ENGINEERING RESEARCH INSTITUTE UNIVERSITY OF MICHIGAN ANN ARBOR EFFECT OF ROOT WALL THICKNESS ON BOND RESISTANCE TO HEAT TRANSFER OF BIMETAL TUBES REPORT NO. 34 EDWIN H. YOUNG Assistant Professor of Onemical Engineering VERN W. WEEKMAN, JR. GAREN BALEKJIAN Research Assistants DENNIS J. WARD STEVE G. GODZAK SHAMSHER S. GROVER Graduate Students Project No. 1592 WOLVERINE TUBE DIVISION CALUMET AND HECLA, INC. DETROIT, MICHIGAN July, 1954

ABSTRACT This investigation of variations in aluminum root-wall thickness on the bond resistance and heat-transfer performance of Al-Cu Bimetal tubes with copper liners indicates no significant effect or consistent trend between the, four experimental tubes at the conditions under which they were investigated. The performance of an equivalent all-aluminum tube under identical test conditions was used as a basis of comparison for evaluation of heattransfer performance and bond resistance of the four Bimetal tubes having nominal aluminum root-wall thicknesses of 0.02, 0.03, 0.04, and 0.05 inch. In general, the magnitude and effect of bond resistance on heat transfer determined by two test methods —one with steam condensing inside the tube and the other with cold water flowing inside the tube —are in agreement with results expected on the basis of thermal-expansion considerations. ii

TABLE OF CONTENTS ABSTRACT LIST OF TABLES INTRODUCTION ii iv 1 APPARATUS AND PROCEDURE A. Test Procedure for Steam B. Test Procedure with Cold ANALYSIS OF RESULTS A. General Discussion B. Analysis of Test Results C. Analysis of Test Results the Tube D. Discussion of Results of Condensing Inside the Tube Water Flowing Inside the Tube with Steam Condensing Inside the Tube Obtained with Water Flowing Inside the Two Tests 3 5 6 7 7 7 13 16 CONCLUSIONS AND RECOMMENDATIONS 16 APPENDICES Appendix A Sample Calculations Determination of Tube Characteristics for 0,05-Inch Tube Appendix B Sample Calculations Determination of Tube Characteristics for All-Aluminum Tube Appendix C Sample Calculations Run No. 93 with Steam Condensing Inside the Tube Appendix D Sample Calculation of Bond Resistance for 25 psig Steam Condensing Inside the 0,03-Inch Tube Appendix E Sample Calculation of a Wilson Plot Point for Figure 7 Appendix F Sample Calculation of Bond Resistance for Water Flowing Inside 0,03-Inch Tube 18 19 20 21 22 22 23 FIGURES 1 THROUGH 19 NOMENCLATURE 25-43 44 iii

LIST OF TABLES Table Page I TUBE CHARACTERISTICS 4 II TYPICAL TEST DATA OBTAINED ON THE 0,03-INCH ROOT-WALL TUBE WITH STEAM CONDENSING INSIDE THE TUBE (RUN NO. 54) 6 III TYPICAL TEST DATA OBTAINED ON THE 0,03-INCH ROOT-WALL TUBE WITH WATER FLOWING INSIDE THE TUBE (RUN NO,; 93) 7 IV SUMMARY OF TEST RESULTS WITH STEAM CONDENSING INStIDE TUBES 8 V SUMMARY OF TEST RESULTS FOR WATER FLOWING INSIDE TUBES 9 VI SUMMARY OF TEST RESULTS OBTAINED BY CONDENSING STEAM INSIDE THE TUBES 12 VII PERCENTAGE PERFORMANCE OF BIMETAL TUBES COMPARED TO THE ALL-ALUMINUM TUBE FOR CONDENSING STEAM 13 VIII WILSON PLOT DATA AND EVALUATED BOND RESISTANCES 14 IX CALCULATED BOND RESISTANCES FOR TESTS WITH STEAM INSIDE TUBE AND COMPARISON WITH TESTS WITH WATER FLOWING INSIDE TUBE 15 iv

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN EFFECT OF ROOT-WALL THICKNESS ON BOND RESISTANCE TO SEAT TRANSFER OF BIMETAL TUBES INTRODUCTION Bond resistance is defined as the resistance to heat transfer at the contact surfaces between the liner metal and the fin-root metal of a Bimetal finned tube. It is believed that an extremely thin film of air in the metal-to-metal bond zone constitutes an additional resistance to heat transfer which varies with the particular use of the tube. The variation in bond resistance is due to the thermal contraction and expansion of the two dissimilar metals due to the difference in thermal-expansion coefficients of the two metals, e.g., the thermal-expansion coefficient of a copper liner is 66 percent of that of aluminum so that, depending on the relative temperatures of the inner and outer surfaces of the bond' zone, the clearance constituting the bond resistance may expand or contract, resulting in a corresponding variation in bond resistance. Although it may appear that the thickness variations of the air gap are slight, the effect of the variation on heat transfer is great because the thermal conductivity of air is approximately one ten-thousandth that of copper. The heat-transfer performance of a finned tube is related to the outside heat-transfer area and mean overall temperature difference driving force by the following relationship: Q = UoAA, (1) where Q = Heat transferred, Btu per hour Uo 0 Overall heat-transfer coefficient Btu/(hr)(F)(ft2) outside surface A = Total outside heat-transfer area, sq ft AT = Mean temperature difference, F. If no fouling is present, the heat-transfer coefficient, Uo0 is further defined as follows: I 1

I ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN - I U uo 1 - + ho rm + Lhi ( Ai (2) where ho = Outside film coefficient for a finned tube, Btu/(hr)(~F)(ft2) hj = Inside film coefficient, Btu/(hr) (~F) (ft2) Ai = Inside tube-surface area, ft2/ft of tube length At = Outside tube-surface area, ft2/ft of tube length. The resistance of the tube-metal wall to heat transfer, rm, may be defined as follows: (a) for monometal finned-tube construction: Xe AU ) km Ame) where Xe = equivalent finned-tube wall thickness based on the equiv tube diameter, ft (see nomenclature list) km = Thermal conductivity of the metal wall, Btu/(ft) (hr)(~F).Ame = Average tube-metal area between di and de, ft2/ft (3) alent (b) for copper-aluminum Bimetal tube construction: rm (X).al (A kal me Al 1 + Rb (o) + (Aob)Cu (x{ (Ad) Iku (AA)Cuh (4) where Rb = Bond resistance, (ft2)(hr)(~F)/Btu.. The bond resistance of a Bimetal finned tube may be determined by assuming that the outside film coefficient at infinite inside water velocity is the same as the outside film coefficient of the all-aluminum tube (computed from a Wilson plot intercept using equation (2)), and by using this value of ho as the value for the Bimetal tube in equation (2). The inside water film resistance (Ao/Aihj) in equation (2) is zero at infinite water velocity. When Wilson plot data cannot be obtained for Bimetal and all-aluminum tubes, the bond resistance of the Bimetal tube may be determined from the overall heat-transfer coefficient of the Bimetal tube and the corresponding overall heat-transfer coefficient of an equivalent all-aluminum tube under duplicate test conditions by assuming the inside and outside film coefficients of the Bimetal tube to be respectively equal to the inside and outside film coefficients of the all-aluminum tube, and then using the following relationship: -i L U O/Bimetal \ (all -aluminum = Rb - (A(o)Cu (5) A correction may be applied for the difference in the tube-wall resistance of the Bimetal and all-aluminum tubes by combining the answer obtained in equation (5) with equations (3) and (4)* - 2

