Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Experimental study of transient cooling of a hot steel plate by an impinging circular jet Meng, Qi 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_2002-0494.pdf [ 4.3MB ]
Metadata
JSON: 831-1.0090617.json
JSON-LD: 831-1.0090617-ld.json
RDF/XML (Pretty): 831-1.0090617-rdf.xml
RDF/JSON: 831-1.0090617-rdf.json
Turtle: 831-1.0090617-turtle.txt
N-Triples: 831-1.0090617-rdf-ntriples.txt
Original Record: 831-1.0090617-source.json
Full Text
831-1.0090617-fulltext.txt
Citation
831-1.0090617.ris

Full Text

E X P E R I M E N T A L S T U D Y OF T R A N S I E N T C O O L I N G OF A H O T STEEL P L A T E BY A N IMPINGING C I R C U L A R JET By Q i M e n g B. Eng. (Mechanical Eng.), Tsinghua University, 1994 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T F O R T H E D E G R E E O F M A S T E R O F A P P L I E D S C I E N C E IN T H E F A C U L T Y O F G R A D U A T E STUDIES D E P A R T M E N T O F M E C H A N I C A L E N G I N E E R I N G We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A October 2002 © Q i Meng , 2002 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. Department The University of British Columbia Vancouver, Canada (Oct. <] , 1^*7^ DE-6 (2/88) Abstract ii Abstract The steel industry produces approximately 200 million metric tons of hot-rolled steel strip annually. During the production, runout table cooling process is crucial because it strongly influences the final mechanical properties of a steel strip in a hot strip mill. The heat transfer during runout table cooling itself is considered one of the most complicated processes in the industrial world. Over the past decades much effort has been mounted worldwide to control runout table cooling operation. However, little is known about what is really happening during the water jet impingement and how far the operations are from the optimum even today. In the current study, transient heat transfer experiments of single water jet impingement on a static steel plate were achieved. The initial temperature of the steel plate, the flow rate of the water jet and the size of the nozzle were set close to the runout table condition. Jet water temperatures were set relatively high so that the film boiling phenomena can be observed. Film boiling data on the hot steel plate was experimentally and theoretically analyzed to obtain fundamental knowledge that is helpful to predict the runout table cooling process. 11 Table of Contents iii Table of Contents ABSTRACT i i TABLE OF CONTENTS i i i LIST OF TABLES vi LIST OF FIGURES vii NOMENCLATURE x ACKNOWLEDGMENTS xiii Chapter 1 INTRODUCTION 1 Chapter 2 LITERATURE REVIEW 3 2.1 Industrial runout table 3 2.2 Hydrodynamics of jet impingement 5 2.3 Jet impingement boiling heat transfer 8 2.3.1 Boiling curve 9 2.3.2 Nucleate boiling 11 2.3.3 Critical heat flux 14 2.3.4 Transition boiling and Leidenfrost point 18 2.3.5 Film boiling 24 Chapter 3 EXPERIMENTAL APPARATUS AND PROCEDURE 29 3.1 Apparatus 29 3.1.1 Pilot runout table facility 29 3.1.2 Test samples 32 iii Table of Contents iv 3.1.3 Thermocouples 32 3.1.4 Data acquisition and visual recording facilities 34 3.1.5 Flow rate calibration 35 3.2 Experimental procedures and uncertainty analysis 35 3.2.1 Preparation of test samples 35 3.2.2 Running the test 37 3.2.3 Uncertainty analysis 37 Chapter 4 DATA P R O C E S S I N G 39 4.1 Numerical method for surface heat flux calculation 39 4.1.1 Finite difference solution for temperature profiles 40 4.1.2 Evaluation of surface temperature gradient 41 4.1.3 Investigation of space and time discretizations 41 4.2 Data filtering 43 4.3 Thermal and physical properties of DQSK steel 44 4.4 Water flow parameters 46 Chapter 5 R E S U L T S A N D DISCUSSION 48 5.1 Visual observations 48 5.2 Cooling curves 52 5.3 Boiling curves 56 5.4 Effect of subcooling 62 5.5 Effect of jet flow rate 70 5.6 Effect of surface oxidation 76 5.7 Rewetting Front 78 i v Table of Contents 5.8 Correlations Chapter 6 C O N C L U S I O N S A N D F U T U R E W O R K 6.1 Conclusions 6.2 Future work B I B L I O G R A P H Y List of Tables vi List of Tables Table 3-1 Chemical composition of test samples 32 Table 3-2 Uncertainty in measurements 38 Table 4-1 Thermal conductivity of AISI 1008 steel [46] 45 Table 4-2 Experimental flow parameters at the stagnation point 47 Table 5-1 List of experiments 48 Table 5-2 Correlations for saturated convective film boiling 85 Table 5-3 Effects of factors to the comparison results 91 VI List of Figures vii List of Figures Figure 2-1 Three typical cooling systems [4] 4. Figure 2-2 Schematic of the various jet configurations [7] 6 Figure 2-3 Invisid pressure and velocity profile for a planar jet [7] 8 Figure 2-4 Schematic of the boiling curve for a saturation liquid [7] 9 Figure 2-5 Subcooled impinging water boiling curve [27] 18 Figure 2-6 Wet area fractions measured in transition boiling [32] 20 Figure 2-7 Effects of subcooling and jet velocity on boiling curves [34] 22 Figure 2-8 Boiling curve with a shoulder in the transition boiling region [34] 23 Figure 2-9 (a) For subcooling of 25 K (b) For jet velocity of 3 m/s Effects of jet velocity and subcooling on film boiling [39] 26 Figure 3-1 Schematic layout of the runout table facility at U B C 30 Figure 3-2 Test locations on the sample plate 33 Figure 3-3 Thermocouple installation for one test location 34 Figure 3-4 Schematic view of plate installation 36 Figure 4-1 Comparison of space discretization 42 Figure 4-2 Comparison of time discretization 43 Figure 4-3 Comparison of filter effect on boiling curve 44 Figure 5-1 Image at 0.5 s after the impingement of test #5 49 Figure 5-2 Image at 5.0 s after the impingement of test #5 50 Figure 5-3 Image at 14.0 s after the impingement of test #5 51 Figure 5-4 Cooling curve of test #5 53 Figure 5-5 Cooling curve of test #13 54 vii List of Figures viii Figure 5-6 Surface and internal temperature in test #5 56 Figure 5-7 Boiling curve of test #5 57 Figure 5-8 (a) cooling curve and (b) boiling curve of TC7 in test #5 58 Figure 5-9 Magnitude of surface temperature fluctuation at location 7 of test #5 .... 60 Figure 5-10 (a) Cooling curve and (b) boiling curve of stagnation in test #11 61 Figure 5-11 Magnitude of surface temperature fluctuation at stagnation of test #11 .. 62 Figure 5-12 Comparison of boiling curves of TC2 for different subcoolings 63 Figure 5-13 Comparison of boiling curves of TC4 for different subcoolings 64 Figure 5-14 Comparison of critical heat flux for different subcoolings 66 Figure 5-15 Comparison of minimum heat flux for different subcoolings 66 Figure 5-16 Comparison of initial peak of cooling rate for different subcoolings 67 Figure 5-17 Comparison of maximum cooling rate for different subcoolings 68 Figure 5-18 Time of maximum cooling rate for different subcoolings 70 Figure 5-19 Comparison of boiling curves of TC2 for different flow rates 71 Figure 5-20 Comparison of boiling curves of TC6 for different flow rates 72 Figure 5-21 Comparison of critical heat flux for different jet flow rates 73 Figure 5-22 Comparison of minimum heat flux for different jet flow rates 73 Figure 5-23 Initial peak of cooling rate for different jet flow rates 74 Figure 5-24 Maximum cooling rate for different jet flow rates 75 Figure 5-25 Effects of surface condition on the cooling curve 76 Figure 5-26 Effects of surface condition on the boiling curve 77 Figure 5-27 (a) Image at 19.2 s after the impingement in test #5 (b) Image at 22.3 s after the impingement in test #5 79 viii List of Figures ix Figure 5-28 (a) Surface temperature with progress of rewetting front in test #5 (b) Heat flux with progress of rewetting front in test #5 80 Figure 5-29 (a) Image at 30.4 s after the impingement in test #5 (b) Image at 33.7 s after the impingement in test #5 81 Figure 5-30 Rewetting front velocity at different locations 83 Figure 5-31 Rewetting front velocity with different subcooling and jet velocity (TC4, r=47.6 mm) 84 Figure 5-32 Comparison between the predicted values by film boiling correlations and the experimental data for location 8 of test #11 (ATSUb=5°C, Flow rate =151/min) 86 Figure 5-33 Comparison between the predicted values and the experimental data for parallel zone of test # 11 (AT s u b=5 °C, Flow rate= 15 1/min) 89 Figure 5-34 Comparison between the predicted values and the experimental data for parallel zone of test #5 (AT s u b=30°C, Flow rate=30 1/min) 90 Figure 5-35 Corrected subcooling at each location of test #5 (AT s u b=30°C, Flow rate =30 1/min) 91 Figure 5-36 Corrected subcooling at each location of test #9 (AT s u b=20°C, Flow rate =15 1/min) 92 ix Nomenclature x Nomenclature A Heat transfer area of the object; C H F Critical heat flux; c p, c Specific heat; D Heater diameter; D n Nozzle diameter; Dj Diameter of water jet at the stagnation point; dj Circular jet diameter or Planar jet width; F Fraction; F N B Fully developed nucleate boiling; h Heat transfer coefficient; hfg Latent heat of evaporation from liquid to vapor; hfg* Effective heat of evaporation from liquid to vapor; H Distance between nozzle exit and the surface; k Thermal conductivity; l c r Critical wavelength of two-dimensional instability; m Mass of object; M H F Min imum heat flux; O N B Onset of nucleate boiling; P N B Partially developed nucleate boiling; p Static pressure; p s The pressure at the stagnation; X Nomenclature XI p a , p»o Ambient pressure; Q Flow rate; q" Surface heat flux; q"fiim_t The total film boiling heat flux; q"f i im_c F i lm boiling heat flux via convection; q"fiim_r F i lm boiling heat flux via radiation; r Radial coordinate in plate; T w e t Rewetting temperature; t Time; T Temperature; T s a l Saturation temperature of fluid; A T s a t The difference between wall temperature and fluid saturation temperature; A T s u b The difference between fluid saturation temperatur and fluid temperature; U o o x component of free-stream velocity; u s Interfacial velocity between vapor and liquid; V n Nozzle velocity; V j Impingement velocity of the jet; v r Velocity of the rewetting front; x Streamwise coordinate along the impingement surface with origin at the stagnation point; X Property of D Q S K steel; z Thermocouple depth; p Density; xi Nomenclature o~ Surface tension; G S B Stefan-Boltzmann constant, 5.6697xl0"8 W/(m 2 K 4 ); E s u r Emissivity of the surface; ESUb Correction factor for the effect of subcooling; es Emissivity of vapor-liquid interface; a Thermal diffusivity (k/(pcp)); u. Dynamic or absolute viscosity; v Kinematic viscosity (u/p); 8 Boundary layer thickness of liquid; P Subcooling parameter; Ga Galileo number (g • ); Nu Nusselt number (hx/ki); Pr Prandtl number (v/cc); Ra Modified Rayleigh number (Ga • Pr- (p, /pv -1)) ; Re Reynolds number (ux/v); Subscripts f, 1 Fluid or liquid; im Impingement zone; j Jet; sur Surface of the plate; v, g Vapor or gas; w Heated surface wall. xii Acknowledgments X l l l Acknowledgments Thanks to Dr. D. Fraser, Dr. V. Prodanovic, Dr. M . Gadala, Dr. D. McAdam, Dr. M . Militzer, Dr. I.V. Samarasekera for granting me the opportunity to work on this project and giving me guidance and support throughout the study. Thanks to Agust Torfi Hauksson, Pat Wenman, Ross Mcleod, Gary Lockhart, Grant Caffery, Pierre Constantineau and Carl Ng for helping me with the experiments. Thanks to my husband Qing Shao, my parents Xianzhao Meng and Qin Liu, my sister Lin Meng, and my friend Greg Brown for supporting me with their encouragement and love during this project. Special thanks to my newborn daughter Alyssa Mengyi Shao for keeping me happy during the thesis writing. X l l l Chapter 1 Introduction 1 Chapter 1 Introduction Approximately 750 million metric tons of crude steel is produced worldwide annually, and around 200 million metric tons of them are hot rolled to flat products for various applications, such as in automotive, construction, or manufacturing industries. The quality of hot rolled steel strip is thus of economic importance for a broad range of industries. The steel strip is produced in hot strip mills. A typical hot strip mi l l usually consists of a reheating furnace, descaling units, roughing and finishing stands, runout table and down coiler. The finishing temperature in controlled hot rolling lies in the range of 800 - 950°C and the coiling temperature is typically within the range of 500 - 750°C [1]. The steel is cooled along the runout table. Arrays of water jets are mounted at both the top and the bottom of the moving strip. The cooling water temperature is around 18 -50°C. It has been widely recognized that the cooling process along the runout table strongly influences the final mechanical properties of the steel strip [2]. Heat transfer during water jet impingement on a hot steel plat is one of the most complicated industrial heat transfer processes. Over the past three decades researchers have taken very significant efforts attempting to quantify the thermal phenomena during runout table cooling operation and trying to control the cooling path to optimize the microstructure of the steel strip. However, even today the cooling data for industrial conditions is still very limited. In order to obtain cooling data under real runout table cooling conditions, an industrial pilot runout table facility has been built at the Centre for Metallurgical Process l Chapter 1 Introduction 2 Engineering of the University of British Columbia (UBC) . Various tests were carried out with plain carbon D Q S K steel and stainless steel SS316. Measurements were taken under the jet and at locations with different distances from the jet. The initial temperature of the hot steel plate is around 860°C. Flow rates and subcooling can be varied for different experiments. A l l tests were transient and the boiling curves within this range were obtained. The purpose of the current study is to gain more fundamental knowledge on transient boiling heat transfer occurring on runout tables. Higher cooling water temperature was used and heat flux was calculated to get complete boiling curves which consists of nucleate boiling, transition boiling, film boiling regimes, as well as maximum and minimum heat flux points. Other experimental parameters were kept consistent with the previous research in U B C in order to simulate the real runout table cooling operation. 2 Chapter 2 Literature Review 3 Chapter 2 Literature Review It is well known today that the mechanical properties of steel are strongly influenced by the cooling path. For example, to obtain the required properties, it is usually necessary to control the temperature at which the strip is coiled. In addition, it may be necessary to interrupt the cooling and hold the strip temperature for some time. These are mostly achieved by effectively controlling the runout table cooling process in hot strip mills. Several cooling methods are usually employed in hot strip mills such as laminar flow cooling, spray cooling, and water curtain cooling [3]. Some of these methods are shown on Figure 2-1. Over the past decades, numerous studies have been performed on laminar flow cooling because of its advantages of high-heat-flux applications and intermediate cooling capacity. Extensive theoretical and experimental work on jet impingement cooling has been described in various reviews [5-8]. It should be noted, however, that very limited research has been done on transient processes or near industrial conditions. 2.1 Industrial Runout Table In a typical hot strip mill, the finishing temperature of hot rolling lies in the range of 800 ~ 950°C and the coiling temperature is in the range of 500 ~ 750°C [1]. The strip is cooled, while being transported along the runout table, by the application of water jets. The runout table may be as long as 150 m in a large continuous mill. Cooling jets are usually arranged in groups or banks for convenience of control and mounted above and 3 Chapter 2 Literature Review 4 under the strip, as shown in Figure 2-1. Figure 2-1 Three typical cooling systems [4]. The cooling banks consist of headers, which apply water to the array of jet lines and, circular nozzles from which the water impinges on the steel strip. The bottom headers are normally installed less than 100 mm below the moving strip. The top headers are mounted at least 1200 mm above the strip to avoid damage by bending of the moving strip. The strip is conveyed through the runout table by arrays of rolls. These rolls are driven by D-C motors and the speed is adjustable. Water is supplied to large storage tanks that are normally higher than the headers. Water flows (or pumped, if tanks are lower than headers) to the headers and then to the nozzles by gravity. Fairly constant pressure is required to maintain sufficient and stable water flow. After application, the water is filtered and reused. The cooling water temperature is around 18 ~ 40°C. 4 Chapter 2 Literature Review 5 Figure 2-1 shows three typical water cooling systems. For laminar tube cooling, the water flows as a smooth column and creates a circular area of cooling. The tubes are usually spaced about 50 mm apart [3]. In water curtain the water flow is a smooth column of "curtain" with the thickness of 6-12 mm and creates a line of cooling on the strip. In water spray cooling, the water reaches the strip as very fine droplets and leaves a relatively larger cooling area on the hot surface. Because of the lower heat removal ability, water spray is not used as often as the other two cooling types in strip mills. 