e aas e 4w e ks caei i - OH | %) W MN 2 T LN 2 IV P \ Q = - L = Ev - = o ’ O Sow ““““ - o 5 . 2 o == 2 O (= n o e=——= =0 m = ) U o- mm e m = o & > : m = - 2 itV o = R . Wnu = m o & s i o= g ot chofd b = % <3 B (D e o @ < g A S § £ 28 f o C - = - B .m M < = = £%o¢ 5 & & 2 £ & N 10 14 14 19 19 20 21 23 23 23 30 31 35 36 36 40 40 40 43 48 49 52 54 55 57 58 58 61 62 65 66 67 68 69 69 73 78 81 O TR Ty e i e o REGC O FrANSIENES oo e ‘ 87 " Calculated power change resulting from a regulating rod movement ... 90 Reactor temperature differential as a function of helium blower speed............................... 90 The phenomenon of the 1ime 1ag ..o, 91 Reactivity effects of transients in the sodium system ... 92 Reactivity following @ SCPAM ... e 93 Final Operation and SRUtdOWN ... 95 7. RECOMMEND AT IONS et 97 APPENDIXES B. SUMMARY OF DESIGN AND OPERATIONAL DATA e 107 DS CrIP ION oo e 107 MatEriQls oo e 111 REACOr PRy Sies oot e 114 S TG oo 115 Reactor Control .o e e e 116 System Operating Conditions ... o e, 120 MiSCEIIANEOUS ..o e e 123 C. CONTROL SYSTEMDESCRIPTION AND OPERATION .. 125 Control System Design ..o, 125 Instruments DesCriprion ..o 125 Controls DesCriPHOn ..o 126 Console and Control Board Description ... e 129 Control OPerations ..ot e 130 REAEEOr OP@ration .......ooo it 135 D. NUCLEAR OPERATING PROCEDURES ... e, 144 Addition of Fuel Concentrate ..o e, 144 Suberitical Experiments ... 145 Initial Criticality oo e e, 147 Rod Calibration vs Fuel Addition ... 147 Low-Power Experiments ..., 147 Approach to POWer ..o, e, 150 Experiments at Power ..., 152 E. MATHEMATICAL ANALYSIS OF APPROACH TO CRITICALITY .o, 156 F. COLD, CLEAN CRITICAL MASS .. oo e, 158 G. FLUX AND POWER DISTRIBUTIONS ..o oo, 160 Neutron Flux Distributions ..., 160 Fission-Neutron Flux Distributions oo 161 Power Distribution ..o e 162 H. POWER DETERMINATION FROM FUEL ACTIVATION ...l 165 R OrY e e 165 Experimental Procedure ... ..o e, 166 [. INHOUR FORMULA FOR A CIRCULATING-FUEL REACTOR WITH SLUG FLOW ... 168 viii AT AT Sl S TR W L e TN e g o s o e e e Mo P e ERTA T, s g e B TETY » BT e T T WPen. b o et -~ eI WSOy R b B & v R e L otk o Beie ok L TR kb i o i o o J. CALIBRATION OF THE SHIM RODS .o ittt ettt e 172 Calibration from Critical Experiment Data ..o 172 Calibration Against the Regulating Rod ..., 174 Calibration by Using the Fission Chambers ..., 177 K. CORRELATION OF REACTOR AND LINE TEMPERATURES ... 180 L. POWER DETERMINATION FROM HEAT EXTRACTION ..o 184 M. THERMODYNAMIC ANALY S S it oottt 188 Insulation Losses, Heater Power Input, and Space Cooler Performance ... 188 Experimental Yalues of Heat Transfer Coefficients ... 188 N. COMPARISON OF REACTOR POWER DETERMINATIONS ... 190 0. ANALYSIS OF TEMPERATURE COEFFICIENT MEASUREMENTS ... et 192 Importance of the Fuel Temperature Coefficient ... 192 Effect of Geometry in the ARE .. 192 Time Lag Considerations ..., e e 193 Suberitical Measurement of Temperature Coefficient ... 193 Low-Power Measurements of Temperature Coefficients of Reactivity ..., 197 High-Power Measurements of Temperature Coefficients of Reactivity ... 199 P. THEORETICAL XENON POISONING ... et 200 Q. OPERATIONAL DIFFICULTEIES oot e 202 E i MENT SYS ORI oottt 202 Process INStrumentation ..o e e e 204 Nueclear Instrumentation and Controls .o e 205 ANUNCIGEOTS oo ettt e s e e, 205 Heaters and Heater Comtrols o oo et ettt ettt s e 205 System COmPONENTS ... e e 205 L OK S oo e e 206 R. INTEGRATED POWER oo et 208 AP ACEEA P OWET oot et e e 208 N EAE P OWET o oo e oo ettt e et e et e e e et a e e 210 S. INTERPRETATION OF OBSERVED REACTOR PERIODS DURING TRANSIENTS ... 212 T NUCLE AR LG oo oo oottt et e ettt b ettt ettt b e sttt 213 U, THE ARE BUILDING.............co e bR R 232 B B L IO G R A PHY oot s 235 ix E k k I3 - Y }&gm,, SRR 5 AR s el G P ol S o git ik ol B v ks d o VA mn e o eenl “ ol e, bbb .“...m OPERATION OF THE AIRCRAFT REACTOR EXPERIMENT SUMMARY The Aircraft Reactor Experiment (ARE) was oper- ated successfully and without untoward difficulty in November 1954. The following statements sum- marize the notable information obtained from the experiment. 1. The reactor became critical with a mass of 32.8 Ib of U?35, which gave a concentration of 23.9 Ib of U233 per cubic foot of fluoride fuel. For operation at power, the U233 content of the fuel mixture was increased to 26.0 Ib/ft3, and thus the final composition of the fuel mixture was 53.09 mole % NaF, 40.73 mole % ZrF,, and 6.18 mole % UF,. 2. The maximum power level for sustained oper- ation was 2.5 Mw, with a temperature gradient of 355°F; the maximum fuel temperature at this level was 1580°F. Temperatures as high as 1620°F were recorded during transients. 3. From the time the reactor first went critical until the final shutdown, 221 hr had elapsed, and for the final 74 hr the power was in the megawatt range (0.1 to 2.5 Mw). The total integrated power was about 96 Mw-hr. 4. While at power the reactor exhibited excellent stability and it was easily controlled because of its high negative temperature coefficient of re- activity, which made the reactor a slave to the load placed upon it. The fuel temperature coef- ficient was -9.8 x 10™° (Ak/k)/°F, and the over- all coefficient for the reactor was -6.1 x 1077, 5. Practically all the gaseous fission products and probably some of the other volatile fission products were removed from the circulating fuel. In a 25-hr run at 2.12 Mw the upper limit of the reactor poisoning due to xenon was 0.01% Ak/k. No more than 5% of the xenon stayed in the molten fluoride fuel. 6. The total time of operation at high tempera- ture {1000 to 1600°F) for the sodium circuit was 635 hr, and, for the fluoride fuel system, 462 hr. During most of the operating period the sodium was circulated at 150 gpm and the fuel at 46 gpm. 7. The fabricability and compatibility of the materials system, i.e., fluoride fuel, sodium coolant, and Inconel structure, were demonstrated, at least for the operating times, temperatures, and flux levels present. 8. All components and, with few exceptions, all instrumentation performed according to design specifications. The performance of the pumps was particularly gratifying, and the low incidence of instrumentation failure was remarkable in view of the quantity and complexity of the instruments used. R i g ? 1. INTRODUCTION The Aircraft Nuclear Propulsion (ANP) project at the Oak Ridge National Laboratory was formed in the fall of 1949, at the request of the Atomic Energy Commission, to provide technical support to existing Air Force endeavors in the field. The ORNL effort gradually expanded and, following the recommendation of the Technical Advisory Board in the summer of 1950, was directed toward the construction and operation of an aircraft reactor experiment. A complete description of the ARE falls naturally into three categories that correspond to the three phases of the project: (1) design and installation, (2) operation, and (3) postoperative examination. Each of these phases is covered by a separate report, ORNL-1844, ORNL-1845, and ORNL-1868, respectively. Much detailed infor- mation pertaining to the selection of the reactor type and to the design, construction, and pre- nuclear operation of the reactor experiment will be presented in ORNL-1844. As the title of this report (ORNL-1845) indicates it is concerned primarily with the operation of the experiment, and only insofar as they are necessary or useful to the understanding or evaluation of the nuclear oper- ation are design and preliminary operational data included herein. The third report {ORNL-1868) will describe the aftermath of the experiment, with particular reference to corrosion, radiation effects, and the decay of activity ~ effects that cannot be evaluated at this time because of the high level of the radioactivity of the equipment. The specific operating objectives were to attain a fuel temperature of 1500°F, with a 350°F temperature rise across the reactor, and to operate the system for approximately 100 Mw-hr. Other objectives of the experiment were to obtain as much experimental data as possible on the reactor operational characteristics. The extent to which each of these objectives was fulfilled is described herein, and a measure of the success of the program is thus provided. Although it was initiclly planned to use a sodium-cooled, solid-fuel-element reactor, the reactor design evolved first to that of a sodium- cooled, stationary-liquid-fuel reactor and, finally, to that of a circulating-fuel reactor employing sodium as a reflector coolant. These evolutionary processes left their mark on the experiment, par- ticularly in that the reactor had to incorporate a moderator geometry that was originally specified and ordered for the sodium-cooled reactor. The adaptation of this moderator geometry to the circulating-fuel reactor resuited in a reactor in which the fuel stream was divided into six parallel circuits, each of which made numerous passes through the core. These fuel passages were not drainable — a condition which caused considerable concern throughout the course of the experiment. Although it is not the purpose of this report to give a detailed history of the design, construction, or preliminary testing of the reactor system, the final design and pertinent prenuclear operation are briefly described. The bulk of the report concerns the operation of the experiment from the time uranium was added to the fuel system on October 30 until the evening of November 12, when the reactor was shut down for the last time. In addition to the description of the various experiments and onalyses of the data which are presented in the body of the report, the entire nuclear operation, as recorded in the “Nuclear Log,'” is given in Appendix T. The report also includes a number of recommendations based on the operating ex- perience, and much detailed supporting data and information not appropriate for inclusion in the body of the report are given in the other ap- pendixes. The various experiments that were performed on the ARE were designated as E, L, or H series, depending upon whether they occurred during the critical experiment, low-power operation, or high- power operation, respectively. Each of these experiments, as listed below, is discussed in this report. -, » b b B 4 ; ' T mT e B e e e g g Tt g B R e - ek ke i o ko ek btk && S L el e BB o ki i i ML [y kb diy, Experiment Series No. E-1 E.2 L-1 L-2 L-3 L-4 L-5 L-6 L-7 L-8 L-9 H-1 H-2 H-3 H-4 H-5 H-6 H-7 H-8 H-9 H-10 H-11 H-12 H-13 H- 14 Type of Experiment Critical experiment Subcritical measurement of reactor temperature coefficient Power determination at 1 w {nominal) Regulating rod calibration vs fuel addition Fuel system characteristics Power determination at 10 w (nominal) Regulating rod calibration vs reactor period Calibration of shim rod vs regulating rod Effect of fuel flow on reactivity L ow-power measurement of reactor temperature coefficient Adjustment of chamber position Approach to power: 10-kw run Test of off-gas system Approach to powef: 100-kw to 1-Mw runs High-power measurement of the fue! temperature coefficient High-power measurement of the reactor temperature coefficient Reactor startup on temperature coefficient Sodium temperature coefficient Effect of a dollar of reactivity High-power measurement of reactor temperature coefficient Moderator temperature coefficient Xenon run at full power Reéc_fiv'ity effecis‘ of sodium flow Xenon buildup at. one-tenth full pofier Operation at maximum power St o - e e T e T T T T TR e i RO 2. DESCRIPTION OF THE REACTOR EXPERIMENT The ARE consisted of the circulating-fuel reactor and the associated pumps, heat transfer equipment, controls, and instrumentationt required for its safe operation. A schematic arrangement of the reactor system is shown in Fig. 2.1. The major functional parts of the system are discussed briefly below. The physical plant is described in Appendix U. A detailed description of the reactor and the associated system may be found in the design and installation report.’ A summary of the design and operational data, including a detailed flow sheet of the experiment, is given in Ap- pendix B. REACTOR The reactor assembly consisted of a 2-in.-thick Inconel pressure sheli in which beryllium oxide moderator and reflector blocks were stacked around fuel tubes, reflector cooling tubes, and control assemblies. Elevation and plan sections of the reactor are shown in Figs. 2.2 and 2.3. The innermost region of the lattice assembly was the core, which was a cylinder approximately 3 ft in diameter and 3 ft fong. The beryllium oxide was machined into small hexagonal blocks which were split axially and stacked to effect the cylindrical core and reflector. Each beryliium oxide block in the core had a 1.25-in. hole drilled axially through ]Desz'gn and Installation of the Aircraft Reactor Ex- periment, ORNL-1844 (to be issued). HELIUM BLOWER ABSORBER ROD EAT EXCHANGER HEAT EXCHANGER b4 . REFLECTOR COCLANT ROD ACTUATOR its center for the passage of the fuel tubes. The outer 7.5 in. of beryllium oxide served as the reflector and was located between the pressure shell ond the cylindrical surface of the core. The reflector consisted of hexagonal beryllium oxide blocks, similar to the moderator blocks, but with 0.5-in. holes. The fuel stream was divided into six parallel circuits at the inlet fuel header, which was located above the top of the core and outside the pressure shell. These circuits each made 1] series passes through the core, starting close to the core axis, and progressing in serpentine fashion to the periphery of the core, and finally leaving the core through the bottom of the reactor. The six circuits were connected to the outlet header. Each tube was of 1.235-in.-OD seamless Inconel tubing with a 60-mil wall. The combination of parallel and series fuel passes through the core was largely the result of the need for assuring turbulent flow in a system in which the fluid properties and tube dimensions were fixed. The reflector coolant, i.e., sodium, was admitted into the pressure shell through the bottom. The sodium then passed up through the reflector tubes, bathed the inside walls of the pressure shell, filled the moderator interstices, and left from the plenum chamber ot the top of the pressure shell. The sodium, in addition to cooling the reflector and pressure shell, acted as a heat transfer medium DwG 145624 HELIUM BLOWER HEAT EXCHBANGER HEAT EXCHANGER = m 2 HELIUM Fig. 2.1. Schematic Diagram of the Aircraft Reactor Experiment. € T r g R ot e EET R e s - T s s AR 4w BB e o i e o e Saadi ik n o Nl sl i b . - " in the core by which moderator heat was readily transmitted to the fuel stream. FUEL SYSTEM The fuel was a mixture of the fluorides of sodium and zirconium, with sufficient uranium fluoride added to make the reactor critical. While the fuel ultimately employed for the experiment was the NaF-ZrF4-UF4 mixture with a composition of 53.09-40.73-6.18 mole %, respectively, most pre- liminary experimental work (i.e., pump tests, corrosion tests) employed a fuel containing some- what more UF,. The fuel was circulated around a closed loop from the pump to the reactor, to the heat exchanger, and back to the pump. An isometric drawing of the fuel system is given in Fig. 2.4. The fuel pump was a centrifugal pump with a vertical shaft and a gas seal, as shown in Fig. 2.5. DWG. 6336 REGULATING ROD ‘ ASSEMBLY T SAFETY RO | il ——————TUBE EXTENSION D d GUIDE SLEEVE j THERMAL SHIELD CAP T | THERMAL SHIELD TOP il SAFETY ROD ASSEMBLY S M; / L SHECD 107 500N RS T SN TR ! i e 7 THERMGCOUPLE 2 O“ sl ) TOP HEADER LAYOUT i %%%J CORE ASSEMBLY — RN §%§ TOP TUBE SHEET Ay n/ REFLECTOR COOLANT . o HEATERS TUBES 3 b 7 S / % 5 %%% BeO MODERATOR 4 L AND REFLECTOR FUEL TUBES é % e % | .—THERMAL SHIELD w Z 7 ASSEMBLY & PRESSURE SHELL 7N £ | 2 < ] - | 7 & i e l N N2 NRV BOTTOM TUBE A Z %/ SHEET f ] e ) ‘ ‘ - sTUD ! 