ENGINEERING RESEARCH INSTITUTE THE UNIVERSITY OF MICHIGAN ANN ARBOR Report No. 44 AN INVESTIGATION OF THE FOULING OF 19-FIN-PER-INCH ADMIRALTY TUBES IN THREE HEAT TRANSFER UNITS LOCATED AT THE AURORA GASOLINE COMPANY REFINERY Edwin H. Young Associate Professor of Chemical and Metallurgical Engineering Marvin L. Katz Dennis J. Ward James, R. Wall William F. Conroy Walter R. Gutchess Clifford T. Terry Research Assistants Project 1592 CALUMET AID HECLA, INCORPORATED WOLVERINE TUBE DIVISION DETROIT, MICHIGAN October 1956

The University of Michigan Engineering Research Institute TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES OBJECTIVE ABSTRACT I. INTRODUCTION II. DESCRIPTION OF UNITS A. High-Pressure Gas Cooler B. Debutanizer Overhead Condenser C. Debutanizer Bottoms Cooler III. FIELD TEST PROCEDURE IVo THEORETICAL CONSIDERATIONS V. PRELIMINARY CALCULATIONS FOR ANALYSIS OF TEST DATA A. Finned-Tube Specifications B. External Heat Transfer Areas C. Water Flow-Rate Relationships D. Fouling-Resistance Equations VI. DISCUSSION OF RESULTS A. High-Pressure Gas Cooler B. Debutanizer Overhead Condenser C. Debutanizer Bottoms Cooler VII. CONCLUSIONS REFERENCES APPENDIX A - THE GRISCOM-RUSSELL CO. HEAT EXCHANGER SP] APPENDIX B - GRISCOM-RUSSELL DESIGN BLUEPRINTS APPENDIX C - SUMMARY OF TEST DATA ON THE HIGH-PRESSURE APPENDIX D - SUMMARY OF TEST DATA ON THE DEBUTANIZER 0' Page iv v V vi vi 1 1 1 3 15 3 3 9 12 15 14 17 22 22 27 31 355 37 38 42 49 51 I ECIFICATIONS GAS COOLER VERHEAD CONDENSER ii

The University of Michigan T Engineering Research Institute TABLE OF CONTENTS (Concl.) APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX I. II. APPENDIX APPENDIX PE E - SUMMARY OF TEST DATA FOR THE DEBUTANIZER BOTTOMS COOLER F - EXAMPLE CALCULATION OF TEST DATA ON THE HIGH-PRESSURE GAS COOLER G - EXAMPLE CALCULATION OF TEST DATA ON THE DEBUTANIZER OVERHEAD CONDENSER H - EXAMPLE CALCULATION OF TEST DATA ON THE DEBUTANIZER BOTTOMS COOLER J - SUMMARY OF CALCULATED RESULTS FOR THE HIGH-PRESSURE GAS COOLER K - SUMMARY OF CALCULATED RESULTS FOR THE DEBUTANIZER OVERHEAD CONDENSER L - SUMMARY OF CALCULATED RESULTS FOR THE DEBUTANIZER BOTTOMS COOLER M - DEBUTANIZER OVERHEAD CONDENSER-COMPUTATION OF THE CONDENSING RANGE OF THE HYDROCARBON STREAM Bubble -Point Calculations Dew-Point Calculations N - DEBUTANIZER OVERHEAD CONDENSER-CHECK ON THE OVERALL HEAT TRANSFER COEFFICIENT ASSUMING A NONFLOODED TOTAL CONDENSER 0 - CORRECTION FACTOR TO BE APPLIED TO THE LOG-MEAN TEMPERATURE DIFFERENCE IN A NONFLOODED, SPLIT FLOW CONDENSER age 55 55 58 6o 63 65 67 69 70 71 75 79 L iii

The University of Michigan * Engineering Research Institute LJST OF TABLES Table Page I Typical Test Data on t-e HigIh-Pressure Gas Cooler 7 II Typical Test Data on the Debutanizer. Overhead Condenser 8 III Typical. Test Data on thie 1Debu-tanizer Bottoms Cooler 9 IV Fin Resistance of a 3/4-Inch, 19-Fins-Per-Inch Admiralty Tuiec i1 V Admiralty T~ube Dimensions 13 VI Dimensions of Outlet Water Pipes and iOrifices for [i'nits Tested 15 VII Sumrlary of Equations Used in Analysis of Test Data 21 VIII Comparison of Field Load Conditions with1 Specif:ications for the High-Press >ure Gas Cooler 27 TX Comparison of Field Load Conditions with Specifications for the Debutanizer Orverh..ead Condenlse,' 35 X Comparison of Field Load Conditions wit- > Specifica-lti ons for the Debutanizer Bottorms Coolier: 35 iv

- The University of Michigan * Engineering Research Institute LI ST OF FIGTUES Figure Page 1. Flow arrangement of the high-pressiure gas cooler. 2 2. Flow arrangement of the debutanizer overhead condenser. 4 3. Flow arrangement of th:e delbutanizer bottoms cooler. 5 4. High-pressure gas cooler; variation of the computed fouling resistance with days of operation. 23 5. High-pressure gas cooler; variation of th.e comr.puted fouling resistance witn- heat dtlty. 24 6. Highi-pressur.e gas cooler; variation of the computed effective outside coefficient with h-eat duty. 26 7. Debutanizer overheead condenser; variation of the computed overall heat transfer coefficient with days of operation. 28 8. Debutanizer bottoms cooler; variation of t;e computed fouling resistance withi days of operation. 32 9. Wilson plots on th:e deb-utanizer bottoms cooler. 33 i0. Dew-point and bubble-poin- curves for a mixture of 50 inol, propane and 50 m;ol n —butane. 74 11. Condensing rarLge of a mix-t ure of 5{' mooL% propane and 50 nmol n -butane. 74 v

The University of Michigan * Engineering Research Institute r I OBJECTIVE The purpose of this investigation was to study the fouling rates of 3/4-inch, 19-fin-per-inch admiralty tubes in three heat-exchanger units located at the Detroit, Michigan, Refinery of the Aurora Gasoline Company. The units involved are identified as a debutanizer overhead condenser, a debutanizer bottoms cooler, and a high-pressure gas cooler. ABSTRACT Field test data collected over a period of nine months on a debutanizer overhead condenser, a debutanizer bottoms cooler, and a high-pressure gas cooler are presented. The heat transfer performance variations and fouling rates of the three units were computed from the test data and the results tabulated. The fouling results obtained on the debutanizer overhead condenser are inconclusive because of the effect of partial flooding and fluctuation in the amount of flooding on the overall heat transfer coefficients. The fouling rate of the debutanizer bottoms cooler was determined. The results on this unit indicate that the water-side fouling rate greatly affects the overall fouling of the tube. The fouling results obtained on the high-pressure gas cooler are inconclusive because this unit was operated as a partial condenser and it was not possible to determine the shell-side heat transfer coefficients from the test data. The composition of the shell-side vapor-liquid feed varied during the tests. J - vi

The University of Michigan T Engineering Research Institute I. INTRODUCTION Wolverine Trufin tubing is rapidly becoming accepted for certain oil refinery heat transfer applications. This report presents the results of an extensive field investigation of the fouling characteristics of Trufin tubes in a partial condenser, a total condenser, and a bottoms cooler in a local oil refinery. The data for this report were obtained from three shell and tube heat-exchanger units located at the Aurora Gasoline Company refinery, Detroit, Michigan. The field tests were made on these units over a period of nine months, starting August 21, 1955. The tests were started the fourth day after the units were first placed on stream. Field test data were obtained on twenty-seven occasions. The first five tests were conducted at about three-day intervals. This was followed by seven tests taken at approximately weekly intervals. After the third month the tests were conducted at intervals depending upon the plant operating schedule and the observed fouling trend of the units. II. DESCRIPTION OF UNITS The three heat-exchanger units on which test data were obtained are located in the gas concentration unit of the Aurora refinery. All the exchangers were initially tubed with 3/4-inch (diameter over the fins) admiralty Trufin tubes having 19 fins per inch. The specification sheets for the units are reproduced in Appendix A. The Griscom-Russell design blueprints are shown in Appendix B. Line diagrams showing the flow arrangements in the units are presented in Figs. 1, 2, and 3. A. HIGH-PRESSURE GAS COOIER The high-pressure gas cooler consisted of two heat exchangers operating in series as indicated in Fig. 1. This unit operated as a partial condenser. Midway through the test period the severe corrosive conditions existing in the high-pressure gas cooler necessitated retubing the upper exchanger of this unit. To gain better corrosion resistance, the exchanger was retubed with 3/4-inch-OD steel Trufin tubes having 19 fins per inch (Wolverine Catalog No. 195065-63). Thermowells were located in the following positions: (a) in the hydrocarbon inlet pipe line just above the inlet vapor nozzle of theupper condenser, (b) in the hydrocarbon outlet pipe line from the bottom condenser 1

_ HYDROCARBON INLET I I I I. H20 OUTLET U —— n I A-I, Illl. I I I -r I - ORI FICE -t 3 Zr _3. I'< 0 =r CI do I( q- 4-. —-- - -.- ~-1 —-. -- -I r) I, - Ill.- - - - - - — r LEGEND -- — WWATER FLOW - HYDROCARBON I --- THERMOMETER TEST POINTS iYDROCARB( OUTLET n Il JI' 4 H20 INLET, WATER MAIN rn 00 Sr m rp S" 00 70 m i n r+% f,+ c+ r+ m Fig. 1o Flow arrangement of the high-pressure gas cooler. I I I

The University of Michigan Engineering Research Institute about ten feet downstream and (c) in the outlet water line from the top condenser The piping arrangement on, the water side was such that water could be bled from the line connecting the water headers of the exchangers (water flowed in series) and from the water main supplying the cooling water to all the exchangers which were tested. A sharp-edged orifice was located in the outlet water line for determining the water flow rate through the exchangers. B. DEBUTANIZER OVERHEAD CONDENSER The debutanizer overhead condenser also consisted of two exchangers, but with the exchangers arranged in parallel as indicated in Fig. 2. This unit operated as a total condenser. Thermowells were located (a) in the inlet hydrocarbon line approximately ten feet before the Tee split the stream between the two heat exchangers, (b) in the hydrocarbon line leaving the condensate receiver, and (c) in the water line downstream from the Tee at which the outlet water streams from the two heat exchangers were joined. The water leaving each heat exchanger could be bled from the system before the Tee in order to obtain the individual exit water temperatures. A sharp-edged orifice was located in the outlet water line for determining the total quantity of water passing through both heat exchangers. C. DEBUTANIZER BOTTOMS COOLER The debutanizer bottoms cooler consisted of a single liquid-to-liquid exchanger, having two passes on both the shell and tube sides as indicated in Fig. 35 Its function was to cool the bottoms product from the debutanizer column, with no change of phase being involved. Thermowells were located in (a) the hydrocarbon inlet pipeline to the cooler, (b) the hydrocarbon outlet pipeline leaving the cooler, and (c) the outlet water pipeline leaving the cooler. A sharp-edged orifice was located in the outlet water line for measuring the total quantity of cooling water flowing through the heat exchanger. III. FIELD TEST PROCEDURE The industrial thermometers provided in the thermowells were not sufficiently sensitive for the test measurements. It was therefore necessary to contrive some means by which laboratory-grade, mercury-in-glass thermometers could be used for temperature measurements. The technique which was used consisted of filling the thermowells with grease, followed by insertion of the thermometers into the wells. The comparatively low thermal conductivity of the grease tended to insulate the the trmometer bulb from the effect of the outside air temperature. However, the grease also tended to damp out variations in the fluid temperature during a test run, due to the time lag required 3

The University of Michigan Engineering Research Institute r —--- I I,OUT - ORIFICE I "" I i I TVI I HYDR N I T - I HYDROCARBON INLET I I I I I!i.. I III I 1111 I I I L —-; -,J_ (?[t"" --- \Zz 4 I I I I 0~~ 0w I I 4 I L_ —- __________; A H20 I INLET WATER MAIN ( CONDEN SATE COLLECTING TANK ) LEGEND - - - - WATER FLOW - ---- HYDROCARBON -1-1 —- THERMOMETER TEST POINTS <CZ~ THERMOCOUPLE TEST ri- I POINTS Eo Not1|HlH1. —, —, Fig. 2. Flow arrangement of the debutanizer overhead condenser. 4

--- The University of Michigan Engineering Research Institute I - -ORIFICE H - I ~ A. % I 0 H2U I UTLET A I HYDROCARBON INLET H20 1, INLET HYDROCARBON I OUTLET WATER MAIN LEGEND I ----- WATER FLOW ---- HYDROCARBON FLOW +Htl — THERMOMETER TEST POINTS Fig. 3. Flow arrangement of the debutanizer bottoms cooler. 5

The University of Michigan * Engineering Research Institute for the grease to reach a new stasble temperature. At some of the test points, such as the water intermediate between the two exchangers on the high-pressure gas cooler, water was bled from the line, with a mercury-in-glass thermometer being immrersed in the water stream for a temperature measurement. A problem was encountered on the debutanizer ovexrh-lead condenser in that the thermowell for measurement of the outlet hydrocarbon temperature was located a considerable distance downstream from the unit, with a large condensate collecting tank being situated between the thermometer and the test unit. The condensate residence time in this tank had some effect on the fluid temperature, due to heat losses. Therefore; there existed some question as to the actual temperature of the hydrocarbon stream leaving the test unit. In an attempt to overcome this difficulty, the.rmocouples were embedded in the wall of the hydrocarbon outlet pipes from each exchanger, and a layer of insulation was placed over the thermocouples. However, due to various sources of error, the temperature measurements taken with these thermocouples did not appear to be reliable and were not used in the analysis of the test data. The inlet water temperature for, al-l units was measured by bleeding from the inlet water main feeding all three test units. This test point was located slightly upstream from all the units. The water flow rate through the units was determined by measurement of the pressure'drop across the orifices in the outlet water lines. This measurement was made with liquid-filled manometers. For most of the runs, mercury was used as the manometer fluid. However, to gain accuracy at the low flow rates such as we.re used in thLe Wilson Plot runs on the debutanizer bottoms cooler, dyed cartbon tetrachlliloride was used as the manometer fluid. During the first phase of the investigation all three units were tested. The temperature-s and flow rates of the units were not adjusted, but were recorded as they occurred in normal plant operation. It soon b-ecame apparent thlat the debu-tanize:r bottoms cooler gave thte most accurate indication of a fouling trend. It was t..erefore decided that the majority of the test data should be taken on this unit. Permission was received to vary the water flow rate thirough thne exchanger for short periods of time so that Wilson Plot type data could be taken. This allowed a more accurate analysis of the fouling trend on the unit. No Wilson Plot test runs were made on the other two test units. Afte- 2-1/2 months of operation the test: data indicated that the bottoms-cooler fouling:resistance had exceded ceede te design fouling resistance In order to improve the heat transfer performance of this and thle other units, all the units were cleaned during the next plant shutdown early in November, 1955. Testing of this unit was resumed on Novemrite: 18, 1955. After receiving i j 6

