1、 International Journal of Heating, Ventilating, Air-conditioning and Refrigerating Research Editor John W.Mitchell, Ph.D., P.E. Professor of Mechanical Engineering University of Wisconsin-Madison, USA Associate Editors James E. Braun, Ph.D., P.E., Associate Professor, Ray W. Herrick Laboratories, Al
2、berto Cavallini, Ph.D., Professor, Dipartmento di Fisicia Tecnica, University of Padova, Italy Arthur L. Dexter, D.Phil., C.Eng., Reader in Engineering Science, Department of Leon R. Gicksman, Ph.D., Professor, Departments of Architecture and Ralph Goldman, Ph.D., Chief Scientist, Comfort Technology
3、, Inc., Framingham, Massachusetts, USA Hugo Hens, Dr.Ir., Professor, Department of Civil Engineering, Laboratory of Building Physics, Katholieke Universiteit, Belgium Anthony M. Jacobi, Ph.D. Associate Professor and Associate Director ACRC, Department of Mechanical and Industrial Engineering, Univer
4、sity of Illinois, Urbana-Champaign, USA Jean J. Lebrun, Ph.D., Professor, Laboratoire de Thermodynamique, Universit de Lige, Belgium Reinhard Radermacher, Ph.D., Professor and Director, Center for Environmental Energy Engineering, Department of Mechanical Engineering, University of Maryland, College
5、 Park, USA Jean Christophe Visier, Ph.D., Head, Centre Scientifique et Technique du Btiment Energy Management Automatic Controller Division, Mame La Valle, France School of Mechanical Engineering, Purdue University, West Lafayette, Indiana, USA Engineering Science, University of Oxford, United Kingd
6、om Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, USA Policv Committee Editorial Assistant Richard H. Rooley, chair Jack B. Chaddock Mario Costantino John W. Mitchell Frank M. Coda W. Stephen Comstock Jennifer A. Haukohi W. Stephen Comstock Robert A. Parsons, Handbook Edit
7、or Scott A. Zeh, Publishing Services Manager Nancy F. Thysell, Typographer Publisher ASHRAE Staff 02000 by the American Society of Heating. Rehigcra!ing and IS1 (instimte for Scientific information) Web Science and Rescarch Alert; and BSRiA (Buiidmg Services Rexarch and most importantly, that the co
8、ncept when integrated into the building is finan- cially feasible. One tantalizing benefit of improved environmental design is the improved pro- ductivity of the occupants. Although reliable quantitative results are scarce, even a one percent improvement in productivity will more than justi investme
9、nts in improved designs. It is distressing to American researchers and practitioners that we must look to Europe for the development of many new innovative concepts and their initial demonstration in large-scale building projects. Certainly Europeans as a whole have a higher level of environmental c
10、oncern. The substantially higher costs of energy encourage the use of energy-efficient designs. On a rel- ative basis, there is stronger support for research on improved technologies for building energy 21 1 212 HVAC e.g. the solid-liquid interface movement, the shape of the solid-liquid interface,
11、the effect of air inside the ball, etc. If a regression analysis of experimental data is used to predict the global heat transfer coeffi- cient, it can not represent the heat transfer mechanisms and the model has poor applicability. The heat flux from the ice ball to the fluid medium can be written
12、as Q = KAAr. During the sensible heat transfer stage, the resistance, 1/K, includes the external convective resistance and the conductive resistance of ball shell Rb. During the latent heat transfer stage, the resistance 1/K includes the resistances on both external and internal sides of the ice bal
13、l shell and Rb: 218 HVAC 2 + 0.732 z* = 1.75 3 ArPrp ClC2 c3 where so* = s,/d, T* = SteFo a z/dL, and z is the time since the start of discharge. The term Ar is the Archinedes number. The parameter so includes the thickness of the melted region below the ice, although it was very small. Therefore, s
14、 - so and s* - so* and Equation (10) can be written as: T* = 1.75 3 s* - 0.119* + O.73* ArRP C,C, c3 where C2 is constant, C3 depends on the temperature, and C1 = (1 - For an ice ball with diameter d 2 20 mm, Cl varies by less than 5% with d so that it can be considered to be nearly independent of d
15、. VOLUME 6, NUMBER 3, JULY 2000 22 1 Therefore, in Equation (18), only Ar depends on the ice ball diameter d. The Archinedes num- ber is proportional to d3 so that non-dimensional time is proportional to d-34. According to the definition of z*, z is proportional to z*d2 and z is proportional to d5I4
16、. From the heat energy balance during the melting process in the inside surface of the ball shell: 2 4 d3 3 8 Kiflxd Atz = -picex-( 1 - 1PF)L -0.25 where Kin is the average Kin over time T. Therefore: Ei, = d . The diameter modification coefficient , during the discharge process, can then be express
17、ed as: d 0.25 = ($) The heat exchange coefficient determined by Equations (8)-(16) and (20) applies to a wide variety of ice tanks including both vertical and horizontal with vertical flow. It also applies to different types of ice balls such as ice balls with coarse external surfaces or ice balls w
18、ith metal cores. In most cases, the external thermal resistance of the ice ball is much smaller than the internal resistance. Accordingly, the coarse external surface has little effect on the global heat transfer coefficient. Usually, the metal core is a metal bar with a diameter of less than 10 mm.
