. ---{{OAK RIDGE NATIONAI. I.ABORATORY TR . operated by __;__;umou CARBIDE conpounon for the | u s ATOMIC ENERGY comwssaou | ORNL TM 79 --ii!i:;_iuo-'ncs L _This document cuntums mformqtmn of a preluminary nature ond was prepcred S ~ primarily for internal use at the Oak Ridge National Laboratory, It is. subiect : : to revision or correction and therefore does not represent a final report, The -~ - information is not ‘to -be abstracted, reprinted or otherwise given public dis- = semination ‘without the approval - cf the ORNL pctent branch, Legai and Enfor- S ~ mation Conh'ol Deparfment. ST T ST MASTER' NNl - LTI o e ermemr e i v e e e AR5 58 s e - o Rt ~LEGAL NOTICE - This report was preparsd as on account of Gonmmcm sponsored work., Neither the Unlnd Sfchs,'- : nor the Commission, nor any person acting on behalf of the Commission: "A. Makes any warranty or representation, expressed or implied, with respsct to the uccuracy, completeness, or usefulness of the lnformatwn contained Iin this report, or that the use of any information, apparatus, mefhod or proctss disclosed in !hus report may not infnnge privately owned rights; or - - - : B. Assumes any liabilities with nspccf lo the use of, or for dcmcgos resulting from the use of any information, apporatus, method, or process disclosed in this report. As used in the above, “person acting on behalf of the Commission® includes any employes or contractor of the Commission, or employss of such contractor, fo the extent that such employee or contractor of the Commission, or omplono of such contractor prepares, disseminates, or provides access to, any informuiion pursvaat to Ms .mploymem or contract with the Commiuion, or his nmpioymcnt with such contractor. ORNL-TM-T9 Copy Lo. Contract No. W-T405-eng-26 . Reactor Division WATER TEST DEVELOPMENT OF THE FUEL PUMP FOR THE MSRE el MG 1 bt P. G. Smith | DATE ISSUED . - | MAR 2 7 1862 OAK RIDGE NATIONAL LABORATORY Oak Ridge, Tennessee i : operated by j | UNION CARBIDE CORPORATION ! _ - for the U. S. ATOMIC ENERGY COMMISSION ot ’\ * » iii 'CONTENTS List of Figures Abstract . ¢« v v v 4 4 4 e e e e e e e Introduction . .+ « ¢ ¢ o v v 4« 4 e s . . Experimental Apparatus . . . . . .+ . . . . Pmp » e . r - ” . » - » - & - . - Test LOOP « + « ¢ s o ¢ o ¢ & o & s » & Carbon Dioxide Stripping Devices . . . Instrumentation . . . . « + . ¢ o 4 . . ... Description of Tests .+ « « « « « « « + & Head-Flow-Power-Speed Performance Carbon Dioxide Stripping Tests . . . Pump Tank Liquid and Gas Behavior . . Fountain Flow . . . . . « « « « . Stripper Flow . . « « . « « & Gas Bubble Behavior in the Pump Tank Priming . « « « ¢ « ¢« « ¢« o o o Coastdown Characteristics . . . . . Test Results . . . ¢« « ¢« « « « « « + & Head-Flow-Power-Speed Performance . Carbon Dioxide Stripping Effectiveness Pump Tank Liquid and Gas Behavior Fountain Flow . . . . . « . . Stripper Flow . . + « + « ¢ ¢ ¢ o & Gas Bubble Behavior in the Pump Tank Priming . . . . . . . Coastdown Characteristices . . . . . Conclusions . « « ¢ o & o o o« s 2 o s o o Acknowledgments . . . « + « + o o . . 0 . REferenc eS - o & . . . - * - *« 8 - * » . * Vblumer . Page No. vii O OO N 12 12 12 1k 1k 1k 1k 14 15 15 15 19 22 22 27 27 29 31 31 32 32 Appendix . . « & Nomenclature Table I . . Table II . . Teble III * * * Computations for: Table I « ... Table IT Table III . . iv ‘Page No. , " 35 36 37 .38 N b 0 . O FPig. No. O O~ O o= D o O 11, 12. 130 1k, 15. 16. 17. 18. 19. LIST OF FIGURES Captions Cross Section of Pump Discharge Connection to Loop .« « « .