2.2 Hydrodynamics of Jet Impingement With respect to the continuous cross sections of water jets, the jet impingement can be delineated into five categories. They are free-surface jets, plunging jets, submerged jets, confined jets, and wall jets. These five configurations are presented in Figure 2-2. The free-surface jet is the one with the water that travels relatively unimpeded to the impingement surface. This happens on the runout table where the steel strip is cooled by the first rows of water jets. As the steel strip moves along the runout table, water is accumulated and a water layer is formed on the surface. In this stage, jets act as plunging type. The plunging jet is defined as a water jet with water that impinges into a pool of water covering the surface. The depth of the pool should be less than the nozzle-to-surface spacing. The other three types of jets do not commonly exist in industrial hot strip mills. 5 Chapter 2 Literature Review 6 Nozzle Gas Gas Liquid Liquid a. Free-surface b. Plunging Gas Liquid Nozzle plate -Gas Liquid c. Submerged d. Confined Nozzle Gas Liquid e. Wall (free-surface) Figure 2-2 Schematic of the various jet configurations [7] Figure 2-3 presents representative conditions for a planar, free surface jet. At the time when the jet hits the surface, the pressure is the ambient value P a and the flow velocity is the jet velocity Vj. The flow on the surface is distinguished into three hydrodynamic regions. These three regions are the stagnation region, the acceleration Chapter 2 Literature Review 7 region and the parallel flow region. The stagnation region is defined as the locations where X/WJ < 0.5, where x is the distance from the stagnation point and vv7 is the width of the jet. The streamwise velocity is zero at the stagnation point and increases almost linearly in the stagnation region. Conversely, the pressure is maximum at the stagnation point due to the dynamic contribution and decreases to the ambient value. The acceleration region includes locations where 0.5 <X/WJ < 2. In this region the pressure decreases and approaches to the ambient value within a few percent. The fluid, on the other hand, continues to accelerate to nearly the jet velocity. In the parallel flow region where X/WJ > 2, the pressure reaches the ambient value and the streamwise velocity is the same as the jet velocity. The hydrodynamic effects of impingement are no longer realized. The combination of the stagnation region and acceleration region is also named as the impingement region or impingement zone. For free surface jets, gravity accelerates the flow for downward impingement and decelerates it for upward impingement. The jet impingement velocity on the surface can be deduced from the Bernoulli equation. Furthermore, the jet diameter on the surface, which strongly influences the size of the heat transfer impingement zone, can be evaluated by the continuity equation. The gravity effect on jet velocity is significant if the nozzle-to-surface spacing is large. The pressure at the impingement region controls the local saturation temperature. 7 Chapter 2 Literature Review Nozzle Free surface Stagnation Point u«(x) A. Stagnation Region B. Acceleration Region C. Parallel-now Region B Figure 2-3 Invisid pressure and velocity profile for a planar jet [7]. 2.3 Jet Impingement Boiling Heat Transfer Water jet impingement has been used in many industrial applications. For single-phase convection, the heat transfer coefficients could exceed 10,000 W/m2-°C [7], with much higher values when boiling exists. In spite of the numerous numerical simulations and experimental tests in this area, fundamental understanding and knowledge are still 8 Chapter 2 Literature Review 9 limited. 2.3.1 Boiling Curves Figure 2-4 shows the pool boiling curve for a saturated liquid. It is obtained by plotting the surface heat flux, q", as a function of the wall superheat, A T s a t (difference between the wall temperature and the saturation temperature). cr to Single-Phase - Forced — Convection Regime Nucleate -Boiling Regime Transition "Boiling *" Regime Onset of Nocleaic Boiling Minimum Heat Flu* Wall Superheat log AT„ Figure 2-4 Schematic of the boiling curve for a saturated liquid [7]. The single-phase forced convection represents heat transfer in the absence of boiling. This occurs when the wall temperature is lower or equal to the saturation temperature of the liquid. On the industrial runout table, single-phase convective heat transfer is most likely to occur in the entrance or exit regions (air cooling) and at all Chapter 2 Literature Review 10 locations where the surface temperature is below saturations. The heat flux (q", having the units of W/m 2) of singe-phase convection is governed by Newton's law of cooling: q" = h ( T w - T f ) (2.1) where h is the heat transfer coefficient and has the units of W/m -°C, T w is the wall temperature and Tf is the fluid temperature, both have the units of °C. Nucleate boiling heat transfer starts when the wall temperature exceeds that of saturation. Discrete vapor bubbles are formed in cavities and detach from the surface in this region. The bubble motion causes the heat flux to increase by bubble evaporation and coalescence until the curve reaches the point of maximum or critical heat flux where the transition boiling regime starts (point B on the curve). In this regime, heat transfer is weakened because the large degree of bubble coalescence ultimately prevents liquid from contacting with the surface. In transition boiling region, the vapor blankets are unstable. They collapse accompanied by intermittent wetting of the surface. As the vapor film is more and more stable, the heat flux decreases to the minimum point. Heat transfer in film boiling regime is from the surface to the liquid across a vapor film. The mode of heat transfer is mainly forced convection of the vapor, with radiation becoming dominant at higher surface temperature. The complete boiling curve may be difficult to obtain when the surface is heated but not quenched because it is difficult to control the temperature in transition boiling region. The dashed line B to B ' indicates that the surface boundary condition is heat flux controlled. However, the transition boiling can be ended at any time when the vapor blanket is complete and stable. In this case it changes to film boiling through a straight line like BB ' . For transient jet impingement heat transfer, one regime of boiling may often be 10 Chapter 2 Literature Review 11 succeeded by other regimes on the same surface. For instance, at low subcoolings and high plate temperatures, the jet might be isolated from the surface by the vapor layer. As the plate temperature declines, the jet penetrates the vapor and wets the surface surrounding the stagnation point while film boiling persists at the locations farther downstream [14]. Observations also reveal that both single-phase forced convection and nucleate boiling may occur simultaneously at different locations on one surface [15]. 2.3.2 Nucleate Boiling As shown in Figure 2-4, nucleate boiling exists in the temperature range between points A and B. Point A represents the onset of nucleate boiling (ONB), where the wall superheat becomes sufficient to cause vapor nucleation at the surface. Around point A , boiling is initiated. Research on ONB is limited. In some work on planar free-surface wall jet, visual observations agree with a marked change in the slope of the q"-AT s a t data [16]. Correlations for ONB have been made and work well but were not successful to generally predict the boiling incipience. The overall accuracy of these findings may be limited because of the difficulties associated with visually resolving the smallest bubbles [17]. Miyasaka and Inada [18] obtained surface temperature and heat flux measurements at the stagnation point of a planar, free surface water jet for both single-phase convection and fully developed nucleate boiling. They defined boiling incipience to occur under conditions for which the correlated heat transfer coefficients for the respective regimes were equivalent. For subcoolings in the range of 85 < ATsub < 108°C and the three velocities investigated (1.1, 3.2, 15.3 m/s), they gave the following expression to correlate the data: i i Chapter 2 Literature Review 12 q " O N B = 1 . 4 0 x l 0 6 V n 0 5 6 (2.2) where q"oNB is the heat flux at the onset of nucleate boiling and has the units of W/m 2 , V n is the nozzle velocity and has the units of m/s. Furthermore, researchers [18] note that independent increases in jet velocity and subcooling delay the ONB to higher heat fluxes and surface temperatures. In Figure 2-4, the regime between point A and A ' is defined as partially developed nucleate boiling (PNB). It is the transition region between single-phase convection and nucleate boiling. The PNB zone consists of comparatively few nucleation sites. Some heat is transferred by single-phase processes between patches of bubbles. As the superheat increases, the number of bubble sites increases and the area for single-phase heat transfer decreases. The regime A ' to B is defined as fully developed nucleate boiling (FNB) region. Point A ' has been chosen because it marks the beginning of the linear variation of q" with A T s a t and the end of the transition from single-phase convection. In this region bubble sites cover the whole surface and the single-phase component reduces to zero. Heat flux increases mainly by bubble evaporation and coalescence. The rate of bubble formation and growth is a very sensitive function of the surface temperature. Document about FNB region has shown a very important feature that for a given fluid the heat flux in fully developed nucleate boiling depends only on the wall superheat and is independent of the jet velocity and the subcooling. Katto and Monde [19] studied FNB region over a large range of jet velocity from 5.3 to 60 m/s. They found that with saturated water, the fully developed nucleate boiling curve was independent of the jet velocity and it simply represents an extension of the data for pool boiling to higher heat 12 Chapter 2 Literature Review 13 fluxes and wall superheats. Monde and Furukawa [20] also extended the range of velocities to 0.67 m/s. Copeland [21] studied FNB with a downward-facing heated surface. He also reported that the heat flux was independent of impingement velocity (0.79-6.4 m/s). Copeland [21] investigated the effects of subcooling on FNB heat flux as well. No differences in boiling curves for water at subcoolings of 4°C and 78°C were detected. Monde and Katto [22] also examined the effects of subccoling for water with A T s u b < 30°C. They found that at higher wall superheats, the heat flux was independent of subcooling and coincided with the data of saturated boiling. They reported some deviation from the saturated results at lower wall superheats. However, the deviation could be thought of within the band of scatter. In the meanwhile, investigations on the effects of nozzle diameter and nozzle-to-surface spacing showed no dependence of fully developed nucleate boiling heat transfer on these parameters. On the other hand, the wall temperature affects the heat flux significantly in fully developed nucleate boiling regime. Monde and Katto's data [22] was correlated to the following equation: q"FNB = 4 5 0 x A T s a t 2 7 (2.3) where q "FNB is the fully developed nucleate boiling heat flux and has the units of W/m 2 , ATsat is the wall superheat and has the units of °C. Copeland's correlation for an upward facing water jet was given by q"FNB = 740 x A T s a t 2 3 (2.4) where q "FNB and A T s a t have the same meanings and the same units as those in equation (2.3). Equation (2.4) was said to be valid for wall superheats in the range of 8 ~ 31°C. 13 Chapter 2 Literature Review 14 Correlations of fully developed nucleate boiling heat flux for planar jets and submerged jets have the same form as Equation (2.3) or Equation (2.4). 2.3.3 Critical Heat Flux Point B in Figure 2-4 represents the maximum heat flux or critical heat flux (CHF). Since accurate knowledge of critical heat flux plays an important role in controlled cooling operations, it has been studied rather extensively over the past decades. CHF was assumed to occur when a vapor blanket isolates the heated surface from bulk liquid cooling leaving a thin liquid layer (sublayer or macrolayer) at the wall, and liquid entering the sublayer falls short of balancing the rate of sublayer dryout by evaporation. In free-surface single circular jet impingement, the existence of four different CHF regimes was proposed by several researchers [7]. The four regimes are V-, I-, L - and HP-regimes, depending on the specific flow conditions. In each regime, the relationship between the CHF and various system parameters is different remarkably. Among them, the I- and HP-regimes have been reported only at pressures above the atmospheric. The L-regime (limiting condition of CHF) exists when the mass flow rate is low and the ratio of heater to nozzle diameters is high. Under this condition the latent heat of a saturated jet is approximately balanced by the heat extracted from the surface. The dependence of CHF on jet velocity is of the form q " c H F ~ V„. By contrast, the liquid consumed by evaporation in V-regime is small. V-regime, which occurs at large mass flow rate, includes most of flow conditions at atmospheric pressure. Therefore it is the most widely investigated regime. The dependence of CHF on jet velocity in V-regime is of the form q " c H F ~ V n 1 / 3 . Because V-regime exists in the runout table cooling case, it will be reviewed in the 14 Chapter 2 Literature Review 15 following parts. Monde [23] presented a correlation given by Equation (2.5) after refining many of his previous studies. _ 9 a i ^ = 0 .0757(^-)° 7 2 5 (—V)1/3 Pg h fg Vj P g P f v 2 D where q"cHFis the critical heat flux and has the units of W/m 2 , p g and Pf, with the units of kg/m3, are densities of water in gas and liquid phase at saturated temperature respectively, hfg is the enthalpy of evaporation and has the units of J/kg, Vj is jet velocity at impingement and has the units of m/s, a is surface tension and has the units of N/m, D is heater diameter and dj is the jet diameter, both D and dj have the units of m. The heater diameter D is the diameter of heated impingement surface in steady state studies. This expression is valid when the saturated water jet velocity is in the range of 0.3 to 15 m/s and the heater-to-nozzle diameters is lower than 36.4. Data from other researchers is correlated well by Equation (2.5) and confirms the cube-root power dependence of CFIF on velocity. Subsequently Monde [24] developed a new correlation for a wider range of density and diameter ratios than those reported in the preceding investigations. The expression is based on a general model proposed by Haramura and Katto [25] and is expressed by: - 5 C H F _ = o.280(^)° 6 4 5 ( ) ° 3 4 3 (1 + D / d =)"°364 (2.6) P g h f g V j p g P l v 2 ( D - d j ) The definition of variables in Equation (2.6) is the same as those in Equation (2.5). It is reported that Equation (2.6) successfully correlated about 94% of the previous V-regime data to within ±20% [7]. 15 i+o.ooi nip/d^f (2.5) Chapter 2 Literature Review 16 Ishigai and Mizuno [26] found that for jet velocities between 1.3 and 9.0 m/s and subcooling between 45 to 80°C, the dependence of critical heat flux on jet velocity is of the form q " c H F ~ V n ° 3 4 and may be expressed by: q ' W = 1-42 x 104 (V„/ d / 3 4 ATsub1 1 5 (2.7) where q " c H F is the critical heat flux with the units of W/m 2 , A T s u b is the subcooling temperature with the units of °C, V n is the nozzle velocity with the units of m/s and dj is the jet diameter with the units of m. Equation (2.7) provided a direct dependence on ATSUb. but could not be used successfully for near-saturated jets. Furthermore, the inverse relationship between CFfF and the nozzle diameter is inconsistent with much of the CF£F impingement literature. The effect of subcooling on the critical heat flux is significant. The increase in subcooling leads to the increase of maximum heat flux. Equation (2.7) revealed the dependence of q " c H F ~ ATSUb 1 ' 1 5 . Ochi [55] studied water jets quenching at a velocity of 3m/s and reported a six-fold increase in CFfF with the subcooling changing from 5 to 80°C but no correlation was given. Monde and Katto [22] developed the following: = 0 . 