19、 Therefore, the area of the metal bar contacting the ball shell is rather small and does not increase the heat flux through the shell very much. However, for non-spherical capsules, Equations (1 1) and (12) may not apply. In this case, if the manufacturer provides curves of inlet and outlet tem- per
20、atures during charging and discharging processes, new empirical equations for Kin can be developed by a linear regression analysis of the data. Equations (1) and (2) were solved using the finite difference method with the heat exchange coefficient K determined from Equations (8)-(16) and (20) to cal
21、culate the ice tank outlet tem- perature and the frozen fraction of the ice ball (IPF) as a function of time. Decreasing the time step and increasing the number of element layers in the tank increased the accuracy of the results. Comparison of results using different time steps and number of layers
22、showed that the layer number should be such that the temperature difference between neighboring layers was no more than 0.1“C. For a normal size tank, good simulation accuracy no higher than 1% was obtained with time steps of 10 minutes. However, a larger tank thermal mass would allow a longer time
23、step. Ice Tank Model Validation Three sets of experimental data were used to validate the model. The first set was obtained from Li (1997), the second from Zhao et al. (1995), and the third from Arnold (199 1). The mea- sured inlet flow rate and inlet temperatures of the ice tank were used as input
24、conditions, and the predicted outlet temperature was compared with the test data. The three comparisons were made to evaluate the accuracy the adaptability of the model to different flow velocities, different ice ball and tank configuration, and different ice ball diameters. In the experiment by Li
25、(1997), the flow rate and inlet temperature were varied. The encapsu- lated ice tank was a horizontal cylinder, with flow from flow distributors at the top and bottom of the cylinder. Experimental data at inlet flow rates of 7, 9, 13, and 15 m3h and at different inlet temperatures are illustrated in
26、 Figures 7 through 10. The accuracy of test was the same as that in Li. Equations (1 1) and (12), determined from experimental data from this tank, were used in all the simulations of different flow rates. The results show that for both charging and dis- 222 HVAC .$:.:.:.:.:. /1 . .p&. 0 8V /,/# Fig
27、ure 1. Definition of static contact angle VOLUME 6, NUMBER 3, JULY 2000 233 Figure 2. Iilustration of falling angle, and definition of advancing and receding contact angles surface, because the test fluid will be absorbed in the porous surface. Hence, this works only for smooth surfaces, and it does
28、 not allow quantitative measurement of the contact angle. EXPERIMENTAL METHODS The five commercial coatings listed in Table 1 were evaluated fist. Then alternative surface treatments, such as oxidized surfaces or rotary brushed surfaces were prepared and tested. Fur- ther descriptions of these surfa
29、ce treatments are given in Appendix B of Hong (1996). The alter- native surface treatments are described below: Oxidization. A bare aluminum sheet 50.8 mm by 50.8 mm (2 in. by 2 in.) was oxidized in an oven at 204C (400F) for one hour. The wettability of the oxidized surface was rated by mea- suring
30、 the contact angle and was compared to that for the hydrophilic coatings. A SEM photo was also taken to assess the physical surface structure change. Thermal Etching. A propane torch flame was applied to the bare aluminum for 15 seconds at a temperature range of 270 to 300“F, and then the sample was
31、 allowed to cool until it reached room temperature. Contact angle was measured on the flame side and an SEM photo was then taken. Unidirectional Grooved Surfaces. Several different levels of brushing were used to impart unidirectional micro roughness to the bare aluminum fin stock. The samples were
32、prepared by a commercial fm in Minneapolis, MN and are described later. The degree of the roughness was measured using a surface profiler. The wettability of the coatings or surface treatments was evaluated using contact angle mea- surements. Durability was tested using a dry/wet cycling of the samp
33、les. After each 50 dry/wet cycles, the contact angle of the samples was measured. The effect of fin press oil on the contact angle was also investigated. The wettability was measured by the sessile drop method using a contact angle goniometer with 23X magnification. A screw pipette graduated from 0.