+ Photo of Test LOOP « ¢ v « o v o o o o o s o+ o Stripper Configuration 1 . . . . . « « « « « «+ « & Stripper Configuration 2 . . + « « « « « « + &« « & Stripper Configuration 5 . .« + « « ¢ « ¢« v o o o & Venturi Calibration . « . « « « v ¢ ¢ ¢« v ¢« + & + & Motor Calibration « ¢« « v ¢ « v « ¢ o o o + o o o o Hydraulic Performance, 13-in. Impeller . . . . . Hydraulic Performance, l1ll-in. Impeller . . . . . . Prerotation Baffle .« « « « « + o o « o o o & Head, Input Power, and Speed Versus Flow, 1ll-in. Impeller * * . - . & . » . . v * » . - - * * - . Hydraulic Performance, 1l-in. Impeller, Efficiency Contours Superimposed . « ¢ o ¢« « o « ¢« o ¢ o o « Relative Effectiveness Versus Sweep Gas Flow Half-Life Versus Stripping Flow . . « . « « + & Relative Effectiveness Versus Jet Velocity Cross Section of Upper Labyrinth . . Fountain Flow Versus Speed, 1ll-in. Impeller . Pump Inlet X-Section « « ¢« ¢ ¢« ¢ ¢ ¢ 4 v o o ¢ &« o Page No. O 1 WV W 10 11 13 16 17 18 20 21 23 24 25 26 28 30 , O €) vii ABSTRACT A vertical centrifugal sump-type pump utilizing commercially available impeller and volute deslgns was selected to circulate the fuel salt in the Molten Salt Reactor Experiment (MSRE). Tests were conducted in water to determine the adequacy of the pump design, to assist design of the prototype fuel pump, and to investigate the effectiveness of xenon removal with high velocity liquid Jjets con- tacting sweep gas in the pump tank. Hydraulic head characteristics were within +1 to -3 ft of manufacturers data for a given constant speed. Adequate and necessary provisions were devised to control the liquid and gas bubble behavior in the pump tank. The results of priming and coastdown tests are reported. During the gas removal tésts, the fuel, xenon, and helium in the MSRE were simulated with distilled water, carbon dioxide, and air, respectively. The best configuration removed carbon dioxide from water at approximately 99% of the ideal removal rate when the stripping flow was 65 gpm and the sweep gas flow rate was 4 scfm., “ [, » WATER TEST DEVELOPMENT OF THE FUEL PUMP FOR THE MSRE P. G. Smith INTRODUCTION The Molten Salt Reactor Experiment (MSRE) is to be a'low-pressure, ‘high-temperature, graphite moderated circulatifig fuel nuclear reactor using fissile and fertile materials dissolved in molten fluoride salts and is designed for a heat generation rate of 10 Mw'(l, 2, and 3). Its goals include proving'the'safe and reliable operation of this nuclear reactor'concept and demonstrating the maintainability of molten salt machinery. The investigation reported herein is concerned with the pump ‘required to circulate the fuel salt in the MSRE. A centrifugal sump-type pump consisting of a rotary element and pump tank was selected for this application. The rotary element in- cludes the vertical shaft and underhung impeller, the shaft bearings, and the means for lubricating and cooling the bearings. The pump tank includes the volute (casing), suction and discharge nozzles, other nozzles for accommodating inert ges purge, fuel sampling and enrichment, liquid level sensing devices, a flange for mounting the rotary element, and various liquid bypass flows for degassing and.rémpving Xxenon poison from the circulating'fuel salt. The device used for removal of xenon will be referred to as a "stripper”. Much of the design of the fuel _pump was‘derived from the past experienée.with similér pumps for elévated temperaturé'service which were developed during the Aircraft ' Nuclear Propulsion Program at Oak Ridge National Laboratory (4, 5, ' an@ié)‘- l'_ | | ' : ‘ ) The initial