0 7 4 5 ( ^ - ) 0 7 2 5 ( - ^ - ) 1 / 3 ( l + e J (2.8) P A K P, PfVnD ( ~ ^°'5fc.,rAZ„ P/ \PSJ 'pf-x±sub V hfs J (2.9) where q"cHF is the critical heat flux with the units of W/m 2 , p g and Pf are densities of cooling liquid in gas and liquid phase at saturated temperature respectively with the units of kg/m3, hfg is the enthalpy of evaporation and has the units of J/kg, V n is the nozzle velocity with the units of m/s, o is surface tension with the units of N/m, D is the heater 16 Chapter 2 Literature Review 17 diameter and has the units of m, cPf is the specific heat of liquid with the units of J/kg°C and eSUb is the dimensionless correction factor for the effect of subcooling. In these above two equations, the effect of subcooling on CHF is given by the correction factor £ s u b . These two equations fit their data well when the nozzle velocity is in the range of 1.7 to 26 m/s and subcooling is in the range of 3 to 30°C. Miyasaka [27] reported his tests of pool boiling and impingement boiling for critical heat flux with a planar jet and a 1.5-mm-diameter heater. The jet boiling curve of CHF is shown in Figure 2-5. The curve revealed a nearly constant heat flux region despite of the substantial increase in the wall superheat. Three regions shown on the boiling curve are defined according to the slope changes. The heat flux at the beginning of the first transition region was recognized as the critical heat flux in other literatures. The correlations for critical heat flux (being thought of the departure from nucleate boiling) was given by the author as follows: q'cHF = ^ .^(i + csev^ Xi + e^ ) (2.io) where V n is the nozzle velocity and has the units of m/s, q "cHF, pool is the critical heat flux for pool boiling with the units of W/m 2 , which is expressed by: . 0.25 CHF,poo! = 0.16p,*A (2.11) and 8 s u b is the dimensionless correction factor for the effect of subcooling: ^=0-1121 0.8 \ b { K J 1.13 (2.12) The expression is valid with the subcooling between 85°C and 108°C. 17 Chapter 2 Literature Review The effects of nozzle-to-surface distance on critical heat flux have been found to be minimal. Katto and Kunihiro [28] investigated the nozzle-to-surface spacing from 0.63 to 42 times of nozzle diameters and reported they had no effect on maximum heat flux. On the other hand, the effects of surface condition have been reported to exist and the critical heat flux is higher with the enhanced surface. 10' -C ci E Cr 10' D = 1.5 mm Tf=15 °C key Uo • 1.5 0 35 A 55.3 J 1 L_LJ_I_UJ J i i L M i l AT, !0Z sat !0 °c Figure 2-5 Subcooled impinging water boiling curve [27]. 2.3.4 Transition Boiling and Leidenfrost point In Figure 2-4 the transition boiling regime is defined by the range from point B to C . In this region with the surface temperature increasing, an unstable vapor blanket is Chapter 2 Literature Review 19 formed over the surface, which reduces the contact area of water with the surface. Large patches of vapor are released from the surface at some intervals and intermittent wetting of the surface occurs. The heat flux decreases to the minimum heat flux point, which is also called Leidenfrost point. Because of its very unstable feature, transition boiling is the least understood regime in all boiling regimes. Berenson [29] obtained a nearly complete boiling curve in his pioneering study on pool boiling heat transfer. He concluded that transition boiling is a combination of unstable film boiling and unstable nucleate boiling alternatively existing at any given location on the heating surface. He addressed the variation of heat transfer rate with temperature difference to be the result of a change in the fraction of time each boiling regime existing at a given location. Based on Berenson's presumption, Kalinin et al. [30] presented the heat flux q" in transition boiling region as follows: where q'\ and q"v are the time averaged heat flux of liquid and vapor contacts, respectively, F is the fraction of the heating surface in contact with liquid. Bjonard and Griffith [31]'s expression for transition boiling heat flux was given as: where q"cHF&nd q^MHF are critical heat flux and minimum heat flux respectively, F is also the fraction of the heating surface in contact with liquid. The experiments by Dhuga and Winterton [32] and Lee et al. [33] in 1980's strongly supported the theory mentioned above. They used a micro thermocouple to measure the fractions of wetted surface. As shown in their results (given here in Figure 2-6), the fraction of wetted surface decreases remarkably with the increase of wall q"=q"iF + q"v(l-F) (2.13) q"= q'cHFF + q"MHF ( 1-F ) (2.14) 19 Chapter 2 Literature Review 20 superheat. 10 ; 10 o 10 Z 10" 3 1 o o 10 • o o • • 0 a 0 A A • .6 A • • • • • l e a dc Chan (1985) O Roghab dc Ch«ng (1979) QAO Dhugo dc Wintarton (1985) 0 50 100 150 200 250 300 W o l I s u p e r h e a t . AT (K ) Figure 2-6 Wet area fractions measured in transition boiling [32]. Ishigai et al. [34] studied transition boiling with a planar, free-surface water jet. The jet quenched a surface with an initial temperature of 1000°C. The subcooling was from 5 to 55°C. He found that with the increase of subcooling, the boiling curves clearly shifted to higher wall superheats and higher heat fluxes, as shown in Figure 2-7(a). Moreover, the characteristics of transition boiling region were quite different along with different subcooling. At lower subcooling, the heat flux decreased sharply as the wall superheat decreased in film boiling region. When it reached the minimum heat flux point, boiling noise was heard. Then the boiling was supposed to enter the transition boiling 20 Chapter 2 Literature Review 21 region. The heat flux increased monotonously to the maximum point and frequent liquid-solid contact was observed. When subcooling was higher than 15°C, the boiling curve had a shoulder in the transition boiling region. That means, the heat flux did not increase monotonously. Instead, it had a nearly constant region despite that the wall superheat still decreased. A more detailed view of the boiling curve with a shoulder is given in Figure 2-8 (between point B and C). As the water subcooling increased, the width of the shoulder increased. In the same paper, the effects of impinging velocity were also investigated. As shown in Figure 2-7(b), for lower subcooling, the heat flux increased with the increase of velocity, but the temperature of the minimum heat flux point did not have obvious changes. For the higher subcooling, the boiling curves shifted to higher superheats and heat flux as the velocity increased. In the meanwhile, the shoulder width was larger with higher velocity. The author the following correlation for minimum heat flux: q"min = 5.40 x 104v j 0 6 0 7 ( 1 + 0.527AT s u b ) (2.15) where q "mi n is the minimum heat flux and has the units of W/m 2 , A T s u b is the subcooling temperature with the units of °C and Vj is the jet velocity at impingement with the units of m/s. 21 Chapter 2 Literature Review 22 (a) CT r ' T T T * « • o v„ - 2 t rrvs 4 r t u 0 «55'C »35 - 0 - o - « > " > f 0 o ° > ° 0 °« 4 15 -ft * 4 ^ " V " 0 0 ft v \ 0 V a \ a k fc "1 I VT a a 5f -i i i i er~ • > 1 1 1 1 1 50 100 200 500 1000 AT s a t (°C) 50 100 200 500 1000 2000 AT s a t (°C) Figure 2-7 Effects of subcooling and jet velocity on boiling curves [34]. Ishigai et al. [35] reported the same trends for a free-surface, circular water jet in another publication for the conditions of 1.5 <Vj < 12 m/s and 20<ATsut, <90°C. Ochi et al. [36] also investigated a free-surface, circular jet and their results were similar to those of Ishigai et al., as shown in Figure 2-8. A shoulder could be found in transition boiling region with higher subcooling. However, none of them explained the reason why 22 Chapter 2 Literature Review 23 the shoulder occurred. 10 £ 10' o ° D * , ••> ° 0 V 5 ( » T 4 u b 2.1 m/s 3S'C 0 o o < . 0 • V / f 3 1 % \ V - • E Ti ', , ,1 20 50 100 200 A T S 0 ( 500 1000 "c Figure 2-8 Boiling curve with a shoulder in the transition boiling region [34]. Ochi et al. [36] took the effects of nozzle diameter into account. Their correlation for minimum heat flux at the stagnation point is given by: 0.828 q m i n = 0 . 3 1 8 x l 0 ( (l + 0.383AT s u b) (2.16) where q" m i n is the minimum heat flux and has the units of W/m , A T s u b is the subcooling temperature with the units of °C, vj is the jet velocity at impingement with the units of m/s and dj is the jet diameter at impingement with the units of m. Equation (2.16) is valid when the velocities, subcooling and jet diameters were in the ranges 2 < Vj < 7 m/s, 5< ATsub ^45 °C and 5<dj <20 mm, respectively. Some researchers examined locations away from the stagnation point. They attempted to quantify the size of the rewetted region while quenching and determine the temperature of wetting front, which is the boundary of wetted and non-wetted area. Rewetting is a critical factor to control the cooling of a hot metal. As mentioned before, 23 Chapter 2 Literature Review 24 rewetting occurs when the surface temperature is low enough to allow the liquid penetrating the vapor film and wet the surface. The surface temperature at that moment is called rewetting temperature. Rewetting temperature is also the temperature at minimum heat flux point or, Leidenfrost point. Kokado et al. [14] heated a stainless steel plate to the initial temperature of 900°C. They then used a 10-mm-diameter impinging water jet to quench it. They measured the temperatures on five locations, which were equally spaced in the radial direction at 20-mm intervals, beginning from the stagnation point. They only correlated the data for the stagnation point by the following equation: where T w e t is the rewetting temperature and Tj is the jet temperature, both T w e t and Tj have the units of °C. Equation (2.17) excludes other important parameters in the correlation and may be used only in the limited condition of 5<T ; <45 °C and 2 <v7 <7 m/s. 2.3.5 Film Boiling When the surface temperature is very high, a stable vapor layer exists between the surface and the cooling liquid. The heat transfer from the solid to liquid across the vapor film is in film boiling regime, as shown in Figure 2-4, point C to point D. It has been widely accepted that film boiling heat transfer consists of forced convection of vapor and radiation through the vapor. Based on this point, Bromley [37] developed a correlation to calculate film boiling heat flux, given in Equation (2.18): where q"fiim_t, q"fiim_c and q"fiim_r represent the total film boiling heat flux, film boiling heat flux via convection and film boiling heat transfer via radiation, respectively. T w e t = 1150-8Tj (2.17) q"film_t = q"film_c + 0.75q"film_r (2.18) 24 Chapter 2 Literature Review 25 However, model predictions were lower than experimental data. Nakanishi et al. [38] presented a correlation that better fit the experimental data as: q"fiim_t = 1.74q"f,im_c + 0.75q"fiim_r (2.19) Equation (2.19) is valid in the range of 1.0 <Vn <3.17 m/s, 5 <ATsub <35 °C. Furthermore, Bromley's model [37] can also be expressed as: Qfilm_t 1 V v'3 "MHF \Qfilm_t J where q " M H F is the minimum heat flux, q"fiim_t is the total film boiling heat flux and q"fiim_r is the film boiling heat flux due to radiation: qfilm r + - - i r ' ( C - 0 (2.2D e e sur s where es is the emissivity of vapor-liquid interface and assumed to be constant at 0.96; e s u r is the emissivity of the plate, which is a function of surface temperature T w , T s a t is the saturated temperature of the liquid, and OSB is Stefan-Boltzman constant whose value is 5.6697xl0"8 W/(m 2 K 4 ) . Ishigai et al. [34] heated a steel plate to an initial temperature of approximately 1000°C, and then quenched it with a planar water jet. Their results are shown in Figure 2-7. The data shows that film boiling is affected by both subcooling and jet velocity. No film boiling was revealed when the subcooling was as high as 55°C. Along with the decrease of subcooling, the boiling curves shift to lower superheat and lower heat flux. Film boiling might be seen at subcooling below 35°C. In addition, film boiling heat flux was found to increase with the increase of jet velocity, as shown in Figure 2-7(b). Liu and Wang [39] studied on film boiling at stagnation point with a circular 25 Chapter 2 Literature Review 26 water jet impinging on a hot steel plate. Their experimental conditions were similar to those of Ishigai et al. The stainless steel was heated to about 1000°C and cooled by a water jet. The results are shown in Figure 2-9. <3* (a) (b) Figure 2-9 Effects of jet velocity and subcooling on film boiling [39]. (A) For subcooling of 25K. (B) For jet velocity of 3 m/s The data reveals the same trends of heat flux with the changes of subcooling and jet velocity as Ishishi et al. reported. In their paper they gave a correlation that is valid for higher subcooling: R e f - P r r - l k . k . A T ^ A T ^ ) 1 \l/2 1 . 4 1 4 X - (2.22) where q" is the heat flux with the units of W/m 2 , ki and k v, with the units of J/m-s-°C, is the thermal conductivity of liquid and vapor, respectively, A T s u b and AT s a t , with the units of °C, is the subcooling and superheat temperature, respectively, dj is the diameter of the jet at the impingement with the units of m, Rej is the Reynolds number of the jet at 26 Chapter 2 Literature Review 27 impingement, Pn. is the Prandtl number of liquid. Data correlated by Equation (2.22) is represented by solid lines in Figure 2-9. The dashed lines are obtained by other simulation (not given here). The experimental data of lower subcooling coincides with the calculated value better than the data of higher subcooling. Filipovic et al. [47] studied subcooled turbulent film boiling on a moving isothermal surface theoretically. He assumed that both the vapor film and the liquid film were stable and continuous, the interface velocity was constant, the thermophysical properties were constant and the effect of radiation was negligible. According to the above assumptions, they gave both the similarity and the integral solutions for location Nusselt number, as shown in Equation (2.23) and (2.24). Nu, ( PiM-i Pv^v (e'v)0(Rex) (2.23) u„, - u c N u . =C — rvm+l (l + m) ^ m+1 X- u , x i i , llv 2u r + n,u„ a u i + P ^ Pi 5 m+1 R e / m + 1 Pr^"p J (2.24) where the subscript x is the streamwise coordinate, v is vapor and 1 is liquid, u<» is the free-stream velocity, u s is the x component of the interfacial velocity, Ui is the largest velocity in the system, p is the mass density, u. is the dynamic viscosity, 5 is the velocity boundary layer thickness in liquid, 8V is the vapor-layer thickness, Re is the Reynolds number and Pr is the Prandtl number. Other variables are constants defined in their 27 Chapter 2 Literature Review 28 model. Filipovic et al. [47] compared the calculated results with the experimental results. In the experiment, a stationary steel strip was heated up to approximately 800°C and cooled by a planar water jet of 30°C. The comparison was made at the location of measurement with 400mm from the stagnation line. The integral results were generally in good agreement with the experimental result. The largest discrepancy was about 30% and existed at the lowest plate temperature. 28 Chapter 3 Experimental Apparatus and Procedure 29 Chapter 3 Experimental Apparatus and Procedure Over the past decades a lot of research work has been done on investigating jet impingement boiling heat transfer. Many experiments were performed under steady state conditions [27, 40-42]. However, in order to develop practical models for runout table, real life industrial transient conditions should be considered. Initiated by the above need, an industrial small-scale runout table facility has been designed and constructed by Liu et al. [1] in the Advanced Materials and Process Engineering Laboratory (AMPEL) at UBC. Liu [1] and Hauksson [43] performed experiments with SS316 stainless steel and DQSK (Drawing Quality Special Killed) carbon steel. They heated the steel sample up to around 930°C, and impinged water onto the metal surface. The cooling water temperatures of their experiments were 30, 40 and 50°C. In present research the cooling water temperature was increased to get a wider view of boiling heat transfer on runout table using the existing UBC facilities. 3.1 Apparatus 3.1.1 Pilot Runout Table Facility The overall view of the UBC runout table facility is presented in Figure 3-1. 29 Chapter 3 Experimental Apparatus and Procedure 30 Fump. Upper tank •Heater Orifice pressure1 gage Thermccoup! Hi  i o   p le w ire s Lower tank Figure 3-1 Schematic layout of the run out table facility at U B C The facility consists of an upper tank, a lower tank, a heater in the upper tank, a pump, a header, a nozzle, a furnace and the necessary pipe and pipe fittings. The overall height of the facility is approximately 6.5 m. The upper tank is located on a 6.5-m tower. The size of the tank is 1.5x1.5x1.0 m (length x width x height). The tank is equipped with inlet and outlet pipes and valves, as well as the overflow and drain pipes. The maximum capacity of the upper tank is 1350 1. Water is usually filled up to a depth of 0.55m and heated. A 30-kW heater lies inside the 30 Chapter 3 Experimental Apparatus and Procedure 31 upper tank. The heater can heat water up to about 65°C. In order to get much-higher water temperature for current research, an insulated lid was made of wood and polystyrene insulation and used to cover the tank. With the tank lid, a water temperature of 95°C can be achieved. An orifice fitting is installed upstream of the header and connected with a pressure gage. Flow rate is measured by the reading on the gage. The header could hold up to three nozzles. The spacing between the nozzles can be adjusted from 50 to 90 mm. In the current research only one nozzle is employed. The height of the nozzles from plate surface can be changed from 0.6 to 2.0 m. A 19-mm-diameter nozzle was made from stainless steel 304 and provided by an industrial company. A stainless steel valve is installed on the pipeline ahead of the header. The flow rate can be adjusted up to 90 1/min. The lower tank is located on the ground. It acts as a containment tank during the test period. A water pump was installed on the lower tank outlet pipeline and controlled by the water level of the lower tank. The pump is used to recycle the cooling water to the upper tank when necessary. A steel trail is mounted, between the nozzle and the lower tank, to allow the hot plate to be taken out of the furnace and settled in the right position under the nozzle. The heating furnace (240 V, 25 A , 5.8 kW, 60 Hz) was made by Lindberg/Blue M . The furnace has a high-efficiency and it may heat a steel plate to 1000°C within three hours. Nitrogen is compressed into the furnace while heating to avoid oxidation of the plate surface. 