I -- ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN Information on the relative performance of Al-Cu. Bimetal, all-copper and all-aluminum tubes had been compiled from test data provided by eleven organizations on thirty different units for crossflow of air on the outside of tube banksI. Except for a few Bimetallic units, heat-transfer data on the all-copper, all-aluminum and the Al-Cu. tubes were correlated satisfactorily. The correlation indicates no appreciable difference in the heat-transfer performance of monometallic and Bimetallic tubes for applications of low heat flux. It was proposed to investigate the effect of the aluminum root-wall thickness on the bond resistance to heat transfer of four Bimetal tubes having aluminum fins and copper liners over a range of heat fluxes. The four tubes were to have nominal root-wallthickness of 0.02, 0.035 0.04, and 0.05 inch. The heat-transfer performance of a similar all-aluminum tube was to be used as a basis of comparison. Table I gives the characteristics of the allaluminum tube and four Bimetallic finned tubes. APPARATUS- AND PROCEDURE Two test procedures were used to cover the range of heat fluxes in this investigation. The basic- difference between the two test arrangements was that of reversing the temperature gradient directions. In one arrangement, cold water was passed through the inside of the copper liner and a hotwater bath was maintained on the fin side of the tube. Under this condition the air-gap thickness increased, since the thermal-expansion coefficient of aluminum is considerably greater than that of copper. In the other arrangement, steam was condensed inside the copper liner and a water bath was maintained on the fin side. This reversal of temperature gradient resulted in a decrease in the thickness of the air gap with a corresponding decrease in bond resistance. Both of the above tests were repeatedon the all-alum~fium tube to obtain comparable performance with a tube having no bond resistance. Wilson plots were obtained by varying the water flow rates through the tubes in order to evaluate the outside film coefficients of heat transfer and the bond resistance of the Bimetallic tubes. Katz, D. L., et al;. "Correlation of Heat Transfer and Pressure Drop for Air Flowing Across Banks of Finned Tubes", University of Michigan Engineering Research Institute Project Report for Wolverine Tube Division of Calumet and Hecla, Inc.^ Augusta 1955* L 3

TABLE I TUBE CHARACTERISTICS Bimetallic Finned Tubes AllCharacteristic Nominal Aluminum Root-Wall Thickness Aluminum ~-; - -;- -1;-. —- - -— ^ -Tube 0.02 in. 0.03 in. 0.04 in. 0.05 in. Aluminum Root-Wall Thickness, in. 0.0291 0.0o48 0.0401 0.0571 0.0730 Average Fin Thickness, in. 0.0198 0.0198 0.0164 0.0174 0.0190 Inside Diameter, Cu Liner, in. 0.928 0.928 0.928 0.928 1.010 Outside Diameter, Copper Liner, in. 0.988 0.987 0.988 0.988 Aluminum Root Diameter, in. 1.046 1.057 1.068 1.102 1.156 Diameter over Fins, in. 1.989 2.010 2.001 2.014 1.999 Fins per inch 9.15 9.15 9.15 9.10 9.28 Length of Finned Section, in. 66.0 66.0 66.0 66.0 64.75 Finned Area/ft, ft2/ft 3.709 3.781 3.705 3.684 3.538 Equivalent Diameter, de, in. 1.217 1.230 1.208 1.246 1.305 Total Heat-Transfer Area in Bath, ft2 20.89 21.25 20.85 20.64 19.78 4z

I I I - ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN A. Test Procedure for Steam Condensing Inside the Tube The equipment used in the present experiment was that originally developed for preliminary bond-resistance tests, completed at an earlier date.. As shown in Figure 1, the apparatus consisted of a steam supply to a finned tube submerged in a water trough. The steam was first passed through a condensate separator. prior to feeding to the finned tube in order to insure a dry saturated steam feed. The steam was condensed on the inside of the tube, the rate of condensate formation being a measure of the rate of heat transfer. The inlet steam-pressure gage was calibrated within + 0.1 psi, and the outlet gage within + 0.05 psi. The water-bath temperature was measured at three points by thermometers which were calibrated to + 0.1 C. The rate of heat transfer was measured by timing the collection of 1000 ml of condensate Water was admitted to the trough at a constant rate, and was drained through underflows at the opposite end of the.trough,. A five-foot-long steam sparger, resting on the- tank bottom, was used to maintain the bath water at a constant temperature. The sparger provided agitation of the bath water resulting in a more uniform bath temperature. The condensate from the finned tube was run through a Jerguson gage in which a constant liquid level was maintained, The condensate was bled from the Jerguson gage outlet through a subcooling coil and then collected in a 1000-ml volumetric flask. The subcooling of the condensate prevented partial flashing: as the condensate pressure was dropped to atmospheric. Before each run, steam was bled from the steam purge valve located at the base of the downstream compound gage to remove noncondensable gas accumulation. A test run consisted essentially of timing the collecting of 100D ml of condensate after the equipment had reached a steady state of operations During the test run the inlet and outlet steam pressures and the three waterbath thermometers were read at regular intervals. Two series of test runs were carried out at different steam pressure levels, one series with an inlet steam pressure of 25 psig and the other with an inlet steam pressure of 35 psig. Table II presents typical test data obtained in test run No. 54 on the 0.03-inch-thick root-wall tube with steam at an inlet pressure of 26 psig condensing inside the tube. 5

- ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN - r I I TABLE II TYPICAL TEST DATA OBTAINED ON TIE 0,03-INCH ROOT-WALL 1TUE WITH STEAM CONDENSING INSIDE THE TUBE (RUN NO. 54) Steam Pressure to Steam Condensate Water Bath Tube, psig Pressure, psi Temperature,~C TB T2 T3 26.0 4,2 62,8 65.0 69.2 26.0 4.3 62,5 65.8 69.2 26.0 4.3 64.2 65.0 69.2 Time to collect 1000 ml of condensate = 1280 minutes. 7 B. Test Procedure with Cold Water Flowing Inside the Tube The equipment previously described was converted from a steam feed to a cold-water feed to the tube under test. Figure 2 shows the apparatus as modified for the cold-water test procedure. The cold-water inlet and outlet temperatures were measured by thermometers which had been calibrated to + 0.1~F A pressure gage on the upstream end of the tube provided a visual means for controlling the water flow rate, A weigh barrel was used to measure accurately the tube-side water flow rate. Provisions were made for preheating the inlet cold water to the tube by direct steam sparging in a separate surge tank. This arrangement permitted Wilson plot tests to be carried out, since a constant average water temperature could be maintained on the tube side. The trough was filled with water to the overflow level and steam was sparged to heat and maintain the bath water at the desired temperature level. The bath-water temperature was measured by the three calibrated centigrade thermometers used in the previous test procedure. A test run consisted of adjusting the inlet cold-water temperature to such a level that the average of the inlet and outlet temperatures resulted in the desired average temperature level. In flowing through the tube, the cold-water temperature was increased by heat transferred from the bath, which was maintained at a higher temperature level by sparging steam into it. After the system had been operating at constant conditions for approximately 15 minutes, a test run was made. The test run consisted of timing the collecting of a quantity of water in the weigh barrel, During this period all thermometers were read and recorded at regular intervals. Table III presents typical test data obtained in test run No. 93 on the 0.03-inch-thick root-wall tube. r I -a. 6 I

I - ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN TABLE III TYPICAL TEST DATA OBTAINED ON THE 0,.05-INCH ROOT-WALL TUBE WITH WATER FLOWING INSIDE THE TUBE (RUN NO. 93) Inlet Water Outlet Water.Temperature, Temperature, Bath Temperatures, ~C OF ~F T1 Ta T3 65.530 81,40 81. 82.9 85.1 63.10 81.40 81,7 83.0 83.6 62.70 81 50 81.8 85.1 83.4 63.50 81.50 81.7 83.1 83.3 63.70 81.70 81.5 83.2 83.6 Time to collect 230 lb of water = 1.948 minutes. _~~~~~~~~-..-, -., I.......,.....;.,.I,;.................. -- _.~~......... i..,.^... ANALYSIS OF RESULTS A, General Discussion The characteristics of the tubes investigated were summarized in Table I. The calculations of the characteristics for the 0,03-inch tube and the all-aluminum tube are given in Appendices A and B respectlvely. Table IV s-ummarizes the bond-resistance-test results obtained with steam condensing inside the tubes, and Table V summarizes the bond-resistance test results obtained with water flowing through the finned tubes. B Analysis of Test Results with Steam Condensing Inside the Tube _4-. -e:-:.- - I. Table VI contains a summary of test results obtained by condensing steam inside the tubes at 25 and 35 paig inlet steam pressures. A sample calculation for the Q053-inch root-wall tube with steam condensing at 25 psig is given in Appendix C. Figure 3 graphically compares the overall coefficients given in Table VI as a function of the nominal root-wall thickness. It may be seen that the 0.035- and 0,04-inch nominal root-wall tubes did not perform as well as the 0.02- and 0.05-inch root-wall tubes. The all-aluminum tube, of course, gave the best performance. 7

TABLE IV SUMMARY OF TEST RESULTS WITH STEAM CONDENSING INSIDE TUBES Nominal Run Average Overall Time to Total Heat Transfer Overall Heat-Transfer Al Root-Wall nBath AT Collect 25 psig steam 5Bu5 psig steam Coefficient, U Thickness, in. N. Temp,OF OF 1000 ml, sec (267~F) Btu/hr (280.5~F) Btu/(hr)(~F)(ft2) (outside).05 54 149.9 106.7 74.0 100,900 45.8.05 55 151.0 105.9 75.6 101,900 46.5.05 36 151.7 104.9 74.0 100,800 46.6.05 57 150.2 105.7 74.8 100,000 45.7.05 58 150.0 105.9 74.4 100,200 45.9 05 59 149.0 120.4 59.5 124,500 50.1.05 40 149.2 120.1 61.0 122,000 49.1.05 41 150.6 118.9 61.5 120,500 49.1.05 42 152.0 117.5 61.4 120,500 49.6.04 43 151.3 105.7 76.7 97,200 44.1.04 44 151.3 105.8 76.7 97,200 44.0.04 45 151.9 104.9 77.1 96,700 44.1.04 46 149.7 121.4 63.5 116,000 45.8.04 47 148.4 122.5 65.7 115,500 45.2.04 48 149.9 120.8 62.5 117,900 46.8.04 49 151.3 119.7 62.2 118,500 47.5.04 50 150.6 119.8 62.0 120,000 48.0.04 51 150.5 119.7 62.9 118,700 47.6.03 53 151.1 104.9 77.2 99,200 44.5.03 54 150.6 105.5 76.8 99,700 44.5.03 55 150.7 105.5 76.1 100,5300 44.7.03 56 151.2 104.9 77.0 99,700 44.8.03 57 150.9 115.6 61.5 1235,500 49.0.03 58 150.6 119.1 61.5 125,000 48.5.03 59 151.6 118.1 61.4 125,500 49.1.02 61 151.5 104.1 74.4 102,800 46.7.02 62 151.4 1035.5 75.6 102,000 46.6.02 63 151.0 104.0 77.2 102,500 46.7.02 64 150.6 104.4 75.7 100,800 45.7.02 65 150.9 104.6 74.8 102,200 46.4.02 66 152.0 117.6 61.1 124,000 50.0.02 67 149.0 120.5 61.1 124,100 48.8.02 68 151.9 118.1 61.4 1235,500 49.5.02 69 152.4 -117.4 61.5 125,000 49.6 All-Aluminum Tube 155 1535.3 104.5 81.8 935,500 46.0 156 150.8 104.4 79.6 96,400 46.6 157 149.9 105.3 78.6 97,400 46.9 158 149.9 104.7 78.6 97,500 47.1 159 151.2 119.7 64.0 117,900 50.0 160 151.5 119.7 61.8 121,800 51.5 161 151.5 118.9 63.0 120,000 51.1 162 150.1 120.9 62.5 120.500 50.4

TABLE V SUMMARY OF TEST RESULTS FOR WATER FLOWING INSIDE TUBES Nominal Average Inlet Outlet Overall Flow- Flow- Total Heat Overall Heat-Transfer Run Al Root-Wall Bath Water, Water, AT, Rate Rate, Transfer, Coefficient, Uo Thickness, in. o Temp.,F ~FF ~F ~F lb/hr ft/sec Btu/hr Btu/(hr) (~F) (ft) (outside).02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.02.03.05.0o.03.03.o05.05.053.053.05.03.05.03.03.03.03 121 181.4 122 181.7 123 181.3 124 181.9 125 181.8 70 183.5 71 180.7 72 182.8 75 181.1 74 180.7 75 181.0 76 181.4 77 181.0 126 182.0 127 181.4 128 181.9 205 181.8 206 181.9 207 181.8 208 181.9 209 181.9 210 182.0 211 181.3 212 181.8 213 181.5 88 181.8 89 182.0 96 181.5 97 181.5 98 182.5 99 181.4 90 182.3 91 182.3 92 180.8 93 181.9 94 182.6 95 182.4 100 182.6 101 180.8 102 181.9 103 181.7 56.08 55.42 58.06 58.43 58.98 60.08 59.51 60.70 60.98 62.21 62.85 63.18 63.73 64.50 65.27 65.72 90.18 92.01 92.93 95.34 94.03 95.00 94.95 86.36 88.84 53.35 51.10 54.80 55.13 58.04 58.08 59.28 58.70 61.15 62.98 62.32 64.14 62.98 63.91 64.76 64.89 87.62 87.15 86.34 84.82 85.05 86.62 85.92 83.95 83.48 81.64 79.35 79.97 78.54 78.92 78.55 78.73 109.58 108.50 107.06 106.34 105.80 105.03 104.67 113.36 110.52 91.80 90.17 88.21 88.63 86.32 86.15 84.90 84.61 82.11 81.30 80;79 80.51.79.41 78.62 78.43 78.26 109.5 110.4 109.1 110.3 109.8 109.9 109.1 111.6 110.0 109.9 109.9 109.8 109.9 110.3 109.6 109.7 81.9 81.7 81.8 82.1 82.0 82.0 81.5 82.0 81.8 109.2 111.4 110.0 109.6 110.3 109.3 110.2 110.7 109.2 109.8 110.0 110.2 111.4 109.5 110.3 110.1 2545 2495 5110 3590 3600 3700 3670 4940 4980 6250 7980 7640 9440 9680 10,890 10,900 3295 4295 5520 6200 7350 9240 9200 1838 2690 1945 1960 2415 2420 3340 3340 5965 5880 5430 7090 7060 8450 8670 9670 10,950 11,010 2.42 2.37 2.95 3.41 3.42 3.51 3.49 4.69 4.73 5.94 7.58 7.25 8.96 9.20 10.32 10.35 5.13 4.08 5.05 5.88 6.98 8.77 8.73 1.74 2.56 1.85 1.86 2.29 2.30 3.17 3.17 3.77 5.69 5.16 6.74 6.71 8.03 8.23 9.19 10.40 10.46 80,200 79,300 88,000 94,800 93,800 98,100 97,000 115,000 112,000 121,500 131,800 128,200 139,900 139,000 142,200 142,200 63,900 70,800 75,300 80,600 86,500 92,600 89,500 49,600 58,400 74,800 76,500 80,700 81,100 94,500 93,700 101,800 100,600 113,800 130,000 130,500 138,200 142,400 142,300 149,800 147,100 34.7 33.9 57.9 40.6 40.5 42.3 42.1 48.8 48.5 48.3 52.4 56.9 55.4 -60o.3 59.6 61.5 61.4 37.5 41.4 44.o 46.9 50.4 54.1 52.6 28.9 34.2 32.3 51.9 34.5 34.8 40.3 40.3 43.4 45.0 48.9 55.7 55.8 58.8 60.1 61.2 63.7 62.8