phase of development and testing of the fuel pump was conducted with water to aSceitain_the capability of the pump to meet the hydraulic requirements of the fuel circuit and to remove from the circu- -lating-fuel the xenon whichlwiil-bé generated by the fisslioning process. Data were taken on the headéfiow-power-speed performance of the pump for twb impeller outside diametérs; 13_and 11 inches. Various baffles were devised to control splash, spray, and gas bubbles caused by the operation of the bypass flows in the pump tank. The ability -of the pump to prime was determined at various liquid levels of interest. The coastdown characteristics of the pump were measured from various speeds and\flows. Attempts were made to measure indirectly the effectiveness with which xenon poison might be removedufrdm the circulating fuel using high velocity liquid jJets in contact with gas in the pump tank. During this particular test the fuel and xenon were simulated, respéctively, with distilled water and carbon dioxide; this gas is much more soluble in water than xenon is in molten salts of interest and. in addition provides for convenient measurement of solubility. Pertinent information from these water tests were incorporated. in the desigfi of the prototype fuel pump and will be subjected to elevated temperature testing at MSRE design conditions. EXPERIMENTAL APPARATUS The experimental apparatus includes the pump, the test loop, and the stripper configurations. A description of each follows: Pump The pump is shown in Fig. 1 and includes a centrifugal impeller and volute with the impeller supported at the lower end of a vertical shaft, . grease~lubricated bearings for supporting the.éhaft, bearing housing, pump tank bowl, and volute support. The pump tank bowl was fabricated of plexiglas to-permit visual observation of the behavior of the liquid and the gas bubbles. Labyrinth-type seals were utilized on the impeller inlet shroud and on the impeller support shroud. The Impeller suppoft shroud labyrinth seal was supported on the 1mpeller'cover'plgte,which wvas sealed to the volute by an elastomeric O-ring. The volute discharge vas connected to the pump tank discharge nozzle through a flexibly . mounted bridge tube. The connection arrangement is shown in Fig. 2. Test Bnqp The test loop is shown in Fig. 3, which consists of the pump, piping, venturl flowmeter, throttle valve (globe type), stripper flow circuits - O, o €} ) LOWER LABYRINTH |~ ————— %t PREROTATION BAFFLE Fig. 1. Cross Section of Pump, UNCLASSIFIED ORNL-LR-DWG 60816 MOTOR i i | BEARING HOUSING GREASE LUBRICATED SHAFT BEARINGS jam| = = N VOLUTE SUPPORT "SLINGER 5;‘&;?‘3 FOUNTAIN| | B IMPELLER FLow\ ' STRIPPER PUMP UPPER : h BAFFLE ‘ : CONFIGURATI - TANK LABYRINTH GURATION § | - Pt LIQUID LEVEL —FOUNTAIN J - FLOW— ) = _ . | = . BAFFLE O. UNCLASSIFIED ’ ORNL-LR-DWG 60817 o . ‘ ; O NY ANANTS ANNT MY b ‘ \7 1. ! Y PUMP TANK DISCHARGE NOZZLE — BRIDGE TUBE— T Fig. 2. Discharge Connection to Loop. , © ¢ RN 3. Photo of Test Loop. UNCLASSIFIED PHOTO 36279 (not shown), and a cooler. The pump was driven with a 60 hp d.c. variable speed motor. The vei'tical inlet pipe to the pump was fabri- cated of plexigla.s to permit visual observation of the inlet flow.con- ditions. A bundle of l-in. diameter thin-wall tubes, 6-in. long, was ‘added to the lower end of this pipe to reduce rotation of the water column. The cooler was installed in parallel with the main loop throttle velve. A pert of the main loop flow.was bypessed through the cooler to control the system temperature. The