31 Chapter 3 Experimental Apparatus and Procedure 32 3.1.2 Test Samples DQSK carbon steel plates are used in the current research. The steel sheets were supplied by an industrial company. Test samples were cut from the as-received sheets to a size of 280 x 280 x 7.6 mm (Length x Width x Thickness). A l l test samples were used in the as rolled condition. They were cleaned with methanol but not polished. Each test plate was used only once. The chemical composition of the DQSK is listed in Table 3-1. Table 3-1 Chemical composition of test samples, wt% Steel Grades C Mn P s Si Cr Ni Mo Al N DQSK 0.06 0.24 0.005 0.011 0.006 - - - 0.041 0.0035 3.1.3 Thermocouples In most of the experiments, no surface temperature measurement was taken because installation of thermocouples on the surface may result in disturbance of the fluid, and may affect bubble generation and temperature data, etc. In the current study, however, surface temperature was successfully measured by carefully installing the thermocouples and properly filtering the data. Regular thermocouples entail the problem of thermal mass (the junction of joining two dissimilar wires). When the thermal mass touches an object, a small voltage is given, which can be converted to a temperature value. The larger the thermal mass is, the longer the response time. Due to the high-rate transient feature of the tests (300 °C/s), even a fraction of a second response time may not be acceptable. For this reason intrinsic thermocouples are used. For intrinsic thermocouples, the two wires are connected to the testing surface separately but close to each other. The surface between the two junctions 32 Chapter 3 Experimental Apparatus and Procedure 33 acts as the joint point. Intrinsic thermocouples do not have thermal mass. Therefore no delay time exists. Type K chromel-alumel thermocouples (Omega 304-K-Mo-1.5 mm) were employed in current study since their temperature limit is as high as 1150°C. The two metal wires of the thermocouple are insulated from each other. The diameter of the wire is 0.05 mm. A metal coat is outside the insulation, whose diameter is approximately 1.5875 mm. Eight locations on each test plate have been selected. They were distributed with a 15.87 mm intervals in a ray form as suggested by Hauksson [43] starting from the center of the plate, which is the stagnation point. The locations are shown in Figure 3-2. This distribution minimizes the effect of small flow disturbance around the surface thermocouples. #4 r = 79.4 #6 Figure 3-2 Test locations on the sample plate (units: mm) 33 Chapter 3 Experimental Apparatus and Procedure 34 For each location, two thermocouples are welded to the surface and the flat bottom of the hole under the surface point respectively. Figure 3-3 gives a schematic view of the combination of two thermocouples for one location. Once the surface and internal temperature values are obtained, heat flux at this location could be worked out. Insulation stick Figure 3-3 Thermocouple installation for one test location. Measuring surface temperature directly was not commonly used. In most of the literature [14, 34, 36, 39-40, 44], thermocouples were installed on the back surface of test samples. Then extrapolation models were employed to deduce the surface temperatures. The internal thermocouple, however, has a delay at the very beginning. Moreover, it can not catch two-phase heat transfer information on the surface. Therefore, measuring surface temperature will produce more accurate results and insight into the heat transfer process. 3.1.4 Data Acquisition and Visual recording facilities Information from thermocouples has been collected and converted to temperature 34 Chapter 3 Experimental Apparatus and Procedure 3 5 values by a PC-based data acquisition system. InstruNet sampling software package was selected for the current study. The highest frequency for 16 data samples that may be achieved by the software is 120 Hz . Due to the very high rate changes of the temperature, some peak information might be missed i f the frequency is too low. From this point, 100 H z is adopted to sample the data. A Sony video camera was used to record each test visually. The visual record is very helpful for data analysis and for determining the size of impingement zone and how soon the rewetting front spreads. 3.1.5 Flow rate calibration A s shown in Figure 3-1, the flow rate is measured by reading the pressure drop of flow after an orifice plate. The orifice plate is made of stainless steel. The diameter of the orifice is 17 mm and the corresponding range of pressure drop is 0-1.8 inches of water. Calibration experiments had been done before the tests. The result is presented by Equation 3.1 with the correlation coefficient of 0.9946. Q - 3 7 . 3 I P 0 5 3 2 7 (3.1) Where Q is the flow rate with unit of 1/min and P is the pressure drop which is expressed as inches of water. 3.2 Experimental Procedures and Uncertainty Analysis 3.2.1 Preparation of Test samples Test samples should be well prepared before the tests. At first, they were cleaned 3 5 Chapter 3 Experimental Apparatus and Procedure 36 with methyl alcohol. The depths of all flat-bottom holes were measured with a micrometer. The plate was fixed on one end of a steel pipe. Thermocouple wires went through the pipe and were connected with the InstruNet data acquisition system on the other end. Eight of the sixteen thermocouples were spot welded on the surface while the other eight were spot welded to the flat bottom of the holes. Insulations were put between the test plate and the angle iron to avoid undesired heat transfer. Figure 3-4 gives a schematic view of the connection. Plate Insulation Steel pipe Thermocouple wires Angle steel Figure 3-4 Schematic view of plate installation. The next step is spot-welding the thermocouples. The eight go-through holes lie approximately 1.6 mm downstream from the flat bottom holes. The diameter of the holes is almost the same as the diameter of the wires. The wires went through the holes until the end of the coat was at the same level as the plate surface. The wires were fixed in the holes. The two very fine metal wires (0.05 mm in diameter), whose coats have been stripped off, were led along the ray direction and welded to the surface right above the flat bottom hole separately with a distance of 1 mm. It is thought that this arrangement minimizes the effect of holes and limits the flow disturbance by the wires. As to the internal thermocouple welding, a piece of insulation stick was inserted into the hole in 36 Chapter 3 Experimental Apparatus and Procedure 37 advance. After the wire was fixed on the reverse side of the plate, the two fine metal wires were put inside the two holes of the insulation stick. This secures the separation of the two wires from each other in the hole. When the end of the wire touched the bottom of the hole, it gave a spark that initiated the welding of the wire to the bottom of the hole. This internal welding is actually more challenging than the surface one since the welding surface is hidden. 3.2.2 Running the Test The test sample was heated in the electrical furnace to the initial temperature of 940°C. When the plate reached the initial temperature, it was pulled out of the furnace and laid on the right location under the nozzle. It took approximately 10 seconds for the plate to be settled down. The plate temperature dropped slightly due to radiation and natural convection during this period. When the plate temperature reached the desired temperature (around 860°C), cooling water was turned on. Cooling water has been set to the desired flow rate and temperature before the test. InstruNet program was run once the plate was pulled out of the furnace such that the transient process can be recorded from the very beginning. 3.2.3 Error analysis Results of error analysis for the measurements of temperatures, jet flow rate and positions of thermocouple are tabulated in Table 3-2. The errors were estimated according to the error limits provided in the equipment specifications. 37 Chapter 3 Experimental Apparatus and Procedure 38 Table 3-2 Measurements errors Quantity Relative errors Temperature 5T / T < ±3% (T < 277°C ) 5 T / T = ±0.75% (T > 277°C ) Flow rate 5 Q / Q = ±0.3% Thermocouple position 8r / r = ±0.6% Thermocouple depth 5z / z = ±0.3% 38 Chapter 4 Data processing 39 Chapter 4 Data Processing Most of the previous studies in the literature used the inverse heat conduction method to evaluate the surface temperature and heat flux according to the measured internal or back-face temperature. In the current study, however, at each location of measurement, two thermocouples were installed. The surface temperature measurement provides not only direct information of the heat transfer mode on the surface, but also one more boundary condition to develop the temperature profile inside the plate. Therefore, inverse heat conduction method is not required for data processing in this study. Crank-Nicolson method is used to obtain the temperature profile inside the plate in current study. 4.1 Numerical Method for Surface Heat Flux Calculation The heat flux through the surface of the test plate can be calculated by Equation where c/"is the surface heat flux, dT/dx is the temperature gradient at the surface and k(T) is the temperature dependant conductivity of the steel. It should be noted that the temperature profile between the surface and the internal measurements is nonlinear, especially within the first second of the test. This assumption has been verified by Hauksson [43]. (4.1). (4.1) 39 Chapter 4 Data processing 40 4.1.1 Finite difference solution for temperature profiles From the definition of heat flux the surface heat flux can be expressed as: ,, rnxc dT „ s q" = —-x— (4.2) A dt where m is the mass of the object, c is the specific heat of solid, A is the heat transfer area of the object, Tis the temperature and t is the time. Furthermore, m can also be given by: m = pAdx (4.3) where pis the density of the solid. The thermal diffusivity oris defined as: a = — (4.4) pc where k is the heat conductivity of the plate. Combining Equations (4.1) to Equation (4.4), the following typical parabolic differential equation is obtained: d T d2T — = a ~ T T (4-5) dt dx2 The Crank-Nicolson method is commonly used for solving transient state problems. Once the space discretization Ax and the time discretization At are determined, Equation(4.5) can be presented in a finite difference form as follows: 7 1 + 1 " T " = " 2 T " + 7 ; " 1 ) + 2 ^ ( r - t + ' " ^ + ^ ' ( 4 ' 6 ) where k and k+1 denote successive timesteps and n is the space index. A uniform initial temperature is assumed as the necessary initial condition to solve the above equation. The measured surface and internal temperatures are considered as boundary condition for each timestep. Equation (4.6) can be rearranged as: 2 A x 2 " 1 1, Ax2)" 2 A x 2 " + 1 2AX 2 I A x 2 J " 2 A x 2 " + 1 40 Chapter 4 Data processing 41 Equation (4.7) represents a tri-diagonal matrix. Thus for every instantaneous time k of the test, an array of temperature values with space step n can be obtained by evaluating the tri-diagonal matrix. 4.1.2 Evaluation of surface temperature gradient Once the temperature profile inside the plate is available, Taylor's second order expansion is used to evaluate the surface temperature gradient. Using three temperature values to evaluate the gradient renders accurate simulation. If Ti denotes the surface temperature, and T 2 , T 3 denote the calculated temperature values at depth of Ax and 2Ax at the same time respectively, the surface temperature gradient is expressed as: dTjTl-4T2+T3 ( 4 g ) dx 2Ax 4.1.3 Investigation of space and time discretizations Figure 4-1 shows the effect of space increments on the calculated heat flux. Different space increments from 3 to 100 between the top surface reading point and the bottom of the hole reading point have been examined. In order to present the effect clearly in one figure, only three values are selected to represent the trend. It may be seen from Figure 4-1 that when n takes a smaller value, the calculated heat flux is more likely to miss some peak points. On the other hand, when n is too large, fluctuations of the result increase. According to the above investigation, and to be consistent with the previous work [43], a value of n equal to 20 is used in this work. 41 Chapter 4 Data processing 42 x10 7 3 r 2 2.5 3 3.5 4 Time [s] Figure 4-1 Comparison of space discretization (T W =70°C, Flowrate=30 1/min, Stagnation point) The effect of different time discretizations can be seen in Figure 4-2. Since the sample frequency of the data acquisition system is 100 Hz , the smallest time step which can be obtained from raw data is 0.01 s. When smaller time steps are needed, linear interpolation is applied. Figure 4-2 shows that the fluctuation from raw data can be significantly decreased when half of the test time step, 0.005 s, is used for the Crack-Nicolson method. If the time step is shortened further, no effect on decreasing the fluctuation is observed. With decreasing the At from 0.005 s to 0.0002 s, no obvious difference on the calculated heat flux values could be seen. As a result, time discretization of 0.005 s is necessary and sufficient to analyze the temperature profile and this value is employed in the current research. Chapter 4 Data processing 4? At=0.01s At=0.005s and 0.001s 5 6 Time[s] Figure 4-2 Comparison of time discretization (TW=70°C, Flowrate=30 1/min, Stagnation point) 4.2 Data filtering Figure 4-2 shows that although decrease of the time step reduces the fluctuation of surface heat flux to some degree, much undesired fluctuation still exists. Filtering of the data is, therefore, necessary. Since the internal temperature values are quite stable, filtering is applied only to surface ones. Figure 4-3 shows the difference between the effects of mean and median filter on the boiling curve. The scheme of mean filter is described by: Data(i) _ filtered = Dataii) _ unfiltered (2m + 1) (4.9) To use median filter, at first, m values before and m values after each data point, i , and data(i) itself, are sorted. Then, the result after filtering is: Data(i) _ filtered = Data(i) _ sorted (4.10) 43 Chapter 4 Data processing 44 Figure 4-3 shows that median filter reduces the fluctuation significantly, while keeping the peak information. The median value is calculated using 20 values before and after one data respectively. x 10 ? 