7 ] %%% SUPPORT ASSEMBLY BOTTOM HEADER | Gl FUEL OUTLET T m—— 1 MANIFOLD | 7N E W | \ | NS / 4 — __THERMAL SHIELD / THERMAL SHIELD CAP N / BOTTOM N ’ / v b [aed N any £ N o ———— ||} o™ 1l = @ i j SUPPORT ASSEMBLY L 1 § SCALE IN INGHES Fig. 2.2. The Reactor (Elevation Section). - TR TETTIT T TR SpETRTIIRT O T, R T EEETIT T MRe v e T Lo I T - e ...lgfl & TUBE COIL B {INCONEL } TUSE COIL A (INCONEL) DWG. 15647 REFLECTOR EDGE BLOCK { BERYLLIUM OXIDE) P REFLEGTOR BLOCK ERYLLUM OXIDE) CORE BLOCK o (BERYLLIUM OXIDE) A REFLECTOR COOLANT l TUSES (INCONEL) CAN {INCONEL) CORE SLEEVE (INCONEL) TT—SAFETY ROD GUIDE SLEEVE (INCONEL) SCALE IN INCHES Fig. 2.3. The Reactor (Plan Section). The fuel expansion volume around the impeller cavity provided the only liquid-to-gas interface in the fuel system. While speeds up to 2000 rpm could be attained with the 15-hp d-c pump motor, the desired fuel flow of 46 gpm was attained at a speed of 1080 rpm. Although only one pump was used in the experiment, a spare fuel pump, isolated from the operating pump and in parallel with it, was provided. From the pump the fuel flowed to the reactor,’ where it was heated, then to two parallel fuel-to- helium heat exchangers, and back to the pump. The cycle time was about 47 sec at full flow, of which approximately 8 sec was the time required for the fuel to pass through the core. The two fuel-to-helium heat exchangers were each coupled to a helium-to-water heat exchanger. The heat extracted from the fuel was transferred via the helium to water, and the water — the ultimate heat sink ~ was discharged. The helium flow rate in the fuel-to-helium heat exchanger loop was controlled through a magnetic clutch that coupled the blower to a 50-hp motor. Control of the helium tlow rate in this manner permitted smooth control at any reactor power at which the heat generation was great enough for the temperature coefficient to be the controlling factor. ' o - CoatRE R e s W T Ty o P e . e R r p ’me o . AT o - - »omEE o meer S R ek el b o ek L Rl & Eh - S b k. L wek o ok Nl oo BB e & 5 3 ‘ _-MAIN FUEL PUMP STANOBY FUEL PUMP -~ o FRANGIBLE DISK VALVE U-3 . o N3 i FRANGIBLE 3 §. DISK VALVE U-2 y—ROTAMETER 1 - PRESSURE TR HEAT EXCHANGER NO.1 - | TRANSMITTERS ’ - HEAT ,/ Loy ™\ EXCHANGER - LR SN No.2 VALVE U-1 > PR - VALVE U-62 REACTOR - vaLve -5 ® . 5 vl \ /% ) -~ - ¥ ¥ !’l,/ T . - > HOT FUEL 3 % fis DUMP TANK D% : VALVE B-55 " r ~ ] VALVE B-56 ' . \CARRIER M FILL LINE f ~~ RESERVE TANK NO.{ FILL TANK NO. 2 Fig. 2.4. Isometric Drawing of Fuel System, TP T— BT g T e e o e w e KEY! £33 CROSS BARS ON LINES INDICATE WELDS. ALL FUEL PIPE LINES NUMBERED IN 400 SERIES. T e T ORNL-LR-DWG 6247 LT S e A e & DRIVE SHEAVE CLAMP RING —— BEARING HOUSING TOP SHAFT SEAL ASSEMBLY SEAL FACE RING INTERNAL SHAFT COOLING LUBE OIL GUIDE TUBE -——— SEAL FACE RING SLINGER LOWER SEAL ASSEMBLY SPACER AND HEAT DAM OIL DRAIN HEADER CONNECTION PUMP BODY ASSEMBLY THERMOCOUPLE GLAND DISCHARGE ELBOW INLET DISCHARGE GRAVITY ORNL-LR-DWG €218 SEAL Ol RETAINER OIL BLEED NIPPLE (UPPER SEAL LEAKAGE DRAIN) PRESSURE SLEEVE SPACER SLEEVE OIL SHIELD == OIL INLET B.H. BREATKER AND SIGHT GLASS GAS EQUALIZER BEARINGS FUEL INJECTION SYSTEM NOZZLE BEARING SPACER BEARING SEAL RING OVERFLOW RESISTANCE HEATER CONNECTICN O LAVA SPAGER -~ __ — RESISTANGE HEATER CONNEGTION OIL DRAIN PUMP BODY CONNECTION PUMP TANK COVER FLANGE 3/—16 NC HEX HD CAP SCREW 1-in. LONG STHELING WELL DRAIN TROUGH PUMP TANK FLOAT LEVEL INDICATOR HOUSING - PLATE g:., va__ DIP LEG Fig. 2.5. Fuel Pump, As shown in Fig. 2.4, the fuel system was connected with two fill tanks (only one of which was used) and one dump tank which had provisions for removing the afterheat from the fuel. The relatively high melting point of the circu- lating fuel (about 1000°F for the NaF-ZrF ,-UF, fuel containing 6.18% UF,) required that all equipment within which it was circulated be heated sufficiently to permit loading, unloading, and low-power operation. This heating was ac- complished by means of electrical heaters attached to all components of the fuel and sodium systems; i.e., pressure shell, heat exchanger, pumps, and tanks, as well as all fuel and sodium piping. In addition to the heaters and insulation, all fuel and sodium piping was surrounded by a 1- to ]l/2-in. annulus (inside the heaters) through which helium was circulated. The helium circulated through the annuli was monitored at various stations around the system for evidence of leaks and was, in addition, a safety factor in that it could be expected to keep hot any spots at which heater failures occurred. SODIUM SYSTEM The sodium circuit external to the reactor, shown in Fig. 2.6, was similar to that of the fuel. The sodium flowed from pump to reactor, to heat h e oeray - E‘ e i » - r o e TR e o w - & . i, & b 5 ke i b g o g Bl idl Ko T T T e i Al i b Gt . ol i i g "“MMMMM_m o i e s bl il et a2 S B B b ORNL-LR-DWG 6219 AT EXCHANGER L o NO - S ' i 4 v )« FUEL i s = = /4 PoINTS z — —k—} fRa3 T / 525 SoDIUM He 7O WATER TRY 8 / Y o HELIUM GD _— - EHE?\T EXEH 4POINTS 39 D8 3 He TO WATER I— - P o o « ——-@— WATER HEAT EXCH. MaTO He Na TO He HEAT EXCHANGERS| T |nos | & |vot | T |Noz | . HEAT EXCH. HEAT EXCH. e g o £ VENT 5[ | g | 1R |z 18 Q z Y v o e . —— AUXILLARY PROCESS LINES w = H [ 2 = £ He) TRC-3 _ ¥ . e — A—— —— AIR CONTROL LINES z > % 1_ 114 —— ——E— ELECTRICAL CONTROL LINES o & & & —— ——— — INSTRUMENT SENSING LINES » Q & = E-M FLOWMETER £-M FLOWMETER | j_‘ | | S blg T IZZZT g ] 303 J_l il ) n2 . TEMPERATURE RECORDER CONTROLLER ¥ |« i 0 LTI B I = , o g | : l b £ N 205 321 {vo) TO ORAIN I | J o) (U @ © | . TEMPERATURE INDICATOR ALARM - 8 sl | s T N b - ] TNy & o o oRan = wi wl wl i @ h T ‘y b o) =| 3 F= < <+ | !-—HYDRAULIC glg 103 TEMPERATURE INODICATOR 2 &) 2 L1l —' 22 CONTROL = w o & 2 2 5 l 6| | o 3!l 9 o 2 T 801 T w al = ol 13 8 69 by " | g (FRRY ¢ ow RECORDER ALARM ; N 1 ] @eH ‘-“ | 7Q X o X 3 P 31 \3/ (LR iquip LeveL REcoRveR @« U-44 ' - PC5 o £ sl w5 11 ‘ | ‘ WA - .flh FLOW INDICATOR ALARM 8-301 8-66 B-68 1o/ @ - o : 869 P16 ‘ C) | ‘ ¥ | \_/ He ] pC-13p-, SUPPLY HEADER | 23 () | (Da ¥ /ATRY 69 (PIAY PRESSURE INDICATOR ALARM 876 4 | 1 s ’ ALY ‘ ‘ G . ‘ _J - = g 5 (e © 1 | w o ® ‘ (ATR o 1 - PRESSURE INDICATOR s @ © B, z HOT FUEL 1 2 SODIUM FILL AND SODIUM FILL @ 3 | TANK NO. 2 OUMP TaNK | 1 LEVEL INDICATOR DUMP TANK NO. 4 He He AND DUMP TANK + i g CARRIER FILL I_fl | LIQUID LE N | L] [ex CHIQH : k P-213 ‘ | l ‘ p-241 I | 8-231 ’ |8—232 | | I fl 2 ‘ | ST y @ AMMETER ™ o ) ) | () 88 1 | i R [N | Sy | TN [ L 1] ) /amy | ! e i | | Y |B-38 | B — A\ T C TEMPERATURE DIFFERENTIAL RECORDER - - | | 11 o @ g H BT s B _ : 2 z W ‘ 5 8 | FILL AND Zjo R e sl ¥ slc o SHAFT SPEED INDICATOR S W z z DUMP LINE alg LT S 3Jla S Sjw EW why! m| K‘E " o o2 Jjo THOT FUEL 58 Jz 38 + g slg o w Z\x w - |o A E: ) |- 1 REMOVAL LINES wg wlg £ Wi [ m 7 i 2 3|y 2 2iE = o & P . IS 55 @ E XS @ RADIATION INDICATOR ALARM o o 2l& 2 ol¥ 3|z <7 o S 1o ravaTion £ o2 r ool® ol ¥ ol ol W o5 8 & o ii5 4R & - 2 MONITORS 2 = 2 3 e Flw ¥ Fla o Flwo 3 B-74 @'y o iR wlo & a = 2 -2 = it [ = 2= & ok + == & a = Q a8 3 = G wla a W a Gl 5 < @ - @ W & << o — w ciy %5 3w S w Bw 3o ¢ WIND SPEED INDICATOR ALARM ® 56 * = = = « s 7 3% 3 &y Ty [Ty BT EF X 2T BT o« \__/ CED LB302 1w T, Z PXT | PR TRANSMITTER a Q g Q QN Qa6 g O @ g g Q (PX7] PRESSURE 8 oo} 8 oo L2 °9 29 PRESSURE SWITCH 12 HELIUM BOTTLES w - B OHOOOOHEOLIHOOOLOEOOLES @608 \E/ &/ N N NN N e e B N NE NS NG NS N2 &) s ) & & \i/ & NE N\ T N NS X g wasal vawe -E==3 rrecze vawve FROM HELIUM REMOTE MANUAL & tiil supaLy AR ¢ MANUAL SPOT SWITCH p-d NORWALLY CLOSED @-(}-@ ANGLE VALVE SOLENOID VALVE PRESSURE REGULATOR ¥ INSTRUMENTS LOCATED IN BASEMENT o U= mManual DPDT SWITCH [XI NORMALLY OPEN @-(]—X REMOTE MANUAL VALVE I% RELIEF VALVE —(3) PUSH BUTTON CONTROL - [4 NORMALLY THROTTLING [b RHEOSTAT CONTROL I} CHECK vALVE -@ FRANGIBLE VALVE F . P I . ige 2.7. Process Instrumentation. T TR T VIR T TR e YT S - s st e T e T T T Ty e Tt T TSI NAT R T e O T YT - T o TR " T - i - e T TR T T e ey e " i T S o e = R R e e il 4 ™~ (’l A2} w w _m F £ e bkl a e et e o e e AR A G SARe AN, Bt s A s el - = o & @ S N @ = o w wi < © z = ~ a5 W x = w g _ b g 2 2 x AM.” 5 o %W 1 - C o ot — z = a = s — 3 3 = U g a8 - z i | @ W > & L g g S £ & T & E 2 C& 2 2 @ = g = = x o Let _ g [ x << 5 n Bm a = 2 s Q i = > @ & = Q <« o ] w » = T S o T 3 - o < - 3 A fi?\\\\\\\a\\;\m ___ , o e, § A | =l T ) /VWhr/ ‘ | ] & N A 7 o E. £, 5 N - - L - = = I I /nm__ . / &\ 1 / z > 2 o T o O C. > 8 N B, z = g w > =L S5e = 22 T Lt o« t 3 v W § 13 Fig. 2.8. lsometric Drawing of Control Rods and Nuclear Instrumentation. center of the reactor. Each rod was made up of slugs of a hot-pressed mixture of boron carbide and iron; the slugs were canned in stainless steel. The canned sections were slipped over a flexible tube. The total rod travel was 36 in., and 13 min was required for withdrawal of the rod. Thus, the 5.8% Ak/E in each rod (total for the three was approximately 17% At/E) could be withdrawn at a rate of 0.45% (Ak/k)/min per rod. A small amount of the helium used to cool the fission chambers was directed through the shim and regulating rod holes, but the helium flow was so low that not much cooling was effected. This helium was circulated in an independent circuit, known as the rod cooling system, which had two two-speed helium blowers in parallel and three helium-to-water heat exchangers. The helium circulated from the blowers through the rod sleeves and fission chamber sleeves to the heat ex- changers and back to the blowers, and it removed, in the process, up to 35 kw of reactor heat. Part of the helium returning from the reactor was diverted through pipe annuli before reaching the heat exchanger. OFF-GAS SYSTEM The off-gas system was designed to permit the collection, holdup, and controlled dischgrge of the radioactive fission products that were evolved as a consequence of reactor operation. Although only the gases above the fuel system were ex- pected to have significant activity, the gas from the sodium system was also discharged through the off-gas system. As shown in Fig. 2.9, there was, in addition to the primary off-gas system, an auxiliary system for discharging gases from the pit through nitrogen-cooled charcoal tanks. In the primary system the off gases from both the fuel and the sodium were directed to a common vent header. From this header, helium plus the volatile fission gases passed through a NaK scrubber, which removed bromine and iodine, and then into two holdup tanks, which were connected in series, to permit the decay of xenon and krypton. From these the gases were released up the stack. The release of gases to the stack was dependent 14 upon two conditions: (1) a wind velocity greater than 5 mph and (2) radioactivity of less than 0.8 pc/ecm®. The activity was sensed by a monitor which was located between the two holdup tanks. FUEL ENRICHMENT SYSTEM The fluoride fuel used in the ARE was amenablie to a convenient enrichment technique in which the fuel system was first filled with a mixture of the fluorides of sodium and zirconium. When this mixture, NaZrFs, had been circulated for a suf- ficient time to ascertain that the reactor was ready to be taken critical (no leaks, etc.), uranium in the form of molten Na,UF, was added to the fluorides then in the system. The temperature contour diagram of the NaF-ZrF -UF, system is shown in Fig. 2.10. The melting point of Na,UF was 1200°F, and there were no mixtures of higher melting points on the join between Na,UF, and NaZrF ., which is shown in Fig.2.11. The addition of Na,UF raised the melting point of the mixture only slightly above that of NaZ¢F . (955°F). It was initially intended that the enrichment operation would consist of the remote addition of Na,UF, to the system from a large tank which contained all the concentrate. From the large tank, the concentrate was to pass through an intermediate transfer tank which would transfer about 1 gt or less at a time. The intermediate tank was suspended from a load beam so that the weight of the concentrate could be determined before each addition of enriched material. However, this system was discarded when preoperational tests proved the temperature control of the system to be inadequate; in addition, the accuracy of the weight-measuring instrumentation on the transfer tank was uncertain. In lieu of the original enrichment system, a less elaborate, more direct method of concentrate ad- dition was employed. The enrichment procedure actually used involved the successive connection of numerous small concentrate containers to an intermediate transfer pot, which, in turn, was connected to the fuel system by a line which injected the concentrate into the pump tank above the liquid level. PO T e mm—" P TR T R e T R T P TR T el en Al e i e e et o et i e il At o RN, i o N oyt i AR . . RS, & o e Y, oA i i ORNL-LR- DWG 6222 PRESSURE TRANSMITTER “sTanpBY 8! ODIUM PUMP -84 FRO - SODIUM ~ M HELIUM SUPPLY - . - VAPOR TRAP fi% = 1 1 <3 & ; I-u-e3 N PRESSURE TRANSMITTER - L =1-91 T R MAIN SODIUM { , B-1 4 ‘ 0 PUMP SODIUM VAPOR TRAP - § 314 i B 5 B ) u-83 804 m SODIUM 1y QJ ‘ ' 313 SIIIIIW \@ e RN U-8t - vAPCR | | LU/ ol catcH 878 W TRap \/ BASN LINE FROM HELIM SUPPLY HEADER TO OIL. CATCH ~U-110 BASIN OFF -GAS HOLOUP TANKS L2 Q’%"" MAIN FUEL PUMP OIL CATCH BASIN ZrF, VAPOR TRAP NC.A KINNEY PUMPS DUMP TANK NO. 6 ?\ ‘ ' 809 SODIUM FILL AND SODIUM FILL AND DUMP TANK NO.5 HOT FUEL DUMP TANK SODIUM FILL AND DUMP TANK NO. 4 RESERVE TANK NO. 1 TANK NO. 3 (NOT USED) CARRIER FILL TANK NO. 2 Fig. 2.9. Isometric Drawing of Off-Gas System. 15 I I W TN ™ WY T T T W T T gy RIS DT Y TR T o TR T T S 1T ™ T R T T © T o T T T WY TRy S weewmm—mp < T akoig i o | e i e, i -, DWG. 244534 780 30 0 . 20 N o\ ?oo 675 | >’22> 625 650 S A 5 X % e S OO ANNCES NGy [ 7T A E N, /NN NN AN e e 0D A SN SN N\ e A WL NGNS A TN AN o 7 s wio Ceps O A * % ° ’ NazZrF, NapZrFy NaZrFg NagzZrgFyg Fig. 2.10. The System NaF-Z¢F .UF . 17 e e n - e e e - ¢ g T e e e v T Ry TR 675 ORNL—LR-DWG 6223 ‘. L . ’ 4/ b 650 s ~® b ./ - L~ y / o o N ® L 625 g ® ~ & & £ 600 | ) . w / : o« 3 o ® b = ¥ <——— VENT FLOOR LEVEL AUXILIARY VENT SAMPLING LINE FUEL PUMP DISCHARGE SUCTION Fig. 4.2. Equipment for Addition of Fue! Concentrate to Fuel System, TABLE 4.1. FUEL CONCENTRATE BATCHES ADDED DURING CRITICAL EXPERIMENT Batch Concentrate Added Uranium Added Weight of U235 Added Date Time Can —_— No. g Ib wt % g g b 10/30 1625 A-6 13,609 30.002 59.548 8104 7569 16.687 1N 1415 A-7 11,510 25,375 59.513 6850 6398 14,105 1804 A-8 13,852 30.538 59.530 8246 7702 16.980 2203 A-9 13,806 30.437 59.587 8227 7684 16.940 11/2 0213 A0 13,638 30.066 59.454 8108 7573 16,695 - 0610 A-12 13,241 29.191 59,531 7882 7362 16.230 0831 A-5 8,221 18.124 59.637 4903 4579 10.095 11/3 0523 A-11 4,505 9.932 59.671 2688 2511 5.536 0941 B-22 6,432 14,180 59.702 3840 3587 7.908 1245 B.31 6,312 13.915 59.637 3764 3516 7.751 1536 B-20 4,567 10.068 59.529 2719 2540 5.600 27 e ————— v R T P T T —T e WY AT e T = B d A st o Fig. 4.3. Chemists Preparing for Enrichment Operation During Critical Experiment, TR ———— I B X o p-ray oy R e mr oo R R porre e R e P e v F?WW‘ ooy, e e T WE o g g RS R U FEE AP R S gm0 ET D gty T e P ol e e Trr e g g il kBN e i Bk L e i, e B R . R L 4wt g kb b i, e PN R, el . g -y bbbk, A MR R R, M kS o ki g . TRt time the volume of carrier in the fuel system was calculated to be 4.82 %, and the weight was 927 ib. At 1507, concentrate transfer was started from batch can A-6 to the intermediate tank, and, at 1539, transfer of the first 5.5 Ib of fuel to the system was accomplished. At 1554, the second 5.51b of fuel had been added. These fuel injections very noticeably affected the fission chamber re- corders. Mixing of the concentrate with the carrier did not occur rapidly, and therefore each time the enriched slug entered the reactor it produced a multiplication that was observable on the fission chamber count-rate recorder. Figure 4.4 is a photo- graph of the trace of fission chamber No, 1; the pips that occurred during addition of the first ond second slugs of fuel are readily observable. From the time interval between pips, which was almost COUNTING RATE (arbitrary scate) 2The first five samples taken prior to this time were for carrier impurity analysis. These five analyses were discussed in chap. 3, **Prenuclear Operation,”’ TRACE OF FISSION CHAMBER NO. § RECORDER OCT. 30,1954 - — = * exactly 2 min, and the volume of the fuel system, a check on the rate of fuel flow was obtained: 4.82 (%) x 7.48 (gal/#) f {gpm) = 3 {min) = 18.03 . This checked fairly well with the value of about 20 gpm read from the fuel flow recorder. The remaining four injections from can A-6 were accomplished smoothly. The pumps were speeded up to an observed fuel flow rate of 46 gpm to obtain better mixing of the fuel. At 1720, fuel sample 6 was removed for analysis.? At 1920 the first 5.5-lb injection from can A-7 was started, but the transfer line from the inter- mediate pot to the pump clogged due to concentrate freezing in it. After several hours of unsuccessful attempts to free the line, a gas leak occurred in the transter line at the intermediate tank pot, and ORNL~LR-DWG 3853 PUMP AT MINIMUM SPEED SECOND SLUG THROUGH SECOND TIME SECOND SLUG THROUGH FIRST TIME ! i e U'FIRST SLUG THROUGH SECOND TIME & ,5 oL ul o z g ] w 45 Lz u- —e————— TIME Fig. 4.4. Passage of Enriched Slugs Through Reactor, 29 ¢ r S T T e it was then decided to install a new transfer line and transfer pot. At this time it was reported that no transfer of concentrate had been made to the pump from can A-7. The first 5.5 Ib of can A-7 was lost to the experiment through the leaks and in the plugged line. At 2000 on November 1, the new transfer line and tank were installed and checked out ready for use. At 2006, fuel sample 7 was drawn for analysis, and, at 2300, the critical experiment was resumed. When the first 5.5-Ib slug was injected, periodic pips were again observed on the fission chamber recorder. These pips had a period of 47 sec which, together with a pump speed of 1080 rpm, yielded a calculated fuel flow of 46 gpm, which corre- sponded to the flow observed on the fuel flowmeter. At 2315, while attempting the next 5.5-lb transfer, the line again froze. The line was disconnected and found to be plugged at the injection fitting through the pump flange. |t was therefore decided to increase the current to the resistance-heated fitting and to add a separate vent line from the pump so that the chemists could continue to *‘blow through’’ the transfer line and out the new vent line without raising the system pressure. By 1320 on November 1 the new vent line had been instalied and the transfer system was again in operating condition. The transfer of the re- maining 20 |b of concentrate in four batches from can A-7 took place with no difficuity, and the transfer was completed by 1415. At 1617, fuel sample 8 was taken for analysis, and, at 1707, the first batch from can A-8 was transferred into the system. The remaining 5 batches were trans- ferred smoothly, and the transfer from can A-8 was completed by 1804. At 2028, fuel sample 9 was removed for analysis. The next three additions of fuel from cans A-9, A-10, and A-12 were accom- plished with little difficulty. At 0830 on November 2, while transferring the first 5-1b batch from can A-5, a leck occurred in the injection line just below the floor level of the loading station {see Fig. 4.3). Examination of the injection line showed the leak to have occurred at a Swagelok fitting. Approximately 0.2 Ib of concen- trate was lost from the experiment as a result of the leak. Since the line had clogged, it was neces- sary to install a new section of line. During the time the repairs were being made to the injection system, the original schedule was changed, and a subcritical measurement of the temperature coef- ficient of reactivity was made. In addition, samples 10 and 11 were taken for analysis. 