34、000 to 2.000 ml in 0.002 ml incre- ments was used to place 0.002 ml of 18 MW deionized water on the specimen. The static contact 234 Table 1. Commercial Coatings Evaluated HVACBiR RESEARCH Surface Name of Coating Coating Type 1 Bare None 2 Sama Inorganic-Organic (silicate based) 3 SP411 Inorganic-Or
35、ganic (water glass) 4 Pol y green Organic 5 Hypercore Organic 6 Aqua-Shed Inorganic-Organic (Zr based) angle was measured within one minute after the expelled drop was positioned on the sample. The results reported are for the average contact angle measured at three positions on the sample. The cont
36、act angles typically varied in the range of So and the uncertainty of contact angle mea- surement was approximately two degrees. A scanning electronic microscope (SEM) was used to examine the physical surface character- istics at each stage of the durability tests. The SEM photos of all samples were
37、 taken at 10,000 magnification. The base sample to be photographed was gold-coated to be an electrical conduc- tor that is required for the SEM. The SEM picture was taken before and after coating durability tests. To measure the surface roughness of the brushed samples, a surface profiler was used.
38、The stylus of the profiler was moved across and along the grooves of the sample. The sampling dis- tances were 1 mm. Durability of the surface treatment to maintain wetting was determined by wetidry cycling tests. A dry/wet cycle consisted of immersing the samples in a beaker containing condensate (
39、collected from wind tunnel tests), and then the samples were allowed to au dry for 30 minutes at room temperature. The contact angle was measured every 50 dry/wet cycles. The condensate in the beaker was replaced at 50 cycle intervals to prevent contamination of the water. A labora- tory analyzed th
40、e condensate properties and found that the electrical resistivity and pH were close to that of distilled water. Simulation of the evaporative oil applied to the fin surface during the fin forming operation was done to determine its effect on contact angle and on surface durability. The treated surfa
41、ce was dipped in the evaporative oil container for 20 seconds. Two different oils were used as described in Tables 4.3a and 4.3b of Hong (1996). The oils are described here as Oil A and Oil B. For Oil A, the oil contaminated surface was suspended in air for 72 hours to allow the oil to evaporate bas
42、ed on the manufacturers recommendation. For Oil B, the oil was removed by heat- ing the oily sample for 30 seconds. After measuring the contact angle, the samples were set up for dry/wet cycles. The key difference between Oil A and Oil B is that Oil A contained a surfac- tant, while Oil B did not. R
43、ESULTS Figure 3 shows static contact angle as a function of the number of wetidry cycles for two types of fin samples. The first type is the as-received material. The second type is with fin press oil applied to the as-received samples. Figure 3 shows that the initial contact angle of the as-receive
44、d hydrophilic coated fins was less than 10“. Initially the uncoated, as-received sam- ples had an 85“ contact angle. However, the type A oil contaminated sample caused a large decrease in the contact angle from 85“ to 16“. This is because the oil contains a surfactant. The contact angle on the oil c
45、ontaminated sample is very close to that on the hydrophilic coatings. Figure 3 shows the change of contact angles for up to loo0 dry/wet cycles. All samples except the oil-free, bare aluminum sample show increasing contact angle with number of drylwet VOLUME 6, NUMBER 3, JULY 2000 235 Bare AhmMum O
46、100 200 300 400 500 600 700 800 900 1000 Number of Drywet Cycles Figure 3. Contact agie as a function of number of dry/wet cycles cycles. The contact angle of the as-received Aqua-Shed sample increased to 65“. A smaller con- tact angle was observed on the oil contaminated Aqua-Shed sample after 600
47、cycles (50“), pos- sibly because the surfactant helps retain hydrophilicity. Figure 3 also shows weddry cycle results for uncoated aluminum samples. Initially the uncoated as-received samples had an 85“ contact angle. The contact angle decreases as the drylwet cycling progresses. After 600 dry/wet c
48、ycles, the contact angle for the uncoated, as-received sample is 50“. The contact angle for the untreated surface is expected to decrease with time because of oxidation. After 600 cycles, the contact angle of the uncoated surface was approximately 15“ lower than the as-received Aqua-Shed sample (65“
49、). The oil contaminated (Oil A) Aqua-Shed sample had a lower final contact angle of 55“ while the as-received Aqua-Shed sample attained a contact angle of 65“. Oil A contamination on bare aluminum sample reduces the initial contact angle to 15“. However, the contact angle on the Oil A contaminated bare aluminum sample rapidly increases as a result of dry/wet cycling. After 150 cycles, the contact angle on the oil contaminated bare aluminum sample was 40“ higher than the Aqua-Shed contaminated sampl