25T i 1 unfiltered data Surface temperature fC] Figure 4-3 Comparison of filter effect on boiling curve (T W =70°C, Flowrate=30 1/min, Stagnation point) 4.3 Thermal and physical properties of DQSK steel To develop the heat flux values with Equation 3.1, the thermal conductivity k of the test material has to be known. In the current work all test samples are cut from D Q S K carbon steel sheets, which goes under the austenite-to-ferrite phase transformation during the test temperature range, making the calculation of thermal and physical properties more complicated. For D Q S K steel, austenite is the stable phase at high temperatures. As it is cooled down, phase transformation occurs resulting in ferrite and pearlite structure at 44 Chapter 4 Data processing 45 lower temperature. The fraction of the microconstituents depends on the cooling rate and temperature of the phase change. Associated with the phase changes, latent heat generation also occurs in the steel. However, Investigation of phase transformation and latent heat is beyond the scope of this study. Thus, in the current study latent heat from steel's phase changes is not considered for calculation. Theoretically the thermo-physical properties of DQSK steel should be calculated by Equation (4.11) [1]: m X=i^FiXi (4.11) where X is the property of DQSK steel, m is the total number of microconstituents, F, is the volume fraction of microconstituent i, X, is the property of microconstituent /. Details of phase changes and calculation of F, as function of temperatures are, however, beyond the purpose of the current study. Fortunately, the properties of the AISI 1008, plain low carbon steel, are in good agreement with the properties of DQSK steel [1]. And the thermo-physical properties of the AISI steel can be obtained from standard handbooks. Table 4-1 gives the thermal conductivity of AISI 1008 steel with different temperatures: Table 4-1 Thermal conductivity of AISI 1008 steel [46] Temp. (°C) 0 100 200 300 400 500 600 700 800 1000 k (W/m°C) 59.5 57.8 53.2 49.4 45.6 41.0 36.8 33.1 28.5 27.6 Based on the data of Table 4-1, a linear correlation function of conductivity k with temperature T has been developed by Hauksson [43]: k=60.571-0.03849T (4.12) The density and specific heat of AISI 1008 steel can be considered constant within the 45 Chapter 4 Data processing 46 experimental temperature range. The values are given as: density p = 7800 kg/m , and specific heat c = 470 J/kg°C. 4.4 Water flow parameters As described in Chapter 2, in the region near the stagnation point during water jet impingement, a pressure gradient is developed which leads to the changes of water flow velocity. This region where the radial velocity u(r) is proportional to the distance from the stagnation point is defined as the hydraulic impingement zone. The flow parameters in the impingement zone may be evaluated with Equation (4.13) through Equation (4.15). From the momentum and energy equation, the jet impact velocity at the stagnation point is expressed as: v j=A/v n 2+2gH (4.13) where V„ is the nozzle velocity, and H is the distance between nozzle exit and the plate. The diameter of water jet at the stagnation point is: D J = D n ^ (4.14) where Dn is the nozzle diameter. The pressure at the stagnation point is: P s = P ~ + ^ P | V ? (4.15) where is the atmospheric pressure and pi is the density of water. The pressure and normal component of the velocity are maximum at the stagnation point and they decrease to zero at the edge of hydraulic impingement zone. The pressure in the impingement zone 46 Chapter 4 Data processing 47 is determined by: Pjr)=p„+±p,v;m{r) (4.16) where v,,„ is the velocity in normal direction. Usually v,,„ or Pim distribution is obtained from experimental measurement. The saturation temperature T s a t , which depends on the pressure value, can be obtained from the saturation table for water. Applying Equation (4.13) to (4.15) in the current study, the flow parameters at the stagnation point is presented in Table 4-2. Table 4-2 Experimental flow parameters at the stagnation point Flowrate D n V n Ps T s a t (1/min) (mm) (m/s) (m/s) (mm) (Pa) (°C) 15 19 0.88 5.49 7.6 116389 103.6 30 19 1.76 5.70 10.6 117556 103.9 45 19 2.64 6.03 12.6 119502 104.4 The region outside the impingement zone is called parallel flow zone. In this region the flow is fully developed and the flow velocity is equal to the jet velocity, which is expressed as u<x,=Vj. 47 Chapter 5 Results and Discussion 48 Chapter 5 Results and Discussion Fourteen tests have been done. They are presented in Table 5-1. Test #3 and test #10 are used to examine the repeatability and test #4 and Test #4a to analyze the effect of surface condition. Table 5-1 List of Experiments 60°C 70°C 80°C 95°C 15 1/min #2 #4, #4a #9 #11 30 1/min #1 #5 #8 #12 45 1/min #3, #10 #6 #7 #13 Visual observation, as well as the cooling curves and boiling curves obtained from experimental data are discussed in this chapter. The tests in Table 5-1, together with tests under higher subcooling conditions run by Hauksson [43], constitute a complete test data matrix for DQSK carbon steel, which presents boiling heat transfer from highly subcooled to nearly saturated. The comparison of test results with lower subcooling and higher subcooling is provided to give a better understanding of jet impingement boiling heat transfer. The effects of jet velocity on transient boiling heat transfer are also presented. 5.1 Visual Observations The cooling process of each test has been recorded by a video camera. The video signal was converted to digital image and compared with the data. Test #5 is selected to represent a typical cooling process. The test was done with the flow rate of 30 1/min and 48 Chapter 5 Results and Discussion 4') cooling water temperature of 70°C. The heated plate was brightly red when withdrawn from the furnace. The temperature was approximately 860°C and uniform all over the plate before the water was turned on. During the cooling process at least three distinct zones are observed. At the time when the water hit the plate, the stagnation point turned dark gray immediately, as shown in Figure 5-1. A relatively light gray zone, indicating the existence of bubbles or patches of vapor in a layer, spread outward rapidly and its size kept almost constant with its radius of about the distance from the stagnation point to the seventh thermocouple location (95 mm). A large amount of steam was generated. The stagnation zone was always darker than the gray zone, indicating more direct solid-liquid contacts there. The surface out of the gray zone kept the original red color of the plate. The gray zone is shown in Figure 5-2. Figure 5-1 Image at 0.5 s after the impingement of test #5 49 Chapter 5 Results and Discussion 50 Figure 5-2 Image at 5.0 s after the impingement of test #5 (Flow rate=30 I/min, AT s u b =30°C) Along with the formation of the initial gray zone, surface temperatures at all locations had a sudden drop and performed relatively large fluctuation during the gray zone period. It is speculated that the initial temperature drop at all locations may be associated with the initial wetting of the plate and the build-up of the vapor layer. However, the vapor fi lm was not stable and the considerable degree of fluctuation demonstrates that some liquid-solid contact occurs in the gray zone. The subcooling strongly influences the duration of this initial gray zone. The lower the subcooling, the longer the duration of the gray zone. It can be seen from the experiments with subcooling of 5°C, that the plate was not wetted even after one minute of film boiling. Around 6.8 s after the water jet first hit the plate, the zone around the stagnation point turned from dark gray to black, indicating that a complete contact of the water jet with the plate surface. Once the black zone was formed, it spread outward rapidly, and 50 Chapter 5 Results and Discussion 5 1 water splashed out of the plate from the edge of the black zone. The edge of the black zone is commonly called the rewetting front. The spreading of the black zone is shown in Figure 5-3. The progression of the rewetting front became slower with the distance from the stagnation point. After the black zone coincided with the initial gray zone, the expanding speed was remarkably lower than before. The boundary of the wetting and non-wetting zones was not sharp. Between the black zone and the almost-red surface, there was still a gray ring about 5 mm wide. When the whole plate turned black the temperature of all thermocouples were uniform and the cooling process was over. The process lasted about 1 to 3 minutes depending on the subcooling temperature. Figure 5-3 Image at 14.0 s after the impingement of test #5 (Flow rate=30 1/min, AT s u b=30°C) This is typical for all tests, especially the tests with higher subcooling (AT s u b>20°C). For tests with lower subcooling, the gray zone is lighter and almost white 51 Chapter 5 Results and Discussion 52 and stays on for a long time before the stagnation zone is rewetted. It can be speculated that the water film is insulated from the solid surface by a vapor film in this period. 5.2 Cooling Curves Cooling curves provide the information of temperature change with time. The cooling curve of surface temperature for test #5 is shown in Figure 5-4. The curves in Figure 5-4 represents the experimental data for the surface temperature measured at eight different locations on the test plate. The radial distances from the stagnation point to the measurement points are also given in Figure 5-4. The time on the horizontal axis in Figure 5-4 starts from the moment when data was first collected, and it might not be exactly the same as the onset of the impingement. It can be seen in Figure 5-4 that before cooling water hit the plate, the temperature of the plate was fairly uniform. Curves of all eight locations coincided well. The temperature dropped down gradually due to radiation and air convection heat transfer. As soon as the water hit the plate, the surface temperatures of all locations had an immediate and remarkable drop. The behavior of thermocouples at location 1 and 2 were similar and quite different from all the other points. The temperature drop at locations 1 and 2 has the largest value and indicates that both locations are in the impingement zone where heat flux is highest. The magnitudes of initial drops at locations 3-8 were getting smaller with increased distance from the stagnation. Visual observation indicates that surface boiling occurs as soon as the water impinges on the hot plate. Instantaneous wetting of surface resulted in the sudden initial drop of temperature. 52 Chapter 5 Results and Discussion S3 10 - onset of impingement t1 - 0.5 s after impingement t2 - 5.0 s after impingement t3 - 6.8 s after impingement t4 - 14 s after impingement TC1 - stagnation TC2- r=15.9mm TC3 - r=31.8 mm TC4 - r=47.6 mm TC5 - r=63.5 mm TC6 - r=79.8 mm TC7 - r=95.3 mm TC8- r=1t1.1 mm 40 50 60 Time [s] 100 Figure 5-4 Cooling curve of test #5 (Flow rate=30 1/min, AT s u b =30°C) After the initial drop, temperature fluctuation can be seen at all locations. It is believed that after a finite time, unstable vapor fi lm is generated and water penetrated the vapor film occasionally. If the subcooling is lower (i.e. 5°C) this period lasts much longer, as shown in Figure 5-5. The magnitude is lower (Figure 5-5) for lower subcooling. When the surface temperature in the impingement zone was low enough not to support the existence of a vapor film, the temperatures at the first two locations showed a second sudden drop. According to the visual record, this indicates the rewetting of the surface. From the visual information, it was seen that after a very short time (approximately 2 s), the impingement zone turned from gray into black at the time when the curves at locations 1 and 2 changed their slopes. Single phase convection heat transfer was the dominant heat transfer mode in the black zone. The critical heat flux occurred in the gray zone right before the black zone was formed. Nucleate boiling existed after it. 53 Chapter 5 Results and Discussion S4 TC1 - stagnation TC2 - r=15.9 mm TC3 - r=31.8 mm TC4 - r=47.6 mm TC5 - r=63.5 mm TC6 - r=79.8 mm TC7 - r=95.3 mm TC8- r=111.1 mm 01 I I i I I I I I I 0 20 40 60 80 100 120 140 160 180 Time [s] Figure 5-5 Cooling curve of test #13 (Flow rate=45 1/min, AT s u b=5°C) Once the stagnation was rewetted, there was less water on the surface outside the rewetting zone due to water splashing from the rewetting front. This could be the reason why the curves of other locations showed a trend of slight recovery. The water film remaining in this area became thinner and the temperature slightly increased, making it easier for a stable vapor film to be formed. After the short recovery, the temperature started decreasing smoothly. No solid-liquid contact existed and heat was transferred by radiation and convection. After a period of smooth and stable flat drop of surface temperature, a second extensive drop occurred at locations 3-8, indicating the arrival of rewetting front. The closer the distance of the location from stagnation point, the earlier the second drop occurred. Upon the arrival of the rewetting front, a temperature drop to approximately Chapter 5 Results and Discussion 55 150°C occurred within a few seconds. Transition boiling heat transfer and critical heat flux is believed to be occurring in this short period. The curves of all locations changed their slopes to much smaller values when the surface temperature reached about 150°C, suggesting a change of heat transfer mode to nucleate boiling and single phase forced convection. Finally all eight curves coincided to a horizontal line and the whole plate had a uniform surface temperature. The trend of curves discussed above is typical for all tests with cooling water temperature of 60°C to 95°C. Test #13, with subcooling temperature of 5°C and flow rate of 45 1/min (Figure 5-5), has a slightly different trend and is selected to demonstrate the general trends of nearly saturated cooling. Similar to the above tests, Figure 5-5 shows an initial temperature drop at all locations with a smaller magnitude because the lower water temperature delayed the vapor build-up. Then, a period of surface temperature fluctuation with a steady decrease was observed at all locations. Film or transition boiling occurred in this period. The fluctuation extended to a much longer period with smaller magnitude than that observed for lower cooling water temperature. The fluctuation ended with the second drop for the first two locations. For other locations, it ended with a slight recovery in temperature. The behavior of curves after the stagnation was rewetted was almost the same as that for higher subcooling. Figure 5-6 represents the typical behavior of surface and internal thermocouples at the same location of measurement. It is plotted from data of the second (representing TCs in the stagnation zone without the surface temperature recovery) and third location (representing TCs outside the stagnation zone with the surface temperature recovery) in 55 Chapter 5 Results and Discussion 5 6 test #5. The depths of the internal thermocouples were 1.03 mm and 1.02 mm respectively. The internal temperature had a more stable response with a very short delay at the onset of impingement and more gradual drop due to impingement (indicating very high heat fluxes at the beginning). The stable response is believed to be due to the fact that the internal location did not experience the two-phase heat transfer conditions evident on the surface. 1 - surface TC2 2 - internal TC2 3 - surface TC3 4 - internal TC3 20 25 Time [s] Figure 5-6 Surface and internal temperature in test #5 (Flow rate=30 1/min, AT s u b =30°C) 5.3 Boiling Curves Boil ing curve is used in boiling heat transfer to represent the heat flux as a function of the wall superheat A T s a t (difference between the surface temperature and saturation temperature). Figure 5-7 shows the boiling curves for five different locations in test #5. It represents the typical trend observed in all tests with different subcooling (AT s u b >30°C) . 56 Chapter 5 Results and Discussion 5 7 In Figure 5-7, location 1 shows the highest heat flux values but film and transition boiling regions are not clear at this location. Location 2 shows similar trend to that of location 1 with lower values of the heat flux. As expected, heat flux values decrease as the distance from the stagnation point increases because the water film thickness decrease and the water temperature rises. Due to the temperature recovery at locations 3-8, heat flux shows a peak at approximately A T s a t = 600°C, then a sudden drop occurs before bounding back to a second peak. Figure 5-8 shows both the cooling curve and the boiling curve for location 7 in test #5. The points A , B , C, D , E in both curves correspond to the same time instant. 2.5 x 10 1.5 1= | 1 0.5 TC7 TC5 • TC4 TC2 TC1 r=95.3 mm r=63.5 mm r=47.6 mm r=15.9 mm stagnation -0.5 -100 500 600 700 Figure 5-7 Boi l ing curve of test #5 (Flow rate=30 1/min, AT s u b =30°C) 5 7 Chapter 5 Results and Discussion (a) 1000 r -100 0 1 00 200 300 400 500 600 700 ATsat F°Cl Figure 5-8 (a) Cooling curve and (b) boiling curve of TC7 in test #5 (Flow rate=30 1/min, AT s u b=30°C, r=95.3 mm) 58 Chapter 5 Results and Discussion 59 The temperature at location 7 shows an initial sudden drop and the heat flux increases rapidly. Then, the hot plate starts to heat up the water on the surface showing a region of nucleate boiling followed by a region of transition boiling with unstable vapor film and large fluctuation of surface temperature accompanied with nearly constant heat flux (from point A to point B). After point B, the temperature at location 7 starts to increase and heat flux starts to decrease until the vapor film stabilized at point C where heat transfer was due to convection and radiation until point D which marks the start of the rewetting front at this location. A l l locations changed to nucleate boiling around the temperature of point E. Slopes changes in both cooling curve and boiling curve after point E indicate the beginning of single phase heat transfer mode. Xu et al. [48] indicated that the magnitude of surface temperature fluctuation may be used to identify the boiling mode. Xu et al.'s method of fluctuation analysis has been used in the current research to produce the magnitude of surface temperature fluctuation for location 7 of test #5 and the results are shown in Figure5-9. 59 Chapter 5 Results and Discussion 60 0 10 20 30 40 50 60 70 80 90 100 Time fsl Figure 5-9 Magnitude of surface temperature fluctuation at location 7 of test #5 (Flow rate=30 1/min, AT s u b=30°C, r=95.3 mm) Figure 5-10 shows both the cooling curve and the boiling curve at the stagnation in test #11. The subcooling for this test was 5°C and the flow rate was 15 1/min. Points A, B, C, D in both drawings correspond to the same time instant. In this test, the stagnation point was instantaneously wetted by the water jet at the onset of impingement. Surface temperature at location 1 dropped immediately to around 540°C (point A). At the same time, heat flux went up sharply until point A. From point A to B, the water was quickly heated by the hot plate, nucleate boiling started and vapor film generated near the surface became gradually stable. Surface temperature showed considerable fluctuation indicating some intermittent solid-liquid contacts. In the boiling curve, the heat flux began to decrease. 