure drop
for both single component flow and two-component flow. The correlation considered fluid proper-
ties, vapor quality and mass velocity.
The method of calculation employs two separate sets of curves. The first of these plots, Fig. (4.3),
is a plot of the two-phase frictional multiplier φ2f o as a function of a physical properties index
[(µf /µg )0.2 (vf /vg )] with vapor quality x as a parameter for a reference mass velocity of 1356 kg/m2 s.
The second, Fig. (4.4), is a plot of a correction factor Ω expressed as a function of the same
physical property index for mass velocities of 339, 678, 2712, and 4068 kg/m2 s with vapor quality
as a parameter. This plot serves to correct the value of φ2f o obtained from Fig. (4.3) to the
appropriate value of mass velocity. Thus,

 
dp 2ff o G2 νf 2
= φf o(G=1356) Ω (4.21)
dz f rict D

where νf (νL ) and νg (νG ) are the respective vapor and liquid specific volumes and ff o is obtained
with Eq. (4.13).
The correlation was based on data for steam, water-air, and mercury-nitrogen for a wide range
of vapor qualities and mass velocities. His method was tested against a wide range of systems
including both liquid metals and refrigerants with satisfactory agreement between the measured
and calculated values. This correlation has the disadvantage of being graphic in nature.

4.2.7 Chawla [27]

Chawla suggested the following method based on the vapor pressure gradient:

   
dp dp
= φChaw (4.22)
dz f rict dz G0
60 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

Figure 4.3: Two-phase frictional pressure drop correlation (Baroczy 1965).

4.2. EMPIRICAL METHODS 61

Figure 4.4: Mass velocity correction vs property index (Baroczy 1965).

62 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

The vapor frictional pressure gradient is calculated from:

 
dp 2G2
= fG0 (4.23)
dz G0 DρG

where the vapor friction factor fG0 is obtained as follows:

0.079 GD
fG0 = where ReG0 = (4.24)
Re0.25
G0 µG

His two-phase multiplier is

  2.375
1.75 1 − x ρG
φChaw = x 1+S (4.25)
x ρL

and his slip ratio S is:

uG 1
S= = 
−0.9
−0.5  (4.26)
uL 1−x −0.167 ρL µL
9.1 x (ReG0 F rH ) ρG µG

where ReG0 = GD/µG and F rH is determining using the following expression:

G2
F rH = (4.27)
gDρ2h

using the following expression to determine the homogeneous density ρh

 −1
x 1−x
ρh = + (4.28)
ρG ρL

His method is valid for vapor qualities from 0 < x < 1. Due to fact that this model uses the vapor
phase as the starting point, it works reasonably well for high vapor qualities in the annular flow
regime.

4.2.8 Chisholm [31]

Chisholm transformed the graphical procedure of Baroczy [11] to enable a more convenient applica-
tion to the case evaporating turbulent flow of two-phase mixtures in smooth tubes. His two-phase
frictional pressure gradient is given as

∆pf rict = φ2Ch ∆pL0 (4.29)

The frictional pressure drops for the liquid and vapor phases are calculated as follows:
4.2. EMPIRICAL METHODS 63

   
L 2 1
∆pL0 = 4fL0 G (4.30)
D 2ρL

   
L 2 1
∆pG0 = 4fG0 G (4.31)
D 2ρG

The friction factors are obtained for turbulent flows as:

0.079 GD
fL0 = where ReL0 = (4.32)
Re0.25
L0 µL

0.079 GD
fG0 = where ReG0 = (4.33)
Re0.25
G0 µG

while for laminar flows (Re < 2000) the corresponding friction factors are calculated with the
following expression:

16
f= (4.34)
Re

The flow is considered here fully turbulent at Re ≥ 2000 to avoid an undefined transition flow
interval interval in his method.
The parameter Y is obtained from the ratio of the frictional pressure drops as follows:

∆pG
Y2 = (4.35)
∆pL

Then, his two-phase multiplier is determined as:

φ2Ch = 1 + (Y 2 − 1)[Bx(2−n)/2 (1 − x)(2−n)/2 + x2−n ] (4.36)

where n is the exponent from the friction factor expression of Blasius (n = 0.25).
Finally, Chisholm’s parameter B is then determined following the procedure described below:
For 0 < Y < 9.5, B is calculated as:

55
B= for G ≥ 1900 kg/m2 s
G1/2
2400
B= for 500 < G < 1900 kg/m2 s (4.37)
G
B = 4.8 for G ≤ 500 kg/m2 s

For 9.5 < Y < 28, B is calculated as:

64 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

520
B= for G ≤ 600 kg/m2 s
Y G1/2
21
B= for G > 600 kg/m2 s (4.38)
Y

For Y > 28, B is calculated as:

1500
B= (4.39)
Y 2 G1/2

Once the value of B is determined, the two-phase multiplier can be calculated and therefore the
frictional pressure drop can be determined.
It is important to note that, while the procedures developed for smooth tubes gave satisfactory
agreement with steam/water mixtures in vertical channels, these procedures considerably under-
estimated experimental air-water flow data in horizontal tubes when the mass velocity was below
700 kg/m2 s.

4.2.9 Friedel [46]

One of the most accurate two-phase pressure drop correlations is said to be that of Friedel. It was
obtained by optimizing an equation for φ2f o using a large data base of two-phase drop measurements.
This method is for vapor qualities from 0 ≤ x < 1 and utilizes a two-phase multiplier as:

∆pf rict = ∆pL0 φ2f 0 (4.40)

where ∆pL0 is determined from Eq. (4.30) and the corresponding values of liquid friction factor
and liquid Reynolds numbers are obtained from Eq. (4.32). His two-phase multiplier is correlated
as:

3.24F H
φ2f 0 = E + 0.045 W e0.035 (4.41)
F rH L

where F rH , E, F and H are as follows:

G2
F rH = (4.42)
gDρ2h

ρL fG0
E = (1 − x)2 + x2 (4.43)
ρG fL0

F = x0.78 (1 − x)0.224 (4.44)

4.2. EMPIRICAL METHODS 65

 0.91  0.19  
ρL µG µG 0.7
H= 1− (4.45)
ρG µL µL

His liquid Weber W eL is defined as:

G2 D
W eL = (4.46)
σρh

and his definition of the homogeneous density ρh is:

 −1
x 1−x
ρh = + (4.47)
ρG ρL

The correlation is applicable to vertical upflow and to horizontal flow. This method is known to
work well when the ratio µL /µG < 1000, which is the case for most working fluids and operating
conditions.

4.2.10 Grönnerud [53]

This method was developed specifically for refrigerants and is as follows:

∆pf rict = φgd ∆pL0 (4.48)

and his two-phase multiplier is

⎡
⎤
  ρL
dp ⎢ ρG ⎥
φgd = 1 + ⎣
0.25 − 1⎦ (4.49)
dz Fr µL
µG

where Eq. (4.30) is used for ∆pL0 . His frictional pressure gradient depends on the Froude number
and is given as:

 
dp
= fF r [x + 4(x1.8 − x10 fF0.5r )] (4.50)
dz Fr

When applying this expression, if the liquid Froude number F rL ≥ 1, then the friction factor
fF r = 1.0, or if F rL ≤ 1, then:

 2
1
fF r = F rL0.3 + 0.0055 ln (4.51)
F rL

where

G2
F rL = (4.52)
gDρ2L
66 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

The correlation of Grönnerud is applicable to vapor qualities from 0 ≤ x < 1.

In a recent experimental study, Moreno Quibén and Thome [93] compared experimental results
from three different refrigerants and a wide range of operating conditions to three leading existing
methods (Friedel, Grönnerud, Müller-Steinhagen and Heck) recommended by Ould Didi, Kattan
and Thome [105]. Moreno Quibén and Thome showed that while all three provided a reasonable
agreement at one of the mass velocities studied, they either significantly overpredicted or under-
predicted the data at the other mass velocities. Also, the methods did not reliably capture the
variation in two-phase frictional pressure drop versus vapor quality. Despite these drawbacks, the
Grönnerud method was found to provide the most accurate predictions.

4.2.11 Müller-Steinhagen and Heck [97]

This method [97] proposed a two-phase frictional pressure gradient correlation that is in essence a
clever empirical interpolation between all liquid flow and all vapor flow:
 
dp
= F (1 − x)1/3 + Bx3 (4.53)
dz f rict

where the factor F is

F = A + 2(B − A)x (4.54)

The factors A and B are the frictional pressure gradients for all liquid (dp/dz)L0 and all vapor
(dp/dz)G0 flow and are defined as:

 
dp 2G2
= fL0 (4.55)
dz L0 DρL
 
dp 2G2
= fG0 (4.56)
dz G0 DρG

The friction factors are obtained with Eqs. (4.32) and (4.33).
This new correlation for the prediction of frictional pressure drop for two-phase flow in pipes
is simple and more convenient to use than other methods. To determine their reliabilities, the
authors checked the present correlation, as well as fourteen correlations from the literature, against
an extensive data bank with measurements of frictional pressure drop in pipes. They found that
the best agreement between predicted and measured values was obtained by the correlation of
Bandel [8] which, however, is quite complex to implement and use. The correlation suggested
in their paper still predicted the frictional pressure drop with reasonable accuracy. It includes
single-phase liquid and gas pressure drops and predicts correctly the influence of flow parameters.

4.3 Analytical methods

Two-phase pressure drop models developed following this approach are general models since no
empirical information is used in their development. Besides this obvious positive aspect, the corre-
4.4. PHENOMENOLOGICAL METHODS 67

sponding mathematical models are very complex and often iterative and numerical procedures are
required resulting in time consuming calculations.
Several two-phase pressure drop models developed following this approach can be found in litera-
ture. Nevertheless the complete set of equations describing these models, as well as the complex
resolutions procedures, will not be given in this section due to their size.

4.4 Phenomenological methods

Two-phase pressure drop models developed following a phenomenological approach are theoretical
based methods as the interfacial structure is taken into account. Thus, they are not blind to the
different flow regimes resulting in general applicability models. Despite this obvious advantage,
an important drawback of this approach is that some empiricism is still required in order to close
these models and be able to solve for the two-phase frictional pressure gradent. Another important
aspect is that no general flow pattern based model is yet available. In fact, they are only available
for individual flow patterns or flow structures. This observation preludes one of the major difficulty
in following this approach,that is they need a very reliable flow pattern map in order to be able
predict the different interfacial structures.
In what follows in this section, the existing two-phase pressure drop models developed pursuing
this approach are presented and critically reviewed.

4.4.1 Bandel [8]

Bandel assumed that the pressure drop is dependent only upon three different flow regimes, the
annular flow regime, the stratified flow regime and the transitional flow regime. He derived an
iterative procedure to calculate the frictional pressure drop for the annular flow regime and the
stratified flow regime. In these iterative procedures conditions for the different flow regimes were set
up, and from these conditions the flow regime was determined. If the flow is within the transitional
region, the pressure drop is determined via an interpolation from the results of the annular and the
stratified flow calculations.
The complete set of equations will not be given here due to their complexity and length but the
calculating procedure is given in a simple form in the paper by Müller-Steinhagen and Heck [97].

4.4.2 Beattie and Whalley [14]

Beattie and Whalley proposed a theoretically based flow pattern dependent calculation method. In
fact, they adapted an existing mixing-length theory model of two-phase pressure drop (Beattie [13])
to give a simple calculation method, of the same form as that used for single-phase flow, in which
flow pattern influences are partially allowed for in an implicit manner and therefore need not be
explicitly taken into account when using the method. They postulated that if it is assumed that
coefficients appearing in mixing-length theory are the same as those in the single-phase application
of the theory, loss of generality occurs. However the familiar Colebrook-White equation relating
friction factor f , Reynolds number Re, and surface roughness/diameter ratio k/D:
68 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

 
1 2k 9.35
√ = 3.48 − 4 log + √ (4.57)
f D Re f

remains valid for many two-phase situations provided flow pattern effects are considered in the
definition of the friction factor and Reynolds number. For Re and f they used the same defintions
as in the homogeneous model:

 
dp 2f G2
= (4.58)
dz f rict ρD

GD
Re = (4.59)
µ
−
and ρ is the homogeneous density given by:

 −1
− x 1−x
ρ= + (4.60)
ρG ρL

They proposed the following hybrid definition for the viscosity µ, allows flow pattern influences to
be taken into account

µ = µL (1 − β)(1 + 2.5β) + µG β (4.61)

where β is the homogeneous void fraction defined as:

ρL x
β= (4.62)
ρL x + ρG (1 − x)

A special feature of this model is the use of Eq. (4.57) for all values of Re, even for those where
single-phase behaviour would suggest its replacement by a laminar flow relation. This is because
turbulent-like characteristics can extend to very low Reynolds numbers in two-phase flows. This
is not surprising because even in the absence of turbulence, Reynolds stresses can be expected
in two-phase flows due to the interaction of gas-liquid interfaces with the flow field. The authors
compared their method to an extensive adiabatic round tube data bank and it showed to be, despite
its simplicity, as good as most alternative, more complex methods.

4.4.3 Hashizume et al. [61]

Based on experimental two-phase flow refrigerant data in a horizontal pipe, Hashizume [60] pro-
posed a modified version of Baker flow pattern map. The boundaries for the flow transitions in
their modified version were approximated by the following correlations:

Stratified to wavy: Y = [(13.6X −0.1776 )−1 + (972X −1.429 )−1 ]−1 (4.63)
−1.429
Wavy to slug: Y = 972X (4.64)
−0.1436
Wavy to annular: Y = 28.05X (4.65)
4.4. PHENOMENOLOGICAL METHODS 69

Here, X, Y, λ and ψ are

(1 − x)

X= λψ (4.66)
x

Gx
Y = (4.67)
λ

  1/2
ρG ρL
λ= (4.68)
ρAA ρW A

 1/4   2 1/3

σW A µL ρW A
ψ = (4.69)
σL µW A ρL

where σ represents the surface tension and the subscripts AA and W A denote the air and water
at ambient conditions, respectively. A graphical version of this modified Baker map was already
introduced in Chapter 2.
The authors argued that typical flow patterns in horizontal two-phase flow are annular and strat-
ified. On this basis, simplified models for annular and stratified flow were proposed. Since both
configurations consist of a continuous liquid (film) flow and a gas (core) flow, and their both have
a free surface as a boundary, their characteristics were supposed to be different from those of
single-phase flow. In fact, the corresponding velocity profiles were described with the Prandtl mix-
ing theory. From this analysis, and once the flow pattern configurations were determined using
the proposed map, the frictional pressure drop for both types of regimes were determined. The
intermediate region, i.e. wavy flow, was interpolated between annular and stratified flow. The
proposed simplified models for annular and stratified flows are presented below.
Annular flow
The proposed simplified model for annular flow consisted of a liquid film and a gas core. Entrain-
ment was neglected and the liquid film was assumed to be smooth and have uniform thickness in
the circumferential direction. With these assumptions and the mentioned mixing length theory,
the following equations were deduced:

Re+ /2
1
ReL = u+ dy + , for liquid phase (4.70)
4 0

Re+ /2  
1 x 2y +
ReL Γ= 1− u+ dy + , for gas phase (4.71)
4 1−x δ+ Re+

the two unknowns Re+ and δ+ can be solved for given values of ReL (= G(1 − x)D/ρL ), x and
Γ(= ρL /ρG ) using an iterative procedure. The frictional Reynolds number Re+ and dimensionless
liquid film thickness δ+ definitions are:

ρL u∗ D
Re+ = (4.72)
µL
70 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

ρL u∗ δ
δ+ = (4.73)
ρL
√
Here, u∗ is the friction velocity u∗ = gδ, u+ is the dimensionless velocity u+ = u/u∗ , and y +
dimensionless distance y + = ρL u∗ y/ρL .
Once Re+ value is known from the equations above, the pressure drop can be obtained from the
relationship between wall shear stress and pressure drop:

dp 4 4(Re+ )2 µ2L
= τw = (4.74)
dz D ρL D3

Stratified flow
In the same way as for annular flow, Re+ and δ+ can be calculated from Eqs. (4.70) and (4.71) for
given values of ReL , x and Γ.
In the case of stratified flow, flow in the pipe was modeled as flow between two parallel plates.
Therefore, the relationship between the distance between the upper and lower plates De and the
pipe diameter D has to be found. Comparison was made with experimental data on refrigerants in
the stratified region, presented in Hashizume [60], to seek a suitable equivalent value with which
the analytically obtained pressure drop agreed with experimental data. The following expression
was found:

De
= [(x5.0/Γ )−2 + (1.22x−1.42 )−2 ]−0.5 (4.75)
D

Transition flow
The variation of pressure drop versus mass flux under a constant quality showed characteristic
slopes on a log log plot. Each of these slopes corresponds to stratified, transient and annular flow.
Therefore, the expression below was proposed to calculate pressure drop in the transient region by
interpolation between stratified and annular flow

     log[(dp/dL)AC /(dp/dL)SC ] log[GAC /GSC ]

dp dp G
= (4.76)
dL transient dL SC GSC

where the subscripts SC and AC denote the critical values (at fixed vapor quality) for stratified
and annular flow. To predict GSC and GAC the modified Baker map is used.
Comparisons of this analysis with existing experimental data on refrigerants showed good agree-
ment.

4.4.4 Olujić [104]

Olujić proposed a general correlation based on the fact that in horizontal two-phase flow two
extremely different flow regimes exist: a β region, in which the velocities of the two phases are
nearly equal (bubble and plug flow), and an α region, in which the velocity of the gas phase
is higher than that of the liquid phase (wavy, slug and annular flow). In his approach, separate
4.4. PHENOMENOLOGICAL METHODS 71

Figure 4.5: Linear interpolation for transient flow (Hashizume 1985).

correlations were developed for the two regions, with the demarcation of the valid range expressed in
terms of a relationship for the ratio of phase-volume flow rates, and a dimensionless ratio containing
a modified Froude number.
Pressure drop in β-region:
In the range of low quality flow, where the average liquid velocity, uL , is approximately equal to
the average gas velocity, uG , the simplest approach is to consider the gas-liquid mixture to be a
homogeneous flow. However, the simple homogeneous model does not ensure accurate predictions
of frictional pressure drop. Hence, he used the following expression:

(dp/dz)f ric,β
= [1 − x(R − 1)(K2 − 1)] (4.77)
(dp/dz)f ric,h

The homogeneous frictional pressure drop (dp/dz)f ric,h is calculated from:

 
G2
(dp/dz)f ric,h = f [1 + x(R − 1)] (4.78)
2DρL

where R is the density ratio of the phases R = ρL /ρG , and f is the Darcy friction factor for smooth
and rough tubes estimated from:
   
k/D 5.02 k/d 14.5
f= −2 log − log + (4.79)
3.7 Re 3.7 Re

Here, k/D = relative roughness; Re = GD/µtp ; and µtp = µL [1 − x(1 − θ)] with θ = µL /µG .
72 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

The parameter K2 in Eq. (4.77) is determined from

(7 + 8n)(7 + 15n)
K2 = 1.2 (4.80)
(7 + 9n)(7 + 16n)

where n is calculated using

n = (0.617/β)[1 + (1 + 0.907β)1/2 ] (4.81)

and β is the volume-flow ratio of the phases

ρG ṀG
β= (4.82)
ρL ṀL

Pressure drop in α-region:

In the range uG >> uL , the frictional pressure drop, as well as the void fraction, is affected by the
momentum exchange between the phases. For this region the proposed correlation is:

 19/8
(1 − x)
(dp/dz)f ric,α = (dp/dz)SG 1+ (4.83)
xR
T

where the superficial frictional gas pressure drop (dp/dz)SG is calculated as follows:

 
(Gx)2
(dp/dz)SG =f (4.84)
2DρG

where the parameter
T is determined from the following equations:

T = (
−3 −3 −1/3
1 +
2 ) (4.85)

0.057 )

1 = 0.77R−0.55 Γ(0.266R (4.86)

0.078 )

2 = 2.19R−0.61 Γ(1.78R (4.87)

   −1/2  −1/8
1−x G2 (1 − x)2 ρL µL
Γ= (4.88)
x ρ2L gD ρG µG

ρL
R= (4.89)
ρG

The proposed method was compared with experimental data from different sources. In general, the
predictions agreed well with a variety of published data for adiabatic gas-liquid flows in horizontal
pipes.
4.4. PHENOMENOLOGICAL METHODS 73

4.4.5 ARS Model [58]

Hart, Hamersma and Fortuin [58] presented a phenomenological model, called ARS (Apparent
Rough Surface), to calculate the value of the frictional pressure drop in gas-liquid pipe flow with a
small liquid holdup,
L < 0.06 (
> 0.94), covering the stratified, wavy and annular flow regimes.
For this kind of flow it was shown that liquid, flowing along the tube wall, may be considered as
a local roughness over a fraction θ = α/2π of the tube wall, see Fig. (4.6). The interfacial shear
stress τi exerted by the gas on the liquid film may considered as shear stress exerted by the gas
phase on a rough wall.