30 SUBCRITICAL MEASUREMENT OF THE REACTOR TEMPERATURE COEFFICIENT Approximately 100 Ib of U?3° (180 Ib of Na,UF ) had been added at the time the leak occurred on November 2, at 0831, and it was estimated that this was about 80% of the total fuel needed. Al- though it had been planned to make a preliminary measurement of the temperature coefficient of reactivity with about 90% of the fuel added (see “*“Nuclear Operating Procedures,’”’ Appendix D), the plans were revised and the measurement was made at this time. At the start of the experiment the reactor mean temperature was 1306°F. The fuel heat exchanger barrier doors were raised (at 1003); two minutes later the fuel helium blower was started and its speed increased to 275 rpm. The resultant cooling, with time, as traced by the reactor fuel mean temperature recorder, is shown in Fig. 4.5, with the counting rate from fission chamber No. 2 super- imposed. The data for both fission chambers are given in Table 4.2. The counting rate of the fission chambers monitors the neutron flux. The counting rate was observed to rise with decreasing temperature and thus indicated a nega- tive temperature coefficient of reactivity, Counts were taken for 40 sec during every minute. At 1010, when the reactor temperature reached 1250°F, the helium blower was stopped, and soon thereafter the reactor temperature began to rise again, with a corresponding decrease in counting rate. ORNL-LR-DWG 6396 1400 . —I 'L.‘:‘ i i 1 lmI— & Z .~ x| £ % |08 <8 e = i 5&. 050, 2 5 P Fig S 1350 |-—- et G Bt SN E e = m < = @ o ' T — o Mmoo 8 o 5 -1 800 o % + | 3 Z 1300 [ —tobe et R N —— - 780 £ 8 g b EE — = = 760 O 5 Led gfl: z -1 740 @ 2 g 1250 (-1~ e = e — 720 5 g o 2 S o 1200 1006 —— 1004 1002 ® e ¥ w9 e 2 & g 2 TiME OF DAY Fig. 4.5. Subcritical Measurement of the Reactor Temperature Coefficient. (Plot of fission chamber No. 2 counting rate superimposed on fuel mean temperature chart.) Lo e pre T e e et gy F P e v ¥ Creem o B - vt R e prmr Y o T T [ e ot o W T L e 1 EEreRTY i e K . - Tt b o b B LA B o, e Rk we molbgle s i, il s TABLE 4.2, SUBCRITICAL MEASUREMENT OF THE REACTOR TEMPERATURE COEFFICIENT Fission Fission * No. 1 No. 2 ) (counts/sec) (counts/sec) 1004 1306 486.4 723 1005 1304 492.8 736 1006 1302 512.0 755 1007 1290 531.2 780.8 1008 1275 537.6 793.6 1009 1260 537.6 793.6 1010 1252 544,0 800.0 1011 1244 531.2 787.2 1012 1240 531.2 774.4 1013 1246 518.4 768.0 1014 1252 518.4 768.0 1015 1258 512 761.6 1154 1275 502 736.0 1319 1284 490.9 725.0 1422 1290 484 713.0 The reactor temperature coefficient, as estimated from the data presented in Fig. 4.5, was of the order of ~5 x 10~3 (Ak/k)/°F. The data were adequate to show that the reactor would be easy to control; therefcre the experiment was allowed to proceed. |t will subsequently be observed that the temperature data recorded in the data room did not exhibit the full temperature drop across the reactor in any run where heat was being abstracted from the system. Consequently where the ab- stracted heat is one of the parameters of the experiment, as in the above temperature coefficient measurement, it would be expected that some temperature correction would be in order. The existing data for the case in question are, how- ever, too meager to justify correction, although based on correlations (app. K) with data taken during the low- and the high-power runs, the con- trol room temperature indications were lower than the actual temperature, and therefore the estimated temperature coefficient given above is too high. It may aiso be noted from examination of Fig. 4.5 that the maximum counting rate and the temper- ature minimum did not occur simultaneously. A- lag of about 2.5 min in the response of the reactor temperature was observed. The reason for this time lag is not well understood, but it may have some connection with the fact that the thermo- couples were outside the reactor and hence did not ‘‘see’’ the changes immediately. Details of this and later measurements of the fuel and reactor temperature coefficients are given in Appendix O. APPROACH TO CRITICALITY At 0130 on November 3, the injection system was again in operation, and the fuel additions during . the remainder of the critical experiment occurred without mishap. The remaining portions of can A-7 and all of cans B-22, B-31, and B-20 were added during this time. Throughout the course of the experiment, the progress toward criticality was observed on the neutron detectors. With every fuel injection the counting rates of the two fission chambers and the BF, counter were simultaneously clocked and recorded. The approach to criticality could readily be seen by plotting the reactivity, &, against the concentration of the fuel in the system. The re- activity was obtained from the relationship 1 k=1-—, M where M is the subcritical multiplication, which is determined from the expression where N is the counting rate after a fuel injection and Ny is the initial counting rate. As criticality is approached, 1/M approaches zero, and therefore at criticality, & = 1. Figure 4.6 shows a plot of & vs U235 concen- tration, where & is determined from three neutron detectors, i.e., two fission chambers and o BF3 counter. The reactivity, as observed on the two fission chambers, showed a rapid increase during the early stages of the experiment and then leveled off as the critical condition was near. The BF, counter, on the other hand, showed a more uniform approach to criticality. The differences in the responses of the two types of detectors are dis- cussed in Appendix E. Table 4.3 presents the data from which Fig. 4.6 was drawn. A condensed, running uranium inventory during the critical experiment is given in Table 4.4, The 31 ORNL-LR-DWG 3852B -~ | | FISSION CHAMBER NO. 2 -~ EXP. E-f, NOV. 3 , | b, 0.9 //A , FISSION CHAMBER NO. { i // V8 1.0 0.8 7’ / ' // 0.98 0.7 /. e e e e 7 r; /f ~ > / * ~ / o 0.96 — ; 0.6 - £ g : gg ) —-BF5 COUNTER @ : 0.94 e 05 0.92 _ 0.4 / A 16 i8 20 22 U235 CONCENTRATION IN FUEL (In/f3) 03 ——- S 0.2 0 4 8 12 16 20 24 28 32 36 40 U235 CONCENTRATION IN FUEL (ib/ft3) Fig. 4.6. Approach to Criticality: Reactivity vs Fuel Concentration. figures of columns 4, 6, and 11 were supplied by the chemists.® At 1545 on November 3 upon the addition of the second batch of fuel from can B-20, a sustaining chain reaction was attained - the reactor was critical. Figure 4.7 shows a photo of the control room just as the critical condition was reached. At 1547 the reactor was given over to the servo mechanism at an estimated power of 1 watt, and during the next ¥ hr a brief radiation survey of the reactor and heat exchanger pits was made (cf., chap. 5, “Low-Power Experiments’”). At 1604 the reactor was shut down, and thus the initial phase of the ARE operation program was completed. 3). P. Blekely, Uranium in the ARE, ORNL CF-55- 1-43 (Jan. 7, 1955). 32 The uranium inventory (Table 4.4) showed that the critical concentration of U232 was 23.94 b/, which corresponded to a total weight of 133.8 Ib of U233 in the fuel system. At this time the total weight of the fuel was 1156 Ib, and its volume was 5.587 #°. The per cent by weight of U?3% was calculated to be 11.57. The critical mass of in the reactor was calculated from the ratio of volumes: U235 Vv 7 M =M . r 5 V 5 where M, = critical mass of U235 in reactor core, M, = mass of U233 in system, V = critical volume of reactor, NPT VR & WEITTT T e b T TERTEON T e, TR M e o ey * o B ETr NS e wTRCMmcR MR g Meow SRR o FETY T ry T e, (TSN S 7 TR e A W i, . CebiRe Ra R ~ i ke i Ca el e sgieha. Yy L [ - TABLE 4.3. APPROACH TO CRITICALITY: REACTIVITY FROM VARIQUS NEUTRON DETECTORS vs FUEL CONCENTRATION Fission Chamber No. 1 Fission Chambet No. 2 BF, Counter I&:i Concli::rf;fion Counting Multiplication, R . Counting Multiplication, R . Counting Multiplication, R o No. (b/f%) Rate, N M ecc:vlty, Rate, N M eacway, Rate, N M eqclzlvufy, {counts/sec} (N/Ng) {counts/sec} (N/Ng) (counts/sec) {N/NG) 1 0 9.46* 1 0 16.28* i 0 4.49* 1 0 2 3.39 26.7 2.82 0.645 42,5 2,61 0.617 7.04 1.57 0.381 3 6.16 47.44 5.02 0,801 71.54 4.39 0.772 9.19 2.05 0.510 4 9.37 86,96 9.19 0.891 128.5 7.90 0.873 13.43 2.99 0.667 5 12.45 147.2 15.6 0.936 217.2 13.34 0.925 19.73 4,39 0.780 6 15.36 251.9 26.6 0.962 367.6 22,6 0.956 29.5 6.57 0.848 7 18.09 450.9 47.6 0.97% 649.9 39.9 0.975 49,0 10.3 0.908 8 19.74 648.2 68.5 0.985 977.4 60,0 0.983 70.5 15.7 0.936 9 20.63 859.3 90.8 0.989 1317 80.9 0.988 93.7 20.9 0.952 10 21.88 1489 157.4 0.994 2316 142.3 0.993 161.3 35.9 0.972 n 23.08 4028 425.6 0.998 6730 413.5 0.998 442.0 98.5 0.990 12 23,94 Critical *Initicl counting rate, N, TABLE 4.4. URANIUM INVENTORY DURING CRITICAL EXPERIMENT Samples Removed Fuel Mixture Uranium-235 Run Fuel Concentrate Added for Analysis Total Weight ol . Total Date Time No. Weight Yolume Fuel y23s Concentrate Tm Density v’i:'jh; Weight Cencentration (Ib) (£t3) Removed Removed Plus Carrier o"éme (tb/Ft3) 8¢ in System |b/f% Wt % (1b) (1) (Ib) (#) () (i) 10/30 1425 1 927.3 4.820 1924 0 1625 2 30.00 0.1046 957.3 4,925 194.4 16.687 16.687 3.388 1.743 1740 2.447 0.0427 954.8 4,912 16.644 10/31 2013 253 0.0441 952.3 4,900 16.600 1171 1415 3 25.37 0.0885 $77.7 4.988 196.0 14,105 30.705 6.156 3.141 1635 2.286 0.0716 977.5 4.976 30.633 1804 4 30.54 0.1065 1008 5,082 198.3 14,980 47.613 9.368 4.723 2040 2.670 0.1261 1005 5.069 47.487 2203 5 30.44 0.1062 1036 5175 200.2 16.940 64.427 12,45 6.220 12 0213 6 30.07 0.1049 1066 5.280 2019 16.695 81.122 1536 7.610 0610 7 29.19 0.1018 1095 5.382 203.5 16,230 97.352 18.09 8.890 0831 8a 5.50 0.0192 1101 5.401 203.8 3.058 100.410 1859 9.123 1625 2.562 0.2337 1098 5.388 100.176 2025 2.716 02478 1095 5.375 99.928 /3 0256 85 1262 0.0440 ’ 1108 5.419 2045 7.037 106965 19.74 9.654 0523 9 9.93 0.0346 1118 5.454 205.0 5,536 112,501 20,63 10.06 0941 10 14,18 0,0495 1132 5503 205.7 7.908 120,409 21.88 10.64 1245 W 13.92 0.0486 1146 5.552 2064 7751 128,160 23.08 11.18 1536 12 10.07 0.0351 1156 5.587 206.9 5.600 133.760 23,94 1157 1545 Criticality reached 7 7 /4 0910 2.679 0.3100 1153 5.574 133.450 0915 2718 0.3145 15 5.561 133,135 1310 2.482 0.2872 1148 5.549 132,848 1315 0.732 0.0847 1147 5.545 132,763 33 TR T B g < e ik b 34 R e e R icality, Control Room at Cri ig. 4.7. F mr vvos " g s by - T W ATIARENTERT R B Mo el v @ T FET L t o PR SRR BT e FT e peert nnpe e e o m———— ™y "o y T TR R ee FEER V_ = system volume. Therefore 1.37 #2 M, = 133.81b [—— | =3281ib . 5.587 f+° The calculated, cold, clean critical mass of the reactor, as obtained from the subsequent rod cali- brations, was 32.75 Ib (cf., app. F). The reactor was not instrumented to permit the measurement of the flux or power distributions through the reactor, but measurements of these distributions were made on a critical mockup of the reactor at the ORNL Critical Experiment Fa- cility about two years before the operation of the ARE.? These measurements represent the best 4p. Callihan and D. Scott, Preliminary Critical As- sembly for the Aircraft Reactor Experiment, ORNL.-1634 (Oct. 28, 1953). information that is available and, because of the general interest therein, typical axial and radial flux and power distribution curves are given in Appendix G. ANALYSES OF FUEL SAMPLES in addition to the fuel samples taken, as noted, during the critical experiment, four more samples were removed on November 4 after initial criticality was reached. A list of all samples taken and the results of the chemical analyses are presented in Table 4.5. Besides showing the analyses of the percentages of U and U?3% by weight in the fuel system and a comparison with the U233 (wt %) content obtained from the criticality data, the table also lists the results of the analyses for the impurities and corrosion products Fe, Cr, and Ni to provide information on the purity of the fuel and the corrosion rate of the fuel system. TABLE 4.5. CHEMICAL ANALYSES OF FUEL SAMPLES U235 ¢ 235 vure Tim, Sample Impurities and Corrosion Products (ppm) UT::L'M Chemical _from Critical No. Cr Fe Ni (wt %) Analysis Experiment Data (wt %) (wt %) 10/25 1650 1 90 200 50 10/26 1414 2 81 <5 40 1911 3 81 <5 6 10/27 1919 4 90 18 12 10/29 0935 5 102 30 20 10/30 1740 6 100 20 10 2.47 + 0.01 2.31 1.74 2013 7 150 15 10 1.84 * 0.04 1.72 1.74 11/1 1635 8 190 15 15 3.45 * 0.01 3.22 3.14 2040 9 200 10 17 5.43 * 0,01 5.07 4.72 /2 1625 10 210 9.58 + 0.08 8.95 9.12 2025 n 205 20 20 9.54 + 0,08 8.91 9.12 1174 0910 12 310 | 12,11 20,10 11,32 11.57 0915 13 300 12,21 £ 0,12 11.41 11,57 1310 14 320 12.27 £ 0,08 11.46 11.57 1315 15 310 | 12,24 £ 0,12 11.43 11.57 1/5 1100 16 372 5 <5 1254 £ 0.07 1172 11.72 11/6 17 f‘ 1257 2007 1175 11.79 3 0535 18 420 12.59 £ 0,12 11.77 11.79 11/7 0423 19 445 13.59 + 0.08 12.70 12.38 35 3 13 . it . A 4 :i The first five samples were removed from the system prior to enrichment, and samples 16 through 19 were taken after the critical experiment and during the low power runs at the time of the cali- bration of the regulating rods against fuel addition. For the most part, the U233 chemical analysis agreed to within a few per cent with that obtained from the running inventory. When the sample for analysis was withdrawn from the system too soon after a fuel addition, there was a tendency for the analysis to be low, un- doubtedly, because of inadequate mixing of the additive with the bulk of the fuel. The chemical analysis of sample 1, however, was obviously in error, whereas that of sample 9 gave a uranium content that was about 7% higher than the inventory showed. The other sample analysis figures were within about 2% of the percentages given by the inventory. The analyses of samples 12, 13, and 14, which were taken after the critical experiment, agreed to within about 1% with the inventory calcu- fation. ' The increase in the buildup of chromium in the fuel system with time is shown in Fig. 4.8 to give an indication of the corrosion rate of the system. The initial corrosion rate was quite small, but after enriched fuel had been added to the system, the analyses showed a chromium content increase of about 50 ppm/day. content as given by CALIBRATION OF THE SHIM RODS As outlined in the ‘‘Nuclear Operating Pro- cedures,”” Appendix D, during the first fuel ad- ditions the shim rods were withdrawn all the way out of the reactor to obtain the multiplication. After about 100 Ib of U%%® had been added to the system, the procedure was altered in order to obtain a shim rod calibration. After each fuel addition, all three rods were simultaneously with- drawn to positions of 20, 25, 30, and 35 in. out; total movement was 36 in. The counting rate of each of the neutron detectors was recorded for each rod position. Figure 4.9 shows the reciprocal multiplication and the reactivity as a function of U?35 content of the fuel system for the various rod positions. The data ysed for plotting Fig. 4.9 were obtained from the BF3 counter and are pre- sented in Table 4.6. A cross-plot of reactivity vs shim rod position determined