I I - The University of Michigan * Engineering Research Institute permission to vary the water flow rate through thlebottoms cooler for short periods of time, the first Wilson Plot run was made on December 23, 1955. At the time the units were cleaned an evaluation of the test procedure was undertaken. It was decided that since thermowells could not be installed in the hydrocarbon outlet lines leaving each of the debutanizer overhead condensers, sufficiently accurate test data for determining the fouling trend of this unit could not be obtained. Therefore, no further test data were taken on this unit after October 21, 1955. Also at the top heat exchanger of caused by corrosion. exchanger was retubed December 10, 1955. time of cleaning of the units over fifty tubes in the the high-pressure cooler were plugged due to leakage No further test data were taken until after this heat with steel tubes. The next test data were taken on Excessive fouling of the debutanizer bottoms cooler was apparent from the test data taken on February 1, 7, and 18, 1956. This fouling condition was discussed with the plant engineering office personnel on March 1, 1956. The project group was informed at this meeting that the high fouling resistance calculated from the field test data was due to oil contamination of the recirculated cooling tower water. On March 3, 1956, the tube side of this unit was cleaned by plant personnel. Test data were obtained just before and immediately after the cleaning. The testing was then continued at regular intervals until May 26, 1956, at which time all field tests were terminated. Typical test data for the throee units are presented in Tables I, II, and III. TABLE I ITYPICAL TEST DATA ON TIE HIGH-PRESSURE GAS COOLER I I Date: May 26, 1956 Run No. 21 Time Hydrocarbon Water (~F) Manometer (p.m.) (~F) Inter- (Inches of CC14) In.Out.In Mediate Out Left Right 12:25 139.5 113.2 70.6 90.6 116.7 +13.4 -12.6 12:25 139.2 113.5 70.6 91.6 116.6 +13.4 -12.6 12:27 139.0 113.5 70.3 91.0 116.7 +13.3 -12.6 12:29 139.0 113.5 70.6 90.8 116.7 +13.2 -12.6 12:31 136.5 113.2 70.6 90.4 116.6 +13.2 -12.6 12:33 135-5 113.4 70.6 90.5 116.6 +13.3 -12.7 12:35 135.3 113.5 70.7 90.8 116.6 413.3 -12.7 12:37 135.0 11.36 70.7 91.5 116.6 +13.4 -12.6 Average 137.58 113.43 70.59 90.83 116.66 +151 — -12.63 Manometer Temperature = 68 F 7

l TABLE II TYPICAL TEST DATA ON THE DEBUTANIZER OVERHEAD CONDENSER Date: October 14k, 1955 Run No.: 10 Time Water Water Out Hydrocarbon Hydrocarbon Out Thermocouple Readings Manometer In Exchanger A Exchanger B Mixed Mean In After Collecting Exchanger A Exchanger B (inches of Hg) (p.m.) (~F) (~C) (~C) ( F) (~F) Tank (~F) mv ~F mv ~F Left Right 3:50 76.0 33.7 34.5 92.4 144.4 102.0 1.210 64.30 -3.0 +3.0 3:52 75.9 33-7 34.3 92.3 144.2 101.0 -3.0 +5.0 3:54 75.7 33.6 34.5 92.3 144.0 99.2 -3.0 +2.9 3:56 75.7 34.2 35.0 93.0 144.0 98.0 1.020 67.50 -2.9 +2.8 3:58 75.5 34.1 35.2 93.4 144.1 98.2 -2.8 +2.8 4:00 75.5 33.9 35.1 93.3 144.2 99.2 -2.9 +2.8 4:02 75.3 34.2 35.1 93.0 144.5 100.0 1.110 64.30 -2.9 +2.8 4:04 75.5 34.3 35.0 93.4 144.6 100.9 -3.1 +2.8 4:06 75.5 34.3 35.3 93.5 144.7 101.8 -2.8 +2.7 4:08 75.2 34.0 35.0 92.8 144.3 102.0 1.130 67.50 -2.5 +2.4 4:10 75.1 34.4 35.6 94.3 144.2 99.7 -2.3 +2.2 4:12 75.0 55.0 35.7 94.5 144.5 98.6 1.200 66.00 -2.3 +2.2 4:14 75.0 55.0 35.8 94.5 144.7 98.9 -2.3 +2.2 4:16 74.8 34.8 35.8 94.5 145.0 100.0 -2.5 +2.4 4:18 75.0 34.0 35.1 93.0 145.2 101.0 -2.9 +2.8 4:20 75.0 33.9 34.8 92.5 145.2 101.2 -3.1 +3.1 Average 75.37 34.19 35.11 93.29 144.49 100.10 1.170 64.87 1.070 67.50 -2.77 +2.68 ~~~~~~~6. 0 -. 7+.6 r_ M ro w, -'A _. S _. __ 3 m, 3 m_ 3 _. 3* fI S _. Ko n^ 31 CO' 2) ro M S" M go

The University of Michigan T Engineering Research Institute TABLE III TYPICAL TEST DATA ON THE DEBUTANIZER BOTTOMS COOLER Date: January. 13, 1956 Run No.: 27 Time Hydrocarbon Water Manometer (p. n) (~F) (~F) (inches of Hg) In Out In Out Left Right 2:05 177.0 75.0 70.1 86.8 -1.45 -3.85 2:07 177.0 74.8 70.3 86.5 -1.45 -3.85 2:09 178.0 75.0 70.4 86.8 -1.45 -3.85 2:11 177.0 75.2 70.5 86.7 -1.40 -3.80 2:13 178.0 75.7 70.4 86.6 -1.40 -3.80 Average 177.40 75.14 70.37 86.68 -1.43 -3.83 A summary of all the test data collected on the high-pressure gas cooler, the debutanizer overhead is presented in Appendices C, D, of the test data given in Tables F, G, and H, respectively. condenser, and the debutanizer bottoms cooler and E, respectively. Example calculations I, II, and III are presented in Appendices IV. THEORETICAL CONSIDERATIONS The overall coefficient of heat transfer is defined by the equation Q = UO Ao ATM, (1) in which Q = rate of heat transfer, Btu/hr, Uo = overall coefficient of heat transfer, Btu/hr-~F-ft2 outside area Ao = outside heat transfer area, ft2, and ATM = mean temperature difference driving force, ~F. The overall coefficient of heat transfer is related to the individual heat transfer coefficients by 1 1 AO Ao AO = =+O'+ rf0 + rf + A m + Am Uo ho, Aihi Am mi (2) 9

The University of Michigan * Engineering Research Institute in which ho' = outside film coefficient of heat transfer for a finned tube, Btu/hr-~F-ft2 outside area, ro' = outside fouling film resistance for a finned tube, hr-~F-ft2 outside area/Btu, rf = fin resistance as defined by Equation 3, hr-~F-ft2 outside area/Btu, hi = inside film coefficient of heat transfer, Btu/hr-~F-ft2 inside area, rm = root-wall metal resistance, hr-~F-ft2 mean metal area/Btu, ri = inside fouling film resistance, hr-~F-ft2 inside area/Btu, Ai = inside heat transfer area, ft2 and Am = logarithmic mean metal area, ft2. The fin resistance1 is defined by the equation rf =. — ro A1 1 Ef (3) ho' Ar L+ Ef in which Ef = fin efficiency of Gardner.2 The fin resistance of the admiralty tubes used in the heat exchangers is given in Table IV. Examination of Table IV indicates that the fin resistance may be assumed constant over the range of coefficients encountered in the heat exchangers tested. TABLE IV FIN RESISTANCE OF A 3/4-INCH, 19-FINS-PER-INCH ADMIRALTY TUBE 1 -1 r. rf 0.0001128 50.0001127 100.0001125 500.00011121000.0001097 The heat transfer rate for a unit may be obtained from a heat balance on the water side of the unit, so that Q = W Cp AtH20 (4 10

r The University of Michigan * Engineering Research Institute i 1 in which W = water flow rate, lb/hr, Cp = heat capacity of water, Btu/lb-~F, and AtHO = temperature rise of the water, OF. The inside coefficient for water inside tubes may be calculated from the equation3 h 150(1 + 0.011 tw) Vt08 hi = tw) Vt(5) d 0.2 di~.a in which tw = average water temperature, ~F, Vt = water velocity inside the tubes, ft/sec, and di = inside tube diameter, inches. Since the flow rate of the shell-side fluids was not directly measured and the composition of the shell-side stream was unknown for the majority of the test points, the outside (fin side) coefficient could not be directly calculated. In the case of the debutanizer bottoms cooler, assuming a constant heat capacity for the shell-side fluid, the shell-side flow rate is proportion al to the ratio of the rate of heat transfer, Q, to the temperature drop of the hydrocarbon stream. Also, as indicated in the Williams-Katz report,4 the shell-side coefficient is proportional to the shell-side mass flow rate to the 0.65 power. Therefore, it was assumed that 0.65 hot c Ct —- 1, (6) thydrocarbon where ho' = fin-side coefficient in the bottoms cooler and C = a constant (to be experimentally determined). The constant was determined from analysis of the data taken with the tubes in a nonfouled condition. In the case of the high-pressure gas cooler the shell-side coefficient could not be determined in the manner indicated above since the method is only applicable to units in which no phase changes occur on the shell side. The shell-side coefficients for this unit could be determined from the test data taken on the unit in a nonfouled condition by use of Equation 2 (assuming rot and ri to be zero). This coefficient was assumed to be constant throughout the test period. i - 11

- The University of Michigan * Engineering Research Institute The shell-side coefficient on the high-pressure gas cooler could not be calculated due to reasons to be explained later in the report. The fouling resistances for the bottoms cooler were obtained from the field test data by 1. computing Uo from the data, using Equation 1, 2. computing hi from the data, using Equation 5, 3. computing hot from the data, using Equation 6, and 4. solving Equation 2 for the fouling resistance, (0 + Ao ri The factors ro' and ri in this group cannot be separated and evaluated using the field test data. The fouling resistances for the high-pressure gas cooler were obtained from the field test data by 1. computing Uo from the data using Equation 1, 2. computing hi from the data using Equation 5, 3. assuming ho' to be a constant equal to the value determined for the unit in the nonfouled condition, and 4. solving Equation 2 for the fouling resistance, (r +.r.r The fouling resistances for the debutanizer overhead condenser could not be calculated (see Section VI-B). V. PRELIMINARY CALCULATIONS FOR ANALYSIS OF TEST DATA A. FINNED-TUBE SPECIFICATIONS The dimensions of the admiralty tubes used in all thre inunits are tabulated in Table V. These specifications were taken from the Wolverine Trufin Catalog C for tube No. 195065-26. This catalog number corresponds to the tube specified for the units on the blueprints shown in Appendix B. 12

The University of Michigan E Engineering Research Institute TABLE V ADMIRALTY TUBE DIMENSIONS Catalog Number: 195065-26 Specified Outside Diameter: Alloy: Admiralty 3/4 inch Plain End Outside Diameter: Wall Thickness: Finned Section Diameter over Fins: Root Diameter: Wall Thickness: Outside Area: 0.747 - 0.755 inch 0.078 - 0.086 inch 0.732 0.633 0.061 - 0.742 - o.647 - 0.069 inch inch inch 0.438 ft2/ft of length From correspondence with Wolverine Tube, the determined to be area ratio for the tubes was Ao/Ai =.28 The following calculations present transfer areas, (b) the flow-rate equations, equations which were used in the analysis of the determination of (a);the heat and (c) the fouling-resistance the field test data. B. EXTERNAL HEAT TRANSFER AREAS 1. High-Pressure Gas Cooler.-From the heat-exchanger blueprints (Appendix B), tube length = 196 inches, finned length = 185-1/2 inches, number of tubes = 1300 (650 per shell), Ao = 0.438 ft2/ft of length (from Table V) total outside area = (1855) (1300)(0.438) = 8802 ft2. The Griscom-Russell specification sheet (Appendix A) gives 5425 ft2 of surface area per shell for a total area of 10,850 ft2. The value of 8802 ft2 computed on the basis of the blueprint notation calling for Wolverine Trufin No. 19506526 tube was used in analyzing the test data taken before December 5, 1955. The top heat exchanger in this unit was retubed on December 5, 1955, with steel finned tubes (Catalog No. 195065-63) having an Ao (outside area per 13

L The University of Michigan * Engineering Research Institute foot of length) of 0.496 ft2/foot. The total external area of this unit after the top exchanger was retubed was calculated as A = (1 5)(650)(0.438) + (1815)(650)(0.496) = 4401 + 4984 = 9385ft2. This external heat transfer area was used in analyzing all data taken on this unit after December 5, 1955. 2. Debutanizer Overhead Condenser.-From the heat-exchanger blueprints (Appendix B), tube length = 196 inches, finned length = 185-1/2 inches, number of tubes = 1410 (705 per shell), Ao = 0.438 ft2/ft of length (from Table V), total outside area = 8- 5 (1410)(0.438) = 9570 ft2. The Griscom-Russell specification sheet (see Appendix A) gives 5870 ft2 of surface area per shell for a total area of 11,740 ft2. The value of 9570 ft2 computed on the basis of the blueprint notation calling for Wolverine Trufin No. 195065-26 tube was used in analyzing the test data. 3. Debutanizer Bottoms Cooler.-From the heat-exchanger blueprints (Appendix B), tube length = 192 inches, finned length = 185-1/2 inches, number of tubes = 370, Ao = 0.438 ft2/ft of length (from Table V), total outside area = (185 (570)(0.438) = 2505 ft2. The GriScom-Russell specification sheet (see Appendix A) gives 3050 ft2 of surface area for this unit. The value of 2505 ft2 computed on the basis of the blueprint notation calling for Wolverine Trufin No. 195065-26 tube was used in analyzing the test data. C. WATER FLOW-RATE RELATIONSHIPS The outlet water pipe diameters and water orifice diameters for the three units are given in Table VI. The water flow rate can be computed from the pressure drop across the orifice by use of5 W = CAo 2gp (-P) (7) V " UAi/ I 14

The University of Michigan E Engineering Research Institute TABLE VI DIMENSIONS OF OUTLET WATER PIPES AND ORIFICES FOR UNITS TESTED Outlet Water Pipe Unit Nominal Diameter Inside Diameter Orifice Diameter (inches) -(inches) (inches) High-Pressure Gas Cooler 8 7.981 5.147 Debutanizer Overhead Condenser 10 10.020 6.543 Debutani zer Bottoms Cooler 6 6.065 4.457 in which W Co Ao gc p P -AP Ai water flow rate, lb/hr, =orifice coefficient, dimensionless, flow area of the orifice, ft2 =conversion factor, (lb mass) ft/ (lb force) hr2, =water density, lb/ft3, =pressure drop across the orifice, lb/ft2, and =inside flow area of the pipe, ft2. The orifice coefficient, CO, in Equation 7 is a function of the Reynolds number in the orifice. For Reynolds numbers greater than 30,000, the coefficient can be assumed to have a constant value of 0.61.5 During the tests the orifice Reynolds numbers were all computed to be greater than 30,000. 1. High-Pressure Gas Cooler.-Substituting 0.61 for the value of the orifice coefficient and the dimensions of the orifice and pipe (for this unit) from Table VI into Equation 7 gives W = 178,000 N-AP, (8) where -AP' is the manometer reading in inches of mercury. In the case of the test data which were taken with carbon tetrachloride as the manometer fluid, AP = 20 at 4P" -AP' - -j — at 400~F (, 20?a) and (9b ) -At = iat ~F -AP' = 2 at 70`F, 15

