UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Experimental investigation of boiling heat transfer to R-134a refrigerant Podut, Alexandru Ion 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2002-0209.pdf [ 5.97MB ]
Metadata
JSON: 831-1.0080993.json
JSON-LD: 831-1.0080993-ld.json
RDF/XML (Pretty): 831-1.0080993-rdf.xml
RDF/JSON: 831-1.0080993-rdf.json
Turtle: 831-1.0080993-turtle.txt
N-Triples: 831-1.0080993-rdf-ntriples.txt
Original Record: 831-1.0080993-source.json
Full Text
831-1.0080993-fulltext.txt
Citation
831-1.0080993.ris

Full Text

EXPERIMENTAL INVESTIGATION OF BOILING HEAT TRANSFER TO R-134a REFRIGERANT by ALEXANDRU ION PODUT B.Sc, Technical University of Cluj-Napoca, 1986 M.Sc, International Technological University, 1994 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF MECHANICAL ENGINEERING We accept this thesis as conforming to the^equjred standard THE UNIVERSITY OF BRITISH COLUMBIA November 2001 © Alexandru Ion Podut, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. The University of British Columbia Vancouver, Canada Department Date he^HA^ %OQ ( DE-6 (2/88) Abstract The aim of the present investigation was to improve the understanding of the boiling heat transfer to alternative refrigerants, specifically to tetrafluoroethane (R-134a). The refrigerants are mainly used in heat transfer applications, and due to environmental limitations, a need for assessing the suitability from a heat transfer efficiency perspective of new candidates has become a priority. The heat transfer coefficient that mainly controls the heat transfer phenomenon is very dependent on the operating conditions and on the geometry of the equipment and therefore an experimental assessments of proposed alternative refrigerants are required. In order to evaluate the performance of R-134a, a significant part of the project consisted in designing and building an experimental apparatus for the heat transfer study. A closed circulation loop with heat transfer and hydraulic equipment was constructed. The apparatus was equipped with state of art instrumentation, control and data acquisition equipment. Further, subcooled boiling experiments were pursued. These were carried out based on an experimental strategy that led to a significant amount of data, limited only by the maximum mass flow rate that could be achieved. Saturated boiling experiments that followed generated a limited data bank due to technical factors. For assessing the results, several theoretical methods for predicting the heat transfer coefficient were employed. Chen's (1966) correlation was used for both subcooled and saturated boiling data validation. A different method of extending Chen's correlation to subcooled boiling, when used for verifying experimental data was proposed. Liu and Winterton (1997) and Gungor and Winterton (1986 and 1987) correlations were also employed for validating the experimental data. This exercise confirmed that the measured vs. calculated heat transfer coefficients were within the range of accuracy claimed by correlations authors and/or acceptable from an engineering perspective. Discussions and conclusions for both subcooled and saturated boiling were part of the data analysis. u Table of Contents Abstract ii Table of Contents iii List of Tables • vii List of Figures viii Nomenclature >x Acknowledgments xi Dedication x i i Chapter 1: Introduction 1 1.1 Purpose of the study 1 1.2 Refrigeration industry 1 1.2.1 Development of refrigeration 2 1.2.2 Refrigeration process 2 1.2.3 Refrigerants 3 1.3 The Ozone 5 1.3.1 Ozone formation 6 1.3.2 Stratospheric ozone 6 1.3.3 Ozone depletion 7 1.3.4 The effects of ozone depletion 8 1.4 CFC regulations 11 1.4.1 Early actions to control the CFCs 11 1.4.2 1987 Montreal Protocol 12 1.4.3 Substances covered by the Montreal Protocol 13 1.4.4 1990 Update to the Montreal Protocol 13 1.4.5 1990 Amendment to the U.S. Clean Air Act 14 1.4.6 EPA regulations 14 iii Chapter 2: Literature Review of Forced Convective Boiling Correlations 15 2.1 The Chen correlation, 1966 15 2.2 The Shah correlation, 1976 and 1982 19 2.3 The Gungor and Winterton correlation, 1986 and 1987 24 2.4 The Liu and Winterton correlation, 1991 27 2.5 Darabi et al correlations review 29 Chapter 3: Review of Experimental Studies for R-134a 33 3.1 M. Pate and S. Eckels, 1991 33 3.2 K. Torikoshi and T. Ebisu, 1993 36 3.3 H. Takamatsu, S. Momoki and T. Fujii, 1993 37 3.4 J.P. Wattelet and J.C. Chato, 1993 39 3.5 J.Y. Shin, M.S. Kin and S.T. Ro, 1996 40 3.6 X. Liu, 1997 41 3.7 N. Kattan, J.R. Thome and D. Favrat, 1998 42 Chapter 4: Experimental Apparatus 46 4.1 General description 46 4.2 Circulating pump 47 4.3 Flow meters 51 4.4 Preheater 53 4.5 Test section 53 4.6 Separator, condenser and subcooler 53 4.7 Power supply ..55 4.8 Accumulator / pressurizer 57 4.9 Thermal insulation 57 4.10 Optical compensation box 58 4.11 Safety equipment and procedures 58 iv Chapter 5: Test Section 59 5.1 Inner tube 59 5.2 Heat flux 61 5.3 Electrical resistance 61 5.4 Hydrodynamic entry length 62 5.5 Outer tube 63 5.6 End cap fittings 63 Chapter 6: Instrumentation and Control 65 6.1 Flow measurement and control 65 6.2 Power measurement and control 66 6.3 Pressure measurement and control 66 6.4 Temperature measurements 68 6.5 Data acquisition system 70 Chapter 7: R-134a Subcooled Boiling 71 7.1 Experimental procedure 71 7.2 Data validation 72 7.3 Data reduction 76 7.4 Data processing 78 7.4.1 Chen's correlation applied to subcooled boiling 79 7.4.2 Liu & Winterton correlation applied to subcooled boiling 82 7.5 Experimental results for subcooled boiling experiments 83 7.6 Discussion and conclusions 89 v Chapter 8: R-134a Saturated Boiling 93 8.1 Experimental procedure and data 93 8.2 Experimental results 94 8.3 Analysis of saturated data against theoretical correlations 97 8.3.1 Chen's correlation applied to saturated boiling 98 8.3.2 Gungor (1986) correlation applied to saturated boiling 98 8.3.3 Gungor (1987) correlation applied to saturated boiling 100 8.4 Results 100 8.5 Discussion 102 Chapter 9: Conclusions and Future work 104 9.1 Conclusions of the present investigation 104 9.1.1 Experimental set-up 104 9.1.2 Experimental data base 105 9.1.3 Data analysis 105 9.2 Future work 107 References 108 Appendix No. 1: Experimental apparatus components 112 Appendix No. 2: Calculation of electrical resistance of the test section 118 Appendix No. 3: Instruments calibration data sheets 119 Appendix No. 4: Tabulated subcooled boiling experimental data 123 Appendix No. 5: Tabulated saturated boiling experimental data 136 Appendix No. 6: KLEA 134a physical property data fde 138 vi List of Tables Table 1.1: Environmental properties of refrigerants 4 Table 2.1: Mean deviation [%] of theoretical correlations vs. experimental data 32 Table 3.1: Comparison of experimental set-ups and conditions for R-l 34a studies..44 Table 5.1: Electrical resistance of test section components 62 Table 6.1: DAS-1602 data acquisition features 70 Table 7.1: Accuracy of tested theoretical correlations 89 Table 8.1: Experimental conditions for R-l34a saturated boiling 93 Table A4: Tabulated subcooled boiling experimental data 123 Table A5: Tabulated saturated boiling experimental data 136 vii List of Figures Figure 1.1: Elementary mechanical refrigeration system 3 Figure 1.2: Previous industry applications of chlorofluorocarbons 5 Figure 2.1: Reynolds number factor, F, and Suppression factor, S 17 Figure 2.2: The CHART Correlation (Shah 1976) 23 F i gure 4.1: Experimental apparatus 48 Figure 4.2: Aurora F05 pump head curve 49 Figure 4.3: Diagram of VFD connection to the pump 50 Figure 4.4: Turbine flow-meter construction 52 Figure 4.5: Capacity characteristics and diagram of pressure relief valve 55 Figure 4.6: Schematic diagram of system power supply 56 Figure 5.1: Test section schematic 60 Figure 6.1: Electric voltage transducer calibration set-up 68 Figure 6.2: Data acquisition system hardware configuration 70 Figure 7.1: Saturation pressure-temperature readings with system at rest (no flow)...74 Figure 7.2: Data validation: measured vs. predicted temperature differences 75 Figure 7.3: Distribution of the thermocouples along the test section 77 Figure 7.4: Typical experimental results for subcooled boiling experiments 85 Figure 7.5: Subcooled boiling selected experimental results 86 Figure 7.6: Accuracy of tested theoretical correlations for subcooled boiling 90 Figure 8.1: Measured heat transfer coefficient for 1.0 GPM vs. predicted quality 94 Figure 8.2: Flow boiling data of Wattelet et al [39] at 5°C for R-l34a 95 Figure 8.3: J.Y. Shin et al [32] heat transfer coefficients for pure refrigerants 96 Figure 8.4: Saturated boiling experimental data for 1.0, 2.0 and 3.0 GPM at thermocouple location #1 (discharge from the test section) 101 Figure 8.5: Accuracy of the tested correlations for saturated boiling 103 V l l l Nomenclature Bo = Boiling number=q/(Ghfg) Co = Convection number = (1 /x-1)° 8(pg/pi)°5 Cp = Specific heat, (J/kgK) d = Diameter, m F = Forced convection (Reynolds Number) factor 2 2 F r = Froude number =G /( pi gd) g = Acceleration due to gravity, (m/s ) G = Mass flux, (kg/m2s) h = Heat transfer coefficient, (W/m K) hfg - Latent heat of vaporization, (J/kg) k = Thermal conductivity, (W/mK) M - Molecular weight Pr = Prandtl number = u.|Cpi/k| Pr - Reduced pressure Q = Heat flux, (W/m2) q = Specific heat flux, (W/m2s) Re = Reynolds number = G(l -x)d/u,| S = Suppression factor T = Temperature, (K) A: = Quality of the mixture Xtt - Martinelli's coefficient z = Weight fraction liquid = 1 -x ix Greek Letters: a = Coefficient of thermal expansion Y = Dimensionless parameter (Shah) = hTp/h| <j> = Heat flux, (W/m2) // = Viscosity, Ns/m2 a = Surface tension, (N/m) p = Density, (kg/m3) Subscripts: b = bulk bs = bubble suppression cb = convective boiling FC = force convective I = liquid phase mac = macroconvective mic = microconvective NB = nucleate boiling nb = nucleate boiling pool = pool (boiling) sat = saturation SC = subcooled tp = two-phase w = wall x Acknowledgments During my M.A.Sc. Program I achieved new and valuable skills, especially thanks to the practical aspect of the project.. "/ hear, and I forget. I see, and I remember. I do, and I understand. " Confucius (551-479 BC) I express my sincere appreciation to Dr. Dan Fraser for his critical evaluation, insightful suggestions, invaluable guidance, encouragement and support during the entire course on the M.A.Sc. Project. It has been a great and already rewarding experience. I believe that the skills that I developed during this program made me not only a better engineer but also a better person. Special thanks to Dr. Hill, Dr. McAdam and Dr. Green of Mechanical Engineering Department, to Dr. Atwater of Civil / Environmental Engineering Department and to Dr. Sheraton of the Chemical Engineering Department, who through their courses have enriched my knowledge. I am grateful to Dr. Steve Rogak, who offered me the opportunity to work at the SCWO project. This helped me gain valuable experience and fair appreciation, which in turn opened the door to a new career opportunity. I would like to thank to all the staff employed by the Mechanical Engineering Office and Machine Shop, who assisted me during my program at UBC. To all the friends who made my days enjoyable, Vladan, Majid, Danijela, Tazim, Anne-Marie, Sanja, Al, Petro, Silviu and many others, thank you. xi D e d i c a t e d to my m o t h e r D r . A u r e I i a G a h r i e l a P o d u t and to my w i f e E l i s a b e t a M a g y a r P o d u t xn Chapter 1 Introduction 1.1 Purpose of the study The primary objective of the present study was to design and build an experimental facility for testing the heat transfer characteristics of environmentally acceptable refrigerants. This objective was undertaken as part of the broader goal of improving the general knowledge of boiling heat transfer to alternative refrigerants. A second objective was to experimentally determine the local boiling heat transfer coefficient of the chosen refrigerant, for both subcooled and saturated boiling regimes encountered in practice. An important part of the experimental program was to generate a heat transfer database, to be used as reference for future studies. Finally, in the course of study, theoretical models for predicting the heat transfer coefficient were tested against the collected data and their suitability and accuracy assessed. Recommendations for the use of the correlations were made, based on this assessment. 1.2 Refrigeration industry The term of refrigeration is used to describe a thermal system that maintains a process space or material at a temperature less than available from the ambient conditions [21]. During refrigeration, the heat is transferred from the space or materials to be cooled into a lower temperature substance referred as refrigerant. The refrigerant is cooled by physical processes, which transfer the heat to the ambient environment. 1 1.2.1 Development of refrigeration The mechanical refrigeration process first appeared about 1910, when J.M. Larsen first produced a manually operated household machine [2]. In 1918 Kelvinator produced the first automatic refrigerator for the American market. The first of the sealed or "hermetic" automatic refrigeration units was introduced in 1928 by General Electric. Beginning with 1920, domestic refrigeration became one of the North-American important industries. Fast freezing to preserve the food for extended periods was developed about 1923. Mechanical refrigeration systems were first connected to heating plants to provide summer cooling in the late 1920's. From a slow start in 1930's, air conditioning of automobiles has grown rapidly. In 1935, Frederick McKinley Jones produced an automatic refrigeration system for long-haul trucks [2]. 1.2.2 Refrigeration process Most of the refrigeration systems operate in a continuous cycle, using a recirculating fluid that boils at low temperatures to remove the heat, Figure 1.1. The vapor is compressed to a pressure such that the condensing temperature is elevated. The vapor is cooled and condensed by a heat exchanger that rejects the heat into the environment. The condensed liquid is passed through a restricting valve or orifice, where it changes into a mixture of liquid and vapors which is colder than the refrigerated space. This expansion at constant enthalpy is known as the Joule-Thompson effect [21]. The equipment used in refrigeration processes is referred to as follows: high-pressure vapor heat exchangers are called condensers, low-pressure heat exchangers are called evaporators, throttling devices are called expansion valves or capillaries, and the pressure-raising devices are called compressors. 2 Figure 1.1 Elementary mechanical refrigeration system CABINET 1.2.3 Refrigerants The heat transport fluids, which convey the heat energy from the low-temperature enclosure to the high-temperature environment, are called refrigerants. Refrigerants are designated by an identifying number, used in conjunction with the trade name. For example, Tetrafluoroethane ( C F 3 C H 2 F ) identified with the refrigerant number 134a can also be referred to as R-134a, HFC-134a, or SUVA-134a . The number designation and the safety classification are covered by A N S I / A S H R A E Standard 34-1992. 3 Recent evidence indicates that the refrigerants are posing an environmental threat, especially the decomposition of chlorofluorocarbons (CFC) which depletes the atmospheric ozone layer. This effect wi l l be detailed in a subsequent paragraph. Several indicators are used to define the relative environmental destructiveness of refrigerants. The ozone depletion potential (ODP) is the ozone-destroying capacity of the refrigerant relative to Refrigerant 11 (CFC-11). The global warming potential ( G W P ) is a relative measure of the ability of a substance to cause an increase in the temperature of the atmosphere by absorbing solar and earth radiation, relative to the effect of Refrigerant 11. Table 1.1 shows the environmental properties of the most common refrigerants [21 ]. Table 1.1 Environmental properties of refrigerants Refrigerant and number R-11/(CFC-11) R-12ACFC-12) R-502 R-l 14 R-402B (SUVA HP81) HCFC-123 HFC-134a HFC-125 HCFC-124 R-22/(HCFC-22) HFC-23 HFC-32 Ozone depletion potential Global warming potential Toxicity, ppm (v/v) AEL,* TLVf Flash point, lower/upper vol. % 1.0 1.0 1000 TLV -1.0 2.8 1000 AEL — 0.23 3.75 1000 TLV — 0.7 3.9 1000 TLV — 0.03 0.52 1000 AEL None 0.02 0.02 30 AEL None 0.0 0.26 1000 AEL None 0.0 0.84 1000 AEL None 0.02 0.1 500 AEL None 0.05 0.3 1000 TLV None 0.0 6.0 1000 AEL None 0.0 0.11 1000 TLV 14/31 * AEL is the recommended time-weighted average concentration of an airborne chemical to which nearly all workers may be exposed during an 8-h and/or 12-h day, 40-h week without adverse effect. Determined by the du Pont Company for compounds that do not have a TLV. . . . t TLV established for industrial chemicals by the American Conference of Governmental Industrial Hygiemsls. is the recommended time-weighted average concentration of an airborne chemical to which nearly all workers may be exposed during an 8-h day, 40-h week without adverse effect. S O U R C E : Data from E. I. du Pont de Nemours & Co., Inc. Bulletin AG-1. The chlorofluorocarbons were widely used in the refrigeration industry due to their high-suitability, as it was believed. Science writer Lemonick, Michael D. sums up the characteristics of C F C s [24]: 4 When (hey were first synthesized in the late 1920s, fCFCsJ seemed too good to he true. These remarkable chemicals...are non-toxic and inert, meaning that they do not combine easily with other substances. ...CFCs are perfect as coolants in refrigerators and propellanl gases for spray cans. Since CFCs are good insulators, they are standard ingredients in plastic-foam materials like Slyrofoam. Best of all, the most commonly used CFCs are simple, and therefore cheap, to manufacture. Next figure shows the various applications of CFCs [24], before the environmental threat became evident to the scientific community and the general public. Figure 1.2 Previous industry applications for chlorofluorocarbons USES OF CHLOROFLUOROCARBONS* 2% 2% 2 0 0 / 0 Refrigerants in automobile air-conditioning systems J Z % 17% Rigid toam insulation in homes and offices 12% Solvents for cleaning electronic components ^3% 10% Blown foam insulating material in refrigerators, storage tanks, etc 9% Refrigerants in commercial refrigeration systems 8% Solvents for degreasing metal parts 4% Aerosol sprays 4% Refrigerants in large commercial air-conditioning systems 4% Miscellaneous uses 3 % Disinfectants or sterilants in hospitals 3 % Cushioning foams in automobiles, airplane seats, furniture 2% Refrigerants in home refrigerators ^ ^ ^ ^ ^ 2% Polystyrene foam plastics, such as fast-food I'WZ*^^^^^ containers 2% Food freezants *700 million pounds of C F C s produced annually by U.S. manufacturers 1.3 The Ozone The credit for the discovery of ozone is given to Friedrich Schonbein, a German professor of chemistry at the University of Basel, Switzerland. Performing laboratory experiments in order to find the substances in the earth's atmosphere, once he passed an 5 electric charge through a beaker of water. Every time he repeated the experiment he noticed a strange smell. His first assumption was that the smell came from the electrically charged oxygen. Finally, he realized that he found a new substance, with three atoms of oxygen in the molecule. He named it ozone, from the Greek word ozein, "to smell" [24]. 1.3.1 Ozone formation There are two general types of ozone. One exists near the earth's surface and is called low-level ozone or tropospheric ozone. Usually ozone is produced when sunlight reacts with chemicals in the air. Substances from incomplete combustion, emissions from coal, gas or oil-burning utility plants, all react with each other and with the sunlight creating ozone in this process. Fire, whether it occurs naturally or is set by human beings is also a source for the low-level ozone. Other sources of tropospheric ozone are completely natural. Storm lighting, methane given off by decaying plants and animal tissue or gaseous wastes of grazing animals is another source of low-level ozone. In 1987, William Chameides, Ronald Lindsay and Jennifer Richardson from the Georgia Institute of Technology were studying the ozone concentrations around Atlanta, Georgia. Based on their measurements, they found that the ozone levels were to high to be attributed to vehicles and factories alone. They found that the trees released hydrocarbons during photosynthesis. When the hydrocarbons reacted with the sunlight and pollutants, they created ozone [24]. 1.3.2 Stratospheric ozone The stratosphere is a part of earth's upper atmosphere, extending from approximately 8 to 30 miles above the earth's surface. The ozone found in the stratosphere is formed continuously as the result of reaction between oxygen molecules O 2 and solar radiation. 6 This reaction breaks the oxygen molecules into two single atoms, which being very reactive quickly combine to form new oxygen O2 molecule. However, in the presence of solar radiation, the single atoms will occasionally combine with an oxygen molecule to form ozone O 3 . Ozone is constantly broken apart by solar radiation that it absorbs. This reaction does not result in a net loss of ozone, because the process generates single oxygen atoms, which will combine again to form more ozone. 1.3.3 Ozone depletion The atmosphere is filled with several chemicals that have high potential of destroying the ozone. One of the most reactive ones is chlorine, which is a component of both CFCs and HCFCs and as well is naturally emitted by oceans. In 1974, the first connection between halocarbons (which include CFCs and HCFCs) was established. Long before the ozone hole was discovered, two scientists began to have second thoughts about CFCs. The two chemists, F. Sherwood Rowland and Mario J. Molina, from the University of California, Irvine, wondered if CFCs were as stable high up in the atmosphere as they were at low altitudes. In 1974, they published a scientific paper in which they outline their concerns about CFCs [15]. In their paper they have explained how they thought that halocarbons could damage the ozone layer. CFCs are liquids that evaporate very quickly, changing from a liquid to a lighter-than air gas. When released into the environment, they evaporate and get into the atmosphere. Since they are so stable, the scientists reasoned that they wouldn't combine with other molecules in the air, but gradually rise through the troposphere until they reached the stratosphere. Once they reach the stratosphere, these molecules are impacted by the solar radiation that forms ozone. The chemical bonds that hold the CFC molecules together are broken, releasing single chlorine ions, Cl". When this free and highly reactive ion encounters a relatively unstable ozone molecule, it breaks the ozone molecule forming molecular 7 oxygen O2 and chlorine monoxide CIO. The chlorine monoxide is itself unstable and will readily break down when it comes in contact with an other ozone molecule. Researchers believe that it takes CFC molecules about 7 to 10 years to reach the upper stratosphere. With half-life of 50 to 100 years, this means that CFCs that were released over the last decade are only now beginning to affect the ozone layer. This would also mean that control measures that were put in place within the last years would not have a measurable impact on stratospheric ozone levels until some time in the future [23]. It has been also estimated that a chlorine molecule can destroy approximately 1,000,000 ozone molecules [19]. 1.3.4 The effects of ozone depletion Scientists have studied in detail the adverse effects of ozone layer thinning. One of the primary motivation behind policy addressing the ozone-depletion issue is the fact that a decrease in column ozone allows greater amounts of biological damaging radiation to reach the surface of the planet. Not all radiation is equally damaging to the living systems. For instance, exposure to UV radiation is needed for the formation of Vitamin D in humans. UV-A radiation (wavelengths above 320 nm) may cause some biological damage, but also stimulates repair mechanisms. Some of the damaging effects are presented as follows. (a) DNA damage Extended exposure to UV-B radiation (290 - 320 nm) can cause damage to DNA. Exposure to substances that damage DNA increases the probability that some DNA will not be repaired and that cells will reproduce the altered form of DNA. Such exposure is classified as mutagenic, because it mutates genes. Production of cells injured in this way can lead to cancer or other diseases [22]. 