Figure 4.6: Schematic representation of gas-liquid flow with a small liquid holdup
L in straight
smooth tubes (ARS 1989).

These assumptions led to the following correlation for the two-phase pressure drop ∆PT P :
 
L 1
∆PT P = 4fT P ρG u2G for uG  ui (4.90)
D 2

where fT P is the two-phase friction factor and uG is the vapor mean velocity defined in Chapter 2.
The complete procedure for the calculation of fT P and afterwards ∆PT P is as follows:

1. The value of the liquid holdup
L is calculated from:

   1/2 

L jL ρL
= 1 + 10.4Re−0.363 (4.91)
1 −
L jG SL
ρG

where superficial velocities jG and jL are calculated from Eqs. (2.9) and (2.10) introduced in
Chapter 2 and ReSL (= jL D/µL ) is the liquid Reynolds number based on superficial liquid
velocity.
2. Vapor mean velocity uG and liquid mean velocity uL are calculated using Eqs. (2.5) and (2.6).
3. Modified F r and the Re of the gas phase are calculated with
u2L ρL uG D
Fr = ReG = (4.92)
gD(ρL − ρG ) µG
74 CHAPTER 4. TWO-PHASE PRESSURE DROP MODELS

4. The value of the wetted wall fraction θ is calculated by means of the following correlation:
θ = θo + 0.26F r 0.58 , in which θo = 0.52
0.374
L (4.93)
If θ > 1 then θ = 1.
5. Apparent relative roughness of the liquid film k/D is determined from:
 

k δ L
= 2.3  2.3 (4.94)
D D 4θ
6. Interfacial friction factor fi is calculated with
0.0625
fi = 
2 (4.95)
15 k
log ReG + 3.715D

7. The value of the single-phase friction factor f is calculated with

0.07725
fG = 
2 (4.96)
log Re7G

which is valid for 2100 < ReG < 108 .

8. The two-phase friction factor fT P is calculated from proration of the respective wet and dry
perimeters as
fT P = (1 − θ)f + θfi (4.97)
9. Finally, the value of the two-phase pressure drop ∆PT P is calculated with Eq. (4.90).

This model has been verified with experimental data. These data refer to steady-state gas-liquid
flow through horizontal straight smooth tubes and have been obtained both from the literature and
from authors’ own experiments.

4.5 Conclusions

Existing two-phase pressure drop model have been presented and critically reviewed. The models
were classified according to the different approaches used for their development, that is empirical,
analytical and phenomenological. The main advantages and disadvantages of those approaches and
methods were emphasized. While empirical based model are the easiest to implement and provide
a reasonable accuracy within the range of database used in their implementation, their reliability
seemed to be the weakness link due to the amount of empirical information used. In contrast,
the analytical approach provides general models, since no empirical information is used, but their
implementation is very complex and time consuming just to do one calculation. A good compromise
between purely empirical and purely analytical seems to be the phenomenological approach. This
later preserves their general applicability, as they are theoretically based, and introducing some
empiricism they overcome complex calculations resulting in somewhat easy to use models. Despite
this, one needs to know in advance the interfacial structure, introducing the need of a reliable
flow pattern map to allow the prediction of flow patterns or flow regimes. Apparently, no flow
pattern based two-phase frictional pressure model that covers the important flow regimes occurring
in horizontal refrigerant evaporator tubes, is available in the literature. Thus, this proves the
analytical objective of the present work.
Chapter 5

Description of Experimental Test

Facility

The existing LTCM intube refrigerant test loop has been modified and adapted to the new test con-
ditions and measurement methods. Two new test sections have been successively implemented into
the modified test rig. In order to determine pressure drop separately for the different flow regimes,
the new test sections consist of both diabatic and adiabatic zones. The present configuration allows
tests to be run under diabatic and adiabatic conditions simultaneously. All modifications have been
made on the original test rig developed by Kattan [75] and modified afterwards by Zürcher [162].

5.1 General description

The objective of the experimental part of this study was to run in-tube evaporation tests over
a wide range of experimental conditions in order to obtain accurate values of two-phase pressure
drops. The ranges of the experimental conditions are shown in Table (5.1).

Experimental test conditions

Test fluids R134a, R22, R410A
Saturation temperature (Tsat ) 5o C
Internal diameter (d) 8.00, 13.8 mm
Vapor quality (x) 0 − 1.0 [−]
Mass velocity (G) 70 − 700 kg/m2 s
Heat flux (q) 6.0 − 57.5 kW/m2

Table 5.1: Experimental conditions for in-tube evaporation tests.

The existing LTCM test loop, developed by Kattan [75] and modified afterwards by Zürcher [162],
has been adapted to be able to run tests over a wide range of test variables. Two new test sections
have also been implemented into the modified test rig. The new configuration results in very
accurate values of two phase pressure drops. The test facility consists of a refrigerant circuit and
a heating water circuit. An overall view of the modified test facility is shown in Fig. (5.1).

75
76 CHAPTER 5. DESCRIPTION OF EXPERIMENTAL TEST FACILITY

Figure 5.1: Overall view of the modified test facility including the new two-zone test section.

5.2 Refrigerant circuit

A schematic of the refrigerant circuit is depicted in Fig. (5.2). The refrigerant passes first through
a series of electrical preheaters and then though an insulated tube before it enters the two zone test
section. In the first zone (called diabatic zone) refrigerant is heated by a counter-current flow of hot
water in the annulus of the double pipe system. Then, refrigerant passes through a visualization
zone and through the second part of the test section (called diabatic zone) where no heat is added
to or removed from the refrigerant. Refrigerant exits the test section and goes through a condenser,
a magnetically driven gear type pump and finally a Coriolis mass flow meter. The refrigerant circuit
also includes a vapor-liquid reservoir for controlling the amount of refrigerant circulating in the test
rig and thus the operating pressure.
A particularly detailed description of the new two-zone test section is given in Section (5.4) in this
chapter, while a description of other main components of the refrigerant circuit are given below:

Preheater

The electrical preheater is made of six horizontal stainless steel tubes. Internal and external di-
ameters of the tubes are 13 mm and 16 mm, respectively. The length of each tube is 1.40 m and
a 3 mm diameter and 10 m long hot wire is coiled around the tube. The total maximum heat flux
provided by the preheater is 61.2 kW/m2 . Thermocouples are installed on each tube to prevent
operation from reaching the critical heat flux (CHF).
5.2. REFRIGERANT CIRCUIT 77

P re h e a te r

F lo w m e te r 1 F lo w m e te r 2

H o t W a te r o u t H o t W a te r in
O il F r e e P u m p

S ig h t
g la s s
D ia b a tic T e s t S e c tio n A d ia b a tic T e s t S e c tio n
C o n d e n s e r R e fr ig e r a tio n
m a c h in e r y

R e s e r v o ir

Figure 5.2: Schematic of the refrigerant circuit.

Condenser

Refrigerant is condensed and subcooled by means of a stainless-steel shell and tube condenser.
Condensation takes place inside the tubes and the chilled liquid receives the heat. The condenser
is composed of 165 stainless steel tubes of 1.5 m length. External and internal tube diameters are
6.35 mm and 4.55 mm, respectively. The total area of heat exchange is 5 m2 .

Gear pump

To avoid the presence of oil in the experimental test rig, a magnetic drive gear pump has been
used. Both gears are made of nickel-molybdenum-chromium alloy compatible with refrigerants used
during the experimental campaign. The pump provided stable flow in the range from 0.15 kg/min
to 10 kg/min. The regulation of the flow can be done either by a by-pass around the pump or by
pump motor frequency variation.

Flow meters

Two Coriolis type flow meters were installed on the modified test rig. The two flow meters are
fixed securely to the wall to insure steady measurement conditions. A selection of valves allowed
switching between them depending on operating flow rates. The nominal measurement points are
1.5 kg/min and 10 kg/min for the small and big flow meter, respectively. The accuracy of the flow
meters is 0.15 % of the measured value. The channels connecting the pump to the flow meter and
flow meter to the preheater are heavily insulated to avoid heat gain from the surroundings.
78 CHAPTER 5. DESCRIPTION OF EXPERIMENTAL TEST FACILITY

5.3 Hot water circuit

A schematic of the hot water circuit is depicted in Fig. (5.3). The hot water circuit reheats water
after it passes through the diabatic zone of the two-zone test section. The hot water flows counter
currently to the refrigerant in the annulus of the diabatic zone of the test section. The water flow
rate is measured by a Coriolis flow meter and the water is circulated by a stainless steel pump. The
operating pressure of water flow is fixed to 2 bars. The inlet water temperature (which determines
heat flux) is controlled by two secondary water circuits using two plate heat exchangers connected to
an industrial cold water circuit. To insure the mixing of the heating water in the annular chamber,
the laminar condition has been avoided. The Reynolds number of flowing water was always higher
than 4500 for the experimental conditions described in Table (5.1).

Figure 5.3: Schematic of the hot water circuit ([75]).

5.4 Test sections

Two test sections have been constructed and implemented into the modified test rig. Each test
section consists of two separate zones called diabatic and adiabatic, respectively. In the first zone
of the test section (diabatic zone), refrigerant is heated by a counter-current flow of hot water in
the annulus of the double pipe system. In the second zone of the test section (adiabatic zone) no
heat is added or removed from the refrigerant. Both test sections were made of copper and have
plain smooth interiors. Internal diameters were 8.00 mm (TS1) and 13.8 mm (TS2), respectively.
The outer stainless steel annulus has an internal diameter of 14.00 mm for TS1 and 20.00 mm for
TS2. The internal tubes (only diabatic zone) are perfectly centered within the outer annulus tube
5.4. TEST SECTIONS 79

P P P

D p
D p

D p
D p
T *
W a te r o u t W a te r in
T u b e w a ll

R e fr ig e r a n t in R e fr ig e r a n t o u t

T r T w T w T w T r

F lo w v is u a liz a tio n
T w T w z o n e

D ia b a tic z o n e A d ia b a tic z o n e

T r R e fr ig e r a n t te m p e r a tu r e m e a s u r e m e n t
P P re s s u re m e a s u re m e n t
T w W a te r te m p e ra tu re m e a s u re m e n t
D p D iffe r e n tia l p r e s s u r e m e a s u r e m e n t
T * W a ll te m p e r a tu r e m e a s u r e m e n t

Figure 5.4: New two-zone test section.

in five positions using centering screws. The external surfaces of both zones of the test section
were heavily insulated to avoid heat gain from surroundings. The main physical properties and
geometrical dimensions of the implemented test sections are shown in Table (5.2), where: D is the
internal diameter of the tube, Dext is the external diameter of the tube, zdiab. is the length of the
diabatic zone, zadiab. is the length of the adiabatic zone, λ is the thermal conductivity and δwat is
the annular gap for the heating water in the diabatic zone.

Internal tube External tube

Mat. Dext D zdiab. zadiab. λ Mat. Dext D δwat
[mm] [mm] [mm] [mm] [W/m◦ K] [mm] [mm] [mm]
TS1 Cu 9.53 8.00 2035 980 339 SS 17.40 14.00 2.235
TS2 Cu 15.87 13.8 2026 989 339 SS 23.00 20.00 2.065

Table 5.2: Main properties and geometrical dimensions of the implemented test sections.

A schematic of the two-zone test section is depicted in Fig. (5.4). Refrigerant flows into the internal
tube from left to right, and hot water flows counter-currently into the annulus from right to left.
80 CHAPTER 5. DESCRIPTION OF EXPERIMENTAL TEST FACILITY

5.5 Experimental procedure and data acquisition

All the measurements are made with a computer equipped with a data acquisition system. The data
acquisition system is a National Instruments SCXI. The acquisition card is a PCI MIO 16XE 50
installed in the computer. The resolution of this card is 16 bits and the maximum acquisition
frequency on a single channel is 10 kHz. A SCXI 1000 module with four bays is connected to this
card. Each of the four bays has a 32 channel voltage measurement card (type 1102). The total
number of acquisition channels is thus 128. Each channel has a computer programmable gain:
1 for 0 to 10V signal (pressure transducer and mass flow meter) and 100 for low voltage signal
(thermocouples). The signals can be adjusted to the 0 to 10V range of the acquisition card in the
computer. A 2 Hz low pass frequency filter is also included in the card for each channel. This helps
to diminish the measurement noise and does not affect the steady-state measurement. At the end
of the acquisition chain, a terminal block with 32 sockets is connected to the 1102 card. Each card
has its own terminal block. The cold junction for every thermocouple is made in this terminal block
at the socket. The material for this socket is copper for both poles (+ and -), the continuity of the
two different specific materials of the thermocouple is then broken at this point located inside the
terminal block. The temperature of the 32 cold junctions is maintained uniform with a metallic
plate and is measured via a RTD installed in the middle. Additionally, all the terminal blocks are
placed in a close cupboard away from external thermal influences.
In order to measure a test parameter in a channel, 100 acquisitions are made in 0.02 s (50 Hz
electrical period) and the average of these 100 values is calculated during the acquisition. The
result is the measured value of the channel. By this way, the noise from alternating current on
the measured signal is removed. This value is stored and the system goes to the next channel.
With this measurement method, the theoretical channel measurement frequency is 50 channels per
second, but due to the switching time between channels, the actual frequency is 30 channels per
second. In total, it takes 4.3 s to measure all the channels of the acquisition system once. To obtain
one experimental point, 10 of such acquisition cycles are recorded and averaged.

5.6 Measurements

The major objective of the experimental part of this work was to measure two-phase pressure drops
over the vapor quality range from 0 to 1. Two-phase pressure drop values were directly obtained
from the differential pressure transducers. Meanwhile to completely establish the experimental
conditions, some others parameters need to be determined by calculation from measured values.

5.6.1 Pressure drop

The two-phase pressure drops across the diabatic test sections were each measured by a selection of
sensors depending on the level of the pressure drop being measured. For the diabatic test section
two differential pressure transducers (0 − 40 mbar and 0 − 500 mbar) were used for the small and
intermediate pressure drops and a pair of absolute pressure sensors for pressure drops larger than
500 mbar. For the adiabatic test section, two differential pressure transducers (0 − 20 and 0 − 160
mbar) were used.
5.6. MEASUREMENTS 81

5.6.2 Refrigerant temperature

The saturation temperature of the refrigerant Tsat was calculated using the the pressure measure-
ments at the inlet and outlet of the diabatic part of the test section and by the assumption of a
linear pressure distribution over the length of the diabatic test section as is typical in such tests.
Then, knowing the saturation pressure Psat the saturation temperature is obtained based on the
thermodynamical properties calculated using EES (linked to REFPROP 6.0 of NIST).

5.6.3 Heat Flux

Heat is transferred from the hot water to the refrigerant. The temperature of the water is measured
at five positions (represented as Tw in Fig. (5.4)) with a total of 26 thermocouples giving five local
water temperature values Tw . The enthalpy of the flowing water, assuming that the pressure drop
along the annulus is negligible is given for any location along the heated tube by the following
equation:

hw = cpwater Tw (5.1)

From Eq. (5.1), five discrete enthalpy values were obtained from the five water temperature mea-
surement points. These points were used to determine the enthalpy profile over the diabatic test
section. For the enthalpy profile determination, a second order polynomial fit has been used.
Then, the heat transferred from the water between two points of the test section can be defined as:

Q12 = (hw2 − hw1 )Ṁw (5.2)

The derivative of Eq. (5.2) with respect to axial position along the divided by the tube perimeter
gives the external heat flux qext along an elementary length dz:

 
1 dQ Ṁwat dhw (z)
qext = = (5.3)
πDext dz πDext dz

Thus, knowing the enthalpy profile hw (z), the external heat flux can be calculated at any point
along the diabatic test section. The internal heat flux provided to the refrigerant is determined as:

   
Dext Ṁw dhw (z)
q(z) = qext = (5.4)
D πD dz

5.6.4 Vapor quality

The vapor quality is calculated by an energy balance over the preheater and the test section. Hence,
the vapor at any test section position can be calculated fom the following equation:

P + Ṁw cw (Tw(inlet) − Tw (z))

x(z) = (5.5)
Ṁref hLV
82 CHAPTER 5. DESCRIPTION OF EXPERIMENTAL TEST FACILITY

where P is the electrical power provided from the preheater, Tw(inlet) is the temperature of the
heating water at the inlet of the diabatic test section, Tw (z) is the water temperature at any position
along the test section, cw is the water specific heat, Ṁw , Ṁref are the water and refrigerant mass
flow, respectively.

5.7 Accuracy of measurements

In order to sure that the entire measurement system is working correctly, liquid-liquid tests have
been done with subcooled R22 and R410A versus water for both test sections. The accuracy of
the energy balance between the heating and cooling fluids was thus determined. The results of this
tests are not shown here but Qw /Qref varied from 0.98 to 1.02, with a mean error of ±0.98% for
R410A and ±0.96% for R22 in the 13.8 mm test section. Similar results were obtained for the 8 mm
test section. This is very accurate for this type of measurement and confirms that all measurement
instruments were precisely calibrated and that the system was working correctly.

5.7.1 Pressure drop measurement accuracy

Two different types of pressure transducers were used to determine the two-phase pressure drop,
namely: absolute and differential. The pair of absolute transducers used in this study ranged
1 to 25 bars. They were calibrated with a very accurate balance and their accuracy was found
to be ±20 mbar. Four differential transducers were used for the small and intermediate range in
this study. They ranged 0 − 20 mbar and 0 − 40 mbar for the low range and 0 − 160 mbar and
0 − 500 mbar for the intermediate. An accuracy of ±0.05% F.S. was given by the supplier. After
inhouse calibration against a water column, this value was realistic and taken as the accuracy of
the transducers.

5.7.2 Heat flux measurement accuracy

For the local heat flux absolute error determination, first the derivative of the enthalpy was con-
verted into the derivative of the temperature of the heating water

dhw (z) dTw (z)

= cpw (5.6)
dz dz

Then, substituting into Eq. (5.4), local heat flux in each point of the test section can be calculated
with:

Ṁw dTw (z)

q(z) = cpw (5.7)
πD dz

and assuming that the measurements of the water temperature are linearly independent, the abso-
lute error of the heat flux ∆q can be determined with the following expression:

 2  2
dq(z) dq(z)
∆q(z) = ∆Ṁw + ∆Tw (5.8)
dṀw dTw
5.8. CONCLUSIONS 83

After calculation, the mean relative error does not exceed 2% and 1% for the 13.8 mm and 8 mm
test section, respectively.

5.7.3 Vapor quality measurement accuracy

Assuming that all measured values are linearly independent, the absolute vapor quality error at
each position of the evaporator based on the general definition applied to Eq. (5.5) can be expressed
as:


  2
2
2
2
 ∂x
 ∂P ∆P + ∂T 1∂x ∆T 1 + ∂x
∆T (z) + ∂x
∆ Ṁ
∆x(z) = 
water ∂Twat (z) wat wat
 ∂x
water
2
2 ∂ Ṁwat
(5.9)
+ ∂ Ṁ ∆Ṁref + ∂P∂xsat ∆Psat
ref

The term with the saturation pressure Psat in the vapor quality error calculation is encountered
because the latent heat of the refrigerant is determined from the measurement of the saturation
pressure.
The absolute uncertainty for the heating power measurements is ±20 W and the accuracy of the
temperature measurements is ±0.02o C. The absolute error of the absolute pressure transducers
after calibration has been estimated to be ±20 mbar. The refrigerant and water mass flow rates are
measured with an accuracy of ±0.15%. After calculating of partial derivatives and substituting all
values in Eq. (5.9), the maximum absolute errors for each tested mass velocity have been obtained.
The calculated mean relative errors does not exceed 2% and 3% for the 13.8 mm and 8 mm test
section, respectively.

5.8 Conclusions

The existing LTCM flow boiling test facility has been successfully modified and adapted to the
different test conditions and measurement methods. Two new test sections, 8.00 mm and 13.8 mm
plain horizontal tubes, have been successively implemented into the modified test rig. The new
test section consists in two zones: diabatic and adiabatic. This configuration allows tests to be run
that obtain experimental two-phase pressure drop values under diabatic and adiabatic conditions
simultaneously. Moreover, the addition of an adiabatic test section allows data to be obtained with
the same flow regime at the inlet and outlet of the test section.
84 CHAPTER 5. DESCRIPTION OF EXPERIMENTAL TEST FACILITY
Chapter 6

Experimental Results

This chapter focuses on experimental results obtained during the experimental campaign. First,
the data reduction procedure to obtain the two-phase frictional component of the measured values
is described and validated. Next, the influence of the different test variables is depicted for some
representative sets of experimental conditions. Finally, the entire results of the two-phase pressure
drop measurements are presented.