6 0 Chapter 5 Results and Discussion 61 (a) 900 100 10 100 ATsat [°C] 200 _ l , , u 400 600 760 Figure 5-10 (a) Cooling curve and (b) boiling curve of stagnation in test #11 (Flow rate=15 1/min, AT s u b=5°C) Around point B, the vapor film became stable. Water film was insulated from 61 Chapter 5 Results and Discussion 62 the surface for approximately one minute. The very small surface temperature fluctuation also showed that film boiling existed in this period. Heat flux dropped down to the minimum point around point C. Around point C, the surface temperature was too low to keep the vapor film and the stagnation was rewetted. After a very short transition boiling period with big temperature drop, nucleate boiling (at point D) occurred with significant heat flux decrease. Uti l izing X u et al.'s method [48], the magnitude of surface temperature fluctuation at the stagnation in test #11 is given in Figure 5-11 and it supports the above analysis. Time [s] Figure 5-11 Magnitude of surface temperature fluctuation at stagnation of test #11 (Flow rate=15 1/min, A T s u b = 5 ° C ) From the above analysis, it may be concluded that unlike tests with higher subcooling, film boiling heat transfer exists on the whole surface of the plate after the instantaneous wetting. 62 Chapter 5 Results and Discussion 6 3 5.4 Effect of Subcooling For each jet flow rate, the subcooling was varied from 5°C to 40°C in current study. Combined with the experimental data with subcooling of 50-70°C from Hauksson [43], the effect of subcooling is discussed in this section. Figure 5-12 shows the boiling curves of stagnation with fixed flow rate of 45 1/min and various subcoolings from 5 to 70°C. 7^ 50 100 ATsat [°C1 200 300 500 750 Figure 5-12 Comparison of boiling curves of T C 2 for different subcoolings (Flow rate=45 1/min) From Figure 5-12, it can be seen that when subcooling is higher than 30°C, no film boiling occurs at the impingement zone. Heat flux is higher with the higher subcooling, but the difference is small. When subcooling is lower than 30°C, film boiling is observed. The period of film boiling is longer with the decrease of subcooling. The difference between heat flux in transition and film boiling regions is significant. Kokado et al. [14] gave similar results in their study of film boiling phenomena with subcooled 6 3 Chapter 5 Results and Discussion 6 4 water. They suggested that the water at the interface could not attain the saturation temperature when the cooling water temperature was below 68°C. The data in Figure 5-12 clearly shows the shift of boiling curves to higher wall superheats and higher heat fluxes with the increase of subcooling. As described in Chapter 2, Ishigai et al. [34], Ochi et al. [36] and L i u et al. [39] presented the same trend of boiling curves at the stagnation. However, their experimental condition and data processing method are different. The absolute values cannot be compared. In Figure 5-13, the effect of subcooling on boiling curves at the location outside the impingement zone is given. ATsat f°Cl Figure 5-13 Comparison of boiling curves of T C 4 for different subcoolings (Flow rate=45 1/min) Generally, the heat flux decreases with the decrease of subcooling. The peak of the heat flux is blunter than that in the impingement zone. Even with high subcooling of 6 4 Chapter 5 Results and Discussion 65 70°C, film and transition boiling can be seen. Because of the water splashing from the rewetting front, the heat flux shows a sharp drop and ascends suddenly when the re-wetting front reaches the point. The shifting of boiling curves is not as clear as in the impingement zone. Ochi et al. [36] also obtained data outside the impingement zone in their study. Their boiling curves of those points were similar to Figure 5-13. They also observed that the splashing of water resulted in the heat flux sudden drop. The relation of the critical heat flux (CFfF) vs. the degree of jet subcooling at different locations is shown in Figure 5-14. As expected, CFfF increases with the increase of subcooling at all locations. At the stagnation region, CHF increases almost linearly. For locations outside the stagnation region, CHF increases slowly when the subcooling is low and faster as the subcooling is higher than 20°C. It indicates that the heat transfer in the outer region gets enhanced with increasing the jet subcooling. Kumagai et al. [49] in their study of transient cooling by an impinging jet reported the similar trend of CHF changes at different locations. The minimum heat flux can be seen at the stagnation region with the subcooling lower than 40°C. The effect of subcooling on the M H F is shown in Figure 5-15. Only the data of locations 1-3 were plotted in the figure because the other locations were affected by the water splashing. Their minimum heat flux points were associated with less water film and not comparable. In Figure 5-15, M H F increases with the water subcooling. There is not much difference between the M H F of location 1 and 2. M H F of location 3 is obviously lower than that of the stagnation. 65 Chapter 5 Results and Discussion 66 x107 1.8 i — 0.4 1 1 1 1 1 ' 1 5 10 15 20 25 30 35 40 A T s u b F ° C l Figure 5-14 Comparison of critical heat flux for different subcoolings (Flow rate=45 1/min) 2 I 1 I I ! ! 1 1 5 10 15 20 25 30 35 40 ATsub r°Cl Figure 5-15 Comparison of minimum heat flux for different subcoolings (Flow rate=45 1/min) Cooling rate was calculated in current study with a 0.5 s time interval as the computing time step. For all tests and all locations, there was an initial peak of cooling 66 Chapter 5 Results and Discussion 67 rate just when the water jet first hit the plate. The initial maximum cooling rate results from the instantaneous wetting of the surface. At some locations in some tests, a second peak of cooling rate existed which can be associated with the critical heat flux. Since the magnitudes of both peaks are varied with the change of subcooling, they are shown in Figure 5-16 and Figure 5-17. 1600 -o 1400 -CD CO o 1200 -°B i _ 1000 -D ) ool 800 -o E 600 -E X CC E 4 0 0 n GO -©- TC1 - stagnation -B TC2 - r=15.9 - • - TC3 - r=31.8 - A - TC5 - r=63.5 mm TC7 - r=95.3 mm 20O Figure 5-16 Comparison of initial peak of cooling rate for different subcoolings (Flow rate=45 1/min) In Figure 5-16, the relation of the peak of initial cooling rate vs. subcooling at various locations is given. The flow rate of these tests was at 45 1/min. It can be seen that the locations 1 and 2, which are assumed to be in the impingement zone, have the same trend. The initial cooling rates of these two locations increase with the subcooling. When the subcooling is below 30°C, their initial cooling rate goes up slowly from 400°C/s to 600°C/s approximately. With subcooling higher than 30°C, it increases rapidly from 600°C/s for subcooling of 30°C to 1400°C/s for subcooling of 70°C. For the locations outside the impingement zone, however, the trend is quite different. Generally, initial 67 Chapter 5 Results and Discussion 68 cooling rate decreases with distance from the stagnation at each subcooling. When subcooling is lower than 30°C, the initial cooling rates of all locations increase with subcooling. The trend is similar to that at the stagnation. As the subcooling keeps going up, the curves become flat. Figure 5-16 shows that the higher subcooling has strong effect on the cooling rate in the impingement zone and little effect outside the impingement zone. When the subcooling is relatively low, it affects the cooling rate to a lesser extent. o a) 03 b 03 -•—» CO CT .C O O o E E 'x CO E TD C o o 03 CO 550 500 450 400 350 300 250 200 1504> L i 100O 50 o TC1 - stagnation • TC2 - r=15.9 mm • TC3 - r=31.8 mm A TC5 - r=63.5 mm v TC7 - r=95.3 mm v o • o A A V • O 5 10 20 30 40 A T s u b r°Cl 50 60 70 Figure 5-17 Comparison of maximum cooling rate for different subcoolings (Flow rate=45 1/min) Figure 5-17 shows the change of the second maximum cooling rates with the subcooling. Location 1 and 2 did not show the second maximum values at all when the subcooling is higher than 40°C, so they are not given on the figure. In Figure 5-17, the second maximum cooling rate of all locations increases with subcooling. The way of the changing is not regularly. The second maximum cooling rate of location 1 and 2 is much lower than the initial value. The other locations do not show much difference compared 68 Chapter 5 Results and Discussion 69 with the initial cooling rate. Since the second maximum cooling rate is associated with the transition boiling heat transfer mode, it can be seen that the nucleate boiling is enhanced with the increase of subcooling all over the plate. The time at which the second maximum cooling rate occurred is plotted in Figure 5-18. At each location, values at various subcoolings are given. The flow rate is 45 1/min. The same as in Figure 5-17, location 1 and 2 do not have values when the subcooling is higher than 40°C. Since the time when the cooling rate reaches its second maximum value is associated with the with the arrival time of rewetting front, in Figure 5-18, the slope of each curve represents the reciprocal of the spreading velocity of rewetting front in that test. A higher gradient of the curve represents lower velocity. It can be seen from Figure 5-18 that along with the decrease of subcooling, the time before the surface is rewetted gets longer, which means longer period of film boiling and transition boiling mode. When the water temperature is nearly saturated (AT sub=5°C), the surface was not completely rewetted for approximately 80 s. As to the subcooled cooling tests, the impingement zone was rewetted within 20 s, indicating very short film and transition boiling. The slope of the curve becomes steeper as the subcooling decreases, showing that the rewetting front spreads outward more slowly when the cooling water temperature is higher. This is because the less effective cooling for low subcooling. However, with the subcooling of 5°C, the slope turns flatter again. The rewetting front advances faster when the subcooling is very low because film boiling keeps for a long time until the surface temperature all over the plate is low. Also, the effect of subcooling on the spreading of rewetting front is stronger with lower subcoolings. 69 Chapter 5 Results and Discussion 70 160 r 0 20 40 60 80 100 120 Diatance from the stagnation [mm] Figure 5-18 Time of second maximum cooling rate for different subcoolings (Flow rate=45 1/min) 5.5 Effect of Jet Flow Rate For each subcooling, the jet flow rate was varied from 15 to 45 1/min similar to the industrial range of 20-40 1/min. Hauksson [43] has demonstrated the effect of jet flow rate with high subcoolings. In this section, the effect of water flow rate on heat transfer under low subcoolings is discussed. In Figure 5-19 the boiling curves of location 2 with different jet flow rates are given. The subcooling was fixed at 5°C. It can be seen that the heat flux in film and transition boiling regions increases with the jet flow rate (velocity). The boiling curve shifted to high superheat temperature and high heat flux region with higher flow rates. Chapter 5 Results and Discussion 7 1 ATsat [°C] Figure 5-19 Comparison of boiling curves of TC2 for different flow rates. (AT s u b=5°C) Figure 5-20 shows the effect of jet flow rate on the boiling curve of location outside the stagnation zone. The subcooling is 5°C. The heat flux of transition and film boiling right after the onset of the impingement increases significantly with the jet flow rate. When the flow rate is higher, the surface temperature drops to lower value and then the heat flux sudden drop is less. This indicates that increasing the flow rate leads to higher cooling ability in the area outside the stagnation zone before the rewetting front comes. Heat transfer after the arrival of rewetting front shows lesser difference with various flow rates. 7 1 Chapter 5 Results and Discussion 72 x10c 300 500 750 100 200 ATsat [°C] Figure 5-20 Comparison of boiling curves of TC6 for different flow rates. (ATSUb=5°C) The relation of the critical heat flux vs. jet flow rate is shown in Figure 5-21. When the flow rate increases from 15 1/min to 30 1/min, the CHF at all locations rises significantly. But as the flow rate goes up to 45 1/min, the CHF decreases. This phenomenon is not only observed for the subcooling of 5°C. Comparison of the effect of flow rate with other subcoolings shows the same trend. It points out that the critical heat flux does not always increase with the jet flow rate. 13 x10" -e- TC1 stagnation 12 - a - TC3 r=31.8 mm TC5 r=63.5 mm 11 EL - V - TC7 r=95.3 mm 10 9 I" * 7 o cr 6 5 r 4 3 r 10 15 20 25 30 35 40 45 50 Jet flow rate fl/minl Figure 5-21 Comparison of critical heat flux for different jet flow rates. (AT s u b=5°C) 72 Chapter 5 Results and Discussion 73 The relation of the minimum heat flux vs. jet flow rate is given in Figure 5-22. Only the curves of location 1 and 2 are showed because the minimum heat flux at other locations is impacted by the water splashing and not comparable. Figure 5-22 shows that the minimum heat flux increases with the jet flow rate. x 1 0 u 4.5 5 3.5 2.5 - & ~ TC1 - stagnation - B - TC2 - r= 15.9 mm 10 15 20 25 30 35 40 45 50 Jet flow rate fl/minl Figure 5-22 Comparison of minimum heat flux for different jet flow rates. (ATSUb=50C) The jet flow rate affects the cooling rate as well. Cooling rate was calculated with the time interval of 0.5 s. Figure 5-23 gives the effect of jet flow rate on the initial peak of cooling rate right after the onset of impingement. At the stagnation zone, the effect is not clear. At the area outside the stagnation zone, however, the initial peak of cooling rate increases with the jet flow rate. As the flow rate was changed from 15 1/min to 30 1/min, the initial cooling rate was about double. It demonstrates that higher flow rate results in higher initial cooling intensity in the parallel zone. 73 Chapter 5 Results and Discussion 74 700 600 •500 O ° C D 400 ra E 300 [-IS 200 100 10 -e- TC1 -stagnation — E r - TC2- r=15.9 mm —A- TC3- r=31.8 mm TC4- r=47.6 mm TC5 -r=63.5 mm — TC6- r=79.8 mm 15 20 25 30 35 Row rate fl/minl 40 45 50 Figure 5-23 Initial peak of cooling rate for different jet flow rates. (AT s u b=5 0C) The second maximum cooling rate is associated with the transition boiling heat transfer. Figure 5-24 shows the change of the second cooling rate with various jet flow rate. The subcooling is 5°C. Similar to the CFfF shown in Figure 5-21, the second maximum cooling rate increases significantly when the jet flow rate is changed from 15 1/min to 30 1/min. As the flow rate keeps going up to 45 1/min, the cooling rate decreases instead of increasing with the flow rate. It indicates again that simply increasing the jet flow rate does not always enhance the heat transfer. There should be an optimum value of jet flow rate which makes the transition boiling heat transfer most intensely. When the jet flow rate is higher than that value, it will not benefit the heat transfer any more. 74 Chapter 5 Results and Discussion 75 450 400 350 a> w 300 (3 °V 15 250 c "o o 200 O 150 100 50 10 -e- TC2 - r=15.9 mm -a- TC3 - r=31.8 mm - A - TC4 - r=47.6 mm -v- TC5 - r=63.5 mm - • - TC6 - r=79.8 mm - • - TC7 - r=95.3 mm 15 20 25 30 35 Row rate il/minl 40 45 50 Figure 5-24 Maximum cooling rate for different jet flow rates. (AT s u b=5°C) The effects of jet flow rate discussed above are not particular for the subcooling of 5°C. The tests with lower subcoolings show the same trend. 5.6 Effect of Surface Oxidation In the current study test 4 and test 4a were done to examine the effect of surface oxidation on heat transfer. The same test sample plate was used for both tests. Test 4 was done first. After the plate was cooled by water, it was put in the furnace again and heated. A l l experimental conditions were the same except the surface condition of the plate. The surface of the plate was oxidized by the water after test 4 and before test 4a. An oxide existed on the surface but at some spots it was peeled off. Figure 5-25 gives the cooling curves of location 1 and location 7 in test 4 and test 4a respectively. It can be seen that the surface temperature of location 1 in test 4a had a larger initial drop than in test 4. For location 7, this difference is not clear. After the initial drop, it took a shorter time for the surface temperatures of both locations in test 4a 75 Chapter 5 Results and Discussion 7 6 to be rewetted. The surface temperature fluctuation is smaller in test 4a. After the second sudden temperature drop, the cooling curves of these two tests do not show any difference. The comparison of cooling curves demonstrates that the existence of the oxide on the surface accelerates the cooling process to some degree, especially the transition and film boiling period. 