8 (b) Effects on human immune system The effect of UV radiation on the immune system is considered one of the most important aspects of ozone depletion. Apart from the effect of increased susceptibility to disease on individuals, severe ozone loss may cause widespread immune-system impairment and thereby increase risk of epidemics [22]. (c) UV-B and skin cancer There is scientific consensus that UV radiation has a role in causing two of the three types of skin cancer: squamous-cell type and basal cell cancer. The third and most frequently deadly type, malignant melanoma, is linked to solar radiation mainly by epidemiological evidence [22]. (d) The threat to plant life Plants as well can suffer from increased doses of ultraviolet radiation. According to Robert C. Worrest, researcher of the EPA, the scientists have tested about two hundred species of land plants, mostly food crops, for sensitivity to ultraviolet light. They have concluded that the most sensitive plants are the ones related to peas and beans, melons, mustard and cabbage. The effect of UV radiation is noticed in the reduction of the leafs and steam growth and lower total dry weight. In addition to other factors, increased levels of UV radiation may reduce the crop yield [22]. (e) UV and aquatic life Ozone depletion can pose a threat to marine life. Zooplankton and phytoplankton are highly sensitive to UV-B radiation. Much of the ocean's life is protected from the UV radiation, which penetrates only the upper 10 m of water. However, zooplankton and phytoplankton spend either all or critical stages of their life cycles in sunlight. Since they play crucial roles in complex ecological food webs, damage to these organisms may have important ramifications for all oceanic life [22]. 9 (f) The Greenhouse Effect The nature of ozone seams to be very contradictory. On one hand there is low-level ozone, which is classified as a pollutant, and then there is the upper-level ozone that is essential for the existence of life. The amounts of both kinds are changing. The stratospheric ozone levels are increasing while the tropospheric levels of ozone are decreasing. The low-level ozone, along with other gases like CFCs, C O 2 and water vapors are blamed for the greenhouse effect. We should note that there is a natural greenhouse effect that is vital for the life on earth. If the absorption of solar radiation were the only process of heating the earth, the temperature was below freezing, about -18°C, if the planet reflectivity remained as at present. The much higher temperature actually sustained, about +15°C on average, arises because certain gases in the atmosphere absorb and re-emit a substantial fraction of the infrared radiation which the surface emits in response to solar heating and which in absence of these gases would escape directly to space. The downward component of this re-emitted radiation warms the surface and the lower levels of atmosphere [37]. As the concentration of these gases, known also as greenhouse gases, increases, more radiation is re-emitted back to the earth surface, leading (theoretically) to an increase in temperature. At present, no one is absolutely certain how dangerous global warming might be. Nevertheless, some scientists have tried to predict what global warming could mean to the earth and its inhabitants. They say that climate changes that would accompany global warming would transform life, as we know it today. Humanity has survived the worldwide temperature drop of 5°C during the Ice Ages, but some predictions show an increase of few degrees within 100 years. The impact on life on earth might be considerable. 10 1.4 CFC regulations Acknowledging the effect of CFCs as ozone depleting substances, the chemical companies started looking for alternatives in the late 70's. The progress made was quite consistent, when, in the early 1980's the threat of more government regulations began to fade. At that time the search for refrigerant substitutes technically stopped. The world use of CFCs continued to grow [19]. In 1985, scientists discovered a loss of ozone as large as North America over part of the Southern Hemisphere. Later data has revealed losses that exceed 50% in the total column and greater than 95% at an altitude of 9 to 12 miles above the earth. This newly discovered hole in the ozone layer gave importance to international efforts to protect the ozone layer. 1.4.1 Early actions to control the CFCs One of the first restrictions placed on CFCs by EPA (Environmental Protection Agency) and FDA (Food and Drug Administration) was implemented in 1978. Then restrictions on the use of CFCs for aerosol propellant applications, such as spray paint, household cleaners, etc., as it was determined that approximately 50% of the CFCs produced were used for that purpose. The EPA ruled for further restrictions on the production of CFCs in the 1980s. The EPA made two proposals to reduce the use of CFCs other than aerosols: Mandatory regulations and Economic Incentives from storing and recycling. However, at that time the scientific community was losing the interest in the ozone depletion phenomenon. Because of that, EPA did not pursue these regulations at that time. Public attention focused again on the two proposed regulations when the "greenhouse effect" and the new evidence that CFCs are in fact depleting the ozone layer captured the 11 media in 1986. It has been determined that CFCs were causing a "hole "in the ozone layer over Antarctica. The major concern throughout the world led to what is now known as the Montreal Protocol on Substances that Deplete the Ozone Layer. 1.4.2 1987 Montreal Protocol In September 1987, more than two dozen nations gathered and signed the Montreal Protocol on Substances that Deplete the Ozone Layer. The decisions were put in place in January 1989. At that time 68 nations have adhered to the Protocols decisions. Surprisingly, after the Protocol was signed, the scientific observations lead to the conclusion that the loss of ozone is on a worldwide scale and not limited only to remote areas over the Antarctica. The measured data were significant higher than the computer model predictions. That lead to the doubt around the adequacy of control measures set in the original form of the Protocol. There were four areas that further polarized the attention of the signing parties and national legislatures. These are: • accelerating the CFC and methyl chloroform phaseout schedule • controlling and finally eliminating the production and use of hydrochlorofluorocarbons (HCFC) • eliminating the emissions of ozone-destroying compounds • implementing effective trade sanctions These areas are covered by the Clean Air Amendments of 1990. As part of this act, title VI, "Stratospheric Ozone Protection" represents one more link in the worldwide effort to save the critical part of the atmosphere that protects the earth from the harmful ultraviolet rays from the sun. 12 1.4.3 Substances covered by the Montreal Protocol The chemical substances discussed at the meeting were separated into two groups. The first one contains the fully halogenated chlorofluorocarbons, the second one consists of the halons. This is the list of chemicals controlled by the original Montreal Protocol: Group 1: Fully Halogenated Chlorofluorocarbons CFC - 11 CFC - 12 CFC-113 CFC - 114 CFC - 115 Group 2 - Halons (Fire extinguisher agents) Halon- 1211 Halon- 1301 Halon - 240 The Montreal Protocol, in the original form, called for the manufacturing of Group 1 CFCs to be frozen at their 1986 consumption quantities. Additional restrictions called for a 20% reduction from the 1986 consumption level by July 1, 1993, with further reductions of 50% by July 1, 1998. The manufacture of halons was to be frozen at 1986 consumption levels by 1992 or three years after the law was placed into force. 1.4.4 1990 Update to the Montreal Protocol It was on June 29, 1990, when the signatory parties at the Montreal Protocol met in London to work on making the original protocol more stringent. The 56 participating countries called for the phase-out of HCFCs in addition to CFCs decided at the previous 13 meeting. The decision was to completely phase-out HCFCs by year 2040, or if possible by July 2000. At the same meeting there has been decided that halons will be reduced by half by the year 1995 and completely phased-out by the year 2000. 1.4.5 1990 Amendment to the U.S. Clean Air Act On November 1990, U.S. President signed into law the clean Air Act Amendment. Title VI, Stratospheric Ozone Protection, established a detailed method for phasing out ozone-depleting substances. The EPA was to develop regulations to implement these amendments to the Clean Air Act. Title VII is the regulatory portion of the act. The amended Clean Air Act does not require individual states to develop regulations to protect the stratospheric ozone layer. Federal regulations will override state or local regulations unless the state proves to the EPA that its regulations are at least as stringent as the federal regulations and that the state is enforcing them. 1.4.6 EPA regulations The intentional venting of CFCs and HCFCs after July 1, 1992 carries a fine of up to $25,000 per day per offense. Each kilogram of refrigerant is considered as an offense. There is also an Awards Provision that allows anyone to report an offender. If the offender is convicted, the person who reported the violation receives up to $10,000 [19]. 14 Chapter 2 Literature review of forced convective boiling correlations On the topic of forced convective boiling, in particular to refrigerants, there is an abundance of published information, which makes the task of surveying the literature appear as rather challenging. General reviews of correlations and experiments focused on refrigerants thermal behavior have been made by many knowledgeable authors, and is not the intention of this review to repeat them. However, there were considered as more relevant for the present study the publications that treated predicting models for the flow boiling heat transfer and the reports on experimental work that used R-134a as the heat transfer fluid. The heat transfer through boiling and in particular through forced convective boiling is of particular importance for the nuclear and power generation industry, for the refrigeration and air-conditioning applications and for the chemical processes, just to mention a few. In order to optimize the efficiency of any modern heat transfer equipment, the designers need to have reliable and as accurate as possible methods to predict the boiling heat transfer coefficient. This led to a strong interest from the scientific community to develop correlations that are valid over all of the forced convective heat transfer regimes and applicable to a wider range of fluids. 2.1 The Chen correlation, 1966 One of the most established and quoted correlation for saturated flow boiling is the one developed by Chen in 1966. He started with the assumption that saturated boiling can be described as an addition of macroconvective mechanism corresponding to heat transferred to the flowing fluid and a microconvective mechanism which takes into account the bubbles nucleation and growth [4]. The proposed expression for calculating the two-phase heat transfer coefficient was: 15 h = h + h ntp nmac T nmic (2.1) where, h,p is the two-phase boiling heat transfer coefficient, hmac and hmic are the respectively the macroconvective and the microconvective contributions. Equation (2.1) is often found in the literature also expressed as: hB = hFC + hm or hlp = h,F + hpoo,S (2.2) where the indexes FC and NB stand for forced convective and nucleate boiling. Further, he proposed the macroconvective term to be evaluated with the well known Dittus-Boelter equation corrected by an enhancement factor F, a flow parameter taking into account the enhancement of heat transfer due to an increase in vapour quality. The parameter F, also known as Reynolds number factor, was determined empirically from the experimental data and plotted as a function of reciprocal Martinelli's parameter. Its value is always greater than unity, showing that the flow velocities are always greater in two phases than in single phase for a given specific mass flux. The microconvective term represents the pool boiling contribution to the two-phase heat transfer coefficient. It can be estimated from the Foster and Zuber pool boiling correlation corrected with a suppression factor S. The factor S quantifies the suppression in nucleate boiling, i.e. the reduction in the thermal boundary layer thickness, due to an increased contribution of the forced convection effect. It's value approaches unity at low values of the two-phase Reynolds number and decreases asymptotically to zero at high values of the two-phase Reynolds number. Both parameters F and S were originally given in a graphical form, Figure 2.1. 16 Figure 2.1: Reynolds number factor, F, and Suppression factor, S, [4] I O ' L _ | I I 1 1 1 1 1 1 00 o • A P P R O X I M A T E REGION OF D A T A ,D^ I ' i t • / m l t i i t t i i i l I IO-I I IO 10* I I  i/xtt=(xyz)0-9(Pl/pv)a5(Ltv/Li,)a 1.0 .9 Os as B o P 7 < ,6 > * .4 H < 3 II 2 .1 0 T— I 1 I I 1 1 1 | T t i l l APPROXIMATE-REGION OF DATA ' T i l l 10* Re = Re, F 17 Collier in 1981 suggested a set of equations to fit Chen's empirically determined F and S original curves [5]. He also suggested how to apply the equation for subcooled boiling heat transfer. In this case a reasonable assumption for calculating the heat flux at the wall, (j), is proposed in the following form: <f> = h,c{Tw-Th)+hm(Tw-Tsal) (2.3) where hpc is the forced convective or macroconvective term, hm is the nucleate boiling or microconvective term, Tw is the wall temperature, TB is the bulk temperature and Tsat is the saturation temperature. The factor F is taken equal to unity and the factor S is found from the graphical form or calculated with the proposed correlations. More computational details are presented in Chapters 7 and 8. The correlation was tested against over 600 data points from ten experimental cases, for vertical flow through tubes and annuli of water and organic fluids, such as methanol, cyclohexane, pentane, heptane and benzene. Other experimental conditions worth mentioning are: absolute pressures 0.5-34.8 bar, flow velocities (liquid) 0.06-4.5 m/s, mixture qualities 1-71% and specific heat fluxes 6.3-2394 kJ/m2s. For these data points the average deviation between calculated and measured boiling heat transfer coefficients was ±12%. 18 2.2 The Shah correlation, 1976,1982 The interest in the design of boilers and evaporators and the lack of a general correlation (Chen's correlation is limited to vertical flows) were determining factors in Shah's efforts to develop a new method for heat transfer prediction. In 1976 he successfully presented a chart correlation constructed based on about 800 data points from 18 independent experimental studies. The data included the most used refrigerants in their entire range of application and boiling water between 15 to 2500 psia pressure, practically covering all the boiler operating conditions. Horizontal and vertical flow orientations, upward and downward flow directions, different pipe materials and wide ranges of heat and mass fluxes were covered. At this stage the subcooled boiling and the stable film boiling were not considered, the chart being applicable for saturated boiling at subcritical heat fluxes. Shah [29] suggested that the flow-boiling regime be divided into three regions on the chart: a convective boiling with partly dry surface regime, a bubble suppression regime and a nucleate boiling regime. The correlation employs four dimensionless parameters defined as follows: (2.4) (2.5) (2.6) (2.7) Here Bo represents the Boiling number, Fri is the Froude number assuming all the mass flowing in liquid form and Co is the Convection number, the only new parameter introduced by Shah. In the first equation (2.4) hi is calculated with the Dittus-Boelter correlation. V = hyhi=f(Co,Bo,Fri) where 0.5 V / H I ) Bo= % Gh •Jg / Pi gd 19 The Shah correlation was presented in a graphical form (Figure 2.2) in terms of the non-dimensional parameters and was evaluated by taking the larger of the three heat transfer coefficients corresponding to nucleate boiling, suppression and convective boiling. Later, Shah developed a new chart correlation for subcooled boiling heat transfer in pipes and annuli [31], which successfully completes the original correlation. It is expressed through separate equations for the low subcooling region, ¥ = % (2.8) and for high subcooling region: = x¥0 + ^ sc/^j (2.9) where ATsc=Tsa, - Th and ATsal=Tw - Tsa,. In equation (2.9) *F = y^T h (2.10) To is the value of T at zero vapour quality and hi is the heat transfer coefficient for all fluid mass flowing as liquid only, calculated with Dittus-Boelter or Sieder-Tate equations. To is a function of Boiling number as given by: Bo>0.3(10"4) T0=230Bo05 (2.11) Bo<0.3(10"4) vF0=l+46Bo05 (2.12) Equations (2.11) and (2.12) correspond to saturated boiling at zero quality and were developed based on more than 11 different liquid-surface combinations and a wide range of pressures, heat and mass fluxes. Furthermore, the two-phase heat transfer coefficient h,p is defined as: 20 h^qAT,-T,ry(ATx-ATj <2-13> For annular cross section the hydraulic diameter (four times the ratio of cross section area and the wetted perimeter) provided excellent agreement with all the single phase and subcooled boiling data used in developing the correlation, except with a case where the annulus clearance was 2.2 mm. In this last case where only the inner pipe was heated the wetted perimeter was replaced by the heated perimeter. The subcooled boiling correlation was developed based on about 500 data points from 18 independent experimental studies, covering wide ranges of parameters. The data were for water, a few common refrigerants and other organic fluids and alcohols, for horizontal and vertical pipes and annuli. The accuracy was found to be ±30%. At this moment, for most encountered practical situations the method of reading the CHART for was analogous to the method for reading the Moody chart for friction factors in pipe flows. Later, Shah suggested a set of analytical equations to fit his original CHART curves [29]. He defines a parameter TV as follows: N = Co (2.14) For vertical tubes at all values of Fri and for horizontal tubes with Fr/ >0.04, and N = 0.38 Frf0J Co (2.15) for horizontal tubes with Fri <0.04. The equations were again developed for the three distinct regions of the boiling regime as similar with the CHART: 21 For N > 1.0 (nucleate boiling regime), T n b = 230Bon5 for Bo > 0.3 (JO'4) T n b = 1 + 46Bons for Bo < 0.3 (1CT4) (2.16) T c b = 1.8/N08 and T = the larger of^ and T c b (2.17) For 0.1 < N < 1.0 (right section on the CHART of bubble suppression regime) T b s = F Bon5 exp (2.74 ATft 10) (2.18) For N < 0.1 (left section on the CHART of bubble suppression regime) T b s = FBo05 exp (2.74 N~a,5) (2.19) and T = the larger o/"%s and 4^ (2.20) The constant F in the above equations was defined as follows: F = 14.7 for Bo >11 (W4) and F = 15.4 for Bo < 11 (W4) (2.21) For annular cross section Shah recommends an hydraulic diameter according to the size gap as follows: for gaps greater than 4 mm - four times flow area divided by wetted perimeter, and for gaps smaller than 4 mm - four times flow area divided by the heated perimeter. This new correlation was verified with about 3,000 data points for 12 different fluids and a wide range of experimental conditions. The predictions were found to be in good agreement with the experimental data. 22 23 2.3 The Gungor and Winterton correlation, 1986 and 1987 The quest for a general correlation for boiling heat transfer was also pursued by Gungor and Winterton [12]. Their objective was to develop a new correlation, using simple equations, tested against a large data bank, extendible to subcooled as well as saturated boiling and applicable to tubes and annuli for both vertical and horizontal flows. They have collected data from a wide range of sources covering a broad range of experimental conditions. Their data points were measured values of the heat transfer coefficients and wall temperatures as function of pressure or saturation temperature, mass flux, heat flux and quality. For subcooled boiling the bulk temperature was recorded in place of quality. The data consisted of 4300 data points for water, refrigerants and ethylene glycol, covering seven fluids and 28 authors. Their correlation is a modification of Chen correlation, written as: hlp=Ehl+Shpml (2.22) where h/ is given by the Dittus-Boelter equation for liquid only flowing in the duct. For the hpooi term, after investigating several expressions, the Cooper correlation was selected. The forced convection heat transfer enhancement term F is replaced by a factor E, dependent on quality and vapour to liquid density ratio and expressed, as the common practice, by the Martinelli parameter X t t. To incorporate the effect of disturbance of the boundary layer next to the heat transfer surface by the vapour generation in the boding process, the boiling number was also considered for the final correlation. Using all the available data, Gungor and Winterton defined the E and S factors, as detailed in Chapter 8, for the saturated boiling heat transfer calculations. 24 For horizontal tubes and Froude numbers less than 0.05, the authors recommended to multiply the factor E by Fr(OI~2Fr) and S by Fr05. All the properties used in correlation are estimated at saturation temperature. For boiling in annuli with only one of the walls heated the hydraulic diameter is used. Its values is given by the classical definition of "four times flow area, divided by the wetted perimeter" for annulus gaps greater than 4 mm, whereas for gaps less than 4 mm the wetted perimeter is replaced by the heated perimeter. Subcooled boiling correlation was expressed considering that the temperature differences behind the convective and boiling contributions are different, i.e. q = h,(Tw-Th)+ Shpool(Tw -Tsal) (2.23) No enhancement factor is considered since no net vapour is generated, but the suppression factor remains in effect. Comparing the measured heat transfer coefficients with the coefficients predicted with their equations, Gungor and Winterton found mean deviations of 21.4% for saturated boiling and 25.0% for subcooled boiling. They also compared other previous eight correlations with their database such as Chen (1966) and Shah (1982). They found that Chen correlation overpredicts the saturated boiling heat transfer coefficient of refrigerants (mean deviation 91.7%), but provides acceptable results for all fluids under subcooled boiling conditions (mean deviation 27%). Shah correlation acceptably agreed with the data for all fluids (mean deviation 21.9% for saturated boiling and 36% for subcooled boiling). One year later, in 1987, Gungor and Winterton proposed a new simplified general correlation for the saturated flow boiling [13]. Compared to other existing accurate correlations, the number of equations required for calculating the heat transfer coefficient was halved. The new equation was expressed in the following form: 25 ^ - = 1 + 3 0 0 0 x Bo0M + 1 . 1 2 J — 1 h, [l-xj [pv where hi is the single-phase heat transfer coefficient calculated with Dittus-Boelter equation and Bo is the Boiling number. For horizontal tubes and Froude number less than 0 . 0 5 they recommended E to be multiplied by Fr(0''~2Fr). Further, they recommended for saturated boiling in annuli the use of heated equivalent diameter rather than the usual hydraulic diameter. This method was also tested for saturated boiling heat transfer, see Chapter 8. For validating and comparing their correlation with the data bank they have also analysed Shah, Chen and Kandlikar correlations. Previous tested correlations which proved unsuccessful (i.e. Bennett-Chen 1 9 8 0 , Mumm 1 9 5 4 , Bjorge et al. 1 9 8 2 , and the modified versions recommended by Stephen and Auracher in 1981 of Rohsenow, Chawla and Kutateladze correlations) were not included anymore in their analysis. They found that for saturated boiling, only their correlation and that of Shah showed good agreement, which does not vary with different ranges of dimensionless parameters. The analysis was systematically undertaken for different ranges of liquid Reynolds number, Boiling number, Froude number, quality, reduced pressure and Prandtl number. For these ranges the mean deviation for Gungor-Winterton correlation varied between 1 0 . 5 % and 4 3 . 7 % and that of Shah between 1 2 . 4 % and 4 8 . 5 % , respectively. Gungor-Winterton correlation for subcooled boiling proved unsuccessful. However, based on their data they pursued the analysis of Chen correlation and found that it gave the smallest error with the data. The mean deviation was 1 2 . 4 % for all the data and 1 5 . 7 % and 1 4 . 7 % for R-l 1 and R-12 refrigerants, respectively. This can be considered as an excellent agreement for heat transfer calculations. 