the rod calibration detail s, which are given in Appendix J. The rods 36 ORNL-LR-DWG 63387 500 . = o . — o 12.4 % £ = 2 s 400 O .8 z O O g Hr 2 [ < H = 4 w w! e £ E 300 2 = o 5 2 2 2 200 [ > 51 © = Eo 3y @ o g dom & < 17 °4 X o g © 00 *D_: f QO 0 25 28 34| 4 4 8 OCT, = NOV. Fig. 4.8. Increase in Chromium Concentration in Fuel as a Function of Time. were found to be nonlinear, with (Ak/k)/in. and total Ak/k varying in the manner shown in Appen- dix D. From the calibration it was found that each rod was worth a total of about 5.8% Ak/k. As a check on the general shape of the reactivity curves of the shim rods, a series of counts were taken on the BF3 counter and the fission chambers for various rod positions for two different uvranium concentrations. These data are also discussed in Appendix J. MEASUREMENT OF THE REACTIVITY-MASS RATIO Prior to the ARE operation it had been estimated that when the critical mass (assumed to be 30 Ib) was reached, the value of the ratio (Ak/R)/(AM/M) should be 0.232. This valve was obtained from a calculation of the ratio for various amounts of U233, as shown in Fig. D.1 of Appendix D. From the data taken during the critical experiment, it was possible to establish an experimental curve for the ratio and to verify the value given above for the ratio at the critical mass. In order to find (Ak/k)/(AM/M) experimentally, use was made of the curve of Fig. 4.9 which gives E in terms of the U233 content of the fuel in the system for various shim rod positions; i.e., for any Ak/k (/_\.k M < AM/M \NAM/ \&/ ' which is the reactivity-mass ratio. The value point, v poer mops s g TABLE 4.6. REACTIVITY vs U235 CONTENT OF FUEL SYSTEM EOR VARIOUS SHIM ROD POSITIONS E xperiment u23s5 Shim Rod Counting Rate, E-1 in System Positions™ BF3 Counter NO/N k=1-— (NO/N) Run No. (1b) (in. out) {counts/sec) 8 A 100.41 20 32.0 0.140 0,860 25 38.4 0.117 0.883 30 46.9 0.0956 0.904 35 54,3 0.082 0.918 8B 103.13 20 34.1 0.131 0.869 25 40.5 0.111 0.889 30 51.2 0.0877 0.912 35 57.8 0.0774 0.923 8 C 106.18 20 36.3 0.123 0.877 25 46.9 0.0954 0,905 30 57.8 0.0774 0.923 35 66,1 0,0676 0.932 8 D 106.97 20 36.3 0.123 0.877 25 46,9 0.0954 0.905 30 59.7 0.0749 0.925 35 68.3 0.0655 0,935 9 A 110,14 20 38.4 0.117 0.883 25 51,2 0.0874 0.913 30 68.3 0.0655 0,934 35 83.2 0.0538 0.946 9B 112,50 20 40.5 0,111 0.889 25 55.5 0.0808 0.919 30 76,8 0,0582 0.942 35 93.9 0.0479 0.952 10 A 115,58 20 44.8 0.0998 0.900 25 59.7 0.0749 0.925 30 83.2 0.0538 0.946 35 110.9 0.0403 0.960 10 B 118.63 20 42.7 0.105 0.895 25 68,3 0.0655 0.934 30 93,3 0.0479 0.952 35 136.5 0,0327 0.967 0C 120.41 20 51,2 0.0873 0.913 ‘ 25 68.3 0.0655 0.934 30 110.9 0,0403 0,960 35 162,1 0.0276 0,972 1A 123.43 20 51.2 0.0873 0.913 25 85,3 0.0524 0.948 30 128.0 0,0349 0,965 35 - 213.3 0.0209 0.979 *Position of all three shim rods. 37 Y i : | TABLE 4.6 (continued) Experiment u23s Shim Rod Counting Rate, E-1 in System Positions* BFS Counter Ng/N k=1-—- (NO/N) Run No. (Ib) (in. out) (counts/sec) 11 B 126.48 20 59.7 0.0749 0.925 25 93.9 0.0476 0.952 30 162.1 0.0275 0.973 35 307.2 0.0145 0.985 11cC 128,16 20 59.7 0.0749 0.925 25 93.9 0.0476 0.952 30 196.3 0.0228 0.977 35 443.7 0.0101 0.990 12 A 131.08 20 59.8 0.0748 10,925 25 110.9 0.0403 0.960 30 256.0 0.0175 0.983 35 955.7 0.00468 0,995 12B 133,76 20 68.3 0.0655 0,935 25 136.5 0.0327 0.967 30 375.5 0.0119 0.988 35 Critical *Position of all three shim rods. RUN NUMBER ORNL—LR—-DWG 6398 8A 88 8C SD 9B {0A 10B 10C 11A B 14C 12A 12B — J e 0.6 ' l | | | ! 0 | l’l ioolS l\‘\’ ?lss'm-__,____-L 1 N EXPERIMENT E-1, NOV 3, 1954 - s\—\“\gR :I;rl . ~ BF, COUNTER DATA —T = 014 //..“/ l . 100 0.g8 \O -8 e ,’-f'.’. ODS AT = |_le=—Toum R l L. < o012 .t l - 096 - onS AT F’ ‘[“‘ = e R x S 040 ] 9"‘]‘ML | ow. |- 094 - o ol o) |SHME ] S - i - Qo8 sl 092 é ® //f/ OSH - — 5 = f M ROpg AT 20‘;"‘--.3~_. @ g 006 Tl SHIM| e 0.90 5 004 ol | ‘? "“r‘ n__~___ 088 3 ol LTl s g — | ‘ T - ‘(:)“ES AT 30in ' 002 SOLID LINES SHOW REACTIVITY - Ol | SHiMg F?oo?,fi’"*i% 086 e — T 35 . I DASHED LINES SHOW RECIPROCAL MULTIPLICATION n. 1 1 L] i ? L1 ll_l_ ] - 1 Ull_ ! JL 1 llo ] ’ TNI\T'?"'TE-- 0.84 0 100 102 104 106 108 110 12 144 416 418 120 422 124 126 128 130 432 134 136 >IN FUEL SYSTEM {ib) Fig. 4.9. Reactivity and Reciprocal Multiplication vs U235 Content of Fuel System for Yarious Shim Rod Positions, 38 » ! R T B ¢ i T T R T E e W T IR O R e s TE R PR R CURE T P T ¥ v B g R L A L L A ORNL~LR~DWG 6399 (AR/AM) is the slope of the & vs M curves shown in Fig. 4.9. For various small sections of the curves the slopes approximated straight lines. From the average values of M and & over these increments of the curves and the measured slopes, 0.35 g 0.30 the ratio R was determined for each increment. . 3 Figure 4.10 shows a plot of R vs the U23% content > of the fuel in the system. 3 < 0.25 The value of R for the experimental critical mass (133.8 ib) was calculated to be Ak/k AM/M for shim rod positions of 25, 30, and 35 in. Since e e e el s s, e e = 0.236 0.20 25 26 27 28 29 30 3 32 33 34 235 MASS U IN REACTOR (10) this value (0.236) was in excellent agreement with : the calculated value (0.232), it was used through- Fig. 4.10. Reactivity-Mass Ratic as a Function out the experiment. The data are tabulated in of the U235 Content of the Fuel in the Reactor, Table 4.7. TABLE 4.7. REACTIVITY-MASS RATIO AS A FUNCTION OF THE AMOUNT OF U235 IN THE REACTOR FOR VARIOUS SHIM ROD POSITIONS Reactivity-Mass Mass Shim Rod M, Mass of . ' . Ratio, ® Range* Position** U235 i Reactor An Reactivity, Ak AL/k (1b) (in.) (1b) (1b) k AM/M - 100 10 106 25 25,24 1.47 0,891 0.0172 0.331 30 25.24 1.47 0.913 0.0175 0.329 35 25,24 1.47 0.924 0.0174 0.322 ~_ | Average 0,327 i 114 +o 120 25 ) | 28,66 1.47 0,929 0.0137 0.287 E 30 28,66 - 1.47 0.951 0.0136 - 0.279 35 28.66 1.47 0.963 0.0135 0,273 Average 0,280 ] 128 t0 134 25 32,10 44 0,961 6.0110 0,257 ' : 3O 32,10 | 1.49 0,982 0.0100 0.219 35 : - 32,10 1.40 0.995 0.0100 0,231 | 7 . Average 0,236 *These rfiass rarfgeg refer to the curves of Flg 4.9 on .\n'.?hich these calculations are based. | . **Position of all three rods. | E P T 39 5. LOW-POWER EXPERIMENTS The reactor was operated, after initial criticality was attained, for about 20 min at a nominal power of 1 w and then shut down. The next morning a series of low-power experiments was started. This series of experiments lasted from the morning of November 4 until the morning of November 8. The chronology of this low-power operation is given in Fig. 5.1. Two of the early experiments, L-1 and L.-4, were devoted to a determination of the reactor power from measurements of the fuel activation. These runs were each of 1 hr duration with the reactor at a nominal power of 1 and 10 w, respectively. After each of these runs, fuel samples were taken and the activity counts were made from which the reactor power was determined., The method was inaccurate, as subsequently evidenced by power calibrations from the extracted power, but did indicate that the nominal power estimates were low, Radiation surveys of the entire experimental system were made during both the nominal 1- and [0-w runs except that a survey of the reactor pit was not attainable during the 10-w run because the pit was sealed at that time. Most of the equip- ment and many of the instruments were, however, located in the heat exchanger pits to which access was maintained throughout the low-power operation. Most of the time of the low-power operation was devoted to calibration of the regulating and shim rods. The two essentially independent methods used to calibrate the regulating rods were cali- bration against fuel addition and calibration against reactor periods by using the inhour equation (app. I}). The shim rods were then calibrated against the regulating rod. Also as an integral part of the low-power opera- tion the earlier measurement of the temperature coefficient was checked, and the fuel system per- formance characteristics were determined. Finally, in preparation for the high-power experiments, the ion chambers, which had been positioned close to the reactor during the critical and low-power ex- periments for greater sensitivity, were withdrawn so that they would register sensibly at the higher powers, POWER DETERMINATION FROM FUEL ACTIVATION Reactor power measurements are usually made by exposing gold or indium foils to the neutron 40 flux in a reactor soon after initial criticality is attained. In the ARE it was simpler to draw off a sample of the uranium-bearing fuel after operation and measure its activity with an ion chamber. Com- parison with a similar sample of known activity gave a determination of the flux level and, hence, the power level. The procedure is described in Appendix H. In run L-1, the reactor was operated for 1 hr at an estimated power level of 1 w, and a sample was drawn off for measurement of the activity. The specific activity was too low for a reliable de- termination and, accordingly, the experiment was repeated in run L-4 at a nominal 10-w power level, It was subsequently found from the heat balance at high power that the actual power was 27w during run L-4, However, the fuel sample activity was: only on the order of one-half to two-thirds the activity to be expected from operation at 27 w. Apparently a considerable amount of the gasecus fission products was being given off from the fuel. A comparison of the data obtcined from fuel acti- vation with those obtained from heat balances is given in Appendix N, RADIATION SURVEYS The primary purpose of experiments L-1 and L-4 was to calibrate the estimated reactor power against the power determined from the radioactivity of fuel samples, but, at the same time, radiation dose levels were measured at various locations around the reactor and the tank and heat exchanger pits. These data will subsequently be of interest in evaluating the radiation damage to various com- ponents of the system. Most of the radiation measurements were made with a “‘Cutie-Pie’’ (a gamma-sensitive ionization chamber), but a GM survey meter, a methane pro- portional counter, an electroscope, and a boron- coated electroscope were used for some measure- ments, in particular, those made close to the reactor. While the GM survey meter only provided a check on the gamma dose as measured by the ‘‘Cutie-Pie,” the other instruments provided meas- urements of the fast- and thermal-neutron doses. The proportional counter measured the fast-neutron dose, and the thermal-neutron flux was calculated from the difference of the two electroscope readings. I PR T T TIME OF DAY T 0000 0300 0600 Q900 1200 1500 1800 2100 2400 M .o, Raiaieiii u., it e e NOVEMBER 4, 1954 L] I SAMPLES 12 AND 13 REMOVED FOR ANALYSIS REACTOR MADE CRITICAL AND POWER INCREASED TO 1t w REACTOR AT~! w FOR DETERMINATION OF POWER LEVEL SAMPLES 14 AND 15 REMOVED FOR ANALYSIS START OF CALIBRATION OF REGULATING ROD FROM FUEL ADDITION RUN O PENGUIN NO. 11 INJECTED} RUN 1} REACTOR AT 1 w EXP. PENGUIN NO. 19 INJECTED RUN 2 REACTOR AT 1 w REGULATING ROD FOUND TO 8E TOO LIGHT; CHANGED TC ROD WITH MORE kex FUEL SYSTEM CHARACTERISTICS RUN; EXP, L-3 } EXP. L-1 L-2 CHRONOLOGY OF THE LOW-POWER EXPERIMENTS NOVEMBER 5, 1954 o~ :_/ REACTOR AT 1 w o PENG UIN NO. 10 INJECTED; GAS RUN 11| - LEAK IN INJECTION SYSTEM > - DEVELOPED 5 REACTOR AT | w PENGUIN NO. 4 INJECTED GAS LEAK REPAIRED REACTOR AT 1 w RUN 34 REACTOR AT 1 w PENGUIN FOUND TO HAVE NO DIP LINE; NO FUEL INJECTED POWER INCREASED REACTOR AT 1 w REACTOR AT 10 w; 1-hr RUN FOR POWER PENGUIN NO. 14 INJECTED DETERMINATION; RADIATION SURVEY © (NJECTED + RUN 38 o MADE; EXP. L-4 REACTOR AT 1 w o REACTOR AT 1 w L~ SAMPLE 17 REMOVED FOR AMALYSIS o PENGUIN NO. 16 INJECTED b RUN 4 % SAMPLE 18 REMOVED FOR ANALYSIS —\ REACTOR AT 1 w — - REACTOR AT | w FOR RUN 1 OF EXP. L-5: 1 REACTOR AT ‘”] ROD CALIBRATION AGAINST PILE PERIOD PENGUIN NO. 13 INJECTED r RUN 5 PERIGD EXCURSION TO ~100 w; RUN 2 REACTOR AT 1 w REACTOR AT T w F—1— INJECTION LINE PLUGGED APPARENTLY “ PENGUIN NO., 5 INJECTED » RUN 6 AT FITTING INTO PUMP o REACTOR AT 1 w p ] REACTOR AT 1 w - > DURING RUN 6, HIGH GAMMA RADIATION -J FUEL PUMP STOPPED & NOTICED AT PUMP; REACTOR MADE REACTOR EXCURSION TO ~50 w RUN 3 SUBCRITICAL DURING INVESTIGATION PUMP STARTED [} PUMP LEVEL TRIMMED BY REMOVING = ABOUT 71 1b OF FUEL MIXTURE N#gkiAIRNoPPIETe;AHON STOPPED FOR FINAL [ S SAMPLE 16 REMOVED FOR ANALYSIS INJECTION FITTING AT PUMP FOUND TO BE _/ PLUGGED BECAUSE OF ELECTRICAL SHORT - — IN RESISTANCE-HEATED FITTING; FINAL INJECTION TO BE MADE THROUGH FUEL SAMPLE LINE REACTOR MADE CRITICAL — 10-min RUN AT 1 w FOR RADIATION MR SyURVEY; ACTIVITY AT PUMP FOUND TO BE DUE TO PRESENCE OF GAS LINE ‘ — NEAR DETECTOR F— ; —1— REACTOR PIT SEALED 1 : i . . EXP. L-2 RESUMED L REACTOR AT 1 w 1 > PREPARATIONS BEING MADE FOR FINAL SHIM RODS RESET ADDITION OF FUEL CONCENTRATE RUN 7 THROUGH SAMPLE LINE; FINAL CHECKS REACTOR AT 1 w MADE PREPARATORY TO SEALING PITS PENGUIN NO. 15 INJECTED REACTOR AT 1 w o~ REACTOR AT | w 0 PENGUIN NO, 18 INJECTED [ RUN 8 p . a. REACTOR AT 1 w X REACTOR AT 1 w PENGUIN NO, 17 INJECTED [ RUN 9 REACTOR AT 1 w REACTOR AT 1 w PENGUIN NO. 12 INJECTED p RUN 10 REACTOR AT 1 w y Fig. 5.1. NOVEMBER 6,1954 " T T T m— T i b, it e il e s e o o Biie ki i e e e ik dsdmass o e i b eanm NOVEMBER 7, 1954 FINAL FUEL ADDITIDON; REACTOR AT 1 . w START OF INJECTION OF 22.16 |b OF CONCENTRATE FROM CANS 25 AND 32 PLUS 38.2 |b FROM THE 70 Ib WITHDRAWN COMPLETION OF INJECTION ) ] | EXP. L-é | | FINAL FUEL SAMPLE (19) REMOVED FOR ANALYSIS ALL INJECTION AND SAMPLING LINES AND EQUIPMENT REMOVED FROM PUMP AND PITS TANK PIT SEALED ' REACTOR PERIOD EXCURSION TO 100 w; RUN 4 REACTOR PERIOD EXCURSION TO 100 w; RUN 5 ALL PIT WORK COMPLETED REACTOR AT 1 w PUMP STOPPED REACTOR PERIOD EXCURSION TO 460 w === PUMP STARTED ~ HEAT EXCHANGER PIT SEALED; ALL PITS NOW SEALED REACTOR PERIOD EXCURSION TO 100 w; RUN 7 g EXP., L-6: SHIM ROD CALIBRATION AT | w PERIOD EXCURSION TO 200 w; RUN 8; END OF EXP, L-5 REACTOR AT 1 w FOR SHIM-ROD CALIBRATION i e il - L-5 EXP. e i - sad oy ambia " REGULATING ROD POSITION ORNL-LR~-DWG 6400 NOVEMBER 8,1954 EXP, L-7: EFFECT OF FUEL FLOW RATE ON DELAYED NEUTRONS; REACTOR AT | w; FUEL FLOW VARIED O TO 46 gpm START EXP. L-8: MEASUREMENT OF 7 TEMPERATURE COQEFFICIENT AT LOW POWER; REACTOR AT 1 w START HELIUM BLOWERS RAISE THERMAL BARRIER DOORS INCREASE HELIUM BLOWER SPEED TO 255 rpm READIJUST SHIM RODS; LOW HEAT EXCHANGER REVERSE; HELIUM BLOWER OFF READJUST SHIM RODS L-8 EXP. OBSERVE HEAT UP AND CHANGE OF END OF EXP, L-8; START OF EXP, L-9: READJUSTMENT OF CHAMBERS FOR HIGH POWER LEVEL ACCIDENTAL SCRAM ' POWER 40 TO 500 w DURING ADJUSTMENT OF CHAMBERS PERIOD SCRAM TO CHECK SAFETY CHAMBERS; END OF EXP. L-9 END OF LOW-POWER EXPERIMENTS . EXP 20-kw RUN; APPROACH TOQ POWER; EXP, H-1 REACTOR SCRAMMED; ACTIVE GASES IN BASEMENT 4] | e T = i s i s i D . i e b s s i eiiia i s oo, .l ol S o s 1) i s 1, ., TRY TR T R T Y WV T e T T T Y e, T " W AR B oy kb e, e iy i b SRR L e w 4 m.‘k Lo it i SR Bl ok 3 L The radiation dose data from the two runs during which the dose in the pits was measured are pre- sented in Tables 5.1 and 5.2, Table 5.1 gives the gamma and the fast- and thermal-neutron doses in the reactor pit during the 2.7-w run; Table 5.2 gives the gamma doses at various stations in the heat exchanger pit during both the 2.7- and 27-w runs. Within the accuracy of the measurements, there is excellent agreement between the doses per watt from the two runs. REGULATING ROD CALIBRATION FROM FUEL ADDITIONS Twelve small cans, or “‘penguins” (so-called because of their shape), had been prepared for adding small amounts of fuel to the system in order to calibrate the regulating rod (exp. L-2) after initial eriticality had been reached. Each penguin contained from 0.2 to 0.5 |Ib of fuel concentrate (0.1 to 0.3 [b of U233), The contents of the pen- guins were injected directly into the pump rather than through the intermediate transfer pot that was used in the critical experiments. Table 5.3 lists the penguins in the chronological order in which they were used, together with the amount of yranium injected from each can, The method used to calibrate the regulating rod was to note the difference between the position of the rod at criticality before and after each injection of fuel, with the shim rods held constant. Before each injection the reactor was brought from sub- critical to a nominal power of 1 w, and the shim and regulating rod positions were recorded. The reactor was then brought subcritical on the regu- lating rod, with the shim rods in position, and the fuel was injected from the penguin, Upon comple- tion of the injection the reactor was again brought critical to l-watt power with the regulating rod. The positions of the rods were again recorded. The worth of the rod was then obtained from the relation Ak AM k M and the known increment of fuel added to the sys- tem; the proportionality factor, 0.236, was obtained experimentally during the critical experiment (cf., chap. 4). The first run of experiment L-2 started at 1651 on November 4. After two injections had been completed, it was apparent that the worth of the 12 in. of vertical movement of the regulating rod in terms of Ak/k was about 0.24%. It was desirable, for reasons of safety and also for convenience in conducting high-power transient experiments, fto have a rod worth about one dollar of reactivity (which for the ARE was 0.4% rather than 0.76% as in stationary-fuel reactors, because of delayed- neutron loss in the circulating fuel). A number of spare rods of varying weights had been made up to take care of such a contingency, and one was selected and installed which had more nearly the desired weight. The original rod weighed 19.2 g/em of length, and the new rod weighed 36 g/cm. While the rod was being changed (a delay of about 5 hr), the fuel system characteristics of pressure head vs flow were obtained. These measurements are described in a following section of this chapter, TABLE 5.1. RADIATION LEVELS IN REACTOR PIT DURING 2,7-w RUN Gamma Fast-Neutron Thermal-Neutron Total Position Dose Rate Dose Rate Dose Rate Dose Rate* {mr/hr per w) (mrep/hr per w) {mrep/hr per w) (me/hr) Space cooler No. 1 48 67 8 760 Top center of reactor : 280 280 35 (est.) 3200 (1'in. above thermal shield) Side of reactor at mid-plane 440 180 22 {est.) 2350 (at surface of thermal shield) Manhole into pits 12 10 2 130 *The total dose rate was obtained from the weighted sum of the preceding columns by using an RBE of 10 for fast neutrons and 5 for thermal neutrons. 