TABLE V (cont.) SUMMARY OF TEST RESULTS FOR WATER FLOWING INSIDE TUBES Nominal Average Inlet Outlet Overall Flow- Flow- Total Heat Overall Heat-Transfer Run Al Root-Wall Rn Bath Water, Water, AT, Rate, Rate, Transfer, Coefficient, Uo Thickness, in. Temp.,F F F Fb/rft/secBt/hrBt/(hr)F)(ft2) ide).04.04.04.04.04.04.04.04.04.04.04.04.04.04.04.04 104 105 106 107 108 109 110 111 112 113 114 117 118 115 116 119 182.1 182.1 181.8 181.9 181.7 182.0 181.8. 181.5 181.8 181.8 182.0 181.6 181.5 182.2 181.6 182.0.05.05.05.05.05.05.05.05.05.05.05.05.05.05.05 All-Aluminum Tube 78 181.0 79 181.5 129 181.6 130 181.6 131 181.3 80 182.5 81 182.4 132 181.5 133 181.3 134 181.9 83 181.5 84 181.3 85 181.5 86 181.4 87 181.1 135 181.9 136 182.0 157 181.9 138 182.1 139 181.8 140 181.7 141 181.7 142 182.3 143 181.6 144 182.3 145 182.4 146 182.4 53.11 52.50 56.26 57.50 58.61 60.58 60.25 62.98 63.02 64.05 63.93 64.68 64.74 65.99 65.50 65.26 53.58 53.52 55.43 58.71 59.95 60.71 60.36 62.15 62.13 62.91 63.44 64.58 64.84 65.67 65.69 61.62 61.69 61.71 61.71 61.73 61.79 61.71 61.69 61.99 61.91 62.30 62.36 91.08 90.41 86.96 84.80 84.09 82.85 82.81 81.52 81.58 80.08 79.67 80.40 78.96 79.21 78.55 78.41 90.63 90.44 87.76 85.28 83.18 83.90 83.60 82.00 79.51 80.41 79.86 79.03 79.21 78.50 78.51 95.87 95.71 91.92 91.90 89.32 88.34 84.08 84.20 82.10 82.31 80.94 81.14 108.9 109.5 110.0 109.6 109.7 110.2 110.4 109.4 110.5 110.3 109.9 109.2 109.5 109.3 109.0 103.2 103.3 105.1 105.3 1063. 106.6 108.8 109.4 109.6 110.2 110.8 110.7 110.0 110.7 110.2 110.7 110.3 110.3 110.3 109.3 109.3 109.7 110.2 109.1 109.7 109.6 109.6 110.2 2040 2050 2885 3520 3965 4920 5220 6590 6560 8200 8360 8480 9700 10,960 10,980 10,970 2100 2100 2400 3420 4270 4480 4470 5640 6860 6840 7720 9220 9450 10,950 10,850 2400 2410 2950 2950 3450 3710 4920 4890 5890 5980 6900 6910 1,94 1.95 2.74 3.34 3.77 4.66 4.96 6.25 6.24 7.78 7.95 8.05 9.22 10.40 10.43 10.43 1.99 1.99 2.38 3.25 4.05 4.26 4.25 5.35 6.52 6.49 7.33 8.75 8.98 10.40 10.31 1.92 1.93 2.36 2.36 2.76 2.97 35.94 3.91 4.71 4,79 5.52 5.53 77,900 77,500 77,600 90,800 99,200 103,800 104,000 112,000 119,200 119,600 126,900 133,000 135,800 140,500 139,200 82,450 82,100 89,400 89,000 95,200 98,500 110,000 110,000 118,400 122,000 128,500 129,600 77,500 77,600 88,500 96,200 101,000 109,800 116,200 122,200 121,900 131,500 131,800 133,500 138,000 145,000 145,000 144,000 33.9 33.7 38.4 41.6 43.9 47.7 50.5 53.5 535.4 57.5 57.3 58.5 60.2 63.4 62.5 62.5 34.7 34.4 34.2 40.2 43.8 45.5 45.6 49.5 52.2 52.5 55.9 58.9 60.0 62.2 61.9 40.4 40.1 42.9 42.7 45.3 46.7 51.1 50.8 54.7 55.9 58.6 59.1

TABLE V (concl.) SUMMARY OF TEST RESULTS FOR WATER FLOWING INSIDE TUBES Nominal un Average Inlet Outletverall Flow- Flow Total Heat Overall Heat-Transfer Al Root-Wall NOBath Water, Water, AT, Rate, Rate, Transfer, Coefficient, Uo Thickness, in. o Temp.,~F OF OF ~F lb/hr ft/sec Btu/hr Btu/(hr)(OF)(ft2)(outside) All-Aluminum Tube 147 181.8 61.71 80.82 110 ( xi lxCin nnn 7n 148 149 150 151 152 153 154 182.0 182.2 182.1 182.1 181.9 182.1 181.7 63.26 63.49 64.03 63.77 64.50 64.53 65.10 195 181.9 196 182.0 197 182.6 198 181.8 199 181.6 200 181.9 201 182.0 202 182.1 203 182.3 204 182.1 94.71 94.21 93.45 93.52 93.42 92.77 91.65 91.29 87.80 90.03 81.40 80.18 80.01 79.71 78.53 78.79 78.55 105.44 105.91 106.88 107.28 106.41 107.71 108.54 109.15 112.46 110.08 109.7 110.4 110.1 110.4 110.4 110.4 109.9 7330 8550 9000 9020 10,450 10,550 11,400 5.86 6.84 7.20 7.22 8.37 8.45 9.13 153,000 142,500 143,700 143,800 146,800 150,500 1535,500 DY.D 61.3 65.2 65.9 65.8 67.2 68.6 70.4 81.8 81.9 82.4 81.4 81.7 81.6 81.9 81.9 82.2 82.1 9490 8360 6840 6390 7040 5420 4660 4220 2380 3420 7.67 6.76 5.53 5.16 5.69 4.39 3.77 3.42 1.92 2.77 102,000 97,900 92,000 88,000 91,500 81,000 78,700 75,400 58,700 68,600 63.0 60.4 56.4 54.6 56.6 50.1 48.6 46.5 36.1 42.2 2.77 68,600 42.2

TABLE VI SUM[MARY OF TEST RESUlTS OBTAINED BY CONDENSING.STEAM INSIDE THE TUBES 2)' H,) Average Average Average Average Average Uo Tube Runs Bath Temp. T F Time (1000 ml),,Btu/hr Btu/hr=F-ft2 out Al-Root Thickin. Averaged. sF avavsec 25 psig steam inside tubes 0.02 61 - 65 151.0 104.1 75.5 10.2,060 46.4 0,03 53 - 56 150.9 105.2 76.8 99,7350 44.6 0.04 43 — 45 151.5 105.5 76.8 97,030 44,7 0.05 354 - 38 150.8 105.8 74.2 100,760 46.1 All Al 155 - 158 151.0 104.7 79.6 96,200 46.7 35 psig 0.02 0.05 0.04 0.05 All A1 steam inside tubes 66 - 69 57 - 59 46 -51 39 - 42 159 - 162 151.5 151.0 15o0.1 150.2 151.0 1 T.i 118.4 117.6 120.6 119.2 119.8 61.5 61.5 62.8 60.9 62.8 123,650 125,550 117,770 -1.21,870 120,000 49.5 48.9 46.8 49.5 50.75

- ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN - The relative performance of the Bimetal tubes compared to the allaluminum tube is given in Table VII. The bona resistance was determined by comparing the performance of the four Bimetal tubes with the performance of the all-aluminum tube by the use of equation (5) under identical conditions. The calculation of the bond resistance for the nominal 0.05-inch root-wall tube with 25 psig steam condensing inside the tube is given in Appendix D. Figure 4 graphically presents the variation of the bond resistance of the four Bimetal tubes with condensing steam pressure., A comparison of this performance with that obtained from the Wilson plot results made with water flowing througL the inside of the tubes is also given in this figure, a discussion of which is given in section C. TABLE VII PERCENTAGE PERFORMANCE OF BIMETAL TUBES COMPARED TO THE ALL-AlMINUM TUBE FOR CONDENSING STEAM Nominal Root-Wall 25 psig Steam 55 psig Steam Thickness. in. 0.02 99.4 97.5 0.03 95-5 96.2 0.04 95.7 92.1 0.0Q5 98.6 97.5 C. Analysis of Test Results Obtained with Water Flowing Inside the Tube Figures 5 through 9 present the Wilson plots for the all-aluminum and Bimetal finned tubes. Appendix E presents a sample calculation of a Wilson plot point for the nominal 0.03-inch tube utilizing the test data given in Table.III (Run No. 95 of Table V). The intercepts of the Wilson plots are tabulated along with the computed overall coefficients in Table VIII. The bond resistances are also presented in this table and were obtained by first computing the outside film coefficient for the all-aluminum tube,. second, assuming that this same coefficient existed on the outside of the four Bimetal tubes at an infinite water velocity inside the tubey and third, back calculating the bond resistance by use of equations (2) and (4). Appendix F contains the calculation of bond resistance for the nominal 0.05-inch toot-wall tube by this method. It should be emphasized that the calculation of bond resistance by this method is highly sensitive to slight variations of the outside film coefficient from that assumed. There iLs, therefore, some question as to how significant the decimal digits 058 are in the value 0I000158 for the 0,02-inch tube for the 719'F water in Table VIII. 13

TABLE VIII WILSON PLOT DATA AND EVALUATED BOND RESISTANCES tw = 71.9 ~F tw = 100.0 "F Tube Al Root'' ailson Al t Wilson Plot Total Heat- overan l. Thickness, in.,. Coefficient iIntercept Transfer Bond* Resistance Wilson Plot Uo at.i Bond at, Water hr-~PF/Btu Area, ft Velocity hr-F-ft2/Btu Intercept Velocity Resistance 0.02 0,00049 20,89 97.6 0.000138 0.00064 74.8 0.00023 0.05 0.000492 21.85 95.5 0.000148 0.04 0.00052 20.85 92.2 0.000185 0:05 0.00051 20.64 95.0 0.000158 All-Al Tube 0,000418 19.78 121.0 - 0.00051 99.5 Calculated on the assumption that the outside coefficient between fins and bath water are the.same for Bimetal and all-aluminum tubes; at 71.9~F, ho = 142y and at 1000~F ho = 113 Btu/hr-OF-sq ft0 H

I ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN Figures 10 through 15 present the individual experimental overall coefficients as a function of water velocity for the four Bimetal finned tubes Figure 14 again presents the four curves given in Figures 11 through 13 for ease of comparison. This figure indicates that the effect of variation of root-wall thickness on bond resistance is of minor importance, At low heat fluxes the performances tend to merge, while at high heat fluxes the effect of root-wall thickness becomes more appreciable. Figure 15 presents a comparison of the performance of the four Bimetal tubes with the performance of the all-aluminum tube with an average tube-side water temperature of 71.9~F. This figure clearly depicts the influence of bond resistance on heat-transfer performance. Figure 16 presents a similar comparison at an average tube-side water temperature of 100,O~F for the O.02-inch tube. The 10OO.0F tests were made to determine the effect of increasing the inside water temperature on bond resistance. Increasing the inside water temperature by 28~F resulted in bond resistances.of 0,000138 and 0.,00023 at 71i,9F and 100.0~F for the 0.02inch tube as indicated in Table VIII. This corresponds to a 67 percent increase in the numerical value of the bond resistance, but the overall performance remained about the same, Figure 17 graphically indicates the relative performance of the 0,02-inch tube as indicated in Figures 15 and 16. It should be pointed out that the difference between the two curves of Figure 17 is due to the 28-~F difference in temperature level between the average tube-side water temperatures, The bath-water temperature was the same for both curves, It is interesting to note that the relative effect of bond resistance decreased with decreasing water flow rates below 5 ft/sec. On the other hand, the curves indicate that for velocities of 5 ft/see and higher, the relative effect of bond resistance levels off to a constant performance of 85 percent for 71,9~F water and 82,5 percent for lOO100 F water* TABLE IX CALCUJLATEB BOND RESISTANCES FOR TESTS WITH STEAM INSIDE TUBE AND COMPARISON WITH TESTS WITH WATER FLOWING IMSIDE TUBE is I.I Nominal Al Bond Resistance - hr-F'-f/t2/Bitu Root-Wall Steam Inside Tubes,. Water Flowing Insiide Tubes Thickness,in. 25 psig'35 psig tw = 719 F tw = 100,0 ~F 0.02 0,00001 0.000055 0.000158 0.00025 0.03.000oooo68 0.000051 0.oO0148 No data 0,.04.oooo000066.0001L5 0.w000185 obtained for 0,05 0.000013 0.000035 0.000158 these tubes -t,..u;b11e..s. 15

,- ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN - D, Discussion of Results of the Two Tests An examination of Figure 4 and Table IX indicates the range of bond resistances obtained in the two tests. The staam-condensing bond resistance varied from 0000001 to 0.000115 hr-~F-ft2/Btu, whereas the cold water tests resulted in bond resistances varying from 0*000138 to 0.000185 for 71,91F water. A value of 0.00023 was obtained for the 0.02-inch tube with the 100,00F water. The lower bond resistances obtained with steam inside the tube as compared to 71.90F water inside the tube is what would normally be expected, since the hotter fluid would thermally expand the inner liner and thereby tighten up the bond existing between the liner and the finned-tube metal.. A comparison of the value of 0.00023, obtained with l00,OF water, with the value of 0.000138, obtained with 71.9~F water, indicates an inconsistency} since the 100.0~F water should thermally expand the liner metal and thereby reduce the bond resistance over that obtained with 71,9~F water. The overall coefficients obtained at infinite water velocity from the Wilson plot intercepts of Figures 5 and 6 and tabulated in Table VIII for these two conditions are 74.8 and 97.6 Btu/hr-~F-sq ft, respectively, for the 0.02-inch tube. The corresponding values for the all-aluminum tube are 99.3 and 121.0 respectively, The corresponding outside film coefficients for the all-aluminum tube are 113.0 and 142.0 Btu/hr-~F-ft2 respectively. The above information, when put togethe] with the computed bond resistances1 indicates that the higher inside water temperature resulted not only in a reduced overall AT driving force, but also in reduced overall heat-transfer coefficients and reduced outside film coefficients. The net effect was an increase in the bond resistance contrary to what would normally be expected. This discrepancy may be due to the assumption of equal outside film coefficients for the Bimetallic and all-aluminum tubes under the same test conditions* It must be emphasized that a small variation in the value of the outside film coefficient of the Bimetal tube appreciably affects the magnitude of the corresponding bond resistance, CONCLUSIONS AND RECOMMENDATIONS This investigation of the effect of the aluminum root-wall thickness on bond resistance to heat transfer for the tubes studied at the conditions under which they were investigated does not indicate any significant trend or effect. This is clearly indicated by Figures 14 and 15 which show the relative performances decreasing in the following orders 0.053 0.04, 0,02, with the 0,05 giving the poorest performance. It should be emphasized, however, that the differences in performance are of a minor nature, as the overall coefficients for the four Bimetal tubes corresponding to any fixed water velocity