6.1 Reduction of experimental data

The general expression for describing the total two-phase pressure drop ∆ptotal is:

∆ptotal = ∆pstatic + ∆pmom + ∆pf rict (6.1)

where ∆pstatic is the elevation head pressure drop (presently ∆ptotal = 0 for a horizontal tube),
∆pmom is the momentum pressure drop created by acceleration of the flow in a diabatic process
and ∆pf ric is the two-phase frictional pressure drop.
The new two-zone test sections yields the two-phase pressure drops in the diabatic (first zone) and
adiabatic (second zone) simultaneously. The adiabatic two-phase pressure drops were obtained
at the vapor quality leaving the diabatic test section and for the reason mention above (adiabatic
process) the momentum pressure drop ∆pmom = 0. Hence, one can directly determine the two-phase
frictional pressure drops for the adiabatic test section from the measured values:

∆pf rict = ∆ptotal (6.2)

For the diabatic test section one determines the two-phase frictional pressure drops from the fol-
lowing equation:

∆pf rict = ∆ptotal − ∆pmom (6.3)

Hence, one must first evaluate the momentum pressure drop ∆pmom to obtain the frictional pressure
drop values. The momentum pressure drop in the diabatic test section reflects the increase in the
kinetic energy of the flow during the evaporation process and is given by the following expression:

85
86 CHAPTER 6. EXPERIMENTAL RESULTS

    
2 (1 − x)2 x2 (1 − x)2 x2
∆pmom = G + − + (6.4)
ρL (1 −
) ρG
out ρL (1 −
) ρG
in

where G is the total mass velocity of liquid plus vapor, x is the vapor quality at the inlet or outlet
of the test section,
is the vapor cross-sectional void fraction at the inlet or outlet of the test
section, ρL and ρV are the liquid and vapor densities and g is 9.81 m/s2 . Based on a recent work
by Wojtan-Ursenbacher-Thome [154], they found the Steiner version of the Rouhani-Axelsson drift
flux model to be very accurate for predicting void fractions in similar experimental conditions.
Hence the following expression is used to calculate the void fractions:

   −1
x x (1 − x) 1.18(1 − x)[gσ(ρL − ρG )]0.25

= (1 + 0.12(1 − x)) + + (6.5)
ρG ρG ρL G2 ρ0.5
L

This expression is used in Eq. (6.4) to determine the momentum pressure from the experimental
conditions.
Using water in the annulus of the double pipe system to evaporate the refrigerant, the water
undergoes a temperature change while the phase changing refrigerant stays at nearly the same
saturation temperature. This causes a change in the local heat flux as the temperature difference
between the water and the refrigerant decreases during the evaporation process along the length
of the tube. In order completely determine the momentum pressure drop created by acceleration
of the flow ∆pmom , one must known the vapor quality. As this is changing during the evaporation
process, the results are reported at the mean vapor quality in the diabatic test section based on an
evaluation (piecewise decomposition) of the enthalpy profile obtained from the water-side.
Figs. (6.1) and (6.2) show, for the 8 mm and 13.8 mm tube respectively, the total, momentum
and frictional pressure drops for a representative set of experimental conditions. The momentum
pressure drops were calculated following the procedure detailed previously and the total pressure
drops correspond to the measured values. As expected, the momentum pressure drop is larger for
high heat fluxes reflecting the increase in the kinetic energy due to the evaporation process and
vary from around 3% of the total pressure drop for the lowest heat fluxes up to 40% for the highest.
Fig. (6.3) shows a comparison, for a representative set of experimental conditions, of the two-phase
pressure gradients for the adiabatic test section versus those for the diabatic test section. The
agreement between the results is quite remarkable, attesting to the accuracy and reliability of the
measurements as well as the appropriate data reduction procedure. At high vapor qualities, the
difference between the two types of measurements is thought to be caused by the vapor quality
variation along the diabatic test section around the peak, which averages out the pressure gradient
rather than giving the ”local” value at a particular vapor quality as in the adiabatic test section.
For instance, in the diabatic test section at the vapor quality at the peak, part of the test section
will be operating at local conditions before the peak and part at local conditions after the peak,
resulting in a lower value than for the adiabatic test section at the same vapor quality. It is for this
range of test conditions that the adiabatic test section was added to the loop in order to obtain
data representative of the real pressure gradients at high vapor qualities near and after the peak.
6.1. REDUCTION OF EXPERIMENTAL DATA 87

40 40

o 2 o 2
35 R134a, d=13 mm, T=5 C, G=150 kg/m s 35 R134a, d=13 mm, T=5 C, G=150 kg/m s
2 2
q=06-09 kW/m , L=2.035 m q=15-20 kW/m , L=2.035 m
30 30
pressure drop [mbar]

pressure drop [mbar]

25 25
total total
momentum momentum
20 20
frictional frictional

15 15

10 10

5 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

30 30

o 2 o 2
R22, d=13 mm, T=5 C, G=150 kg/m s R22, d=13 mm, T=5 C, G=150 kg/m s
25 2 25 2
q=15-20 kW/m , L=2.035 m q=30-40 kW/m , L=2.035 m
pressure drop [mbar]

pressure drop [mbar]

20 20

total total
momentum momentum
15 15
frictional frictional

10 10

5 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

140 140

o 2 o 2
R410A, d=13 mm, T=5 C, G=500 kg/m s R410A, d=13 mm, T=5 C, G=500 kg/m s
120 120
2 2
q=30-40 kW/m , L=2.035 m q=55-60 kW/m , L=2.035 m
100 100
pressure drop [mbar]

pressure drop [mbar]

80 total 80 total
momentum momentum
frictional frictional
60 60

40 40

20 20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.1: Total, frictional and momentum pressure drops vs. vapor quality at different experi-
mental conditions (d = 13.8 mm).
88 CHAPTER 6. EXPERIMENTAL RESULTS

100 300

o 2 o 2
90 R22, d=8 mm, T=5 C, G=200 kg/m s R22, d=8 mm, T=5 C, G=350 kg/m s
2 250 2
80 q=06-09 kW/m , L=2.035 m q=06-09 kW/m , L=2.035 m
70
pressure drop [mbar]

pressure drop [mbar]

200
60
total total
momentum momentum
50 150
frictional frictional

40
100
30

20
50
10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

300 350

o 2 o 2
R22, d=8 mm, T=5 C, G=350 kg/m s R410A, d=8 mm, T=5 C, G=500 kg/m s
300
250 2 2
q=15-20 kW/m , L=2.035 m q=06-09 kW/m , L=2.035 m
250
pressure drop [mbar]

pressure drop [mbar]

200

total 200 total

momentum momentum
150
frictional frictional
150

100
100

50
50

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

350 350

o 2 o 2
R410A, d=8 mm, T=5 C, G=500 kg/m s R410A, d=8 mm, T=5 C, G=500 kg/m s
300 300
2 2
q=15-20 kW/m , L=2.035 m q=30-40 kW/m , L=2.035 m
250 250
pressure drop [mbar]

pressure drop [mbar]

200 total 200 total

momentum momentum
frictional frictional
150 150

100 100

50 50

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.2: Total, frictional and momentum pressure drops vs. vapor quality at different experi-
mental conditions (d = 8 mm).
6.2. TWO-PHASE PRESSURE DROP MEASUREMENTS 89

30 30

o 2 2 o 2 2
R410A, d=13 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m R410A, d=13 mm, T=5 C, G=300 kg/m s, q=15-20 kW/m
25 25
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

.. . .. . . ..
20
. . .. 20
. .. ..
..
Diabatic Diabatic
Adiabatic

. . ....
Adiabatic

...
15

.
15
.
. .
. ..
..
10 10

. .
5

.. .
5
. ..
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

30 30

o 2 2 o 2 2
R410A, d=13 mm, T=5 C, G=300 kg/m s, q=30-40 kW/m R410A, d=13 mm, T=5 C, G=300 kg/m s, q=55-60 kW/m
25 25
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

. . . .. . . .. .
20
. Diabatic

.. ..
20
. Diabatic
. .
. ... .. .. .... .
Adiabatic Adiabatic

..
15 15

..
10
. 10

5
.. 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.3: Adiabatic vs. diabatic test section results for R410A.

6.2 Two-phase pressure drop measurements

First, some comparisons of two-phase pressure drop measurements are presented here in order
to determine the influence of different test variables. Then, the entire database is depicted in a
graphical form.

6.2.1 Some comparisons for different experimental parameters

Diameter

Fig. (6.4) depicts a comparison of the two-phase frictional pressure drops for the internal diameters
tested at a particular set of test variables. The smaller diameter tube induces a larger two-phase
pressure gradient.
90 CHAPTER 6. EXPERIMENTAL RESULTS

60 60

o 2 2 o 2 2
R22, T=5 C, G=200 kg/m s, q=06-09 kW/m R410A, T=5 C, G=300 kg/m s, q=06-09 kW/m
50 50
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

40 40

d=8 mm d=8 mm
30 30
d= 13.8 mm d= 13.8 mm

20 20

10 10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.4: Frictional pressure gradients vs. vapor quality for R22 and R410A.

Fluid

Fig. (6.5) shows that the frictional two-phase pressure gradients are largest for R134a and small-
est for R410A with R22 in the middle. The data for each fluid shows the characteristic rise in
two-phase frictional pressure gradient with rising vapor quality, a peak at high vapor quality, and
the subsequent falloff as the vapor quality approaches 100%. The trends in the data are seen to be
very clear with little scatter in the data because of the accuracy of the transducers, the accuracy
of the energy balances and the fine control on the steady-state operating conditions.

20 60

o 2 2 o 2 2
18 d=13 mm, T=5 C, G=150 kg/m s, q=15-20 kW/m d=13 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m
50
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

16

14
40
12
R134a R134a
R22 R22
10 30
R410A R410A

8
20
6

4
10
2

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.5: Comparison of the frictional pressure gradients for R134a,R22 and R410A at two set
of experimental conditions.

Mass velocity

Fig. (6.6) depicts the experimental results for R22 and R410A at several mass velocities for a
particular set of experimental conditions. As expected, the frictional pressure gradient increases
with mass velocity.
6.2. TWO-PHASE PRESSURE DROP MEASUREMENTS 91

140 140

o 2 o 2
R22, d=13 mm, T=5 C, q=06-09 kW/m R410, d=8 mm, T=5 C, q=06-09 kW/m
120 120
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

100 100

2 2
G=600 kg/m s G=500 kg/m s
2 2
80 G=500 kg/m s 80 G=350 kg/m s
2 2
G=300 kg/m s G=300 kg/m s
2 2
G=150 kg/m s G=200 kg/m s
60 60

40 40

20 20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.6: Frictional pressure gradients vs. vapor quality for different mass velocities.

Heat flux

Fig. (6.7) illustrates the frictional pressure drop data obtained at different heat fluxes, from rather
low to quite high values. The effect of the evaporation heat flux is seen to be of only minor
importance at vapor qualities before the peak. At high vapor qualities, the peak is shifted to lower
vapor qualities by the increase in the heat flux and this is the more important influence of the
heat flux observed. The location of the peaks coincides with the onset of dryout at the top of the
tube and the top of the tube is dry after the peak. For the highest heat fluxes, based on visual
observations in the sight glass at the end of the diabatic test section, the flow was observed to
convert to mist flow, and these are the data at high vapor quality that depict an increasing trend
with vapor quality after the falloff, e.g. such as those in the left graph at q = 55 − 60 kW/m2 .

30 40

2 o 2 o
R410, d=13 mm, G= 300 kg/m s, T=5 C 35 R22, d=8 mm, G= 200 kg/m s, T=5 C
25
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

30

20
q=06-09 kW/m
2 25
2 2
q=15-20 kW/m q=06-09 kW/m
2 2
q=30-40 kW/m q=15-20 kW/m
15 2 20 2
q=55-60 kW/m q=30-40 kW/m

15
10

10

5
5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.7: Frictional pressure gradients vs. vapor quality at different heat fluxes.
92 CHAPTER 6. EXPERIMENTAL RESULTS

6.2.2 Results for R134a

d=13.8 mm

All the results for R134a are shown in Fig. (6.8). Only tests with the 13.8 mm test section were
made for this refrigerant.

18 18

o 2 2 o 2 2
16 R134a, d=13 mm, T=5 C, G=150 kg/m s, q=06-09 kW/m 16 R134a, d=13 mm, T=5 C, G=150 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

14 14

12 12
Diabatic Diabatic
10 10

8 8

6 6

4 4

2 2

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

18 60

o 2 2 o 2 2
16 R134a, d=13 mm, T=5 C, G=150 kg/m s, q=30-40 kW/m R134a, d=13 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m
50
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

14

12 40
Diabatic Diabatic
10
30
8

6 20

4
10
2

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.8: Frictional pressure gradients vs. vapor quality for R134a at different experimental
conditions (d = 13.8 mm).
6.2. TWO-PHASE PRESSURE DROP MEASUREMENTS 93

6.2.3 Result for R22

d=13.8 mm

All the results obtained for R22 in the 13.8 mm test section are shown in Figs. (6.9)-(6.12).

3.0 3.0

o 2 2 o 2 2
R22, d=13 mm, T=5 C, G=70 kg/m s, q=06-09 kW/m R22, d=13 mm, T=5 C, G=70 kg/m s, q=15-20 kW/m
2.5 2.5
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

2.0 2.0
Diabatic Diabatic

1.5 1.5

1.0 1.0

0.5 0.5

0.0 0.0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

10 10

o 2 2 o 2 2
9 R22, d=13 mm, T=5 C, G=150 kg/m s, q=06-09 kW/m 9 R22, d=13 mm, T=5 C, G=150 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

8 8

.
. .. . .
7 7

. Diabatic
Adiabatic

..
6 6 Diabatic

4
.. 5

3 3

2 2

1 1

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.9: Frictional pressure gradients vs. vapor quality for R22 at different experimental condi-
tions (d = 13.8 mm) – (I).
94 CHAPTER 6. EXPERIMENTAL RESULTS

10 30

o 2 2 o 2 2
9 R22, d=13 mm, T=5 C, G=150 kg/m s, q=30-40 kW/m R22, d=13 mm, T=5 C, G=200 kg/m s, q=06-09 kW/m
25
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

8

7
20
6 Diabatic Diabatic

5 15

4
10
3

2
5
1

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

60 60

o 2 2 o 2 2
R22, d=13 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m R22, d=13 mm, T=5 C, G=300 kg/m s, q=15-20 kW/m
50 50
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

40
. Diabatic
Adiabatic
40
Diabatic

30 30

..
..
20 20

10

. . . 10

0
. 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

60 60

o 2 2 o 2 2
R22, d=13 mm, T=5 C, G=300 kg/m s, q=30-40 kW/m R22, d=13 mm, T=5 C, G=300 kg/m s, q=55-60 kW/m
50 50
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

40
. Diabatic
Adiabatic
40
Diabatic

30 ... 30

20 .. . 20

10 10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.10: Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 13.8 mm) – (II).
6.2. TWO-PHASE PRESSURE DROP MEASUREMENTS 95

60 80

o 2 2 o 2 2
R22, d=13 mm, T=5 C, G=350 kg/m s, q=30-40 kW/m 70 R22, d=13 mm, T=5 C, G=400 kg/m s, q=15-20 kW/m
50
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

60

40
50
Diabatic Diabatic

30 40

30
20

20

10
10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

80 80

o 2 2 o 2 2
70 R22, d=13 mm, T=5 C, G=400 kg/m s, q=30-40 kW/m 70 R22, d=13 mm, T=5 C, G=400 kg/m s, q=55-60 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

60 60

. ... ..
...
Diabatic
50 50
Adiabatic

. . Diabatic

..
40 40

. ..
.
30 30

20
.. 20

10 10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

100 100

o 2 2 o 2 2
90 R22, d=13 mm, T=5 C, G=500 kg/m s, q=06-09 kW/m 90 R22, d=13 mm, T=5 C, G=500 kg/m s, q=15-20 kW/m

.
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

.. . ..
80 80

70 70

60 Diabatic 60 . Diabatic
Adiabatic

50 50

40 40

30 30

20 20

10 10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.11: Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 13.8 mm) – (III).
96 CHAPTER 6. EXPERIMENTAL RESULTS

100 100

o 2 2 o 2 2
90 R22, d=13 mm, T=5 C, G=500 kg/m s, q=30-40 kW/m 90 R22, d=13 mm, T=5 C, G=500 kg/m s, q=55-60 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

80 80

70

. . . 70

. .
..
. . ..
. .
Diabatic Diabatic

. .
60 Adiabatic 60 Adiabatic

50 50
.
40 40
..
30

20
. 30

20

10 10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

140 140

o 2 2 o 2 2
120
R22, d=13 mm, T=5 C, G=500 kg/m s, q=06-09 kW/m 120
R22, d=13 mm, T=5 C, G=600 kg/m s, q=30-40 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

100 100

Diabatic Diabatic
80 80

60 60

40 40

20 20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

140 160

o 2 2 o 2 2
120
R22, d=13 mm, T=5 C, G=600 kg/m s, q=55-60 kW/m 140 R22, d=13 mm, T=5 C, G=700 kg/m s, q=55-60 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

.
... .
120

.
. .. ... .
100

. Diabatic
Adiabatic
100 . Diabatic
Adiabatic
.
..
80

..
80
60

..
60

40
40

20 20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.12: Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 13.8 mm) – (IV).
6.2. TWO-PHASE PRESSURE DROP MEASUREMENTS 97

d=8 mm

All the results obtained for R22 in the 8 mm test section are shown in Figs. (6.13)-(6.14).

10 10

o 2 2 o 2 2
9 R22, d=8 mm, T=5 C, G=100 kg/m s, q=06-09 kW/m 9 R22, d=8 mm, T=5 C, G=100 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

8 8

7 7
. .
6 . Diabatic
Adiabatic 6 . Diabatic
Adiabatic
..
. .
5

...
5
..
.
4 4

..
3 3

1
. 2

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

30 50

o 2 2 o 2 2
R22, d=8 mm, T=5 C, G=150 kg/m s, q=06-09 kW/m 45 R22, d=8 mm, T=5 C, G=200 kg/m s, q=06-09 kW/m
25
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

40

. .. .. .
35
20
. Diabatic
. Diabatic

.. .
Adiabatic 30 Adiabatic

..
..
. .
15 25

. .
..
20
10

. .. 15

.
. .. 10
. ..
. .....
5

. 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.13: Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 8 mm) – (I).
98 CHAPTER 6. EXPERIMENTAL RESULTS

50 50

o 2 2 o 2 2
45 R22, d=8 mm, T=5 C, G=200 kg/m s, q=15-20 kW/m 45 R22, d=8 mm, T=5 C, G=200 kg/m s, q=30-40 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

40 40

35

. . . . . .. 35

.
. .
Diabatic Diabatic
30 Adiabatic 30 Adiabatic

. . . . . ..
..
25 25

20 20

15 15

10 10

5 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

140 140

o 2 2 o 2 2
120
R22, d=8 mm, T=5 C, G=350 kg/m s, q=06-09 kW/m 120
R22, d=8 mm, T=5 C, G=350 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

100
.
. . . . ...
100
. .. ..
. Diabatic

. . . Diabatic
.. . .
. .
Adiabatic Adiabatic

.. ..
80 80

.. . .
60
. 60

. .
. .
40
. 40

20 20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

140

o 2 2
120
R22, d=8 mm, T=5 C, G=350 kg/m s, q=30-40 kW/m
frictional pressure gradient [mbar/m]

. .. . ..
100

. Diabatic
. .
... ..
Adiabatic
80

60 ..
40

20

0
0 10 20 30 40 50 60 70 80 90 100

vapor quality [%]

Figure 6.14: Frictional pressure gradients vs. vapor quality for R22 at different experimental
conditions (d = 8 mm) – (II).
6.2. TWO-PHASE PRESSURE DROP MEASUREMENTS 99

6.2.4 Result for R410A

d=13.8 mm

All the results obtained for R410A in the 13.8 mm test section are shown in Figs. (6.15)-(6.18).