2 6 2.4 The Liu and Winterton correlation, 1991 In 1990 a new correlation, based on data bank larger than the one used by Gungor, 4202 saturated data points and 991 subcooled boiling data points, is reported at University of Birmingham [20]. The authors, Liu and Winterton find the correlation applicable for subcooled and saturated boiling, and "more reliable than any of the other correlations tested". The correlation uses Chen's basic postulate that both forced convective and nucleate boiling heat transfer mechanisms have a separate contribution in flow boiling. The method of combining the two contributions suggested by S. S. Kutateladze in 1961 was used: The advantage of this method is that the nucleate boiling is further suppressed once the (F hL) term is appreciably larger than (S h p 0 0 i ) . In the above equation F hL is the forced convection contribution, with F - forced convection heat transfer enhancement factor and hL calculated from Dittus-Boelter equation with the entire mass flow rate flowing as liquid only. S hp o oi is the contribution of the nucleate boiling mechanism, S - suppression factor and h p 0 O i calculated from Cooper's pool boiling correlation. Using all the available data, the authors proposed mathematical expressions for E and S, see Chapter 7 for detailed use of the correlations. For known heat flux, the equations can be used directly to calculate the boiling heat transfer coefficient, and subsequently the wall temperature. As an alternative method, if the wall temperature is known, the heat transfer coefficient can be found from: (2.25) 27 hTP = F x hL x qt 3/2 (2.26) where q* is the single (real) root of the standard cubic equation: ql-Cxql -1 = 0 (2.27) Coefficient C is found as: C = (ApS/FhL)2q4L 4 / 3 (2.28) and, ^ = 5 5 P ; i 2 ( - l o g l 0 P r r 5 M - 0 .5 (2.29) In equation (2.29) Pr is the reduced pressure and M the molecular mass. For subcooled boiling, the driving temperature differences for nucleate boiling and for forced convection are different, therefore the basic equation can be written as: where ATD - Twaii - TDUik and ATS = Xwaii - Tsat- For subcooled boiling the quality x is set to zero and hence F = 1.0. If the temperature difference is known, then htp can be computed from the above equations with qL defined as (F h L ATb). If the temperature difference is unknown, then rearranging equation (2.30) gives: (2.30) (2.31) 28 where: and Shpi>ol(Ts-Th) (2.32) and the two phase heat transfer coefficient is calculated with: AT; (2.33) Except those related to the liquid Reynolds number and the liquid Prandtl number which are calculated at the bulk temperature TD, all the other thermo-physical properties are calculated at the saturation temperature Ts. Compared to the more simple addition models, an advantage of this method is that the nucleate boiling is fully suppressed as the vapour quality increases. Liu and Winterton found a mean deviation to their data of 20.5%. 2.5 Darabi et al correlations review The existing correlations for the prediction of flow boiling heat transfer were comprehensively reviewed by Darabi et. al. in 1995 [6]. They have noted the importance of reliable prediction of the heat transfer coefficient for the successful design of modern heat transfer equipment. As they mentioned, "in spite of the enormous volume of published literature, the prediction of heat transfer remains essentially empirical due to the complex hydrodynamics and heat transfer processes involved". They have classified the flow boiling models in four different categories: • models based on dimensional analysis • models which assume an additive contribution of the nucleate boiling and the convective evaporation to the boiling heat transfer coefficient 29 • models that take the greater of the nucleate and convective components as the final heat transfer coefficient • asymptotic models based on the power-type addition of the two components Out of all the published information on boiling heat transfer they focused only on convective boiling heat transfer of pure and mixed fluids for both smooth and augmented tubes. Correlations like Chen (1966), Shah (1976, 1982) and Gungor-Winterton (1986) have been treated in depth. Other correlation presented is the one developed by Bjorge et al. in 1982. It is a superposition correlation for saturated boiling in tubes at qualities above 5%. The pool boiling term has incorporated a dimensional constant determined experimentally, with specific values for given fluids. Its value for water was given by the authors, with the recommendation that for other fluids a specific constant has to be determined. The mean deviation for the correlation was 15%. The 1988 and 1990 correlation of Klimenko, also reviewed by Darabi et al., is constructed in terms of a dimensionless number and is valid for both vertical and horizontal channels, with the tube walls wet, meaning that the dry-out condition does not occur. The mean deviation was of 14.4% for 21 different fluids from 75 different sources. Kandlikar also introduced in 1990 a general correlation for flow boiling in vertical and horizontal tubes with Froude number greater than 0.04. The two-phase heat transfer coefficient is chosen as the greater of the nucleate boiling and the forced convective terms. The Dittus-Boelter equation is employed in the calculation of single-phase liquid only heat transfer. A fluid-dependent parameter, which is used for calculation, is tabulated for several fluids including seven refrigerants. Since the correlation was developed at the early stages of R-134a introduction in the refrigeration market, its parameter value is not provided. Based on more than 5000 data points, the correlation 30 was found to provide a mean deviation of 15.9% for water and 18.8% for most used at that time refrigerants. Steiner and Taborek in 1992 developed a general correlation for flow boiling in vertical tubes based again on Kutateladze's power-type addition model. His model, also known as the asymptotic model was considered as the most logical model. It is expressed as: It has to be noted that i f n approaches infinity the correlation becomes of the "greater of the two components" type, such as Shah's and Kandlikar's models. For n=l represents the superposition model, characteristic for Chen's and Gungor's correlation and i f n=2 becomes the L i u and Winterton's correlation. Performing a regression analysis on a large data bank of about 13000 points, Steiner and Taborek found an exponent « = i . Apart from traditional expressions for the nucleate pool boiling and convective boiling terms, their correlation includes correction factors that are complicated functions of reduced pressure and molecular weight. Testing the method, 92% of the results for ethanol, 71% for R - l 2 and 88% for N H 3 were found to be within an error band of 20%, without reporting the mean deviation. Darabi et al. synthesis (Table 2.1) shows how the correlation's authors compared their results with the previous analysis, in terms of mean deviations. It can be noticed that the correlations are much dependent on the database used. (2.34) 31 Kandlikar • i I i oo oo • Klimenko I • i • • 14.4 I Liu-Winterton I • • I 20.5 l 1 i Gungor-Winterton I • I 21.4 24.3 l 24.5 I Shah I • 14.0 21.9 23.1 • 23.7 I Bjorge et al. I 15.0 • 59.5 l 46.5 I Chen 12.0 17.4 i 57.7 37.2 • 43.5 • Data Points m VO o o oo o oo r- 3693 4183 3125 4970 13000 Correlation, year Chen, 1966 Bjorge etal., 1982 Shah, 1982 Gungor-W., 1986 Liu-W., 1991 Klimenko, 1990 Kandlikar, 1990 Steiner-T., 1992 32 Chapter 3 Review of Experimental Studies for R-134a The refrigeration industry and other related fields (air-conditioning, heat pumps) have obtained significant results in the process of conversion from CFCs to more environmentally sound alternatives. Some of the replacements are the hydrochlorofluorocarbons (R22 and R123), the hydrofluorocarbons (R134a and R152a, etc) and the natural refrigerants (ammonia and hydrocarbons). These changes involved the redesign and testing of new equipment as well as the retrofitting of existing industrial equipment. The refrigeration heat transfer research has been actively involved in the process and will remain the same since the goal of finding more efficient and environmentally friendly refrigerants will always be of actuality. For the last twenty years and especially after the Montreal Protocol many experimental studies of refrigerant boiling heat transfer have been reported. In the following review we tried to focus on experimental programs which present a certain level of similitude with our program. In that respect, we have selected the experimental programs using R134a as the heat transfer fluid, or employing similar facilities. The interests in R134a can be dated after 1990, and that is due to the need of finding a replacement for R12. 3.1 M. Pate and S. Eckels, 1991 The Department of Mechanical Engineering of the Iowa State University with their Professor Michael Pate and his assistant Steven Eckels were probably the first to report experimental studies of R134a, in 1991 [8, 9]. They measured heat transfer coefficients for HFC-134a and CFC-12 during in-tube single-phase flow, evaporation and condensation. Their goal was to determine the evaporation and condensation heat transfer coefficients of HFC-134a, using measurements for CFC-12 as a baseline for evaluating the performance. 33 The test facility consisted of four main parts: a refrigerant loop, a water loop, a water-glycol loop and a data acquisition system. The refrigerant loop contained the test section, an after-condenser, a positive displacement pump, an accumulator bladder, a boiler and superheater. The test section was a horizontal smooth tube surrounded by an annulus. The inner tube in which the refrigerant flows was a 3.67 m long smooth tube with an outer diameter of 9.25 mm and an inner diameter of 8.0 mm. The refrigerant was heated or cooled during the tests by the water flowing in the outer annulus. The after-condenser was a heat exchanger that subcools the refrigerant discharged from the test section to -15°C. A positive displacement pump, the flow rate being regulated by valves and a bypass line, then pumped the refrigerant. The accumulator set the system pressure and dampened any pressure fluctuations. The quality entering the test section was controlled by a boiler and a superheater located just before the test section. The boiler was a 12.7 mm OD, 2.63 m long stainless-steel tube electrically heated, whereas the superheater was a 12.7 mm OD, 1.83 m long copper tube wrapped with ceramic bead electric heater. The flow rate was measured by a positive displacement flow-meter with an uncertainty of +/- 1%, the refrigerant pressure was measured at the test section inlet by a strain-gage pressure transducer, accurate to +/- 9 kPa. The pressure drop through the test section was measured again with a strain-gage type differential pressure transducer, accurate to +/- 0.2 kPa. A pair of thermocouples was located at the inlet and one at the outlet of the test section. The water loop was used to heat or cool the refrigerant flowing in the test tube. The loop consisted of the test section annulus, a centrifugal pump, a magnetic flow-meter, and a heat exchanger. The temperature was controlled by the heat exchanger, which was supplied with hot or cold water from the building taps. The water flow rate was measured with +/- 2% uncertainty and the temperature with +/- 0.2°C uncertainty. The water-glycol loop contained a storage tank of 209 1 capacity, a centrifugal pump, a coaxial heat exchanger and a 17.5 kW refrigeration unit. The 50:50 mixture was cooled down to -20°C and was used to condense the refrigerant leaving the test section. 34 Data acquisition system consisted of a controller, a 40-channel scanner, and a multimeter. The system monitored 20 thermocouples, two pressure transducers, two flowmeters and the voltage and current to the boiler. The ambient temperature and pressure were entered manually. The data reduction was based on equations given by the energy balances of the water flowing through the annulus and of the refrigerant flowing through the tube, for single phase. The vapour quality entering the test section was calculated from an energy balance on the boiler and superheater. The saturation temperature was determined from the saturation pressure at the inlet of the test section. The quality change in the test section was calculated from the heat transferred in the test section and the refrigerant mass flow-rate. The refrigerant side heat transfer coefficient was determined from the overall and the annulus side heat transfer coefficients, using the LMTD method. This method determined the average heat transfer coefficient over the length of the tube. The researchers reported experimental heat transfer coefficients for HFC-134a and CFC-12 during evaporation, condensation and single-phase flow. The range of test conditions was selected to reflect the actual operation of refrigerating systems. It was found that in single phase the heat transfer coefficients for HFC-134a were 33% higher than for CFC-12. For evaporation, at similar mass fluxes, HFC-134a heat transfer coefficients were 35 to 45% higher than for CFC-12, and for condensation 25 to 35% higher than CFC-12 coefficients. The evaporation experimental data was also compared with predictions from the correlations of Shah, Kandlikar, Chaddock and Brunemann, and Gungor and Winterton. For HFC-134a all the correlations showed an agreement of +/- 25% or less, with Chaddock-Brunemann and Kandlikar correlations predicting the heat transfer coefficients by less than +/- 15%. For CFC-12 the most accurate was the Shah's correlation, with less than +/- 15% deviation. 35 3.2 K. Torikoshi and T. Ebisu, 1993 Another important contribution to the understanding of heat transfer behaviour of R-134a, R-32 (HCFC as well) and a mixture of R-134a/R-32 was brought by K. Torikoshi and T. Ebisu, from the Mechanical Engineering Laboratory of Daikin Industries Ltd., Japan [38]. Their experimental results reported in 1993 were based on a comparison with HCFC-22 in a horizontal smooth tube during evaporation and condensation. The experimental rig was mainly composed of a refrigerant loop and a water loop. The refrigerant loop consisted of a test section, a circulating pump, flowmeter, a heating section and two heat exchangers. The flow meter connected upstream of the test section measured the flow rate, regulated by a controlling valve. The heating section installed just before the inlet into the test section controlled the refrigerant quality. A differential pressure transducer connected to pressure taps at the inlet and outlet measured the pressure drop across the test section. The main heat exchanger was installed in the refrigerant loop to regulate the system pressure. The auxiliary heat exchanger was used to completely liquefy the refrigerant entering the circulating pump. The water loop was used to heat or cool the refrigerant flowing through the tube side of a heat exchanger. The water loop includes the shell side of the heat exchanger, a pump, a flow meter, and a constant-temperature water bath. The flow rate was controlled by the pump. Platinum resistance sensors measured the inlet and outlet temperatures of the water and refrigerant. The test section was equipped with a vacuum layer to minimize the heat loss. The heat transfer coefficient on the refrigerant side over the length of the tube was found using the overall heat transfer coefficient given by the LMTD method, the water (shell) side heat transfer coefficient from the Dittus-Boelter correlation, and the surface areas and the thermal resistance of the inner tube. 36 It was found for R-l34a that the evaporation heat transfer coefficient is about 15% smaller than for R-22. The condensation heat transfer coefficient for R-l34a was found about 10%> larger than for R-22. The pressure drop results for R-l34a during evaporation and condensation were larger than for R-22. 3.3 H. Takamatsu, S. Momoki and T. Fujii, 1993 An other group of researchers from Japan, H. Takamatsu, S. Momoki and T. Fujii from the Kyushu University and Nagasaki University have reported in 1993 an experimental study of HFC-134a, HCFC-22, CFC-114 and CFC-12 flowing inside a 7.9 mm ID horizontal smooth tube [35]. Their experimental apparatus was made up of four loops: a refrigerant loop, two water loops and a brine loop. The refrigerant lop consisted of a positive displacement pump, a preheater, a test section a rear heater, a condenser and an auxiliary condenser. The water loops were used for supplying the heating and cooling water to the test section and the condenser. The brine loop was used for cooling the refrigerant at the auxiliary condenser. The test section used in experiments was a double-tube type evaporator with the refrigerant flowing inside the inner tube and the hot water flowing through the outer annulus. The inner part was a copper straight smooth tube 6 m long, of 7.9 mm inside diameter and 9.5 mm outside diameter. The outer tube was built out of two polycarbonate resin blocks shaped in such way as to provide an annular gap for water of 3.25 mm. The outer annulus was divided along the tube into twelve subsections by brass blocks to measure the local heat transfer rate. The effective heated length of each subsection is 0.46 m. The direction of the refrigerant flow was switched with valves so as to obtain data for both parallel and counter flow conditions. The bulk temperatures of the refrigerant were measured at the mixing chambers from the inlet and outlet of the test section. The wall temperatures were measured at the center of the each subsection with four thermocouples attached at the bottom, top and sides of the outside surface of the 37 inner tube. The saturation temperature of the refrigerant was measured with thirteen thermocouples inserted into the inner tube at the end of each subsection. The pressure at the inlet of the test section and the pressure drop were measured with a pressure transducer and four differential pressure transducers. For the flow rates, a mass flow meter was used for refrigerant and a gear type for water. The experiments were performed with pure refrigerants, at mass velocities of 100, 200 and 350 kg/m2s, with reduced pressures of 0.13 to 0.23. The data was collected for three different regimes: subcooled liquid at the inlet with superheated vapour at discharge, vapour quality at inlet of x=0.3 with superheated vapour at discharge and subcooled liquid at the inlet with x=.0.7 quality mixture at discharge. The local heat transfer coefficient was defined as: where Tsat was calculated from the measured pressure and Twi the average inner wall temperature calculated from the measured outside wall temperature by radial heat conduction equation. The bulk enthalpy and the vapour quality were calculated from a heat balance equation. The researchers compared their data with correlations proposed by Yoshida et al (1990), Jung et al (1989), Dembi et al (1978), Dhar et al (1979), Shah (1982) and Gungor and Winterton (1986). Among these equations, that of Yoshida et al showed a better agreement with the experimental data. They have also developed a correlation based on Chen's superposition method. For their experimental data, the correlation predicted the heat transfer coefficient with a mean deviation of 12.2%. Data available from previous papers was predicted with 9.5% accuracy. (3.1) 38 3.4 J.P. Wattelet and J.C. Chato, 1993 In 1993 a group of researchers from the Department of Mechanical and Industrial Engineering, University of Illinois, conducted by J.P. Wattelet and J.C. Chato reported experimental heat transfer coefficients for R-l2, R-l34a and a mixture of R-22/R-124/R-152a [39]. The refrigerant loop used for experiments consisted of a variable-speed gear pump, a 6 kW variac controlled preheater was used to control the inlet quality, two parallel condensers cooled with ethylene-glycol-water mixture from a chiller for removing the heat. The flow rates were measured with a Coriolis-type flowmeter, the absolute pressures with strain-gauge pressure transducers and the temperatures with T-type thermocouples. The test section was a thick horizontal copper tube of 2.43 m long and 7.04 mm inside diameter. The heat input was provided by surface-wrapped heaters, controlled by variacs. The heat rate was measured by a watt transducer. The test section was insulated in order to reduce the heat loss. The surface temperature was measured by sixteen thermocouples soldered in longitudinal grooves along the test section. Bulk fluid temperatures were measured at the inlet and outlet of the test section, using T thermocouple probes. Differential pressure across the test section was measured using a pressure transducer. Sight glass with the same inside diameter was installed at the inlet of the test section to allow for flow visualization. Data collection was performed using a data acquisition system. Parameters controlled during test were mass flux, heat flux, inlet quality, and saturation temperature. The test parameters were varied as follows: mass flux from 25 to 100 kg/m2s; heat flux from 2 to 10 kW/m2; quality from 10 to 90% and saturation temperature from -15 to 5°C. Experimental local heat transfer coefficients were determined by the convective law of cooling using the circumferentially averaged surface temperatures, the linearly interpolated bulk fluid temperatures, the area of the test section and the heat input rate. The axial heat conduction along the length of the tube and the radial conductance through the wall were neglected. The results were compared with the Shah (1982) correlation and 39 with Kandlikar (1990) correlation. It was found that both correlations predicted the heat transfer coefficients with less than 20% deviation. Comparing the experimental results was found that the heat transfer coefficients for R-l 34a were on average 25% higher than for R-l2, for equivalent conditions. They have also proposed a heat transfer correlation using an asymptotic form. The new correlation predicted both annular and wavy-stratified flow well. 3.5 J.Y. Shin, M.S. Kin and S.T. Ro, 1996 Interest in experimental analysis of refrigerants has been developed at Department of Mechanical Engineering of Seoul National University. Jee Young Shin, Min Soo Kin and Sung Tack Ro have reported in 1996 results for R22, R32, R134a, R290 and R600a and a series of refrigerant mixtures [32]. As experimental set-up they used a refrigerant loop with a magnetic pump for circulation, a mass flow meter, a preheater, a test section (evaporator), and a heat exchanger for subcooling the refrigerant, a liquid receiver and an power supply system. The test section was an electrically heated seamless stainless steel horizontal tube, of 7.7 mm inner diameter, with an effective heated length of 5.9 m. The absolute pressures were measured at inlet and at locations along the test section, 1 m apart. T-type thermocouples were attached to the test section at 12 locations along the axis. At each location the temperature was measured in four section, 90 degrees apart. The experimental conditions covered were as follows: heat flux from 10 to 30 kW/m2, mass fluxes set in discrete values from 424 to 742 kg/m2s. To obtain the heat transfer coefficients the outside wall temperature were measured and then the inside wall temperature calculated considering the radial thermal conductance. The saturation pressure was found from the measured pressure. The power was measured with a wattmeter, and the flow rate with the flow meter. The heat transfer coefficient was found 40 from the ratio of heat flux and the temperature difference between wall and the saturated fluid. A detailed comparison of the experimental data is presented. The researchers have also tried comparing their data with Gungor and Winterton Correlations of 1986 and 1987. They found deviations from the experimental data of -44.7% for all data with 1986 correlation and -30.5%wifh the 1987 revision. For R-l34a they found a deviation of -50.8% and -47.5% with the two variations of the correlation. 