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN have the same order of magnitude, This is further substantiated by Figure 3 for the- condensing-steam tests, which do not indicate a consistent performance order for the four Bimetal tubes at the two steam condensing pressures. The lower bond-resistance values from steam-condensing tests compared to the coldwater tests are in agreement with the results expected from thermal-expansion considerations The effect of decreasing the overall temperature difference by raising the inside water temperature resulted in a decrease -in the overall coefficient as indicated in Figures 15 and 16, and an increase in the bond resistance as indicated in Table VIII, for the 0.02-inch tube. It should be emphasized that the relative performance of the 0.02-inch tube to the all-aluminum tube at these two water temperatures and at normal water flow rates is approximately 84 percent as shown in Figure 17. It was not possible to develop a general correlation for predicting bond resistance for Bimetal tubes over a wide range of temperature levels and heat fluxes on the basis of the above observation. The need for a correlation for predicting vond resistances over a wide range of temperature levels and heat fluxes has been established, It is recommended that an investigation be undertaken to establish a method for predicting bond resistances under these conditions and that such an investigation be made a part of the investigation of cyclic operation on bond resistance. 17

APPENDICES

- ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN - APPENDIX A SAMPE CALCULATIONS DETERINATION OF TUBE CHARACTERISTICS FOR 0.03-INCH TUBE Data from-Table I: Aluminum itoot-wall thickness = 0.0548 in. Aluminum toot'diameter = 1.057 in. Diameter over fins = 2.010 in. Fins ~per inch = 9.15 Length of'finned dection = 66.0 in. Length of Bare dopper Section = 6 in, Length of Bare iron pipe = 16 in* A =.(1.057) + (2) ( ) (2.010 l.057 )9. A, = -';:',... _.. +...,,-..,..:.:.'.-...',,:::'.._::...-.. = 3*78l ft2/ft ~u 12 (4) (12) Total outside area, finned section: 7181 -x 66 = 20.75 ft2 12 Outside area of bare sections: OD Copper liner = 0988 in. (6 in, long) 3,1.4 x 0.988 6..., X988' = 0.1295 ft" 12 12 OD Iron-pipe sections = 1.050 in. (16 in. long) 3.14 x 1.00 x 16 = 0367 ft2 12 12 Total heat-transfer area (outside): A = (20.75 + 0.1295 + 03.67) = 21.247 ft2 19

L - ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN APPEMDIX B SAMPIE CALCULATIONS DETEENRMATION OF TUBE CIEARACTERISTICS FOR ALL-ALUMItNUM TBE Data from Table I: I = 1,010 in. Root diameter = 1.156 in. Root-Wall thickness = 0*073 in, Diameter over fins = 1999 in, Average fin thickness =.019 in, Length of finned section = 64*75 in, Fins per inch 9,28 A x 1.156 =t =.1 -2." + 12 2 [ (1.999)2 - (1.156)2] 9*28 = 3538 ft2/ft. 4 x 12 x x 1.01 Ai = 1 -.. 12 0,264 ft2/ft Ao/Ai 3 -538 1354.264 Total nonfitred 0,*680 ft2. kheat-transfer area in bath was measured and computed to be Total are obf finned section:. 3.538.x -64 12 = 19,10 ft2 Total heat-transfer area in bath: 19.10 + 0.68 = 19.78 ft2 Equivalent diameter: de = 1,156 +.019 (1,999 - 1.156) 928 = 1.305 in. Average aluminum diameter: 1.305 + 1.01 2 1 157 in. Average aluminum conduction.length:: 1.03o: - 1,o010 = 0.148 in, -I~~~~~~~af L 20

- ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN APPENDIX C SAMPLE CALCULATIONS FOR RUN NO,.3 WITH STEAM CONDENSING INSIDE THE TUBE The calculations are based on the data contained in Tables II and IV, The values in Table IV are the values in Table II corrected by the appropriate calibrations., Data: Average bath temperature = 150.6F Inlet steam pressure = 59.70 psia (266*8-"F) Outlet steam pressure = 18.67 psia (224,3.5F) Enthalpy of vapor in 1169.6 Btu/lb Enthalpy of liquid condensate out = 192.5 Btu/lb Calculation of the water flow rate. 1000 ml x 60.mn x.987./l. = 102.0 lb/hr 1,280 min x 1 hr x 45356 gm/lb Calculation of heat tr6nfer: AH = 1169*6 - 192*5 = 977.1 Btu/lb Q = (977.1)(102) = 99,770 Btu/hr Calculation of average overall AT: To determine the effective average overall AT it was necessary to make a preliminary calculation for the tube under a specific set of conditions^ It was assumed that the inlet steam entered at 25 psig and 267F., The condensing length was divided into eight zones as shown in Figure 18, The effece tive AT existing at the end of each zone was computed by a trial-and-error procedure so that the heat duty in each zone could be matched by the correspon ding heat-transfer rate, zone area, and AT driving force. Figure 18 indicates the effective ATts for each zone. Figure 19 was prepared to indicate the variation of the' metal area available for condensing as a result of condensate buildup, It was found that the effective AT could be determined by summing up 75 percent of the inlet AT and 25 percent of the outlet ATW AT = [224.53 + 75 (266,8 - 224.5)] - 150.6 = 105.6~F 21

- ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN Calculatioin of the overall coefficient: Using equation (1), 99,700 Uo = = 44,5 Btu/hr-~F-ft2. (21.247) (105.6) Note: all overall coefficients are based on the total area in the bath; ie, finned-tube plus bare-tube areas, APPENDIX D SAMPLE CALCULATION-OF BOND RESISTANCE FOR 2a-PSIG STEAM CONDENSING INSIDE TEE 0,.-IO. TUBE From Table VI the overall heat-transfer coefficient, Uo, for the all-aluminum tube is 4667'Btu./hr-ft2-F, and the Uo for the 0,03-inch tube is 44.6 Btu./hr-ft2-F. From equation (5) the bond resistance is obtained as follows: Rb - 7 )1 D/.10,259 ~7 44.6 - i781 __j = 0,0000684 hr-FF-ft2/Btu. APPENDIX E SAMPLE CALCULATION OF-A WILSQN PLOT POINT FOR FIGURE _ The following calculations are based on the data for Run No.. 93 contained in Tables III and V. All temperatures listed have been corrected by the appropriate calibration, Calculated data: Tav = average bath temperature Average inlet water temperature Average outlet water temperature Flow rate = 7090 lb/hr Average H20 temperature = 72,11.= 181.9~F (corrected) = 62,98F: (corrected) = 81.30'F (corrected) 4OF 22

-- ENGINEERING RESEARCH INSTITUTE UNIVERSITY OF MICHIGAN r Calculation of Wilson plot coordinates: 104 y = -. b A where i X = [1+ O tl.. - W' ['1 + oil31 t, Iw Vo UQ = overall coefficient A = total outside area (finned and bare) t.w = average tube-side water temperature. OF W = flow rate Ib/hr Overall temperature difference: AT- = 181.9 - 72,14 = 109.8oF Tube-side At: At = 81,30 - 62.98 = 18,32 ~F Total heat transferred: L. Q = w C At Q = 7090 (l)(18.52) = 130,000 Btu/hr 1UoA y - =-UA U A AT x 104 Q 109.8 x 104 130,oo0 = 8-45 7 J 104 1.04 X = = 4.62 [1 +.o01l t] WO.8 [1 +,o01(72,14)]70900*8 Calculation of the overall coefficient7 Uo, for Figure 11 U = 0 Q AAT - 130,000,, I.9 = 55 7 Btu./hr-'F-'ft2. (21.247) (109.8) APPENDIX F SAMPLE CALCULATION OF' BOND RESISTANCE FOR WATER FLOWING INSIDE O'. 0,'OjNCH TUBE Computation of Outside Heat-Transfer Coefficient Based on All-Aluminum Tube for Water Flowing Inside Tube From Figure 5 (Wilson plot for the all-aluminum tube)7 the intercept at infinite water velocity is 4.18 with 71.90F tube-side water 23

-- ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN 104 = 4.18 UoA Therefore,.U = 104..... = 121 Btu./hr-~F-ft2 0 (19,78)(4.18) At infinite water velocity the following equation holds: - + (X)i A (6)A U0 ho ~Al() 1. = 1 _ (.0123)(3.553),00704. ho 121 117(.303) Therefore ho = 142 Btu/hr-ft2-F.. Evaluation of Bond Resistance From Figure 7 (Wilson Plot for 0.03-inch tube). the intercept at infinite water velocity is 4.92w 104 = 4,92 UoA Therefore Uo =.l = 95.5 Btu/hr-ft2-" F (21.247) (4*92) Again, at infinite water velocity the following equation may be written for the 0,03-inch Bimetallic tube: ( 1 Xe)A1 + RbAo. XCuA G —: -- +........ (7) UO ho (k)Al \ Al (Ao)Cu ku(m)C Since the all-aluminmn tube has approximately the same total outside area and the tests wqre controlled under the same conditions, ho calculated: for the all-aluminum tube is assumed to be the same as that of the Bimetallic tubes tested. Therefore, R 259 -i /._,. (0101,)(5.781).0025(31,781,) Rb 53781 i95.5 142 (117)(.290) (220)(.290)/.000148 hr'-F-ft2/Btu, i 24

_ __ ~ C 0- 60 PSIG GAGE 60 PSIG STEAM PIGTAIL COMPOUND TUBING CONTROL ( ) OAGE VALVES iE'l <i=i?, THERMOMETERS - - LN HE BATH- HOS STEAM INLET - - PRE WATE"R -— _ SPECIMEN TUBE -'' TEST BATH STEAM SPARGER JERGUSON DRAIN ULINE WATER ENTRAINMENT SEPARATOR ~~~~~~~~~~~~~~~~~SEPARATOR =- CONDENSATE cl'-. COOLER DRAIN LIQUID LEVEL CONTROL VALVE,, —---,. —------------- A —-----— I -1,I 1 0 0 0 M L r.~,. E l x AOD^A ^ ATI I C r A ik^"rc r,A Km C'"'IO, Ik FLASK r I k3u V I - rr CAl MUZ o AMAlM LL FIJ Lr JUV - DENSING STEAM INSIDE THE TUBE I

BATH INLET WATER 60 PSIG STEAM 0-30 P81G BACK PRESSURE GAGE BATH THERMOMETERS CONTROL INLET THERMOMETER OUTLET THERMOMETER BI-PASS TO I M —-1x= DRAIN TUBE SIDE WATER INLET TEST BATH STEAM SPARGER BATH WATER OUTLET TO WEIGH BARREL FIGURE 2 WATER APPARATUS ARRANGED FOR COLD FLOWING INSIDE THE TUBE.M..~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

52 a 0 a Io N IIL 0 I I. mu o AL 51 50 49 48 47 46,,. 45 - 44 43 A2 i - 0.02" I -METAL - TUBE 0.03" BI-METAL TUBE 0.04" Bl-METAL TUBE 0.05" B1-METAL TUBE ALL ALUMINUM TUBE 0.02" BI-METAL TUBE 0.03" B -METAL TUBE 0.04" BI-METAL TUBE 0.05" BI-METAL TUBE ALL ALUMINUM TUBE a1 I a A_ 0. _ _ _ _ __I'-l &, 25 PSIG STEAM CONDENSING INSIDE TUBES __ ~ 35 PSIG STEAM CONDENSING INSIDE TUBES FIGURE 3 - COMPARISON OF OVERALL HEAT TRANSFER COEFFICIENT FOR ALL TUBES TESTED WITH STEAM CONDENSING INSIDE TUBES

0.0002 0.00018 a. 0.00016 0 (o S 0.00014 i Z 0.00012 z 0.0001 0 z 0.00006 0 m 0.00004 0.00002 0 0 0.02 ACTUAL FIGURE WALL 0.03 0.04 ALUMINUM ROOT 4 - EFFECT THICKNESS ON 0.05 0.06 WALL THICKNESS, INCHES OF ALUMINUM ROOT BOND RESISTANCE 0.07

16 15 14 13 12 II 10 9 8 dC N'o / - AVE TUBE SIDE WATER TEMP. ~ 72.0 *F - < -i 1 -0- - AVE TUBE SIDE WATER TEMP. a 100.0 *F - AVE BATH TEMP. 182.0 OF ~|^..~<'^~~,- * AVE BATH TEMP. 181.5 *F FIGURE 5 - WILSON PLOTS FOR ALL ALUMINUM TUBE I-. --.,^ *. -~.. A.,,, ~ i. II I* I I I I I IB lIt Iu Il IA II 7 6 5 4 3! 0 1 I 0 4q a 10 + *oVD~ + 9 Itu IW 0.011OT W0'8 14e *r I o sW gv i. 1iv NW

17 16 14 13 12 II 1 1 09 8 7 6 5 4 _ _ __ _ _ _ _{___ 0 6 - AVE TUBE SIDE WATER TEMP. * 71.9 *F ] -G|_- -e- AVE TUBE SIDE WATER TEMP. * 99.9 *F I.I / I 1,- 1 1 ~/ I I i 6 - AVE BATH TEMP. a 181.4 *F,,,,,<@0'eX' - AVE BATH TEMP. * 181.8 *F _ _.G-_ _ _ 3/ ~ 5~FIGURE 6 - WILSON PLOTS FOR.02 INCH TUBE 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19,o4/E + O.OIIT W0~8 4!

I I I I I I o 0 mm mI [ m -- 14, e 3 2 X l| / TAVE TUBE SIDE WATER TEMP. * 71.7 F AVE BATH TEMP. ~ 181.9 *F 10 4 —. -aI, —,,,I FIGURE 7 - WILSON PLOT FOR 0.03 INCH TUBE 4' 3 i 0 I 2 3 4 5 6 7 8 9 10 II 104/! + 0.0IIT WO.S 12 1 14 15 16 17 18 19

16 15 14 13 12 II 10 AVE TUBE SIDE WATER TEMP * 71.9 *F >^i[ I fAVE BATH TEMP * 181.8 OF FIGURE 8 - WILSON PLOT FOR 0.04 INCH TUBE 4 9 0 a 7 6 5 4 3 k 0 1 2 3 4 5 6 7 8 9 10 II 104/E O.OIIT]W0~8 12 13 14 15 16 IT I1 I19

16 15 14 13 12 II 10?o 9 o 8 6 5 4 AVE TUBE SIDE WATER TEMP = 71.8 *F i 0 tAVE BATH TEMP = 181.5 ~F FIGURE 9 - WILSON PLOT FOR 0.05 INCH TUBE........ 11 i i.~ IB ~ I It al oI It I0 l 3 I 0 I 2 3 4 5 6 7 8 9 10 10 4/ + o.011o TW W.8 II IZ 13 14 10 10 Ir la IV

It Id a ro 0 0 -J 4 I CM IL IL to Iw 0 I&. IL 0 0 4 66 64 62 60 58 56 54 52 50 48 46 44 42 40 38. 36 34 32 - 30 28 0 F___ _____ 0_______ _ __ I I I I 0 E 0I Q/ AVE TUBE SIDE WATER TEMP ~ 71.9 *F AVE BATH TEMP * 181.4 *F JI Q 0 O / O 0 /0 FIGURE 10 - OVERALL COEFFICIENT WATER VELOCITY, 0.02 INCH TUBE Vs. I I II I I I I I I 2 3 4 5 6 WATER VELOCITY, 7 8 FT/SEC 9 10 11 12 13