2.0 2.0

o 2 2 o 2 2
1.8 R410A, d=13 mm, T=5 C, G=70 kg/m s, q=06-09 kW/m 1.8 R410A, d=13 mm, T=5 C, G=70 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

1.6 1.6

1.4 1.4

1.2 . Diabatic
Adiabatic 1.2 . Diabatic
Adiabatic

1.0 1.0

0.8

. . .. .. ... . . 0.8
.. . ... ... .
..
. . .. . .
0.6 0.6

0.4 0.4

0.2 0.2

0.0 0.0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

8 8

o 2 2 o 2 2
7 R410A, d=13 mm, T=5 C, G=150 kg/m s, q=06-09 kW/m 7 R410A, d=13 mm, T=5 C, G=150 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

6 6

5 . Diabatic

.. . 5 . Diabatic

... .
. . . ....
Adiabatic Adiabatic

. . .
. ..
4 4

. . .
3
. 3

..
. . .. .
. .. .
2 2

1
. . .. 1

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.15: Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 13.8 mm) – (I).
100 CHAPTER 6. EXPERIMENTAL RESULTS

8 14

o 2 2 o 2 2
7 R410A, d=13 mm, T=5 C, G=150 kg/m s, q=30-40 kW/m R410A, d=13 mm, T=5 C, G=200 kg/m s, q=06-09 kW/m
12
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

6

.. . ..
10

. Diabatic
. Diabatic
.
. . .... ... ..
5

.
Adiabatic Adiabatic
8

. ...
. .. .
4

. 6
.
..
3

.
.. ..
4
2

. . .
..... .... .
1 2

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

30 30

o 2 2 o 2 2
R410A, d=13 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m R410A, d=13 mm, T=5 C, G=300 kg/m s, q=15-20 kW/m
25 25
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

.. . .. . . ..
20
. . .. 20
. .. ..
..
Diabatic Diabatic
Adiabatic

. . ....
Adiabatic

...
15

.
15
.
10
. .. 10
. ..
.. .
5

.. .
5
. ..
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

30 30

o 2 2 o 2 2
R410A, d=13 mm, T=5 C, G=300 kg/m s, q=30-40 kW/m R410A, d=13 mm, T=5 C, G=300 kg/m s, q=55-60 kW/m
25 25
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

. . . .. . . .. .
20
. Diabatic

. . ..
20
. Diabatic
. .
. ... .. .. .... .
Adiabatic Adiabatic

..
15 15

..
10
. 10

5
.. 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.16: Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 13.8 mm) – (II).
6.2. TWO-PHASE PRESSURE DROP MEASUREMENTS 101

50 50

o 2 2 o 2 2
45 R410A, d=13 mm, T=5 C, G=400 kg/m s, q=30-40 kW/m 45 R410A, d=13 mm, T=5 C, G=400 kg/m s, q=55-60 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

40 40

.. . . . ......
..
35 35

. Diabatic
. . Diabatic

. ..
30 Adiabatic 30 Adiabatic

.. . .
25 25

..
20 20

15 15

10 10

5 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

80 80

o 2 2 o 2 2
70 R410A, d=13 mm, T=5 C, G=500 kg/m s, q=06-09 kW/m 70 R410A, d=13 mm, T=5 C, G=500 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

. . ..
60 60

. . . . .. . ...
..
Diabatic Diabatic
50 50

.
Adiabatic Adiabatic

. .
.. .
40 40

30
. 30
. ..
... ..
.
..
20 20

.. .
10
.. . 10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

80 80

o 2 2 o 2 2
70 R410A, d=13 mm, T=5 C, G=500 kg/m s, q=30-40 kW/m 70 R410A, d=13 mm, T=5 C, G=500 kg/m s, q=55-60 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

60 60

50 . Diabatic
Adiabatic
. . . .... . 50 . Diabatic
Adiabatic

. . . .... . .
40
. . 40
.. .
. .. . .
.. .
30 30

.. .
...
20 20

10
. 10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.17: Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 13.8 mm) – (III).
102 CHAPTER 6. EXPERIMENTAL RESULTS

100 100

o 2 2 o 2 2
90 R410A, d=13 mm, T=5 C, G=600 kg/m s, q=30-40 kW/m 90 R410A, d=13 mm, T=5 C, G=600 kg/m s, q=55-60 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

80 80

. ...
.........
70 70

. Diabatic
. . Diabatic
60 Adiabatic

..
60 Adiabatic
.
50

. . 50

. . ..
40
. 40
.
30

. . . 30

. .
.
20

10
.. 20

10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

120

o 2 2
R410A, d=13 mm, T=5 C, G=700 kg/m s, q=55-60 kW/m
100
frictional pressure gradient [mbar/m]

..
. . ....
80

.
Diabatic
Adiabatic

. .
..
60

. .
. ..
40

20

0
0 10 20 30 40 50 60 70 80 90 100

vapor quality [%]

Figure 6.18: Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 13.8 mm) – (IV).
6.2. TWO-PHASE PRESSURE DROP MEASUREMENTS 103

d=8 mm

All the results obtained for R410A in the 8 mm test section are shown in Figs. (6.19)-(6.22).

6 8

o 2 2 o 2 2
R410A, d=8 mm, T=5 C, G=100 kg/m s, q=06-09 kW/m 7 R410A, d=8 mm, T=5 C, G=100 kg/m s, q=15-20 kW/m
5
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

6

4
. Diabatic
5 . Diabatic

.
Adiabatic Adiabatic

. . . ..
.
3 4

. . .
.. .
3

..
2

1 .. .. 2

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

10 30

o 2 2 o 2 2
9 R410A, d=8 mm, T=5 C, G=150 kg/m s, q=15-20 kW/m R410A, d=8 mm, T=5 C, G=200 kg/m s, q=06-09 kW/m

.
25
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

. .. . . .
8

6 . Diabatic
Adiabatic
...
20
. Diabatic
Adiabatic
.. .
5 . 15
. .
4

. ..
.
10

.
3

. . . .
1
5

. . . .
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.19: Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 8 mm) – (I).
104 CHAPTER 6. EXPERIMENTAL RESULTS

30 30

o 2 2 o 2 2
R410A, d=8 mm, T=5 C, G=200 kg/m s, q=15-20 kW/m R410A, d=8 mm, T=5 C, G=200 kg/m s, q=30-40 kW/m

..
25 25
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

.. .. . .. .. . . . .
20
. Diabatic
20
. Diabatic
.
.. .
Adiabatic Adiabatic

15
. . 15 .
10 . . 10

5 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

50 50

o 2 2 o 2 2
45 R410A, d=8 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m 45 R410A, d=8 mm, T=5 C, G=300 kg/m s, q=15-20 kW/m

... ..
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

. ..
40 40

. . .. .
.. .
35 35

. Diabatic
. . . Diabatic
.
..
30 Adiabatic 30 Adiabatic

.. . .....
.. .
25 25

..
. .
20 20

15 15

.
..
10 10

5 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

50 90

o 2 2 o 2 2
45 R410A, d=8 mm, T=5 C, G=300 kg/m s, q=30-40 kW/m 80 R410A, d=8 mm, T=5 C, G=350 kg/m s, q=06-09 kW/m

. . ... .. . . ..
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

40

.
70

35

. .. 60
. . .
.
Diabatic Diabatic
30 Adiabatic
.. . Adiabatic

.
....
50
25
40
. .
20

. .
.
30
15

.
..
20
10

5 10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.20: Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 8 mm) – (II).
6.2. TWO-PHASE PRESSURE DROP MEASUREMENTS 105

90 90

o 2 2 o 2 2
80 R410A, d=8 mm, T=5 C, G=350 kg/m s, q=15-20 kW/m 80 R410A, d=8 mm, T=5 C, G=350 kg/m s, q=30-40 kW/m

. ..
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

.. . . .. . .
70 70

. .
. . .. . ..
60 60

.
Diabatic Diabatic

. ..
Adiabatic Adiabatic
50

. . .. 50

40
. 40
.. .
. .
..
30 30

20

10
. 20

10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

90 90

o 2 2 o 2 2
80 R410A, d=8 mm, T=5 C, G=400 kg/m s, q=30-40 kW/m 80 R410A, d=8 mm, T=5 C, G=400 kg/m s, q=55-60 kW/m

..
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

70
. .. 70

60
. Diabatic
.. 60
. Diabatic .......
.
.......
Adiabatic Adiabatic

.... .
50 50

40 40

30 30

20 20

10 10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

140 140

o 2 2 o 2 2
R410A, d=8 mm, T=5 C, G=500 kg/m s, q=06-09 kW/m R410A, d=8 mm, T=5 C, G=500 kg/m s, q=15-20 kW/m
120 120
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

100

. .. ..
. ..
.
100

. . . . . . . ...
..
Diabatic Diabatic

. .. ..
Adiabatic Adiabatic

..
80 80

. . . .
60
. 60

.
40
. . 40
. .
.
20 .. 20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 6.21: Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 8 mm) – (III).
106 CHAPTER 6. EXPERIMENTAL RESULTS

140 180

o 2 2 o 2 2
R410A, d=8 mm, T=5 C, G=500 kg/m s, q=30-40 kW/m 160 R410A, d=8 mm, T=5 C, G=600 kg/m s, q=30-40 kW/m

. . . .....
120
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

. . . .........
140
100

. . .
.. .
120
Diabatic Diabatic

.
. ..
Adiabatic Adiabatic
80 100

60 80

60
40
40

20
20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

180 200

o 2 2 o 2 2
160 R410A, d=8 mm, T=5 C, G=600 kg/m s, q=55-60 kW/m 180 R410A, d=8 mm, T=5 C, G=700 kg/m s, q=30-40 kW/m

... ... .
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

160

.. .. . . . .....
140

140
120
. . .
. . ...
Diabatic Diabatic

.. .
Adiabatic 120 Adiabatic

.
100
100
80
80
60
60

40
40

20 20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

200

o 2 2
180 R410A, d=8 mm, T=5 C, G=700 kg/m s, q=55-60 kW/m
frictional pressure gradient [mbar/m]

160

. ... .
.. .. ..
140

. Diabatic

. . ...
120 Adiabatic

100

80

60

40

20

0
0 10 20 30 40 50 60 70 80 90 100

vapor quality [%]

Figure 6.22: Frictional pressure gradients vs. vapor quality for R410A at different experimental
conditions (d = 8 mm) – (IV).
6.3. CONCLUSIONS 107

6.3 Conclusions

The procedure to obtain the two-phase frictional component from the measured values has been
detailed. As the test section consisted of two separate zones, the reduction of the experimental
data was different depending on the zone. While for the adiabatic part of the test section there
is no momentum pressure drop created by acceleration of the flow and the frictional component
corresponded to the measured value, for the diabatic zone this term needs to be evaluated and
subtracted from the measured values. In order to evaluate the significance of this component a
set of graphs, for shake of space even if the reduction of experimental data was applied to the
entire database, showed the different components of the two-phase pressure drops for a selected
set of experimental conditions. By this, it was shown that the momentum pressure drop became
larger as the evaporation process was more intense, i.e. high heat fluxes. Furthermore, as a
validation of the data reduction procedure a set of graphs showed comparisons, for a representative
set of experimental conditions, of the two-phase frictional pressure gradients for the adiabatic test
section versus those for the diabatic test section. The agreement between was shown to be quite
remarkable attesting to the accuracy and reliability of the measurements as well as the aptness of
the data reduction protocol.
Next, and before the entire presentation of the entire database, a parametric study to evaluated
the influence of the test variables was undertaken. The test variables were for the present study:
fluid, diameter, mass velocities and heat fluxes as well as vapor quality representing the evaporation
process. A set of graphs, for a selected representative values of the test variables, showed that while
the fluid, diameter and mass velocity had a strong effect over the entire range of vapor quality, the
heat flux influenced the measured values only for a particular range of vapor qualities. This latter
was the reason to introduce the new two-zone test section and by this way the appropiateness of
that choice has been justified.
Finally, the entire set of experimental values is presented. The results are segregated first by fluid
and secondly by diameter in order to be presented is a systematic and easily readable way. More
than 2500 experimental values of the two-phase pressure drops were obtained, and just to claim the
significance of the presented results it is important to signal that around 25 minutes were necessary
to reach the steady-state conditions.
108 CHAPTER 6. EXPERIMENTAL RESULTS
Chapter 7

New Two-Phase Frictional Pressure

Drop Model Based on Flow Pattern
Map

In this chapter, first the experimental results obtained during the experimental campaign are com-
pared to the three leading existing two-phase frictional pressure drop correlations. Second, and as
a consequence of their incapability to reliably predict the present experimental database, a new
two-phase frictional pressure drop model is proposed and described. The new model is developed
following a phenomenological approach, that is physically based on a simplified interfacial flow
structure using the recent Wojtan-Ursenbacher-Thome [155] flow pattern map. Finally, all exper-
imental data are compared with the new prediction methods on a statistical basis. The present
work completes the fourth basic steps in LTCM’s flow pattern based work on two-phase and heat
transfer inside horizontal tubes: (i) generalized flow pattern map, (ii) flow boiling heat transfer
model, (iii) convective condensation model and (iv) two-phase frictional pressure drop model.

7.1 Comparison to existing methods

Tribbe and Müller-Steinhagen [134] compared some of the leading two-phase frictional pressure
drop correlations to a large database, including the following fluid combinations: air-oil, air-water,
water-steam and several refrigerants. They found that statistically the method of Müller-Steinhagen
and Heck [97] gave the best and most reliable results. In another recent comparison, Ould Didi,
Kattan and Thome [105] compared leading methods to experimental pressures drops obtained for
five different refrigerants over a wide range of experimental conditions. Overall, they found the
Grönnerud [53] and the Müller-Steinhagen and Heck [97] methods to be equally best while the
Friedel [46] method was the third best in a comparison to seven leading methods.
As a first step in the present analysis, the experimental results were compared to the three previously
”best” methods. The popular Lockhart-Martinelli correlation is not shown here in this comparison
in spite of its extensive use and continued historical references because it was not well ranked in
previous studies (nor does it compare well to the present data). A detailed description of these
methods can be found in Chapter 4.
Figs. (7.1) and (7.2) show a representative comparison of the three prediction methods (Friedel,

109
110 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

Grönnerud and Müller-Steinhagen and Heck) for different experimental conditions. Starting first
with the Grönnerud method, it works quite well at low mass velocities from low to medium vapor
qualities and it also captures the maximum in the data at high vapor quality, although not in
magnitude but reasonably well the location of the peak. Instead, at the highest mass velocities the
Grönnerud method very significantly overpredicts all the results except at low vapor qualities. The
Müller-Steinhagen and Heck correlation does not follow the trends in the data well, underpredicting
most of them and does not capture the peak in the data at all. The Friedel method is seen to
significantly overpredict at the lowest mass velocity while underpredicting most of the data at the
highest and does not follow the variation in the data with vapor quality, again missing the peak.
For sake of simplicity, comparisons for refrigerant R314a are not shown here but the situation is
essentially the same.
Figs. (7.3), (7.4) and (7.5) show a comparison of the complete database to the three prediction
methods (Friedel, Grönnerud and Müller-Steinhagen and Heck). Certainly of the three, the Müller-
Steinhagen and Heck comes out the best overall. However, it is only able to predict one-half of the
database within ±20%.
Hence, it is clear that none of these three methods are reliable for optimizing the thermal-hydraulics
of a direct-expansion evaporator, even though statistically they are not that bad when considering
the normal situation for the prediction of two-phase frictional pressure drops.
In general, the existing two-phase frictional pressure models available in the literature for horizontal
flows have some or all of the following deficiencies:

• They do not account for flow pattern effects on the process;

• They do not account explicitly for the influence of interfacial waves;

• They do not account for the upper dry perimeter of stratified types of flows;

• They do not use the actual velocities of the vapor and liquid by introduction of the local void
fraction into the method;

• They use tubular flow expressions to represent annular film flows;

• They do not capture the peak in the pressure gradient at high vapor qualities (or not its
location or magnitude) nor give a good representation of the pressure gradient trend versus
vapor quality.

• They do not go to acceptable limits at x = 0 and x = 1.

Hence, this state of affairs is the justification to develop a new model. Secondly, the present flow
pattern based pressure drop model fits into the unified method of Thome and coworkers to model
evaporation and condensation inside horizontal tubes.
7.1. COMPARISON TO EXISTING METHODS 111

12 50

o 2 2 o 2 2
R22, d=13 mm, T=5 C, G=150 kg/m s, q=06-09 kW/m 45 R22, d=13 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m
10
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

40
Experimental Experimental
Friedel 35 Friedel
8
Muller-Steinhagen and Heck Muller-Steinhagen and Heck
Gronnerud 30 Gronnerud

6 25

20
4
15

10
2
5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

160 160

o 2 2 o 2 2
140 R22, d=13 mm, T=5 C, G=500 kg/m s, q=15-20 kW/m 140 R22, d=13 mm, T=5 C, G=500 kg/m s, q=30-40 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

120 Experimental 120 Experimental

Friedel Friedel
Muller-Steinhagen and Heck Muller-Steinhagen and Heck
100 100
Gronnerud Gronnerud

80 80

60 60

40 40

20 20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

50 160

o 2 2 o 2 2
45 R22, d=8 mm, T=5 C, G=200 kg/m s, q=06-09 kW/m 140 R22, d=8 mm, T=5 C, G=350 kg/m s, q=06-09 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

40
Experimental 120 Experimental
35 Friedel Friedel
Muller-Steinhagen and Heck Muller-Steinhagen and Heck
100
30 Gronnerud Gronnerud

25 80

20
60
15
40
10

20
5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 7.1: Frictional pressure gradients vs. three prediction methods for R22 at different experi-
mental conditions.
112 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

8 40

o 2 2 o 2 2
7 R410A, d=13 mm, T=5 C, G=150 kg/m s, q=06-09 kW/m 35 R410A, d=13 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

6 Experimental 30 Experimental
Friedel Friedel
Muller-Steinhagen and Heck Muller-Steinhagen and Heck
5 25
Gronnerud Gronnerud

4 20

3 15

2 10

1 5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

120 120

o 2 2 o 2 2
R410A, d=13 mm, T=5 C, G=500 kg/m s, q=06-09 kW/m R410A, d=13 mm, T=5 C, G=500 kg/m s, q=30-40 kW/m
100 100
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

Experimental Experimental
Friedel Friedel
80 80
Muller-Steinhagen and Heck Muller-Steinhagen and Heck
Gronnerud Gronnerud

60 60

40 40

20 20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

40 120

o 2 2 o 2 2
35 R410A, d=8 mm, T=5 C, G=200 kg/m s, q=06-09 kW/m R410A, d=8 mm, T=5 C, G=350 kg/m s, q=06-09 kW/m
100
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

30 Experimental Experimental
Friedel Friedel
80
Muller-Steinhagen and Heck Muller-Steinhagen and Heck
25
Gronnerud Gronnerud

20 60

15
40

10

20
5

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 7.2: Frictional pressure gradients vs. three prediction methods for R410A at different
experimental conditions.
7.1. COMPARISON TO EXISTING METHODS 113

200

+ 30 %
180

... ..
a)
. .. . .. . .........
160
Predicted frictional pressure gradient [mbar/m]

140
. . .
Friedel
120 ...... .... ......
.
. . . .. - 30 %
100
.. .. . . . ....... .. ..
. . ...... . .
... . ..
.. ... . .........................
80

. ... .
60 . ......... ................................................
. ..... ....... ....
......................................................................
....................
40
..
................................... Experimental versus predicted values
20
..
...
....
........ .
.....
In range 67.33%
0
0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m]

200

+ 20 %
180

... ..
b)
. .. . .. . .........
160
Predicted frictional pressure gradient [mbar/m]

140
. . .
Friedel
120 ...... .... ......
.
. . . .. - 20 %
100
.... . . . ....... .. ..
. . .... . . .
... . ..
.. ... . .........................
80

. . ... .
60 . ........ ..............................................
. ..... ....... ....
.......................................................................
........................
40

.......
............... Experimental versus predicted values
... ..
..
.
..
. .
. ..
.......... .
20

......
In range 51.81%
0
0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m]

Figure 7.3: Comparisons of experimental to predicted values using Friedel correlation for the entire
database: a) 67.33% are predicted within ±30%, b) 51.81% are predicted within ±20%.
114 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

350

.. .. . . ....
+ 30 %
300
a)
.
Predicted frictional pressure gradient [mbar/m]

Gronnerud
..
.... . ..
250

.. .
200 .. .... . . . . .
. . ...
.. . . ............. .. . - 30 %
150 . . .. ... ...
. . ....
................................ ..
.. ............ . .
... ......................................... ..
100

........... ...
....................................... .
..... . Experimental versus predicted values
.............................
50

...........
In range 46.15%
0
0 50 100 150 200 250 300 350

Experimental frictional pressure gradient [mbar/m]

350

... . . ....
+ 20 %
300
b) .
.
Predicted frictional pressure gradient [mbar/m]

Gronnerud
..
.... . ..
250

.. .
. .... . . . . .
.
200
. . ...
.. . . ............. .. . - 20 %
150 . . .. ... ...
. . ....
................................. ..
.. ............ . .
... ........................................ ..
100

.......... ...
........................................... .
....... . Experimental versus predicted values
............................
50

. ..
.........
In range 40.45%
0
0 50 100 150 200 250 300 350

Experimental frictional pressure gradient [mbar/m]

Figure 7.4: Comparisons of experimental to predicted values using Grönnerud correlation for the
entire database: a) 46.15% are predicted within ±30%, b) 40.45% are predicted within ±20%.
7.1. COMPARISON TO EXISTING METHODS 115

200

+ 30 %
180

a) ... ...
160
. .. .. . . .........
Predicted frictional pressure gradient [mbar/m]

140
. . .
Muller-Steinhagen .
120 ........... ......
.. . . ..
100
.... .. . - 30 %
.. . .. ....... .. .
. ... .... .
.. .. .. . ................... .
80

... . . . .. . .. .
60
.. . .... .....................................................
...... ..................................................
40
. .
..................................................... Experimental versus predicted values
. .
......
.
..
..................................
.........
20
. .
......
In range 75.79%
0
0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m]

200

+ 20 %
180

b) ... ...
160
. .. .. . . .........
Predicted frictional pressure gradient [mbar/m]

140
. . .
Muller-Steinhagen .
120 ........... ......
.. . . ..
100
.... .. . - 20 %
.. . .. ....... .. .
. ... .... .
.. .. .. . ... ... .............. .
80

.. . . . .. .. .
60
.. . .... ...................................................
...... ................................................ .
40
. .
................................................... Experimental versus predicted values
. .
......
.
..
..
................................ .
.......
20

........
In range 49.64%
0
0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m]

Figure 7.5: Comparisons of experimental to predicted values using Müller-Steinhagen and Heck
correlation for the entire database: a) 75.79% are predicted within ±30%, b) 49.64% are predicted
within ±20%.
116 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

7.2 New two-phase pressure drop model

7.2.1 Introduction

The new two-phase frictional pressure drop model is developed following a phenomenological ap-
proach, that is physically based on a simplified interfacial two-phase flow structure. The correspond-
ing interfacial flow structure is determined using the recent Wojtan-Ursenbacher-Thome [155] flow
pattern map described in Chapter 3. The main reasons to use this flow pattern map are:

• First, this flow pattern map has been experimentally validated.