3.6 X. Liu, 1997 X. Liu of the Carrier Corporation pursued a comprehensive research of R-l34a and HCFC-22, publishing results in 1997 [20]. They used a four-loops facility: a refrigerant cycle containing a reciprocating-type compressor for condensing or evaporating tests; a heated controlled water loop to supply the load for the evaporator; a controlled cooling water loop that removed the heat load from the condenser and a cooling tower brine circuit that served to remove heat from the system. The test section consisted of six identical horizontal passes, constructed as double tube heat exchangers. Each pass was about 2.2 m long and connected in order by U-bends. The heat transfer section of each pass with the water pass in the annular side was 1.8 m long. The inner (refrigerant) tube was a 9.5 mm copper tube with 72 axial fins of 0.185 mm height on its inner surface. The annulus-side water providing the heating or the cooling was supplied into the first and fourth section to provide a more uniform heat transfer. The water temperature was measured at the entrance and exit of each pass using platinum RTDs. The amount of heat transferred in each pass was determined from the waterside flow rate and temperature rise in each pass. The enthalpies of the refrigerant superheated vapour and subcooled liquid were taken from thermodynamic properties at the appropriate saturation conditions and the measured superheat and subcool temperatures. The flow rate of refrigerant was based on the net heat transferred and the 41 specific enthalpy change across the test section. The specific enthalpy change was apportioned according to the fraction of the total heat transferred by each pass. The vapour quality at each pass was defined in terms of the specific enthalpy at the chosen points and the enthalpies of the saturated vapour and liquid. The overall heat transfer coefficient for the "n" pass was determined from the LMTD method, which then led to calculation of the quasi-local refrigerant side heat transfer coefficient for the "n" pass. A detailed analysis was done based on the experimental results. The variation of the heat transfer coefficient and the frictional pressure drop were analysed. The results were also compared to some other experimental data, yet being limited by the different geometrical characteristics of the test section. 3.7 N. Kattan, J.R. Thome and D. Favrat, 1998 The latest reference found in our review of similar experimental programs was the work done by N. Kattan, J.R. Thome and D. Favrat, published in 1998 [17]. The program used five refrigerants (R134a, R123, R402A, R404A and R502) for analysing the heat transfer during evaporation inside a plain, horizontal, copper tube test section. The test data were obtained for 12.00 mm and 10.92 mm diameters using hot water as a heating source. The test rig had four double-pipe test sections grouped in two pairs, with each pair having two double pipe test sections connected in series in a U-tube type configuration. One pair was used with the test fluid flowing inside the inner tube, and the other pair was used with the fluid flowing through the annulus. The water was directed to either pair of the test sections. The construction using flexible and insulated stainless steel hoses allowed for experiments with test section mounted horizontal, vertical or inclined. Before reaching the test section, the refrigerant was passing through a 6 kW preheater, then through on pair of U-tube test sections. Then a condenser cooled down the fluid and a magnetically driven pump completed the circuit. The speed controller on 42 the pump adjusted the flow rate, measured by a Coriolis flow meter connected into the circuit. The hot water heater circuit reheated the water passing through the test sections. The flow rate again was measured by a Coriolis flowmeter and the flow adjusted by the by-pass system of a stainless-steel pump. The condenser used a water-glycol solution maintained at a fixed temperature by a refrigeration system (-15 to +15°C), with a 200 1 receiver being connected into the circuit for eliminating the temperature fluctuations. The test sections consisted of hard PVC outside tubes machined into two halves. A copper tube centred by three screws in several cross sections completed the construction. Thermocouples inserted and sealed through the plastic tube, at four locations 0, 60, 120 and 180 degrees from the top of the annulus, and installed at four measurement locations were used for determining the heat transfer coefficients. Two thermocouples at the second test section were used for measuring the wall temperature, in particular between the top and the bottom of the test section. Tubular sight-glass sections were installed inline with the test sections to visualise the flow at four locations (inlet and outlet of each test section). A computer data acquisition system was used for acquiring, analysing and storing the data. For R134a, the experimental data was presented in terms of vapour quality influence on the heat transfer coefficient. The experimental conditions reported for R134a covered mass fluxes of 102.3, 201.2 and 301.6 kg/m2s, heat flux range of 5 to 25 kW/m2, with a saturation temperature of 10.3°C and tube diameter of 12 mm, for the first set of data. The second set covered mass velocities of 100, 199 and 299 kg/m2s, heat flux range of 4 to 14 kW/m2, at a saturation temperature of 4.4°C and tube diameter of 10.92 mm. A more in detail analysis was done for R502, R402A and R404A. A synthesis of the experimental facilities is presented in Table 3.1. 43 fi © 13 a o U B a a -o o T t CN **? 0 s 00 oo O i 1 o oo IT ) II i - II 3 II -t-» J O 3 ON * — i o T t >, -4—->. * 1 "3 13 ro a CN © oo J*i o oo C N I o T t oo r-CN CD II 3 a cu .t i 3 ^ o 3 O 03 CD J3 U . CU a , 3 00 00 J*( o >n ro ©" o CN © " O CN CU s-3 (/) c« o <=> C II H T3 "o O o JO 3 - a cu CU -3 - a CU 15 cu _c cu a 3 cn _CD -t-» 3 o 00 J4 o o CN J4 © i CN u o o o o ON 15 3 15 3 3 a-fi o H ex 93 fi CU i -CU a 3 .3 "o o o ~So 3 "3 cu J 3 o s r -ro C o 3 CT o 3 » C3 -a D, o o B B ,2 00 s S - 5 0 0 u CU 00 o _g c+J cu cn D . o o 3 O CU H—» C3 00 c c cu "3 3 o o o o e C« 13 +^  3 o o 03 00 _c "o o o "oo _3 %—» 03 <U J 3 I cu .2? !^H <H—I cu cn a , o _o o cu D H OH O O CU J O 3 cu 3 3 cn _3 3 3 C cd 4-* 3 O _N 'C o O i—i CN CN CN oo co I cu J O 3 3 O Q" o CN ON 00 _3 "o O cj 03 00 .3 13 CU J 3 cu oo cn D H O O 3 O OH O _o CU 3 00 _3 °cn 3 cu -3 3 O cu cu cu OH OH O o o o s cn CU J O 3 cu 3 3 3 3 3 3 13 +-» 3 O N "EH o cu J O 3 cu 3 o d B Q" O ON s s ON CU JD 3 00 CN c/5 J»< O _3 izi <U H3 CN o o 00 03 5 -3 CU 3 O O o .2 "C > 00 _s 13 cu J 3 CJ CU w OH o _o 1-<u cn 3 cu -a 3 o CJ B B CN s s o cu J O 3 cu O H D H o CJ +^ G O o a ro CN 03 5 •r s CU S o <Z5 cd ^r ro i <u J ^ CJ W r/3 cu —^» cd &H CN cn k H cu > "S cu 13 03 O Os ON CQ a ^ cd ro • ^5 J 3 cn O J ^ o H CN ro i Ct, C^ r l -r o s 3 cd DH 03 cn <U CN ro 3 T t ro 5 CN CN a. J O W cn 3 T3 3 ro ON ON 3 J3 '3 Q 3 cn 15 s cd H ad 3 UH H o e o 00 U IX, U j>ti 3 cn 3 00 3 3 J3 cn 3 CN u t i n ro ON ON CU 'o CN 3 T t ro * C£i T t CN 2 15 o 3 o F= 3 J3 U o cu > '2 ZD ro ON ON 44 00 CN T T f -i T T (N T T O cn U o I T ) O i + CU o s 2 oo o o r-H—» _ f i (U T3 <U -4—» CS CU X t-oo MD IT ) 00 CN 3 o oo C4 a MD o cn U o r o U o T T >0 ON £ CN —' -4-T T 1 e o o o 3 ts 00 cS 3 C/3 CN s a ~oo M O M i—l O CO O 1 CN i cn oo X CS cS X CN I/O u "o o o X 3 00 13 I/O CN X T3 JU "o o o 3 oo 3 o Q. O O c cS eu 00 0) 3 a, <u fi 00 cS a oo a CU X O H o o 00 .s *o o o as Is a a Os O <u | C/3 oo 00 eu a cS <U O o o cS T3 <U ts CU X a CS fcH 1) op )£H C4-H CU 00 O H O O 3 O 3 Q <u C CU o _e "o o o a, o _o CU ts 00 . c 'oo a CU a o o 1/0 Os" c %—» o CU 00 s-J S 3 o 1—1 a 00 c IS CN oo CU 00 00 cd D . -4—» a o _N '(C O X MD cS a a 00 T T cn cS CU oo CU 1-cx a 3 On CU c 00 CS a o ts S-I o > CU S-i CU ts o X O H o o o © CN oo G _o *-t—» o CU oo -t—» 00 CU O - cS 00 CU X U > O H 3 O 00 eu > ts X CU ts -o eu 2 o (U 00 ~ CS CN cS CJ ••s (U > 3 2 ob <u S-i d, o o c o N o CN as 00 ^ cS T3 C cS 3 3 00 "o o o CS & 00 c '"5 CU cu c O o CJ .2 cs > "cS o o CU W oo cn cs CU X i "eu fi O o _e Q O CN oo 3 cS *cS o <u > o -d eu ts <U X MD O CU X 3 00 00 J S 00 X CU I— >, s a CN CN cS cn cn o "eu O as 2 3 CN Pi cS Ti-r o s CN s < CN O T t oi < o 2 CN O I T ) Pi x CS c o o CU Q0 CD H ty5 MD as as <u > "c ON ON c o 3 >< cS O & o U cS U CU s o X H Pi •> cs <U -o CU C CS •4—' ts C/3 CS 1-> CS fc Q A 00 o o eu H o <u —^» 3 eso as as eu oo cS i-i fc Q 3 o O H CS Ti-r o X oo PQ <+H o CDs Os cS >. X ^ H 00 <u > " f i _3 "o 45 Chapter 4 Experimental apparatus The experimental investigation covered by the present study was accomplished in order to contribute to the general understanding of boiling heat transfer of R-134a, and as well to generate a reference data base for testing and comparing the performance of other proposed alternative refrigerants. The design and construction of the present apparatus was determined by the following requirement: • Exact measurement of the local boiling heat transfer coefficient • Possibility of visualising the boiling pattern • Necessity of studying the influence of mass flux, heat flux and system pressure on the heat transfer coefficient The experiments were performed using a facility developed by A . H . Abdelmessih in 1973 [1] and used for the experimental study of CFC-11 . The facility was suitably modified by the addition of state-of-the-art data acquisition system and hydraulic equipment, and by the addition of a newly designed test section, as presented in the subsequent sections. 4.1 General description The main circuit of the apparatus consists of a closed circulation loop for refrigerant. Heat transfer and hydraulic components made of stainless steel are installed along the flow lines. The connections between loop components are achieved by 3/8" schedule 40 piping and fittings and by '/z" O D x 0.035" wall thickness 304 stainless steel tubing and Swagelok tube fittings. 46 The test section of the loop is designed to operate up to a maximum pressure of 180 psia and a fluid temperature of 47°C, corresponding to the saturation temperature of R-l 34a, chosen as testing refrigerant. As shown schematically in Figure 4.1, the entire loop is attached to a steel structure with overall dimensions of 10 ft high, 6 ft wide and 2 ft deep. 4.2 Circulating pump The circulating pump, component #3 in Figure 4.1, is an important part of the forced convection loop and it must meet the following requirements: (i) Under steady state conditions the pump outlet pulsation (both in flow and pressure) should be negligibly small compared to the hydrodynamic oscillations in boiling two-phase flow. (ii) The pump should have negligible flow oscillations due to a change in head. This means that a steep head flow characteristic is required. A positive displacement pump does have a steep head flow characteristic but the large pressure pulsation generated at the pump outlet and the fast wear in mechanical parts render such pump as undesirable. A.H. Abdelmessih [1] has demonstrated the above-mentioned shortcomings associated with positive displacement pumps in a series of tests in the early phases of construction of the original apparatus. Their final solution was achieved by using an Aurora DO 3 turbine pump, with a bypass circuit of negligible restriction together with the use of sufficient throttling to ensure that the test section pressure drop is a small part of the total pressure drop. As shown in the literature review, many researchers have used magnetically driven pumps, that operates without any lubricating oil on the fluid side, thus eliminating the 47 48 possibility of oil entering the refrigerant flow. Due to prohibitive prices, this option was excluded. The pump finally selected for the loop was an Aurora FO 5-turbine pump direct coupled to a 230V/3PH/60HZ, 3 HP electric motor. The pump was selected in such way as to provide a flow up to 10 GPM (Figure 4.2), with a discharge head high enough to compensate for the pressure drop along the liquid lines and fittings, through the flow-meters, preheater, and the test section itself. The estimated pressure drop was 300 feet. Figure 4.2 Aurora FO 5 pump head curve CAPACITY IN G. P. M. 33PC - I 1532 One other desired feature of this pump was the mechanical seal along the shaft, with carbon against Ni-Resist face with 303 stainless steel parts and "Buna-N" elastomers, chemically compatible with R-l 34a. 49 Also the fact that the turbine impellers can handle gases and vapours up to 20% along with the liquid and thus reducing the possibility of cavitation and vapour lock formation within the pump, was an important consideration. To evacuate the vapours trapped in the pump, after a longer period of inactivity, we fitted the pump with a purge valve mounted on top of the case. A 72" stainless steel globe valve was connected in the pump bypass. The pump inlet and outlet are connected to the loop with the aid of two 18" long stainless steel wire clayed Teflon hoses to isolate the pump-motor unit from the rest of the loop and to dampen the resulting mechanical vibrations. The pump volumetric flow was regulated by adjusting the RPM using a Parametrics variable frequency drive, model Parajust A6031, connected as suggested in the original documentation and shown in Figure 4.3. Figure 4.3 Diagram of the VFD connection to the pump HOTt.-SZ /<Z> . 5O/C0H7 o 7732 o -azi> TBI CD-CD" © — & 04-© - - I L . A/ /// XftlR p ro D C 4r 'over* MCOWJ£ _ ran DISCHARGE ftes/sroR OISCHMOC FJ Discharge Fuse. RS rn TEH Ct'OKE M0OULS5 — • — I KZ\ • *"1 DlO/TXt. \Courncn. noot/ce To OhtlVC/4 50 Parajust is a motor speed controller for 3-phase AC motors. It furnishes variable frequency, variable power to convert a fixed speed motor to an adjustable speed motor. In the control, as the frequency is increased, the voltage is also increased to maintain constant motor torque. In the Parajust A6031, single-phase 230 VAC power is converted to variable voltage DC power by the DC Power Module, smoothed out by the filter choke and filter capacitor, then inverted into 3-phase AC power by the six Driver Modules. The proper voltage to frequency relationship is maintained by the Analog Control Module. The Digital Control Module sequences the Driver Modules to create 3-phase variable frequency power. 4.3 Flow meters The flow rate is measured by two flow meters connected in series. The low range flow rates are measured by a Brooks model 111 2A (Tube no. 8-RM-25-25) rotameter, component #10 in Figure 4.1, rated to 300 psia and 395°C. The rotameter is accurate from 0.078 to 0.78 GPM (see calibration data sheet in Appendix 3). The higher flow rates are measured with turbine flow meter Omega FTB-101, Figure 4.4, component # 9 in Figure 4.1. The FTB-100 Series Turbine Meters are volumetric measuring flowmeters. The flowing fluid engages the vaned rotor causing it to rotate at an angular velocity proportional to the liquid flow rate. The angular velocity of the rotor results in the generation of an electrical signal (ac sine wave type). Summation of the pulsing electrical signals relates directly to the total flow throughput. Frequency of the signals relates directly to the flow rate. The vaned rotor is the only moving part of the flowmeter. The deflector cones eliminate downstream thrust on the rotor and allows for hydrodynamic positioning of the rotor between deflector cones. Integral flow straightening tubes minimize the effects of upstream flow turbulence. 51 Figure 4.4 Turbine flow-meter construction The calibration characteristics provided by the supplier show a linear response within 0.35 to 3.4 GPM. Our calibration, Appendix no. 3, showed a linear response up to 4.0 GPM. The accuracy of the turbine flow meter is +/- 0.5% of reading, repeatability +/-0.1% of reading, temperature range: -267°C to 232°C and maximum pressure 5000 psig. The electrical signal is conditioned by an FLSC-28 High accuracy Integral Signal Conditioner, designed for direct mounting onto the OMEGA Series FTB-100 Turbine Meter. The input circuitry has been designed to receive and condition the low level turbine meter signals while rejecting unwanted noise. A signal threshold control is provided which allows for setting the input sensitivity above the noise level, thereby eliminating any false signal on the output. The unit runs off a 10 VDC power supply and provides a 0-5 V linear output signal of 0 to 5 VDC, signal converted into flow rate reading by the data acquisition system. 52 4.4 Preheater The preheater, component #2 on Figure 4.1, located upstream of the test section is used to control the inlet fluid temperature. The details of the preheater design are shown in Appendix no. 1. The container of the preheater is fabricated from a 2.375" OD (0.154" wall) stainless steel pipe with NPT threads on both ends for mounting the immersion heaters. The immersion heaters consist of two U-shaped heating elements with a full load power of 2000 W. The continuous power control of the preheaters is accomplished by a General Radio W50H variac, as described in the subsequent sections. 4.5 Test section The test section, component #17 in Figure 4.1, is presented in Chapter 5. 4.6 Separator, condenser and subcooler For experiments in the subcooled boiling regime it is essential to control the degree of subcooling of the fluid entering the test section. This requires that the liquid entering the preheater should be at a temperature substantially below the saturation condition so that effective control can be achieved at the preheater outlet and before the liquid enters the test section. To accomplish this, a separator, a condenser and a subcooler were installed in the loop to remove the energy generated in the preheater and test section and to lower the liquid temperature. In addition from separating the liquid from vapour, the separator, component #14 in Figure 4.1, also acts as an expansion tank. The separator is fabricated from a 20 inches long 3.5in. OD (0.216 in. wall) stainless steel pipe with welded caps. The two-phase 53 flow leaving the test section continues its passage through a 3 feet riser and is radially discharged into the upper part of the separator through thirty-six 0.125 in. holes at the end of raiser. A 12.4 inches long high pressure (0.375 in. OD) Pyrex sight glass is installed outside the separator tank to indicate the liquid level. Separator details are shown in Appendix no. 1. The refrigerant accumulated at the bottom of the separator is discharged through the side opening to the subcooler for further cooling. The refrigerant vapour flows upward through the top discharge line of the separator to a water-cooled condenser, component #13 in Figure 4.1. Since the glass tube portion of the test section is rated for a maximum pressure of 180 psia, a RL4 series low pressure proportional relief valve, component #12 in Figure 4.1, was installed at the highest elevation of the loop, at the discharge side of the separator. The relief valve was sized to allow for a discharge capacity at 150 psig cranking pressure of 10 Ft3/min. vapour flow, corresponding to the maximum liquid phase flow rate (4.0 GPM) scheduled for testing. The capacity curves for both liquid and vapour phases are shown in Figure 4.5. The relief valve was custom equipped with Ethylene Propylene seal, chemically compatible with R-l34a. The condenser is a commercially available stainless steel tube and shell heat exchanger (Young's Radiator Co.) with a total heat transfer area of 3.9 square feet. The condensate from the condenser is discharged through the condensate returning line connected to the suction side of the pump, where it is mixed with the cooler fluid coming from the subcooler. The subcooler is also a commercially available stainless steel tube and shell heat exchanger (Young's Radiator Co.) with a greater heat transfer area of 11.9 square feet. 54 Figure 4.5 Capacity characteristics and diagram of the pressure relief valve AIR Fiow rate, std L/min 50 100 200 500 1000 2000 3000 Flow rate, gal/min Cooling water for the condenser and subcooler is directly drawn from the lab city water supply. Fluctuations of the cooling water flow rate due to demands in other part of the building were noticed. However, experiences have shown that the disturbances occur for a short period of time and do not effect the general performance of the heat exchangers and of the whole loop. 4.7 Power supply The power to the test section, preheater and the pump-motor unit is supplied and controlled though a General Radio Model 1582-A voltage regulator. The voltage regulator is of electro-mechanical type with a 230 V (+/-10%) output, single phase, rated at 19.8 kW and 85 circuit amperes, providing an accuracy of +/-0.25%. 55 A 3.2 kW, single phase, 60 cycles transformer with a primary voltage of 240 V and a secondary of 8 V is connected with the power supply to the test section. A schematic diagram of the system power supply and control is given in Figure 4.6. Figure 4.6 Schematic diagram of the system power supply General Radio 15 82A Voltage Regulator 208 V, 50 A R W50H Variac 0-240 V,25 A Preheater, 2kW Variable Frequency Pump Drive Transformer, =dW| Maximum recorded 5 V, 850 A I Test Section 4.3 kW W50H Variac Preheater, 0-240 V, 25 A W 2kW 3 HP Turbine Pump A General Radio W50H open type variac rated at 6.0 kW and with an output range of 0-240 V is connected between the voltage regulator and the primary windings of the transformer to provide continuos control of power to the test section. Two lengths of welding cables rated at 600 A are used as power leads and are connected between the secondary windings of the transformer and the copper connectors on the test section. 56 The power to the preheater is connected so that one of the immersion heaters in the preheater is directly wired to the 230 V output of the voltage regulator while the other is connected to a second W50H variac before being wired to the voltage regulator. This arrangement renders a continuos and complete control of the power supply to the preheater. The turbine pump is powered through a VFD as described before, which allows for a continuous adjustment of the pump RPM, respectively of the pump flow rate. The RPM adjustment is achieved with a multi-turn (10 full turns) potentiometer providing a flow rate adjustment accurate to 0.01 GPM 4.8 Accumulator / pressurizer The accumulator is of a commercially available pneumatic-hydraulic operation, bladder type construction. Compressed nitrogen gas supplied from a liquefied nitrogen tank through a pressure reducer acts on one side of a synthetic rubber bag. The rubber bag expands and contracts in pace with the surrounding refrigerant liquid to absorb any pressure fluctuations. For adjusting the system pressure the pressure on the nitrogen side is increased or decreased using the pressure regulator or a discharge valve. The change in volume of the rubber bag leads to a change in operating pressure. In this way the system pressure can be adjusted with an accuracy of 0.01 psi. 4.9 Thermal insulation The system is thermally insulated to minimize the heat loss to the surroundings. This is also a factor in obtaining thermal stabilisation of the system and results in shorter waiting time for steady state conditions. 57 The loop piping, preheater and separator are insulated with commercially available piping insulation. The test section is not insulated since the visualisation of the boiling process is desired. However, since the temperature difference between the fluid bulk temperature and the lab temperature is not very high, a negligible heat transfer to or from surroundings is assumed. 4.10 Optical compensation box During the experiments, photographic images of the boiling and bubble formation were acquired for a collateral research topic (Prodanovic 2001). It was desired to eliminate the distortion due to light reflection and refraction. An in depth analysis was performed [11] and the design was based on its results. The optical compensation box was built as a square cross-sectional glass box with top and bottom aluminium caps and O-ring seals. The box was mounted on the test section at the testing location and filled with water. The design is shown in Appendix no. 1. 4.11 Safety equipment and procedures A clear Plexiglas shield was mounted around the glass test section to protect the operating personnel in case of failure of the glass tube. The safety relief valve was set at 165 psig cranking pressure and the discharge line piped to a safe area. The entire power equipment and Dexion frame have been properly electrically grounded. The commencement of tests was announced and the experiments were conducted only when other grad students were present in the lab. 58 Chapter 5 Test Section The test section is a vertical, concentrically arranged annulus with an inner tube consisting of three sections and an outer glass tube. The refrigerant flows between the two tubes. The heater and the glass tube are supported by two brass assemblies, designed so that they ensure the concentricity of the tubes, the connection to the fittings and support and alignment for the Pitot tube passing the test section. Details of each component are subsequently presented. 5.1 Inner Tube The inner tube is constructed of three sections of same diameter, one above the other: a thin wall copper tube, the stainless-steel heater and again a thin wall copper tube, pressure-fitted together. They are fabricated from commercially available tubing of V2 in. (12.7 mm) OD with 0.065 in. wall thickness. The size was selected in order to: • allow for the guiding of measurement devices inside the tubes • generate sufficient heat flux • ensure remaining wall-thickness for pressure fitting to connect the three sections The electrical heated length of the test section is 560 mm compared with the unheated length of 360 mm, equally spaced at the inlet and outlet. The unheated section is required to reduce the hydrodynamic entrance effects and to assure a turbulent flow profile at the heater. The ends of the inner tube are connected to the power leads. 