The University of Michigan * Engineering Research Institute where -AP" is the manometer reading in inches of carbon tetrachloride. The water velocity inside the tubes was computed using thy inside diameter obtained from the dimensions given in Table V. The average inside diameter is computed from the dimensions in Table V as di = dr - 2(Xwall) = 0.64 - 0.13 = 0.51 inch. The inside flow area = (05) N = 000142 N ft2 (10) 4 x 144 where N is the average number of tubes per pass on the water side. Therefore, the total inside flow area = 0.00142 x 100 = 0.462 ft2. Assuming an average water density of 62 lb/ft3, W W Vt = (62)(3600)(0.462) = 103000 ft/sec (11) 2. Debutanizer Overhead Condenser.-Substituting the value of the orifice coefficient. and the dimensions given in Table VI for this unit into Equation 7, W = 288,000 1-AP, (12) where -AP' is the manometer reading in inches of mercury. For the test data obtained on this unit using carbon tetrachloride as the manometer fluid, Equations 9a and 9b were used to obtain -AP'. The average number of tubes per pass on the tube side of this unit is trom the blueprints) (1410/2) or 705 tubes. The total water flow area is flow area = 0.00142 N = (0.00142)(705) = 1.002 ft2. Assuming an average water density of 62 lb/ft3, Vt = -- W IW ft/sec. (1) (62)(3600)(1.002) 223,000 3. Debutanizer Bottoms Cooler.-Substituting the values of the orifice coefficient and the dimensions given in Table VI for this unit into Equation 7, W = 144,000, — AP', (14) 16

I The University of Michigan T Engineering Research Institute where -AP' is in inches of mercury. For the test data obtained on this unit with carbon tetrachloride in the manometer, Equations 9a and 9b were used to convert the pressure drop to inches of mercury. The average number of tubes per pass for this exchanger is (from the blueprints) (370/2) = 185 tubes per pass. Substituting in Equation 10, flow area = 0.00142 N = 0.00142 (185) = 0.262 ft2. Again assuming an average water density of 62 lb/ft3, Vt (62)( = = o W ft/sec. (15) (62)(3600)(0.262) 58,500 D.i FOULING-RESISTANCE EQUATIONS 1. High-Pressure Gas Cooler.-From the initial test data obtained on August 2, 1955, the overall heat transfer coefficient was determined as (see Appendix J) Uo = 24.35 Btu/hr-~F-ft2. The computed average water velocity was Vt = 3.60 ft/sec with an average water temperature of 101.92~F. Substituting the values of the tube inside diameter (see Section V-B-1), the average water velocity, and the average water temperature into Equation 5, the inside-water film coefficient is computed as hi _ 150[1 + 0.011(101.92)] (3.60)0-8 hi (0.51)0.2 = 1013 Btu/hr-~F-ft2. The corresponding inside water film resistance is Ao = 3.28 = 0.003235 Ai hi 1013 Assuming negligible fouling present on August 2, 1955, the sum of the fin, root metal, and outside film resistance can be calculated from Equation 2 as 17

The University of Michigan * Engineering Research Institute 1 Ao 1 Ao --- + rf +r = --. (16a) ho' r mAm Uo Ai hi Substituting the above values of UO and Ao/Ai hi into Equation 16a gives 1 Ao l - + rf + rm = 24 - 0.00323 = 0.0379. (16b) ho, + +m 24.35 Assuming a constant outside film coefficient, the fouling resistance can be computed by use of the following relationships obtained from a rearrangement of Equation 2: A 1 O 1 Ao 1A r[i + = o hi + rf + rm (17a) or, substituting from Equation 16b, AO 1 _ Ao'. [A~ ri + ro = U -A hi 0. (17b) For the data obtained after the top exchanger had been retubed, the value of the constant given in Equation 17b was revised to include the added heat transfer resistance of the steel tubes. The resulting relationship was [A ri + ro = U A- + 0.0382 (17c) LAi ] U j Ai hi The steel tubes had an (Ao/Ai) ratio of 3.86 compared to the value of 3.28 for the admiralty tubes. Therefore,an average value of Ao/Ai = 3.57 must be used to compute the value of Ao/Ai hi in Equation 17c. 2. Debutanizer Overhead Condenser.-No fouling results could be computed from the data taken on this unit (see Section VI-B). 3. Debutanizer Bottoms Cooler.-From the initial test data obtained on August 2, 1955, the overall heat transfer coefficient was determined as (see Appendix L) Uo = 64.8 Btu/hr-~F-ft2. The average water velocity was computed as Vt = 4.83 ft/sec with an average water temperature of 110.73~F. 18

L The University of Michigan T Engineering Research Institute Substituting the values of the tube inside diameter, the average water velocity and the average water temperature in Equation 5, the inside water film is computed as hi = 150[1 + 0.011(110.75)] (4.83)0.'8 hi=-1- ~ (0.51)0.2 = 1346 The inside water film resistance is then AO = 3.28 0.00244 Ai hi 1346 The root metal resistance is computed as Ao X Ao o.o65 (0.141) rmA kim Am (12)(65 )(0.438) = 0.00027 hr-~F-ft2/Btu From Table IV, the fin resistance is rf = 0.000113 hr-~F-ft2/Btu Assuming that the degree of fouling on August 2, 1955, was negligible, the outside film coefficient is computed as ho1 1 [A + r1A + rf' — - - UO Ai +j MTM = 648 - [0.00244 + 0.00027 + 0.000113] 64.8 = 0.01272 or h' 0.01272 =78.5 Btu/hr-~F-ft2. Using the values of Q (heat transfer rate) and Athydrocarbon (temperature drop of bottoms fluid) obtained on August 2, 1955, with the value of ho' given above, the value of the constant C in Equation 6 can be evaluated: Q 0.65 ho = C [At -] ho, = C[rAthydrocarbon or 19

The University of Michigan * Engineering Research Institute 78.5 78.5 9.37 x 10o 887= 0. 273.12 Therefore, for the debutanizer bottoms cooler, h o' = 0.0886 t Qr *. (18) AthydrocarbonJ For this unit, the relationships used to determine the fouling resistance were i + ro U [AiA hi+ rmA + rf + h' (17a) - Ai Uo A+m o t A0 Ao or substituting in the values of rm - and rf given above (rm ~ = 0.00027 and rf = 0.000115), m m 1 AO ho1 [ 1-1i + ro' = o- [Ai + - + 0.00038 (17d) Ai] UO [Ai hi ho' where ho' is determined from Equation 18. The Wilson Plot test data were analyzed by rearranging Equation 17d to give 1[_ h1 ] A~+ - -ri + ro] + 0.00038.(19) L. O ho: Ai hi ai A plot of the value of [7 - ho vs A. gives a straight line which has a slope of one and an intercept value equal to H i + roJ + 0.00038. [Ai J The intercept value can be used to obtain the combined fouling resistance. 4. Summary of Relationships.-The areas and relationships which were used in analyzing the field test data are summarized in Table VII. 20

TABLE VII SUMMARY OF EQUATIONS USED IN ANALYSIS OF TEST DATA ro H Total Outside Water Flow Water Velocity Fouling Resistance Unit Heat Transfer Rate in the Tubes. ri + ro' - hr-ft2OF/Btu Area, ft2 lb/hr, ft/secLi.1....i. High-Pressure A Gas Cooler 8802 W = 178,000 (-AP' Vt = 1 = + 0.0379 (before retubing) 103, u0 1 High-Pressure W Gas Cooler 9385 W = 178,000 T1 (_Ap ) Vt - +0.082 (after retubing) )1 0 u hi Debutanizer Over- Cannot be computed from the test head Condenser 9570 W = 288,000 (-AP) Vt = data 225,000 Debutanizer Bottoms r Cooler 2505 W = 144,000 (-^P) Vt = W - + h + 0.000358 58,500 U0 LAihi ho'0 where ho = 0.o886 - --— oab [Abhydrocarbon j 3m C, m 3 0') cm m -S _c m 23 00 3* rI "^

I The University of Michigan * Engineering Research Institute VI. DISCUSSION OF RESULTS A. HIGH-PRESSURE GAS COOLER A summary of the calculated results for the high-pressure gas cooler is presented in Appendix J. The fouling resistances computed for this unit are plotted in Fig. 4 vs the number of days of operation and are plotted in Fig. 5 vs the heat duty of the unit. No test data were obtained on this unit during the two-month period from October 14, 1955, to December 10, 1955, because of tube failures. Approximately one-half of one percent of the tubes had been plugged by October 14, 1955. Thereafter, a number of additional tubes were plugged, and arrangements were made by Aurora Gasoline Company for retubing of the upper heat exchanger. No further test data were taken until the retubed bundle was installed and the unit was placed back on stream. Arrangements had been made to have two thermowells installed in the hydrocarbon outlet nozzles of the upper heat exchanger during this retubing period. This modification was not feasible because the nozzle-wall thicknesses were too thin to take screwed fittings and still meet code requirements. The testing of this unit was resumed on December 10, 1955. The fouling resistance of this unit was computed from the field test data, using the assumption that the outside film coefficient remained constant and was equal to the value determined for the unit in the nonfouled condition (see Section IV). This assumption would be essentially correct if (1) the total mass flow rate of the hydrocarbon gas stream and the ratio of condensables to noncondensables remained constant, or (2) a fortuitous combination of the above two variables occurred, such that the shell-side coefficient remained constant. Examination of the field test data for this unit (see Appendix C) indicates that the shell-side heat transfer conditions varied widely during the course of the investigation. This indicates that the above assumption of a constant outside coefficient was not true for many of the test runs. The outside coefficient in a partial condenser is a function of (1) the mass flow rate of the hydrocarbon noncondensable gas stream, (2) the amount of condensables present in the stream, (3) the physical properties of the fluids, and (4) the tube-wall temperature. An increase in the heat duty of a partial condenser (for approximately the' same inlet hydrocarbon temperature) indicates either (1) an increase in the flow rate of noncondensables, (2) an increase in the flow rate of condensables, or (5) an increase in both of these items. Since all the above three conditions increase the shell-side film coefficient, it would be expected that for the runs in which the heat duty exceeded that of August 2, 1955 (Q>6.13 x 1l0 Btu/hr), the computed fouling resistance would be less than the actual fouling resistance. In the case of J I 22 -

The University of Michigan * Engineering Research Institute 0 ra) 0 *H 4o *H L 0 0 o.0 0 0 0 0 o. C OD.,~~ 0 ~~o.0 0 03__I3NHX3 dO__ 0 O _3NV- _Nn ~0 O0 0 0 _ _CD 0 0 0 0 (0 0 N 0 ---------- 0 -- -------- -^ - -- - -- -- 0o i_,....... I -------- --- -- 1 -- \ -- a -- I -- I -- 1 -- I -- L ---- I 0 CD Jrd r.. U) *H 0 C) V(U) F-1 a, CH cz a) r - L 0 o 0 n. o~ o'e L O'H I s b a) Cl) a) *d O CO bO (Ul a, *HI - ojN N - - 0 0 - o o o o o. 0 + + - 4- +' I nei /oz l H - H[, t01J a I ov~~~~~~ i) 0 i) - CJN N 00 0 I I I v bD *H Pr

L The University of Michigan Engineering Research Institute =Z) \I I I I U I- +.006 - 0 +.004 gr h 0 -.002 -0 2 4 6____ 0___ -__ ___ 0 0 - -.002 --- - 006 0 0 -.008 -.010 -.012 0 -.014 -.016 -.018 -.020 0 0 2 4 6 8 10 12 [Q x I0'] -BTU/HR Fig. 5. High-pressure gas cooler; variation of the computed fouling resistance with heat duty. 24 I

The University of Michigan * Engineering Research Institute 1 the runs in which the heat duty of this unit was less than 6.13 x 106 Btu/hr, the computed fouling resistance would be expected to be above the actual fouling resistance. Thus, the variation in the computed fouling resistance shown in Fig. 5 includes a variation in the shell-side coefficient. The logarithmic mean temperature difference was used in computing all the overall coefficients given in Appendix J for this unit. The use of this temperature difference implies the assumption that the heat transferred from the hydrocarbon stream per degree of temperature drop is a constant. The composition of the hydrocarbon stream for this unit was not available to check this assumption. Using the following approximate relationship for the shell-side heat transfer coefficient in a partial condenser,l heff. = ho QT (20) in which heff = effective outside coefficient for partial condensers, ho' = gas film coefficient, QT = total heat duty, and QS = sensible heat duty (heat duty not associated with a change of phase), and assuming a constant QS, sensible heat transfer, and a constant gas film coefficient, ho', the effective coefficient is related to the total heat duty of the exchanger by heff. = KQT' (21) in which K = a constant = QS A plot of Equation 20 on rectangular coordinates will result in a straight line passing through the origin. This is illustrated in Fig. 6, where the computed shell-side coefficients (assuming no fouling-) are plotted vs the heat duty of the exchanger. The straight line given on this figure was drawn through the origin and through the test point computed froih the data taken on the unit in a nonfouled condition (data of August 2, 1955). It is significant to note that, although the experimental data points plotted in Fig. 6 scatter considerably, the majority of the test points lie below the line. If the inlet conditions for the high-pressure gas cooler were such that the assumption of a constant QS and ho' proved to be valid, all the points computed from data taken with the tubes ina- fouled condition would necessarily fall below the line. The scatter shown on the graph indicates that the inlet conditions varied sufficiently to cause QS and ho' to change somewhat during the course of the investigation. The quantitative effect of the varying inlet conditions on the computed fouling resistance could not be computed due to the lack of the hydrocarbon stream analysis. However, the.1 I 25

I L The University of Michigan 60 55 50 45 40 00 Ift: 30 I 0 2C 5 I ------------ -- ---- Engineering Research Institute [Qx1O6] -BTU/HR Fig. 6. High-pressure gas cooler; variation of the computed effective outside coefficient with heat duty. 26