• Second, this flow pattern map does not require any iterative calculations and it can be easily
used for flow regime identification.

• Finally, the use of this flow pattern map fits the present flow pattern based pressure drop
model into the unified approach of Thome and coworkers to model evaporation and conden-
sation inside horizontal tubes.

7.2.2 Segregation of experimental data

The entire experimental data are segregated by flow regime as a previous step in the new model
development. For the reasons mentioned above, the recent Wojtan-Ursenbacher-Thome is used for
this task. For sake of simplicity and knowing that for a given set of experimental conditions a
flow pattern map needs to be generated in order to accomplish the segregation process, it has been
chosen to include in this section the flow pattern map for some representative sets of experimental
parameters. The procedure to generate the flow pattern and the complete set of equations can be
found in Chapter 3.
Table (7.1) shows the entire database segregated by flow regime using the recent Wojtan et al.
flow pattern map. It has to be pointed out here that only experimental values where the complete
test section was in the same flow regime are taken into account. This restriction is always verified
for the adiabatic test section as no heat is added to the refrigerant, while for the diabatic test
section the inlet and outlet vapor qualities were checked to verify that the entire test section
was in the same regime. For this reason only 1745 experimental points have been used in new
model development from the more than 2500 experimental values obtained during the experimental
campaign (representing about 50% diabatic data and 50% adiabatic data).
It has to be pointed out here that the fixed value of the mass velocity appearing in the graphs de-
picted in Figs. (7.6)-(7.8) represents the value used for the void fraction calculation using Eq. (3.35).
As the variation of this value practically does not affect the void fraction calculation for mass ve-
locities above 50 kg/m2 s, a design condition for the mass velocity is assumed for simplicity in
generating these graphs and this is the value appearing in the title of the graphs. In actual appli-
cation of the of the flow pattern map and the new frictional pressure drop model, the actual mass
velocity is used in calculations.
7.2. NEW TWO-PHASE PRESSURE DROP MODEL 117

R−22, G=300kg/m2s, Tsat=5 oC, D=13.84mm, q=7.5kW/m2 2

R−22, G=300kg/m s, Tsat=5 C, D=13.84mm, q=17.5kW/m
o 2
700 700

a) b)
M M
600 600

500 500
Mass Velocity [kg/m s]

Mass Velocity [kg/m s]

2

2
400
I A 400
I A
D
300 300
Slug D Slug
200 200

100
Slug+SW SW
100
Slug+SW SW
S S
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Vapor quality [−] Vapor quality [−]
2 o 2 2 o 2
R−22, G=300kg/m s, Tsat=5 C, D=13.84mm, q=37.5kW/m R−22, G=300kg/m s, Tsat=5 C, D=13.84mm, q=57.5kW/m
700 700
c) d)
600
M 600
M

500 500
Mass Velocity [kg/m s]

Mass Velocity [kg/m s]

2

400
I A D 400
I A D

300 300

Slug Slug
200 200

100
Slug+SW SW 100
Slug+SW SW
S S
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Vapor quality [−] Vapor quality [−]

Figure 7.6: Flow pattern maps for R-22, Tsat = 5o C, D = 13.84 mm at G = 300 kg/m2 s and heat
fluxes: a) 7.5 kW/m2 , b) 17.5 kW/m2 , c) 37.5 kW/m2 , d) 57.5 kW/m2 .

R−410A, G=400kg/m2s, Tsat=5 oC, D=8.00mm, q=37.5kW/m2 R−410A, G=400kg/m2s, Tsat=5 oC, D=8.00mm, q=57.5kW/m2
700 700
a) b)
600
M 600
M

500 500
Mass Velocity [kg/m s]

Mass Velocity [kg/m s]

2

400
I A 400 I A
300
D 300
D
Slug
200 200
Slug

100 Slug+SW SW 100 Slug+SW SW

0
S 0
S
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Vapor quality [−] Vapor quality [−]

Figure 7.7: Flow pattern maps for R-410A, Tsat = 5o C, D = 8.00 mm at G = 400 kg/m2 s and
initial heat fluxes: a) 37.5 kW/m2 , b) 57.5 kW/m2 .
118 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

2 o 2 2 o 2
R−410A, G=300kg/m s, Tsat=5 C, D=13.84mm, q=7.5kW/m R−410A, G=300kg/m s, Tsat=5 C, D=13.84mm, q=17.5kW/m
700 700

a) b)
M M
600 600

500 500
Mass Velocity [kg/m s]

Mass Velocity [kg/m2s]

2

400
I A 400
I A
D
D
300 300

Slug Slug
200 200

100
Slug+SW SW 100
Slug+SW SW

S S
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Vapor quality [−] Vapor quality [−]
2 o 2 R−410A, G=300kg/m2s, Tsat=5 oC, D=13.84mm, q=57.5kW/m2
R−410A, G=300kg/m s, Tsat=5 C, D=13.84mm, q=37.5kW/m
700 700

c) d)
M M
600 600

500 500
Mass Velocity [kg/m s]

Mass Velocity [kg/m2s]

2

400
I A 400
I A
D D
300 300
Slug Slug
200 200

100
Slug+SW SW Slug+SW SW
100

S S
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Vapor quality [−] Vapor quality [−]

Figure 7.8: Flow pattern maps for R-410A, Tsat = 5o C, D = 13.84 mm at G = 300 kg/m2 s and
initial heat fluxes: a) 7.5 kW/m2 , b) 17.5 kW/m2 , c) 37.5 kW/m2 , d) 57.5 kW/m2 .

Segregated experimental values by flow regime

Fluid D[mm] Slug+SW SW Slug I A D M S
R134a 13.8 6 3 5 2 13 0 0 0
R22 8 28 20 18 12 75 21 17 0
R22 13.8 18 54 15 37 162 44 47 0
R410A 8 26 19 24 55 167 121 110 0
R410A 13.8 78 115 17 77 166 104 69 0
Total 156 211 79 183 583 290 243 0

Table 7.1: Segregated experimental values by flow regime using Wojtan et al. flow pattern map,
where S= stratified flow, SW= stratified-wavy flow, I= intermittent flow, A= annular flow, M=
mist flow and D represents the transition zone between annular and mist flow.
7.2. NEW TWO-PHASE PRESSURE DROP MODEL 119

7.2.3 Model development

Hence, once the flow regime is known, a simplified interfacial structure can be assumed, and then
a set of equations is proposed to determine two-phase pressure drop for that regime, while at
the same time taking care not to have any jumps in the predicted values when crossing a flow
transition boundary. Below, the new model and its calculation procedure for the different flow
regimes is proposed and described.

Annular region (A)

According to the flow pattern map, about one-third of the experimental data were in the annular
flow regime. Therefore, this regime will be treated first. The model for analysis is a simplified
annular flow consisting of a liquid film and a gas core, as shown Fig. (7.9). Entrainment is ne-
glected and the liquid film is assumed to be smooth and to have a uniform thickness δ around
the circumference. These assumptions are not always valid but they allow a much more simplified
model to be proposed.

Liquid

Vapor
Dcore
D

Figure 7.9: Simplified annular flow configuration.

The model is based on the steady-state gas and liquid phase momentum balance equations. Assum-
ing equal average pressure gradients in the gas and liquid, which is generally valid in the absence
of interfacial liquid level gradients, this leads to the following equation:

∆p τi
=4 (7.1)
L (D − 2δ)

Taking into account that generally D >> δ, Eq. (7.1) can be written as follows:

∆p τi
=4 (7.2)
L D

The expression above states that the frictional pressure gradient is directly related to the interfacial
shear stress τi and the tube diameter D. The interfacial shear stress represents the shear stress
exerted by the gas on the liquid and is given by:

1 1
τi = fi ρG (uG − uL )2 ≈ fi ρG u2G if uL  uG (7.3)
2 2
120 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

where fi is the interfacial friction factor and uG and uL are the true average velocities defined in
Chapter 2. The interfacial friction factor fi remains unknown and is as always the most difficult
parameter to model in two-phase flow. From Eqs. (7.2) and (7.3) and the present two-phase
frictional pressure drop database in the annular flow regime obtained during the experimental
campaign, the following new film flow correlation for fi is proposed:

   −0.4  0.08
δ 1.2 (ρL − ρG )gδ2 µG
(fi )annular = 0.67 [W eL ]−0.034 (7.4)
2R σ µL  ! "
 ! " ! " ! " 4
1 2 3

where the non-dimensional groups include the following effects:

− Group 1: Film thickness effect relative to the internal radius of the tube R.

− Group 2: Related to the formation of interfacial waves with the characteristic wavelength
similar to the film thickness. A more detailed description of this effect can be found in
Appendix B.

− Group 3: The viscosity ratio of the gas and liquid phases.

− Group 4: The Surface tension effect is introduced by means of the liquid Weber W eL number
(Eq. (2.30)).

The empirical constant and exponents were obtained statistically from the annular flow database.
Hence, the frictional two-phase pressure drop can be now calculated with:

 
L ρG u2G
(∆p)annular = 4 (fi )annular (7.5)
D 2

Slug + Intermittent (Slug+I)

These two flow regimes were treated together as was shown in the previous chapter, the respective
experimental frictional two-phase pressure drops follow similar trends. This behaviour is consistent
with Wojtan [153] observations regarding flow boiling heat transfer coefficients. Moreover, due
to the unsteadiness of the flow characterizing these two regimes trying to capture this behaviour
its quite complex. A common characteristic of both regimes is that all the tube perimeter is
continuously wetted, explaining perhaps why their data have similar trends.
For the reasons mentioned above, and to avoid a jump at the transition with annular flow while
also respecting the natural limit at x = 0, a good compromise was found by using the following
expression:

 0.25  0.25



(∆p)slug+intermittent = ∆pL0 1 − + (∆p)annular (7.6)

IA
IA

where ∆pL0 is single-phase frictional liquid pressure drop (evaluated at x = 0),
IA is the void
fraction at the intermittent to annular transition boundary xIA and (∆p)annular is the two-phase
frictional pressure drop evaluated at the actual vapor quality using annular flow equations.
7.2. NEW TWO-PHASE PRESSURE DROP MODEL 121

Eq. (7.6) is an interpolation between the single-phase frictional liquid pressure drop (x = 0) and the
two-phase frictional pressure drop (∆p)annular that would exist if the regime were annular for these
flow conditions. Using void fraction for the interpolation procedure, this expression accomplishes
two important purposes: first, it matches the correct limit at x = 0 (single-phase liquid flow).
Second, it insures a smooth transition at the intermittent to annular transition without introducing
any jump. In doing so, the model is in agreement with the experimental observations and bypasses
these two drawbacks of some existing models.

Stratified-Wavy (SW)

In this regime, the parameter that defines the flow structure and the ratio of the tube perimeter in
contact with the liquid and gas is the dry angle θdry shown in Fig. (7.10).

Vapor PV

AV
qdry

D
qwet
Pi
AL dcrit
Liquid
PL
Figure 7.10: Simplified stratified flow configuration.

It is an experimental fact in stratified-wavy flow that liquid creeps up the sides of the tube to a
varying extend. This effect significantly affects the interfacial perimeters Pi , PL and PG , as well as
the interfacial friction factor fi . In fact, for a fixed vapor quality x, θdry varies from 0 for Gwavy (x)
at the annular flow transition and to its maximun value of θstrat for Gstrat (x) at the fully stratified
flow transition. Several expressions have been proposed to capture this variation: a linear variation
by Kattan et al. [78] and a quadratic interpolation by El Hajal [40]. In the recent work of Wojtan et
al. [156] based on experimental heat transfer data for this region, they have proposed the following
expression:

 0.61
(Gwavy − G)
θdry = θstrat (7.7)
(Gwavy − Gstrat )

where θstrat is determined using the approximate expression, evaluated in terms of void fraction,
proposed by Biberg [16],
  3π 1/3 
θstrat = 2π − 2 π(1 −
) + 2 [1 − 2(1 −
) + (1 −
)1/3
−
1/3 ] (7.8)
1
− 200 (1 −
)
[1 − 2(1 −
)[1 + 4((1 −
)2 +
2 ]

Using Eqs. (7.7) and (7.8) to determine θdry , the following equation is now proposed for the
two-phase friction factor:

∗ ∗
(ftp )stratif ied−wavy = θdry fG + (1 − θdry ) (fi )annular (7.9)
122 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

∗ =θ
where θdry dry /2π, fG is single-phase gas friction factor calculated using the classical definition

0.079 GxD
fG = where ReG = (7.10)
Re0.25
G µG

and (fi )annular is the interfacial friction factor for the annular flow regime described previously
evaluated at the actual vapor quality.
The frictional two-phase pressure drop can now be calculated as follows:

 
L ρG u2G
(∆p)stratif ied−wavy = 4 (ftp )stratif ied−wavy (7.11)
D 2

Slug+Stratified-Wavy (Slug+SW)

In this regime both low amplitude waves (which do not reach the top of the tube) and liquid slugs
washing the top of the tube are observed. With increasing vapor quality in this zone, the frequency
of slugs decreases and the waves become more dominant. This behaviour is highly chaotic and as a
consequence is quite difficult to capture physically within a simplified model. A good compromise
was found, as for the slug and intermittent regimes, by using the following expression:

 0.25  0.25



(∆p)slug+SW = ∆pL0 1 − + (∆p)stratif ied−wavy (7.12)

IA
IA

where ∆pL0 is the single-phase frictional liquid pressure drop (evaluated at x = 0),
IA is the void
fraction at the intermittent to annular transition boundary and (∆p)stratif ied−wavy is the two-phase
frictional pressure drop evaluated at the actual vapor quality using stratified-wavy flow equations.
Eq. (7.12) again insures a smooth transition at the intermittent to annular boundary and matches
the correct limit for x = 0. The interpolation exponent is again 0.25 as in Eq. (7.6) and gives the
best representation of the data.

Mist (M)

This regime is encountered when all the liquid is entrained in the gas core by the high velocity gas.
The vapor phase is the continuous phase and the liquid flows in the form of droplets. In fact,it is not
far from reality to consider the two phases to flow as a single phase possessing mean fluid properties.
Under these conditions one can use the homogeneous flow theory in order to accurately predict
two-phase frictional pressure drop values. Hence, the following expression is used to calculate the
two-phase frictional pressure drop in this regime:

 
L G2
(∆p)mist = 2 fm (7.13)
D ρm

where ρm is the homogeneous density and is determined as

ρm = ρL (1 −
H ) + ρG
H (7.14)
7.2. NEW TWO-PHASE PRESSURE DROP MODEL 123

and the homogeneous void fraction
H is

1

H = (1−x) ρG
(7.15)
1+ x ρL

The friction factor fm is calculated introducing an homogeneous viscosity µm using single-phase

equations as follows,

0.079 GD
fm = where Rem = (7.16)
Re0.25
m µm

and µm is determined using the Cicchitti (1960) expression:

µm = xµG + (1 − x)µL (7.17)

This expression (Eq. 7.13) goes to the ideal limit of all gas flow at x = 1.

Dryout (D)

Analyzing the experimental results, it is obvious that there is not usually a step-wise jump from
annular(A) to mist(M) flow. This transition regime between them is called dryout. The process of
dryout starts at the top of the tube, where the liquid film is thinner, and takes place over a range
of vapor qualities (from the inception of dryout at xdi at the top of the tube to the completion of
dryout at xde at the bottom of the tube) and thus ends when the fully developed mist flow regime
is reached. This process is depicted in Fig. (7.11).

Dryout zone
a)
A xdi B C xde

Flow

A B C
Liquid Vapor
A-A B-B C-C
b)

Figure 7.11: Dryout zone during evaporation in horizontal tube.

The following linear interpolation captures this variation and does not introduce any jump in the
frictional pressure gradient being in agreement with the experimental observations:

x − xdi  
(∆p)dryout = (∆p)tp (xdi ) − (∆p)tp (xdi ) − (∆p)mist (xde ) (7.18)
xde − xdi
where xde is the dryout completion quality and xdi is the dryout inception quality calculated, ac-
cording with the flow pattern map, using the following expressions proposed by Wojtan et al. [153]:
124 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

0.17 F r 0.37 (ρ /ρ )0.25 (q/q 0.70 ]

xdi = 0.58e[0.52−0.235W eG G G L crit )
(7.19)

−3 W e0.38 F r 0.15 (ρ /ρ )−0.09 (q/q 0.27 ]

xde = 0.61e[0.57−5.8·10 G G G L crit )
(7.20)

that are equivalent to Eqs. (3.54) and (3.55).

(∆p)tp (xdi ) is the frictional pressure drop at the inception quality calculated either with Eq. (7.5)
for annular flow or Eq. (7.11) for stratified-wavy flow and (∆p)mist (xde ) is the frictional pressure
drop at the completion quality calculated with Eq. (7.13). At high mass velocities the value of
xde intersects that of xdi ; at these conditions, the dryout zone (D) no longer exits and a jump is
contemplated between annular and mist flow.

Stratified (S)

The stratified flow regime was out of the possibilities of the present test facility as it occurs at
very low mass velocities. Thus, no experimental values were obtained in this regime. In spite of an
experimental verification the following calculation procedure is proposed for this regime:

− if x ≥ xIA then

∗ ∗
(ftp )stratif ied = θstrat fG + (1 − θstrat ) (fi )annular (7.21)
∗
where θstrat = θstrat /2π, fG is single-phase gas friction factor calculated with Eq. (7.10) and
and (fi )annular is the interfacial friction factor for the annular flow regime described previously
evaluated at the actual vapor quality.
The frictional two-phase pressure drop can now be calculated as follows:
 
L ρG u2G
(∆p)stratif ied (x≥xIA ) = 4 (ftp )stratif ied (7.22)
D 2

− if x < xIA then

 0.25  0.25



(∆p)stratif ied (x<xIA ) = ∆pL0 1 − + (∆p)stratif ied (x≥xIA ) (7.23)

IA
IA

where ∆pL0 is single-phase frictional liquid pressure drop (at x = 0),
IA is the void fraction
at the intermittent to annular transition boundary and (∆p)stratif ied (x≥xIA ) is the two-phase
frictional pressure drop evaluated at the actual vapor quality using stratified flow equations
for x ≥ xIA .

Bubbly (B)

Bubbly flows are not addressed these occur at mass velocities that are beyond the range of present
interest. The bubbly flow transition from intermittent flow remains the same as in the original
map, see Wojtan [155].
7.2. NEW TWO-PHASE PRESSURE DROP MODEL 125

7.2.4 Comparisons of experimental to predicted values

This section presents comparisons of experimental to predicted values obtained using the new
two-phase frictional pressure drop model.
Fig. (7.12) depicts comparisons of the entire database to predicted values. The new method predicts
82.3% within ±30% and 64.71% within ±20%. Taking into account that the range of experimental
conditions included a wide range of experimental conditions (fluids, diameter, mass velocities, heat
fluxes), the method predicts very accurately the experimental data. The values for all the database
versus the new and previous methods are summarized in Table (7.2).