59 Figure 5.1 Test section schematic DC 4.3 kW Power source Brass body Heater, 1/2" OD Cu/Inconel/Cu Glass Tube 1.5" ID, 3' LG Pitot tube, 1/8" OD R134a - • Surface Temperatures • Outlet Bulk Temperature • Outlet Pressure • Outlet Mixing Temperature - • Inlet Mixing Temperature • Inlet Pressure • Inlet Bulk Temperature 60 A few of the design characteristics of the inner tube are analysed in the following paragraphs. These were considered the most relevant in the design process of the test -section. 5.2 Heat flux The maximum heat flux generated by the heater, can be calculated with: where q"= heat flux [W/m2], P= electrical power [W] and A= heater surface area [m2] Based on the measured power during experiments, the maximum heat flux generated by the heater is approximately 200 kW/m2. During the subcooled boiling experiments, verifications of the energy balances were undertaken in order to assess the accuracy of the measured heat flux and temperature readings. Details are presented in Chapter 7. 5.3 Electrical resistance It is important to know the resistance of the test section in order to asses how much of the heat input into the test section is dissipated through the Inconel (SS 625) portion, where the measurements are done, and how much through the copper tube connections. For a cylindrical element the resistance can be calculated as follows: q"=P/A, [W/m2] (5.1) (5.2) where: p = resistivity [u.Qm] 1 - length [m] A = area [mm2] 61 The resistivity at a given temperature "t" can be calculated as: p,=p20[l + a(t-20'C)\ (5.3) where: a = thermal expansion coefficient [1/ °C] The results of the calculations for the expected temperature range (see Appendix no. 2) are given in Table 5.1 Table 5.1.Electrical resistance of test section components Tube Material Resistance at 20°C [Q] Resistance at 60°C [Q] Inconel 16.427 x 10"J 16.435 x IO"3 Copper 0.448 x 10"4 0.518 x 10"4 It can be seen that the electrical resistance of the inconel portion of the heating element is more than 300 times higher than the resistance of the copper connections. Therefore we may consider that the power (heat) dissipated through the copper portion of the heater is negligible compared to the amount of heat generated in the inconel portion. 5.4 Hydrodynamic entry length The heat generated in the test section is removed by the refrigerant flowing through the annulus. In order to assure a better heat transfer it is desirable to develop an ideal turbulent flow behaviour. Moreover, the effects from the radial inlet velocities at the entrance (bottom) of the test section have to be compensated. For average pipe velocities corresponding to turbulent flow Reynolds numbers (>2100), the entrance effects disappear about 10 to 20 diameters from the inlet [18]. 62 The hydraulic diameter Dh of the test section is defined as for most of the hydraulic problems as: Dh=4 x Flow Cross-Sectional Area / Wetted Perimeter (5.4) For our particular case the relevant dimensions are as follows: Heater OD: 12.7 mm (1/2 in) 3.175 mm (1/8 in.) 38.1 mm(l Vi in.) Pitot-tube OD: Glass tube ID: The hydraulic diameter, based on the wetted area, is 23.72mm. Furthermore the hydraulic entry length is found to be 237.2 to 474.4 mm. The entry length was chosen approximately 350mm, based also on the overall length of the test section heater. 5.5 Outer tube The actual design of the test section allows for visualisation of the flow boiling. This contributes to a better understanding of the boiling process and bubble behaviour. In support of quantitative analysis, data can be collected by means of photographic and high speed filming. The selection of the glass tube was mainly based on the working pressures in the test section. The maximum rated pressure for the glass tube is 180 psig (safety factor included), as confirmed by the scientific glass company, who supplied the component. 63 5.6 End cap fittings The assembly and sealing of the main components of the test section are achieved by specially designed fittings. The top and bottom fittings are machined from solid brass (see Appendix no.l). Teflon sleeves with Buna-N "O" rings are employed to provide the proper seal around the glass tube and along the inside face of the fitting. For sealing the copper connectors of the test section Teflon compression ferrules were employed. They provided a good seal without deforming the test section and allowing for replacement when required. This assembly arrangement ensured that the current carrying test section is electrically segregated from the rest of the loop. As well the O-ring contact on the glass tube surface allows for small displacements due to different coefficients of expansion between glass tube and heating element. The top and bottom fittings were also equipped with two 3/8" Swagelok compression fittings, which connected the test section into the hydraulic circuit through clear Teflon tubes, allowing to monitor the state of the fluid at the discharge. The tubes were mounted 180° apart, so that the entrance swirl effect is reduced as much as possible. 64 Chapter 6 Instrumentation and Control 6.1 Flow Measurement and Control For steady state flow measurement, both the Brooks rotameter (model 1112A) and the Omega FTB 101 turbine flow meter can be used. Calibration consisted in timing the flow rate into graduated vessels. Several runs were performed, covering the entire range of desired flows. The calibration charts for both flowmeters are found in Appendix no. 3. As seen in the calibration chart, the rotameter can be accurately used for flow ranges between 0.1 and 2.5 1/min. (0.078-0.78 GPM), corresponding to low flow conditions for the current set of experiments. The Omega turbine flowmeter is accurate within the range of 0 to 3.4 GPM (12.9 1/min.). The flowmeter calibration showed a linear response up to 4 GPM (15 1/min.), see Appendix no. 3. The flow range covers the complete mass flow spectrum for the proposed subcooled and saturated boiling experiments. The millivolts output signal of the turbine flowmeter conditioner was converted into flowrate by the data acquisition board and displayed on the computer monitor, as described in a subsequent paragraph. The system flowrate can be adjusted by a Whitney needle valve near the inlet into the test section or by regulating the pump RPM controlled by the Variable Frequency Drive. In practice was found that this second method provided a more precise control of the flowrate, to an accuracy of 0.1 GPM. 65 6.2 Power measurement and control The power input into the heated test section can be determined from the product of the voltage drop across the test section and the electric current passing through it. The voltage drop was measured by means of a Omega True RMS voltmeter, model DPI 8-RV2, with a range of 0-20 VAC, a resolution of 10 mV and an accuracy of 0.2% of the full scale. The current passing through the power leads was first reduced by an Omega Hall Effect transformer model CTL-156050 with a 500:5 ratio and an accuracy of 1.0%. The current reading was displayed by an Omega True RMS ammeter model DP-18-RT1, providing an accuracy of 0.2% of the full scale. The power input to the preheater is not required for the data reduction, therefore, no measurements were taken for this unit. The preheater power control variac dial was adjusted manually (between 0-240 V) until the desired fluid temperature at the inlet into the test section was established. The power regulation of the entire system was controlled by a General Radio Model 1582-A voltage regulator in conjunction with the two W50H variacs as previously described. 6.3 Pressure measurement and control The measurement of the static system pressure is extremely important for these experiments since the saturation temperature for each run is derived from this pressure. Since, as already mentioned, the surface temperature was measured at 6 different locations along the test section and it was desirable to have also the possibility of measuring the pressure at each location. Therefore, a long Pitot tube of 1/8" OD with two static pressure measuring stations was designed, see Appendix 1. The tube was mounted inside the test section, in the annular section, parallel with the heated tube. The connections through the top and bottom fittings were sealed with Swagelok fittings 66 equipped with nylon compression ferrules. The additional length at both sides allowed for sliding the tube in parallel with the test section, in such way that the top orifices could be aligned with the top three thermocouple locations and the bottom orifices with the remaining three bottom locations. This would allow for minimising the errors in determining physical property and local heat transfer. The Pitot tube was connected at each end to a Setra 204 pressure cell, via a 3.5 mm ID Teflon tube. The two ends were connected as well to a Validyne differential pressure cell, with a bypass circuit for protection. A bleed valve used to release trapped vapours within the tubing was also connected to the transducer. The pressures cells, equipped with individual DC power supplies, were connected to the data acquisition board, which converted the signal to a flow rate reading displayed on the computer monitor. Prior to and during the experimental stage the pressure transducers were calibrated with an Omega PCL-200 series portable calibrator with a pressure range of 0.0-500 kPa, (0-72 psig) and a specified accuracy of 0.25 kPa (0.036 psi). The Omega PCL-200 provides a simultaneous dual display indicating pressure and electronic transducer output on eight key selectable engineering units. The calibration kit includes also a precision PCL-2HP hand pump. The pressure calibration set-up is shown in Figure 6.1. Using the hand pump, pressure was applied in small increments of the voltage output of the transducer. The pressure corresponding to each voltage increment was recorded. The calibration data sheets are presented in Appendix no. 3. The liquefied nitrogen pressurizer unit was used to control the system pressure during experiments. The adjustment of the system pressure was achieved by operating the pressure-regulating valve on the liquefied nitrogen tank. 67 Figure 6.1 Electronic voltage transducer calibration set-up C O M STATIC PRESSURE OR DIFFERENTIAL PRESSURE TRANSMITTER LT LOOP SUPPLY mAmps PCL-200 H A N D PUMP PCL-2HP 6.4 Temperature measurements During the design process of the test section we have established that several temperatures have to be recorded in order to provide in depth information about the boiling process. For all locations we have used K-type thermocouples, the most suitable for our application. The heating element of the test section was equipped with 6 K-type thermocouples welded on the surface, Figure 7.3. The wires were guided through the inner tube (heating element), to avoid disturbing the flow profile. In order to measure the surface temperatures, holes were drilled on the surface of the heater, and the junctions silver-soldered as close as possible to the surface. The wires are isolated inside the heater and the end of the heater is filled with epoxy resin to minimize an accidental refrigerant leak. 68 The mixing cup temperatures were measured with sheathed 1/16" K-type ungrounded thermocouples mounted in the top and bottom test section fittings with specially modified Swagelok fittings. The thermocouples were aligned with the flow and centred in order to assure a good accuracy of the readings. The bulk temperatures were measured with sheathed 1/64" K-type ungrounded thermocouples, specially manufactured with a length of 3 feet. The thermocouple lengths were conducted through inside of the Pitot tube (1/8" OD) and then guided out through one of the pressure tap holes. The length outside the Pitot tube orifice (inside the test section) was determined so the junction is close to the centre of the annulus. The connection outside the Pitot tube was achieved with Swagelok fittings with a modified Teflon ferrule to ensure a leak-free connection. Since the Pitot tube measuring stations have the flexibility to slide in front of each surface temperature thermocouple, bulk temperature as well can be read at each test location. This could assure that the measured values are actual bulk temperatures and pressures corresponding to each surface temperature, and no interpolations would be required. The temperature values were measured (scanned) with a hand-held Omega temperature reader, HH-20 series. This temperature reader is equipped with dual inputs, which allowed us to continuously monitor the inlet temperature, while scanning the surface temperatures. The temperature reader assures a resolution of 0.1 °C throughout the entire range and an accuracy of 0.1 % of the reading. Calibration of thermocouples was performed using a Thermo Electric Model 136 Ice Point Reference Chamber for the 0.0°C reference. The ice point cell contains six thermoelectric cooling modules with a specified accuracy of 0.0+/-0.05°C and a stability of +/-0.01°C. In order to calibrate the thermocouples at higher temperatures, boiling water was chosen for convenience. 69 6.5 Data acquisition system The data acquisition system consisted of a DAS-1602 high-speed analog and digital I/O board, connected to a 486 series computer. The major features are outlined in the following table: Table 6.1 DAS-1602 Data acquisition hardware features Analog Inputs: 16 single-ended or 8 differential Range Selection: Programmable Maximum Throughput: 100 kS/s Gains: 1,2,4,8 Resolution: 12 bits D/A Channels: 2 Input:, Unipolar Bipolar 0to+10 V +/- 10 V Digital I/O Lines: 32 For connecting the signals we used STA-16 screw terminal board and attempted to use EXP-16, a 16-channel multiplexer accessory board, with selectable gain, filtering, cold junction compensation and electrical isolation. The data acquisition software did not recognise the expansion board as part of the compatible hardware and therefore we operated the system exclusively with the STA-16 board. A typical configuration is presented in Figure 6.2. Figure 6.2 Data acquisition system hardware configuration DAS-1602 & 486 PC STA-16 EXP-16 EXP-16 C-1800 S-1600 S-1600 The data acquisition system was used for displaying and recording the turbine flow meter reading, the inlet and outlet system pressure and the differential pressure across the test section. 70 Chapter 7 R-134a Subcooled Boiling For investigating the heat transfer behavior of R-l34a, an ozone-friendly refrigerant, under forced convective sub cooled boiling conditions, a large experimental data bank was acquired with the use of purposefully built experimental rig, in UBC Mechanical Engineering department. The experiments were performed at system working pressures of 80 to 160 psig in increments of 20 psi. This corresponds to an absolute pressure range of 94.7 to 174.6 psia or 45 to 83 kPa, respectively. The flow rates were in range of 1.0 to 4.0 GPM with increments of 0.5 GPM, corresponding, based on cross-section area to mass fluxes of 78.9 to 309.06 kg/m2s. The heat fluxes covered a range of 3 kW/m2 to 200 kW/m2. The inlet temperatures of the refrigerant were between 10.1 and 18.4°C. The developed data bank for R-l34a subcooled boiling consists of 1872 data points from 34 experimental runs. 7.1 Experimental procedure The experiments were carried with decreasing heat flux in order to avoid the hysteresis effect observed prior to this set of experiments. In a typical run, the inlet temperature, flow rate and system static pressures were held constant while decreasing the heat flux in small steps. First, the test section was brought to vigorous boiling while controlling the outlet subcooling by means of pressure-temperature readings and visual observations. The measurements were taken after steady state conditions were achieved, at each power input setting. The system was considered at steady state when pressure fluctuations were less than 0.3 psi and the temperature reading fluctuations at the inlet and outlet of the test 71 section were less than 0.1°C. In general the steady state conditions were achieved 3 to 5 minutes following the power input change. Within each experimental run we measured the system inlet and outlet pressure with the pressure cells. The outlet pressure was measured in parallel with the Heise pressure gage. The flow rate from the Omega turbine flow meter as well as the system pressures were read on the data acquisition computer monitor. The atmospheric pressure was read on the existing barometer in the wind tunnel lab. The inlet/outlet bulk temperatures, inlet/outlet mixing cup temperatures and the surface temperatures were measured with the hand-held Omega temperature reader. 7.2 Data validation An analysis of energy balances for each run was performed in order to assess the accuracy of the temperature-power input readings. To accomplish that, the predicted temperature gain along the test section was calculated as: AT V x I n n G x , 9 , x tp, where V is the voltage input (W), I is the current input (A), G is the flow rate (m Is), pi is the liquid density (kg/m3) and Cpi is the liquid specific heat (J/kgK). The liquid density and the liquid specific heat were estimated at the mixing cup temperature between inlet and outlet of the test section, for each different power input. We assumed here that since we deal with subcooled boiling all the bubbles are condensing in the bulk fluid and their latent heat is transferred back to the fluid contributing to the overall increase of the bulk temperature. This was practically confirmed by our visual observations of the boiling process, at the discharge from the test section. The mixing cup temperature difference was found to be more accurate compared to the predicted temperature difference, than the bulk temperatures. This may be due to the fact 72 that some heat from the inconel portion of the test section was conducted into the copper connecting rods. Since it would be difficult to accurately estimate how much of the heat was "lost" through conductance we decide to further proceed with our analysis based on the mixing cup temperature readings. The average and mean deviations were calculated, as follows: 1 ^  (Estimated Value - Measured Value) x 100 Average deviation - —y, n i 1 " i Absolute mean deviation = — ^ Measured Value (Estimated Value - Measured Value) x 100 Measured Value (7.2) (7.3) It was found that for all the subcooled data the average deviation is -1.97% and the absolute mean deviation 7.72%. This was considered an acceptable agreement. However, at low heat fluxes the deviations were found to be larger due to limitations in temperature measurements. These data points were eliminated from any further calculations. Some typical variations of measured vs. predicted temperature differences are presented in Figure 7.2. We have also checked the relationship between measured pressure and temperature at saturation conditions. The data were taken after the system was at rest for a minimum of 16 hours and then we slowly increased the system temperature with the test section heater or alternatively with a heat gun. The measured data was compared with the DuPont reference data for R-l34a, and found that bulk temperature measurements are more consistent. This can be explained by the fact that the mixing cup thermocouples were located in the end caps of the test section and heat being radiated from the brass bodies led to a small drop in temperature. The readings from the bulk thermocouples were used as a cross-reference check for measured operating static pressures at rest. The results are shown in Figure 7.1. 73 74 Figure 7.2 Data validation (selected) - measured vs. predicted temperature differences 1.0 G P M , 100 PSIG, 10.0°C Inlet temperature, 0.096 Absolute mean deviation 0 0 ~y 0 ^y 0 ^y 0 * ^y * y^ 0 0 ^ 0 0 ^y «•» / * ^y 0 ^y 0 ^y * ^y predicted - measured * 10000 20000 30000 40000 50000 60000 80000 Heat Flux, W / m ' 75 7.3 Data Reduction The heat transfer coefficient was defined as in the following equation: 71 -T„ (7.4) wall bulk 2 where q" represents the heat flux (W/m ) calculated from the power readings divided by the surface area of the heater and Twau is the surface temperature measured at each of the six thermocouple locations along the test section. Ttnik is the bulk temperature of the fluid, estimated through linear interpolation between the inlet and outlet mixing cup temperature readings. This appears to be a valid assumption since a plot of liquid enthalpy of R-l 34a against temperature shows a perfect linear variation. The bulk temperatures were calculated as follows: where Tbx represents the bulk temperature at each thermocouple location, and x the heated length (mm) for each location. These are for thermocouples 6 to 1 (inlet to outlet) 40, 120, 200, 280, 360 and 440 mm, respectively (see Figure 7.3). Tbx=Tmi + x/480(Tmo - Tm$ (7.5) 76 77 7.4 Data processing In order to assess the performance of newly proposed heat transfer fluids, all of the research experimental programs that we have examined were able to collect data for both refrigerants, the replaced one and its replacement. The replaced refrigerant data provided a base line for analysis and therefore facilitated a lenient comparison of the behaviour under same conditions and geometry of the test sections. The performance of the new refrigerant can then be easily expressed as an increase/decrease in heat transfer, COP or other indicators. In our case we were limited in this respect. Since R-l34a replaces R-l2 (CFC) it would have been relevant to directly compare both refrigerants performing under the same conditions and using the same test section. Unfortunately, the legislation controlling the use of CFCs is very strict with the handling procedure. Only a certified technician is allowed to purchase and work with CFCs. Further, the design and construction of the test rig would have to go through a complex process of testing and certification. R-l34a, being ozone-friendly, does not impose any of the above "inconveniences". Since we had no access to any subcooled boiling heat transfer data for R-l34a, we chose to compare the experimental data with existing predicting correlations. A review of existing correlations for predicting forced convective boiling heat transfer was undertaken. It was found that for subcooled boiling the Chen (1961) correlation predicts the heat transfer coefficient with the highest accuracy, followed by Gungor correlation (1987). Liu and Winterton (1991) correlation presumably provides an even better fit but other researchers, according to our sources have not verified this fact. Chen's correlation is also the most widely quoted in textbooks and scientific publications. 78 7.4.1 Chen's correlation applied to subcooled boiling Considering that the Chen's correlation was developed for saturated boiling, we followed Collier's recommendations [5] as how to use the model for the subcooled boiling regime. Collier's notes are very brief (one third of page) and referred to a private communication with D. Butterworth, dated 1970. As suggested by Collier (1981), it can be assumed that the total surface heat flux is expressed as: <P = KB (T»all ~ Tsa, ) + Kc (T»all ~ Thulk ) (7-6) The value of the single-phase forced convection heat transfer coefficient, hFc (hmac), was obtained from the Dittus-Boelter equation setting F equal to unity. 