bypass flow was controlled by & -throttle velve located in the bypass flow circuit. Stripper configu- ration flow was supplied through & tep located. just dovnstreem of the pump tenk discherge nozzle. The stripper flow as well as the flow from the impeller upper labyrinth pa,ssed’, through the pump tenk and re-entered the system at the impeller ‘inlet. Throttle velves were used to control _the stripper flow. Following the initial tests an orifice was added to the nearly vertical section of the loop between the discharge and the venturi flow meter to decrease the pressure drop through the maip throttle valve. Carbon Dioxide Stripping Devices Tests were conducted wherein a portion of the pizmp_discha:rge flow was introduced into the gas volume of the pump tank t_hrough high velocity Jets (strippers). A number of configurations were investigated, starting with a single stream and progressing to configurations vwhich gave in- creasingly more fresh liquid-gas interface.- | The strippers tested .and identified in Table I (Appendix) are described as follows (:I.n each test two strippers were used): 1. Configuration 1 is shown in Fig. 4. The flow discharged from one side of the can through 1/L-in. holes. For this test the holes were submerged belofi the liquid surface in the pump tank. 2. Configuration 2 is shown in Fig. 5. The lower end of the entry tube was éfl.bsed. and the beaker was packed with Inconel wool. The strip; pifig_"flow,entered the pump 'tank gas space in tangential direction as a spray. One beaker contai_ned.Bh spray holes, 1/8-111. .in diameter, and the.-other contained 30 spray holes , 1/b-in. in diameter. " O. b ” UNCLASSIFIED ORNL-LR-DWG &608ie ase I/J \\ PUMP TANK LIQUID LEVEL = | ST SIXTY Y,—in. HOLES — o i ! I 1 3in1Ps ? 5 in. 6% in. GRAVITY ™ \ J - . - 5in. Fig. 4. Stripper Configuration 1. ' B ; ¥ | UNCLASSIFIED ORNL-LR-DWG 60819 L PUMP TANK LIQUID LEVEL PACKED WITH INCONEL WOOL DRAIN HOLES Fig. 5. Stripper Configuration 2. ' 131 .in rpm. 3. Configuration 3 was the same as No..2, except for the size of spray holes, and the number of holes. Each stripper contalned 162 spray holes, 1/16-in. in diameter, with the beaker suspended such that the spray was circumferential. L. Configuration 4 was the same as No. 3, except the number of holes was reduced by a factor of two and the spray was directed radially ‘invards towsrd the pump shaft. 5. Configuration 5 was a toroid constructed of pipe as shown in Fig. 6, and located in the pump tank as shown in Fig. 1. Each stripper contained two rows of 80 holes each, 1/16-in. in diameter., INSTRUMENTATION Insttumentation was provided to measure venturi pressure drop, discharge pressure, pump shaft speed, water temperature, motor input power, fountain flow, stripper flow, pH value of the water, and pump tank liquid level. _ ; Three diffefent methods were used in measuring the vetturi pressure drop: mercury manometer, difference between individual pressures measured at the inlet and.throat,.and,by¢differential pressure transmitter. Cali- bration of the venturi was provided by the vendor, and it 1s shown in Fig. T. Individual pressures at the inlet and throat were indicated on Bourdon tube gages, 0-30 psi range, 1/8 psi subdivision, and 1/4% ac- curacy. The differential pressure~transm1tter was read out on a dif- ferential gage, 0-50 psi range, 1/2 psi subdivision, 1/4% accuracy. The flow is estimated to be accurate within + 3%. The discharge pressure'was measured on a Bourdon tube gage, 0-100 psi range, 1/2 psi subdivision, and 1/4% accuracy. The'pump shaft speed was measured by use of a 60-tooth gear mounted on the shaft, a magnetic pickup, and a counter which indicated directly The water temperature was ‘méasured with audialétype thermometer, 0 to 240 F range, 2F subdivision. Motor input power data was. dbtained by two methods pover recorder, 0 to 40 kw range, 0.8 kw subdivision and power analyzer which indicated 10 Fig. 6. Stripper Configuration 5. O. . O AP, DIFFERENTIAL PRESSURE (cm Hg) 11 UNCLASSIFIED ORNL-LR~-DWG 60820 / 120 /. 