Compilation of statistical results

±20% ±30%
Friedel 67.33% 51.81%
Grönnerud 46.15% 40.45%
Müller-Steinhagen and Heck 75.79% 49.64%
New Method 82.30% 64.71%

Table 7.2: Compilation of percentages of the database within an specific range for the new proposed
model and three existing methods.

In order to check the reliability of the new model, Figs. (7.13) and (7.14) depict comparisons
of experimental to predicted values when the fluid is changed for both internal tube diameters.
Figs. (7.15) and (7.16) show comparisons for both fluids when the tube internal diameter is changed.
It is deduced from the figures that changing these two important parameters does not have an effect
on the accuracy of the predictions attesting for the reliability of the new method.
126 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

200

+ 30 %
180

a) .
160
.
Predicted frictional pressure gradient [mbar/m]

.
140 ..
..
. . .
120
. .......... ..... ... .... . ........
........ . ..
100
. .. . . - 30 %
. .................................... . ......
.... .. .........................
80

.... ............................ . .
60
.. ......
............................................................ .
........ .
........................................... ....Experimental versus predicted values
40

............
.......................................
20
In range 82.3%
0
....
0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m]

200

+ 20 %
180

b) .
160
.
Predicted frictional pressure gradient [mbar/m]

.
140 ..
..
. . .
120
. .......... ..... ... ... . . ........
........ . ..
100
. .. . . - 20 %
. .................................. . ......
.... .. .........................
80

.... ........................... . .
60
.. .......
........................................................... .
.......
....................................................Experimental versus predicted values
40

...........
................................. .....
20
In range 64.71%
0 ......
0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m]

Figure 7.12: Comparisons of experimental to predicted values for the entire database: a) 82.30%
are predicted within ±30%, b) 64.71% are predicted within ±20%.
7.2. NEW TWO-PHASE PRESSURE DROP MODEL 127

200 200

+ 30 % + 20 %
180 180

a) b)
160 160
Predicted frictional pressure gradient [mbar/m]

Predicted frictional pressure gradient [mbar/m]

140 140
R22 R22
120 120

100 - 30 % 100 - 20 %
. ... ...................... . . ... ...................... .
80
... . .... 80
... . ....
.... .. ........................ .... .. ........................
.. ...... .. ......
. ... ........................... . . ... ........................... .
60 60

. .. . ..
.. ........... . .. ........... .
................................ . Experimental versus predicted values ................................ . Experimental versus predicted values
40 40

............... ...............
......................... .........................
20 20
In range 82.67% In range 60.26%
0.
. 0.
.
0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m] Experimental frictional pressure gradient [mbar/m]

Figure 7.13: Comparisons of experimental to predicted values for R22 and both tested diameters:
a) 82.67% are predicted within ±30%, b) 60.26% are predicted within ±20%.

200 200

+ 30 % + 20 %
180 180

a) . b) .
160
. 160
.
Predicted frictional pressure gradient [mbar/m]

Predicted frictional pressure gradient [mbar/m]

. .
140 .. 140 ..
R410A .. R410A ..
. . . . . .
. ..... ...... ...
. . ..... ...... ...
.
. .... . .. .. . . ........ . .... . .. .. . . ........
120 120

.. .... .. ....
.. ............. .. .............
100 - 30 % 100 - 20 %

.. . . ...
..
. .. . . ...
..
.
. .... .. .. .. . .... .. .. ..
... ............ . . ... ............ . .
80 80

60
................. .... 60
................. ....
.. ........................... ......
. . . . .. .. ........................... ......
. . . . ..
. .
...............................Experimental versus predicted values ...............................Experimental versus predicted values
40 40

.. .
..
.
..
. ...
. . .. .
..
.
..
. ...
. .
............................ ............................
20 20
In range 83.02% In range 69.46%
0.
..... 0.
.....
0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m] Experimental frictional pressure gradient [mbar/m]

Figure 7.14: Comparisons of experimental to predicted values for R410A and both tested diameters:
a) 83.02% are predicted within ±30%, b) 69.46% are predicted within ±20%.
128 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

200 200

+ 30 % + 20 %
180 180

a) . b) .
160
. 160
.
Predicted frictional pressure gradient [mbar/m]

Predicted frictional pressure gradient [mbar/m]

. .
140 .. 140 ..
d=8 mm .. d=8 mm ..
. .. . ... . . .. . ... .
. . .. . . ..
. .. ....... .. ..... .... . . ...... . .. ....... .. ..... .... . . ......
120 120

. . . .
.. .. .. ..
............................. .............................
100 - 30 % 100 - 20 %
...... ......
. ... ..... . . ... ..... .
... .................... . ... .................... .
80 80

. .......... ... . .......... ...

60
. .. .... 60
. .. ....
.................................. . .. . .................................. . .. .
.. .. .. ..
........................... .....Experimental versus predicted values ........................... .....Experimental versus predicted values
40 40

... . ... .
.............. ..............
................. .................
20 20
In range 82.06% In range 67.64%
0.
. . 0.
. .
0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m] Experimental frictional pressure gradient [mbar/m]

Figure 7.15: Comparisons of experimental to predicted values for d = 8 mm and both tested fluids:
a) 82.06% are predicted within ±30%, b) 67.64% are predicted within ±20%.

200 200

+ 30 % + 20 %
180 180

a) b)
160 160
Predicted frictional pressure gradient [mbar/m]

Predicted frictional pressure gradient [mbar/m]

140 140
d=13.8 mm d= 13.8 mm
120 120

100 - 30 % 100 - 20 %
. ... ...................... . . ....... .
... ... . ................. ... ... . . ......
80 80

. .. . ....................
....................... . . . ...
......................................
60 60
. . . .
. ................. .... . . .
.
........
.................................. . Experimental versus predicted values ............
...................................... Experimental versus predicted values
40 40

. . . .
............
..................... ...............................
......................... .
20 20
In range 82.24% .
...
..
.
. In range 62.86%
0. 0.
....
0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m] Experimental frictional pressure gradient [mbar/m]

Figure 7.16: Comparisons of experimental to predicted values for d = 13 mm and both tested
diameters: a) 82.24% are predicted within ±30%, b) 62.86% are predicted within ±20%.
7.2. NEW TWO-PHASE PRESSURE DROP MODEL 129

Fig. (7.17) shows comparisons of experimental to predicted values for the entire database segregated
by flow regime. For the sake of simplicity, the mist and dryout results are not shown in this figure but
a similar behaviour was observed. It has to be pointed that pressure drop experimental values in the
stratified-wavy (SW) and slug+stratified-wavy (Slug+SW) regimes were in the range of 0 − 8 mbar
and trying to accurately predict such low values is a difficult task because small deviations can
induce large errors. This later feature assesses the quality of the proposed model.

200 200

+ 30 % + 30 %
180 180

a) . b)
160
. 160
Predicted frictional pressure gradient [mbar/m]

Predicted frictional pressure gradient [mbar/m]

.
140 .. 140
Annular .. Slug+Intermittent
. . .
120
. . .... 120
....
.. ... ..
.. ..............
100 - 30 % 100 - 30 %
.
. . ..... . .. ..
..... .........................
80 80

. ...
60
.... ............................ 60 .... .
..... .
.. .
............................................... ..
40
................................ . Experimental versus predicted values
40
. .................. ..
....... .
. .......................... ..
Experimental versus predicted values
. ..
...................
.
. ..
.. ...........
20 20
. .... .
.............
In range 94.70% In range 70.63%
0 0
0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200

Experimental frictional pressure gradient [mbar/m] Experimental frictional pressure gradient [mbar/m]

20 20

+ 30 % + 30 %
18 18

c) d)
16 . 16
.
Predicted frictional pressure gradient [mbar/m]

Predicted frictional pressure gradient [mbar/m]

14 14
SW . Slug+SW
12 ...... .. 12

.
10 . - 30 % 10 - 30 %
. . . .
8 .
.. ... .. . .. 8

.. ............. . .
6
. . .................... .
. 6 .. .
..
........... . .
4
............................ Experimental versus predicted values 4

........... .
.. . . . .
............ .
Experimental versus predicted values
.. ......................
.......
2 2

...... In range 76.16%

. In range 69.47%
0 0
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20

Experimental frictional pressure gradient [mbar/m] Experimental frictional pressure gradient [mbar/m]

Figure 7.17: Comparisons of experimental to predicted values for the entire database segregated
by flow regime: a) annular, b) Slug+Intermittent, c) SW, d)Slug+SW.

Thus, it has been shown that the new two-phase frictional prediction model from a statistical point
of view is able to very accurately predict the experimental values. The reliability of the proposed
model has also been proven but still a question remains open: Does the new model capture the
130 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

trends observed in the two-phase frictional pressure drops vs. vapor quality? This latter is crucial
for the thermal optimization of evaporators. Figs. (7.18)-(7.22) show experimental and predicted
values plotted vs. quality for a selected set of experimental conditions. The new model follows the
experimental trends quite well and it is able to capture reasonably well the position and a little
less so the magnitude of the characteristic peak at high vapor qualities.

20 20

o 2 2 o 2 2
18 R134a, d=13 mm, T=5 C, G=150 kg/m s, q=06-09 kW/m 18 R134a, d=13 mm, T=5 C, G=150 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

16 16

14 14
Experimental Experimental
12 Predicted 12 Predicted

10 10

8 8

6 6

4 4
87.45% in 30% 57.89% in 30%
2 2

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

20

o 2 2
18 R134a, d=13 mm, T=5 C, G=150 kg/m s, q=30-40 kW/m
frictional pressure gradient [mbar/m]

16

14
Experimental
12 Predicted

10

4
100% in 30%
2

0
0 10 20 30 40 50 60 70 80 90 100

vapor quality [%]

Figure 7.18: Experimental and predicted values vs. vapor quality for R134a at different experi-
mental conditions.
7.2. NEW TWO-PHASE PRESSURE DROP MODEL 131

20 20

o 2 2 o 2 2
18 R22, d=13 mm, T=5 C, G=150 kg/m s, q=06-09 kW/m 18 R22, d=13 mm, T=5 C, G=150 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

16 16

14 14
Experimental Experimental
12 Predicted 12 Predicted

10 10

8 8

6 6

4 4
92.86% in 30% 95.45% in 30%
2 2

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

80 80

o 2 2 o 2 2
70 R22, d=13 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m 70 R22, d=13 mm, T=5 C, G=300 kg/m s, q=30-40 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

60 60

Experimental Experimental
50 50
Predicted Predicted

40 40

30 30

20 20

10 87.50% in 30% 10 95.24% in 30%

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

120 160

o 2 2 o 2 2
R22, d=13 mm, T=5 C, G=500 kg/m s, q=06-09 kW/m 140 R22, d=13 mm, T=5 C, G=600 kg/m s, q=06-09 kW/m
100
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

120

80
Experimental Experimental
100
Predicted Predicted

60 80

60
40

40

20
70.83% in 30% 20 63.16% in 30%

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 7.19: Experimental and predicted values vs. vapor quality for R22 at different experimental
conditions and d = 13.8 mm.
132 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

80 80

o 2 2 o 2 2
70 R22, d=8 mm, T=5 C, G=200 kg/m s, q=06-09 kW/m 70 R22, d=8 mm, T=5 C, G=200 kg/m s, q=15-20 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

60 60

Experimental Experimental
50 50
Predicted Predicted

40 40

30 30

20 20

10 88.00% in 30% 10 100.00% in 30%

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

140 140

o 2 2 o 2 2
R22, d=8 mm, T=5 C, G=350 kg/m s, q=06-09 kW/m R22, d=8 mm, T=5 C, G=350 kg/m s, q=35-40 kW/m
120 120
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

100 100
Experimental Experimental
Predicted Predicted
80 80

60 60

40 40

20 88.89% in 30% 20 84.62% in 30%

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 7.20: Experimental and predicted values vs. vapor quality for R22 at different experimental
conditions and d = 8 mm.
7.2. NEW TWO-PHASE PRESSURE DROP MODEL 133

40
14 o 2 2 o 2 2
R410A, d=13 mm, T=5 C, G=200 kg/m s, q=06-09 kW/m 35 R410A, d=13 mm, T=5 C, G=300 kg/m s, q=06-09 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

12
30

10
Experimental Experimental
25
Predicted Predicted
8
20

6
15

4 10

2 96.88% in 30% 5 89.98% in 30%

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

40 70

o 2 2 o 2 2
35 R410A, d=13 mm, T=5 C, G=300 kg/m s, q=15-20 kW/m R410A, d=13 mm, T=5 C, G=500 kg/m s, q=06-09 kW/m
60
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

30
50
Experimental Experimental
25
Predicted Predicted
40
20
30
15

20
10

5 91.67% in 30% 10 100.00% in 30%

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

70 80

o 2 2 o 2 2
R410A, d=13 mm, T=5 C, G=500 kg/m s, q=15-20 kW/m 70 R410A, d=13 mm, T=5 C, G=600 kg/m s, q=30-40 kW/m
60
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

60
50
Experimental Experimental
50
Predicted Predicted
40
40
30
30

20
20

10 95.00% in 30% 10 78.57% in 30%

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 7.21: Experimental and predicted values vs. vapor quality for R410A at different experi-
mental conditions and d = 13.8 mm.
134 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL

40 80

o 2 2 o 2 2
35 R410A, d=8 mm, T=5 C, G=200 kg/m s, q=06-09 kW/m 70 R410A, d=8 mm, T=5 C, G=350 kg/m s, q=06-09 kW/m
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

30 60

Experimental Experimental
25 50
Predicted Predicted

20 40

15 30

10 20

5 73.91% in 30% 10 94.12% in 30%

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

140 180

o 2 2 o 2 2
R410A, d=8 mm, T=5 C, G=500 kg/m s, q=06-09 kW/m 160 R410A, d=8 mm, T=5 C, G=600 kg/m s, q=30-40 kW/m
120
frictional pressure gradient [mbar/m]

frictional pressure gradient [mbar/m]

140
100
120
Experimental Experimental
Predicted Predicted
80 100

60 80

60
40
40

20 83.33% in 30% 100.00% in 30%

20

0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

vapor quality [%] vapor quality [%]

Figure 7.22: Experimental and predicted values vs. vapor quality for R410A at different experi-
mental conditions and d = 8 mm.

Comments on the attributes of the new flow pattern based method:

− It is more accurate (according to Table (7.2)).

− It better follows the variation in pressure gradient with vapor quality.

− It captures the peak in the pressure gradient at high vapor qualities.

− It handles mist flow and the dryout region.

− It goes to the correct limits at x = 0 (single-phase liquid flow) and at x = 1 (single-phase gas
flow).

− It explicitly includes the effect of interfacial waves.

− It includes the effect of interfacial flow structure (such as partially dry perimeter) via the flow
pattern map.

− It is based on the actual mean velocities of the phases via the void fraction function rather
than superficial velocities.
7.3. CONCLUSIONS 135

Furthermore, in thermal design with a vapor quality change from x = 0.2 at the inlet to x = 1 at
the outlet, it is most important that the two-phase pressure gradient is calculated accurately where
the gradients are high. For example, a 20% error at x = 0.2 is insignificant in the total pressure
drop calculation. Hence, the present method that works very well in annular flow (see Fig. (7.17a))
and that captures the position of the peak will give much better total pressure drop predictions
than the other methods.

7.3 Conclusions

The experimental results were compared to three recommended two-phase frictional pressure drop
correlations: Friedel, Grönnerud and Müller-Steinhagen and Heck. While all three provided a
reasonable agreement for some particular set of experimental conditions, they either significantly
overpredicted or underpredicted the data for most of the others. Also, the methods did not reliably
capture the variation in two-phase frictional pressure drop versus vapor quality, which is necessary
for such methods to be useful for the thermal optimization of evaporators. Hence,the development
of new two-phase frictional pressure drop model based on flow pattern map is justified. Then, the
experimental data were segregated by flow regime using the recent Wojtan-Ursenbacher-Thome flow
pattern map in order to develop the two-phase frictional pressure drop model based on flow pattern
map, the new model is introduced and the development procedure is detailed. Comparisons with
the database show that the new model is able to accurately and reliably predict experimental values
and, furthermore, the variation of the frictional pressure drop vs vapor quality is well captured.
136 CHAPTER 7. NEW TWO-PHASE FRICTIONAL PRESSURE DROP MODEL
Chapter 8

Conclusions

As the first step in this work, a comprehensive experimental study was undertaken in order to ob-
tain accurate two-phase pressure drop values during evaporation of refrigerants in horizontal tubes.
The experimental conditions were chosen to obtain experimental values over a wide range of test
parameters so that the effect of each parameter could be easily identified. The range of experi-
mental conditions covered were: three refrigerants (low, medium and high pressure, respectively),
two internal tube diameters, eight mass velocities and four heat fluxes. In addition, using water as
heating fluid we were able to cover the entire range of vapor quality. Fulfilling these experimental
requirements imposed a modification of the existing LTCM experimental facility and the imple-
mentation of new test sections. The new two-zone test sections allowed tests to (i) be run under
both diabatic and adiabatic conditions simultaneously, (ii) to obtain two-phase pressure drop values
for nearly every flow regime and (iii) to validate the data reduction procedure used to obtain the
frictional component of the pressure drop. The campaign acquired 2543 experimental two-phase
pressure drop values covering a wide range of experimental conditions and five flow regimes.
Next, based on a comprehensive state-of-the-art review and comparison versus leading two-phase
frictional pressure drop prediction methods, it was proven that none of these methods were able to
accurately and reliably predict the present experimental values. While leading methods provided
partial agreement for a particular set of experimental parameters, they either significantly overpre-
dicted or underpredicted the data for a different set of test variables. Furthermore, they did not
adequately capture the trends observed in the two-phase frictional pressure drop vs. vapor quality
nor the characteristic peak in the data at high vapor quality. Hence, it has been shown that none
of the existing methods was reliable for optimizing the thermal-hydraulics of a direct-expansion
evaporator, even though statistically they are not that bad when considering the normal situation
for the prediction of two-phase frictional pressure drops.
For the reasons mentioned above, an analytical study was undertaken in order to develop a new
two-phase frictional pressure drop prediction method. It has been shown in the literature that
models developed based on a so called ”phenomenological approach” could provide more accurate
and realistic predictions as the interfacial structure between the phases is taken into account. Based
on that, a phenomenological approach was chosen in the present study to develop a new two-phase
pressure drop prediction method. To do this, the interfacial structure must be known in advance,
which can be determined based on a flow pattern map. Hence, based on a comprehensive literature
review of existing flow pattern maps, the recent Wojtan-Ursenbacher-Thome map was chosen to
provide the corresponding interfacial structure needed for the development of the new model.

137
138 CHAPTER 8. CONCLUSIONS

Hence, the entire experimental data were segregated by flow regime using this flow pattern map.
Because about one-third of the experimental data were in the annular flow regime, this particular
geometry was used as a starting point for model development and then extended to other flow
regimes. The new model treats each flow regime (i.e. interfacial structure) separately and then
ensures a smooth transition at the flow regime boundaries (no jump in pressure gradient), being in
agreement with the experimental observations. Another important feature of the proposed model
is that it matches the corrects limits at x = 0 (single-phase liquid flow) and at x = 1 (single-phase
gas flow).
Finally, based on a statistical analysis, it is concluded that the New Two-Phase Frictional Pressure
Drop Model Based on Flow Pattern Map successfully predicts the present experimental data and is
able to provide much better predictions that existing methods. It is recommended for optimizing
the thermal-hydraulics of a direct-expansion evaporators.
The main advantages of the new method relative to previous methods are summarized below:

− It has been developed based on a ’phenomenological approach’ as the interfacial structure of

the various flow regimes is taking into account.

− It includes explicitly the effects of interfacial waves and flow patterns.

− It ensures smooth transitions (no jumps) between flow regimes, being in agreement with the
experimental observations.

− It matches the correct limits for single-phase flow (single-phase liquid flow at x = 0 and
single-phase gas flow at x = 1) .

− It presents a reasonable degree of simplicity as no iteration is required for its application.

− It successfully predicts experimental data, based on a statistically based, and their change of
trend with flow pattern.

− It captures the variation of the frictional pressure drop versus vapor quality, which is crucial
for optimizing the thermal-hydraulics of evaporators.