I The University of Michigan * Engineering Research Institute fact that the majority of the test points lie below the line indicates that the tubes fouled somewhat during the course of the test investigation. The equation relating heff and Q, which was plotted in Fig. 6, was also used to determine a reference line for Fig. 5. The line shown in Fig. 5 represents the curve which would be obtained if the variation in Q for the unit which was observed during the test period were due entirely to variations in the outside coefficient of the unit (no fouling of the tubes). Consequently, all points on this figure corresponding to tubes in the fouled condition should fall below the line. The scatter of the test points (with some points falling above the line) indicates that the assumption of a constant h.' and Q was not entirely valid. The fouling results obtained from the field test data on this unit are not significant. Therefore, no conclusions can be drawn in this case. The computed results for this unit are summarized in Appendix J and are useful for comparing the actual performance and load conditions against the design specifications for the unit. For convenience, the average values of the test data and computed results appear at the bottom of the summary sheets in Appendices C and J. The test values are summarized in Table VIII for comparison with specifications. TABLE VIII COMPARISON OF FIELD LOAD CONDITIONS WITH SPECIFICATIONS FOR THE HIGH-PRESSURE GAS COOLER Item Specifications Field Test Average Inlet water Outlet water AT water Inlet hydrocarbon Outlet hydrocarbon AT hydrocarbon Heat duty, Q Water velocity Uo MTD (Uncorrected) MTD (G.R. correctic Ao 85 ~F 110~F 25 OF 155 ~F 100 OF 55 OF 15,790,000 Btu/hr 5.2 ft/sec 56 Btu/hr-ft2-~F 27,3~F )n) 23.2~F 10,850 ft2 81.o4~F 107.28~F 26.24~F 144.67~F 110.01~F 34.66~F 7,720,000 Btu/hr 2.90 ft/sec 27.0 Btu/hr-ft2-~F 32.23~F 27.40~F 8,802 ft2 B. DEBUTANIZER OVERHEAD CONDENSER A summary of the calculated results for the debutanizer overhead condenser is presented in Appendix K. The variation of the calculated overall heat transfer coefficient with days of operation is shown in Fig. 7. -Due to 27

100 90 80 LL 0 C4Q I o o O D 70 60 50 40 I W I > ss~~~l~. l l. II Tu^_ l a==..!!/~~~~~~~l~ V~~~~~~ -I ro _m, < "r -< 0 30 20 10 ~( im 3 -m 3 m::r "' 3,l go s r+ c VI+ fO I 10 20 30 40 50 60 70 80 90 DAYS OF OPERATION Figo 7o Debutanizer overhead condenser; variation of the computed overall heat transfer coefficient with days of operationo

The University of Michigan * Engineering Research Institute several factors which are discussed later in this section, it was impossible to compute any fouling results from the field test data obtained on this unit. A total of 11 test runs were made on this unit from August 2, 1955, to October 21, 1955. At this time it was believed that the unit operated as a nonflooded total condenser. The test data were therefore used to compute a series of fouling resistances, assuming a constant outside coefficient of heat transfer. The results of this analysis showed no fouling trend, with several of the computed fouling resistances being negative. The lack of a fouling trend was at this time attributed to (1) the invalidity of the assumption of a constant outside heat transfer coefficient for the unit, and/or (2) the inaccuracy of the outlet hydrocarbon temperature measurement due to the position of the measuring thermometer (see Section III). Since it was impossible to install thermowells in the hydrocarbon outlet line leaving the exchangers (see Section III), it was decided that sufficiently accurate test data for a fouling analysis could not be obtained on this unit. Therefore, the tests on this unit were discontinued after approximately eighty-four days of operation. A study of the specifications and the final designs was undertaken to ascertain why no fouling trend could be obtained from the field test data. A series of calculations, based on the specification sheet for the debutanizer overhead condenser, were made to determine the condensing range of the hydrocarbon stream for the unit. These calculations are presented in Appendix M. The calculations indicated that, at the design pressure, the condensing range of the hydrocarbon stream would be approximately 25~F. The specification sheet for the unit specifies a 25 temperature drop for the hydrocarbon stream4 Calculations were also made to predict the overall heat transfer coefficient for the unit, assuming that all the tube area was used for condensing (no flooding of the exchanger). These calculations are presented in Appendix N. The predicted overall coefficient using this assumption checked within seven percent of the specified design coefficient. Therefore, since (1) the specification sheet allowed for no subcooling (25~F temperature drop of the hydrocarbon stream), and (2) the computed overall coefficient, assuming no flooding checked reasonably well with the design value, it appeared that the unit was originally designed to operate with no flooding of the tubes. This seemingly confirmed our original thinking on this matter. The average observed temperature drop of the hydrocarbon stream was 57.33~F (see Appendix K). This represented 32.33~F of subcooling if the calculated 25~F condensing range were correct and the hydrocarbon inlet stream had no superheat. This amount of subcooling could not be accounted for in the condenser itself unless the unit was flooded. However, the observed subcooling was assumed to be due to the condensate residence time in the overhead receiver since the measured outlet hydrocarbon temperature was always:slightly higher than the ambient-air temperature at the time of the test. It was impossible to check this assumption since the actual hydrocarbon outlet temperature from the condensers could not be experimentally measured. This was one 29

The University of Michigan * Engineering Research Institute of the more important reasons why thermowells in the outlet hydrocarbon lines leaving the exchangers would have been very valuable. In order to determine more accurately the actual condensing range of the hydrocarbon stream entering the unit, stream analysis data giving the composition of samples of the hydrocarbon stream were obtained from the Aurora Gasoline Company. Calculations of the condensing range, based on these compositions, checked closely with the value calculated indirectly from the specification sheets, i.e., between 20~ and 25~F in all cases. In industrial applications involving an overhead condenser operating off a fractionating tower, some method of control is necessary to prevent the condenser from affecting the operating pressure of the tower. The two most commonly used methods of control are (1) throttling the vapor line between the tower and the condenser and (2) flooding the condenser to the point where only sufficient surface is exposed to vapor condensation to give the desired total condensation. The first of these methods permits a fouling investigation to be made on the condenser. With the second case the actual operational heat transfer area used for condensing is unknown and is variable. Due to the extreme scattering of the test data and the large apparent amount of subcooling, it seemed likely that, although thire unit had apparently not been specifically designed for operating in a partially flooded condition, the control system used by the Aurora Gasoline Company might have resulted in partial flooding of the unit. Subsequent discussion with the Aurora Gasoline Company confirmed that the control system did involve partial flooding of both exchangers. Since the operational condensing area of this flooded condenser cannot be determined, no meaningful fouling results could be computed from the field test data. The overall coefficients (Uo) plotted in Fig. 7 are based on terminaJ temperatures, the total exchanger area, and an uncorrected log-mean temperature difference. This method of analysis would yield values of Uo which would normally correspond to the design value of UO if the exchanger were not flooded. The computed overall coefficients in Fig- 7 do not represent either the condensing coefficient or the subcooling coefficient and serve only to demonstrate the scattering of the results computed in the above manner. Some of the factors causing errors in the analysis are 1. the heat transfer area of the exchanger, when flooded, is distributed between condensing and subcooling in an unknown manner, 2. the log-mean temperature difference computed using the terminal temperatures is not applicable when the area is split between condensing and subcooling, and 5. the log-mean temperature difference for a nonflooded condenser, in which case the terminal temperatures may be used in computing the correct 350c

The University of Michigan * Engineering Research Institute temperature difference, must be corrected according to the equation shown in Appendix 0. For convenience, the summaries of field test data (Appendix D) and computed results (Appendix K) have been averaged, with the average appearing at the bottom of the respective appendices. These averages are summarized in Table IX for comparison with the specifications for the unit. TABLE IX COMPARISON OF FIELD LOAD CONDITIONS WITH SPECIFICATIONS FOR THE DEBUTANIZER OVERHEAD CONDENSER Item Secifications Field Test Average Inlet water 85 ~F 86.00~F Outlet water 110~F 103.710F ATwater 250F 17.71~F Inlet hydrocarbon 1355F 150.600F Outlet hydrocarbon 110~F 93.28~F AThydrocarbon 25OF 57-32~F Heat duty, Q 21,360,000 Btu/hr 9,830,000 Btu/hr Water velocity 7 ft/sec 2.61 ft/sec A0 11,740 ft2 9,570 ft2 C. DEBUTANIZER BOTTOMS COOLER A summary of the calculated results for the debutanizer bottoms cooler is presented in Appendix L. The variation of the calculated fouling resistance with the number of days of operation is shown in Fig. 8. The Wilson Plots, from which the fouling resistance was determined during the later part of the test period, are shown in Fig. 9. For discussion purposes, Fig. 8 will be divided into two sectionsan initial period of 84 days (from August 2 to October 21, 1955) and a second period of 199 days (from November 8, 1955, to May 26, 1956) which occurred after the tubes had been cleaned on the inside during a plant shutdown. The division separates the data taken with an older catalytic cracking unit on stream (the initial period) from the data taken with a new catalytic cracker on stream. The computed fouling resistances for the first period show a reasonable trend with the operating time, as indicated by the curve shown in Fig. 8. The fouling rate indicated by the data for this period was considerably greater than that normally encountered in fouling studies.8 The unit apparently reached the design fouling condition after about 70 days of operation, with the amount of fouling continuing to rise at a rapid rate. The rapid fouling 31

+.016 +.014 +.012 +:D m (U ItLl i L. 4- ~ o -- <1< — + + I I I I Oil contamination of cooling water began - 002 0 10 20 30 40 50 60 70 80 90 0 I I I I I I I I I 80 9 I I 2I I I I 6 I 10 20 30 40 50 60 70 80 90 100 110 120 BO0 140 150 160 1-70 180 190 200 DAYS OF OPERATION Fig. 8. Debutanizer bottoms cooler; variation of the computed fouling resistance with days of operation. 32

The University of Michigan * Engineering Research Institute rate during this period was attributed to the high inlet hydrocarbon temperature, which ranged from 50~ to 150~F above the design value. Such a high inlet hydrocarbon temperature causes a high tube-wall temperature, which in turn normally increases the precipitation rate of the hardness salts present in the cooling water. It seems apparent that the majority of the fouling accumulated during the first period was on the inside (water side) of the tubes; this fact is shown by the large decrease observed in the fouling resistance when the tubes were cleaned on the inside only. The large outside-to-inside area ratio of the finned tubes used in the exchanger tends to magnify the effect of fouling on the inside of the tubes. The plot showing the second series of test data, after the tubes had been cleaned on the inside, has no data points within the first 10 days of operation. This is due to the fact that the project personnel were not informed that the unit had been cleaned and placed back on stream until nine days after the cleaning. Thus, the exact amount of cleaning could not be determined. The first 70 days of the second period showed a low fouling rate, with the fouling remaining well below the design value. The hydrocarbon inlet temperature during this time was considerably lower than during the first period, falling from an average of 356~F during the first period to less than 200'F for the early part of the second period. This tends to corroborate the theory that the high fouling rate during the first period was due to an excessively high inlet hydrocarbon temperature. The first Wilson Plot runs on the unit were taken on December 23, 1955. After this date, all data taken were of this type. The fouling resistances plotted in Fig. 8 may be obtained by subtracting.o00038 (the sum of rf and As rm) from the corresponding day's intercept value in Fig. 9. Am After approximately 70 days of operation, the fouling resistance of the unit was observed to rise sharply. This trend was first noticed on February 1, 1956, and was confirmed on February 7, 1956. Subsequent conversations with Aurora Gasoline Company plant engineers established that an oil leak had developed somewhere in the plant which was contaminating the recirculated cooling water with oil. Since the observed fouling was considerably higher than the design value, it was decided to attempt to clean the unit while leaving it on stream. By arrangement with the plant, project personnel were present to take data just before and immediately after the cleaning. The results of the cleaning are shown in Fig. 8 (at 116 days of operation). The cleaning succeeded in decreasing the fouling to below the design value. However, on the next occasion of obtaining test data (one week later) the fouling had again exceeded the design condition. The remainder of the fouling-trend curve (from 123 to 200 days of operation) shows the results of the Aurora Gasoline Company's attempt to re move all oil contamination from the cooling water. Without actual cleaning 34

r The University of Michigan * Engineering Research Institute! of the unit, the decrease in the amount of oil contamination of the water caused the observed fouling to decrease. Thus, the unit effectively seemed to clean itself during this period. For convenience, the summaries of field test data (Appendix E) and computed results (Appendix L) have been averaged, with the average appearing at the bottom of the respective appendices. These averages are summarized in Table X for comparison with the specifications for the unit. TABLE X COMPARISON OF FIELD LOAD CONDITIONS WITH SPECIFICATIONS FOR THE DEBUTANIZER BOTTOMS COOLER Item Specifications Field Test Average Inlet water Outlet water ATwater Inlet hydrocarbon Outlet hydrocarbon *AThydrocarbon Heat duty, Q Water velocity, Vt Overall heat transfer coefficient, Uo MTD (ATLM) Ao 85 ~F 115 F 30OF 230~F 100 F 130~F 9,790,000 Btu/hr 6 ft/sec 67 Btu/h r-ft2- F 48 F 3050 ft2 75.75 ~F 108.87~F 33.12~F 253370~F 91.070F 142.653F 5,780,000 Btu/hr 3.29 ft/sec 45.65 Btu/hr-ft2 -F 49.62~F 2505 ft2 VII. CONCLUSIONS No significant conclusions could be drawn concerning the fouling rates of either the high-pressure gas cooler or the debutanizer overhead condenser. In the case of the high-pressure gas cooler the unit was operated as a partial condenser with a liquid-vapor feed containing noncondensables. It was not possible to separate the fouling resistance from the film resistances. The debutanizer overhead condenser was operated in a partially flooded manner. It was not possible to determine the fouling resistance because the extent of flooding was unknown. Significant conclusions can be drawn from the fouling test data obtained on the debutanizer bottoms cooler. These are: (a) if the temperature of the hydrocarbon stream is maintained at the design level, the total fouling can be expected not to exceed the design level over a reasonable length of time, (b) the tube-side fouling (water side) greatly affects the overall foul J 35 I

a The University of Michigan T Engineering Research Institute ing, and (c) cleaning of the tube side restores the unit to below the design fouling level. Discrepancies exist between the specified heat transfer areas and the actual heat transfer areas provided in all three units. The units have about 81% of the specified heat transfer area. In general, the three units were not operating with the design flow rates. 56

- The University of Michigan * Engineering Research Institute REFERENCES 1. Carrier, W. H., and Anderson, S. W., "The Resistance to Heat Flow Through Finned Tubing," Heating, Piping, and Air Conditioning, 16:304-318 (1944). 2. Gardner, K. A., "Variable Heat-Transfer Rate Correction in Multipass Exchangers,Shell-Side Film Controlling," Trans. ASME, 67:31-38 (1945). 3. McAdams, W. H., Heat Transmission. 3rd Edition. New York: McGraw-Hill Book Co. Inc., 1954. 4. Williams, R. B., and Katz, D. L., "Performance of Finned Tubes in Shell and Tube Heat Exchangers'" Engineering Research Institute, University of Michigan, 1951. 5. Brown, G. G. et al., Unit Operations. New York: John Wiley and Sons, Inc., 1950, pp. 157-160. 6. Maxwell, J. B., Data Book on Hydrocarbons. New York: D. Van Nostrand Co., Inc., 1950. 7. NGSMA Data Book, Natural Gasoline Supply Men's Association, 1951. 8. "Fouling Rates for Fuel Oil Heat Exchangers with Plain and Finned Tube Bundles," Engineering Research Institute, University of Michigan, Project M592, July 1953. 9. Perry, J. H., Chemical Engineer's Handbook. New York: McGraw-Hill Book Co., Inc., 1950. p - - 37

The University of Michigan'Engineering Research Institute APPENDIX A THE GRISCOM-RUSSELL CO. HEAT EXCHANGER SPECIFICATIONS 38