43 o T T e T T el L TABLE 5.2. GAMMA-RAY DOSE IN TANK AND HEAT EXCHANGER PITS » i s e e Doses per Watt During 2,7-w Run Doses per Watt During 27-w Run (mr/hr) (me/hr) Locations Surveyed in Dump Tank Pit Motor for space cooler No, 3 0.3 0.3 Vent valve U-113 for tank No. 6 0.1 0.15 - Motor for stack vent on hot fuel dump tank 0.2 0.15 Vent Valve U-112 for hot fuel dump tank 0.1 0.07 : Helium inlet valve U-100 0.07 0.07 Motor for space cooler No. 4 0.07 Air-operated valve for tank No, 5 0.07 Locations Surveyed in Fuel-to-Helium Heat Exchanger Pit Operator’s position at fuel sampler 6 4 Pressure transmitter PXT-6 22 Top plate of main fuel pump 150 120 Top bearing of pump motor 45 80 V-belt at pump . 7 6 V-belt at pump motor 7 4 Line 120 under valve U-3 110 12 Bend No, 2 of line 303 105 45 Motor for space cooler No. 6 10 West side of heat exchanger No. 1 135 120 Bottom of fuel flowmeter 95 105 Line 112 between valves U-63 and U-1 220 200 Lubrication pump for pump 65 90 Top of fuel storage tank 4 Helium analyzer dryers 0.3 1 3 East side of heat exchanger No. 1 230 West side of heat exchanger No. 2 200 Sheet metal can around pump 185 L ocations Surveyed in Scdium-to-Helium Heat Exchanger Pits Top of standby sodium pump 1.5 1.4 Y-belt above standby pump 1.5 1.7 Line 310 under vaive B-141 13 15 Motor for space cooler No. 7 4 Top of main sodium pump 1.5 0.7 V-belt above main pump 0.7 0.8 Line 309 at valve U-21 2.2 1.8 Line 313 at bend No. 5 1.5 1.5 Motor for space cooler No, 8 1.5 North side of front heat exchanger 15 16 Kfnney pump No. 1 4 4 Magnetic clutch on 50-hp motor 1 13 North side of back heat exchanger 1 6 North end of rod-coolant blower No. 1 6 7 EM flowmeter on line 305 3.7 5 Standby pump lubrication system 9 Main pump lubrication system 3 44 . o - T T * T '?'r‘!‘?"*“’&fl_”' . T © g It R T o e e e ety v e e B poy e Pormrr v v v B e 2 e o i & kL -t s b ~anbod R A i LR A . hboeg S R ek K R R el e, ok = - s R, i, o B i TABLE 5.3. FUEL CONCENTRATE BATCHES ADDED DURING LOW-POWER E XPERIMENTS _ Ponguin Concentrate Added Uranium Added U235 Added Date Time No. g wt % g g b 11/4 1710 C-11 227 0.5004 59.671 135 126 0.2778 2009 C-19 722 1.5917 59.671 43 403 0.8885 11/5 0354 C-14 247 0.5445 59.671 147 137 0.3020 0439 C-16 285 0.6283 59.671 170 159 0.3505 0529 C-13 131 0.2888 59.671 78.2 73.0 0.1609 0619 C-5 123 0.2712 59.671 73.3 68.5 0.1510 2045 C-15 92 0.2028 59.671 55 51.4 0.1143 2204 C-18 87 0.1918 59.671 52 48.6 0.1071 2258 C-17 457 1.0075 59.671 273 255 0.5622 2345 C-12 84 0.1852 59.671 50 46.7 0.1030 11/6 0100 C-10 84 0.1852 59.671 50 46.7 0.1030 11/7 0231 120% 21,681 47.798 27.557 5975 5581 12.304 *Container No. 120 was not a penguin but a specially prepared batch of concentrate which provided the excess uranium required for the experiments and to compensate for burnup. At 0142 on November 5 the installation of the second rod was completed and the experiment was resumed. Four more injections of fuel were made and the rod movement was noted. At 0630, during the sixth fuel injection, an unexpectedly high burst of gamma-ray activity (55 mr/hr) was ob- served on a ‘‘Cutie-Pie’’ monitoring the fuel addi- tion at the injection station over the pump. Further fuel addition was stopped until the cause of the high gamma-ray reading could be ascertained. While investigation was under way the level of the pump was trimmed to avoid any possible hazard from the uranium held in the pump. About 71 .Ib of fuel was removed from the system. Later in the day the cause of the high gamma-ray activity was determined to be a vent line passing close to the position of the ‘'Cutie-Pie.”” Apparently, the “‘Cutie-Pie”" had been held close enough to the vent line to read the fission gases (evolved during operation) as they were discharged through the vent line. At 1947 the experiment was resumed and five more injections were made uneventfully, A photo- graph of the control room taken during this time is presented in Fig. 5.2; it shows two members of the ARE operating group examining the fission chamber recorders after a fuel injection for rod calibration; other members of the evening crew are shown in their nominal operating stations. An inventory of the uranium added during this experiment is given in Table 5.4. The final large amount (47.8 Ib) of concentrate containing 12.30 ib U235 that was added after the experiment is also shown in the table, as well as a record of the samples and withdrawals, The final amount of U235 jn the system was 138.55 Ib, with a system volume of 5.33 fi3, The data obtained for calibration of the regulating rod as a function of fuel addition are given in Table 5.5, which lists the values of M, AM, AM/M, the recorded movement of the rod during each fuel addition, the average position of the rod for each fuel addition, and the calculated value of (Ak/k)/in. The results of the calibration are shown in Fig, 5.3, where (Ak/k)/in. is plotted as a function of rod position. The movement of the rod for each point is shown as a horizontal line through the point. The volues of (Ak/k)/in. for both rods were used, the values of reactivity for rod No. 1 being corrected by the ratios of the weights of the two rods. From these data it is evident that there is no good reason for presuming that the value of (Ak/E)/in, over the whole length of rod is not constant. Therefore, the average value taken over the first ten runs gives (Ak/E)/in. = 0.033%/in, 45 N [ — -l R e R 9w rro® coep T e T s egmerrpt T e T g e e T e e B e et b ke el - Bl ., g iRk ke oas s S Coulle k. W . [ b o amRS g " A N om kit g . R LS SR Wb bk TABLE 5.4. URANIUM INVENTORY DURING LOW-POWER EXPERIMENTS TR T ——— Samples Removed Fuel Mixture Uranium-235 Fuel Concentrate Added for Analysis Total Weight _ Total Date Time NL:: Weight Yolume Fuel y23s Concentrate Vi?::wle Density X;:fi}:f Weight Concentration (Ib) (f13) Removed Removed Plus Carrier (F13) (Ib /53 (Ib) in System |b/#3 wt% (ib) {ib) (tb) (b} /4 1315 0 ) 114745 55454 206.9 132,763 23.94 1157 F 1710 1 0.5004 0.0017 1147.95 5.5471 2069 0.2778 133.041 23.98 11.59 2009 2 1.5917 0.0056 1149.54 5.5527 207.0 0.8885 133.930 2412 11.65 ! /5 035¢ 3 0.5445 0.0019 1150.08 5.5546 207.05 0.3020 134232 2417 11.67 ' 0439 4 0.6283 0.0012 1150.71 5.5558 207.1 0.3505 134.583 2422 11.70 0529 5 0.2888 ¢.0010 1151.00 5.5568 207.1 0.1609 134.744 2425 11.71 0619 6 0.2712 0.0009 1151.27 5.5577 207.1 0.1510 134.895 24.27 11.72 ( 1015 53.486* 6.269 1097.78 5.2944 207 128.626 24.27 11.72 E 1020 17.523* 2.064 1080.26 5.2148 207.1 126.562 2427 11.72 ; 1100 2.5807 0.2%1 1077.68 52023 207.1 126.271 24.27 1172 i 2045 7 0.2028 0.0007 1077.88 5,2030 207.2 0.1143 126385 2429 11.73 : 2204 8 0.1918 0.0004 1078.07 52034 207.2 0.1071 126492 2431 1173 2258 9 1.0075 ¢.0020 1079.08 5.2054 207.3 0.5622 127.054 2441 11.77 2345 10 0.1852 0.0004 1079.27 52058 2073 0,030 127,157 2443 1178 11/6 0160 11 0.1852 0.0004 1079.45 5.2062 2073 0.1030 127.260 24.44 1179 ’ 0530 2.686 0.317 1076.76 5.1932 2073 126.943 2444 1179 l 0535 2.839 0.335 1073.92 51795 2073 126.608 24.44 1179 ~ /7 0231 12 47.798** 0.1667 1121.72 5.3462 209.8 12304 138912 2598 1238 . 0423 2.918 0.361 1118.80 5.3323 209.8 138.551 2598 1238 *Two large batches of fluoride mixture were removed from the system in order to frim the liquid level in the pump. **+0f this amount, only 15.419 |b was N02UF6, the remainder being some of the fluoride mixture that was removed from the system when the pump level was trimmed. ORNL-LR-DWG 6401 © e e e e REACTIVITY (% /in)) 0.05 | o —ROD COMPLETELY ROD COMPLETELY INSERTED , WITHDRAWN \ ! 0.04 —~— o - WEIGHTED AVERAGE [0 L~ D AVE 0.03 — it 0.02 C.01 0 : 2 3 4 5 6 7 8 9 10 " i2 13 14 REGULATING ROD PQOSITION (in.) : k L E Fig. 5.3. Calibration of Regulating Rod from Fuel Addition, 47 \ TABLE 5.5. REGULATING ROD REACTIVITY CALIBRATION FROM FUEL ADDITION E (Exp. L-2) F | M, 235 AM, 23 Average Rifl::;?:; F:Zd Reactivity, k Run Weight of U Weight of U233 Am/m Ak/k Regulating Rod During Fuel (Ab/k)in. No. in System Added (%) Position Addition (%/in) .k (Ib) (1b) (in.) , . (in.) ; ] 133,041 0.2778 0.00209 0.0493 11.80 2.4 ©0.0308* . E 2 133.930 0.8885 0.00663 0,156 9.35 7.3 | 0,0400% E 3 134.232 0.3020 0.00225 0.0531 10.70 1.6 0.0335 fi 4 134.583 0.3505 0.00260 0.0614 8.80 1.9 0.0323 E 5 134,744 0.1609 0.00119 0.0281 6.80 0.85 0.0330 E 6 134.895 0.1510 0,00112 0.0264 6.00 1.0 0.,0264 é 7 126.385 0.1143 0.000904 0.0213 6.35 0.7 0.0304 b 8 126,492 0.1070 0.000847 0.0200 6.00 0.6 0.0333 9 127.054 0.5622 0.00442 0.104 5.10 3.6 0.0289 10 127.157 0.1030 0.000810 0.0191 12.25 0.5 0.0383 11 127.260 0.1030 0.000809 0.0191 2.95 0.67** 0.0305%+* Average reactivity, weighted according to rod movement 0.0331 *Corrected by the ratio of the weight of the new rod to the weight of the original rod: 36/19.2 = 1.88, ** Average of three readings. The value from run 11 was not included in this scrammed. From the induced period and the known S average because three different values of rod final flow, the excess reactivity introduced during the £ position were recorded in different places for this run was obtained from the inhour formula, This ' run. number was then divided by the travel of the regu- £ lating rod to find the value of (Ak/k)/in. for the i REGULATING ROD CALIBRATION FROM run, A total of eight reactor excursions of this ¥ REACTOR PERIODS nature were made, six at 48-gpm flow, and two at - The calibration of the regulating rod by use of zero flow. g reactor periods (exp. L-5) utilized the inhour equa- Some of the reactor excursions during this ex- ‘ tion in a form in which reactivity is given in terms periment are shown in Fig. 5.5, as recorded by the # of the reactor period for a given rate of fuel fiow, log N recorder. The straight lines that indicate E Figure 5.4 shows a plot of the reactivity {Ak/k) as the slope of the power increase reflect periods on a function of reactor periods for 0- and 48-gpm the order of 20 to 25 sec. Figure 5.6 presents some F fuel flow, as calculated from the inhour relation, of the period traces recorded on the period recorder : Details of the calculational method used are pre- during the experiment. The initial high peaks for E sented in Appendix I, each rod movement are due to the transient condi- k In the experimental procedure followed, the regu- tion, the interpretation of which is given in ; lating rod was suddenly withdrawn a known dis- Appendix S. In plotting the period for a particular “ tance, while the shim rods were in a set position and the reactor was operating at a nominal power of about 1 w. The reactor was then allowed to rise in power on a constant period to a peak of 50 to 100 w, at which time it was manually 48 rod motion, the average of the periods obtained from the log N and the period recorders was used, A plot of (Ak/k)/in. as a function of rod position as obtained from this experiment is shown in Fig. 5.7. The horizontal lines on each side of the R T U ey - wy R T Cakii . R e e BA. - ol o b B ia ehcd Gomdges.d sl g ORNL—-LR—DWG 8402 0.0035 0.0030 \ 0.0025 . ZERO FLOW L 0.002C \\ 0.0015 \ REACTIVITY, Ak/k 0.0010 ] \\ — 0.0005 T~ \ — 0 0 10 20 30 40 50 60 70 80 20 100 REACTOR PERIOD (sec) Fig. 5.4. Regulating Rod Reactivity vs Reactor Period for Fuel Flow Rates of 0 and 48 gpm. experimental points show the extent of the rod movement for that point. Again, there is no con- clusive evidence that the reactivity of the rod per vnit length was not constant. Therefore, the average of the 48-gpm flow data weighted over the rod movement for each run gives (Ak/R)/in. = 0.029%/in. For the two zero-flow points the rod was moved over the middle 10 in. of its 12-in. travel in two 6-in. overlapping runs covering the upper and lower halves of the rod, respectively, The weighted average of these points gave a value of (Ak/k)/in. = 0.033%/in. , which is in excellent agreement with the fuel addition calibration, This value was settled upon as the best value of reactivity per inch over the whole length of the regulating rod, The values for the reactivity vs rod position are given in Table 5.6. CALIBRATION OF SHIM RODS VS REGULATING ROD With the reactivity of the regulating rod known, the shim rods could be calibrated against it (exp. .-6). This calibration could then be compared with the previous calibration made during the fuel addition. It was assumed that the three shim rods were enough alike that a calibration of one rod would be representative, and rod No. 3 was chosen for the calibration, The calibration procedure consisted of setting shim rods Nos. 1 and 2 in a fixed position, and then, with the reactor at a power level of 1 w on the servo mechanism, rod No. 3 was moved until a specified travel of the regulating rod was obtained. At the start of the experiment the rods were ad- justed so that shim rod No. 3 was nearly all the way out (position indicator at 35 in.) and the regu- lating rod nearly all inserted (position indicator at 3 in.). After recording the position of all rods, 49 = T g tn S ORNL-LR-DWG 6403 1000 2 —— RUN2,NOV. 6, 0715 RUN 3, NOV. 6, 0950 — RUN 5, NOV. 7, 1810 FUEL FLOW, 46 gpm FUEL FLOW, O gpm FUEL FLOW, 46 gpm RUN 8, NOV. 7, 2355 FUEL FLOW, 46 gpm 100 REACTOR POWER (watts) 1 1w, NOMINAL POWER 0.1 —~—— TIME Fig. 5.5. Typical Reactor Excursions Recorded by Log N Recorder During Period Calibration of Regulating Rod. Experiment L-5, L4 ¥ wee s WF R OT P ) M wh o e CEETTE CUpcin e g e, TEPECCRIE T TEehr mertrie R gt ST Yo omEEs “‘"‘!’m"m Cr o Mg e fEEUY R O EENEEUres o EUTTET T e rer Y g S tr. glewr Tt R T rEmpgmes g P » 16 REACTOR PERIOD, (sec) RUN 3, NOV. 8, 0715 FUEL FLOW, O gpm RUN 4, NOV. 7, 1610 8 FUEL FLOW, 46 gpm 10 20 30 100 —{00 -30 RUN 5, NOV. 7, 1808 FUEL FLOW, 46 gpm RUN 6, NOV. 7, 2034 FUEL FLOW, Ogpm - - TIME ORNL-LR-DWG 6404 RUN 7, NOV. 7, 2208 FUEL FLOW, 46 gpm RUN 8, NOV. 7, 23 FUEL FLOW, 46 gpm Fig. 5.6. Reactor Periods Recorded by the Period Recorder During Calibration of Regulating Rod, Experiment L-5. - e T T - R TR =~ T ORNL-LR-DWG 6405 0.40 FUEL FLOW, Cgpm 0.35% REACTIVITY (% /in.} ROD COMPLETELY INSERTED 0.25 I ROD COMPLETELY WITHDRAWN ——_,_ | 0.04 ‘ I WEIGHTED AVERAGE 003 —F—TF—— —— — ’ ! * 0.02 REACTWITY (% fin.) FUEL FLOW, 46 gpm 0.04 2 3 4 5 6 7 8 9 10 1 12 13 14 REGULATING ROD POSITION (in.} Fig. 5.7. Calibration of Regulating Rod from Reactor Period Measurements, shim rod No. 3 was inserted until the servo action had withdrawn the regulating rod a compensating distance of about 10 in. The action was then stopped and the new rod positions were recorded. With shim rod No. 3 set, shim rods Nos. 1 and 2 were then adjusted to bring the regulating rod back to its original position, and again the rod positions were recorded, Again rod No. 3 was inserted until a 10-in. withdrawal of the regulating rod was attained. This process was repeated until rod No. 3 had been inserted about 14 in,, and each new position had been recorded. This calibration agreed well with the calibration against fuel addi- tion. Details of the results are given in Appendix J. FUEL SYSTEM CHARACTERISTICS The fuel system characteristics with the fuel concentrate in the system were determined for the 52 first (and only) time the day after the reactor first became critical (exp. L-3). The final amounts of concentrate (i.e., the excess to provide for burnup and poisoning during the subsequent power runs) had not yet been added to the system, Conse- quently, the fuel density at the time the data were taken was 3.32 g/cm3 as compared with 3.36 g/cm? for the final fuel composition, and 3.08 g/cm® for the carrier; all densities were determined at 1300°F. The measured and calculated flow data as a function of pump speed are shown in Fig. 5.8. The flow was calculated for two pump speeds by using the time between pips on the fission chamber as concentrate was first added to the system. The straight line joining the two calculated points is considered to be a good approximation because of the regime in which the pump was operated. e b oy e gy e "o »e et ey s . P e g Theer ot ot - —p . e TTT F oo oy o e o s wErmeToomes mewer WET Y roeerze MF TRE ey T meey P ke, i bl e g o R S, g, R% T Y el s iR, B R o e, Sk S b A BREREE | TABLE 5.6. REGULATING ROD REACTIVITY CALIBRATION FROM REACTOR PERIOD MEASUREMENTS (Exp. L-5) A Evel FI Average Ak/E R Ivu;-;l:ageR 4 Movement of Reactivity, Run No. U? ow Period from Inhour eg; a .