100 - / # 80 : / : / €0 ',.I 40 /.4 / 20 " < 0 200 400 600 800 {000 {200 1400 1600 4800 2000 2200 @, FLOW (gal /min) Fig. 7. Venturi Calibration. 12 ~current and voltage. The povwer measurements were in error during most of the testing with the 13-in-o.d. impeller which preceded tests with the 11-in. impeller. During'this.period,_investigations were conducted ~to: locate and correct the source of error. Satisfactory power measure- ments were obtained with the 11-in. impeller. The motor calibration curve is shown in Fig. 8. | >The-fountain flow was measuréd by directing the flow through 90° V-notch weirs and measuring the height of the flow column. | The stripper'flow wes measured by use of rotameters. The pH vaelue of the water was indicated with & Beckmen pH meter, Model H-2, renge O to. 1k pH with an accuracy of 0.03 pE. The pump. tank liquid level weas indicated with & scale marked off in O.d=in. divisions. Zero level corfesponded.withlthercenter line of the volute. DESCRIPTION OF TESTS 'Hbad-Flow-Power—Speed Performance Hydraulic perfbrmance data vere cobtained over a wide range of ,oPerating conditions with 1mpellers of 11- and 13-in.. outside diameter. Two methods of operation were used: speed was varied (700 to 1300 rpm) ‘at constant system resistance for several values of resistance with the 13~in. 1mpellef, and system resistance was varied at constant speed for several values of speed (700 to 1300 rpm) with the 1l-in. impeller. Data were obtained for computing head, flow, brake.horsepower, and efficiency. Carbon Dioxide Stripping Tests - A number of tests were>performed with both impeller diameters to ascertain the change in effectiveness of 002 removal caused by various stripper configurations, flow rates, jet velocities, and sweep gas flow rates, Carbon dioxide in dry-ice form was added to the circulating dis- tilled water in the system until saturation .was achieved,,after.vhich time the stripper flow was started. Readings of pH of the water were taken versus time to determine the time required to reduce the CO2 con- centration by a factor of two. A total of 37 tests were performed. O T, O INPUT (kw) 60 50 40 30 20 10 13 UNCLASSIFIED ORNL-LR-DWG 60821 850 rpm 7 1160 rpm ‘ 2400 rpm 60-hp dc G.E.MOTOR SERIAL NO. 7287063 y 10 20 30 40 50 60 70 QUTPUT {hp) Fig. 8. Motor Calibration. 1h An expression was derived to. give the theoreticel time required to .reduce the CO2 concentration by one half. Comparison of the'theoretical and experimental data is reported as relatlve effectiveness of the stripper. Pump Tank Liguid end Ges Behavior Fountain Flow Considerable testing was performed to . observe the flow of weter fram the impeller upper labyrinth (flOW‘up the shaft and return to the system ‘through the pump. tank volnme) and to. develope adequate control of the re- turn of this flow into -the pump tank 11quid keeping the splatter of water “end ges bubble formation to & minimum (see Fig. 1). Clearances were varied. between the shaft and the. impeller wpper labyrinth and the impeller upper shroud. and seal plate. The corres- ponding