Future work should extend comparison of the new model to other fluids and databases. This model
should also be compared to two-phase pressure drops for intube condensing flows and modified
to account for any differences arising from film condensation on the upper perimeter of stratified
types of flow and the flow pattern map for condensation. Furthermore, two-phase pressure drops
in U-bends and downstream of U-bends should also be investigated.
Appendix A

Fluid Physical Properties

Physical Properties
Fluid TSat[o C] ρL [kg/m3 ] ρG [kg/m3 ] µL [kg/ms] µG [kg/ms] hL [kJ/kg] hG [kJ/kg]
0 1295 14.53 0.000331 1.08E-05 49.17 246.6
R134a 5 1279 17.26 0.000312 1.10E-05 55.81 249.4
10 1261 20.38 0.000294 1.12E-05 62.54 252.1
0 1285 21.22 0.000232 1.19E-05 44.59 250.0
R22 5 1268 24.78 0.000226 1.21E-05 50.49 251.7
10 1250 28.81 0.000220 1.23E-05 56.46 253.4
0 1176 30.67 0.000163 1.23E-05 56.46 277.0
R410A 5 1155 35.95 0.000153 1.25E-05 63.89 278.4
10 1133 41.98 0.000144 1.27E-05 71.44 279.6

Physical Properties
Fluid PSat [bar] kL [W/mK] kG [W/mK] cP L [kJ/kgK] cP G [kJ/kgK] σ[N/m] Pred [−]
2.93 0.09941 0.01179 1.308 0.9034 0.01157 0.07208
R134a 3.49 0.09709 0.01222 1.327 0.9286 0.01085 0.08608
4.15 0.09477 0.01264 1.346 0.9553 0.01015 0.1021
4.97 0.1001 0.009396 1.160 0.7229 0.01174 0.09997
R22 5.83 0.09762 0.009696 1.173 0.7436 0.01099 0.1173
6.80 0.09514 0.009996 1.187 0.7660 0.01025 0.13680
7.98 0.1139 0.01243 1.453 1.1180 0.01174 0.1622
R410A 9.33 0.1108 0.01282 1.472 1.1700 0.01099 0.1896
10.85 0.1076 0.01320 1.495 1.2280 0.01025 0.2203

Table A.1: Physical properties for fluids used during the experimental campaign at three different
saturation temperatures.

139
140 APPENDIX A. FLUID PHYSICAL PROPERTIES
Appendix B

Liquid-Vapor Interfaces

Instabilities associated with liquid-vapor interfaces can have a significant impact on the pressure
drop (as well as on heat and mass transfer) during phase-change processes. Often these instabil-
ities cause a change in the morphology of the two-phase system at a particular set of transition
conditions. Altering the interface morphology results in changes in transport properties at the
interface.
P e rtu rb e d
g V a p o r in te r fa c e
u G

L iq u id U n p e rtu rb e d
u in te r fa c e
L

Figure B.1: Perturbed vapor-liquid interface.

The one-dimensional Taylor instability wavelength λT for the unsupported liquid film is:

 1/2
(ρL − ρG ) √
λT = 2π 3 (B.1)
σ

and this is related to the formation of interfacial waves. If the interfacial waves have characteristics
wavelengths similar to the film thickness, then substituting δ for λT means that the interfacial
roughness ∆δi will be approximately scaled as

 k
(ρL − ρG )gδ2
∆δi ∝ (B.2)
σ

where term inside the brackets is non-dimensional.

141
142 APPENDIX B. LIQUID-VAPOR INTERFACES
Bibliography

[1] S. S. Agrawal, G. A. Gregory, and G. W. Govier. An analysis of horizontal stratified two

phase flow in pipes. Can. J. Chem. Eng., 51:280–286, 1973.

[2] N. Andritsos and T. J. Hanratty. Influence of interfacial waves in stratified gas-liquid flows.
AIChE Journal, 33(3):444–454, 1987.

[3] C. Aprea, A. Greco, and G. P. Vanoli. Local heat transfer coefficients and pressure drop
during evaporation in a smooth horizontal tube. In Proceedings of IMECE2002, New Orleans
Louisiana, 2002.

[4] M. M. Awad and Y. S. Muzychka. A simple asymptotic compact model for two-phase frictional
pressure gradient in horizontal pipes. In Proceedings of IMECE2004, Anaheim, California,
2004.

[5] B. J. Azzopardi. Drops in annular two-phase flow. Int. J. Multiphase Flow, 23:1–53, 1997.

[6] S. Badie, C. P. Hale, C. J. Lawrence, and G. H. Hewitt. Pressure gradient and holdup
in horizontal two-phase gas-liquid flows with low liquid loading. Int. J. Multiphase Flow,
26:1525–1543, 2000.

[7] O. Baker. Design of pipe lines for simultaneous flow of oil and gas. Oil and Gas J., 53:185–190,
1954.

[8] J. Bandel. Druckverlust und Wärmeügang bei der verdampfung siedender kältemittel im durch-
strömten waagerechten rohr. PhD thesis, University of Karlsruhe, 1973.

[9] S. G. Bankoff. A variable density single-fluid model two-phase flow with particular reference
to steam-water. J. Heat Transfer, 11(Series B):265–272, 1960.

[10] D. Barnea. A unified model for predicting flow pattern transitions for the hole range of pipe
inclinations. Int. J. Multiphase Flow, 13:1–12, 1987.

[11] C. J. Baroczy. A systematic correlation for two-phase pressure drop. Chem. Eng. Prog. Symp.
Ser., 62(44):232–249, 1965.

[12] D. R. H. Beattie. Two-phase flow structure and mixing length theory. Nucl. Eng. Desing,
21:46–64, 1972.

[13] D. R. H. Beattie. A note on the calculation of two-phase pressure losses. Nucl. Eng. Desing,
25:395–402, 1973.

143
144 BIBLIOGRAPHY

[14] D. R. H. Beattie and P. B. Whalley. A simple two-phase frictional pressure drop calculation
method. Int. J. Multiphase Flow, 8:83–87, 1982.

[15] C. Ben-Salem and H. F. Ellouze. Pressure drop in annular horizontal two-phase flow. In
G. P. Celata, P. di Marco, and R. K. Shah, editors, Proceedings of the 2nd International
Symposium on Two-Phase Flow Modelling and Experimentation, volume 2, pages 1351–1355,
Rome, Italy, 1999. Edizioni ETS, Pisa.

[16] D. Biberg. An explcit equation for the wetted angle in two-phase stratified pipe flow. Can.
J. Chem. Eng., 77:1221–1224, 1999.

[17] D. Biberg. Liquid wall friction in two-phase turbulent gas laminar liquid stratified flow. Can.
J. Chem. Eng., 77:1073–1082, 1999.

[18] D. Biberg. Two-phase stratified pipe flow modelling: a new expression for the interfacial
shear stress. In G. P. Celata, P. di Marco, and R. K. Shah, editors, Proceedings of the
2nd International Symposium on Two-Phase Flow Modelling and Experimentation, volume 2,
pages 1–10, Rome, Italy, 1999. Edizioni ETS, Pisa.

[19] G. Bigot. Étude et conception de systèmes air/air inversables utilisant des mélanges a glisse-
ment de température. PhD thesis, École de Mines de Paris, 2001.

[20] J. A. Bouré. Two-phase flows and heat transfer with applicaton to nuclear design problems.
Hemisphere, New York, 1978.

[21] R. S. Brodkey. The phenomena of fluid motions. Dover publications, New York, 1967.

[22] Van P. Carey. Liquid-vapor phase-change phenomena. Taylor & Francis, Hebron, KY, 1992.

[23] A. Cavallini, G. Censi, D. Del Gol, L. Doretti, G. A. Longo, and L. rossetto. Experimen-
tal investigation on condensation heat transfer and pressure drop of new HFC refrigerants
(R134a, R125, R32, R410A, R236ea) in a horizontal tube. Int. J. Refrigeration, 24:73–87,
2001.

[24] A. Cavallini, D. Del Gol, L. Doretti, G. A. Longo, and L. rossetto. Condensation pressure
losses in smooth tubes. In E. W. P. Hahne, W. Heidemann, and K. Spindler, editors, Proceed-
ings of the 3rd European Thermal Sciences Conference, volume 2, pages 927–932, Heidelberg,
Germany, 2000. Edizioni ETS, Pisa.

[25] G. P. Celata, K. Mishima, and G. Zummo. Critical heat flux prediction for saturated flow
boiling of water in vertical tubes. Int. J. Heat Mass Transfer, 44:4323–4331, 2001.

[26] S. D. Chang and S. T. Ro. Pressure drop of pure HFC refrigerants and their mixtures flowing
in capillary tubes. Int. J. Multiphase Flow, 23(3):551–561, 1996.

[27] J. M. Chawla. Wärmeübergang and druckfall in waagerechten rohren bei der strömung von
verdampfenden kältemitteln, chapter Lg1-Lg2. 523. VDI-Forschungsh, 1967.

[28] I. Y. Chen and K. Yang. Two-phase pressure drop of air-water in small horizontal tubes. J.
Thermophysics Heat Transfer, 15(4):409–415, 2001.

[29] I. Y. Chen, K. Yang, Y. Chang, and C. Wang. Two-phase pressure drop of air-water and
R-410A in small horizontal tubes. Int. J. Multiphase Flow, 27:1293–1299, 2001.
BIBLIOGRAPHY 145

[30] J. J. J. Chen and P. L. Spedding. An extension of the Lockhart-Martinelli theory of two-phase

pressure drop and holdup. Int. J. Multiphase Flow, 7:659–675, 1981.

[31] D. Chisholm. Pressure gradients due to friction during the flow of evaporating two-phase
mixtures in smooth tubes and channels. Int. J. Heat Mass Transfer, 16:347–358, 1973.

[32] T. Y. Choi, Y. J. Kim, M. S. Kim, and S. T. Ro. Evaporation heat transfer of R-32, R-134a,
R-32/134a and R-32/125/134a inside a horizontal smooth tube. Int. J. Heat Mass Transfer,
43:3651–3660, 2000.

[33] G. Chupin, S. Labesque, P. Laux, and O.-J. Nydal. Air-oil-water pipe flow experiments at
low liquid loading. In Proceedings of the 40th ETPFG Meetings, Stockholm, Sweeden, 2002.

[34] A. Cicchitti, C. Lombardi, M. Silvestri, and G. Soldaini R. Zavattarelli. Two-phase cooling

experiments - pressure drop, heat transfer and burnout measurements. Energia Nucleare,
7(6):407–425, 1960.

[35] J. G. Collier and J. R. Thome. Convective boiling and condensation. Oxford University Press,
Oxford, 1984.

[36] M. K. Cooper. Saturated nucleate pool boiling: A simple correlation. In 1st U.K. National
Conference on Heat Transfer, Vol., volume 2, pages 785–793, 1984. (I. Chem. E. Symposium
Series N o 86).

[37] J. M. Delhaye. Two-phase flows and heat transfer in the process industries. Hemisphere, New
York, 1990.

[38] A. E. Dukler, Moye Wick III, and R. G. Cleveland. Frictional pressure drop in two-phase
flow:b. an approach through similarity analysis. AIChE Journal, 10(1):44–51, 1964.

[39] J. el Hajal, J. R. Thome, and A. Cavallini. Condensation in horizontal tubes, part 1:

two-phase flow pattern map. Int. J. Heat Mass Transfer, 46:3349–3363, 2003.

[40] J. el Hajal, J. R. Thome, and A. Cavallini. Condensation in horizontal tubes, part 2: new
heat transfer model based on flow regimes. Int. J. Heat Mass Transfer, 46:3365–3387, 2003.

[41] H. F. Ellouze. Contribution à l’étude de la chute de pression en écoulement diphasique. PhD

thesis, Université Paris-Val de Marne(Paris XII), 1991.

[42] Z. Fangde and L. Weidong. Interfacial shear stress and friction factor of stratified flow. In
G. P. Celata, P. di Marco, and R. K. Shah, editors, Proceedings of the 2nd International
Symposium on Two-Phase Flow Modelling and Experimentation, volume 2, pages 1263–1268,
Rome, Italy, 1999. Edizioni ETS, Pisa.

[43] L. B. Fore, S.G. Beus, and R. C. Bauer. Interfacial friction in gas-liquid annular flow: analogies
to full and transition roughness. Int. J. Multiphase Flow, 26:1755–1769, 2000.

[44] M. Fossa. A simple model to evaluate direct contact heat transfer and flow characteristics in
annular two-phase flow. Int. J. Heat Fluid Flow, 16:272–279, 1995.

[45] F. França and R. T. Lahey. The use of drift-flux tecniques for the analysis of horizontal
two-phase flows. Int. J. Multiphase Flow, 18:787–801, 1992.
146 BIBLIOGRAPHY

[46] L. Friedel. Improved friction drop correlations for horizontal and vertical two-phase pipe flow.
In European Two-phase Flow Group Meeting, paper E2, Ispra, Italy, 1979.

[47] L. Friedel. Pressure drop during gas/vapor-liquid flow in pipes. Int. Chemical Engineering,
20:352–367, 1980.

[48] F. Fu and J. F. Klausner. A separated flow model for predicting two-phase pressure drop
and evaporative heat transfer for vertical annular flow. Int. J. Heat Fluid Flow, 18:541–549,
1997.

[49] L. Galbiati and P. Andreini. The transition between stratified and annular regimes for hori-
zontal two-phase flow in small diameter tubes. Int. Comm. Heat Mass Transfer, 19:185–190,
1992.

[50] E. N. Ganic and W. M. Rohsenow. On the mechanism of liquid drop deposition in two-phase
dispersed flow. J. Heat Transfer, 101:288–294, 1979.

[51] H. Gao, H.-Y. Gu, and L.-J. Guo. Numerical study of stratified oil-water two-phase turbulent
flow in a horizontal tube. Int. J. Heat Mass Transfer, 46:749–754, 2003.

[52] E. Grolman and M. H. Fortuin. Gas-liquid flow in slightly inclined pipes. Chem. Eng. Sci.,
52:4461–4471, 1997.

[53] R. Gronnerud. Investigation of liquid hold-up, flow-resistance and heat transfer in circulation
type of evaporators, part iv: two-phase flow resistance in boiling refrigerans. In Annexe
1972-1, Bull. de l’Inst. du Froid, 1979.

[54] S. E. Haaland. Simple and explicit formulas for the friction factor in turbulent pipe flow. J.
Fluids Engineering, 105:89–90, 1983.

[55] P.J. Hamersma and J. Hart. A pressure drop correlation for gas/liquid pipe flow with a small
liquid holdup. Chem. Eng. Sci., 42(5):1187–1196, 1987.

[56] N. P. Hand, P. L. Speeding, and S. J. Ralph. The effect of surface tension on flow pattern,
holdup and pressure drop during horizontal air-water pipe flow at atmospheric conditions.
Chemical Engineering J., 48:197–210, 1992.

[57] T. J. Hanratty and J. M. Engen. Interaction between a turbulent air stream and a moving
water surface. AIChE Journal, 3:299–279, 1957.

[58] J. Hart, P. J. Hamersma, and J. M. H. Fortuin. Correlations predcting frictional presure drop
and liquid holdup during horizontal gas-liquid pipe flow with a small liquid holdup. Int. J.
Multiphase Flow, 15(6):947–964, 1989.

[59] K. Hashizume. Flow pattern and void fraction of refrigerant two-phase flow in a horizontal
pipe. Bulletin of the JSME, 26:1597–1602, 1983.

[60] K. Hashizume. Flow pattern, void fraction and pressure drop of refrigerant two-phase flow
in a horizontal pipe - I: Experimental data. Int. J. Multiphase Flow, 4(5):399–410, 1983.

[61] K. Hashizume, H. Ogiwara, and H. Taniguchi. Flow pattern, void fraction and pressure drop
of refrigerant two-phase flow in a horizontal pipe - II: Analysis of frictional pressure drop.
Int. J. Multiphase Flow, 11:643–658, 1985.
BIBLIOGRAPHY 147

[62] W. H. Henstock and T. J. Hanratty. The interfacial drag and the height of the wall layer in
annular flows. AIChE Journal, 22:990–1000, 1976.

[63] G. Hetsroni, D. Mewes, C. Enke, M. Gurevich, and A. Mosyak R. Rozenblit. Heat transfer
to two-phase flow in inclined tubes. Int. J. Multiphase Flow, 29:173–194, 2003.

[64] G. F. Hewitt and A. H. Govan. Phenomenological modelling of non-equilibrium flows with

phase change. Int. J. Heat Mass Transfer, 33:229–242, 1990.

[65] E. T. Hurlburt and T. A. Newell. Prediction of the circumferential film thickness distribution
in horizontal annular gas-liquid flow. J. Fluids Engineering, 122:396–402, 2000.

[66] W. Idsinga, N. Todreas, and R. Bowring. An assessment of two-phase pressure drop correla-
tion for steam-water systems. Int. J. Multiphase Flow, 3:401–413, 1977.

[67] M. Ishii. Thermo-fluid dynamics theory of two-phase flow. Eyrolles, Paris, 1975.

[68] T. Johannessen. A theoretical solution of the Lockhart-Martinelli flow mmodel for calculating
two-phase flow pressure and holdup. Int. J. Heat Mass Transfer, 15:1443–1449, 1972.

[69] D. S. Jung, M. McLinden, R. Radermacher, and D. Didion. A study of flow boiling heat
transfer with refrigerant mixtures. Int. J. Heat Mass Transfer, 32:1751–1764, 1989.

[70] D. S. Jung and R. Radermacher. Prediction of pressure drop during horizontal annular flow
boiling of pure and mixed refrigerants. Int. J. Heat Mass Transfer, 32(12):2435–2446, 1989.

[71] D. Juric and G. Tryggvason. Computations of boiling flows. Int. J. Multiphase Flow,
24(3):387–410, 1998.

[72] S. Kabelac and H.J. de Burg. Flow boiling of ammonia in a plain and a low finned horizontal
tube. Int. J. Refrigeration, 24:41–50, 2001.

[73] V. Kadambi. Void fraction and pressure drop in two-phase stratified flow. Can. J. Chem.
Eng., 59:584–589, 1981.

[74] K. G. Kandlikar. Development of a flow boiling map for subcooled and saturated flow boiling
of different fluids inside circular tubes. J. Heat Transfer, 113:190–200, 1991.

[75] N. Kattan. Contribution to the heat transfer analysis of substitube refrigerants in evaporator
tubes with smooth or enhanced surfaces. Thesis n 1498, Swiss Federal Institute of Technology
Lausanne, 1996.

[76] N. Kattan, J. R. Thome, and D. Favrat. Flow boiling in horizontal tubes: Part 1 - develop-
ment of a diabatic two-phase flow pattern map. J. Heat Transfer, 120:140–147, 1998.

[77] N. Kattan, J. R. Thome, and D. Favrat. Flow boiling in horizontal tubes: Part 2 - new heat
transfer data for five refrigerants. J. Heat Transfer, 120:148–155, 1998.

[78] N. Kattan, J. R. Thome, and D. Favrat. Flow boiling in horizontal tubes: Part 3 - devel-
opment of a new heat transfer model based on flow pattern. J. Heat Transfer, 120:156–165,
1998.

[79] D. Kim and A. J. Ghajar. Heat transfer measurements and correlations for air-water flow of
different flow patterns in a horizontal pipe. Exp. Thermal Fluid Sci., 25:659–676, 2002.
148 BIBLIOGRAPHY

[80] Y. Kim, K. Seo, and J. T. Chung. Evaporation heat transfer characteristics of R-410A in 7
and 9.52 mm smooth/microfin tubes. Int. J. Refrigeration, 25:716–730, 2002.

[81] J. F. Klausner, B. T. Chao, and S. L. Soo. An improved method for simultaneous determi-
nation of frictional pressure drop and vapor volume fraction in vertical flow boiling. Exp.
Thermal Fluid Sci., 3:404–415, 1990.

[82] E. Kordyban. Some detals of developing slugs in horizontal two-phase flow. AIChE Journal,
31(3):802–806, 1985.

[83] P. G. Kosky. Thin liquid films under simultaneous shear and gravity forces. Int. J. Heat
Mass Transfer, 14:1220–1224, 1970.

[84] J. E. Kowalski. Wall and interfacial shear stress in stratified flow in a pipe. AIChE Journal,
33(2):274–281, 1987.

[85] S. S. Kutateladze. On the transition to film boiling under natural convection. Kotlotur-
bostronie, 10(3), 1948.

[86] M. Lallemand, C. Branescu, and P. Haberschill. Coefficients d’échange locaux au cours de

l’ébullition du R22 et du R407C dans des tubes horizontaux, lisses ou micro-aileté. Int. J.
Refrigeration, 24:57–72, 2001.

[87] J. E. Laurinat, T. J. Hanratty, and J. C. Dallman. Pressure drop and film height mesurements
for annular gas-liquid flow. Int. J. Multiphase Flow, 10:341–356, 1984.