PECIFICATION ^THE GRISCOM'-RUSSELL CO. SHBET No. I HEAT EXCHANGER SPECIFICATIONS G-R No. ^1 j` ~Job No. 82 5-A| 2 Customer AURORA C1ASOLIrE1 COrAPANY Reference No. 3 j Address Detroit, MTTichi -an,. Inquiry No. 4 Plant Location Gas Concontration Ulit Date 11-11-_Q4 5 Service of Unit Iih?rCS sur e Co00or Item No. E.2 6 Size and Type YTi$2) #33<K24192 PSA ______________ 7 -' Horiz. - Vert: Stat. Hd. 8 Surface per Unit 10, ( 50 sq ft Shells per Unit 2;Seri 2; Parallel Surface per She 9 i PERFORMANCE OF ONE UNIT Shell Side Tube Side 10 Fluid Circ slated W- er 11 Total Fluid Entering _tQra j' - lb per hr - Q lb ht'h 12 _ Vapor Jl09A.- -,-. 2300- lb per hr! __Ib pe 13 Liquid l7b':I-'2 _ _2QQ___.lb per hr. i_ _ b per hr 14.....stearr _ __ _.........._] -. - -.-70 ~_..... _b._. h!.. —..............................P., 1 4 -Steamr -- 700 lb per hr t lb per hr 15 Nor.Condensabtles __ 70 lb per hr I lb per hr 16 Fluid Vapbrized or Condensed __l 29 lb per hr_. lb_ per hr_ 17 Steam Condensed _7f0 __ lb per hr lb per hr 18 Gravity -Liquid-_____ _____ 19 _Density _@ Ave. Cond. lb per cu ft.... ____lb per cu ft 20 Viscosity I Centipoises at O F ~ Centipoises at OF 21 Viscosity.....____ _ Q-3 Centipoises at12__- 0F I Centipoises at ___ ~F 22 M1 olecular Wt. __......__...___ -__-.______ 23 I Conductivity.........______ 24 Specific Heat ___ Btu per lb Btu per lb 25 iLatent Heat - Vapors Btu per lbI Btu per lb 26 Temperature In _ ___F OF 27 Temperature' Out OF 1 -.. l. -- 28 Operating Pressure i _ psig 7 pg 29'Number of Passes. t t........_..t..i..l....... - --.. —30 Velocity -! - fps _.- - 2 fps 31 Pressure Drop_______ ________ _ps psi 32 Heat Exchanged - Btu per hr _1, 790oQQC - MTD (Corrected) - _ —— ___.. -- - 33; Tranlsftr Rate - Service C Clean _ _ Fouling Resistance: Shells 008 34' CONSTRUCTION 35 Design Pressure _psg - 7 psig 36 TestPressure. —L.. Ig... 1 3. psig 37 Design Temp. (Max. Metal Temp.) O F',' 38 T ubces NA t - -* ^O e o.D. ^/pf BWVG 3_ Length J _Pitch 1)2t rcu 39 Shell 33 T.D. Mat'l..- Fins AJj^j.jt -- 40 Shell Cover.-te l...._.._.......... e......d Co -ver 41 Channel S Cot har -el C over 42 Tube Sheets-Stationary _.__ _...t........-...- Floating 43 Baffles-Cross S._t C -l... Typ_.':.e..nt.._ pacing 2.. Thickness- 3 " ole Dia. 44 Baffle-Long Type. Thickness- Impact Baflet Se' 1 45 Tube Supports FLma T _ _. __mi.. No. - 1 Thick-ness / - - 46 Gaskets As R_ ic.'. ed - lan..lacking ____ - 47 Connections-Shell - In 12" ____ut___.2 -_.....1 -____ Dome Yes 48 -- - Channel-In 81? ___ Out 8" - __Series __ _______ _ 49 Corrosion Allowance-Shell, Side 1/, Tube Side.l/~f, _ 50 Weight 18 000 Code 952.AS}1 Customer's Specifications nTej rp -' ThUro l Tema Class - 51 Length _______ _ —Width _Height. _... 52 Ref. Print _____________Inspection by whom 53 Based upon operating conditions as given, maximum temperature of tubes will not exceedl~F OF 54 Note: Indicate after each part whether stress relieved (S.R.) and whether radiographed (X-R) __ t55'Remsarksm: """""'"""""I S. S. No. LITW 59~_.OTL'l2NCX * Total Press.ure Drop for Two Shells * Wolvcrine Trufin. 39

SPECIFICATION THE GRISCOM-RUSSELL CO. SHEET No. 1 HEAT EXCHANGER SPECIFICATIONS G-R No..1 Job No. No 3-E 2 Customer r'.. f..C T COe l''.' Reference No. 3 Address t:O r! it,.':"-':'. Inquiry No. 4 Plant Location c.r:' - T-,' _ Date 111!45 Service of Unit'C:,U -'':;j'-5 —: __ ___, Item No. _._ _______ 6 Size and Type TLWO, ( ) 2i21^ -22~: dQ l *' __. _ __ 7 __ Horiz. - Vama. Stat. l4d; 8 |Surface per Unit 11s7^0 q ft Shells pe Unit 2;Series Parallel 2; Surface per Shell'"70 Iq ft 9 PERFORMANCE OF ONE UNIT Shell Side Tube Side 10 Fluid Circulated _ ________.''.. 11 i Total Fluid Entering )..L' lb per hr __ lb per hr 12 _ Vapor _______ _ lb per hrt _ lb per hr 13 _ Liquid _ _..... lb_ per hr: - lb per hr 14 1 Steam _ __ _ _. - - _ lb per hr __ ___ lb per hr 15 " Non-Condensables lb per hr lb per hr I) 6 Fluid Ut~'ia6d. or Conden.ed i lb per hr lb per hr 17 Steam Condensed _ lb per hr' _b per hr 18 iGravity - Liquid 121 A1-T__ 19 Density _.@ Ave. Cond. l_ _i _b per cu ft lb per cu ft 20 Viscosity _ __ _ Centipoises at 3 _ ____ Centipolses at_ 1 21 I Viscosity Centipoises at.F Centipoises at ____ F 22 I_MolecularWt.__ _____.__.. ___ 23 onductivity.07.'' 24 Specific Heat __ - - Btu- per lb _. -tu per lb 25 Latent Heat - Vapors } Btu per lb_: Btu per lb 26 Temperature In __ __:,. "F' ____ ~F -— 6 -:-?- -- - -.. ___...........'F 27 Temperature' Out - Fi -...-. 28 Operating Pressure __psig I pig 29'Number of Passes _-_____ 30 Velocity.. fp. __ _7 fps.31 i Prcssure Drop - Ii p____ __ __' _____ P5 ____1__ si 32 H eat Exchanged - Btu per hr M' MTD (Corrected).. 33 Transfer Rate - Service' Clean _Fouling Resistance: Shell Tubes..~ V: _ 34 CONSTRUCTION _ 35 Design Pressure psig j psig 36: Tefst Pressure ps3ig pspig 37' Design Temp. (Max. Metal Temp.)..:...':'F -~_F 38 Tubes_:. Y - No.,r,'- O. D. / BWG' L engith 7 c2 Pitch. -.__ 38:~.T~ubes__ _.:........-.... -_-_39 Shell I' I.D.; Mat'l "-'. l'C''i-e Fins. - 40 Shell Cover (Floatirv Heed Cover 41 Channel _ t: e.e Channiel Cover' - 42 Tube Sheets-Stationary'' Floating __ _. t_____ 43 Baffles-Cross:O. _ _ Type _ _ Spacing Thickness Hole Dia. ____ 44 Baffle-Long Type____..... Thckness.Impact Baffle t 45 Tube Supports Sl 1.c ____._. -No.7__...T1^! -Th 3ck____s___ _s _sj____-__ 46 i Gaskets__ -^-^^C y:)e.A_....-..... -- Glan Packing_f___.- ____ ____~__ _ 47 Connections-Shell - In O: ut __:. O ut_ Di k -:-n omes s.. - 48 __C hannel-In___orts___...........2Out _ S.eries........_ 49 _ _Corrosion Allowance-_Shell Side n__Tube Side /a" n_ 50 Weight 17,300 Codel? 1 -T? Customer's Specifications O 0''1. 7,' Tema Class 51 Length W__ _______________idth__ Heig^ht_. _____ 52 Ref. Print Inspection by whom 53 Based upon operating conditions as given, maximum temperature of tubes will not exceed'F 54 Note: Indicate after each part whether stress relieved (S.R.) and whether radiographed (X-R) 55 Rcmarks: i S. S. No...T..'C7J- —... * u'rC2~u"e dron- to <. for two- sh1zlS, ir,pa1rallel performnance, deI9; n arnd'-ie-"it for one l c11 onl T W olverine Tr.ifin to be used 40

SPECIFICATION SHEET No. 1 THE GRISCOM-KUSELL CO. HEAT EXCHANGER SPECIFICATIONS G-R No.! 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24'?5 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 66 Job No._ 3G_ — Customer ~" iR 0.A 3 ",Tt C.', Reference No. Address DPT'r:. T T T?TCTC:T. I C:i I:_, _ Inquiry No. Plant Location GAS C' Ti NIT Date 17. 1j Service of Unit _,_ _ r_'7.'m%,'.... *71 Item No.' I rl_ Size and Type L. I FTA __.._A...... Horiz - r,.ti. Jj.I'I Surface per Unh;QQ sq ft Sheler Unr Unit;. Series 1; Parallel; Surface per Shell 3fq, sq ft PERFORMANCE OF ONE UNIT Shell Side Tube Side Fluid Circulated __l.'r "~.'._,- _... 4 - Total Fluid Entering _ _lb per hr l-__ b per hr Vapor lb per hr_ b perhr Liquid _____ lb per hr I lb per hr _Steam l lb per hr lb per hr __Non-Condensables _ _ _ _ lb per hr lb per hr Fluid Vaporized or Condensed lb per hr_ lb per hr'Steam Condensed lb per hr ______________ lb per hr Gravity- Liquid.1. 7.-, r _.T Density (a! Ave. Cond. lb per cu ft lb per cu ft Viscosity. Centipoises at / F -Centipoises at. F i Viscosity._. Centipoises at 0 F Centipises at Molecular Wt _._ Conductivity - - _- - - Specific Heat _ _Btu per lb Btu per lb Latent Heat - Vapors _ Btu per lb Btu per lb Temperature In ~F F Temperature Out i.F -_.F- I Operating Pressure psig ____. psig Number of Passes_........ Velocity_ - - _ __ fps.... ___fps Pressure Drop.0 psi " _ nI psi Heat Exchanged - Btu per hr r,'r -.,., MTD (Corrected) - - Transfer Rate - Service / Clean Fouling Resistance: Shell ~,- Tubes CONSTRUCTION Design Pressure _. _ psig P-ig Test Pressure psig. psig Design Temp. (Max. Metal Temp.) -.'3 -"F Le h PcF Tubes...... a^... BN.3WG. Length ]_. Pitch.,ae Shell 2_ ID.; Mat' Tube Finsi Shell Cover _-.... - _.Floating Head Cover.-. anl__.....-. Channel Channel Cover_. Tube Sheets-Stationary Fl c _C._ _... _._-_._ -.. F "loating, -. - -... Baffles-Cross _,- L-_ Type. Spacing poT Thickness )/If" Hole Dia. Baffle-Long._- t C C - __ Type _laj - Thickness /142 Impact Baffle c Tube Supports -- __ teNo.' Thickness Gass ^eq-uired - - - G —-- and__acking — -.- - - a c. — -_ Connections-Shell- In _ Out r Series Dome t Channel- In t Out Series....-... T b___ i.de.......__21_ __ __ __ __ __ Corrosion Allowance-Shell Side Tube Side __ Weight fl(n Code 1 AT: Customer's Specifications C.'~ 7 Tema Class Length Width _ _Height Ref. Print. Inspection by whom _. Based upon operating conditions as given, maximum temperature of tubes will not exceed ~F INote: Indicate after each part whether stress relieved (S.R.) and whether radiographed (X-R). Remarks: LS. S..No i20D.GyC-X-. -..... _S_ - yN_0 ~.~.00~O~X......,-.olver-~.no rei a 41

I The University of Michigan Engineering Research Institute APPENDIX B GRISCOM-RUSSELL DESIGN BLUEPRINTS - 42

L I —-- The University of Michigan T Engineering Research Institute APPENDIX C SUMMARY OF TEST DATA ON THE HIGH-PRESSURE GAS COOLER

Date Cooling Water Hydrocarbon Manometer No. of Inlet Intermediate Outlet Inlet Outlet Left Right F1eter Remarks Run OF ~OF~F ~F F Inches Inches Fl 1 Aug. 2, 1955 93.88 98.70 109.70 151.55 108.60 +2.26 -2.16 Hg 2 Aug. 2, 1955 94.46 98.92 109.65 154.75 109.90 +2.20 -2.10 Hg 3 Aug. 5, 1955 93.49 98.39 111.62 167.56 110.46 +2.02 -2.22 Hg 4 Aug. 8, 1955 84.18 88.60 110.91 149.86 101.52 +0.78 -3.69 Hg 5 Aug.13, 1955 88.01 91.17 99.20 145.39 102.68 +2.40 -1.88 Hg 6 Aug.17, 1955 90.82 96.52 107.82 127.18 110.24 +1.73 -1.23 Hg 7 Aug.20, 1955 95.77 102.62 115.03 131.50 117.44 +1.34 -0.84 Hg 8 Sept. 7, 1955 82.57 93.06 108.48 152.46 112.66 -2.75 -3.75 Hg 9 Sept. 20, 1955 83.75 92.17 108.26 139.11 115.88 -2.30 -4.20 Hg 10 Oct. 14, 1955 75.15 80.88 91.46 105.75 106.19 -1.06 +0.93 Hg 11 Dec. 10, 1955 71.64 - 98.-70 132.34 96.64 +0.10 -0.20 Hg Manometer readings questionable. 12 Dec. 17, 1955 70.89 82.88 101.56 140.40 105.65 +0.55 -2.45 Hg 13 Dec. 23, 1955 73.19 83.08 99.80 141.23 108.57 +22.0 -19.5 Hg Manometer reading in inches 14 Dec. 29, 1955 73.37 87.90 109.80 163.77 109.03 +1.35 -1.71 Hg of water. 15 Jan. 7, 1956 71.05 82.35 102.78 139.31 102.05 +1.53 -1.93 Hg 16 Jan. 13, 1956 69.78 81.62 102.55 144.83 100.13 -1.15 -4.13 Hg 17 Feb. 1, 1956 67.66 79.46 99-36 158.52 108.34 -1.10 -4.06 Hg 18 Mar. 3, 1956 78.32 90.50 111.48 167.70 122.94 +1.40 -1.40 Hg 19 April 11, 1956 82.80 100.00 122:20 150.40 124.40 +0.57 -0.88 Hg 20 May 26, 1956 70.59 90.83 116.66 137.38 113.43 +13.31 -12.63 CC14 Manometer Temperature = 68~F. Average 81.04 90.51 107.28 144.67 110.01 (Omitting Run No. 11) -14 _' r_ 3 fO w go r) ro Mo