fountain flows were measured. Strigger’Flow The flow through the various stripper configurations was measured and- baffling was developed to control splatter and ges bubble formation. \ Gas Bubble Behavior in the Pump Tank Volume Throughout all of the testing the formation and behavior of gas bubbles were observed in the pump tank volume. Baffling was devised to prevent entry of gas bubbles into the pump inlet from the pump tank volume. Priming The priming characteristics of the pump were checked at various static liquid. levels -in the pump tank. The abillty of,the.pump-to hold prime -as the liquid level in the pump tank ‘was being lowered was in- vestigated. Data were obtained bf»head—flowhspeed perfbrmence‘and of O. O 15 change in starting level for various starting levels as the pump. was accelerated from zero to design speed. Coastdown Characteristics A number of coastdown tests were made from various pump operating conditions. The power supply to the pump drive motor was interrupted vhile the pump was operating at specific speed and flow conditions, and the time required to reach reduced system flow and pump speed was de- ‘termined. TEST RESULTS Head-Flow-Power-Speed Performance Hydraulic performance data were obtained over a wide range of head and flow conditions at several speeds for the 8 in. x 6 in. volute, using impellers of 13- and 1ll-in. outside diameter. These tests with the 13-in. diameter impeller were conducted without a baffle in the pump inlet. The 13-in. impeller performance is presented in Fig. 9, which is.a plot of head versus flow at various speed?. The flow 1s total flow consisting of system flow, fountain flow, ané stripper flow. The corresponding dats are tabulated in Table IT (Appendix). Allis-Chalmers data are also shown for comparigcon. The heads obtained are increasingly lower than .Ailis—Chalmers data with decreasing flow at constant speed. The performance obtained with the 1l-in. diameter impeller is pre- sented in Fig.'lo,_whidh.is a plot of head versus flow at various speeds. The flow is total fldwficonsisting of system flow, fountain flow, and stripper flow. The correspbnding;data_are.tabulated in Table IIT (Ap- pendix). Data for'thrée différent inlet configurations are shown: in two configurations a prerotatioh\baffle was located at the inlet to the impeller; and the other configuration had none. The baffle consisted of two,plates:arrangednin.a.crbés as shown in Fig. 11; it had the effect of increasing the head at the lower range of flows on a constant speed line. There was essentially no difference in the results obtained with the two g H,HEAD {ft) o o 30 16 UNCLASSIFIED ORNL~LR-DWG 60822 1300 rpms; i-.\ r "50 FPM = "-“—--"—-q / ..‘-L /) '-.,‘-.\ e \ -~ A H50~g %S . \ Z (. J _/ N /) /., T '/ A / / ] / - 4 4;// :;/<:>,/ 700~ 17 7 "~~~ CONSTANT RESISTANCE - === ALLIS-CHALMERS DATA == ——=— ORNL. DATA WITHOUT INLET BAFFLE L ALLIS CHALMERS MFG. CO. IMPELLER AND VOLUTE. SIZE 8x6; TYPE E; IMPELLER P-482,13~in. CD o 200 400 €00 800 1000 {200 1400 @, FLOW (gal/min) 1600 180C 2000 2200 2400 Fig. 9. Hydraulic Performance, 13-in. Impeller. O. £ . O 17 UNCLASSIFIED ORNL-LR-DWG 60823 88 . . . T T . ALLIS CHALMERS MFG. CO. IMPELLER-VOLUTE SIZE 8x6; TYPE E; IMPELLER P-482, 11-in. 0D 80 ——ALLIS CHALMERS DATA —_ ORNL DATA © WITHOUT INLET BAFFLE 4-19-6! 72 ® WITH INLET BAFFLE (4-in. LONG) 4-20-61 _ | ——a] e & WITH INLET BAFFLE (2'2-in. LONG) 6-19-6! &.\ 64 \K\\ \t...‘\ T A ] 1300 rpm E N \__{\\ \4 — ha [~ w _ R S . 40 SaN., T }\ 150 1.&\ 32 _a__-.--.‘_!“- [ ., N ey X B6O -.