“Tg ° Regulating Rod (Ak/R)/in. gpm) (sec) Equation os'tt|§n (in.) (%/in.) . (in.) 1 48 48 0.00036 7.5 1.1 0.0318 2 48 22,2 0.00060 7.6 2.C8 0.0289 3 0 21 0.00206 9.6 6.05 0.0330 4 48 26 0.00054 3.8 2.0 0.0270 S 48 22 0.00061 11.1 2.2 0.0278 6 0 21.6 0.0020 5.6 ‘ 6.1 0.0328 7 48 20.1 0.00065 5.2 2,08 0.0313 8 48 22.2 0.00060 9.3 2.06 0.0291 Average for 0-gpm flow 0.0329 Average for 48-gpm flow 0.029 1200 ORNL-LR-DWG 6408 ORNL-LR—DWG 6407 L 30 FLOW CALCULATED FROM PIPS /O/' 1000 [— AT 1080 AND 520 rpm ASSUMING // CIRCULATING VOLUME TO BE //' 25 ACTUAL PRESSURE DATA (FLOW CALCULATED FR(!M 3 - P . E s00 482 f17{ASIN Fig. 4.4) /e/ TIME BETWEEN PIPS ON FISSION CHAMBER )~ / 2 //; °/ & 600 A& / // v /< 20 7/ % /P/ FLOW FROM FLOWMETER = / o. P ] = 400 v g // WRE .,/ 200 {7~ 7 , . & o / T / 0 S 0 10 20 30 40 50 60 10 7 FLOW RATE (gpm) /‘/ .// ,o/ / // Fig. 5.8. Fuel Flow Rate as a Function of Pump 5 4!\ Speed. ' o | “PRESSURE CORRECTED - (FLOW FROM PIPS) ‘ . A ' The flow data as a function of the system head 0l 0 {0 20 30 40 50 loss from the reactor inlet to the pump tank are given in Fig. 5.9. The flow data are plotted di- rectly vs head (by using the flow rates determined from pump speed as given by the upper curve in Fig. 5.8) and then as corrected for a ‘‘live’’ zero FLOW RATE {gpm) Fig. 5.9. Fuel Flow Rate vs the Pressure Head from the Pump Suction to the Reactor Inlet. 53 on the pressure transmitter from which the head values were obtained. The latter data fit the theoretical curve very well, Furthermore, these system data fit the pump performance datq, if it is assumed that there is a 7-psi drop from the pump discharge to the reactor inlet at design flow (i.e., 46 gpm); there is no experimental measurement of this pressure drop. The pump performance charac- teristics determined in a test stand run with fuel at 1300°F are given in Fig. 5.10, “ EFFECT OF FUEL FLOW ON REACTIVITY , An experiment was performed in which the effect of fuel flow on reactivity was observed (exp.L-7). ’ For this experiment the reactor was brought to a nominal power of 1 watt and given over to the flux servo mechanism after full fuel flow of 46 gpm had been established. After recording the positions of the regulating and shim rods the flow rate was reduced step-wise until zero-flow was reached. Each reduction of the flow was accompanied by a change of position of the regulating rod. Reducing the flow had the effect of allowing more delayed neutrons to contribute to the flux level in the reactor, and therefore the servo mechanism inserted the regulating rod to compensate for the excess reactivity created by the delayed neutrons. Each rate of fuel flow and the corresponding regulating rod position were recorded. The results of the experiment are shown in Fig, 5.11, where AL/k is plotted as a function of flow rate. It is observed that 12 in. of rod movement, or 0.4% Ak/k, was needed to compensate for the reactivity introduced when the flow rate was reduced from full to zero flow. This curve is comparable to that shown in Fig. D.2 of Appendix D. The data for this experi- ment are tabulated in Table 5.7, Although not a part of experiment L-7, additional insight into fuel activity may be obtained from examination of Fig. 5.12, which shows activity as recorded by fission chamber No. 2. Before the data shown in the figure were recorded the reactor had been operating at a power of about 1 w with the fuel pump stopped. As shown, the reactor was ORNL—-LR-DWG 4524 60 60 50 {400 rpm| 50 . \ 20 e EFFICIENCY AT 1200 rpm {200 rpm g’ — ey < T / ; O P — 30 30 o SCJ \ S ’ E ui / 800 rpm // 1000 rpm —— 20 A T \ \‘_ 20 T~ 54 600 rpm \ o - 0 0 0 20 30 40 50 80 70 80 90 ' FLOW (gpm) Fig. 5.10. Fuel Pump Performance Curves., ey P F e T g - T .l. W ;W T W’F"@'W‘”W*’ »W' e T ” "'WF = o SRR e e e P L s o % ORNL-LR-DWG 6408 04 { -+ 105 03t 9.0 <4 7.5 ; \ | a5 \. REACTIVITY CHANGE DUE TO (o] ) SRR REE . STOPPING SODIUM FLOW \. | \.‘1 \l —H 15 i\\a} ® 0 5 10 15 20 25 30 35 40 45 50 FLOW RATE (gpm) 30 EXCESS REACTIVITY Ak fk (%) o av & =} INSERTION OF REGULATING ROD (in) 0 Fig. 5.11. Effect of Fuel Flow Rate on Reac- fiVifYQ TABLE 5.7. EFFECT OF FUEL FLOW RATE ON REACTIVITY (Exp. L-7) Fuel Sodium Movement of Change in Run Flow Flow Regulating Rod Reactivity, e (gpm) (gpm) (in.) Ak/k (%) 1 45,5 149 0 0 2 41 149 0.7 0.0231 3 35 149 1.4 0.0462 4 28.5 149 1.9 0.0627 5 238 149 2.4 " 0.0792 6 18 149 3.0 0.099 7 13.5 149 3.45 0.114 8 0 149 12.05 0.398 9 13.5 149 3.45 0.114 10 0 149 12,05 0.398 11 13.5 149 3.65 0.120 12 13.5 149 3.65 0.120 13 43.5 149 0.9 0.0297 14 41 149 0.96 0.0317 15 41 0 1.12 0.0370 16 41 149 0.96 0.0317 then scrammed! and the fuel pump started, The fuel which had been in the reactor before the scram was more active than the remainder of the fuel and ]Scrum, as used in this report, refers to an intentional shut down of the chain reaction by suddenly dropping the contrdl rods into the core. ORNL-LR-DWG 3854 —T— REACTOR SCRAMMED ‘ . DATA FROM FISSION CHAMBER NO. 2 : 3 EXP. L-5, RUN 6, NOV. 7,1954 == T IEITETT (READ FROM RIGHT TO LEFT) 1) i PUMP STARTED Fig. 5.12. Circulating Fuel Slugs after Starting Pump Following a Reactor Scram. it was this excess activity which showed up as the fuel was circulated after the reactor was scrammed. LOW-POWER MEASUREMENT OF THE TEMPERATURE COEFFICIENT A measurement of the temperature coefficient of reactivity (exp. L-8) was made next. With the reactor isothermal at 1312°F and controlled by the flux servo mechanism, the heat barrier doors were raised and the helium blower turned on to 200 rpm to cool the fuel in the heat exchanger. As the temperature of the reactor decreased, the servo began to drive the regulating rod in to com- pensate for the increased reactivity, The recording of the regulating rod position vs time is shown in Fig. 5.13. Beginning at 0218, the rod was in- serted by the servo quite rapidly, and by 0221 the With the regulating rod on servo, one of the shim rods was inserted as rapidly as possible. This shim rod insertion overcompensated for the increase in reactivity due to the temperature drop, and the rod was approaching its lower limit, 55 - TP P T VT [ T . v g T T T R e T e 4 i i 4 ¢ ORNL—LR-DWG €409 02492 0244 e e o b 0240 | ol WITHDRAWAL 0239 b — e INSERTION —— - G238 0237 0236 0235 EXP. L-8, NOV. 8, 1954 0234 . 0233 0232 0234 0230 0229 0228 0227 TIME OF DAY Q226 0225 0224 ; 0223 ' \‘_\ ‘ L. UPPER END OF | / ROD TRAVEL - | -« BLOWER OFF - I e} ROD WITHDRAWN 0222 i 0220 0219 oz18 0217 0218 by 02145 ONE INCH REGULATING ROD 0214 TRAVEL Q213 0212 ‘ : 5 ; - BY INSERTING SHIMS TO OVERRIDE SERVO | LOWER END OF b\-\‘_L"T\ T ROD TRAVEL |, | O BLOWER ON {255 rpm) T 0 | R — .- oL _,}\START TO RAISE B HEAT BARRIER DOORS " 13 12 i 10 9 8 7 5 4 3 2 1 0 REGULATING ROD INSERTED (in.) Fig. 5.13. Regulating Rod Position During Low-Power Measurement of the Reactor Temperature Coefficient, servo quickly withdrew the regulating rod. In- sertion of the shim rod was stopped when the regulating rod had been withdrawn to near its vpper limit, and the regulating rod was again in- serted by the servo to compensate for the reactivity introduced by the cooling of the fuel. By 0230, the outlet temperature in the heat exchanger was -approaching its lower limit, 1150°F, so the helium blower was stopped and the run was ended. A trace from the chart for recording the mean fuel temperature is shown in Fig. 5.14, and super- imposed on it is a plot of the displacement of the 56 regulating rod as obtained from the previous figure, A measure of the temperature coefficient may be obtained by comparing the slopes of the two curves, At 0228 the rod was moving at exactly 1 in./min, corresponding to an increase in reactivity of 3.3 x 10~4 (Ak/k)/min. At the same time, the mean temperature was dropping at the rate of 5.1°F/min. The resulting reactor temperature coefficient, as obtained from these data, would appear to be ~6.48 x 10=° (Ak/k)/°F. By com- paring the slopes of the curves obtained earlier during the run, a considerably higher coefficient - e T . R T b I g B L BT wer e ?‘$'?'W&'!"’ ot Ll L T o T ., wrv% AT s W Y TN e o LHRE T b W g - i ‘k. b & il i~ S k. Sl i il - . * ORNL-LR-DWG 6410 [ 400 ) £ T T ERRER" Ll o~ - b Le. 3z o LrN) 1 o S« o £XP. L~8 NOV. 8,1954 = T O & = = o i o o “ 2 — ‘3:“135 x Yo —1 1350 U Sea |En - 2 - 2 - I w VES &1 = T ¢ - — - EN\ENL’ N Ll = \’pC - n q 0%~ b 5 16 A \—— -------- : 1300 » (AR/E)/°F. These are the best values for these coefficients that were obtained. It should be noted that the value for the over-all temperature coefficient agrees to within 25% with the value previously obtained from the low-power experiments. However, - PR Pyt R WY T e et T ey e o - Tt e M e e |t E o e T oy 7 T e T ey E & i k - E . K N o AR g ol et ol L, o, el wirli mpy, bRk K, s, gl L G K e R kel e e xR e A o Vb gy ol TIME CF DAY ORNL—-LR—-DWG 64142 2127 ’ ( l ~=—[NSERTIO AL —— WITHORAWAL = OFF SERVO CONTROL 2126 \\‘ 2125 fifi\\\ EXP, H-4. NOV, 9, 1954 \“w\\ UPPER END OF N / ROD TRAVEL ered — | ROD WITHDRAWN | ___J___———————“‘—"J BY INSERTING SHIMS B e TO OVERRIDE SERVO : \ h“"\~=‘ LOWER END OF 2123 < ROD TRAVEL — \\ 2122 T ] \ }_‘ — \ ONE INCH —ea— BLOWER ON {350 rpm) REGULATING ROD TRAVEL 2121 2120 12 f 10 9 8 7 6 5 4 3 2 1 0 REGULATING ROD INSERTED {in.) Fig. 6.2. Regulating Rod Position During Measurement of Reactor Temperature Coefficient. Experi- ment H-4, ROD DISPLACEMEN® (in.} 2121 BL.LOWER ON 2122 ORNL - LR-DWG 6413 1400 EXP. H-4 NOV. 9, 1954 BLOWER OFF -— 1350 1 RO 1300 ME AN TEMPERATURE REACTOR MEAN TEMPERATURE (°F) 1250 2123 2124 2125 2126 2127 2128 TIME OF DAY Fig. 6.3. Regulating Rod Position Supetimposed on Fuel Mean Temperature Chart, Experiment H-4. 63 * T B T T T YT PO e R T T e - e TABLE 6,1, REACTOR TEMPERATURES AND ROD POSITIONS DURING 100-kw MEASUREMENT OF TEMPERATURE COEFFICIENTS (EXP. H-5) El petmmeee o W*W L. A W e TE T E Regulating Rod Rod Reactor Mean Reactor Inlet Reactor Outlet 5 Time Position Withdrawal Temperature Temperature Temperature (in.) (in.) °F) (°F) (°F) . i 2223 3.72 0 1263.5 1262.5 1283 . - 24 3.84 0.12 1265 1263.5 1285 25 4,15 0.43 1266 1264 1286 b 26 4.43 0.71 1267 1265.5 1287 L 27 4.69 0.97 1268 1266 1288 i 28 4.96 1.24 1269 1267.5 1290 ‘ 29 5.23 1.51 1270.3 1269 1291 . 30 5,50 1.78 1271.5 1271 1292 31 5.77 2.05 1273 1272 1293 * 32 6.02 2.30 1274 1273 1294 33 6.28 2.56 1275 1274.5 1296 34 6.49 2.77 1276.5 1275.5 1297 35 6.68 2.96 1277.5 1277 1298 36 6.90 3.18 1278.5 1278 1300 37 7.10 3.38 1279.5 1279 1301 38 7.35 3.65 1281 1280.5 1302 39 7.50 3.78 1282 1281.5 1303 40 7.69 3.97 1283 1282.5 1304 41 7.92 4.20 1283.5 1283.5 1305 42 8.14 4.42 1285 1284.5 1307 | 43 8,32 4.60 1286.3 1285.5 1308 ¥ 44 8.55 4.83 1287.5 1286.5 1309 45 8.75 5.93 1288.5 1287.5 1310 46 8.94 5.22 1289.5 1289 1311 : 47 9.11 5.39 1291 1290 1312 48 9.31 5.59 1292.5 1291 1313 49 9.50 5.78 1293.5 1292 1314 50 9.68 5.96 1294.3 1293 1315 L 51 9.85 6.13 1295 1294 1316 : 52 10.05 6.33 1296.5 1295 1317 g 53 10.23 6.51 1297.5 1296 1318 54 10.45 6.73 1298.8 1297 1319 ‘i" 55 10.60 6.88 1299.5 1298 ' 1320 E 56 10.78 7.06 1300.5 1299 1321 Efi | 57 10.99 7.27 1301.3 1300 1322 58 | 11.17 7.45 1302.5 1301 1323 E - 59 11.41 7.69 1303.5 1302.3 1324 . 2300 11.54 7.82 1305 1303 1325 3 01 11.80 8.08 1306.3 1304 1326 g 02 11.95 8.23 1307 1305 1327 03 12.20 8.48 1308 1306 1328 * f‘ 04 12.38 8.66 1308.8 1307 1329 : 05 12.57 8.85 1309.8 1308 1330 k 06 12.75 9.03 1310.5 1309 1331 | s E 07 12,90 9.18 1311.5 1310 1332.5 5 08 13.19 9.47 1312.5 1311 1333 L 09 13.25 953 | 1313.5 i 0 I 64 E ko Wk e - i B i i ey, e B, Wl i - ©M A S 14 ot b =~ i, b . ORNL-LR-OWG 6414 10 - / 9 - > OVER-ALL TEMPERATURE COEFFICIENT: (é?) - 2N ose3inseF : B fover-aLL S0°F * : 8 Ak -5 @& * (A—f) = ~0.183 x 0.00032 = —6.0 x 107~ Ak /°F OVER-ALL 2 (& ] O @ 2 6 = }-—— < | - | ] | o /‘//‘/“ i i L o § '/ g 4 > O z / = // o 3 / 2 . Ad —9in. . INITIAL TEMPERATURE COEFFICIENT : (—) = = —0.296in./°F 1 /A// B ma - 30-8°F w / : - //0 oL = % = —0.296 x 0.00032 = ~0.98 x 10°% £k/°F ® 7/ 0 o | | | 1260 1265 1270 1275 1280 1285 1290 1295 1300 1305 1310 1315 7, REACTOR MEAN TEMPERATURE (°F) Fig. 6.4. Regulating Rod Movement as a Function of Reactor Mean Temperature., Experiment H-5. the instantaneous value of —9.8 x 1075 (Ak/R)/°F is much more important from the reactor control standpoint, and it was this large fuel temperature coefficient which made the ARE demonstrate the excellent stability described below under the subtitle ‘'Reactor Kinetics.” The value for the reactor temperature coefficient was subsequently confirmed in a later experiment (Exp. H-9) in which the system temperature was changed by gradually cooling the sodium, which in turn, cooled the fuel and the reactor. While the curve obtained from this experiment was very similar to that given in Fig. 6.4, the initial slope could not be that due to the fuel temperature co- efficients, because the fuel is one of the last con- stituents in the reactor to feel the temperature change. The “'equilibrium’’ slope, however, should be that determined by the reactor temperature co- efficient, and a value of —6.3 x 1073 (Ak/E)/°F was obtained. No temperature correction was re- quired because the extracted power in the fuel system did not change (although that in the sodium system did). This value agrees very well with that obtained in the preceding experiment. Sodium Temperature Coefficient In order to measure the sodium temperature co- efficient an experiment {Exp. H-7) was performed in which the sodium was cooled and the correspond- ing changes in sodium temperature and reactivity were measured, [n this experiment it was mando- tory that the sodium be cooled rapidly and the data recorded before the moderator had time to cool. Otherwise, since the sodium bathes the moderator, the moderator temperature would follow the sodium temperature and introduce an extraneous effect due to its (i.e., moderator) temperature co- efficient. Accordingly, with the reactor at about 65 e s s T T R i T P 33 kw and no power being extracted from either the fuel or the sodium, the sodium blower speeds (2 blowers) were quickly raised from 0 to 2000 rpm. The reactor period immediately started to change and went from infinity to 50 sec in 1.5 min, which, as shown in Fig. 5.4, corresponds to a reactivity change of 3.5 x 1074 Ak/k. The con- sequent rate of reactivity change was 2.33 x 1074 (Ak/E)/min. In this experiment the regulating rod, as well as the shim rods, was held in a fixed position. During the 1.5-min interval, the fuel temperature was changing at a rate of about ~1°F/min and the sodium temperature was changing at a rate of —2.3°F/min. (The rate of change of fuel mean temperature was corrected for the dis- crepancy in the recorded mean temperature, as described in App. K, but no comparable correction was necessary in the rate of change of the sodium temperature,) By applying the previously mentioned value of —9.8 x 1072 (AE/E)/°F for the instantane- ous fuel temperature coefficient to give a rate of reactivity change for the fuel of 0.98 x 1074 (Ak/E)/min and subtracting from the rate of re- activity change observed upon cooling the sodium, it is found that the reactivity change caused by the decrease in the sodium temperature is 2.33 x 1074 (Ak/E)/min (observed) - 0.98 x 1074 (Ak/k)/min (fuel) = 1.35 x 1074 (Ak/E)/min . Then, by applying the observed rate of change of the sodium temperature, the sodium temperature co- efficient is found to be 1.35 x 1074 (Ak/k)/min ~2,.3°F/min =~ -5.88 x 1073 (AR/E)°F . This value is valid if it is assumed that the transient took place rapidly enough that there was no appreciable change in the moderator tempera- ture. The measurements were, however, subject to considerable error because the temperature changes involved were so small as to be quite difficult to detect and a correction was necessary in the fuel temperature because of thermal lags. Moderator Temperature Coefficient | An. .