[88] K.-W. Lee, S.-J. Baik, and T.-S. Ro. An utilization of liquid sublayer dryout mechanism in
predicting critical heat flux under low pressure and low velocity conditions in round tubes.
Nuclear Engineering and Design, 200:69–81, 2000.

[89] P. Y. Lin and T. J. Hanratty. Detection of slug flow from pressure measurements. Int. J.
Multiphase Flow, 13:13–21, 1987.

[90] R. W. Lockhart and R. C. Martinelli. Proposed correlation of data for isothermal two-phase,
two-component in pipes. Chem. Eng. Process, 45(1):39–48, 1949.

[91] J. M. Mandhane, G. A. Gregory, and K. Aziz. A flow pattern map for gas-liquid flow in
horizontal pipes. Int. J. Multiphase Flow, pages 537–553, 1974.

[92] P. Melin. Measurements and modelling of convective vaporization for refrigerants in a hori-
zontal tube. PhD thesis, Chalmers university of technology Göteborg, 1996.

[93] J. Moreno-Quiben and J. R. Thome. Two-phase pressure drops in horizontal tubes: new
results for R-410A and R-134a compared to R-22, paper IIC00. In 21st IIR International
Congress of Refrigeration, Washington, D.C., USA, 2003.

[94] H. Mori, S. Yoshida, K. Oshidi, and Y. Kakimoto. Dryout qualiy and post-dryout heat
transfer coefficient in horizontal evaporator tubes. In E. W. P. Hahne, W. Heidermann, and
K. Springler, editors, Proceedings of the 3rd European Thermal Conference. Edizioni ETS,
Pisa, 2000.

[95] A. Mosyak and G. Hetsroni. Analysis of dryout in horizontal and inclined tubes. Int. J.
Multiphase Flow, 25:1521–1543, 1999.
BIBLIOGRAPHY 149

[96] A. A. Mouza, S. V. Paras, and A. J. Karabelas. CFD code application to wavy stratified
gas-liquid flow. Trans IChemE, 79(Part A):561–568, 2001.

[97] H. Müller-Steinhagen and K. Heck. A simple friction pressure correlation for two-phase flow
in pipes. Chem. Eng. Process, 20:297–308, 1986.

[98] C. H. Newton. An experimental and numerical study of stratified gas-liquid flow in horizontal
pipes. PhD thesis, University of New Soth Wales (Australia), 1997.

[99] C. H. Newton and M. Behnia. Estimation of wall shear stress in horizontal stratified gas-liquid
pipe flow. AIChE Journal, 42(8):2369–2373, 1996.

[100] C. H. Newton and M. Behnia. On the use of the stratified momentum balance for the
deduction of shear stress in horizontal gas-liquid pipe flow. Int. J. Multiphase Flow, 24:1407–
1423, 1998.

[101] C. H. Newton and M. Behnia. Numerical calculation of turbulent stratified gas-liquid pipe
flows. Int. J. Multiphase Flow, 26:327–337, 2000.

[102] M. K. Nicholson, K. Aziz, and G. A. Gregory. Intermittent two-phase flow in horizontal pipes:
Predictive methods. Can. J. Chem. Eng., 56:653–663, 1978.

[103] E. Nidegger, J. R. Thome, and D. Favrat. Flow boiling and pressure drop measurements for
R-134a/Oil mixtures. part 1: Evaporation in a microfin tube. HVAC&R Research, 3(1):39–52,
1997.

[104] Z. Olujić. Predicting two-phase flow friction loss in horizonta pipes. Chem. Eng., June
24:45–50, 1985.

[105] M. B. Ould-Didi, N. Kattan, and J. R. Thome. Prediction of two-phase gradients of refrig-

erants in horizontal tubes. Int. J. Refrigeration, 25:935–947, 2002.

[106] S. V. Paras and A. J. Karabelas. Properties of the liquid layer in horizontal annular flow.
Int. J. Multiphase Flow, 17:439–454, 1991.

[107] B. Pierre. Flow resistance with boiling refrigerants - Part 1. ASHRAE Journal, 6(9):58–65,
1964.

[108] W. M. Rohsenow. Post dryout heat transfer prediction. Int. Comm. Heat Mass Transfer,
15:559–569, 1988.

[109] H. Ross, R. Radermacher, and M. di Marzo. Horizontal flow boiling of pure and mixed
refrigerants. Int. J. Heat Mass Transfer, 30:979–992, 1987.

[110] S. Z. Rouhani and E. Axelsson. Calculation of void volume fraction in the subcooled and
quality boiling regions. Int. J. Heat Mass Transfer, 13:383–393, 1970.

[111] T. W. F. Russel, A. W. Etchells, R. H. Jensen, and P. J. Arruda. Pressure drop and holdup
in stratified gas-liquid flow. AIChE Journal, 20(4):664–669, 1974.

[112] J. M. Saiz-Jabardo and E. P. Bandarra-Filho. Convective boiling of halocarbon refrigerants

flowing in a horizontal cooper tube - an experimental study. Exp. Thermal Fluid Sci., 23:93–
104, 2000.
150 BIBLIOGRAPHY

[113] P. K. Sarma, V. D. Rao, T. Subrahmanyan, S. Kakac, and H. T. Liu. A method to predict

two-phase pressure drop using condensation heat transfer data. Int. J. Therm. Sci., 39:184–
190, 2000.

[114] H. Schlichting. Boundary layer theory. McGraw-Hill, New York, 1980.

[115] K. Seo and Y. Kim. Evaporation heat transfer and pressure drop of R-22 in 7 and 9.52 mm
smooth/micro-fin tubes. Int. J. Heat Mass Transfer, 43:2869–2882, 2000.
[116] K. Seo, Y. Kim, K.-J. Lee, and Y. Park. An experimental study on convective boiling of
R-22 and R-410A in horizontal smooth and micro-fin tubes. KSME International Journal,
15(8):1156–1164, 2001.

[117] M. M. Shah. A new correlation for heat transfer during boiling flow through pipes. ASHRAE
Transaction, 82(Part 2), 1976.
[118] O. Shoham and Y. Taitel. Stratified turbulent-turbulent gas-liquid flow in horizontal and
inclined pipes. AIChE Journal, 30(3):377–385, 1984.
[119] T. N. Smith and R. W. F. Tait. Interfacial shear stress and momentum transfer in horizontal
gas-liquid flow. Chem. Eng. Sci., 21:63–75, 1966.

[120] H. M. Soliman. On the annular-to-wavy flow pattern transition during condensation inside
horizontal tubes. Int. J. Heat Mass Transfer, 60:475–481, 1982.

[121] P. L. Spedding. Prediction in stratified gas-liquid co-current flow in horzontal pipes. Int. J.
Heat Mass Transfer, 40(8):1923–1935, 1997.
[122] D. Steiner. VDI-Wärmeatlas (VDI Heat Atlas), chapter Hbb. Verein Deutscher Ingenieure
VDI-Gessellschaft Verfahrenstechnik und Chemieingenieurwessen (GCV), Düsseldorf, 1993.
[123] K. Stephan and M. Abdelsalam. Heat-transfer correlations for natural convection boiling.
Int. J. Heat Mass Transfer, 23:73–87, 1980.

[124] N. D. Sylvester and J. P. Brill. Drag reduction in two-phase annular-mist flow of air and
water. AIChE Journal, 22(3):615–617, 1976.
[125] Y. Taitel and A. E. Dukler. A model for predicting flow regime transitions in horizontal and
near horizontal gas-liquid flow. AIChE Journal, 22:47–55, 1976.
[126] Y. Taitel and A. E. Dukler. A theoretical approach to the Lockhart-Martinelli correlation for
stratified flow. Int. J. Multiphase Flow, 2:591–595, 1976.

[127] T. N. Tandon, H. K. Varma, and Gupta C. P. A void fraction model for annular two-phase
flow. Int. J. Heat Mass Transfer, 28:191–198, 1985.
[128] John R. Taylor. An introduction to error analysis. The study of incertainties in physical
measurements. University Science Books, Sausalito, CA, second edition, 1997.
[129] J. R. S. Thom. Prediction of pressure drop during forced circulation boiling of water. Int. J.
Heat Mass Transfer, 7:709–724, 1964.

[130] J. R. Thome. On recent advances in modelling of two-phase flow and heat transfer. In Pro-
ceedings of the 1st International Conference on Heat Transfer, Fluid Mechanics and Thermo-
dynamics, Kruger Park, South frica, 2002.
BIBLIOGRAPHY 151

[131] J. R. Thome and J. el Hajal. Two-phase flow pattern map for evaporation in horizontal
tubes: latest version. In Proceedings of the 1st International Conference on Heat Transfer,
Fluid Mechanics and Thermodynamics, Kruger Park, South frica, 2002.

[132] J. R. Thome and J. El Hajal. Flow boiling heat transfer to carbon dioxide: general prediction
method. Int. J. Refrigeration, 27:294–301, 2004.

[133] T. N. Tran, M. Chyu, M. W. Wambsganss, and D. M. France. Two-phase pressure drop of re-
frigerants during flow boiling in small channels: an experimental investigation and correlation
development. Int. J. Multiphase Flow, 26:1739–1754, 2000.

[134] C. Tribbe and H. M. Müller-Steinhagen. An evaluation of the performance of phenomeno-

logical models for predicting models for predicting pressure gradient during gas-liquid flow in
horizontal pipelines. Int. J. Multiphase Flow, 26:1019–1036, 2000.

[135] C. P. Tso and S. Sugawara. Film thickness prediction in a horizontal annular two-phase flow.
Int. J. Multiphase Flow, 16:867–884, 1990.

[136] T. Ursenbacher, L. Wojtan, and J. R. Thome. Interfacial measurements in stratified types of

flow. Part I: Mesurements for R22 and R-410A. Int. J. Multiphase Flow, 30:125–137, 2004.

[137] VDI. VDI-Wärmeatlas (VDI Heat Atlas). Verein Deutscher Ingenieure VDI-Gessellschaft
Verfahrenstechnik und Chemieingenieurwessen (GCV), Düsseldorf, 1993.

[138] P. K. Vijayan, A. P. Patil, D. S. Pilhwal, D. Saha, and V. Venkat-Raj. An assessment of

pressure drop and void fraction correlations with data from two-phase natural circulation
loops. Heat Mass Transfer, 36:541–548, 2000.

[139] N. A. Vlachos, S. V. Paras, and A. J. Karabelas. Liquid-to-wall shear stress distribution in

stratified/atomization flow. Int. J. Multiphase Flow, 23(5):845–863, 1997.

[140] N. A. Vlachos, S. V. Paras, and A. J. Karabelas. Prediction of holup, axial pressure gradient
and wall shear stress in wavy stratified and stratified/atomization gas/liquid flow. Int. J.
Multiphase Flow, 25:365–376, 1999.

[141] G. B. Wallis. Annular two-phase flow. part 1: A simple theory. J. Basic Engineering, pages
59–72, 1970.

[142] M. W. Wambsganss, J. A. Jendrzejczyk, and D. M. France. Two-phase flow patterns and

transitions in a small, horizontal, rectangular channel. Int. J. Multiphase Flow, 17:327–342,
1991.

[143] C. Wang and C. Chiang. Two-phase heat transfer characteristics for R-22/R-407C in a 6.5-
mm smooth tube. Int. J. Heat Fluid Flow, 18:550–558, 1997.

[144] C. Wang, C. Chiang, and D. Lu. Visual observation of two-phase flow pattern of R-22, R-134a
and R-407C in a 6.5-mm smooth tube. Exp. Thermal Fluid Sci., 15:395–405, 1997.

[145] C. Wang, C. Chiang, and J. Yu. An experimental study of in-tube evaporation of R-22 inside
a 6.5-mm smooth tube. Int. J. Heat Fluid Flow, 19:259–269, 1998.

[146] C.-C. Wang, S.-K. Chiang, Y.-J. Chang, and T.-W. Chung. Two-phase flow resistence of
refrigerants R-22, R-410A and R-407C in small diameter tubes. Trans. IChemE, 79:553–560,
2001.
152 BIBLIOGRAPHY

[147] C.-C. Wang, J.-G. Yu, S:-P. Lin, and D.-C. Lu. An experimental study of convective boiling
of refrigerants R-22 and R-410A. ASHRAE Trans, 104:1144–1150, 1998.

[148] Z. Wang, K. S. Gabriel, and D. L. Manz. The influences of wave height on the interfacial fric-
tion in annular gas-liquid flow under normal and microgravity conditions. Int. J. Multiphase
Flow, 30:1193–1211, 2004.

[149] J. Weisman, D. Duncan, J. Gibson, and T. Crawford. Effects of fluid properties and pipe
diameter on two-phase flow patterns in horizontal lines. Int. J. Multiphase Flow, 5:437–462,
1979.

[150] P. B. Whalley. Two-phase flow and heat transfer. Oxford University Press, Oxford, 1996.

[151] H. Wijaya and M. W. Spatz. Two-phase flow heat transfer and pressure drop characteristics
of R-22 and R-32/125. ASHRAE Trans, 101:1020–1027, 1995.

[152] M. J. Wilson, T. A. Newell, J. C. Chato, and C. A. Infante Ferreira. Refrigerant charge,

pressure drop, and condensation heat transfer in flattened tubes. Int. J. Multiphase Flow,
26:442–451, 2003.

[153] L. Wojtan. Experimental and analytical investigation of void fraction and heat transfer dur-
ing evaporation in horizontal tubes. Thesis n 2978, Swiss Federal Institute of Technology
Lausanne, 2004.

[154] L. Wojtan, T. Ursenbacher, and J. R. Thome. Interfacial measurements in stratified types of

flow. Part II: Mesurements for R22 and R-410A. Int. J. Multiphase Flow, 30:125–137, 2004.

[155] L. Wojtan, T. Ursenbacher, and J. R. Thome. Investigation of flow boiling in horizontal

tubes: Part I - A new diabatic two-phase flow pattern map. Int. J. Heat Mass Transfer,
48:2955–2969, 2005.

[156] L. Wojtan, T. Ursenbacher, and J. R. Thome. Investigation of flow boiling in horizontal

tubes: Part II - Development of a new heat transfer model for stratified-wavy, dryout and
mist flow regimes. Int. J. Heat Mass Transfer, 48:2970–2985, 2005.

[157] S. Wongwises, T. Wongchang, and J. Kaewon. A visual study of two-phase flow patterns of
HFC-134a and lubricant oil mixtures. Heat Transfer Eng., 23:13–22, 2002.

[158] C. Yang and C. Shieh. Flow pattern of air-water and two-phase R-134a in small circular
tubes. Int. J. Multiphase Flow, 27:1163–1177, 2001.

[159] M. Yu, T. Lin, and C. Tseng. Heat transfer and flow pattern during two-phase flow boilng
of R-134a in horizontal smooth and microfin tubes. Int. J. Refrigeration, 25:789–798, 2002.

[160] W. Yu, D. M. France, M. W. Wanbsganss, and J. R. Hull. Two-phase pressure drop, boiling
heat transfer and critical heat flux in a small-diameter tube. Int. J. Multiphase Flow, 28:927–
941, 2002.

[161] M. Zhang and R. L. Webb. Correlation of two-phase friction for refrigerants in small-diameter
tubes. Exp. Thermal Fluid Sci., 25:131–139, 2001.

[162] O. Zürcher. Contribution to the heat transfer analysis of natural and substitube refrigerants
evaporated in a smooth horizontal tube. Thesis n 2122, Swiss Federal Institute of Technology
Lausanne, 2000.
BIBLIOGRAPHY 153

[163] O. Zürcher, D. Favrat, and J. R. Thome. Development of a adiabatic two-phase flow pattern
map for horizontal flow boiling. Int. J. Heat Mass Transfer, 45:291–301, 2002.

[164] O. Zürcher, J. R. Thome, and D. Favrat. Flow boiling and pressure drop measurements for
R-134a/Oil mixtures. part 1: Evaporation in a plain tube. HVAC&R Research, 3(1):54–64,
1977.

[165] O. Zürcher, J. R. Thome, and D. Favrat. Evaporation of ammonia in a smooth horizontal

tube: heat transfer measurement and predictions. J. Heat Transfer, 121:89–101, 1999.

[166] O. Zürcher, J. R. Thome, and D. Favrat. An onset of nucleate boiling criterion for horizontal
flow boiling. Int. J. Therm. Sci., 39:909–918, 2000.
154 BIBLIOGRAPHY
Nomenclature

Latin

Symbol Description Units

A Cross Section Area m2
AL Cross Section Area of Liquid Phase m2
ALD Non-dimensional Area of Liquid Phase - [AL /D] -
AG Cross Section Area of Vapor Phase m2
AGD Non-dimensional Area of Vapor Phase - [AV /D] -
C Empirical Constant -
cpL Liquid Specific Heat J/kgK
cpG Vapor Specific Heat J/kgK
cpw Heating Water Specific Heat J/kgK
d Vapor Core Diameter m
D Internal Tube Diameter m
Dext External Tube Diameter m
Dh Hydraulic Diameter m
fL Liquid Friction Factor -
fG Vapor Friction Factor -
F rL Liquid Froude Number -
g Acceleration of Gravity m/s2
G Mass Velocity kg/m2 s
hL Liquid Height in the Tube m
hLD Non-dimensional Liquid Height in the Tube - [hL /D] -
hL Liquid Enthalpy J/kg
hLV Latent Heat of Vaporisation J/kg
hw Enthalpy of the Heating Water per Unit of Mass J/kg
j Total Superficial Velocity m/s
jL Liquid Superficial Velocity m/s
jLj Liquid Drift Flux m/s
jLG Liquid Drift Flux Relative to Vapor m/s
jG Vapor Superficial Velocity m/s
L Characteristic Length m
M Molecular Weight g/mol
Ṁ Total Mass Flow Rate kg/s
ṀL Liquid Mass Flow Rate kg/s
ṀG Liquid Mass Flow Rate kg/s

155
156 NOMENCLATURE

Ṁref Refrigerant Mass Flow Rate kg/s

P Pressure bar
P Preheater Power W
Pi Liquid Interface m
PiD Non-dimensional Liquid Interface - [Pi /D] -
Plocal Saturation Pressure on the Local Heat Transfer Measure- bar
ment Position
PL Wetted Perimeter m
PLD Non-dimensional Wetted Perimeter - [PL /D] -
Psat Saturation Pressure bar
PG Dry Perimeter m
PGD Non-dimensional Dry Perimeter - [PV /D] -
q Heat Flux W/m2
qcrit Critical Heat Flux W/m2
qext External Heat Flux W/m2
Q̇L Liquid Volumetric Flow Rate m3 /s
Q̇G Vapor Volumetric Flow Rate m3 /s
R Internal Tube Radius m
Re Reynolds Number -
ReH Homogeneous Reynolds Number -
ReL Reynolds Number in Liquid Phase -
ReG Reynolds Number in Vapor Phase -
Reδ Reynolds Number in Liquid Film -
t Time s
T Temperature K
Tsat Saturation Temperature K
Twall Wall Temperature K
uL Mean Velocity of Liquid Phase m/s
uG Mean Velocity of Vapor Phase m/s
W eL Liquid Weber Number -
x Vapor Quality -
xde Dryout Completion Quality -
xdi Dryout Inception Quality -
xe Equilibrium Vapor Quality -
xIA Vapor Quality at Intermittent-Annular Flow Transition -
Xtt Martinelli Parameter -
NOMENCLATURE 157

Greek

Symbol Description Units

δ Liquid Film Thickness m

Cross-Sectional Void Fraction -
θdry Dry Angle rad
θstrat Stratified Angle rad
µL Dynamic Viscosity of Liquid Pa · s
µG Dynamic Viscosity of Liquid Pa · s
ρL Density of Liquid kg/m3
ρG Density of Vapor kg/m3
σ Surface Tension N/m
τ Shear stress N/m2
158 NOMENCLATURE
 Curriculum Vitae

PERSONAL DATA

Name: Jesús MORENO QUIBÉN

Date of Birth: 08.12.1970
Birthplace: Madrid - Spain
Nationality: Spanish

EDUCATION

Engineering Diploma in Aeronatics. Universidad Politécnica de Madrid, Spain.

DEA ”Dynamique des fluides et des transferts”, Université Pierre et Marie Curie, Paris VI, France.

CAREER

CEMAGREF, Génie des procédés frigorifiques (GPAN), Anthony, France

GROUPE DANONE, Centre International de Recherche Daniel Carasso (Danone Vitapole), Le
Plessis Robinson, France
EPFL, École Polytechnique Fédérale de Lausanne, Laboratoire de Transfert de Chaleur et de Masse
(LTCM), Lausanne, Switzerland.

159

View publication stats