The University of Michigan Engineering Research Institute APPENDIX D SUMMARY OF TEST DATA ON THE DEBUTANIZER OVERHEAD CONDENSER 51

m, -I DEBUTANIZER OVERHEAD CONDENSER O Date ooling ater Hydrocarbon Manometer Run Ref. B Outlet After Left Right Run of Inlet Outlet A Outlet B Outlet Mixed Inlet T.C. A Rcf. A T.C. B Ref. B Outlet After Left Right Manometer No Run ~F ~F ~F Mean, OF ~F mv ~F my F Tank, ~F Inches Inches 1 Aug. 2, 1955 94.03 109.95 105.69 107.20 150.90 +0.025 95.55 +0.027 96355 100.50 -5.05 +5.07 Hg 2 Aug. 5, 1955 95.78 111.95 104.79 107.48 155.62 +0.041 90.80 +0.059 90.80 96.11 -4.28 +4.28 Hg 5 Aug. 8, 1955 84.53 97.89 94.01 96.55 154.45 -87.10 -4.61 +4.61 Hg* 4 Aug. 15, 1955 86.61 107.02 105.17 108.40 151.95-3 88.57 -1.14 +1.05 Hg 5 Aug. 17, 1955 90.96 115.89 115.40 115.69 155.73 +0.114 92.00 +0.158 91.05 95.22 -0.95 +0.95 Hg 6 Aug. 20, 1955 94.52 118.17 119.47 117.45 154.56 +0.034 98.50 +0.027 97.70 98.91 -1.00 +1.00 Hg 7 Aug. 31, 955 84.18 111.03 113.34 110.48 158.53 +0.115 85. +0.240 84.00 90.04 -1.09 +1.09 Hg 8 Sept. 7, 1955 83.51 102.85 101.36 99.25 152.54 85.70 -4.42 +4.42 Hg 9 Sept. 20, 1955 83.14 99.05 99.17 97.25 152.32 - - - 87.35 -3.78 +3.78 Hg 10 Oct. 14, 1955 75.57 95.55 95.20 95.29 144.49 +1.170 64.87 +1.070 67.50 100.10 -2.77 +2.68 Hg 11 Oct. 21, 1955 75.78 86.05 86.69 87.24 125.55 +1.350 56.25 +1.560 57.00 96.43 -5.76 -0.73 Hg Average 86.00 104.85 105.48 105.71 150.60 95.28 0 * Manometer reading in inches of water gage. m 3 m go -I 5% m~

The University of Michigan E Engineering Research Institute I APPENDIX E SUMMARY OF TEST DATA FOR THE DEBUTANIZER BOTTOMS COOLER 55

1 The University of Michigan * Engineering Research Institute I —— TeUiest fMcia Date Cooling Water Hydrocarbon Manometer of Inlet Outlet Inlet Outlet Left Right Manometer Remarks No. Run ~F ~F ~F ~F Inches Inches Fluid 1 Aug. 2, 1955 2 Aug. 5, 1955 3 Aug. 8, 1955 4 Aug. 13, 1955 5 Aug. 17, 1955 6 Aug. 20, 1955 7 Aug. 31, 1955 8 Sept. 7, 1955 9 Sept. 20, 1955 10 Sept. 20, 1955 11 Oct. 7, 1955 12 Oct. 14, 1955 13 Oct. 21, 1955 Average of Runs 1-13 94.17 94.29 84.79 86.64 90.85 95.98 83.75 82.94 83.20 83.35 85.10 76.13 75.77 127.29 124.10 117.15 118.88 131.73 131.40 120.43 123.35 124.95 124.19 125.24 107.21 102.60 371.10 97.98 367.64 97.38 371.53 88.88 370.45 91.35 379.27 96.90 365.40 99.38 334.07 90.55 328.11 93.62 368.30 100.60 359.03 100.74 337.27 103.82 356.00 89.37 317.95 95.07 +29.30 +26.70 +26.80 + 3.51 +2.12 +2.07 +1.96 +1.65 +1.80 +1.80 -1.77 -1.99 -1.83 -22.75 H20 -19.70 H20 -20.80'HO -2.85 Hg -1.70 Hg -1.57 Hg -1.47 Hg -1.18 Hg -1.30 Hg -1.30 Hg -4.76 Hg -4.65 Hg -4.78 Hg 85.77 121.42 355.86 95.82 Tubes Cleaned on the Inside Only During Plant Shutdown 14 Nov. 18, 1955 15 Nov. 28, 1955 16 Dec. 10, 1955 17 Dec. 17, 1955 18 Dec. 23, 1955 19 Dec. 23, 1955 20 Dec. 23, 1955 21 Dec. 29, 1955 22 Dec. 29, 1955 23 Dec. 29, 1955 24 Jan. 7, 1956 25 Jan. 7, 1956 26 Jan. 7, 1956 27 Jan. 13, 1956 28 Jan. 13, 1956 29 Jan. 13, 1956 30 Jan. 13, 1956 31 Jan. 13, 1956 32 Feb. 1, 1956 33 Feb. 1, 1956 34 Feb. 1, 1956 35 Feb. 1, 1956 36 Feb. 1, 1956 3 F 7, 9) 5 6 38 Feb. 7, 1956 39 Feb. 7, 1956 40 Feb. 7, 1956 41 Feb. 7, 1956 42 Feb. 7, 1956 43 Feb. 18, 1956 44 Feb. 18, 1956 45 Feb. 18, 1956 46 Feb. 18, 1956 47 Feb. 18, 1956 48 Feb. 18, 1956 49 March 3, 1956 50 March 3, 1956 75.96 69.19 70.99 73.02 73.50 73.42 73.54 74.08 73.88 70.72 70.44 70.52 70.37 70.38 70.40 70.46 70.50 67.36 66.88 66.61 67.38 67.46 72.20 73.08 73.00 72.78 72.53 72.12 72.18 72.44 72.20 72.16 72.02 78.50 78.48 120.64 121.72 101.29 100.72 91.68 104.24 106.46 95.44 124.80 99.04 114.50 90.58 86.68 83.18 88.89 94.12 108.86 84.96 89.02 115.56 85.72 82.48 G4.6G 87.88 89.54 95.44 114.90 119.48 89.70 93.74 95.00 104.02 111.10 124.79 98.26 104.56 277.00 93.01 283.09 99.17 190.33 85.85 188.00 83.70 189.80 79.76 186.40 82.92 19(.20 91.92 196.60 83.10 196.40 94.28 181.00 84.30 181.00 91.02 184.00 78.94 177.40 75.14 176.40 74.30 177.14 75.70 177.20 77.02 176.00 82.24 177.60 78.78 174.60 78.16 176.43 92.18 190.40 82.12 178.60 77.28 1.60 79.;1Z 187.00 80.68 185.80 81.16 186.80 83.68 187.20 91.42 184.50 92.98 197.80 84.88 198.00 86.70 197.60 85.94 199.20 90.94 198.60 91.34 194.30 98.47 213.00 94.44 213.60 97.34 +1.17 +1.15 +0.61 +10.00 +22.20 +6.00 +0.49 +1.51 +0.12 +13.93 +4.85 +1.66 -1.43 -0.90 +17.85 +11.34 +4.09 +11.00 +6.05 +2.13 -1.35 -1.80 +1.00 +16.20 +9.66 +3.10 +2.26 +1.86 +1.37 +15.57 +9.12 +4.21 +1.93 -1.33 -0.74 -1.42 Hg -1.27 Hg -- - Void (Incorrect manometer reading). -0.75 Hg -9.00 H0 -21.00 H20 -5.00 H20 -0.78 Hg -1.87 Hg -0.39 Hg -13.58 CC14 Manometer temperature ~40OF. -4.72 CC14 Manometer temperature -40~F. -1.82 Hg -3.83 Hg -4.46 Hg -17.36 CC14 Manometer temperature 40~F. -11.01 CC14 Manometer temperature.40~F. -3.79 CC14 Manometer temperaturev 40~F. -10.92 CC14 Manometer temperature. 40~F. -5.76 CC14 Manometer temperature v 40F. -2.04 CC14 Manometer temperature -40~F. -3.80 Hg -3.30 -0.95 Hg -15.52 CC14 Manometer temperature 45~F. -9.06 CC14 Manometer temperature- 45~F. -2.50 CC14 Manometer temperature -45 F. -1.60 CC14 Manometer temperature 45~F. -1.05 g -0.57 Hg -15.59 CC14 Manometer temperature -40~F. -9.02 CC14 Manometer temperature ~40~F. -4.21 CC14 Manometer temperature ~40OF. -2.00 CC14 Manometer temperature -40OF. +1.56 Hg -0.76 Hg I Tubes Partially Cleaned on Inside Only 51 March 3, 1956 52 March 3, 1956 53 March 3, 1956 54 March 3, 1956 55 March 3, 1956 56 March 10, 1956 57 March 10, 1956 58 March 10, 1956 59 March 10, 1956 60 March 10, 1956 61 March 10, 1956 62 March 10, 1956 63 April 21, 1956 64 April 21, 1956 65 April 21, 1956 66 April 21, 1956 67 May 26, 1956 68 May 26, 1956 69 May 26, 1956 70 May 26, 1956 71 May 26, 1956 Average of Runs 14-71 (Omitting Run No. 16) Average of Runs 1-71 (Omitting Run No. 16) 79.08 79.46 79.16 79.04 78.60 70.60 72.08 72.70 72.16 72.01 72.04 72.00 84.00 82.00 81.80 82.20 73.98 74.00 74.76 76.50 76.52 101.00 107.14 115.38 128.20 142.88 92.60 93.86 101.32 113.10 144.90 134.40 107.94 110.20 111.00 123.20 158.60 92.48 97.38 105.58 118.78 142.96 217.00 92.26 245.00 92.20 245.00 98.32 215.20 105.18 214.40 110.00 220.00 85.74 220.60 88.86 223.00 91.44 219.00 94.70 220.50 101.62 219.20 105.74 219.20 93.02 254.00 96.20 254.20 98.20 253.40 102.00 252.60 114.60 215.40 85.36 215.00 87.88 216.60 91.32 217.00 98.94 217.00 111.72 +1.64 +1.06 +16.68 +5.60 +2.79 +1.35 +1.60 +16.75 -8.83 +3.84 +2.60 +8.84 +1.30 +1.31 +14.70 +3.68 -4.40 -5.50 -6.45 +11.42 +3.66 -1.64 Hg -1.06 Hg -16.08 CC14 Manometer temperature -45~F. -5.01 CC14 Manometer temperature - 45 F. -2.30 CC14 Manometer temperature -45~F. -1.15 Hg -1.55 Hg -17.96 CC14 Manometer temperature- 50OF. -10.28 CC14 Manometer temperature ~50OF. -5.31 CC14 Manometer temperature -50~F. - Bubble in manometer line. -4.15 CC14 Manometer temperature 50~F. -10.82 CC14 Manometer temperature -50~F. -1.60 Hg -1.55 Hg -13.92 CC14 Manometer temperature - 70~F. -3.14 CC14 Manometer temperature -70~F. -10.62 Hg -9.50 Hg -8.55 Hg -11.08 CC14 Manometer temperature ~75~F. -3.48 CC14 Manometer temperature -75~F. i 73.47 106.01 205.86 89.99 75.75 108.87 233.70 91.07 54

...... The University of Michigan T Engineering Research Institute APPENDIX F EXAMPLE CALCULATION OF TEST DATA ON THE HIGH-PRESSURE GAS COOLER Date of Run: May 26, 1956 Run No. 20 From the data (-AP") = 13.31 - (-12.63) = 25.94 inches of CC14 Manometer temperature = 68~F. Substituting in Equation 9b, (P') - (-AlP") (-aPt) 21 25.94 21 = 1.235 inches Hg. From Equation 8, w = 178,000oo (-A')' = 178,000 /1.235 = 198,000 lb/hr. From the data; the water temperatures were tH20 out = 116.66~F, tH20 in = 70.590~F ~ AtHO0 46 07~F 46= 46.07 and tw = 70.59 + 7 = 93.62 ~F. Substituting in Equation 4, Q = W Cp AtH20 = (198,ooo)(1)(46.o7) = 9.11 x 106 Btu/hr. 55

The University of Michigan From Equation 11, * Engineering Research Institute Vt = 1035,000 198,000 103,000 =1.92 ft/sec. From the data, the hydrocarbon temperatures were thydrocarbon in = 137.38~F thydrocarbon out = 113.43~F Athydrocarbon = 23.95 F. Computing ATLy AT1 AT2 ATLM I, = 137.38 - 116.66 = 20.72 ~F = 113.43 - 70.59 = 42.84~F AT2 - AT1 ~nA AT1 42.84 - 20.72 42.84 In20,72 = 30.50~F. Rearranging Equation 1, Q Ao ATLM The outside heat transfer area for this unit is (from Table VII) Ao = 9385 ft2. Solving for the overall heat transfer coefficient, 9.11 x 106 (350.o)(9385) = 31.8 Btu/hr-ft2-_F. 56

Then, Then, Substt The University of Michigan * Eng 1 L1 Uo 31.8 = 0.031 From Equation 5, 150[11 + 0.011(t hi = (di)0~2 150[1 + 0.011(9: (O0. = 586 Btu/hr-ft2-' Ao 3-57 Ai hi 586 =0.00{ ituting in Equation 17c, i + rot - Ai Uo Ai;ineering Research Institute L4 hr-ft2-OF/Btu.,)] (Vt)0'8 5.62)] (1.92)0'8 51)o.2 oF 610 hr-ft2-OF/Btu -hi + 0.0382 i hi J = 0.0314 - [0.0061 + 0.0382] =-0.0129 hr-ft2-F/Btu 57

The University of Michigan * Engineering Research Institute APPENDIX G EXAMPLE CALCULATION OF TEST DATA ON THE DEBUTANIZER OVERHEAD CONDENSER Date of Run: Run No. 10 October 14, 1955 From the data, (-AP') = 5.45 inches of Hg. Substituting in Equation 12, W = 288,000 7-Ap = 288,000 N/5.45 = 672,000 lb/hr. From the data, the water temperatures were tH20 out = 95.29~F tH20 in = 75.37~F * AtH20 = 17.920F Substituting in Equation 4, Q = W Cp AtH20 = (672,000) (1)(17.92) = 12.05 x 106 Btu/hr From Equation 13, Vt = 223,000 672,000 225,000 = 3.01 ft/sec 58

The University of Michigan * Engineering Research Institute From the data, the hydrocarbon temperatures were thydrocarbon in = 144.49~F thydrocarbon out (after tank) = 100.10"F ~ Athydrocarbon = 44.39~F Computing ATLM, AT1 = 144.49 - 93.29 = 51.20~F AT2 = 100.10 - 75.37 = 24.73~F AT AT1 - AT ~LM - nAT1 In AT2 51.20 - 24.73 n51.20 24-73 = 36.45~F The outside heat transfer area for this unit is (from Table VII) Ao = 9570 ft2 Solving for the overall heat transfer coefficient, Uo = Ao ATLM 12.05 x 106 (9570)(36.45) = 34.5 Btu/hr-ft2-OF 59