900 r The comparison of boiling curves of location 1 and location 7 in test 4 and test 4a are shown in Figure 5-26. The heat flux of location 1 in test 4a is higher at the moment of impingement and during the transition boiling period before CHF. Boil ing curves of location 7 in both tests are almost the same. For both locations, the magnitude of C H F is about the same for different tests. The curves of the two tests in nucleate boiling and single phase convection heat transfer regions coincide very well in both locations. It is indicated that the surface condition has little effect on heat transfer at locations outside the stagnation zone. In the stagnation zone, the rougher surface can enhance the heat transfer before the surface is rewetted. After the rewetting, it has very little effect. The 1 - test 4 (TC1) 2-test 4a(TC1) 3 - test 4 (TC7) 4 - test 4a(TC7) P CD Qi 1 1 1 1 1 1 I 0 10 20 30 40 50 60 70 Time [s] Figure 5-25 Effects of surface condition on the cooling curve. 7 6 Chapter 5 Results and Discussion 7 7 critical heat flux is not affected much by the surface condition. -100 0 100 200 300 400 500 600 700 ATsat f°Cl Figure 5-26 Effects of surface condition on the boiling curve. 5.7 Rewetting Front After the vapor layer was broken and the cooling surface was rewetted at the stagnation point, the rewetting front advanced outwards gradually. The arrival of rewetting front is associated with the sudden drop of surface temperature and the increase of heat flux to its maximum value at that point. The velocity of rewetting front is discussed in this section. In Figure 5-27 to Figure 5-29, locations 3 and 4 in test 5 are selected to show the change of surface temperature and heat flux associated with the rewetting front. As described in the section of visual observation, there exists a thin gray ring outside the rewetting front and progresses with the front. Figure 5-27(a) shows the moment when the edge of the gray circle reached location 3. In Figure 5-28 the surface temperature and the 77 Chapter 5 Results and Discussion 78 heat flux at that moment is marked. It can be seen that the surface temperature was dropping dramatically and the heat flux was beginning to increase sharply when the gray circle came. Figure 5-27(b) represents the moment when the rewetting front reached location 3. Figure 5-28 shows clearly that at that moment, the slope of surface temperature was changed remarkably and the heat flux just crossed the maximum value and decreased sharply. Figure 5-28 and Figure 5-29 are presented to show the behavior at location 4. The same phenomenon is observed. Actually this is a pattern found at each location. From the analysis above, it can be known transition boiling, including the maximum heat flux, occurs in the small gray ring. As the rewetting front arrives, heat transfer at that point just changes from transition boiling to nucleate boiling. A T s a t at rewetting front is about 58°C. Nucleate boiling and single phase convective heat transfer exist within the rewetted area. If film boiling occurs at one location, it should exist before the gray ring reaches the location. 78 Chapter 5 Results and Discussion Chapter 5 Results and Discussion 1000 -900 t -1 800 o 700 • I a 600 _ w i— tempe 500 face 400 Sur 300 200 100 0 10 t1 12 t3 t4 tO - onset of impingement t1 -19.2 s after impingement t2 - 22.3 s after impingement t3 - 30.4 s after impingement t4 - 33.7 s after impingement 20 30 40 Time fsl 50 60 70 Figure 5-28(a) Surface temperature with the progress of rewetting front in test #5 t3 t4 to - onset of impingement t1 -19.2 s after impingement t2 - 22.3 s after impingement t3 - 30.4 s after impingement t4 - 33.7 s after impingement 30 40 Time [s] Figure 5-28(b) Heat flux with the progress of rewetting front in test #5 so Chapter 5 Results and Discussion 81 Chapter 5 Results and Discussion 82 Once the rewetting front can be located by the slope change of the cooling curve, the velocity of the rewetting front vr can be evaluated by: V r = A r / A f (5.1) where Ar is the distance between the adjacent thermocouples and At is the time for which the rewetting front advanced between the adjacent thermocouples. Since film boiling with vapor layer is outside the gray circle and the liquid inside the rewetting zone spreads outward, it is quite sensible to assume that the thickness of the vapor layer determines the rewetting front velocity to a certain extend. When the vapor layer is thicker, it takes more time for the liquid to push away the vapor, the rewetting front velocity should be slower, vice versa. Therefore parameters affecting the thickness of the vapor layer should have an effect on the rewetting front velocity as well. Figure 5-30 shows the velocity of rewetting front at each location. The curves start from the third location because the first two are in the stagnation zone and rewetted in the same time. As shown in Figure 5-30, with higher subcooling, the velocity of rewetting front was relatively high at the third location, indicating that although the third location was not in the stagnation zone, it was wetted almost instantly after the jet hits the stagnation. The surface temperatures at locations 4 to 8 were still high when the stagnation zone was rewetted. With lower subcooling, the rewetting front advanced quite smoothly from the stagnation zone because most of the surface was cooled down when the stagnation was rewetted. Both curves reach a peak around locations 6 and 7. This coincides with the observation of the initial light gray zone, which represents the initial wetting area. When the rewetting front reached the edge of the initial wetting zone, its velocity went up sharply. After the peak the velocity slowed down significantly. 82 Chapter 5 Results and Discussion 83 10 9 8 I' c o t 6 C % 5 53 o 4 x10"' CD > - A - ATsub=30°C, Flowrate=45 l/min - • - ATsub=60°C, Flowrate=45 l/min 20 40 60 80 Distance from the stagnation [mm. 100 120 Figure 5-30 Rewetting front velocity at different locations The effect of subcooling and jet velocity on the velocity of rewetting front is shown in Figure 5-31. As seen in Figure 5-31, the velocity of rewetting front is affected strongly by both jet velocity and subcooling. The rewetting front velocity increases with the jet velocity. In general, the velocity of rewetting front increases with the subcooling because the effective cooling for higher subcooling makes the vapor layer outside the rewetting front thinner. But when the cooling water is near saturated (ATSUb=50C), film boiling exists for more than one minute and the whole surface is cooled down to low temperature before the rewetting of the stagnation. The vapor layer is very thin outside the rewetting front and then the rewetting front velocity is higher than that with subcooling of 20°C. This trend can be seen for all locations outside the stagnation zone. 83 Chapter 5 Results and Discussion 84 —•- Flowrate=45 1/min - v - Flowrate=301/min 0.61 1 1 1 1 1 1 1 5 10 15 20 25 30 35 40 ATsub [°C] Figure 5-31 Rewetting front velocity with different subcooling and jet velocity (TC4, r=47.6 mm) 5.8 Correlations Accurate prediction of the rate of heat removal from the steel strip is a precondition for successful control of runout table cooling process. Various correlations have been proposed based on different experimental results. Hauksson [43] analyzed the correlations for nucleate boiling and the critical heat flux in his study. In this section correlations for film boiling are discussed. Although film boiling is not a cooling mode with high efficiency, it generally extends over a long distance and is said to account for almost half of the heat removal along the runout table [49]. Usually on a runout table the water jets are highly subcooled. No stable film boiling exists in the impingement zone. As the steel strip moves along the runout table, water layer is formed and gets thicker on the table until being wiped at the 84 Chapter 5 Results and Discussion 85 end of the bank. The water layer is heated up by the hot surface. The subcooling becomes much lower making it possible for film boiling to occur there. Film boiling in the parallel flow zone, especially far from the stagnation point, is not affected by the jet momentum. To some degree it can be seen as the combination of typical convective film boiling heat transfer and the radiation heat transfer [37]. The convective film boiling on a horizontal plate has been the object of investigation since the 1950's. Most of the researchers applied the Rayleigh-Taylor instability theory to their analysis. Some generally accepted correlations for film boiling on the horizontal plate are given in Table 5-2. Table 5-2 Correlations for saturated convective film boiling Chang, Yan-Po [50] Nu = 0.295 f R a h f s* ) c p ( A T s a t + A T s u b ) ^ i / /3 P.J. Berenson [51] Nu = 0.672 ( h > Ra f g * c p ( A T s a t + A T s u b ) ; 1/ /4 J.A. Clark [52] Nu= 0.012 ( R a h f s* ^ c p ( A T s a t + A T s u b ) ^ 1/ /2 Lao et al. [53] i Nu=185Pr v s-0.09 R a - ^ * r c p ( A T s a t + A T s u b ) J V . V . Kimenko [54] Nu =0.19 Gi f v ( h i M l 1 Pr v / 3-0.89 Ra f g l^Pv JJ ^ c p v ( A T s a t + A T s u b ) J In Table 5-2, hfg* is the effective heat of vaporization, Ra is the modified Rayleigh number and Ga is the Galileo number. Results from those correlations compared with experimental data are shown in Figure 5-32. Figure 5-32 shows that all of the equations underpredicted heat flux with relatively large deviation. The authors of the correlations in Table 5-2 considered the 85 Chapter 5 Results and Discussion critical wavelength of the Taylor instability by assuming the vapor film was thick and the vapor velocity is equal to zero. They did not take into account the effects of fluid motion. The correlations in Figure 5-32 are for saturated steady state film boiling without impinging but the tests are subcooled and transient with an impinging jet. Ishigai et al. [34] did both steady state and transient tests for film boiling. They found that the heat flux determined by the steady state experiments was lower by about half. For the cooling tests with an impinging jet, the waviness at the interface of liquid and vapor layer is more violent, which may enhance the heat transfer, but it is neglected in the theoretical analysis. The heat transfer by condensation in subcooled experiments may be large and can not be ignored. 1 - Correlation by Chang, Y. [50] 2 - Correlation by Berenson [51] , 4 400 450 500 550 600 650 ATsat [°C] Figure 5-32 Comparison between the predicted values by fi lm boiling correlations and the experimental data for location 8 of test #11 (AT s u b =5°C, Flow rate = 15 1/min) Fil ipovic et al. [47] developed a correlation for subcooled film boiling heat transfer in parallel flow zone of the water jet. They developed the correlation specifically for the runout table, i.e. the motion of the plate was considered. Theoretically, this 86 Chapter 5 Results and Discussion 87 correlation is most suitable for the current experiments. Thus, it is discussed in detail here. The theoretical model of Filipovic et al. [47] was developed for steady-state film boiling on an isothermal surface. It assumes that (i) The wall temperature is constant; (ii) The boundary layer approximation can be applied for the vapor layer and liquid flow near the interface; (iii) The thermophysical properties of vapor and liquid are constant; (iv) The gravitational force is negligible; (v) The effect of radiation is not taken into account. Based on the two-phase boundary layer theory, the integral solution of the local Nusselt number was developed which was shown in the last part of chapter 2 (Equations 2.23 and 2.24). The correlation for film boiling heat flux by convection can be simplified as: u Re 0 ' 8 i / q"film r = 0.0195x*v x A T ^ x£L*0x(2u, + lf2 x ^ - x P r / * (5.2) My X where kv is the thermal conductivity of vapor, ATsat is the difference between wall and saturated temperature, /// and jUy are viscosity for liquid and vapor respectively, Pn is the Prandtl number of liquid, Rex is the local Reynolds number, x is the distance from the stagnation, us is the dimensionless x component of the interfacial velocity, which can be expressed as: u= -r (5.3) 1 P r 1 + 0X f p r ; V 3 Pr V V where Prv is the Prandtl number of vapor, B is named as subcooling parameter, which is expressed as: P r v xc n , xAT„, h p = _ ^ — p i sub_ ( 5 4 ) P r , x c p v x A T s a t 87 Chapter 5 Results and Discussion where c p is the specific heat. In the current study the heat transfer by radiation from the surface to the vapor-liquid interface is considered and calculated as: " _ 07SB (Tsur — Tsat) lifilm_r ~ i I \ - > - J ) + 1 £ £ sur s where GSB is Stefan-Boltaman constant, £ s u r is the emissivity of the wall and £s is the emissivity of the vapor-liquid interface. Then the total heat flux of film boiling in parallel zone corrected with Bromley's formula [37]: q"film_, = <iffihn.c + 0 - 7 5 X ^ Z m . r ( 5 - 6 ) Figure 5-33 shows the comparison between predicted values from Filipovic et al. 's correlation and the experimental data of test #11. The subcooling of test #11 was 5°C and the flow rate was 15 1/min. Data of location 5 to 8 is presented. Although the location 3 and 4 belongs to the parallel flow zone as well, film boiling was very unstable there because of the momentum of the impinging jet and the early arrival of the rewetting front. In Figure 5-33 the experimental results in film boiling region agree well with the correlated values, especially for location 7 and 8. The largest deviation for location 7 and 8 is 28% and 41.3% for location 5 and 6. The larger discrepancy for locations close to the jet could be attributed to the increased possibility of liquid-solid contact. The same trend was found by Filipovic et al. [49] when they compared the analytical results with the experimental data. They reported that the largest discrepancy was about 40% occurring at locations close to the quench front. Chapter 5 Results and Discussion 89 x 10 x 10 460 480 500 520 540 560 580 ATsat [°C] x 105 460 480 500 520 540 ATsat [°C] 5 x 10 8i • TC8- r=111.1 mm jB 2 400 450 500 550 600 ATsat [°C] 400 450 500 550 600 650 ATsat [°C] Figure 5-33 Comparison between the predicted values and the experimental data for parallel zone of test #11 (AT s u b=5°C, Flow rate = 15 1/min) From Figure 5-33 it can be seen that the predicted results are lower than the experimental data for all locations. The trend coincides with most comparison results reported. Since the theoretical model was developed for steady-state film boiling, it underpredicts the results of transient cooling experiments. It is assumed for developing the correlation that the hot surface is isothermal, but the effects of longitudinal temperature gradients and conduction in the test plate give some departure from the assumed isothermal condition. It makes the experimental results larger than the real heat transfer value. In addition, besides of film boiling, other boiling heat transfer modes may occur simultaneously during the transient cooling process, resulting in stronger heat transfer. Figure 5-33 represents the comparison results for tests with subcooling near saturation. For tests with higher subcoolings, the predicted values do not agree with the experimental, as shown in Figure 5-34. 89 Chapter 5 Results and Discussion 9 0 In Figure 5-34 each line (a) is plotted by the correlated values. It can be seen that the deviation is larger for locations further away from the stagnation. This can result from using the jet subcooling as the local subcooling at those locations. Since the heating effect by the hot plate to the water layer is not negligible but the jet subcooling is applied all over the plate, the subcoolings at further locations were higher than the practical values, which causes overprediction of heat flux at those locations. Some other reasons also cause the deviation. Investigation of other possible errors is made as follows. x 10 x 10 560 x 10 580 600 ATsat [°C] 620 580 600 620 ATsat [°C] 640 x10 *_=1.5 550 600 ATsat [°C] 650 550 600 650 ATsat [°C] 700 Figure 5-34 Comparison between the predicted values and the experimental data for parallel zone of test #5 (AT s u b =30°C, Flow rate = 30 1/min) The main factors that possibly affect the comparison results are (I) subcooling and associated properties of the cooling water; (ii) temperature of vapor layer which is applied in the correlation; (iii) the free-stream velocity which is applied in the correlation. The eighth location in test #9 (AT s u b =20°C) was selected to examine the effects of the three factors mentioned above. The results are shown in Table 5-3. 9 0 Chapter 5 Results and Discussion 91 Table 5-3 Effects of factors to the comparison results Factor Original value used in the correlation Corrected value Deviation to original result AT S U b 20°e 10°C -49.0% T J- vapor 100°C 400°C -3.7% U o o 5.5 m/s 5.0 m/s -7.5% From Table 5-3 it can be seen that compared with the subcooling, inaccurate application of other factors in the correlation equation will not result in significant errors. It is proved that local subcooling instead of jet subcooling should be used when heat flux at each location on the plate is calculated. Some simple corrections of subcooling at different locations of test #5 were made. The corrected subcoolings are shown in Figure 5-35. The corrected analysis results are shown as lines (b) in Figure 5-34. 