——--_._‘_" N% v Jep— \ 18 o p= 700 8 0 0 200 400 600 800 1000 1200 1400 {600 180C @, FLOW (qgal /min) -Fig. 10. Hydraulic- ‘Performance , 11-in, Impeller. 2000 18 UNCLASSIFIED ORNL-LR-DWG 60824 l‘/ - \\\ "'4 - \- - - - ’l ~ ‘I FLOW Fig. 11. Prerotation Baffle. . O -Gy O. - - 19 sizes of baffles. Curves of head and pump input power versus flow at various speeds are shown in Fig. 12. The input power change versus flow for constant speed operation is slight. The prerotation baffle was not fully tested with the 13-in. im- peller. Were such a baffle used with the 13-in. impeller, performance would be more nearly coincident with the published Allis-Chalmers data. From the power data obtalned with the 1l-in. diameter impeller, efficiency contours were computed which are shown in Fig. 13, super- “imposed on a plot of head-flow-speed data. Carbon Dioxide Stripping Effectiveness In the stripping tests, data were obtained to determine the time required to reduce the CO2 concentration by one half (half-life). The change in pH value of the distilled water was measured over a range from 4 to 6 versus time. Fdr plotting purposes the pH values were converted to the logarithm of the molarity of CO2 to determine the half-life. Theoretical half-1ife (t = 0.69 V/QS) was computed for each test and compared with the experimental half-life to give relative effective- ness. The results of the carbon dioxide stripping tests are presented in Teble I (Appendix). Related in the table are data pertaining to the stripper configurations, by-pass flows, liquid level in the pump tank, sweep gas flow rate, system volume, jet velocity, experimental half-life, ideal half-life, and relative effectiveness. | The first six tests fiefe-preliminary; the flow was simply bypassed through the pump tank without passing through strippers. These tests were performed to provide a base from which to proceed with strippers. Values of relative effectiveness ranged from 10 to LO percent. Tests7.through.16'were'éoncerned mainly with varying the stripper configuration. _Other‘variéblesfimay_be noted in the data shown in the table . Values of effectiveness ranged from 15 to 68 percent. ~ From the results of tests through No. 16, configuration 5 (Fig. 6) was_derived and used fbr'thé'reméinder of the tests, 17 through 39. 20 UNCLASSIFIED ORNL~LR-DWG 60825 o Q 1300 rpm — n o PUMP INPUT POWER (hp) S o - n 64 56 48 , \\ ‘c\ 40 L M, HEAD (1) 32 o Tl 24 86‘0 16 |— 50 gal/min STRIPPER FLOW AT 1200 gal /min, 48.5 ft, 1150 rpm; 7.5 gal/min FOUNTAIN FLOW 4-in. IMPELLER, 8 x 6 IMPELLER-VOLUTE 8 — 2% -in. PREROTATION BAFFLE ] c 200 400 600 800 1000 1200 1400 1600 1800 2000 @, FLOW (gal /min) Fig. 12. Head, Input Power, and Speed Versus Flow, 1l-in. Im- peller. ‘ O b 72 64 56 48 40 H,, HEAD (ft) 32 24 16 UNCLASSIFIED ORNL-LR-DWG 60826 70 | ~_ 75 / )'\ — ~<80 £ | e < 87.5 ~ / / o g s / ~ 89.5 EFFICIENCY TSa No CONTOURS ~ ~ _87.5 ..‘\ \. ~ TN.85 \‘.\ g & ~ Aoo \'\ rpm /) 7 // N150 ™ - / > \‘\86 / 70 -—-..“4‘-/ / .\\\ 75 e~ | ~ 80 < S g7+ ™ 860 o *~_V , er.s‘gf- 8 700 FROM DATA WITH 4-in. LONG PREROTATION BAFFLE 0 200 400 600 800 1000 1200 1400 1600 1800 2000 @, FLOW {gol /min) Fig. 13. Hydraulic Performance, ll-in. Impeller, Efficlency Contours Superimposed. 