The University of Michigan T Engineering Research Institute APPENDIX H EXAMPLE CALCULATION OF TEST DATA ON THE DEBUTANIZER BOTTOMS COOLER Date of Run: January 13, 1956 Run No. 27 From the data, (-AP') = Substituting in Equation 14, 2.40 inches of Hg w = 144,o000 -Apt = 144,000 /2.40' = 223,500 lb/hr From the data, the water temperatures were tH2o out = 86.68~F tH2O in = 70.37~F:. AtHa2 = 16.31~F 16.31 and tw = 70.37 + 6 = 78.52 F. 2 Substituting in Equation 4, Q = W Cp AtH20 (225,500)(1)(16.51) 3.64 x 106 Btu/hr. From Equation 15, V =, t 588,500 223,500 58,500 = 3.82 ft/sec 6o

The University of Michigan * Engineering Research Institute - From the data, the hydrocarbon temperatures were hydrocarbon in = 177.40~F thydrocarbon out = 75.14~F A Athydrocarbon = 102.260F Computing ATLM, AT1 = 177.40 - 86.68 = 90.72~F AT2 = 75.14 - 70'37 = 4.77~F ATLM = AT1 - AT2 AT1 inAT2 90.72 - 4.77 90.72 477 = 29.30~F The outside heat transfer area for this unit is (from Table VII) Ao = 2505 ft2 Solving for the overall heat transfer coefficient, U A A Ao ATLM 3.64 x 10 (2505)(29.30) = 49.6 Btu/hr-ft2-F Then, 1 1 UO 49.6 = 0.02015 hr-ft2-F/Btu From Equation 5, 61

The University of Michigan E Engineering Research Institute (150)[1 + O.011(tw)] (Vt)'8 (di) -2 (150)[1 + 0.011(78.52)] (3.82)0.8 (0.51)o*2 = 937 Btu/hr-ft2-OF Then, Ao Ai hi 3.28 957 = 0.00350 hr-ft2-~F/Btu Substituting in Equation 18, ho = 0.0886 [ydrocarbon65.Ahydrocarbon. = 0.0886 35.64 x 106lo165 102.26 = 80.3 Btu/hr-ft2 -F Then, 1 1 0i ~ B^J = 0.01248 hr-ft2-~F/Btu. Substituting in Equation 17d, A — + ro il 1 _ o + uo Ai hi 1 ho + 0.00058 = 0.02018 - [0.00350 + 0.01245 + o.ooo08] = 0.00579 hr-ft2-OF/Btu 62

The University of Michigan T Engineering Research Institute APPENDIX J SUMMARY OF CALCULATED RESULTS FOR THE HIGH-PRESSURE GAS COOLER

0\ F-p Run Date Days AtH20 Athydro- ATLM Q x 10 Vt Uo /Ai hi Ao RunH20e ay^ At~o~thydro- A^ ^ x 10'6 ^b^AO/Ai hi _i + rto No. of of carbon Ai Remarks Run Operation ~F OF OF Btu/hr ft/sec Btu/hr-ft2- _F hr-ft2 -~F/Btu hr-ft2- F/Btu 1 Aug. 2, 1955 4 15.82 42.93 26.15 5-93 3.64 25.8.00319 0 2 Aug. 2, 1955 4 15.19 44.85 27.70 5.60 3.57 22.9.00321 0 5 Aug. 5, 1955 7 18.13 56.90 32.55 6.65 3.55 23.2.00323 +0.0020 4 Aug. 8, 1955 10 26.73 48.36 26.85 10.05 3.65 42.6.00328 -0.177 5 Aug. 13, 1955 15 11.19 42.71 27.41 4.12 3.58 17.1.00340 +0.0172 6 Aug. 17, 1955 19 27.00 16.94 19.89 8.29 2.98 47.4.00370 -0.0205 7 Aug. 20, 1955 22 19.26 13.86 18.78 5.06 2.54 30.6.00421 -0.0094 8 Sept. 7, 1955 40 24.91 19.80 27.50 4.44 1.73 18.3.00600 +0.0107 9 Sept. 20, 1955 53 24.52 23.10 31.45 6.02 2.38 21.8.00465 +0.0033 10 Oct. 14, 1955 77 16.31 -0.47 21.60 4.10 2.44 21.6.00489 +0.0035 Average number of days after cleaning of tubes 11 Dec. 10, 1955 18 27.06 35.70 29.00 2.64 0.95 10.0.01142 +0.0504 Run No. 1l not plotted or in12 Dec. 17, 1955 25 30.67 34.75 36.40 9.45 2.99 27.7.00449 -0.0066 eluded in averages due to 13 Dec. 23, 1955 31 26.61 32.66 38.50 8.26 3.02 22.8.00438 +0.0013 questionable data. 14 Dec. 29, 1955 37 36.43 54.74 45.55 11.35 3.02 27.8.00425 -0.0065 15 Jan. 7, 1956 46 31.73 37.26 33.45 10.51 3.19 33-5.00422 -0.0126 16 Jan. 13, 1956 52 32.77 44.70 35-85 10.05 2.98 29.9.00446 -0.0093 17 Feb. 1, 1956 71 31.70 50.18 49.70 9.70 2.97 20.8.00455 +0.0052 18 Mar. 35, 1956 102 33.16 44.76 50.20 9.57 2.89 20.3.00437 +0.0067 19 April 21, 1956 151 39.4 26.0 54.30 8.43 2.08 26.2.00541 -0.0054 20 May 26, 1956 186 46.07 23.95 30.50 9.11 1.92 31.8.00610 -0.0129 Average: (Omitting Run No. 11) 26.24 54.66 32.23 7.72 2.90 27.0 -a S m~ 3m -o 3 go -1 0 3 w m,

The University of Michigan T Engineering Research Institute APPENDIX K SUMMARY OF CALCULATED RESULTS FOR THE DEBUTANIZER OVERHEAD CONDENSER 65

Run Date No. of Run 1 Aug. 2, 1955 2 Aug. 5, 1955 3 Aug. 8, 1955 4 Aug. 13, 1955 5 Aug. 17, 1955 6 Aug. 20, 1955 h\ 7 Aug. 31, 1955 8 Sept. 7, 1955 9 Sept. 20, 1955 10 Oct. 14, 1955 11 Oct. 21, 1955 Days AtH20 Athydro- ATIL Q x 10-6 Vt U of carbon Operation ~F ~F OF Btu/hr ft/sec Btu/hr,-ft2 ~F XI~~~~- -2'I r7 1 = t 1. 7 -1 rN = ^ A d' 7 " _ - 1 7 10 15 19 22 33 40 53 77 84 13.70 12.00 21.79 24.73 23.13 26.85 15.94 14.11 17.92 11.46 17.71 )U.45 59.51 67.35 63.36 60.51 54.70 68.49 66.84 65-97 44.39 29.12 9Ly.u 15.11 17.79 13.45 15.96 16.60 19.95 16.40 20.10 36.45 28.60 9-37 11.57 2.97 9.28 9.81 9.43 11.42 13.63 11.20 12.05 7.40 3.19 3.62 1.11 1.91 1.78 1.83 1.91 3.84 3.56 3.01 2.90 50.2 79.8 17.4 72.1 64.2 59.4 59.9 86.9 58.3 34.5 27.0 r 3 -m -, I 3- 0 =' fi) 3 Average 57.32 19.99 9.83 2.61 55.4 m -I G" w m I Ir Sb 00 In

The University of Michigan T Engineering Research Institute - APPENDIX L SUMMARY OF CALCULATED RESULTS FOR THE DEBUTANIZER BOTTOMS COOLER ] 67

1 The University of Michigan Engineering Research Institute Run Date AtH20 thydro- ATLM Vt Q x 1- U ho' Ao/Ai hi _10 103 i + r. Days No. of carbon Uo ho. Ai of Run'F'F ~ F ft/sec Btu/hr Btu/hr-ft2-'F Btu/hr-ft2-'F hr-ft2-'F/Btu hr-ft2-~F/Btu hr-ft2-'F/Btu Operation 1 Aug. 2, 1955 33.12 2 Aug. 5, 1955 29.81 3 Aug. 8, 1955 32.36 4 Aug. 13, 1955 32.24 5 Aug. 17, 1955 40.88 6 Aug. 20, 1955 37.42 7 Aug. 31, 1955 36.68 8 Sept. 7, 1955 40.39 9 Sept. 20, 1955 41.75 10 Sept. 20, 1955 40.84 11 Oct. 7, 1955 40.14 12 Oct. 14, 1955 31.08 13 Oct. 21, 1955 26.83 Average of Runs 1-13 35.65 273.12 270.26 282.65 279.12 282.37 266.02 243.52 234.49 267.70 258.29 235.45 266.63 222.88 57.6 4.83 9.37 55.3 4.56 7.96 60.6 4.62 8.75 62.0 6.24 11.75 65.0 4.83 11.54 60.6 4.71 10.32 60.0 4.58 9.82 65.9 4.16 9.82 85.9 4.35 10.62 83.8 4.35 10.38 79.8 4.26 10.00 80.5 4.01 6.59 81.8 4.24 6.66 64.8 78.5 57.5 71.1 57.5 73.4 75.7 89.8 70.8 88.0 68.1 85.0 65.4 87.4 59.5 89.4 49.4 86.4 49.5 87.0 49.9 90.8 32.6 63.4 32.4 71.8 0.00244 0.00265 0.00267 0.00210 0.00246 0.00251 0.00270 0.00291 0.00275 0.00275 0.00278 0.00313 0.00304 2.82 35.5533 3-77 2.06 2.74 2.91 3.83 5.61 8.66 8.70 9.03 14.90 16.92 0.00000 +0.00030 +0.00072 -0.00042 -0.00010 +0.00002 +0.00075 +0.00232 +0.00553 +0.00557 +0.00587 +0.01139 +0.01350 4 7 10 15 19 22 33 40 53 53 70 77 84 260.04 69.1 4.60 9.51 56.4 81.7 0.00268 Unit Cleaned on Tube-Side Only Days After Cleaning 6.00 +0.00257 10 8.14 +0.00458 20 14 Nov. 18, 1955 44.68 15 Nov. 28, 1955 52.53 16 Void 17 Dec. 17, 1955 50.30 18 Dec. 23, 1955 27.70 19 Dec. 23, 1955 18.18 20 Dec. 23, 1955 30.82 21 Dec. 29, 1955 32.92 22 Dec. 29, 1955 21.36 23 Dec. 29, 1955 50.92 24 Jan. 7, 1956 28.32 25 Jan. 7, 1956 44.06 26 Jan. 7, 1956 20.06 *27 Jan. 13, 1956 16.31 28 Jan. 13, 1956 12.80 29 Jan. 13, 1956 18.49 30 Jan. 13, 1956 23.66 31 Jan. 13, 1956 38.36 32 Feb. 1, 1956 16.34 33 Feb. 1, 1956 15.02 34 Feb. 1, 1956 17.60 35 Feb. 1, 1956 22.14 36 Feb. 1, 1956'46.95 37 Feb. 7, 1956 12.50 38 Feb. 7, 1956 15.68 39 Feb. 7, 1956 16.46 40 Feb. 7, 1956 22.44 41 Feb. 7, 1956 42.12 42 Feb. 7, 1956 46.95 43 Feb. 18, 1956 17.58 44 Feb. 18, 1956 21.56 45 Feb. 18, 1956 22.56 46 Feb. 18, 1956 31.82 47 Feb. 18, 1956 38.94 48 Feb. 18, 1956 52.77 49 March 3, 1956 19.76 50 March 3, 1956 26.08 51 March 3, 1956 21.92 52 March 3, 1956 27.68 53 March 3, 1956 36.22 54 March 3, 1956 49.16 55 March 3, 1956 64.28 56 March 10, 1956 22.00 57 March 10, 1956 21.78 58 March 10, 1956 28.62 59 March 10, 1956 40.94 60 March 10, 1956 72.89 61 March 10, 1956 62.36 62 March 10, 1956 35.94 63 April 21, 1956 26.20 64 April 21, 1956 31.00 65 April 21, 1956 43.40 66 April 21, 1956 78.40 67 May 26, 1956 18.50 68 May 26, 1956 23.38 69 May 26, 1956 30.82 70 May 26, 1956 42.28 71 May 26, 1956 66.44 Average of Runs 32.65 14-71 (Omitting Run No. 16) Average of Runs 33.12 1-71 (Omitting Run No. 16) 183.99 65.1 3.97 10.41 183.92 78.2 3.84 11.80 65.8 108.8 60.2 118.1 104.48 104.30 110.04 103.48 104.28 113.50 102.12 96.70 89.98 105.06 102.26 102.10 101.44 100.18 93.76 108.28 101.32 98.82 96.44 84.25 106.68 106.32 104.64 103.12 95.78 91.52 112.92 111.30 111.66 108.26 107.26 95.83 118.56 116.26 124.74 152.80 146.68 112.02 104.40 134.26 131.74 131.55 124.30 118.88 113.46 126.18 157.80 155.00 150.40 137.00 130.04 127.12 125.28 118.06 105.28 115.93 41.5 36.5 33.5 33.7 45.0 38.1 40.7 38.0 39.1 35.3 29.3 28.2 29.5 30.1 31.7 46.5 37.8 38.9 36.8 41.5 35.3 36.9 35.7 37.6 39.6 38.6 44.7 45.5 44.0 46.9 44.9 42.5 50.2 51.4 47.2 52.8 58.0 48.9 48.8 52.8 52.2 55.0 53.6 49.2 55.5 54.2 55.5 60.1 60.8 59.1 47.0 48.5 49.6 51.4 51.9 45.17 2.88 2.92 4.41 2.22 2.78 4.55 1.76 2.90 1.71 4.61 3.82 4.65 3.27 2.61 1.55 3.87 3.02 2.59 1.90 1.12 4.12 3.44 3.11 2.39 1.31 1.08 4.21 3.41 3.08 2.24 1.60 1.09 4.05 3.02 4.46 3.60 3.16 1.80 1.24 3.90 4.23 3.25 2.43 1.67 1.43 2.45 4.21 4.17 2.89 1.41 6.15 4.94 3.57 2.56 1.44 2.99 5.11 4.74 4.68 4.01 5.36 4.55 5.24 4.80 4.40 5.42 3.64 3.49 3.54 3.61 3.48 3.69 2.66 2,66 2.46 3.10 3.01 3.17 3.00 3.15 3.22 2.97 4.33 U.29 4.06 4.17 3.64 3.38 4.68 4.61 5.72 5.82 6.70 5.16 4.66 5.02 5.41 5.44 5.77 7.11 5.23 5.15 6.45 7.57 7.32 6.43 6.66 6.75 6.45 6.32 5.58 4.93 5.78 49.1 51.8 55.9 47.5 47.4 59.3 51.3 50.3 44.9 61.1 49.6 49.3 47.3 47.8 43.3 31.7 23. 27.2 26.6 29.7 4. i 34.3 33.5 33.4 52.5 30.7 33.6 57.7 36.3 355.4 32.4 31.7 37.2 35.8'!'.~s P.'rt.] 1l 48.4 44.1 46.2 42.2 58.o 37.9 41.3 39.5 42.9 57.5 37.6 57.9 46.4 50.1 48.0 43.3 56.5 55.5 51.8 49.1 43.0 43.24 45.65 98.6 94.1 90.4 85.0 101.8 100.4 101.8 99.3 98.5 102.2 80.5 78.3 79.3 81.0 82.5 73.1 65.9 66.9 64.6 82.1 69.0 71.6 69.d 72.7 77.6'6.0 84.5 84.3 81.5 34.u 77.3 79.3 35.9 86.2 35.0 83.6 94.7 95.4 93.1 85.0 88.3 85.7 95.5 112.2 95.4 87.8 87.9 98.6 98.6 96.2 101.9 104.1 102.1 104.8 10i.9 88.90 0.00305 0.00318 0.00420 0.00415 0.00306 0.00510 0.00420 0.00294 0.00577 0.00420 0.00624 0.00307 0.00350 0.00303 0.00392 0.00465 0. 00 75 0.00555 0.00452 0.00490 0.00620 o.oo62f) 0. 00876.00335553 0.00379 0.00407,. 00495 70.00755 0.00871 o.oo8!9 0.00 -19 0.00376 0.00405 0.00510 0.00650 o. oo3o55 0.00855 0.00319 0.00392!:s,[ ilv 0.00290 0.00338 o.oo g58 0.00556 0.00719 0.00339 0.00318 o.oois8 0.00581 0.00466 0.00580 o.oo00670 0.00oo468 0.00294 0.00294 o. oo383 0.00585 0.00629 0.00234 0.00271 0.00346 0.00433 0.00650 0.00455 10.22 8.68 6.82 9.27 11.26 6.92 9.66 9.87 12.13 6.57 7.67 7.52 3.29 8.55 10.78 18.75 20.51 21.30 22.12 21.47 14.85 15.17 15.0o 16.19 17.91 19.38 14.0 o 14.75 14.92 16.43 17.99 19.03 15.22 16.35 10.17 10.68 11.05 13.20 15.58 14.35 12.88 13.67 12.87 8.49 16.11 15.00 10.22 9.81 10.71 12.70 7.48 8.42 9.51 10.80 13.61 +0.00564 +0.00415 +0.00338 +0.00379 +0o.00668 +0.00360 +0.00351 +0.00529 +0.00551 +0.00312 +0.00379 +0.00411 +0.00399 +0.00352 +0.00oo365 +0.01484 +0.01581 +0.01652 +0.01554 +0.01233 +0.01114 +0.01100 +0.01058 +0.01086 +0.00998 +0.01029 +0.01050 +0.01058 +0.01051 +0.01095 +0.01111 +0.01010 +0.01165 +0.01205 +0.00689 +0.00682 +0.00696 +0.00726 +0.00801 +0.01 oo58 +0.00932 +0.00948 +0.00783 +0.00231 +0.00903 +0.00994 +0.00690 +0.00649 +0.00650 +0.00603 +0.00476 +0.00533 +0.00567 +0.00609 +0.00673 39 45 45 45 51 51 51 60 60 60 66 66 66 66 66 85 85 85 85 91 91 91 91 91 91 91 102 102 102 102 102 102 116 116 116 116 116 116 116 123 123 123 123 123 123 123 164 164 164 164 199 199 199 199 199 1 142.63 49.62 3.29 87.56 0.00421 68