4 Id IIQ 0 I l. 3 hIm 8 o k: 0 I 0 66 64 62 60 58 86 54 52 50 48 46 44 42 40 38 36 34 32 30 28 ( 0 AVE TUBE SIDE WATER TEMP * 71.7 *F /~ AVE BATH TEMP * 181.9 / FIGURE II - OVERALL COEFFICIENT VS. T I/ WATER VELOCITY, 0.03 INCH TUBE D I 2 3 4 5 6 7 8 WATER VELOCITY, FT/SEC 9 10 II 12 13

66 64 Id Id 30 W. 4 -:r 0 INm CIL o 0-% mL 0 U 0 a) Id 62 60 58 56 54 52 50 48 46 44 42 40 38 36 34 32 30 28 0 AVE TUBE SIDE WATER TEMP 71.'F / 0 __ _ _ __ __ _ _ __ j____0 __C AVE TUBE SIDE WATER TEMP, 71.9 OF 7 AVE BATH TEMP * 181.8 *F _ / l FIGURE 12 - OVERALL COEFFICIENT VS. WATER VELOCITY, 0.04 INCH TUBE 0 I 2 3 4 5 6 7 8 WATER VELOCITY, FT/SEC 9 10 II 12 13

4 ai 4 0 I- -. I0 0 z u. IL U_ 0 _I *AW a: %so 0 66 64 62 60 58 56 54 52 50 48 46 44 42 40 38 36 34 32 30 28 C - T _ _ _ ________ / | AVE TUBE SIDE WATER TEMP 71.8 F'/ AVE BATH TEMP * 181.5 *F FIGURE 13 - OVERALL COEFFICIENT VS. WATER VELOCITY, 0.05 INCH TUBE I l l l l0 -~~~~~~~~~ I I I ) I 2 3 4 5 6 WATER VELOCITY, 7 8 FT / SEC 9 10 II 12 13

4 w IIL or cu I 0 C Z 0 I_o z w IL I. I1 Ji 66 64 62 60 58 56 54 52 50 48 46 AA -- - - - - - - - -- -- ---- i i __ ___ __ ___ __ __ __ 0.3yi i i i i.0 o/ o/ - / I 0 - 0.02 - 0.03 -0 - 0.04 A - 0.05 INCH TUBE INCH TUBE INCH TUBE INCH TUBE 42 40 38 36 34 32 30 9_ 7 f/ / FIGURE 14 - COMPARISON OF THE FOUR S W R TUBES WITH RESPECT TO OVERALL COEFFICIENT AND WATER VELOCITY 0 1 2 3 4 5 6 WATER VELOCITY, 7 8 FT/ SEC 9 10 11 12 3

72 70 68 66 64 62 60 58 56 54 52 50 4 4 w a iO5 I0 4_ II&. I IL In z Ir z CU. x o o -J o 0 48 AVERAGE TUBE-SIDE WATER TEMPw71I. AVERAGE BATH-WATER TEMP ~ 181.5"F 46 42 40 38 36 32 FIGURE 15 - COMPARISON OF THE PERFORMANCE OF THE ALL ALUMINUM TUBE WITH THE FOUR BIMETALLIC TUBES 30 28 0 I 2 4 5 6 7 8 WATER VELOCITY, FT/SEC 9 10 II 12 13

68 / 4 4c Iz Io a hi 0 0 a hi (3 64 / 62 60 58 56 ---- --- --— ALL ALUMINUM TUBE 54 - L -- 52 _ /- -/0.02 INCH BIMETALLIUC TUBE 5048 -- 46 - AVERAGE TUBE-SIDE WATER TEMPs IOO1. 44 —-— / / AVERAGE BATH-WATER TEMP 181.5SF 42 - 40 38 36 - - 34 _ _ FIGURE 16 - COMPARISON OF THE 32 PERFORMANCE OF THE ALL ALUMINUM 30 TUBE WITH THE 0.02 INCH BIMETALLIC TUBE 28. — 0 2 3 4 5 6 7 8 WATER VELOCITY, FT/SEC 9 10 II 12 13

___ I-l I mI 4 -J a z s a o A 4 e A 0.A D E 1.00 - 0.90 - — r - s^ --— AVE TUBE-SIDE WATER TEMP * 71.9'F AVE BATH-WATER TEMPs 181.5-*F 0.80 A ____VE TUBE-SIDE WATER TEMP- 100.0F AVE BATH-WATER TEMP * 181.8~F I i 0.70 C I1 -1I - 1 I I a - - - - m ) 2 3 4 5 6 7 8 9 10 II 12 WATER VELOCITY, FT/SEC FIGURE 17 - RELATIVE PERFORMANCE OF THE 0.02 INCH TUBE AT TUBE-SIDE WATER TEMPERATURES OF 71.9'F AND IOO.OF

280 260 240 220. 200 U. 180 X ISO c 160. IL I 140 120 100 — FINNED SECTION --- ___ _ 1 CONDENSING STEAM 25 P81G AT" 120*F-AT' 112'-F AT 1041F AT 95F-F AT 86"-F AT 76-F0 0000 —---- _-BATH WATER _________________________________________ __________________________________________ __________________________________________ __________________________________________ _________________________________________ 1~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I., 0 I 2 3 4 5 6 ALL ALUMINUM TUBE LENGTH, FT. FIGURE 18- TEMPERATURE GRADIENTS FOR CONDENSING STEAM TESTS 7 8

2.40 b_ IL d C( z CO w Id 0 z a z 0 220 2.00 1.80 1.60 50 PER CENT. 6 PO' AREA POINT 1.40 - 25 PSIG 8TEAM 1.20 ____ TOTAL AREA UNDER oIo~ L ^ l | |CURVE ~ 6.90 FT. 1.0 0 _ _ _ __ __, 0.8 1 ---- 0.60 I —---- 0.40 I —- -------— I I --- I.j 0 0 I 2 3 4 5 6 FEET OF FINNED SECTION FIGURE 19 - VARIATION OF CONDENSING AREA WITH TUBE LENGTH AS A RESULT OF CONDENSATE BUILDUP

NOMENCLATURE A Total outside tube area in heat exchanger, ft2 Ai Inside tube area, ft2/ft of tube length Am Arithmetic mean metal area between Di and Do for copper liner, ft2/ft Ame Average tube-metal area between Di and De, ft2/ft AO Outside tube-surface area, ft2/ft of tube length Ao/Ai Ratio of outside to inside surface areas Cp Specific heat of fluid, Btu./(lb)(~F) De A diameter for finned tube equal to the diameter of a hypothetical plain tube with the same volumetric displacement (or with the same plane projection) as the finned tube, ft de A diameter for finned tubes equal to the diameter of a hypothetical plain tube with the same volumetric displacement (or with the same plane projection) as the finned tubes, in de d dr + y(do-dr)N Di Inside tube diameter, ft di Inside tube diameter, in. Do Diameter over the fins, ft do Diameter over the fins, in. Dr Finned-tube root diameter, ft dr Finned-tube root diameter, in, AH Change in enthalpy, Btu/lb hi Inside film coefficient, Btu/(hr)(~F)(ft2) ho Outside film coefficient corrected to base of fin, Btu./(hr)(~'F)(ft2) km Thermal conductivity of tube.wall, (Btu)(ft)/(hr)(~F) (ft2) N Number of fins per inch Q Total heat load, Btu/hr Rb Bond resistance,:(hr) (~F) (ft)/Btu. rm Tube metal resistance, (hr) (F)(ft2)/Btu. AT Overall temperature difference, ~F At Temperature difference between outlet and inlet fluid on tube side, (t2 - tx), ~F Tav Average bulk shell-side temperature, ~F tw Average water temperature in tubes, OF Uo Overall coefficient of heat transfer, Btu/(hr) (F) (ft2)outside surface W Total flow through tube side, lb/hr Xe Equivalent finned-tube wall thickness, ft, based on equivalent tube diameter De, ft Xe Equivalent finned-tube wall thickness, in., based on equivalent tube diameter, de, in. y Mean fin thickness, in. 44