0.023Re? 8 Prrf^-V (7.7) The nucleate boiling coefficient h m (hmic) was evaluated from the Foster-Zuber equation with the value of S calculated on the basis that Rerp = Rei using the Collier correlations, or using the charts. h t =0.00122 ( lr ° - 7 V 0 - 4 5 / - I 0 " 4 9 ^ Ki u Pi Pi 0.5 0.29 i 0.24 0.24 *T™6p™S (7.8) Values of wall temperature are assumed for calculating the nucleate boiling coefficient through an iterative procedure. All properties were calculated at saturation temperature and based on ICI Chemicals and Polymers (1996) data file for KLEA 134A, Appendix no. 6. The calculated values were 79 verified with table readings and found to be satisfactory. The first step in performing the calculation was the evaluation of liquid Reynolds number. In the Chen's correlation for saturated boiling the liquid Reynolds number is dependent on quality x as expressed by the equation: G x (1 - x) x d i v a h n l Re, = (7.9) M, Collier expresses the liquid Reynolds number with the quality equal to zero in the above equation. Practically, the forced convection coefficient was calculated as for single-phase heat transfer and the equivalent diameter for the annulus was calculated taking the heated perimeter instead of wetted perimeter, as suggested by Chen as well. Prandtl number was evaluated with the properties at the saturation temperature as follows: Pr,= MiCp,/k, (7.10) Then the liquid only Nusselt number and single-phase heat transfer coefficient were calculated (equation 7.7). The forced convection coefficient hpc is found as equal to hi since F=l as suggested by Collier. The next step was calculating the nucleate boiling heat transfer coefficient, using Foster-Zuber correlation 7.8. The suppression factor S was evaluated from chart or alternatively from the equations proposed by Collier [5], with the liquid only Reynolds number instead of two-phase Reynolds number. The Collier's suggested correlation for S used here is as follows: 80 The Foster-Zuber correlation contains a factor AT s a t = T w a M - Tbuik and Apsat, which is the difference in saturation pressure corresponding to AT s a t. The iterative procedure consisted in guessing a wall temperature, then calculating the AT s a t using the bulk temperature estimated from an energy balance. Further Apsat was calculated and then hN B (equation 7.8). Then the overall boiling coefficient was found as he = hfc + hm • Finally, a new value of Twall was calculated from the heat flux correlation and replaced in the formula for calculating the nucleate boiling coefficient. The iterations will converge when T w a M guessed approaches Twan calculated. For expressing the heat flux Collier recommends equation (7.6) which implies that the whole of the actual temperature difference {Twall - Tbulk) goes into the convective heat flux, and only a part of it into the nucleate component. We tried using this approach with the bulk temperatures as predicted by energy balances for each case, without a satisfactory result. The results were over-predicting the heat transfer coefficient 3 to 7 times, with higher errors corresponding to the higher range of heat fluxes. An explanation could be that the measured heat transfer coefficient calculated from equation (7.4) is very dependent on the bulk temperature of the fluid, dependence that is not accurately modeled by equation (7.6). As mentioned before, during the computational process we have used the suppression factor S suggested by Collier (eqn. 7.11) which models very accurately the original S chart developed by Chen (Figure 2.1). However, Chen's original experimental data was for saturated boiling only, therefore an extension to subcooled boiling would require a different expression for S. In fact, Gungor and Winterton [21] who suggested a similar correlation for subcooled boiling (eqn. 2.23), proposed a different expression for the suppression coefficient S, based on their experimental data bank. 81 Also, when equating the expressions for the experimental heat flux to the predicted heat flux one can find that: (Kc +KH)X (T„all ~ Thu,k ) = Kx: (T*all - Thulk ) + KH (Lall ~ Tsal ) (7-12) and after further reductions: Tsal = Tbulk (7.13) This is true only for the saturated boiling only. In order to be consistent with our experimental procedure we chose to use for the final calculation of Tw an, the heat flux equation that we used for calculating the measured heat transfer coefficient: 0 = (hFC + hNB) x (TwaU - Tbuik) (7.14) This new approach lead to satisfactory results. 7.4.2 Liu & Winterton correlation applied to subcooled boiling As part of the data analysis for the subcooled boiling experiments, we have also compared our results with the predicted heat transfer coefficients by the Liu and Winterton correlation. As already outlined in Chapter 2, the heat transfer coefficient is calculated with: " h]v=(FxhL)2+(Sxhpool)2 (7.15) In the above equation the liquid only heat transfer coefficient was calculated with Dittus-Boelter correlation (7.7) with a forced convection factor as defined by those authors as: 82 F = 1 + xPr \Pv J 0.35 (7.16) Similarly with Colliers suggestions for subcooled boiling application of Chen's correlation, the factor F becomes equal to zero, since the quality is taken equal to zero. The nucleate boiling term was calculated from Cooper's boiling correlation: 0.67 (7.17) where Pr is the reduced pressure, Mthe molecular weight and q the heat flux. The suppression factor S found by Liu and Winterton was expressed as: S = (l + 0 .055 0 , Re° , 6 )~ (7.18) All the properties were estimated again at saturation temperature from the ICI Chemicals and Polymers (1996) data fde for KLEA 134A. The procedure was found to be very convenient to use for our large data bank with acceptable results. 7.5 Experimental results for subcooled boiling experiments Out of the 1872 data points collected during the experimental investigation, 720 were analyzed. These data points corresponded to pressures of 80, 100, 120, 140 and 160 PSIG and flow rates of 1, 2 3 and 4 GPM. The low heat flux data, proved to be less accurate through energy balances, were excluded. The temperature readings from the 83 surface thermocouple #4 were not consistent with the rest of the data and therefore will not be considered in our analysis. Chen's procedure was found to be tedious due to the iterations performed for each point. The iterations were performed until the difference between the guessed value of Twall and the calculated value agreed to less than 0.3 degrees. In comparison, the Liu & Winterton correlation was found convenient and easy to use, especially with a large number of experimental runs. A selected set of experimental results is presented in Figure 7.4. The trends of the heat transfer coefficient vs. heat flux, for the test section locations corresponding to the surface thermocouples, are identical for all the tested conditions. As expected, we observe that for a constant heat flux, while the subcooling (Twall -Tbulk) decreases as the fluid traverses the heated test section, the heat transfer coefficient constantly increases from inlet to discharge. As well we can note that with increasing heat flux while keeping the inlet temperature constant, the value of the heat transfer coefficient at each location increases linearly. For analyzing the influence of system pressure and mass flow rate on the heat transfer coefficient we choose to observe the variation only for the location corresponding to thermocouple # 6, situated at the inlet into the test section. Here the value of the bulk temperature is closer to the measured inlet mixing cup temperature, controlled during experiments, and therefore more reliable. The experimental and predicted results are presented in Figure 7.5. 84 S-i E =3 CO CD o o o o o o o O o o o o o o o o o O o o o in o m o in o in o in in •sr TT CO CO CM CM 85 Figure 7.5 Subcooled boiling selected experimental results 3000 ^ 2 5 0 0 CM E C2000 CD O it= CD o Q1500 a CO c CO 1000 CO CD X 500 Subcooled Boiling, 1.0 GPM, 80 PSIG, 10.8°C Inlet Temperature, Location: Surface Thermocouple #6 . —o— Exper imenta l Liu & Winter ton Cor re la t ion C h e n ' s Cor re la t ion 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 Heat Flux, W/m 2 4500 4000 3500 CM E 3000 •£= 2500 CD o 2000 o o CD 1 5 0 0 CO c 2 1000 CO CD 500 X Subcooled Boiling, 3.0 GPM, 80 PSIG, 13.7°C Inlet Temperature, Location: Surface Thermocouple #6 —o— Exper imenta l Liu & Winter ton Cor re la t ion C h e n ' s Cor re la t ion 10000 20000 30000 40000 50000 60000 Heat Flux, W/m 2 70000 80000 9000° 86 Figure 7.5 - continued 4000 3500 CM E c CD o i t CD O O CO c CO co CD X 3000 2500 2000 1500 1000 500 Subcooled Boiling, 1.0 GPM, 120 PSIG, 10.5°C Inlet Temperature, Location: Surface Thermocouple #6 • Exper imen ta l - - Liu & Winter ton Cor re la t ion • — C h e n ' s Cor re la t ion 20000 40000 60000 80000 Heat Flux, W/m 2 100000 120000 4500 4000 3500 CM E 3000 § 2500 o 51= O 2000 o CD • f t 1 5 0 0 c CO 1— I— 1000 •4—> CO CD X 500 Subcooled Boiling, 3.0 GPM, 120 PSIG, 14.0°C Inlet Temperature, Location: Surface Thermocouple #6 Exper imenta l Liu & Winter ton Cor re la t ion C h e n ' s Cor re la t ion 20000 40000 60000 80000 100000 Heat Flux, W/m 2 120000 140000 87 Figure 7.5 - continued 7000 6000 CM E 5000 c CD O <D O o 4000 £ 3000 2000 CO x 1000 Subcooled Boiling, 1.0 GPM, 160 PSIG, 11.3°C Inlet Temperature, Location: Surface Thermocouple #6 —o— Exper imenta l Liu & Winter ton Cor re la t ion C h e n ' s Cor re la t ion 50000 100000 150000 Heat Flux, W/m 2 200000 250000 CM 7000 6000 5000 c O i t 0 ) o o (/> C CO !~J 2000 CO OJ X 1000 4000 3000 Subcooled Boiling, 3.0 GPM, 160 PSIG, 16.2°C Inlet Temperature, Location: Surface Thermocouple #6 • Exper imen ta l Liu & Winter ton Cor re la t ion C h e n ' s Cor re la t ion 50000 100000 150000 Heat Flux, W/m 2 200000 250000 8 8 For assessing the general accuracy of our calculations we compared the measured values of the boiling heat transfer coefficient with the values predicted by the Chen and Liu-Winterton correlations. The accuracy of tested correlations for the selected test location and experimental conditions is graphically presented in Figure 7.6 and in a condensed form in Table 7.1. Table 7.1 Accuracy of the tested theoretical correlations Correlation Total data points Data Points @ Location #6 Average Deviation, % Absolute Mean Deviation, % Deviation less than Percent of Total, % Liu-Winterton 1068 178 -9.0 10.1 16% 79.2% Chen 732 169 15.9 19.2 19% 80.5% 7.6 Discussion and conclusions Both correlations, Chen and Liu-Winterton, predict the subcooled boiling heat transfer coefficient with an accuracy acceptable for engineering calculations, considered +/- 25%. Based on our experimental data, Chen's correlation overpredicts the subcooled boiling heat transfer coefficient. As illustrated in Figure 7.5, this trend increases with increasing operating pressure. Analyzing Figure 7.6 one can notice that Chen's correlation is more accurate for higher mass flow rates (3 GPM), however the scatter is relatively narrow. The heat transfer coefficients predicted by Liu and Winterton correlation are consistent with our experimental data. The results slightly underpredict the heat transfer coefficient, especially at lower operating pressures, while no influence of flow rates is noticed (Figure 7.5 and 7.6). However, the general accuracy and ease of implementation make the correlation a good candidate for heat transfer calculations. 89 90 91 1 Chapter 8 R 134a Saturated Boiling The saturated boiling experiments with R 134a were performed as part of the experimental investigation of the heat transfer behavior of an alternative refrigerant. The objective was to extend the observation from subcooled boiling to saturated boiling, a regime widely encountered in engineering applications. 8.1 Experimental procedure and data The saturated boiling experiments were conducted in increasing heat flux mode, while keeping the flow rate constant. During the experiments a slight increase in system (saturation) pressure and inlet temperature were observed. Compensating this increase was found to be difficult due to a low condensation / cooling capacity of the system. Therefore, a limited number of data were collected during this stage of experiments. Specifically, for 1.0 GPM thirty-one saturated data points were recorded and sixteen data points (only two thermocouple locations) for 2.0 and 3.0 GPM. The experimental conditions covered for saturated boiling are presented in Table 8.1. Table 8.1 Experimental conditions for R 134a saturated boiling Flow Rate Mass Flux, kg/m2s Current, Amps Heat Flux, kW/m2 Inlet Temp., °C System Press., PSIG 1.0 GPM 73.5-72.6 500 - 700 77.3 - 153.6 34.1 -37.6 118-139 2.0 GPM 151.9-151.4 560 - 640 94.6-216.6 24.2-25.3 98-107 3.0 GPM 220.9-219.9 760 - 860 175.2-225.2 33.6-35.0 138 - 149 93 The measured heat transfer coefficient was calculated as: wall •T... (8 .1) 8.2 Experimental results The experimental results for 1.0 GPM flow rate are summarized in Figure 8.1. The variation of the heat transfer coefficient with the location along the test section and with the calculated quality respectively, shows a general ascending trend with an inflection point. Similar results were obtained for 2.0 and 3.0 GPM, with less saturated data points. Figure 8.1 Measured heat transfer coefficient for 1.0 GPM vs. predicted quality E c CD O it= 0 o O i _ cu c ro ro CD X T3 CD i _ 13 CO ro CD 1300 _ Saturated Boi l ing, 1.0 G P M 1200 ._ 100 10000. _ 900 /f / v_ ^-77.3 kW/m 2 8O0V_ / -a-111.8 kW/m 2 /7000 _ , 1— 1 - A -134 .4 kW/m 2 ^ 1 5 2 . 8 kW/m 2 1 -0.03 0.02 0.07 Quality, x 0.12 0.17 The measured heat transfer coefficient displays a peculiar behavior within the range of qualities covered by experiments. One can notice a strong dependence of the heat transfer coefficient on the heat flux, followed by a drop / plateau region. 94 As noticed by Thome [36], for horizontal flows, some experimental programs found a substantial peak in heat transfer coefficient as a function of quality, attributed to the type of two-phase flow regime encountered and to whether the heated surface is totally or partially wetted by the flow. The visual observations of the boiling process confirmed that the saturated boiling was very intense, with back flows induced by the vapors trapped in the test section, and therefore the possibility of having a partially wetted surface can not be ruled-out. Wattelet et al [39] have studied the effect of flow pattern on the heat transfer of R-l 34a and R-l 2 flowing inside of a horizontal heated tube, defining two flow regimes, annular and stratified-wavy flow. Their results are shown in Figure 8.2. Figure 8.2 Flow boiling data of Wattelet et al [39] at 5°C for R-l 34a o c CC CO A o o (a) 0 R-l 34a <Q 5 kW/m 2 • 10 kW/m 3 • 20kW/m 2 • 30kW/m 2 02 0.8 0.4 06 Vapor quality a) Annular flow at 300 kg/m2s o o C3 1 5 12 0.<» 0.6 03 0.2 (a) R - l 34a in i 0 4 Q 2kW/m-A 3 kW/m 2 • S kW/m 2 0.6 0.8 Vapor quality b) Stratified-wavy flow at 50 kg/m2s 95 The boiling regime was assessed based on flow observations at the discharge of the test section. The magnitude of the heat transfer coefficient is lower than measured during the UBC experiments, a fact which may be attributed to the different experimental conditions, including test section orientation, see Table 8.1. While the mass flux for our experiments (73 kg/m2s) is comparable to the Wattelet et al experiments (50 kg/m2s), the heat fluxes we used were much higher (77 - 153 kW/m2), as required by our higher operating pressures and inlet temperatures. Also, J.Y. Shin et al [32] in their horizontal saturated boiling experiments with pure refrigerants and refrigerant mixtures have obtained similar results, see Figure 8.3. Figure 8.3 J.Y. Shin et al [32] heat transfer coefficients for pure refrigerants. 1 0000 I 8000 J 6000 8 4000 U— I 2000 0 ca cu X i V V V A A — o r, V* A V A A — • o -V O R22 V • 0 R32 R134a 1 A R290 R600a i .0 0.2 0.4 0. 6 0.8 1 Vapor quality a) Mass flux=424 kg/m2s, Heat flux=30 kW/m2, Outlet temperature=12°C E 1 0000 8000 •a 6060r-S 4000 L-i <u | 2000 ca r 0 1 — T — r —1 v4a® o -v v o • R32. « R134a — & R230 V RSOQa i .0 0.2 0.4 0.6 0 8 1 Vapor quality b) Mass flux=583 kg/m2s, Heat flux=30 kW/m2, Outlet temperature=12°C 96 Compared to our experimental conditions, the refrigerant mass flow rates (424 - 742 kg/m 2s) were significantly higher, while the heat fluxes (10-30 kW/m 2 ) were lower. The cooling capacity of the system allowed Shin and his collaborators to run the system at lower temperatures and therefore lower operating pressures. The discharge from the test section was kept constant at 12 (+/-0.5)°C, which corresponds to a saturation pressure of approximately 28 psia. A s they have observed, a strong dependence of the heat transfer coefficients on the heat flux appears in the low quality region. The nucleate boiling is dominant in the initial stage of evaporation, especially for high heat flux conditions. The relatively great heat transfer coefficient in the initial stage decreases as the nucleate boiling effect diminishes. Heat transfer coefficients increase again when the convection effect is prevailing due to an increased velocity of both liquid and vapor phases, assuming no slip. 8.3 Analysis of saturated data against theoretical correlations The comparison of the data with results from other studies is difficult due to insufficient published experimental data and high number of variables involved. These may include different geometry and test conditions, including flow orientation. These limitations led us to attempt comparing the data with predictive correlations. There are very many correlations available for modelling the boiling heat transfer, more or less general and accurate. We have chosen to try Chen's correlation, which is one of the first developed and definitely the most cited by the literature, and the Gungor and Winterton correlation, a relatively new model, which claims to be more accurate and general. 97 8.3.1 Chen's correlation applied to saturated boiling For predicting the saturated boiling heat transfer coefficient, Chen's correlation was used similarly as described for subcooled boiling calculations. Compared to the subcooled boiling calculation procedure of the heat transfer coefficient, for saturated boiling the final verification is done, as stipulated by Chen in his work using: K = T q _ T (8-2) 'wall .sal where q" is the heat flux (W/m2), Tw an is the wall temperature and T s a l is the saturation temperature. The quality was estimated from an energy balance. The factors S and F were calculated with the equations proposed by Collier, and verified with the original charts. A good approximation was found, therefore for the ease of computation, the analytical forms were used. An example of Chen's correlation for saturated boiling from Whalley's "Boiling, Condensation and Gas-Liquid Flow" [40] was consulted. 8.3.2 Gungor and Winterton (1986) correlation applied to saturated boiling The models developed by K.E. Gungor and R.H.S. Winterton were tested against our saturated experimental data. In this procedure, the two-phase heat transfer coefficient is expressed as: hlp=Ehl+Shpool (8.3) First, the liquid only heat transfer coefficient was calculated with the Dittus-Boelter correlation (equation 7.7), where the forced convection enhancement factor is replaced by a factor E, expressed as: 98 ,0 .86 E = 1 + 2 4 0 0 0 Bo i A 6 +1.37 \ / X n J (8.4) In the above equation, X„ is a factor defined by the quality of the mixture and by the liquid - vapour properties, called the Martinelli parameter. \-x V x ) , 0.9 \05r \ 0 A Pi (8.5) Bo is the boiling number, defined as: Bo = % hf8xG (8.6) Further, the nucleate boiling heat transfer coefficient was calculated using the Cooper correlation (equation 7.16). The suppression factor S was calculated as suggested by authors using the following correlation: l + 1 . 1 5 x l 0 - 6 £ 2 R e J 1 7 (8.7) A l l the properties used in the above correlations were estimated at saturation temperature. A s suggested by authors for annular gaps larger than 4 mm (our gap is 12.7 mm.), the hydraulic diameter used in the Re number was based on the wetted perimeter. This approach lead to predicted values of 320 to 700% higher than the measured heat transfer coefficient. Further we tried using the same equations with the hydraulic diameter based on the heated perimeter (Dew c U Cd=0.02372 m, DehCatcd =0.100806 m). The results were significantly improved. However, the correlation overpredicts the saturated boiling heat transfer coefficient, see Figure 8.4. 99 8.3.3 Gungor and Winterton (1987) correlation applied to saturated boiling The revised Gungor and Winterton correlation (equation 2.24) was easily implemented on the existing spreadsheet program. As specified in the 1987 paper, the healed perimeter was used in the equivalent diameter. All properties were again estimated at saturation temperature. The correlation was found to under-predict the heat transfer coefficient, see Figure 8.4. 8.4 Results The results for the tested correlations against the saturated boiling data for one thermocouple location (# 1 -discharge) are presented in Figure 8.4. Chen's correlation generated satisfactory results, the deviation was found to be within -4% to -39% for 1 GPM, with an average of-17.5%. For 2 GPM the average was found to be -31 % and for 3 GPM -36%. The Gungor 86 correlation, used with an equivalent diameter based on the heated perimeter, overpredicted the heat transfer coefficient between 3% to 88%, with lower accuracy corresponding to the higher heat fluxes. Specifically, for 1GPM we obtained an average deviation of 52%, respectively 10% and 52% for 2 and 3 GPM. The Gungor 87 correlation was found to underpredict the heat transfer coefficient with -19% to -42%, however, with values more uniformly distributed over the experimental range. For 1 GPM we obtained -25% deviation, for 2 GPM -28% and for 3 GPM -8%. Figure 8.5 shows the distribution of the deviations for the tested correlations against our experimental data. 100 Figure 8.4 Saturated boiling experimental data for 1.0, 2.0 and 3.0 GPM at thermocouple location #1 (discharge from the test section) 21000 j 19000 • E 17000 • i *= 15000• cu o o 13000 ->S? to c ro 11000 • h-13 9000 • cu X 7000 -5000 • 70000 14000 13000 12000 E ^ 11000 cu 8 10000 o 3> 9000 (/} c ,2 8000 ro CU 7000 X 6000 5000 • 90000 25000 23000 j 21000 19000 17000 • 15000 -13000 11000 9000 7000 5000 flow rate = 1.0 G P M , quality x = 0.091 - 0.226 Gungor 86 HTC 100000 110000 120000 130000 Heat Flux, W/m 2 140000 150000 flow rate = 2.0 G P M , quality x = 0.012 - 0.020 Gungor 86 HTC Chen HTC 95000 100000 105000 110000 115000 120000 125000 130000 Heat Flux, W/m 2 flow rate = 3.0 G P M , quality x = 0.016 - 0.030. Measured HTC Gungor 86 HTC o-Gungor87 HTC Chen HTC 7 200000 Heat Flux, W/m 2 230000 101 8.5 Discussion The trend shown by our saturated data was similar with results from other experimental programs, such as: Wattelet et al (1994) with R - l 34a and R-12, Shin et al (1996) with zeotropic and azeotropic refrigerants, and others for different heat transfer fluids. However, their set-ups used horizontal test sections. The similarity may be attributed to the fact that the very intense boiling regime lead to a partially wetted test section. Nevertheless, the reduced number of data points collected does not allow inferring a solid conclusion. The Chen correlation underpredicts the heat transfer coefficient, with a slightly better accuracy than even Gungor 1987 correlation and an even better distribution of the results. However the procedure is more complex than the one suggested by Gungor. The Gungor correlations show a great difference in results obtained by using both versions. The selection of the equivalent diameter greatly affects the results. The 1986 version overpredicts the heat transfer coefficient as compared to the 1987 version, which still underpredicts but with better accuracy and more uniformity. It is important to note that these conclusions are based on less than 50 data points. However, the predictions appear to be close enough to our experimental data. Chen's correlation seems to give slightly better results, yet the Gungor's 1987 correlation is simpler to use (less than half of the calculation steps required by Chen). 102 103 Chapter 9 Conclusions and future work 9.1 Conclusions of the present investigation This chapter concludes the experimental investigation of boiling heat transfer to R-l 34a refrigerant. The following paragraphs explain the conclusions and achievements of each step pursued during this study. 9.1.1 Experimental set-up The first milestone of this project was the design and construction of an experimental set-up for measuring the local boiling heat transfer coefficient. This task consisted of selecting the appropriate hydraulic, heat transfer and data acquisition equipment, calibrating and testing the experimental loop. During the experimental stage a few operating limitations were noticed, as summarized in the subsequent paragraphs. • The cooling capacity of the closed loop is an important factor in controlling the fixed parameters during experiments. Sizing of the subcooler should be based on the estimated maximum heat input into the test section at the maximum mass flow rate proposed for the experiments. Further, the size of the condenser should be based on the flow rate of refrigerant vapors generated at the maximum heat and mass flux conditions. • For annular cross-sections in vertical flow, the ratio of the outer tube inside diameter and heater outside diameter should be large enough to allow for the generated vapors to disperse into the flow. This will prevent the possibility of having partially wetted areas on the heated surface of the test section. • The data acquisition system should be capable of recording all the operating parameters required by the computational process. Even if only a few parameters have to be recorded manually, this leads to longer and more tedious experiments. Reading errors could also lead to less accurate measurements. 104 9.1.2 Experimental data base During the experimental program, a significant number of data were collected for subcooled and saturated boiling. Specifically, 1872 sets of experimental data were collected for subcooled boiling and a limited number (48 sets) for saturated boiling. The R-134a experimental data can be used as a base line for experimentally assessing the heat transfer performance of other refrigerants. The experimental methodology used proved successful and led to the following conclusions: • During the preliminary stage of experiments, in increasing heat flux mode, the hysteresis effect has been observed during the transition phase between single phase and subcooled boiling. In order to avoid the errors that could be associated with this phenomenon, the subcooled boiling experiments should be conducted in decreasing heat flux mode. • For saturated boiling, conducting the experiments in increasing heat flux mode proved to be convenient, from an operational and parameters control perspective. 