- - The University of Michigan * Engineering Research Institute APPENDIX M DEBUTANIZER OVERHEAD CONDENSERCOMPUTATION OF THE CONDENSING RANGE OF THE HYDROCARBON STREAM Assume that the hydrocarbon stream consists of 50 mol% n-butane and 50 mol% propane. The average molecular weight is then n-Butane 0.50 x 58 Propane 0.50 x 44 = 29 = 22 51. This checks closely with the specified value of 50.8, indicating that the assumption is reasonable. To check the assumption further, the overall heat duty, assuming the above mixture, may be computed. The heat duty may be divided into the latent heat of condensation at 110~F and the sensible heat of cooling the vapor from 135~ to 1100F. n-Butane 29 x 100 =56.8 wt 51 22 Propane = -— x 100 = 51 43.2 wt% At 1100F (data from Maxwell6): Vapor pressure of propane = 14.5 atm Vapor pressure of n-butane = 4.0 atm Latent heat (k) of propane = 130 Btu/lb Latent heat (x) of n-butane = 146 Btu/lb In the range 135~ to 1100F: Cp of propane =.43 Cp of n-butane =.43 W X propane = 143,000 x.432 x 130 W x butane = 143,000 x.568 x 146 W Cp At propane = 143,000 x.432 x 25 x.43 W Cp At butane = 143,000 x.568 x 25 x.43 Total Q BtAi/lb ~F Btu/lb ~F 8,050,000 Btu/hr = 11,850,000 Btu/hr 665,000 Btu/hr = 875,000 Btuhr 210440o000 Btu/hr 69

The University of Michigan * Engineering Research Institute From the NGSMA7 enthalpy charts the value of 21,130,000 Btu/hr may be obtained. The average of these figures checks closely with the specified value of 21,360,000 Btu/hr. Thus, a mixture of 50 mol% propane and 50 mol% butane will be used to compute the condensing range. The procedure used in computing the dew-point and bubble-point curves for the stream is explained in Perry,9 p.587. It involves a trial-and-error solution and uses the equilibrium constants (K) of the components. These values of K were obtained from the NGSMA data book.7 The calculations are presented below. I* BUBBLE-POINT CALCULATIONS. A. At 200 psia Mol4 (M) K140OF M x K140OF K143~F M x K1430F - - r%.1 - - I - C3H8 0.50 1.53.69 1.40 n-C4H10 0.50 0.57.28 0.60 0,97. Bubble point at 200 psia = 1435F. B. At 175 psia.Mo01 (M) K130oF M x K130 F C3H8 0.50 1.450.715 n-C4Hlo 0.50.565 283 0.998. Bubble point at 175 psia = 130~F. C. At 150 psia._~.2.a 0.70 0.50 1.00 Mol% (M) K120OF M x K1200F K118~F M x Kl18~F C3H8 n-C4Hlo 0.50 0.50 1.50 0.55 0.75 0.28 1.05 1.46 0.54 0.73 0.27 1.00.. Bubble point at 150 psia = 118 F. 70

The University of Michigan Engineering Research Institute D. At 12_ psia — Mol% (M) K100OF M x K100oF Klos OF 1.50 0.55 c3H8 n-C4H10 0.50 0.50 1.42 o.48 0.71 0.24 0.95 M x K105F 0.75 0.26 1.01. Bubble point at 125 psia = 104 F. E. At 100 psia — Mo1f (M) K90oF M x KgoOF C5H8 n-C4H10 0.50 0.50 1.55 0.49.775.245 1.020 1.52 0.48 M x K88oF 0.76 0.24 1*00. Bubble point at 100 psia = 880 F. F. At 70 psia. Mol% (M) K6OF M C3H8 n-C4H10 0.50 0.50 1.50 0.42 M x K60qF 0.75 0.21 0o 96 1.55 o.43 K62OF x - K62 F 0.775 0Q215 0.990. Bubble point at 70 psia = 635Fo II. DEW-POINT CALCULATIONS Since both C3H8 and n-C4H10 are error may be simplified so that the value 2.00 for a check. 50 mol% of the total, the trial and of (1/Kc3H8 +1/KnC4H1) must equal A. At 200 psia. — K165 tF 1/K165~F K62oF 1/K1620F C53H8 n-C4Ho1 1.67 0.75 0.60 1.37 1.97 1.62 0.71 0.617 1.408 2.025.. Dew point at 200 psia = 164~F. 71

The University of Michigan * Engineering Research Institute B. At 175 psia.K152~F / 192 F C3H8 1.71 0.585 n-C4Ho1 0.71 0.408 1.993. Dew point at 175 psia = 152~F. C. At 150 psia. — K140 l/K140oF K138OF l/K138oF C3H8 1.78 0.562 1.74 0.575 n-C4H10 0.70 1.425 0.69 1,450 1.987 2.025 Dew point at 150 psia = 139OF. D. At 125 psia, K125 F 1/K125 CF K127F 1/K1270F C C3H n-C4Hlo 1.80 0.68 0.555 1o.85 0.69 1.470 2.025 0.54 1.45 1.99. Dew point at 125 psia = 127~F. E. At 100 psia. Kll0 F 1/Kll F K11 3 F 1/Ki3 OF C 3H8 n-C4H1O 1.90 o.65 0.526 1.559 2. 65 1.95 o0.68 0.515 1.470 1.985. Dew point at 100 psia = 112 F. F. At 70 psia, K90 "F c3H8 n-C4H10 2*12 o.68 1/K900F 0.471 1.470 1.941 K86OF 2.04 0.64 1/K86 F,049 1.56 2.05 ~. Dew point at 70 psia = 88~F. 72

The University of Michigan Engineering Research Institute. The calculated dew-point and bubble-point curves are shown in Fig. 104 The condensing range, which is the vertical distance between the curves in Fig. 10, is shown in Fig. 11. i 73

The University of Michigan 170 160 140 Dew-poin LL 0 a Lr H 120 Wr aI- I H Engineering Research Institute 200 PRESSURE- PSI A Fig. 10. Dew-point and bubble-point curves for propane and 50 molo n-butane. a mixture of 50 mol% LL 0 (D z (D z cr) z w a z 0 0 26 - 24 22 - 20 18 _ 60 80 100 120 140 160 180 200 PRESSURE -PSIA Fig. 11. Condensing range of a mixture of 50 molo propane and 50 molo n-butane. 74

L The University of Michigan E Engineering Research Institute APPENDIX N DEBUTANIZER OVERHEAD CONDENSERCHECK ON THE OVERALL HEAT TRANSFER COEFFICIENT ASSUMING A NONFLOODED TOTAL CONDENSER From the specification sheet: Water inlet temperature Water outlet temperature Water flow rate Hydrocarbon inlet temperature Hydrocarbon outlet temperature Operating pressure (shell side) Hydrocarbon specific gravity Hydrocarbon molecular weight Fouling resistance: Shell side (ro) Tube side (Ari) Ai = 85 ~F 110 F =855,000 lb/hr = 135 F =110OF =133 psig 121~ API =50.8 0.0005 o.o008 I Water velocity: W Vt =22 223,000 855 000 223 000 = 3.83 ft/sec. (Note: this disagrees with the specified value of 7.0 ft/sec.) Inside heat transfer coefficient: 150(1 + 0.011tw) Vt0.8 hi = 0d.2 di tw 5 + 110 975~F = 0.51 inch di = 0.51 inch hi = (150)(1 + 0.011 x 97-5)(5.83)0'8 (0.51). (150) (2.072) (2935) (0.874) 75

! --- The University of Michigan - Engineering Research Institute -- = 1041 Btu/hr-ft2-~F. AThen hi.00315 hr-ft2-~F/Btu. Metal resistance: I A_ rm Am = 0.00027 (see.Section V-D). Fin resistance: rf = 0.00011 (see Section V-D). The condensing coefficient may be evaluated from the equation ho = 0.725 Kf3 Pf g () -) (1 / (1/4 From the specification sheet for the unit: pff = 0.1 centipoise at 123~F Kf = 0.078 Btu/hr-ft2 (~F/ft) pf = 0.56 gm/cc. As shown in Appendix M the hydrocarbon stream has a latent heat'closely approximating a mixture of 50 molf propane and 50 mol% n-butane. Therefore, the value of latent heat to be used in calculating the condensing coefficient is the average of propane and n-butane. Trial No. 1 Assume ho' = 1160 Btu/hr-ft2-~F.1 1/ho't Ao/Ai hi A0 rm/Am rf ro Ao ri/UAi:et 1/ o.00085 hr-ft2- F/Btu = 00315 hr-ft2 -F/Btu.00027 hr-ft2- F/Btu.00011 hr-ft2-OF/Btu =.00800 hr-ft2-~F/Btu =.00050 hr-ft2-~F/Btu.01288 hr-ft -~F/Btu 76

The University of Michigan Then Uo Atc = h (ATLM) (1/At )1/4 tf = tb - (Atc/2) * Engineering Research Institute = 77.6 Btu/hr-ft2-~F 77.6 = 160 x 25 1.67~F = o.880 = (135 + 110 - (1.67/2) 122 F. From Maxwell at 122~F: Vapor pressure propane = 16 atm Vapor pressure butane = 4.6 atm X propane = 128 Btu/lb X butane = 142 Btu/lb X average = 135 Btu/lb (X)1/4 = 40 At ho' = 1160 Btu/hr-ft2-oF: (1/Deq)1/4 -34 From analysis of the blueprints: CN/(N)1/4 = 0.94. Solving for the physical property group: 3 pf2 g1/4 1/4 Pf2 gc / [(0.078)3 (0.56 x 62.4)2 (32.2)(3600)21,f J (0.1 x 2.42) - [9.97 x 108]1/4 = 177.7. I Therefore, hot = (0.725)(177.7)(3.40)(3.34) (0.88) (0.94) = 1210 (no check with assumed value of 1160). 77

The University of Michigan * Engineering Research Institute Trial No. 2 Assume ho' = 1220 Btu/hr-ft2-~F l/ho' =.00082 hr-ft2-OF/Btu. 1/Uo =.01285 hr-ft2-OF/Btu Uo = 77.8 Btu/hr-ft2-oF Atc 77*8 At = - x 25 = 1.59~F 1220 and (l/Atc)1/4 = 0.890. Since the other factors in the hob equation remain essentially constant from Trial No. 1, ho, = 0.890 x 1210 = 1223 Btu/hr-ft2-F. 0,88o0 This checks with the Then Uo = 4.8 This is -- x 100 = 73 assumed value of ho' = 1220. 77.8 Btu/hr-ft2- F. 6.6% higher than the specified value of Uo = 735 78

The University of Michigan Engineering Research Institute APPENDIX 0 CORRECTION FACTOR TO BE APPLIED TO THE LOG-MEAN TEMPERATURE DIFFERENCE IN A NONFLOODED, SPLIT FLOW CONDENSER Tw4 TVl I TV3 Legend - - - - Water Flow -- Vapor Flow Let (ATLM)corrected - (F) (TV, - TW4) (T3 - TW ) Then the eqs (F) is aation to be used in computing the value of the correction factor 1 F = 1 4 [ (TV2 - Tw) - (TV1 - TW3)] (TV1 -TW,) (Tvs - TW1) (TV1 - T4) [(Tv TW) (TV l (TV - TW) [V -^)-( T - )_ 11in ~ [ (TV - TW4 ) - (TV - TW1)] (Tv3 - TW2) (TV - TW1) 79

The University of Michigan * Engineering Research Institute 2 [(Tv Tw) (T -T)] T- T4) (TV3 - TW) 2 [(Tvs - TWa) - (Tv1 - TWs)] (n- Tw) [(TV1 - TW4) - (TV3 - TW1)1 n ( - (TV - TW3) C8 c90