éxpérim'enf'('H-'IO) was conducted in order to determine the temperature coefficient of the moder- 66 ator. During this experiment the fuel temperature was held constant and the speed of the blowers for cooling the sodium was increased to change the temperature of the moderator coolant. Since this was done very slowly, the moderator did cool down, in contrast to the earlier experiment (H-7) in which the sodium temperature was changed so rapidly that the moderator temperature was unable to follow the sodium temperature. The earlier experiment gave information as to the temperature coefficient of the sodium alone, whereas experi- ment H-7 gave the combined effect of sodium and moderator changes. The reactivity was indicated by the position to which the regulating rod was adjusted by the flux servo. The changes in the sodium and the moderator temperatures slightly increased the heat loss of the fuel, and the power had to be increased slightly to keep the fuel mean temperature constant. |In fact, in the middle of the run the reactor mean temperature was lower than at the beginning, but before the final reading was taken, the reactor mean temperature was brought back to its original valve. Table 6,2 gives the pertinent data, The sodium inlet temperature was taken from a record- ing of the temperature of the reflector coolant inlet 4 in. from the bottom of the reactor and the sodium outlet temperature was taken from a recording of the temperature of the reflector coolant outiet 3 in. from the top of the reactor. The time recorded in column one is the time when a reading was taken. The change of the blower speed preceded this time by a few minutes to allow the temperature equilibrium to be established. As an indication of the estab- lishment of the equilibrium, use was made of the leveling off of the trace on the sodium temperature differential recorder. As can be seen from the table, the decrease of the average sodium temperature from 1273 to 1246°F corresponds to a withdrawal of the regu- lating rod from the 8-in. position to the 8,9-in. posi- tion. At the lower temperature the reactor was less reactive and showed that the temperature coefficient was positive. The magnitude of this temperature coefficient was small, (0.9 in./27°F) x 3.33 x 1074 (Ak/EY/ine = +1.1 x 1073 (Ak/R)°F. This co- efficient is, however, the sum of the sodium temper- ature coefficient and the moderator temperature coefficient. Since the sodium temperature coefficient was ~5.9 x 1073, the moderator temperature co- efficient must be +6.9 x 1073 (Ak/E)/°F. This value is, of course, subject to all the inherent < PR wEETr T SRR R ey et - Ty e R g L m vepenr o W EERE R N s = PerEme wgre T s v ey + L reewrere oy - i ¥ I3 ® » b R Lgh . . SR . L T wooEg By all Rl w . wahin e e, Mo b b R - TABLE 6.2 MODERATOR TEMPERATURE COEFFICIENT DATA Fuel Fuel Blower Sodium Regulating Speed Temperature Nuclear ) Mean Temperature P P Rod Time ) (rpm) ©F) o Power Temperature Gradient rem Position (Mw) ' w (°F) (°F) No. 1 No, 2 Inlet Outlet Average (in.) 1735 1313 206 960 1050 1265 1282 1273 8.0 1.98 1745 1314 206 1160 1050 1255 . 1280 1267 8.0 1.98 1752 1313 203 1170 1260 1248 1278 1263 7.6 1.98 1802 1308 200 1340 1250 1238 1272 1255 6.6 1.98 1809 Increased servo demand signal so as to withdraw rod 1812 1311 204 1480 1240 1230 1268 1249 8.4 2.12 1825 1313 205 1470 1480 1225 1268 1246 8.9 2.12 errors of the measurement of the sodium temperature coefficient, as well as the additional errors in the temperatures recorded for this particular experiment. MEASUREMENT OF THE XENON POISONING At 1825 on November 10 a 25-hr run at a power of 2.12 Mw was started for the purpose of measuring the amount of xenon built up in the fuel (Exp. H-11). As discussed in Appendix P, the reactor should have been poisoned by about 2 x 10~ Ak/k after 25 hr, if it is assumed that no xenon escaped from the molten fuel. Not only did the 25-hr run demon- strate that very little of the fission-product gas remained in the fuel but, also, that the reactor possessed phenomenal stability. Except for a minute withdrawal of the regulating rod to compen- sate for a barely detectable drop in the mean re- actor temperature, all readings on both reactor and process instrumentation held constant within ex- perimental error for the 25 hr. It was assumed that the rod withdrawal was due to xenon buildup in the fuel. However, it could have been due to any of a number of minor perturbations, and therefore the experiment demonstrated an absolute upper limit on the xenon poisoning. An abstract of the log book and data sheets during the experiment follows: November 10, 1954 1825 Run started; reactor power, 2.12 Mw; rod position, 9.00 in.; reactor mean temperature, 1311°F November 11, 1954 0635 Reactor mean terfiperature had décreased to 1309°F; rod withdrawn to 9,05 in. 0750 Reactor mean temperature up to 1310°F; rod withdrawn to 9,15 in. 1020 Reactor mean temperature 1310.5°F; rod withdrawn to 9.25 in. 1120 Reactor mean temperature 1311°F 1435 Reactor mean temperature down to 1310°F; rod withdrawn to 9.30 in. 1559 Reactor mean temperature up to 1311°F 1932 Reactor mean temperature up to 1312°F; rod still at 9.30 in.; end of run / 67 g {t is to be noted that the temperature recorded was actually one degree higher at the end of the experiment than at the start; this indicates that the withdrawal of the regulating rod by 0.3 in. may, indeed, have been too much. The withdrawal was unquestionably an upper limit on the compensation needed for xenon poisoning and corresponded to a Ak/k of 1x 1074, This was ), or 5% of the value to be expected if the xenon had not left the fuel {see Appendix P). Removal of the xenon probably occurred by means of the swirling action of the fuel as it went through the pump. As mentioned previously, fission-product gases that probably came from a leak in the gas fittings were detected in the pits early in the experiment. At the conclusion of Exp. H-11 at 1935, the re- actor was operated at various power levels for 2 hr in an experiment (Exp. H-12) for determining the effect of the sodium flow rate on the extracted power. At 2237 the reactor was set at 200 kw (one- tenth full power) and held there by the flux servo for 'IO hr (Exp. H-13). If any appreciable amount of xenon had been built up in the fuel from decay of iodine formed during Exp. H-11, the poisoning effect should have been observed by a compensating rod withdrawal during this 10-hr period. No appre- ciable rod withdrawal was observed, and therefore it was concluded that if there was xenon poisoning from Exp. H-11, it was negligible. POWER DETERMINATION FROM HEAT EXTRACTION The most reliable method for determination of actual reactor power was that based on the energy removed from the reactor in the form of heat. The inlet, outlet, and mean temperatures, the tempera- ture difference of the fuel across the reactor, and the temperature difference of the reflector coolant were continually recorded. The rates of flow of both the fuel and the sodium were recorded and could also be determined from the speeds of the pumps. Since the heat capacities of both the fuel and the sodium were known, the power level of the reactor could be determined by the sum of the values obtained from the following relations: P, =011q,AT, Py, = 0.0343 4, AT, N “where P_ = power from fuel heat extraction (Mw), PNa = power from sodium heat extraction (Mw), 68 volume flow rate (gpm), q AT = temperature gradient across reactor of the fuel or the sodium (°F). Since both the fuel and the sodium were cooled by helium passing across a liguid-to-helium heat exchanger and the helium was cooled by passing it over a helium-to-water heat exchanger, the re- actor power could also be determined from the water Hlow and temperature differences across the helium-to-water heat exchangers. Since the fuel- to-helium and sodium-to-helium heat exchangers are close to their respective helium-to-water heat exchangers, little heat was lost by the helium and therefore practically all of the heat removed from the fuel and sodium was transferred to the water. The heat balance thus obtained was known as the secondary heat balance, while that obtained directly from the fuel and sodium systems was known as the primary heat balance. initial comparisons of the primary and secondary heat balances revealed discrepancies of the order of 50%. Subsequent investigation revealed that the temperature drops of the primary heat balance were in error. 1he temperatures were obtained from a few thermocouples located on fuel and sodium lines within the reactor thermal shield that read considerably lower than several thermocouples located external to the thermal shield on the fuel and sodium lines to and from the reactor. This anomaly is discussed further in Appendix K. All reactor inlet and exit femperatures were subse- quently based on the external thermocouple readings. A detailed analysis of the extracted power was made (App. L) for the 25-hr xenon experiment primarily because of the certainty that equilibrium conditions had been established; also, this high- power run at more than 2 Mw contributed almost two-thirds of the megawatt-hours logged during the entire reactor operation. For this run the primary heat balance showed an extracted power of 2.12 Mw, and the secondary heat balance gave an ex- tracted power of 2.28 Mw. The difference between the two determinations was 7%. Furthermore, for this run about 75% of the power was removed by the fuel, about 24% by the sodium, and about 1% by the rod cooling system. Heat balances and, hence, power determinations were made for a number of other experiments during the high-power operation; these are tabulated in Appendix L. One such heat balance that was of particular interest was obtained during the maxi- § .t - e - ¥ o WrumeRmme MR-@ R R PETUN s othygerr ¥ Eh g T geE T o L -R_,. ?@ - ” sy ETER Y Trr Wm .\,‘Lé-m-m - Py T ey s S . e i b g B e - ol il ko g bR o R sk WAL e i BB, ek &y Ll o B Ay, Lk Ld . ha\ B ke - o Etadli oo B sak 4 LR wibhe & ki bt il R ik Ao ki e o xedu mum power run (Exp. H-14). For this run the mean temperature was raised to 1340°F (to avoid a low- temperature reverse), and the fuel and sodium system blower speeds were increased to their respective maximums, At equilibrium the power levels indicated by the primary and secondary heat balances were 2.45 and 2.53 Mw, respectively. The line temperatures throughout the fuel and sodium systems during this experiment are shown in Figs. 6.5 and 6.6, respectively, The tempera- tures indicated along the lines are the actual thermocouple readings of each point at equilibrium. It may be noted that the outlet fuel line tempera- ture averaged about 1580°F. The temperature measured at the reactor was considerably lower, as shown in Fig. 6.7, which shows a portion of the instrument panel during this experiment. REACTOR KINETICS " A distinctive control characteristic of any circu- lating-fuel reactor is that the reactor power is determined solely by that part of the system which is external to the reactor, i.e., the heat extraction equipment. In the power regime, control rods do not appreciably influence the steady-state power production, but the power extracted from the reactor does influence the reactivity of the reactor and hence renders the reactor controllable. In order that such a system be acceptable as a power reactor, it is requisite that the reactor have a negative temperature coefficient of reactivity. One of the most gratifying results of the ARE operation was the successful demonstration of its large negative temperature coefficient. This temperature coefficient made it possible for the reactor to maintain a balance between the power extracted from the circulating fuel and coolant and the power generated within the reactor. The tem- perature cycling of the fuel was the mechanism by which equilibrium was maintained. ‘ A thorough understanding of control processes in a circulating-fuel reactor with a negative tem- perature coefficient is necessary for an appreciation of the kinetic behavior of such a power reactor in the power regime. An important purpose of the ARE was the observation of the kinetic behavior of the reactor under power coupling to its load when perturbations in the reactivity were intro- duced. Transient conditions could be induced both by control rod motion and by variation of the external power load. Information on the kinetic behavior was obtagined from a number of experiments that were conducted during the period of power operation. These are described in the following sections., Preceding the discussion of these experi- ments is a detailed qualitative description of the temperature cycling of the circulating fuel which is inherent to all kinetic phenomena. Reactor Control by Temperature Coefficient The control of a circulating-fuel reactor with a negative temperature coefficient can best be under- stood by following the course of a *‘slug’’ of fuel as it traverses the ARE system. For a description of the course of a slug, it is assumed that at time zero the system is in equilibrium, with isothermal temperatures throughout; in this condition no power is being extracted. At time t the fuel system helivm blower is turned on. It is also assumed that at this time the slug of fuel under observation is just entering the heat exchanger. In passing through the heat exchanger the slug is cooled by the helium blowing through the heat exchanger. About 20 sec later the cooled slug enters the re- actor and is registered as a decrease on the inlet temperature indicator. Because it is cooler than the fuel it is displacing, its density is greater and therefore the number of uranium atoms per unit volume is greater. This results in a greater fission rate and, hence, a greater reactivity which, in turn, increases the power generated. The temperature of the slug rises as it passes through the reactor because of increased power generation. The rise in temperature of the slug results in its expansion and decrease in reactivity. This, in turn, lowers the rate of power generation. Eight seconds after the slug enters the reactor it passes out into the outlet line and is registered as an increase in temperature on the outlet fuel temperature indi- cator. In 47 sec from the start of its journey it is back at the heat exchanger to be cooled again. The masses of fuel behind the initial slug follow the same pattern so that, since the fuel is a con- tinuvous medium, the power generated in the re- actor® will rise until the increased reactivity due to the incoming fuel and decreased reactivity due to the outgoing fuel attain o balance. At this point equilibrium is reached between power ex- tracted by the helium and the power generated in 3With a power reactor it is appropriate to speak of two types of power: the nuclear power, i.e., the total power generated within the reactor; and the extracted or useful power, i.e., the power removed from the fuel and coolant by cooling. 