22 In tests 17 through 24, the flow and jJet velocity were varied simul- taneously at a constant sweep gas flow rate. The relative effectiveness varied. from 27 to 99 percent. In tests 10, 13,.1k4,.17, 18, and 25 through 29, sweep gas flow rate ,was-varied,with-thé other varisbles held constant, and two stripper con- figurations were used. The relative effectiveness variedrf;om»47 to T2 percent. These data are plotted in Fig. 14, Relative Effectiveness VersuSVSweep Gas Flow, for two configurations. In tests 30, 32, 33, 35, and 39, the stripping‘flow”uas varied with the other varisbles held constant. The relative effectiveness varied from 7O to 90 percent. The'results from these tests are shown in Fig. 15, Half-Life (defined on page 18) Versus Stripping Fléw. Experimental and theoretical curves are shown. In tests 30, 31, 34, and 36, the jet velocity was varied with the other variables held constent. The relative effectiveness varied from 27 t0.90 percent. The results are shown in Fig. 16, Relative Effective~ ness Versus Jet Velocity. Configuration -5 was selected for the MSRE fuel pump, and was .in- corporated in the design of the prototype fuel pump. Tests 37 and 38 yielded effectiveness values of 52 and 55 percent, respectively. These tests were performed at the following conditions, reasonably attainable in the MSRE: stripping flow rate of 65 gpm, and sweep gas flow rates . of 0.05 and 0.07 scfm, respectively. | Pump Tank Liquid,and Gas Behavior Fountain Flow | Observations of the fountaein flow from the impeller upper labyrinth (Fig. 17) revealed the need to control it; the slinger impeller was causing an undesirable spray. This spray was contained and controlled -by use of a cover enclosing the labyrinth and slinger impeller, and ‘having drain ports located at its lower end. | Approximate measurements of the fountain flow were made using weirs located in the windows carrying the flow from the fountain into. the pump 0. . O h L 23 UNCLASSIFIED ORNL~-LR-DWG 60827 100 I I | (smuppsa CONFIGURATION 5 3 80 |[— = 17 & TEST NUMBER ~* 13 3 W _ et g 60 — 28 / / P _4__-—-—"____—-"'" u{é .CT:' _——_—\ STRIPPER CONFIGURATION 3 | i w \10 4 40 o6 5 - W t. 20 JET VELOCITY CONSTANT STRIPPING FLOW CONSTANT . ] 0 1 2 3 4 5 6 7 8 9 ¢, SWEEP GAS FLOW (ft*/min) Fig. l4. Relative Effectiveness Versus Sweep Gas Flow. t, HALF -LIFE (min} {6 - N o ® ° b 24 UNCLASSIFIED ORNL-LR-DWG 60828 T 1 { ] e 35 N 33 o EXPERIMENTAL /\ 39 THEORETICAL P | JET VELOCITY CONSTANT SWEEP GAS FLOW RATE CONSTANT STRIPPING CONFIGURATION CONSTANT 0 10 20 30 40 80 60 70 80 STRIPPING FLOW (gal/min) Fig. 15, Half-Life Versus Stripping Flow. 0. b 100 €, RELATIVE EFFECTIVENESS (%) 80 60 40 20 Fig. 25 UNCLASSIFIED ORNL-LR-DWG 60829 s 30 o L~ 9 18 35 \TEST NUMBER STRIPPING CONFIGURATION CONSTANT SWEEP GAS FLOW RATE CONSTANT STRIPPING FLOW CONSTANT | 16' 5 10 15 20 25 v, JET VELOCITY (ft/sec) 30 35 40 Relative Effectiveness Versus Jet Velocity. 26 SLINGER IMPELLER CLEARANCE "A"— LABYRINTH TOP EDGE e OFf IMPELLER CLEARANCE “C" UNCLASSIFIED ORNL-LR-DWG 60830 SHAFT - L Npid/ CLEARANCE "8" T - Fig. 17. Cross Section of Upper Labyrinth. O. o 27 tank. Values of fountaln flow for several labyrfnth clearances are as follows: 13-in. Diameter Impeller, 1450 gpm, 1030 rpm, 50 ft Head ,Configura&ion Number Clearance "A" Clearance "B", Clearance '"C", Flow, gpm 1l 0.015 0.015 0.090 T.5 - 10 2 0.015 0.0k0 0.090 10 - 12 3 0.015 0.0%0 0.250 10 - 12 b 0.015 0.060 0.250 15 - 17.5 Configuration 4 was used with the 1l1-in. diameter impeller and the fountaln flow was measured at various speeds along a constant resistance line defined by 1300 gpm and 45 ft. The fountain flow versus speed is shown in Fig. 18. Configuration 4 was adopted for use on the prototype MSRE fuel pump. ' The direction of the fountain flow was observed over the range of conditions from which_head-flow-speed.data.were_dbtained with both the 1l-in. and 13