20 40 60 80 1 00 120 Distance from the stagnation [mm] Figure 5-35 Corrected subcooling at each location of test #5 (AT s u b=30°C, Flow rate = 30 1/min) In Figure 5-35 it can be seen that the water was heated up dramatically and the subcooling at location 6 dropped down to around 17°C. Data at location 3 and 4 were not correlated due to the unstability of film boiling there. Correlated results at location 5 are 91 Chapter 5 Results and Discussion 92 difficult to be analyzed. The error between the correlated results and the test data is large at location 5 in test #5. Currently no tools are provided to predict how much error is reasonable. Using the same simple idea of correction, the corrected subcoolings of test #9 are shown in Figure 5-36. 1 1 1 1 1 "0 20 40 60 80 100 120 Distance from the stagnation [mm] Figure 5-36 Corrected subcooling at each location of test #9 (AT s u b=20°C, Flow rate = 15 1/min) The analysis of subcooling above comes just from some very rough correction. From the analysis, in order to predict the heat flux successfully with Filipovic et al. 's correlation, local subcooling should be calculated accurately at different locations on the heating surface. Since the magnitude of local subcooling is associated with the jet flow rate, the flow velocity, which is another big factor affecting the results as shown in Table 5-2, should be evaluated accurately. The minimum heat flux point on the boiling curve is also called the Leidenfrost point. It indicates the end of the film boiling and the onset of the transition boiling. It is a crucial parameter for study. Theoretically, the heat flux at Leidenfrost point increases 92 20f 18 16 O O 1 • -o 14 < 12 10 Chapter 5 Results and Discussion 93 with the jet velocity and with the water subcooling. Some researchers [34] [36] obtained Leidonfrost point in the stagnation. They studied the effects of some system parameters on Leidenfrost point, i.e. subcooling, jet velocity, jet diameter, etc. The correlations they provided, however, were just empirical and in their specific experimental range. These correlations were found not to work under current experimental conditions. Furthermore, as mentioned in the previous sections, due to the initial wetting and the water splashing, Leidenfrost point could not be determined in the parallel zone. Therefore it is not investigated here. 93 Chapter 6 Conclusions and future work 94 Chapter 6 Conclusions and Future Work In the current study experimental work was carried out to investigate water jet impingement on hot steel plates to obtain fundamental knowledge of boiling heat transfer. The subcooling was varied in a comparably low range to achieve film boiling. The effects of system parameters were discussed. The correlated results for film boiling in the parallel flow zone were compared with the experimental data. 6.1 Conclusions The direct measurement of surface and internal temperature was applied for the experiments. The surface heat flux was evaluated based on the finite difference solution (Crank-Nicolson method). The space and time discretizations were investigated and modified to get rid of unwanted fluctuations. Median filter was used to keep the filtered data with the raw data well. Visual observation of the experiments was obtained. The video signal was converted to digital image and compared with the data. The black zone represents the wetting of the heated surface. The gray zone indicates the existence transition boiling heat transfer. In the red zone heat is transferred by radiation and convection. Cooling curves and boiling curves were studied combined with the visual observation and fluctuation analysis. The first two locations were found to be in the impingement zone. They experienced mostly transition and nucleate boiling during the cooling process. At other locations vapor film was formed after the initial liquid-solid contact. Film boiling was achieved especially in the tests with lower subcooling. 94 Chapter 6 Conclusions and future work 95 Subcooling has significant effects on critical heat flux, minimum heat flux and maximum cooling rate. In the stagnation zone boiling curves shifted to higher wall superheats and higher heat fluxes with the increase of subcooling. The critical heat flux, minimum heat flux and maximum cooling rate increase with the subcooling. The effects of jet flow rate were also investigated. The minimum heat flux increased with the increase of jet flow rate. For the critical heat flux and cooling rate, however, it was found that the peak was between 15 1/min and 45 1/min, which means that increasing the jet flow rate does not always enhance the heat transfer. The effects of surface condition were briefly discussed. The rougher surface accelerated the cooling process to some degree and enhanced the heat transfer before the surface was rewetted. The critical heat flux and nucleate boiling were not affected much by the surface condition. Heat transfer before and after the arrival of rewetting front was analyzed by combining the visual observation and the experimental data. It was found that the rewetting front marked the change of the heat transfer mode from transition boiling to nucleate boiling. Transition boiling occurs inside the gray ring outside the rewetting front. Nucleate boiling and single phase convective heat transfer are in the rewetted zone. The effects of subcooling and jet flow rate on the rewetting front velocity were investigated. It was found that in general, rewetting front velocity increased with the jet flow rate and the subcooling. But when the cooling water was nearly saturated, the rewetting front velocity was high. Comparison of correlated results and experimental data was made for film boiling in the parallel flow zone. The correlations of saturated convective film boiling Chapter 6 Conclusions and future work 96 were found to underpredict the results to a considerable degree. The correlated values by Filipovic et al.'s correlation, which was developed for subcooled film boiling on the runout table, agreed with the experimental data well when the cooling water was nearly saturated. For higher subcoolings, the heating effect of the hot surface to the water layer could not be neglected. Local subcooling instead of the jet subcooling should be used in the correlation equation. 6.2 Future work The Current study aims to get more fundamental knowledge of boiling heat transfer, especially in film boiling region. Some future work is recommended as follows: Experiments with a moving bed will be more closely to the real industrial runout table cooling conditions. The flow model will be much different compared to that under a static surface condition. Experiments with water jets impinging on the hot surface covered with water layer are another work close to practical applications. In the steel mills the distance between neighboring banks is as large as 10 m. In between the banks, water layer sitting on the table has significant effects on the heat transfer. Therefore, the effects of water layer temperature, thickness and velocity need to be investigated. Experiments to investigate the size of impingement zone need to be performed. It has been found that the heat transfer in the impingement zone is different with that outside the impingement zone. Predicting the size of impingement zone will benefit the control of cooling process. More experiments need to be done with more densely varying of the 96 Chapter 6 Conclusions and future work 97 experimental conditions. For example, changing the jet flow rate to find the optimum jet flow rate. Furthermore, more accurate measurement of experimental parameters needs to be developed. It would be very helpful to the theoretical analysis. 97 Bibliography 98 Bibliography 1. Z. Liu: "Experiments and mathematical modelling of controlled runout table cooling in a hot rolling mill", Ph.D. Thesis. MMAT, UBC, 2001 2. LV.Samarasekera, D.Q.Jin, and J.K.Brimacombe: "The application of microstructural engineering to the hot-rolling of steel", 38th MWSP Conf. Proc. Vol .XXXIV, ISS-AEVIE, 1996, pp313-328 3. D.T. Blazevic: "Comparison of cooling methods: sprays, water curtains, laminar tubes and aspirated sprays", 35th MWSP Conf. Proc. Vol .XXXI, ISS-AIME, 1994, pp297-310 4. GTacke, H.Litzke, F.Raquet: "Investigations into the efficiency of cooling systems for wide-strip hot rolling mills and computer-aided control of strip cooling", Proc. Symp. on Accelerated Cooling of Steel. The Metallurgical Society of the AIME, Pittsburgh, PA, Aug. 19-21, 1985,pp35-54 5. E. K. Kalinin, I. I. Berlin, V. V. Kostyuk, "Film Boiling Heat Transfer", in Advances in Heat Transfer. Vol. 11, 1975, pp. 51-197 6. E. K. Kalinin, I. I. Berlin, V. V Kostiouk, "Transition Boiling Heat Transfer", Advances in Heat Transfer. Vol. 18, 1987, pp. 241-323 7. D.H.Wolf, F.P.Incropera, R.Viskanta: "Jet impingement boiling", Advances in Heat Transfer. Vol.23, 1993, ppl-131 8. B. W. Webb, C. F. Ma, "Single-Phase Liquid Jet Impingement Heat Transfer", Advances in Heat Transfer. Vol. 26, pp. 105-217, 1995 9. D.Q.Jin, V.H.Hernandez-Avila, LV.Samarasekera, J.K.Brimacombe: "An integrated process model for the hot rolling of plain carbon steel", 2nd Intl. Conf. On Modeling of Metal Rolling Processes. Dec, 1996, London, U K , pp36-45 10. C.Pan, J.Y.Hwang and T.L.Lin: "The mechanism of heat transfer in transition boiling", Int. J. Heat and Mass Transfer. 32-7. 1989, ppl357-1349 11. H.Fujimoto, H.Takuda, N.Hatta and R.Viskanta: "Numerical simulation of transient cooling of a hot solid by an impinging free surface jet", Numerical Heat Transfer. Part A , Vol. 36, 1999, pp767-780 12. J.Filipovic, R.Viskanta, F.P.Incropera and T.A.Veslocki: "Thermal behavior of a mov-ing steel strip cooled by an array of planar water jets", Steel Research (Germany), vol. 63, no. 10, 1992, pp438-446 98 Bibliography 99 13. N.Hatta, Y.Tanaka, H.Takuda and J.Kokado: " A numerical study on cooling process of hot steel plates by a water curtain", ISJI International. Vol. 29, No. 8, 1989, pp673-679 14. J. Kokado, N . Hatta, H . Takuda, J. Harada, N . Yasuhira: "An Analysis of Film Boiling Phenomena of Subcooled Water Spreading Radially on a Hot Steel Plate", Arch. EisenhuttenwestSteel Research). 55, Nr. 3, pp. 113-118, 1984 15. D.T.Vader, F.P.Incropera, R.Viskanta: "Convective nucleate boiling on a heated surface cooled by an impinging, planar jet of water", J. Heat Transfer. 114, 1992, pp152-160 16. S.Toda, H.Uchida: "Study of liquid film cooling with evaporation and boiling", Trans. JSME. 38, (1972), ppl830-1837 17. D.T.Vader, F.P.Incropera, R.Viskanta: " A method for measuring steady local heat transfer to an impinging liquid jet", Exp. Thermal Fliud Sci.. 4, 1991, pp 1-11 18. Y. Miyasaka, S. Inada: "The Effect of Pure Forced Convection on the Boiling Heat Transfer Between a Two-Dimensional Subcooled Water Jet and a Heated Surface", Journal of Chemical Engineering of Japan, Vol. 13, N o . l , 1980, pp. 22-28 19. Y.Katto, M.Monde: "Study of mechanism of burn-out in a high heat flux boiling system with a impingement jet", Proc. 5th Int. Heat Transfer Conf.. IV, B6.2, 1974, pp245-249 20. M . Monde and Y. Furukawa: "Critical heat flux in saturated forced convective boiling with an impinging jet coexistence of pool and forced convective boiling", Heat Trans-fer Japanese Res.. 17(5), 1988, pp31-91 21. R.J. Copeland: "Boiling heat transfer to a water jet impinging on a flat surface", Ph.D. Thesis, Southern Methodist University, Dallas T X , 1970 22. M.Monde, Y.Katto: "Burnout in a high heat flux boiling system with an impinging jet", Int. J. Heat Mass Transfer. 21-3, (1978), pp295-305 23. M . Monde: "Burnout heat flux in saturated forced convection boiling with an imping-ing jet", Heat Transfer Japanese Res.. 9(1), 1980, pp31-41 24. M . Monde: "Critical heat flux in saturated forced convective boiling on a heated disk with an impingement jet, a new generalized correlation.", Trans. JSME. 50B, 1984, ppl392-1396 25. YHaramura, Y.Katto: " A new hydrodynamic model of critical heat flux, applicable widely to both pool and forced convection boiling submerged bodies in saturated 99 Bibliography 100 liquids", Int. J. Heat Mass Transfer, Vol. 26, 1983, pp. 389-399 26. S. Ishigai and M . Mizuno: "Boiling heat transfer with an impinging water jet (about the critical heat flux)", Preprint of JSME. No. 740-16, 1974, pp 139-142 27. Y.Miyasaka, S.Inada and YOwase: "Critical heat flux and subcooled nucleate boiling in transient region between a two-dimensional water jet and a heater surface", Journal of Chemical Engineering of Japan. Vol. 13, N o . l , 1980, pp29-35 28. Y. Katto and M . Kunihiro: "Study of the mechanism of burn-out in boiling system of high burn-out heat flux", Bull. JSME. 16, 1973, ppl357-1366 29. P.J.Berenson: "Experiments on pool-boiling heat transfer'Mnt. J. Heat Mass Tranfer, Vol.5, 1962, pp985-999 30. E.K.Kalinin, I.T.Berlin, V.V.Kostyuk, E.M.Nosova: "Heat transfer during transition boiling of Cryogenic liquids", Advances in Cryogenic Engineering. Vol.21, 1976, pp273-277 31. V.K.Dhir: "Nucleate and transition boiling heat transfer under pool and external flow conditions", Proceedings of the Ninth International Heat Transfer Conference, Vol. 1, KN-9, Jerusalem, Israel, 1990,ppl29-155 32. D.S.Dhuga, R.H.S.Winterton: "Measurement of surface contact in transition boiling", Int. J. Heat Mass Transfer. Vol.28, No. 10, 1985, ppl869-1880 33. L.Y.W.Lee, C.H.Chen: "Liquid-solid contact measurements using a surface thermocouple temperature probe in atmospheric pool boiling water", Int. J. Heat Mass Transfer, Vol.28, No.8, 1985, ppl415-1423 34. S. Ishigai, S. Nakanishi, T. Ochi: "Boiling heat transfer for a plane water jet impinging on a hot surface", Proceedings of the sixth International Heat Transfer Conference. 1978. Vol. 1, FB-30, pp445-450. 35. S. Ishigai, S. Nakanishi, M . Mizuno, M . Sone: "Heat transfer of water on a high temperature surface(2nd report)", Proceedings of the 8thNational Heat Transfer Symposium of Japan, 1971, ppl45-148. 36. T.Ochi, S.Nakanishi, M.Kaji and S.Ishigai: "Cooling of a hot plate with an impinging circular water jet", Muliphase Flow and Heat transfer DI: Part A: Fundamentals, ed. By T.N.Veziroglu et al, Elsevier Science Publishers, B.V., Amsterdam, 1984, pp671-681 37. L.A.Bromley: "Heat transfer in stable film boiling", Chemical Engineering Progress, Vol.44, 1950,pp221-227 100 Bibliography 101 38. S. Nakanishi, S. Ishigai, T. Ochi, I. Morita: "Film boiling heat transfer of impingement plane water jet", Trans. JSME. 1980, 46B, pp955-961 39. Z.Liu and J.Wang: "Study on film boiling heat transfer for water jet impingement on high temperature flat plate", Int. J. Heat and Mass Transfer. 44, 2001, pp2475-2481 40. J.Stevens and B.W.Webb: "Local heat transfer coefficients under an axisymmetric, single-phase liquid jet", J. Heat Transfer, Vol 113, 1991, pp71-78 41. D.H.Wolf, F.P.Incropera and R.Viskanta: "Local jet impingement boiling heat trans-fer", Int. J. Heat and Mass Transfer. Vol 39, No. 7, 1996, pp 105-1406 42. Y.Pan and B.W.Webb: "Heat transfer characteristics of arrays of free-surface liquid jets", J. Heat Transfer. Vol. 117, 1995, pp878-883 43. A . Hauksson: "Experiments and mathematical modelling of controlled runout table cooling in a hot rolling mill", M.A.Sc. Thesis. M E C H , UBC, 2001 44. D. H . Wolf, R. Viskanta and F. P. Incorporeal: "Turbulence dissipation in a free-surface jet of water and its effect on local impingement heat transfer from a heated surface: Part 2 - local heat transfer", J. Heat Transfer. Vol 117, 1995, pp95-103 45. D. Cogan, A. Cogan: Applied numerical modelling for engineers , Oxford University Press Inc., New York , 1997, pp59-62 46. H . E. Boyer and T. L. Gall: Metals handbook desk edition. American society of metals, 1995 47. J. Filipovic, R. Viskanta, F. P. Incropera, "An Analysis of Subcooled Turbulent Film Boiling on a Moving Isothermal Surface", International Journal of Heat and Mass Transfer. Vol. 37, No. 17, 1994, pp. 2661-2673 48. J. J. Xu , K. Adham-Khodaparast and M . Kawaji "study of transition boiling heat transfer in quenching of flat surface using surface temperature and heat flux microsensors", Proceeding of convective flow boiling, USA, 1995, pp237-242 49. J. Filipovic, F. P. Incropera and R. Viskanta: "Quenching phenomena associated with a water wall jet: n. Comparison of experimental and theoretical results for the film boiling region". Experimental heat transfer, 8: 119-130, 1995 50. Y. Chang: "Wave theory of heat transfer in film boiling", J. Heat transfer, 81, 1-12, 1959 51. P.J.Bereson: "Film boiling heat transfer from a horizontal surface", J. Heat Transfer, Vol.83, 1961,pp351-358 101 Bibliography 102 52. J. A . Clark: "Cryogenic heat transfer", Advances in Heat Transfer. Vol.5, Academic Press, New York, 1968 53. Y. J. Lao, R. E. Barry and R. E. Balzhiser: "Study of fdm boiling on a horizontal plate", Proceedings of the fourth international heat transfer conference. Vol. 5, B3.10, Paris, 1970 54. V. V. Klimenko: "Film boiling on a horizontal plate — new correlation", J. Heat Mass Transfer. Vol. 24, pp. 69-79, 1981 55. T. Ochi, S. Nakanishi, M . Kaji, and S. Ishigai: "Cooling of a hot plate with an impinging circular water jet", Multi-Phase Flow and Heat Transfer HI Part A Fundamentals, pp. 671-681,1984 102 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0090617/manifest

Comment

Related Items