9.1.3 Data analysis The experimental data collected during the research program were analyzed and compared with theoretical correlations. Classical heat transfer correlations were tested, as well as newer correlations. • Chen correlation applied to subcooled boiling, when used to validate experimental data, should use the same expression for the heat flux as used in the experimental procedure. This assures consistency when comparing the measured heat transfer coefficient with the predicted one. Using this approach, the average deviation was found to be 15.9%, with 80.5% of the data with an absolute mean deviation of less than 19%. An increase in system pressure leads to less accurate predictions of the heat transfer coefficient, while the mass flux variation does not significantly influence the results. 105 Chen's correlation forced convective and suppression factors, proposed in an analytical form by Collier [4] proved to accurately model the original graphical factors. Liu and Winterton correlation applied to subcooled boiling proved to be very "user-friendly", yet accurate. The average deviation was found to be -9.0% with 79.2% of the data within less than 16% absolute mean deviation. The results were consistent over the whole range of operating pressures and flow rates. Chen's correlation applied to saturated boiling generated acceptable results, the average deviation being -17.5% for 1 GPM set of experiments and -31% and -36% for 2 GPM and respectively 3 GPM. No significant influence of the mass flux and heat flux on the accuracy of the correlation was found. The computation process, using a spreadsheet program, is tedious due to the iterative procedure of finding the wall temperature. When using the 1986 Gungor and Winterton correlation for saturated boiling in annular cross-sections, the equivalent diameter employed by the procedure should use the heated perimeter instead of wetted perimeter, as suggested by authors. Using this approach we found that the correlation overpredicts the heat transfer coefficient with an average of 52% for the 1 GPM set of data, respectively 10% and 52% for 2 GPM and 3 GPM. Increasing the heat flux leads to less accurate results, while a change in flow rates has no significant influence. The revised Gungor and Winterton correlation (1987) was found to underpredict the heat transfer coefficient, but with a deviation more uniformly distributed over the experimental range of data. For 1 GPM the deviation was found to be -25%, for 2 GPM the deviation was -28% and for 3 GPM was -8%. Based on the current assessment, the preferred theoretical correlations for predicting the boiling heat transfer coefficient are Liu and Winterton [17] for subcooled boiling and 1987 Gungor and Winterton [11] for saturated boiling. 106 9.2 Future work The search for an environmental friendly refrigerant is a continuous mission for the refrigeration specialists. This is due to an understandable aim of improving the heat transfer processes, but also due to limitations found for the currently used refrigerants. Most of the limitations are generated by the raising standard of environment acceptability. In this context, the current work aspires to be a good reference for future studies of alternative refrigerants. The equipment used in this study can be easily modified to specifically suit other similar experimental programs. The design information will be useful for the experimentalists, at their initial stage of construction. The presented procedure and equipment can be used for testing other refrigerants with minor modifications, as required. Suggestions for future work include: • Testing other alternative refrigerants, with low or zero global warming potential, needs to be continued. Identifying the potential candidates should be done through a joint effort with a partner providing chemical engineering expertise in evaluating the other relevant properties. • Other boiling heat transfer correlations needs to be tested, against the existing experimental data bank. Existing correlations have to be revised in the light of new available experimental data. This includes developing or revising the forced convection and suppression factors commonly used by correlations to more accurately model the heat transfer behavior of the new refrigerants. • A dialog with manufacturers of refrigeration equipment and with the major refrigerant providers has to be initiated. This will lead to a more focused and problem solving research. Economical support has to be obtained from the interested parties in order to develop better and more versatile testing facilities, which in turn will lead to results covering a broader range of operating conditions encountered in practice. 107 REFERENCES: 1. Abdelmessih A.H., 1973, "A Facility for Experimental Investigation of Forced-Convection Boiling Two-Phase Flow", Technical Publication Series, Department of Mechanical Engineering, University of Toronto, Canada 2. Althouse A.D., Turnquist C.H., Bracciano A.F., 1992, "Modern Refrigeration and Air Conditioning", The Goodheart-Willcox Company Inc. 3. Aounallah Y., Kenning D.B.R., Whalley P.B. and Hewitt G.F., 1982, "Boiling Heat Transfer in Annular Flow", Int. J. of Heat and Mass Transfer, pp. 193-199 4. Chen J.C, 1966, "Correlation for Boiling Heat Transfer to Saturated Fluids in Convective Flow", I&EC Process Design and Development, Vol. 5. pp. 322-329 5. Collier J.G., 1981, "Convective Boiling and Condensation", 2-nd ed., McGraw-Hill, New York 6. Darabi J., Salehi M., Saeedi M.H., Ohadi M.M., 1995, "Review of Available Correlations for Prediction of Flow Boiling Heat Transfer in Smooth and Augmented Tubes", ASHRAE Transactions: Symposia, pp. 965-975 7. Devotta S. and Gopichand S., 1992, "Theoretical Assessments of HFC 134a and Alternatives to CFC12 as Working Fluids for Heat Pumps", Applied Energy, Vol. 41, pp. 285-299 8. Eckels S.J. and Pate M.B., 1991, "Evaporation and Condensation of HFC-134a and CFC-12 in a Smooth Tube and a Micro-fin Tube", ASHRAE Transactions, pp. 71-81 9. Eckels S.J. and Pate M.B., 1991, "An Experimental Comparison of Evaporation and Condensation heat Transfer Coefficients for HFC-134a and CFC-12", Int. J. of Refrigeration, Vol. 14, pp. 70-77 108 10. Fishman J. and Kalish R., 1990, "Global Alert - The Ozone Pollution Crisis", Plenum Press 11. Graillot S, 1997, "Heat Transfer Laboratory", UBC Mechanical Engineering Department Internship Final Report 12. Gungor K.E. and Winterton R.H.S., 1986, "A General Correlation for Flow Boiling in Tubes and Annuli", Int. J. of Heat and Mass Transfer, Vol. 29, pp.351-358 13. Gungor K.E. and Winterton R.H.S., 1987, "Simplified General Correlation for Saturated Flow Boiling and Comparison of Correlations with Data", Chem Eng Res Des, Vol. 65, pp. 148-156 14. Ha S. and Bergles A.E., 1994, "Some Aspects of In-Tube Evaporation", Proceedings at the Tenth International Heat Transfer Conference, Brighton, UK, Vol. 6, pp. 187-192 15. Johnson R.L., 1993, "Investigating the Ozone Hole", Lerner Publications Company 16. Kandlikar S.G., 1990, "A General Correlation for Saturated Two-Phase Flow Boiling Heat Transfer inside Horizontal and Vertical Tubes", Journal of Heal Transfer, Vol.112, pp. 219-228 17. Kattan N . , Thome J.R. and Favrat D., 1998, "Flow Boiling in Horizontal Tubes: Part 2-New Heat Transfer Data for Five Refrigerants", Transactions of the ASME, Vol. 120, pp. 148-155 18. Kreith F., Bohn M., 2001, "Principles of Heat Transfer", 6th ed., Brooks/Cole Publishers 19. Langley B.C., 1994, "Refrigerant Management - The Recovery, recycling, and Reclaiming of CFCs", Delmar Publishers Inc. 20. Liu X., 1997, "Condensing and Evaporating Heat Transfer and Pressure Drop Characteristics of HFC-134a and HCFC-22", Transactions of the ASME, Vol. 119, pp. 158-163 109 21. "Mark's Standard Handbook for Mechanical Engineers", Tenth Edition, McGraw-Hill, New York 22. Makhijani, A. and Gurney, K., 1995, "Mending the Ozone Hole - Science, Technology and Policy", The MIT Press, 1995 Institute for Energy and Environmental Research 23. Miller K.B. et. al., 1995, "Strategies for Managing Ozone-Depleting Refrigerants -Confronting the Future", Battelle Press 24. Nardo D., 1991, "Ozone", Lucent Overview series 25. Rehling T., 1996, "Design of a Heat Exchanger Test section for a Non-Ozone Depleting Refrigerant", UBC Mechanical Engineering Department, MECH 457 Final Report 26. Roan S., 1989, "Ozone Crisis - The 15-Year Evolution of a Sudden Global Emergency", John Willey & Sons 27. Ross H., Radermacher R. and Di Marzo M., "Horizontal Flow Boiling of Pure and Mixed Refrigerants", Int. J. of Heat and Mass Transfer, Vol. 30, pp. 979-992 28. Sami S.M. and Duong T.N., 1993, "Study of Flow Boiling Characteristics of R 134a in Annulus of Enhanced Surface Tubing", International Journal of Energy Research, Vol. 17, pp. 671-688 29. Shah M.M., 1976, "A New Correlation for Heat Transfer During Boiling Flow Through Pipes" 30. Shah M.M., 1980, "A General Correlation for Critical Heat Flux in Annuli", Int. J. of Heat and Mass Transfer, Vol. 23, pp. 225-234 31. Shah M.M., 1986, "A General Correlation for Heat Transfer During Subcooled Boiling in Pipes and Annuli" 32. Shin J.Y., Kim M.S., Ro S.T., 1996, "Experimental Study on Forced Convective Boiling Heat Transfer of Pure Refrigerants and Refrigerant Mixtures in a Horizontal Tube", Int. J. of Refrigeration, Vol. 20, pp. 261-215 110 33. Spindler K., "Flow Boiling", 1994, Proceedings of the Tenth International Heat Transfer Conference, Brighton, UK, Vol. l,pp. 349-368 34. Stephan K. and Auracher H., "Correlation for Nucleate Boiling Heat Transfer in Forced Convection", Int. J. of Heat and Mass Transfer, Vol. 24, pp. 99-107 35. Takamatsu H., Momoki S. and Fujii T., 1993, "A Correlation for Forced Convective Boiling Heat Transfer of Pure Refrigerants in a Horizontal Smooth Tube", Int. J. of Heat and Mass Transfer, Vol. 36, pp. 3351-3360 36. Thome J.R., 1994, "Two Phase Heat Transfer to New Refrigerants", Proceedings of the Tenth International Heat Transfer Conference, Brighton, UK, Vol.T, pp. 19-41 37. Thurlow G., 1990, "Technological Responses to the Greenhouse Effect", Published on behalf of The Watt Committee on Energy by Elsevier Applied Science 38. Torikoshi K. and Ebisu T., 1993, "Heat Transfer and Pressure Drop Characteristics of R-134a, R-32, and a Mixture of R-32/R-134a Inside a Horizontal Tube", ASHRAE Transactions: Research, pp. 90-96 39. Wattelet J.P., Chato J.C, Souza A.L. and Christoffersen B.R., 1992, "Evaporative Characteristics of R-l 2, and a Mixture at Low Mass Fluxes", ASHRAE Transactions: Symposia, Vol. 100, pp.603-615 40. Whalley P.B., 1996, "Two-Phase Flow and Heat Transfer", Oxford University Press 41. Yin S.T., Abdelmessih A.H., 1974, "Measurement of Liquid Superheat, Hysteresis Effect and Incipient Boiling Oscillations of Freon 11 in Forced Convection Vertical Flow", Technical Publications Series, Department of Mechanical Engineering, University of Toronto, Canada 111 APPENDIX NO. 1 Experimental Apparatus Components 112 113 114 115 116 Optical compensation box components Material: Aluminum LOWER PLATE UPPER PLATE 117 APPENDIX NO. 2 Calculation of the Electrical Resistance of the Test Section The heater operates as an electrical resistance and converts the electrical current into heat, further removed by the refrigerant at the tube surface contact. For a cylindrical element the resistance is: / , R = px —, where: p= resistivity [pQm], 1= length [m] and A= area [mm ] A Resistivity at temperature t will be: P t = P20 [1 + a (t - 20°C)], where athermal expansion coefficient [1/°C] For INCONEL625: p 2 0 = 1.26 p,Qm, a= 12.8 um/m°C For Copper: p 2 0 = 0.0179 pQm, a= 0.0039 um/m°C Heater - (Inconel Tube) dimensions are: d o D = 13.06 mm, d|D= 11.02 mm, 1= 0.503 m. R20h = 1.26 x [l +12.8 x 10"6 (20 - 20)]x = 16.427 x 10"3 Q -(l3.062 -11.022) 4V ' RJi = 1.26[l + 12.8 x 10"6(60 - 20)]x = 16.435 x 10"3Q ~(l3.062 -11.022) 4V ' Copper Tube dimensions are: doD= 13.06 mm, d i o = 5.59 mm. 0 302 R20(. = 0.0179[1 + 0.0039(20 - 20)] x : = 0.448 x 10"4 Q -(13.062 -5.592) 4 0 302 Rb0(: = 0.0179[1 + 0.0039(60 - 20)] x — = 0.518 x 10"4 Q -~(13.062 -5.592) 4 The total resistance of the inner tube is: R20total = [16.427xl0"3 + 0.0448x10"3] = 16.4718xl0"3 Q R60total = [16.435xl0"3 + 0.0518xl0"3] = 16.4868xl0"3 Q 118 APPENDIX NO. 3 Instruments Calibration data Sheets 119 o o o 00 o CO o o CM CO o o o o o o o o o o o o o o o o o o c o i ^ c o m ^ r c o c N T - o o o o o o o o o o l/\ld9 'ajeu M O | J o o o CO o CD O o CN ro c o i n CM i n CM m o o UILU/I '8}By M0|J let o o — TO O a> _2 — o > c "a. a> a: CM CO CM CO CD CJ) C M j C M CO lO CM CD CO CM o l d CM mo T - i C O CO o CM CO CD CD i n CM CO CO CO CO CO CO m m i c M l i ^ 00 !CO ! I^ O l O i O d i d i d CM t -CO CO m m d i d CD | r»» 00 CJ) i n i ^ -CO CM d id 00 CO;lO|-<r d j d i d c o CO 00 CD O T -inj tn d o t - i C M CM ; CM i n CM i n N i i n cDiooi-sr i n ; c o ; r -d d d co CM CM: , CM ' CO CM coico iJS Is- cx> <£ i CJ) c o c o i o C O i h - C D C O C O j C O cd co eb •^ j- inico C M C M l C M o o 1^ i n CD CM CJ) CO CJ) CD o CO CO CM 1 2 0 CALIBRATION DATA SHEET FOR THE TURBINE FLOWMETER Calibration for the Range switch selected at 4 Measured Data: Voltmeter Volume Time Flow Vdc GPM [volts] [litres] [min] [l/min] 5.03 4.17 4.06 100 7.83 12.77139 5.01 4.138 5.03 80 5.05 15.84158 4.78 3.943 4.95 100 6.5 15.38462 4.71 3.904 4.48 50 7.23 6.915629 4.6 3.81 3.68 50 8.75 5.714286 4.495 3.72 4.16 100 15.4 6.493506 4.395 3.635 0.3666 1 0.8943 1.118193 4.3 3.569 0.93 1 0.616 1.623377 4 3.309 0.915 18.7 11.8 1.584746 2.995 2.476 0.77 18.6 13.366 1.391591 2.485 2.063 0.85 18.6 12.45 1.493976 2 1.668 1.32 18.6 8.383 2.218776 1.5 1.254 1.63 18.6 6.85 2.715328 1 0.844 1.96 18.6 5.75 3.234783 0.5 0.443 2.51 18.6 4.516 4.118689 2.99 18.6 3.833 4.852596 4.2 18.6 2.75 6.763636 4.03 18.6 2.866 6.489881 3.49 18.6 3.3 5.636364 4.965 18.6 2.333 7.972568 o Q tn 4-1 o > a>~ o m o > 0 0 1 y = 1.2158X-0.0277 2 3 Flow rate, GPM 121 CALIBRATION DATA SHEET FOR PRESSURE CELLS P r e s s u r e c e l l #4 (test section inlet) P r e s s u r e C e l l #2 D i f f . P r e s s . C e l l p s i m V o l t s 0 -37.2 0.004 -38.1 0.004 -38.3 10.57 174.6 10.57 173.3 10.57 172.9 20.37 370.8 20.37 369.7 20.37 369 20.37 369.4 30.04 563 30.04 562 39.2 746 39.2 746 39.2 745 47.74 918 47.74 917 52.14 1007 52.14 1006 60.26 1170 69.13 1349 69.13 1346 p s i m V o l t s 0.004 -0.0251 0.004 -0.0258 10.02 0.1748 10.02 0.1736 19.33 0.3598 19.33 0.3581 26.45 0.5 26.45 0.499 34.99 0.668 35 0.67 43.06 0.831 50.36 0.977 50.36 0.975 57.24 1.114 57.25 1.113 65.03 1.269 65.03 1.268 69 1.348 D P C e l l y = 5.0136x +0.1402 1 1.5 Pressure, psig 2.5 p s i V o l t s 0 0.0004 0.052 0.296 0.103 0.57 0.153 0.832 0.2 1.083 0.301 1.625 0.402 2.166 0.5 2.68 0.599 3.206 0.701 3.737 0.799 4.25 0.9 4.77 0.999 5.27 1.101 5.78 1.201 6.27 1.302 6.76 1.399 7.23 1.5 7.7 1.601 8.18 1.702 8.65 1.799 9.1 1.9 9.57 1.994 9.99 2.001 10.01 y = 20.035X - 38.471 Pressure, psig C e l l #2 y = 0.0199x-0.0259 40 60 Pressure, psig 122 APPENDIX NO. 4 Tabulated Subcooled Boiling Experimental Data 123 Experimental Conditions: Flow Rate= 1.0GPM Pressure3 80PSIG Inlet Temp.= 10.8°C Measured Data: Notes: Current 1 x 0.5, A Voltage V, volts Mixing and Bulk Temperatures °c Surface Thermocouples Temperatures, °C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 190 2.22 10.8 12.6 18.3 18.4 30.9 31 31.2 30.2 31.1 31.8 160 1.86 10.8 12.6 16.3 16.8 28.6 28.6 28.6 27.7 28.5 28.9 130 1.5 10.8 12.6 14.7 15.3 28.3 28.3 28.5 27.9 28.6 28.9 100 1.16 10.8 12.6 13.3 14.2 26.3 26.5 26.6 26.3 26.8 26.8 69 0.8 10.8 12.6 12.3 13.1 24.5 24.8 25 24.9 25.1 25.1 Experimental Conditions: Flow Rate= 2.0GPM Pressure= 80PSIG Inlet Temp.= 12.2°C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 241 2.81 12.2 13.8 18.2 18.9 33.6 33.8 33.9 33 34.8 35.3 210 2.45 12.2 13.8 16.7 17.1 32.1 32.3 32.5 31.5 33.1 33.6 181 2.11 12.2 13.8 15.8 16.2 30.2 30.4 30.5 29.8 31.1 31.4 150 1.75 12.2 13.8 14.7 15.4 28.5 28.5 28.7 28 29.1 29.2 120 1.39 12.2 13.8 14 14.7 27.3 27.6 27.8 27.3 28.1 28 91 1.04 12.2 13.8 13.4 14.1 25.5 25.7 25.8 25.6 25.9 25.6 Experimental Conditions: Flow Rate= 3.0GPM Pressure3 80PSIG Inlet Temp.= 13.7°C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 261 3.04 13.7 15.2 18.3 19.1 34.8 35.1 35 34 36 36.6 230 2.68 13.7 15.2 17.3 18.3 32.9 33 33.2 32.2 34 34.4 200 2.32 13.7 15.2 16.5 17.6 31.2 31.2 31.5 30.5 32 32.3 170 1.98 13.7 15.2 15.9 16.5 29.7 29.8 30.1 29.3 30.5 30.6 140 1.62 13.7 15.2 15.2 16 28.2 28.3 28.5 27.7 28.7 28.6 111 1.27 13.7 15.2 14.7 15.7 26 26.1 26.2 26 26.5 26 80 0.92 13.7 15.2 14.3 15.6 24.7 24.9 25 24.7 25.1 24.9 70 0.8 13.7 15.2 14.2 15.4 24.6 24.8 25 24.7 24.9 23.6 1 2 4 Experimental Conditions: Flow Rate= 1.0GPM Pressure3 100PSIG Inlet Temp. 3 10.0°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C 1 X 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 279 3.27 10 11.7 24.8 24.3 40.9 41.2 41.3 40.2 42.2 42.8 249 2.93 10 11.7 22.2 21.7 39 39.2 39.5 38.5 40.1 40.7 220 2.58 10 11.6 19.6 19.7 37.8 38 38.3 37.5 38.7 39.4 190 2.23 10 11.7 17.3 17.5 36 36 36.5 35.7 36.7 37.2 170 1.99 10 11.6 16 16.4 34.8 34.7 35.2 34.6 35.5 35.9 150 1.75 10 11.6 14.8 15.3 34 34 34.2 33.8 34.5 35 130 1.51 10 11.6 13.6 14.5 32.2 32.3 32.5 32.3 33 33 111 1.29 10 11.6 12.7 13.5 31.6 31.6 31.9 31.8 32.3 32.2 91 1.06 10 11.7 12 12.7 30.3 30.4 30.7 30.7 31.1 30.8 80 0.93 10 11.5 11.5 12.5 29.5 29.7 30.1 30.2 30.3 30.2 Experimental Conditions: Flow Rate= 2.0GPM Pressure3 100PSIG Inlet Temp. 3 12.4°C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 350 4.11 12.4 14.2 24.7 24.7 44.5 45.1 44.7 43.1 46.4 47.4 320 3.76 12.4 14 22.7 23.2 42.8 43.2 43.1 41.6 44.7 45.5 291 3.41 12.4 14 21 21.6 40.8 41.1 41.2 39.7 42.3 43 260 3.05 12.4 14 19.3 20 39.8 39.8 40 38.8 41.1 41.6 230 2.7 12.3 14 17.8 18.4 38.5 38.6 39.1 37.9 39.8 40.2 201 2.35 12.4 13.9 16.6 17.2 36.6 36.7 37.3 36.2 38 38.2 170 1.99 12.4 14 15.4 16.1 .35 35 35.7 34.8 36.2 36.2 140 1.63 12.4 14 14.5 15.4 32.9 33.1 33.6 33 33.8 33.4 111 1.29 12.4 14 13.8 14.4 31.5 31.7 32.1 32 32.2 31.8 100 1.16 12.4 13.8 13.5 14.1 30.8 30.9 31.3 31.2 31.3 30.9 90 1.04 12.4 13.8 13.4 14.1 30.5 30.6 30.9 30.8 30.8 30.4 80 0.93 12.4 13.8 13.2 14.1 29.8 30 30.3 30.2 30.3 29.6 Experimental Conditions: Flow Rate3 3.0GPM Pressure3 100PSIG Inlet Temp. 3 13.6-C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 319 3.7 13.6 15.2 20.2 20.9 42.6 43.1 42.7 41.5 44.5 45.3 290 3.37 13.6 15.2 19.2 19.8 41.1 41.5 41.7 40.3 42.8 43.5 260 3.02 13.6 15.2 18.1 18.8 39.1 39.5 39.8 38.6 40.8 41.2 220 2.55 13 15.1 16.9 17.2 37.3 37.5 37.8 36.8 38.6 39 200 2.32 13.6 15.1 16.3 16.7 36.5 36.6 37.1 36.3 37.8 37.6 170 1.98 13.6 15 15.7 16.2 34.9 35 35.6 34.7 35.7 35.5 140 1.63 13.6 15.1 15.1 15.4 32.7 32.8 33.3 32.7 33.3 32.8 110 1.27 13.6 15.1 14.6 15.4 31.5 31.5 31.8 31.5 31.7 31.5 90 1.04 13.6 15 14.3 15.2 29.8 30 30.2 30.1 30.2 29.6 80 0.92 13.6 15 14.1 15.1 29.8 30.1 30.2 30.1 29.3 25.8 125 Experimental Conditions: Flow Rate= 4.0GPM Pressure= 100PSIG Inlet Temp.= 16.1°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C I Surface Thermocouples Temperatures, °C 1 x 0.5, A V, volts j Tmi Tbi Tmo Tbo i Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 345 4.04 16.1 17.6 22.1 22.6 i 44.9 45.5 46.1 43.8 47 ,48.1 311 3.64 16.1 17.6 21 21.4 j 42.5 42.9 42.9 41.5 44.3 45.2 281 3.29 16.1 17.7 20 20.7 I 40.9 41.2 41.4 40.1 42.6 43.1 240 2.81 16.1 17.6 19 19.6 | 38.5 38.8 39.2 38 39.9 40.3 209 2.45 16.1 17.4 18.4 18.6 | 37.3 37.5 38 36.8 38.5 38.5 167 1.95 16.1 17.6 17.7 18 j 34.3 34.5 35.1 34 35.1 34.8 140 1.63 16.1 17.5 17.2 17.8 | 32.2 32.4 32.8 32.3 32.9 32.6 99 1.15 16.1 17.7 16.8 17.8 | 30.4 30.6 30.8 30.6 30.8 30.7 Experimental Conditions: Flow Rate= 1.0GPM Pressure= 120PSIG Inlet Temp.= 10.5°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I x 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 301 3.55 10.5 12.7 27.5 27.2 47.5 47.5 47.7 46.3 48.7 49.7 280 3.3 10.5 12.5 25.4 24.6 45.6 46 46 44.6 46.7 47.5 250 2.95 10.5 12.5 22.6 22 44.2 44.4 44.6 43.3 45.2 45.9 221 2.6 10.5 12.3 20 20 42.3 42.5 42.6 41.6 43.1 43.7 190 2.23 10.5 12.2 17.7 18.1 41.5 41.5 42 40.9 42.1 42.8 170 2 10.5 12.2 16.5 17.2 40.4 40.3 40.6 39.8 40.9 41.4 150 1.76 10.5 12.2 15.3 15.8 39 38.8 39.1 38.6 39.5 39.9 130 1.52 10.5 12.2 14.1 14.9 38.5 38.6 38.7 38.5 39.1 39.2 111 1.3 10.5 12.2 __, 13.3 14 37.1 37.3 37.4 37.5 38 37.9 Experimental Conditions: Flow Rate= 2.0GPM Pressure= 120PSIG Inlet Temp.= 13.5°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I x 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 348 4.11 13.5 15.3 25.6 26 49.5 50.1 49.7 48.2 51.3 52.4 320 3.78 13.5 15.3 24 24.1 47.8 48.3 49 46.6 49.4 50.5 290 3.42 13.5 15.2 22 22.7 46.7 47.2 47.1 45.7 48.1 49.1 260 3.07 13.5 15.1 20.5 21.4 45 45.3 45.5 44.2 46.2 47.1 230 2.72 13.5 15.1 19 20 43.5 43.6 44.2 42.9 44.9 45.3 201 2.36 13.5 15.1 17.7 18.7 41.8 42.1 42.5 41.4 43 43.3 171 2.01 13.5 15 16.6 16.9 40.2 40.3 40.9 40 41.3 41.3 141 1.65 13.5 15.1 15.7 16.9 38.3 38.5 39 38.6 39.2 39.2 110 1.29 13.5 15.1 15 15.6 36.8 37.1 37.5 37 37.3 37.2 100 1.17 13.5 15.1 14.8 15.5 36.9 37 37.2 37.1 37.5 37.1 90 1.05 13.5 15.1 14.6 15.3 36.3 36.8 36.9 36.7 36.7 35.3 126 Experimental Conditions: Flow Rate= 3.0GPM Pressure3 120PSIG Inlet Temp. 3 14.0°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C J Surface Thermocouples Temperatures, °C 1 x 0.5, A V, volts T Tmi ( Tbi ; Tmo Tbo j Ts1 Ts2 Ts3 j Ts4 Ts5 Ts6 319 3.76 14 15.6 20.9 21.4 47.4 47.9 47.6 46.4 49.1 49.9 290 3.42 14 15.5 19.8 20.1 46.1 46.6 46.7 45.3 47.5 48.4 260 3.06 14 15.6 18.8 19 44.9 45.2 45.3 44.4 46.6 47 240 2.82 14 15.6 18 17.9 43.9 44.1 44.4 43.4 45.1 45.5 220 2.59 14 15.6 17.5 17.4 42.6 42.7 43.2 42.3 44 44.2 201 2.36 14 15.6 17.1 17.3 41.7 41.8 42.1 41.3 42.8 42.7 171 2 14 15.6 16.3 16.2 40.3 40.4 40.9 40 41 40.9 140 1.64 14 15.6 15.6 15.8 38.5 38.6 38.9 38.4 39.1 38.9 110 1.29 14 15.4 15.1 15.6 36.9 37.2 37.3 37.1 37.3 37.2 91 1.06 14 15.4 14.8 15.6 36.2 36.3 36.5 36.3 36 30.6 Experimental Conditions: Flow Rate= 4.0GPM Pressure3 120PSIG Inlet Temp.= 17.1°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I x 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 360 4.25 17.1 18.7 23.8 24 50 50.6 49.9 48.5 51.7 52.9 347 4.1 17.1 18.6 23.3 23.8 49.3 49.9 49.6 48.1 51 52.2 312 3.69 17.1 18.6 22.1 22.7 47.4 47.9 47.7 46.3 49.1 49.9 280 3.31 17.1 18.6 21.2 21.6 45.5 46 46 44.6 47.1 47.1 240 2.83 17.1 18.6 20.2 20.5 44 44.3 44.6 43.3 45.3 45.6 212 2.49 17.1 18.6 19.6 19.6 42.3 42.6 43.1 41.9 43.4 43.7 190 2.23 17.1 18.6 19.1 19.5 40.9 41.1 41.5 40.3 41.7 41.6 171 2.01 17.1 18.6 18.8 19 39.9 40.1 40.5 39.5 40.6 40.4 140 1.64 17.1 18.6 18.3 18.7 38.1 38.1 38.4 37.9 38.4 38.3 121 1.14 17.1 18.6 18 i 18.7 36.8 37 ! 37.1 36.9 37.3 37 Experimental Conditions: Flow Rate3 1.0GPM Pressure3 140PSIG Inlet Temp. 3 10.2°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I x 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 300 3.56 10.2 12.2 28.2 26.8 51.2 51.7 51.4 50 52.3 53.2 280 3.31 10.2 12.2 26.2 25.2 50.6 51 50.8 49.5 51.6 52.5 250 2.95 10.2 12.1 23.1 22.4 48.9 49.1 49.2 48 49.7 50.5 220 2.6 10.2 12 20.3 20.4 47.5 47.6 47.8 46.7 48.3 48.6 190 2.24 10.2 11.9 18 18.3 45.9 46.1 46.2 45.4 46.6 47.2 170 2 10.2 12 16.5 17.2 44.2 44.1 44.3 43.6 44.6 45 150 1.76 10.2 11.9 15.1 15.9 43.7 43.7 43.8 43.4 44.3 44.7 131 1.53 10.2 11.9 14.1 14.8 43.1 43.1 43.5 43.2 43.9 44 111 1.3 10.2 11.9 13.2 14.1 42.1 42.1 42.5 42.5 42.8 43 100 1.17 10.3 12 12.7 13.5 38.2 38.7 40.7 40.7 41.4 41.4 127 Experimental Condit ions: Flow Rate= 2.0GPM Pressure= 140PSIG Inlet Temp.= 12.7°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures .."C , Surface Thermocouples Temperatures, °C 1 x0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 347 4.11 12.7 14.5 24.9 25.7 53.3 54 53.5 52 54.9 56.1 320 3.79 12.7 14.3 23.2 23.5 51.7 52.4 52 50.7 53.4 54.4 289 3.43 12.7 14.5 21.5 22.1 50.4 50.9 50.8 49.7 51.9 52.7 260 3.08 12.7 14.2 19.7 20.3 49.4 49.7 50 48.9 50.8 51.2 230 2.72 12.7 14.4 18.3 18.9 47.7 48 48.2 47.2 48.9 49.4 202 2.38 12.7 14.4 17 17.8 46.2 46.4 46.6 45.9 47.3 47.6 171 2.01 12.7 14.4 15.9 16.4 44.7 45 45.5 44.8 45.9 45.9 140 1.65 12.7 14.3 15 15.7 43.7 43.8 44.2 43.8 44.4 44.5 • 111 1.3 12.7 14.2 14.2 14.8 42.1 42.3 42.5 42.5 42.7 42.3 100 1.17 12.7 14.3 14 14.8 41.4 41.7 42 41.8 42.1 39.7 90 1.06 12.7 14.3 13.8 14.5 40.9 41.5 41.8 41.6 41.6 34.6 Experimental Condit ions: Flow Rate= 3.0GPM Pressure= 140PSIG Inlet Temp.= 14.6°C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 321 3.79 14.6 16.2 21.6 21.8 52.5 52.6 52.8 51.5 54 55 291 3.44 14.6 16.2 20.4 20.4 51.2 51.6 51.6 50.4 52.6 53.5 261 3.08 14.6 16.1 19.3 19.7 49.4 50 50 48.9 50.8 51.4 220 2.6 14.6 16 18.1 17.9 47.2 47.6 47.9 46.8 48.7 48.9 200 2.36 14.6 16.1 17.5 17.2 46.7 47.1 47.4 46.6 47.8 48.1 171 2.01 14.6 16.2 16.9 16.7 44.7 45 45.2 44.4 45.2 45.3 140 1.64 14.6 16.1 16.2 16.3 43.6 43.7 44 43.6 44.1 44.2 110 1.29 14.6 16.2 15.7 16.3 42.3 42.5 42.5 42.3 42.6 39.4 Experimental Condit ions: Flow Rate-= 4.0GPM Pressure= 140PSIG Inlet Temp.= 17.1°C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures , °c Surface Thermocouples Temperatures, °C Tmi Tbi .1...T.T°... Tbo L_I5!. H . . J j ? . . . l . Ts3 Ts4 Ts5 Ts6 347 4.1 17.1 18.6 | 23.1 23.2 53.8 54.5 | 53.9 52.6 55.5 56.6 312 3.68 17.1 18.5 j 22.1 22.2 51.9 52.7 | 52 51.2 53.6 54.4 280 3.31 17 18.4 | 21.2 20.8 50.5 51.1 i 51 50 52 52.4 240 2.83 17.1 18.6 I 20.2 20.5 48.4 48.7 | 48.9 48 49.6 49.9 210 2.48 17.1 18.5 | 19.6 18.9 46.7 47.1 | 47.3 47 47.7 47.6 190 2.23 17.1 18.6 | 19.4 19.1 45.3 45.6 | 45.9 45.1 46.1 46 170 2 17.1 18.5 | 18.7 19.1 44.5 44.6 j 44.9 44.3 45.2 45 140 1.64 17.1 18.5 i 18.2 18.8 43.2 43.4 j 43.5 43.1 43.7 43.5 119 1.4 17.1 18.6 | 18 18.7 42.8 42.8 | 43 42.6 43.5 39.9 128 Experimental Condit ions: Flow Rate= 1.0GPM Pressure= 160PSIG Inlet Temp.= 11 3°C Measured Data: Notes: Current | Voltage [Mixing and Bulk Temperatures, °C < Surface Thermocouples Temperatures, °C 1 x0.5, A ! V, volts !_ Tmi ! Tbi ! Tmo Tbo [ Ts1 i Ts2 i Ts3 i Ts4 ! Ts5 i Ts6 400 4.78 11.3 13.7 42.6 40.8 60.2 61.3 59.8 57.9 61.9 63.6 370 4.42 11.3 13.5 39 37.3 58.7 59.6 58.4 56.7 60.1 61.5 340 4.05 11.3 13.5 34.7 33.3 57.4 58.2 57.4 55.7 58.7 60.1 309 3.69 11.3 13.5 31.1 30 56 56.8 56.2 54.7 57.3 58.6 280 3.33 11.3 13.5 27.6 27 54.7 55.2 54.9 53.5 55.7 56.7 251 2.99 11.3 13.5 24.7 23.7 53.2 53.8 53.6 52.1 54.2 55 220 2.61 11.3 13.4 21.8 21.4 51.4 51.8 51.7 50.5 52.2 52.9 190 2.26 11.3 13.4 19.2 19.9 50.9 51.1 51.1 50.2 51.6 52.2 161 1.91 11.3 13.2 17.2 17.8 49.5 49.7 49.6 49 50 50.6 130 1.54 J 11.3 13.1 15.3 16.4 48.1 48.3 48.1 47.9 48.5 49.1 Experimental Condit ions: Flow Rate= 2.0GPM Pressure= 160PSIG Inlet Temp.= 14.1°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures , °c Surface Thermocouples Temperatures, °C I x 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 400 4.77 14.1 15.5 30.4 30.1 60.6 61.6 60.2 58.6 62.4 64.1 350 4.17 14.1 16 26.7 26.8 57.9 58.6 56.5 56.3 59.8 60.9 331 3.95 14.1 16 25.6 26.1 57 57.8 56 55.6 58.5 59.7 301 3.57 14.1 16 23.6 24.2 55.3 56 54.4 54.1 56.5 57.6 271 3.22 14.1 16 21.8 22.8 54.3 54.9 54.3 53.3 55.6 56.5 240 2.86 14.1 16 20.3 21.8 53 53.2 53.2 52 53.9 54.6 210 2.49 14.1 16 18.9 19.2 51.4 51.9 52 50.7 52.3 52.6 180 2.12 14.1 16 17.7 17.9 50 50.3 50.4 49.6 50.8 51.2 150 1.78 14.1 16 16.7 17.2 49.2 49.5 49.6 49 50 50 120 1.42 14.1 16 15.8 16.6 47.8 48.2 48.3 48.1 48.7 48.3 Experimental Condit ions: Flow Rate= 3.0GPM Pressure= 160PSIG Inlet Temp.= 16.2°C Measurec Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I x 0.5, A V, volts j Tmi Tbi Tmo t Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 400 4.77 16.2 18.2 27.1 27.5 59.6 60.7 58.3 57.3 61.1 62.7 370 4.42 16.2 17.8 25.4 26.5 58.9 59.9 57.5 57.2 60.5 61.9 341 4.06 16.2 17.7 24.1 25.4 57.8 58.7 57.8 56.3 59.3 60.6 309 3.69 16.2 17.8 22.7 23.8 56.2 56.8 55.7 55 57.6 58.7 280 3.33 16.2 17.8 21.6 23 54.5 55.2 53.1 53.5 55.8 56.6 250 2.97 16.2 17.9 20.5 21.3 53.2 53.7 51.9 52.2 54.2 55 220 2.61 16.2 17.7 19.5 20 51.9 52.3 52.4 51.2 52.9 53.4 191 2.26 16.2 17.8 18.7 19.7 50.6 51 51.1 50 51.3 51.8 160 1.89 16.2 17.7 18.1 18.8 49.2 49.4 49.6 48.7 49.6 49.8 130 1.54 16.2 17.8 17.5 18.3 48 48.2 48.3 47.7 48.4 48.2 110 1.3 16.2 17.6 17.2 18 47.1 47.4 47.5 47.1 47.5 41.3 129 Experimental Conditions: Flow Rate= 4.0GPM Pressure3 160PSIG Inlet Temp.= 18.4°C Measurec Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C 1 x0.5, A V, volts Tmi Tbi Tmo t Tbo . . . - M . . . L Ts2 Ts3 Ts4 Ts5 Ts6 401 4.79 18.4 20.1 26.5 27.8 60.1 61.2 59.6 58 62.2 63.7 369 4.4 18.4 20.1 25.3 26 58.9 59.9 58.2 57.3 60.5 61.9 341 4.06 18.4 20.1 24.2 25 57.4 57.9 56.6 55.8 58.9 60 310 3.69 18.4 20 23.3 24 56.4 57.1 56.1 55 57.7 58.6 281 3.33 18.4 20 22.4 23.8 54.8 55.4 53.6 53.7 56 56.8 251 2.99 18.4 20 21.5 23.3 53.4 53.9 51.6 52.5 54.4 55.1 220 2.61 18.4 20 20.8 22.5 51.8 52.4 50.2 50.9 52.8 53.1 190 2.25 18.4 20 20.2 21.6 50.5 50.7 50.8 49.7 51 51.2 160 1.89 18.4 20 19.8 20.4 49.1 49.4 49.5 48.7 49.7 49.8 130 1.53 18.4 19.8 19.3 20.3 47.7 48 48.1 47.7 48.4 46.6 Experimental Conditions: Flow Rate= 4.0GPM Pressure3 160PSIG Inlet Temp.= 18.4°C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, "C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 401 4.79 18.4 20.1 26.5 27.8 60.1 61.2 59.6 58 62.2 63.7 369 4.4 18.4 20.1 25.3 26 58.9 59.9 58.2 57.3 60.5 61.9 341 4.06 18.4 20.1 24.2 25 57.4 57.9 56.6 55.8 58.9 60 310 3.69 18.4 20 23.3 24 56.4 57.1 56.1 55 57.7 58.6 281 3.33 18.4 20 22.4 23.8 54.8 55.4 53.6 53.7 56 56.8 251 2.99 18.4 20 21.5 23.3 53.4 53.9 51.6 52.5 54.4 55.1 220 2.61 18.4 20 20.8 22.5 51.8 52.4 50.2 50.9 52.8 53.1 190 2.25 18.4 20 20.2 21.6 50.5 50.7 50.8 49.7 51 51.2 160 1.89 18.4 20 19.8 20.4 49.1 49.4 49.5 48.7 49.7 49.8 130 1.53 18.4 19.8 19.3 20.3 47.7 48 48.1 47.7 48.4 46.6 Experimental Conditions: Flow Rate= 1.5GPM Pressure3 80PSIG Inlet Temp.= 11 6°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C I Surface Thermocouples Temperatures, °C I x 0.5, A V, volts Tmi Tbi Tmo Tbo ! Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 210 2.45 11.6 13.3 18 18.4 | 32.3 32.5 32.7 31.7 33.1 33.7 181 2.1 11.6 13.3 16.4 17.1 | 30.6 30.6 30.9 30.1 31.3 31.9 149 1.74 11.6 13.3 15 15.6 I 29.3 29.3 29.4 28.7 29.7 30.1 121 1.4 11.6 13.3 14 14.4 | 27.4 27.4 27.7 27.2 28 28.1 90 1.04 11.6 13.3 13 14 | 25.7 25.8 26.1 25.8 26.1 26 81 0.93 11.6 13 12.7 13.6 j 25.3 25.5 25.7 25.5 25.6 25.5 69 0.8 11.6 13.1 12.6 13.5 ! 24.8 25 25.3 25.2 25.3 25.2 130 Experimental Condi lions: Flow Rate= 2.5GPM Pressure 3 80PSIG Inlet Temp.= 12.6-C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, "C 1 x0.5, A V, volts j Tmi i Tbi Tmo Tbo Ts2 Ts3 Ts4 Ts5 Ts6 260 3.02 12.6 14.2 18.1 18.9 34.6 34.7 35.2 33.8 36.1 36.7 240 2.79 12.6 14.2 17.5 18.1 33.4 33.5 34 32.7 34.7 35.3 220 2.57 12.6 14.2 16.6 17.4 32.3 32.2 32.7 31.7 33.6 34 201 2.33 12.6 14.1 16 16.5 31.1 31.2 31.6 30.5 32.5 32.6 171 1.97 12.6 14.1 15.1 15.7 29.4 29.3 30.1 29 30.6 30.7 141 1.63 12.6 14 14.5 15.5 27.6 27.7 28.1 27.4 28.5 28.4 120 1.38 12.6 14.1 13.9 14.5 26.3 26.5 27.1 26.6 27.3 26.9 100 1.16 12.6 13.9 13.6 14.3 25.2 25.3 25.7 25.3 25.7 25.4 80 0.92 12.6 14 13.3 14.3 23.9 24.1 24.3 24.2 24.2 24.1 70 0.8 12.6 14 13.1 14.1 23.3 23.5 23.5 23.6 23.6 23.5 Experimental Conditions: Flow Rate= 3.5GPM Pressure 3 80PSIG Inlet Temp.= 14.2°C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 271 3.16 14.2 15.7 18.5 19.2 36.4 36.7 36.7 35.6 37.8 38.5 240 2.8 14.2 15.8 17.6 18.3 34.2 34.4 34.6 33.6 35.5 35.9 211 2.45 14.2 15.7 16.8 17.5 32.1 32.2 32.5 32.6 33.3 33.7 181 2.1 14.2 15.8 16.3 17.1 30.4 30.4 30.7 30 31.3 31.5 151 1.75 14.2 15.7 15.7 16.5 28.7 28.8 29.2 28.4 29.6 29.4 130 1.51 14.2 15.7 15.4 16.2 27.2 27.9 28.4 28 28.8 28.7 100 1.16 14.2 15.7 14.9 15.8 26.4 26.5 26.8 25.4 26.9 26.7 80 0.92 14.2 15.7 14.7 15.8 24.8 25.1 25.1 24.9 25.1 25 Experimental Conditions: Flow Rate= 1.5GPM Pressure 3 100PSIG Inlet Temp.= 11.6°C Measured Data: Notes: Current Voltage Mixing an d Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I X 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 319 3.76 11.6 13.5 25.4 25.2 43 43.7 43.2 41.6 44.5 45.4 289 3.4 11.6 13.5 23.2 23.2 41.3 41.9 41.8 40.3 42.7 43.6 261 3.06 11.6 13.5 21 21.3 40.3 40.7 40.6 39.2 41.5 42.2 231 2.71 11.6 13.5 19.2 19 38.6 38.8 38.9 37.7 39.8 40.4 201 2.35 11.6 13.5 17.5 18 37.3 37.4 37.5 36.4 38.1 38.6 171 2 11.6 13.5 16 16.3 35.4 35.5 35.7 34.7 36.1 36.4 139 1.62 11.6 13.5 14.6 15.6 33.7 34 34.1 33.5 34.6 34.7 111 1.28 11.6 13.5 13.6 14.5 32.2 32.4 32.5 32.3 33 32.9 89 1.03 11.6 13.5 13.1 14 31.1 31.3 31.5 31.5 31.8 31.6 73 0.84 11.6 13.2 12.7 13.5 30.4 30.7 30.9 30.9 31.1 28.5 131 Experimental Conditions: Flow Rate= 2.5GPM Pressure= 100PSIG Inlet Temp.= 14.4°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, "C 1x0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 349 4.11 14.4 16 24.2 24.6 44.7 45.5 44.8 43.2 46.4 . 47.5 320 3.77 14.4 16 22.7 23.7 43.4 44 43.7 42 45.1 46 291 3.41 14.4 16 21.3 21.8 41.7 42.1 41.9 40.6 43 43.9 260 3.04 14.4 16 20 21 40.3 40.7 40.7 39.4 41.7 42.3 231 2.71 14.4 16 18.9 19.4 38.7 38.9 39.1 37.8 39.8 40.3 201 2.35 14.4 16 17.8 19.2 37 37.2 37.5 36.3 38 38.4 170 1.98 14.4 16 16.9 17.6 35.4 35.7 35.9 35 36.2 36.4 140 1.62 14.4 16 16.2 17.2 33.9 34.2 34.4 33.7 34.6 34.5 110 1.27 14.4 16 15.6 16.4 32.2 32.4 32.6 32.2 32.6 32.6 90 1.04 14.4 v 1 6 15.2 16.2 31.2 31.4 31.4 31.3 31.6 31.3 73 0.84 14.4 16 15 16 30.7 30.8 30.8 30.7 30.7 26.6 Experimental Conditions: Flow Rate= 3.5GPM Pressure= 100PSIG Inlet Temp.= 16.8°C Measured Data: Notes: Current Voltage V, volts Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I x 0.5, A Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 349 4.11 16.8 18.3 24 24.8 44.8 45.6 45 43.5 46.6 48 319 3.75 16.8 18.3 22.8 23.6 42.9 43.6 43.2 41.7 44.6 45.5 290 3.41 16.8 18.3 21.7 22.5 .42 42.4 42.4 40.9 43.4 44.2 262 3.07 16.8 18.3 20.8 21.9 40.4 40.8 40.7 39.4 41.6 42.3 231 2.7 16.8 18.3 20 20.7 38.7 39.1 39.2 37.8 39.8 40.3 201 2.35 16.8 18.3 19.1 19.7 37.1 37.3 37.6 36.4 38 38.5 170 1.99 16.8 18.3 18.6 19.4 35.6 35.8 36.1 34.9 36.2 36.5 140 1.63 16.8 18.3 18.1 18.7 34.4 34.6 34.7 34 34.8 34.8 110 1.28 16.8 18.3 17.6 18.5 32.7 32.8 32.9 32.6 33.1 33 90 1.04 16.8 18.3 17.3 18.4 31.6 31.8 31.9 31.7 32 31.1 Experimental Conditions: Flow Rate= 1.5GPM Pressure= 120PSIG Inlet Temp.= 12.0°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I x 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 349 4.13 12 13.9 28.5 28.2 49.6 50.2 49.6 47.7 50.9 52 320 3.78 12 13.9 26.2 25.8 48.3 48.8 48.6 47 49.6 50.8 291 3.43 12 13.8 23.8 24 46.6 47 47.2 45.6 48 49 260 3.07 12 13.6 21.6 21.7 45.1 45.3 45.4 44.1 46.2 47 231 2.72 12 13.6 19.5 20.2 43.9 44.2 44.4 42.9 44.9 45.6 200 2.35 12 13.5 17.7 18.5 42.1 42.2 42.3 40.7 42.4 42.7 170 2 12 13.6 16.3 17 41.3 41.4 41.9 40.7 42 42.5 151 1.77 12 13.6 15.5 16.2 40.5 40.5 40.7 39.8 41 41.2 131 1.53 12 13.6 14.6 15.4 39.2 39.3 39.7 39 39.6 40.1 111 1.3 12 13.6 14 14.7 38.2 38.5 38.7 38.2 38.9 38.9 91 1.06 12 13.6 13.5 14.5 36.9 37.5 37.8 37.7 38 37.6 132 Experimental Condit ions: Flow Rate= 2.5GPM Pressure 3 120PSIG Inlet Temp.= 14.1°C Measurec Data: Notes: Current Voltage Mixing and Bulk Temperatures, "C Surface Thermocouples Temperatures, °C 1 x 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts3 Ts4 Ts5 Ts6 350 4.12 14.1 15.6 24 24.2 49 49.6 49.2 47.4 50.6 51.9 320 3.78 14.1 15.6 22.4 23.5 48.1 48.7 48.1 46.5 49.4 50.6 290 3.42 14.1 15.6 20.9 21.2 47 47.5 47.4 45.9 48.4 49.2 261 3.07 14.1 15.6 19.7 20.2 45.5 45.8 45.9 44.5 46.8 47.7 230 2.72 14.1 15.6 18.5 18.6 44.2 44.6 44.8 43.5 45.4 46 200 2.36 14.1 15.6 17.6 17.9 42.7 42.9 43.3 42 43.7 44.1 170 2 14.1 15.6 16.8 17.2 41.3 41.4 41.8 40.7 42 42.4 140 1.65 14.1 15.6 15.9 16.2 39.7 39.9 40.3 39.6 40.5 40.4 121 1.41 14.1 15.5 15.6 16.1 39 39.1 39.5 38.9 39.5 39.4 100 1.16 14.1 15.5 15.2 16 38 38.2 38.3 38 38.1 37.5 80 0.93 14.1 15.5 14.8 15.7 37.1 37.3 37.4 37.1 34.6 29.1 Experimental Condit ions: Flow Rate= 3.5GPM Pressure= 120PSIG Inlet Temp.= 16.0°C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 348 4.12 16 17.6 23.1 23.7 49.3 50 49.5 47.8 51.1 52.3 321 3.8 16 17.6 22.2 22.8 48.2 48.9 48.3 46.8 49.7 50.7 291 3.44 16 17.6 21 21.4 46.7 47.1 47.2 45.5 47.1 49 261 3.08 16 17.5 20.2 21.6 45.4 45.8 45.9 44.5 47 47.4 231 2.72 16 17.5 19.2 19.8 43.8 44.2 44.4 43 45 45.5 200 2.36 16 17.5 18.6 19.3 42.5 42.7 43 41.7 43.4 43.8 171 2.01 16 17.5 17.9 18.2 40.9 41.1 41.6 40.5 41.8 41.9 139 1.64 16 17.5 17.3 17.8 39.5 39.7 40 39.1 40.1 40.1 110 1.28 16 17.5 17 17.7 38 38.1 38.3 38.9 38.3 37.9 Experimental Condi l ions: Flow Rate= 1.5GPM Pressure 3 140PSIG Inlet Temp.= 12.4°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I x 0.5, A V, volts j Tmi L Tbi Tmo Tbo Ts3 Ts4 349 4.14 12.4 14.2 29 29.4 54.4 55.4 54.8 53.2 56.3 57.9 320 3.79 12.4 14.2 26.4 26.8 52.5 53.1 52.6 51.1 53.8 55.1 291 3.44 12.4 14.2 24.1 24 51.4 52 51.6 50 52.5 53.6 261 3.09 12.4 14.1 21.9 21.4 49.8 50.3 50.2 48.8 50.8 51.8 230 2.71 12.4 14.1 19.9 20.5 48.3 48.6 48.6 47.4 49.2 50 201 2.36 12.4 14.1 18.2 18.9 46.9 47.1 47.3 46.2 47.6 48.3 170 2 12.4 14.1 16.6 17.2 45.7 45.8 46 45.1 46.3 46.0 141 1.67 12.4 14.1 15.5 16 44.3 44.5 44.7 44 45 45.2 109 1.28 12.4 14.1 14.4 15.6 42.4 42.9 43.3 42.9 43.6 43.7 101 1.17 12.4 14.1 14.1 15 41 42.5 42.5 42.1 42.3 43.2 133 Experimental Conditions: Flow Rate= 2.5GPM Pressure= 140PSIG Inlet Temp.= 14.8°C Measured Data: Notes: Current 1 x 0.5, A Voltage V, volts Mixing and Bulk Temperatures, "C Surface Thermocouples Temperatures, °C Tmi Tbi J Tmo Tbo Ts1 Ts2 Ts3 t Ts4 Ts5 Ts6 348 4.12 14.8 16.6 24.8 25.8 53.9 54.8 54 52.3 55.4 56.8 319 3.79 14.8 16.6 23.2 24 52.7 53.4 52.8 51.5 54.2 55.4 291 3.44 14.8 16.3 21.8 22.1 51.3 51.9 51.5 50.1 52.5 . 53.6 261 3.08 14.8 16.2 20.5 21 49.7 50.1 50.1 48.7 50.8 51.7 230 2.72 14.8 16.3 19.3 19.8 48 48.2 48.3 47.1 48.8 49.5 200 2.36 14.8 16.4 18.3 18.7 46.6 46.9 47 45.9 47.3 48 170 2 14.8 16.2 17.5 18.3 45.5 45.6 45.9 44.9 46.1 46.3 141 1.65 14.8 16.3 16.7 17.6 44.2 44.4 44.6 43.8 44.8 44.8 110 1.29 14.8 16.4 16.1 16.6 43.2 43.5 43.6 43.1 43.6 43.1 Experimental Conditions: Flow Rate= 3.5GPM Pressure= 140PSIG Inlet Temp.= 16.5°C Measured Data: Notes: Current Voltage V, volts Mixing and Bulk Temperatures , °c Surface Thermocouples Temperatures, °C Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 351 4.16 16.5 18.1 23.7 24.2 54.1 54.8 54 52.5 55.7 56.9 318 3.77 16.5 18 22.5 23.6 52.7 53.3 52.8 51.4 54 55.1 290 3.43 16.5 18.1 21.6 22.9 51.1 51.5 51.3 49.9 52.3 53.2 261 3.08 16.5 18.1 20.6 21.2 49.6 50.1 50 48.8 50.9 51.8 230 2.71 16.5 18.1 19.8 20.3 48.2 48.7 48.7 47.3 49.2 49.8 200 2.36 16.5 18.1 19.1 19.8 47.2 47.6 47.7 46.6 47.9 48.3 171 2.01 16.5 18 18.4 18.8 45.8 46.1 46.1 45.2 46.3 46.6 140 1.65 16.5 18 17.8 18.2 44.3 44.5 44.7 43.9 44.8 44.8 111 1.29 16.5 18 17.4 18.1 42.9 43.1 43.1 42.7 43.2 38.5 101 1.17 16.5 18 17.3 18.1 42.5 42.7 42.8 42.3 40.5 34.7 Experimental Conditions: Flow Rate= 1.5GPM Pressure= 160PSIG Inlet Temp.= 12.4°C Measured Data: Notes: Current I x 0.5, A Voltage V, volts Mixing and Bulk Temperatures , °c Surface Thermocouples Temperatures, °C Tmi ™ _, Tmo ..J52...H Ts1 Ts3 Ts4 Ts5 Ts6 351 4.17 12.2 14.1 28.8 28.9 58.2 59.1 58.1 56.5 59.6 60.7 300 3.57 12.4 14.1 24.8 23.8 55.3 56 55.4 54.1 56.7 57.6 270 3.21 12.4 14.1 22.6 23.3 54.5 54.9 54.7 54.5 55.7 56.4 240 2.84 12.4 14.1 20.5 21.1 53.2 53.4 53.4 52.9 54 54.8 210 2.49 12.4 14.1 18.6 19.9 51.6 51.8 51.9 50.7 52.3 52.9 180 2.13 12.4 14.1 17.1 18.2 50.3 50.6 50.6 49.7 51 51.4 150 1.78 12.4 14.1 15.9 16.3 48.9 49 49.2 48.7 49.7 49.9 121 1.42 12.4 14.1 14.8 15.3 47.8 48.1 48.2 47.9 48.6 48.7 134 Experimental Conditions: Flow Rate= 2.5GPM Pressure= 160PSIG Inlet Temp.= 14.8-C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C 1x0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 368 4.37 14.8 16.4 25.9 26.1 58.7 59.4 58.4 56.8 60.2 61.6 342 4.06 14.8 16.5 24.5 25.5 57.3 58.2 56.1 55.9 59 60.3 311 3.7 14.8 16.5 23 23.8 56.1 56.7 54.9 54.8 57.6 58.5 280 3.33 14.8 16.4 21.5 22.3 54.4 55.3 54.9 53.5 55.8 56.7 250 2.96 14.8 16.4 20.1 19.9 53.5 53.9 53.8 52.5 54.5 55.3 220 2.61 14.8 16.4 19 19.7 52.1 52.5 52.6 51.4 53.1 53.5 191 2.26 14.8 16.4 18.1 18.2 51.1 51.4 51.6 51 51.9 52.4 162 1.91 14.8 16.4 17.1 17.9 49.6 50 50.1 49.3 50.3 50.6 131 1.53 14.8 16.4 16.4 17 48.5 48.7 48.9 48.2 49.1 49 100 1.18 14.8 16.4 15.9 16.5 47.1 47.6 47.7 47.3 47.4 38.5 Experimental Conditions: Flow Rate= 3.5GPM Pressure= 160PSIG Inlet Temp.= 16.9°C Measured Data: Notes: Current Voltage Mixing and Bulk Temperatures, °C Surface Thermocouples Temperatures, °C I x 0.5, A V, volts Tmi Tbi Tmo Tbo Ts1 Ts2 Ts3 Ts4 Ts5 Ts6 370 4.4 16.9 18.2 24.9 25.2 58.7 60 58.8 57.2 60.5 61.9 340 4.04 16.9 18.5 23.7 24.1 57.3 58 55.7 55.8 58.8 59.9 311 3.68 16.9 18.4 22.7 23 56.1 56.7 54.6 54.9 57.6 58.5 280 3.32 16.9 18.4 21.6 22.1 54.6 55.3 55 53.6 56 56.7 250 2.96 16.9 18.4 20.7 20.9 53.4 53.8 53.8 52.6 54.6 55.2 220 2.6 16.9 18.4 19.9 19.7 52.4 52.8 52.7 51.5 53.2 53.6 181 2.13 16.9 18.4 19 19 50.6 50.7 50.9 49.9 51.2 51.5 151 1.78 16.9 18.4 18.5 18.8 49 49.3 49.5 48.6 49.6 49.6 131 1.53 16.9 18.4 18.1 18.7 48.2 48.4 48.7 48.1 48.8 48 111 1.3 16.9 18.4 17.8 18.6 47.6 47.7 48 47.4 46.9 39.1 135 APPENDIX NO. 5 Tabulated Saturated Boiling Experimental Data 136 CD 0 0 IS) II cL E CD h --4—» CO CL CD co Q-(3 (0 (0 &. 0) ai E a> H-(0 _a> Q-! 3" O o o E o ro -•—* I co Q[ "ro •*-» c 0 cu CL X LU O) 6 CQ T3 Q) ro ro CO o o (0 cu fc-3 ra o>, QJ E 0) CQ TJ C 3 3 ! SL « ra o a > 5 > ra *J ra Q •o a> 3 tf) ra ~ E I LU I - T ^ 1 0 0 0 CD CO ^ CO CD IO ^ u> in IO m m CM co oo cp uo CN to uo "0 cb T t uo uo uo uo U0 CD r— O T - OO CN oo uo uo uo uo uo T t o uo T t oo UO 0 0 CD oo T t 00 T t uo uo uo OO uo CN CD CN UO UO UO UO UO UO CO 0 0 T t T - : T t m uo uo uo uo uo CD CD 0 0 CO 0 0 CO 5 ci co ^ ci T t TJ - ->a- T t OO I— T — T f • • • CM CM uo CD oo M- o CO CO CO N T - CD UO CD CD T f cd (D CO S N CO CO CO CO CO ^" co r-- uo T t CN CN ^ cd to g co oo co co T t (O N ^ CO O) OO UO CTJ OO T -CN CO CO CO ^ * o oo o o O CM CO UO UO CM CO CO CO CO CO o uo o CD uo UO CO N S N T t T t T t T t <9 eq oq T - CO UO ^j-T t T t T t h- CO CO o ;J T t TT T t T t T t r~- T t CD T t O CO ^ uo T t T t T t T t CN CD CN f-O (M t T t T t T t SL <" O) ±± ra o a > LU H E CD CO •r^ CO T t T t o!oo 00 CO uo £>iai o CN CN r - | C N CO CO CO ---j— — • — o jco CO • E cd o H J C N CO co CO i — i — — — • . . . . !— [CM CD CO 1— ^ T t uo uo i i CN CN CN — — i=ioo CM r- 0 0 I ^ I C M CO co c\i >H jCM CN CM CM S CO f— CD CM If) CO N CO CO CO 0 0 CO Q-! 0 0 O T-r-~ o T -CN CO CO CO x-1 LU I o CQ cu T t T - CD s oi 6 UO UO CD h- oo CM o q N o i d UO UO UO CD « ^ i s . CN T t uo m uo uo uo N N oi i n CO uo uo uo re CL! 3 o o LU CU co D) ±i re o a > .. s > re ra a, £ < 3 j - m ra o ° cu x S -CD 00 uo uo uo uo CD r~-00 o i n CD T f CM CO T -T t CD uo uo h- CD uo uo CO CD I CD CO •<- CN T t T t CD • T -CO O T t T t T f < ° N C £ ) W CO T t T f CO CO CO CO N CO CO CO CM CN CN CO CO CO CO i n CM a-, m T t CD . o T t T t ^ uo r-- i - CO N O) O CO CO T t T t 1 3 7 APPENDIX NO. 6 KLEA 134a Physical Property Data File Source: IC1 Chemicals and Polymers 1996 138 Information contained i n t h i s p u b l i c a t i o n or as otherwise s u p p l i e d to Users i s b e l i e v e d to be accurate and i s given i n good f a i t h , but i t i s for the user to s a t i s f y i t s e l f of the s u i t a b i l i t y of the product f o r i t s own p a r t i c u l a r purpose. ICI gives no warranty as to the f i t n e s s of the product f o r any p a r t i c u l a r purpose and any i m p l i e d warranty or c o n d i t i o n ( s t a t u t o r y or otherwise) i s excluded except to the extent that e x c l u s i o n i s prevented by law. ICI accepts no l i a b i l i t y f o r l o s s or damage (other than that a r i s i n g from death or personal i n j u r y caused by d e f e c t i v e product, i f proved) r e s u l t i n g from r e l i a n c e on t h i s i n f o r m a t i o n . Freedom under Patent, Copyright and Design cannot be assumed. Copyright (C) ICI Chemicals & Polymers 1996 UPDATED DEC91 PHYSICAL PROPERTY DATA FILE DATA FOR KLEA 134A PROPERTY UNITS VALUE Molecular Weight 102 . 03 B o i l i n g Point (latm) *C -26.22 Me l t i n g Point *C -101 C r i t i c a l Temperature *C 101. 0 C r i t i c a l Pressure bar 40.55 C r i t i c a l Density kg/mA3 509.15 Vapour Pressure ' (25*C) bar 6. 621 Latent Heat of V a p o r i s a t i o n at nBpt kJ/kg 216. 9 Saturated Vapour Density at nBpt kg/m^ 5.231 Trouton's Constant kJ/kg.K 0.8783 Coeff. V o l . Therm. Exp. (LIQ,0-20*C) *C/S-1 0.003003 Speed of Sound (sat. LIQ)(25*C) m/sec 702 A d i a b a t i c Exponent (VAP) (25*C/2.9bar) 1.142 A c e n t r i c Factor 0 . 3256 D i e l e c t r i c Constant (VAP 25*C/latm) 1. 014 D i e l e c t r i c Strength (R12=l) 0.5 S p e c i f i c R e s i s t i v i t y (LIQ) AC Mohm.cm 180 S p e c i f i c R e s i s t i v i t y (LIQ) DC Mohm.cm 66000 P u r i t y % WT 99. 98 S o l u b i l i t y In Water (20*C/latm) % wt 0. 0773 139 EQUATION OF STATE (MARTIN-HOU) X*Tr < ( A i + B i * T r + C i * exp(-KTr) ) Pr = + < Vr-B i = l , 4 (Vr-B) where : Tr = T/Tc , Pr = P/Pc, Vr = V/Vc = V*RHOc X = 3.828249 B = 0.2124913803 K = 7.250023581 Tc,Pc,RHOc = 374.15(K), 4 0 . 5 5 ( b a r ) , 509.15(kg/m A3) A1,B1,C1 = -10.72969406, 5.0713498381, -467.1552246 A2,B2,C2 = 14.034313767, -8.6811368103, -679.27189635 A3,B3,C3 = -11.321632367, 8.0404407351, 1661.953119 A4,B4,C4 = 3.201606791, -2.3711553933, -620.50640774 EXTENDED ANTOINE EQUATION B Ln(P) = A + + D*T + E*Ln (T) C + T P = vapour p r e s s u r e b a r a T = t e m p e r a t u r e K A = 123.5423 B = -5763.49 C 0 D = 0.0304565 E = -19.55224 LATENT HEAT VAPORISATION DHvap = A + Bx + Cx"2 + Dx"3 + Ex'4 where x = ( 1 — ( T / T c ) ) A ( 1 / 3 ) A 0 T = Temperature K B = 225.00842 Tc = C r i t . Temperature K C = 194.56068 D = -142.9851 DHvap k j / k g E = 57.587846 1 4 0 IDEAL GAS HEAT CAPACITY C p ( i d e a l ) = A + B*T + C*T A2 + D* T A 3 + E/T A2 A = 0.143897 T = Temperature K B = 0.003005 C =• -2.866246E-6 C p ( i d e a l ) kJ/kg.K D = 1.7812019E-9 E = 0.000000 SATURATED LIQUID ENTHALPY H l i q = A + Bx + Cx A2 + Dx A3 + E x A 4 where x = ( 1 - ( T / T c ) ) A ( 1 / 3 ) A = 289.6766 T = Temperature K B = -120.9303 Tc = C r i t . Temperature K C = 58.162283 H l i q k j / k g D = -610.8266 E = 166.65826 LIQUID DENSITY d l i q = A + Bx + Cx A2 + Dx A3 + Ex A4 where x = (1— (T/Tc) ) A ( 1 / 3 ) A = 509.14972 T = Temperature K B = 902.40197 Tc = C r i t . Temperature K C = 637.09745 d l i q kg/m A3 D = -504.3096 E = 415.76611 LIQUID VISCOSITY Ln(MU) l i q = A + B/T + C/T A2 + D/T A3 A = -8.52527 T = Temperature K B = 2984.6992 MU l i q cP C = -275736 D 0 LIQUID THERMAL CONDUCTIVITY K l i q = A + Bx + Cx A2 + Dx A3 where x = ( 1 - ( T / T c ) ) A ( 1 / 3 ) A = 0.078573 T = Temperature K B = -0.146406 Tc = C r i t . Temperature K C = 0.265347 K l i q W/m.K 1 4 1 0 SURFACE TENSION Sigma = A ( l - ( T / T c ) ) A 1 . 2 6 A = ,60.4 T '= Temperature K Tc = C r i t . Temperature K Sigma mN/m SATURATED VAPOUR DENSITY d vap = A + Bx + Cx A2 + Dx A3 + Ex A4 where x = ( 1 - ( T / T c ) ) A ( 1 / 3 ) -50 TO 0C A = -403.0344 T = Temperature K B = 5002.9036 Tc = C r i t . Temperature K C = -14972.63 d vap kg/m A3 D = 17151.088 E = -6831.522 0C TO +80C A = 455.45416 B = -493.8403 C = -1740 D = 2957.1827 E = -1108.366 VAPOUR VISCOSITY (SAT VAPOUR) MU vap = A + BT + CT A2 + DT A3 A = -.04988113 T = Temperature K B = 6.239809E-04 MU vap cP C = -2.21507E-06 D = 2.822539E-9 VAPOUR THERMAL CONDUCTIVITY (SAT VAPOUR) Kgas = A + BT + CT A2 + DT A 3 A = -.012605 T = Temperature K B = 1.09279E-04 Kgas W/m.K C = -1.604876E-07 D = 3.036155E-10 1 4 2 VAPOUR SPEED OF SOUND (SAT. VAPOUR) u = A + BT + CT A2 + DT A 3 + E/T A = 2619.19 T = Temperature K B = -14.14725 Kgas W/m.K C = 0.037823 D = -3.94551E-05 E = -171017.1 TEMP *C LIQUID ENTH < LATENT HEAT k J / kg SAT VAP LIQUID ID. GAS ENTH Cp Cp > < J/g.K > -50.00 -40.00 -30.00 -20.00 -10.00 0.00 10.00 20 . 00 25.00 30.00 40.00 50.00 60 . 00 70.00 80 . 00 35.26 47.85 60 . 60 73.53 86. 65 100.00 113.59 127.46 134.51 141.65 156.23 171.28 186.93 203.42 221.20 232.00 225.83 219.39 212.66 205.59 198.13 190.23 181.79 177.33 172 . 70 162.80 151.88 139.58 125.30 107 . 88 267.26 273.67 279.98 286.18 292.24 298.13 303.82 309.25 311.84 314 . 35 319.03 323.16 326.51 328 . 72 329.08 1.2507 1.2666 1.2838 1.3026 1.3233 1.3463 1.3723 1.4022 1. 4189 1. 4372 1. 4794 1. 5323 1. 6023 1.7028 1.8694 0.6916 0.7113 0.7308 0.7499 0 .7687 0.7872 0.8055 0.8234 0 . 8323 0.8411 0. 8586 0. 8758 0 . 8928 0. 9096 0.9262 TEMP VAPOUR LIQUID *C PRESS DENSITY Bar kg/m A3 -50 00 0 294 1444 -40 00 0 512 1416 -30 00 0 844 1388 -20 00 1 328 1358 -10 00 2 005 1327 0 00 2 924 1295 10 00 4 136 1261 20 00 5 697 1226 25 00 6 628 1207 30 00 7 669 1188 40 00 10 118 1147 50 00 13 116 1103 60 00 16 742 1053 70 oo • 21 081 997 80 00 26 229 929 LIQUID LIQ.THERM SURF VISCOSITY COND TENSION cps W/m.K raN/m 0.50 0.115 19.3 0.45 0.111 17.7 0.40 0.107 16.1 0.35 0.103 14.6 0.31 0.099 13.1 0.27 0.095 11.6 0.24 0.091 10.2 0.21 0.086 8.78 0.20 0.084 8.11 0.19 0.082 7.44 0.16 0.078 6.14 0.15 0.074 4.90 0.13 0.069 3.72 0.11 0.065 2.62 0.10 0.061 1.60 143 TEMP SAT VAP SAT VAP SAT VAP SPEED *C DENSITY VISCOSITY THERM COND OF SOUND kg/m A3 cps W/M.K m/s -50 00 1. 650 0 0104 0 0072 140 9 -40 00 2. 772 0 0110 0 0080 143 2 -30 00 4 . 431 0 0115 0 0088 144 9 -20 00 6. 791 0 0119 0 0097 146 1 -10 00 10 . 05 0 0124 0 0106 146 6 0 00 14 . 42 0 0128 0 0115 146 7 10 00 20 . 19 0 0133 0 0124 146 2 20 00 27 . 67 0 0138 0 0133 145 0 25 00 32 . 18 0 0141 0 0138 144 1 30 00 37 .26 0 0143 0 0142 143 1 40 00 49 .40 0 0150 0 0152 140 3 50 00 65 . 78 0 0157 0 0162 136 6 60 00 86 . 62 0 0165 0 0172 131 7 70 00 114 .45 0 0175 0 0183 125 7 80 00 153 .46 0 0185 0 0193 118 2 144 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0080993/manifest

Comment

Related Items