69 . ¢ i 3 gt T —_—— e e — ¥ E 0L ORNL=-LR-DWG 6415 MAIN PUMP VALVE U-2 HEAT EXCHANGER W NO. ! Lo M S PRESSURE TRANSMITTER REACTOR —_ T . KEY: . | T 100 SERIES NUMBERS ARE i FUEL LINES. | ALL FOUR DIGIT NUMBERS ARE | 1563 TEMPERATURES IN °F FREEZE VALVE | B-64 FREEZE VALVE —- B-13 S ; —— VALVE U-63 A T Qe \ In N \8 - | Ll o . L E \ | ~ ) X X i ; ! rh e Ty e me L3 e s rrem et 4 S e W T Ty P E TR e I ppe g O TS Mg I npPrttmer TP et Sk B e o ow ok . © e AR e e, o, S by et been effected and equilibrium re-established, the nuclear power and the extracted power will again be equal.4 _ The control of the ARE by extracted power demand is illustrated in Fig. 6.8, which shows the tracings made by several of the control room instrument recorders during power operation between 1258 and 1329 on November 10, 1954. The reactor behavior is demonstrated by reading the tracings right to left, proceeding as follows: At 1258 the reactor was in an equilibrium condition at about a 50-kw power level. At 1259 the fuel system helium blower was turned on and its speed was increased to 500 rpm. The reactor power, as noted on the Log N chart, rose on a 10-sec period to slightly over 1 Mw, ‘‘overshot’’ its mark, fluctuated somewhat in the manner of a damped oscillator, and, finally, some 4 min later, came to an equilibrium power of around 900 kw. Soon,s but not immediately, after the blower was turned on, the reactor fuel inlet temperature began to drop and the fuel outlet tem- perature began to rise, while the reactor mean temperature began to drop slightly. Each of these temperatures showed the oscillatory phenomenon noted with the Log N recorder. The reason for the drop in the mean temperature was that both the decrease and rate of decrease of the inlet tempera- ture were greater than the corresponding increase and rate of increase of the outlet temperature. The inlet temperature dropped from 1335 to 1256°F% at a rate of 1.54°F/sec and the outiet temperature rose from 1405 to 1475°F at a rate of 1.36°F/sec; as a result the mean temperature fell fram 1370 to 1365°F at a rate of about 0.1°F/sec. The fact that the mean temperature dropped {and this was a characteristic phenomenon noted throughout the power operation) rather than remaining constant can be explained, at least in part, by the distortion of the flux patterns within the reactor. The highest fluxes were toward the inlet fuel passages, and the lowest fluxes were toward the outlet fuel passages {(cf., App. O}, At 1310 the blower speed was increased from 500 4On several occasions the nuclear power, as indicated on the Log N recorder, rose temporarily above 2.5 Mw, but the extracted power, limited by the system capacity for removing heat, never exceeded 2.5 Mw. Errors in marking the charts could conceivably ac- count for as much as 1/2 min of the time ditferences noted. This phenomenon of time lag is discussed later in this section, ‘ 8T he temperature readings given here and elsewhere in this section have been corrected by the method given in Appendix K, to 1000 rpm, and the power rose from 1 to 1.8 Mw on a period of 7.5 sec, with corresponding changes of the inlet, outlet, and mean temperatures. Al- though the power increase in this case was compa- rable to the first power increase, the rates of rise of both the power and temperctures were much less, being on the order of one-half as much for the temperatures. The effect of withdrawing the regulating rod 3 in. is demonstrated with the next rise at 1312 in Fig. 6.8. Up to this time the trace of the regu- lating rod position was constant, since it was not on servo. With the withdrawal of the rod the rise of the nuclear power was immediate. After some delay, all of the temperatures showed a rise. Interestingly enough, the outlet temperature rose 16°F, the mean 12°F, and the inlet 9°F. This pattern, with the outlet temperature showing the greatest change, was characteristic of the temper- ature changes incurred by shim rod and regulating rod movement during power operation. After the rod movement the temperatures leveled out at the higher values, but the nuclear power level slowly drifted back to the equilibrium position, At 1321 the blower was shut off and the charts show the result of the shift toward isothermal (no power) conditions. Two minutes later, when the nuclear power had decreased to about 600 kw, the blower was again started and its speed was in- creased to 1500 rpm. The nuclear power level rose initially to 2.6 Mw, and it again went through an oscillatory cycle before settling down to 2.2 Mw. The corresponding temperature oscillations were again noted. These oscillations were of sufficient strength to determine an oscillation period of about 2.25 min. An interesting observation at this point was that the time lag of temperature response for a higher initial power (600 kw compared with 50 kw) was only about one-half the lag during the first rise to power at 1259. This phenomenon had been noted previously in connection with the various temperature coefficient experiments (cf., App. O). Table 6.3 shows the data obtained from the nuclear and process instruments during the typical operation period shown in Fig. 6.8. Startup on Demand for Power (Exp. H-6) To illustrate how completely the reactor was a slave to the load, an experiment was performed in which the reactor was brought to subcritical and then taken to critical on temperature coefficient (i.e., by the power demand and without use of 73 i i —— - - L T FUTRE SR o T T TR R ~{ P R e FUEL QUTLET TEMPERATURE (°F) TEMPERATURE (°F) FUEL MEAN FUEL INLET TEMPERATURE (°F) REACTOR POWER {Mw) REGULATING ROD POSITION (in.) 1300 1350 1300 1300 1250 1200 © mo- N o 0.05 S 7 1328 BLOWER SPEED REDUCED TO LEVEL OUT POWER AT 2 Mw ! BLOWER STARTED; REGULATING ROD BLOWER ON; SPEED UP TO BLOWER WITHDRAWN SPEED RAISED TO 1500 rpm OFF 3in. 1000 rpm ' ! ' ' ORNL~LR-DWG 6556 FUEL SYSTEM HELIUM BLOWER STARTED; SPEED RAISED TO 500 rpm* REACTOR FUEL OUTLET TEMPERATURE -~ REACTOR FUEL MEAN TEMPERA REACTOR FUEL INLET TEMPERATURE - LOG & POWER RECORDER 1327 1325 e R R S T s g - mepmer gne MR s hmes TING ROD POSITION 1324 1323 {322 1321 314 1343 4342 1341 1310 4309 1304 TIME OF DAY, NOV. 10, 1954 Fig. 6.8. Typical Reactor Behavior During Power Operation. FROOCUES M g oW o pWECUTT TN YT v pTR TR T W ITNT R ST TrROmRYT 1303 1302 n 1301 1300 1259 4 TR s W B SRSt M FCE T r e e o ¥ g 1258 O ISt R Gl L s % - B K » TABLE 6.3. NUCLEAR AND PROCESS DATA OBTAINED DURING A TYPICAL OPERATION PERIOD (NOV. 10) Operation Performed Type of Type of Data Fuel System Fuel System Regulating Fuel System Fuel System Measurement Obtained* Blower Speed Blower Speed Rod Blower Speed Blower Speed Increasing Increasing Withdrawn Decreasing Increasing Time 1300 1310 1312 1321 1323 Log N power P, Mw 0.0566 0.99 1.58 1.56 0.57 Py Mw 1.07 1.70 1.94 0.57 2.17 5P, Mw 1.013 0.71 0.36 0.99 1.60 i, sec 28 41 9 60 81 OP/0t, Mw/ sec 0.0363 0.0173 0.04 0.016 0.198 T sec 10 75 33 63 54 Fuel outlet T, 1 °F 1405 1485 1527 1546 - 1491 femperature T, » °F 1474 1527 1543 1491 1566 8T, °F 69 42 16 —55 75 Ot, sec 45 56 24 75 58 8T /61, °F/sec 1.54 0.75 0.67 -0.73 1.29 Fuel inlet T; v °F 1335 1269 1231 1239 1314 temperatore T, o °F 1256 1227 1240 1314 1208 8T, °F ~79 ~42 9 75 -106 St, sec 58 53 42 75 88 6T /1, °F/sec -1.36 ~-0.79 0.21 1.00 ~1.20 Fuel mean T °F 1370 1377 1379 1388 1402 femperature T, o °F 1365 1377 1391 1402 1387 8T ., °F -5 0 12 14 -15 St, sec 52 33 75 73 5Tm/5z, °F/sec —~0.096 0 0.36 0.19 —-0.20 Reactor fuel Ar,, °F 70 216 296 287 177 AT AT,, °F 218 300 303 177 358 XAT), °F 148 84 7 -110 181 Ot, sec 52 54.5 33 75 73 S(AT)/Bt, °F/sec 2.87 1.54 0.21 ~1.47 2.49 Regulating rod dy, in. | 7.60 movement dy, in. 10.37 75 T T e T R £ e epegrm, T T Y W e o o TABLE 6.3 (continued) Operation Performed Type of Type of Data Fuel System Fuel System Regulating Fuel System Fuel System Measurement Obtained* Blower Speed Blower Speed Rod Blower Speed Blower Speed Increasing Increasing Withdrawn Decreasing Increasing Regulating rod Sa’], in. 2,77 ."’°"er_“e”* 01, sec 9 Od/At, in./sec 0.32 Ak/k, % 0.0914 (Ak/k)/St, %/sec 0.011 Fuel system helium §4e TPM 0 500 1000 0 blower speed s_, tpm 500 1000 0 1500 Os, rpm 500 500 -1000 1500 Ot, sec 14 14 60 42 Os/8t, rpm/sec 36 36 ~17 36 *The following symbeols are used: Quantity Subseript P power 1 initial cendition t time 2 final condition T period o outlet condition T temperature i ‘inlet condition d position m mean £ multiplication factor s speed O change in quantity A difference between two quantities The 6's are the times required to go from the initial condition to the final condition at the greatest observed rate of change. The temperatures quoted have all been corrected by the method described in Appendix K. rods}). The experiment consisted of tworuns. In the first run the reactor was taken directly from 100 kw to slightly over 2 Mw. In the second run the reactor was taken to power from a subcritical condition. The progress of these experiments, as recorded by the thermocouple recorders on the inlet and outlet sides of the six individual fuel tubes,is shown in Fig. 6.9. Reading from right to left, the reactor was at 200 kw power at 2300 on November 9 at the beginning of the experiment. The fuel system helium blower speed was increased to 1700 rpm slowly, starting at 2307. Ten minutes later the sodium system coolant blowers were turned on. As the power rose to about 2.5 Mw, a temperature difference of some 320°F appeared across the re- actor tubes. This is shown in Fig. 6.9 by the parting of the inlet and outlet temperature indi- cations of the reactor fuel tubes. This temperature 76 difference remained constant until the blower speed was reduced at 2340, At 2345 the inlet and outlet tube temperatures were nearly the same again at 200 kw power. A sharp drop in the tube tempera- tures at this point corresponded to the insertion of the shim rods. The sharp increase in the temperature differential across the tubes at 2353 was the result of a steep rise in power when the blower speed was increased from 0 to 1700 rpm. At 0005 on November 10, the regulating rod was first entirely inserted (from 7 in. withdrawn) and then fully withdrawn. This motion was followed by a drop and then a sharp rise in the fuel tube temperatures. This action marked the end of Run 1. During Run 2 the reactor was brought subcritical by turning off the fuel system helium blower and inserting the regulating rod, starting at 0016. The sodium system helium blowers had been operating ¥ e AT W T e THE U o S et e T el e e b B Y TR P WM TR et Mo oo T TSSO TS e W Y ¢ e T VTR S rp—: i s WG Uy e ook bt S e R AV - kR ki ikl Wk Ny « Ll : ol e MY002~ H TIVOILIMD v3 ORNL-LR-DWG 3946 i | : T "WNAWIXYIN Ol Q33dS H43M0T8 WNITT3IH W3L1SAS 7304 3SVIUONI ATMOTIS NO ¥3Imong}: WNIT3H W31SAS WNIAOS b | | B 1 "", RN §'¢ 1V NOILVYY3dO t 043z oL [ Q334S YIMOTE WNIMIH G W3LSAS 13n4 430NQ34: LTELL It E LI AT i Ei Lt st g SA0Y WIHS O3LY3SNI {Q@33ds 11n4) 43M0718 WNIT3H W3LSAS 13Nn4 NO @3NHNL ! ! aod ONILVINS3Y Q3A0N H¥3IMO1E WNIN3H 13Nd4 430 Q3NYNL PLE BTV LT AT L WOLLINOENS ¥OLOV 3N 'Q0Y ONILVING3Y QILHISNI (033dS 1IN} ¥43MOT8 WNIT3H /[W3LSAS 13N NO 3NuNL 43345 ¥3mo18 035na3y ATTUDILYWOLNY HOTHILNI FYNLVYIINIL MO |\ 430 syamo18 BN GNV 304 Q3183SNI SA0Y _ _ 1900 1800 1700 t600 ceslodiabtion w_.,_ Vv ONILYY3dO ] | 1500 TUBE OUTLET: TEMPERATURE 253~ 1400 300 - 1200 : TUBE INLET TEMPERATURE = -~ REACTOR COOLED 8& - BROUGHT CRITICAL ¥ 6Geé 00¢g 1600 o= m o Ll = [ ’ @ 2 o | g@8) © oS % 1500 B4 Sel e 80 5L A S Wlmpszl = ] 4 a r 5 ¥31 9§ 1w {5 Z 1400 w52 A 2 100 Wy & z 2T ga @ Lz = 9 =5 mi \ b z Be8 + n . The total effect of both types of reactions was to give a strong multiplication every 47 sec. During the first pip shown in Fig. 6.17, the flux rose by a factor of 2; with succeeding pips, the multipli- cation became less. The average mean decay time of the first four pips was 72 sec. For the two longest lived groups of delayed neutrons the mean lives are 32 and 80 sec.'? Therefore, the attenu- ation of the pips closely followed the theoretical delayed neutron decay. A remarkable feature of Fig. 6.17 is that the pips could be distinguished for about 12 min. Since after the first 3 or 4 min the delayed neutron emitters were gone, the re- maining effect was due solely to photoneutrons. The scram behavior at high power was very similar, as observed with the use of the safety chambers. The fission chambers could not be used at high power, and therefore the first few seconds after a scram could not be observed with them. 1314 experiment L-5 the regulating rod was calibrated by the periods induced by rod motion {cf., chap. 4). Mqussfone and Edlund, op. cit., p 65. 93 - k. 94 COUNTING RATE (arbitrery scale) C.R.M. NO.1; RUN 7, EXP. LS 2212-30; NOV. 7, 1954 THIS FIGURE IS A COMPOSITE OF TWO CHARTS SHOWN = HERE PIECED TOGETHER TO GIVE A GRAPHIC REPRE- = SENTATION OF THE DECREASE IN COUNTING RATE IN THE CIRCULATING SLUGS FOLLOWING A SCRAM, THERE — |$, HOWEVER, SOME ERROR IN BOTH ABSCISSA AND - ORDINATE ACROSS THE JOIN, —~s———— TIME Fig. 6.17. Circulating Slugs After a Scram, ORNL-~LR-DWG 3855 r . : : ¢ k ¢ ‘t. s - e et ST TR .8 S rEY " s err R m porr ot wer e~ T or e e et v ¥ ¥ E. | o ibas B W bk - . Ak ol e . e ki W A o o 4% FINAL OPERATION AND SHUTDOWN The last scheduled experiment conducted on the reactor was the measurement of the xenon buildup following the 25-hr run at 2,12 Mw., The 10-hr period of operation at one-tenth full power was concluded at 0835 on November 12. During the following ]]1/2‘|1r period from 0835 to 2004, when the reactor was shut down the final time, the oper- ation of the reactor was demonstrated for Air Force and ANP personnel who were gathered for the quarterly ORNL-ANP Information Meeting. The demonstrations inciuded repeated cycling of the foad by turning the blowers on and off, and group movement of the three shim rods to change the reactor mean temperature. The information ob- tained during this time on” the dynamic behavior of the reactor is described above under the subtitle “’Reactor Kinetics."”' Also during the final 11%-hr period of operation, two complete surveys of system temperatures were taken with the reactor at maximum power. It was during one of these runs that the equilibrium fuel outlet temperature of 1580°F and the total reactor power of 2.45 Mw was attained (cf., App. L). Since all operational objectives of the experiment had been attained and it was estimated on the basis of time and the assumed reactor power that the desired integrated power of 100 Mwhr would be attained by 2000 on Friday, November 12, it was decided to terminate the experiment at that time. Colonel Clyde D. Gasser, Chief, Nuclear Powered Aircraft Branch of WADC, who was then visiting the Laboratory, was invited to officiate at the termination of the experiment. At 8:04 PM on November 12, with Colonel Gasser at the controls, the reactor was scrammed for the last time and the operation of the Aircraft Reactor Experiment was brought to a close. A photograph taken in the con- trol room at that time is shown as Fig. 6.18. At the time the experiment was terminated it was believed that the total integrated power was more than the 100 Mwhr which had been prescribed as a nominal experimental objective, but subsequent graphical integration of the Log N charts, Ap- pendix R, revealed that the actual total integrated power was about 96 Mwhr. At the time the reactor was scrammed the sodium system had been in operation (circulating sodium) for 635.2 hr and the fluoride system for 462.2 hr. Of this total fluoride circulating time 220.7 hr were obtained with the reactor critical and 73.8 hr after the reactor was first brought to power. Although the nuclear operation was concluded on Friday evening, the fuel and sodium were per- mitted to circulate until the following morning, at which time they were dumped into their respective dump tanks. The final phase of the experiment, including the dumping operation, subsequent analy- sis and recovery of the fuel, and examination of the systems and components for corrosion, wear, radiation damage, etc., are to be discussed in a subsequent report. Since much of the data cannot conveniently be obtained until the radiocactivity has decayed, the last report will not be forthcoming immediately. 95 T g T T e o B o e Egpe T P e n 'S.J CROMER | THTEET Uy YTY Uy g vpomewwent TROTT UUREme wmer sempe r T Yot ompE T b gyt e e oy gk T e - g e Cper - il g SR i, BB ek, e e By k. L ek E 3 7. RECOMMENDATIONS No comprehensive report is complete without conclusions, discussion, and recommendations. In this report the conclusions are presented in the summary at the beginning, and the discussion is incorporated in the text and, especially, in the Appendixes. Furthermore, certain alterations and modifications that would have been desirable in the conduct of the experiment are implicit in the discussions throughout the report. However, the changes, if effected for any similar future ex- periments, could lead to substantial improvements in design, instrumentation, operation, nd interpre- tation. Accordingly, a list of recommendations based on the ARE experience is presented; no significance is inferred by the order of listing. 1. The operating crew should maintain the same shift schedule as the craft labor — electricians, instrument mechanics, pipe fitters, etc. — assigned to the operating crew, and all members of one crew should have their off-day at the same time. In particular, there should be four operating crews that maintain the same rotating schedule as the rest of the plant. 2. The cause of the obvious discrepancy between the temperatures read by the line thermocouples close to the reactor and those further removed from the reactor should be ascertained. No physi- cal phenomena, with the possible exception of radiation, have been proposed, to date, which could account for the observed difference. 3. In order to measure the extracted reactor power from the secondary or tertiary heat transfer mediums, these systems should be adequately instrumented for flow rates and temperatures. Such instrumentation might also make possible a thermodynamic analysis of the performance of the various heat exchangers. _ 4, Therrinocourpkles on the outer surface of fuel and sodium tubes in the heat exchangers should be installed so that they read wall temperatures rather than an intermediate gas temperature. 5. Of the two thermocouples which were required to measure the reactor mean temperature and the two required to measure the temperature gradient, one was electrically insulated from the pipe wall. The insulation effected a time lag between the wall and thermocouple temperatures of about 15 sec. Thermocouple installations for obtaining time-dependent data should be made so that the wall temperature can be read without a time lag. 6. To heat small {up to 3/8 in. OD) lines and associated valves, tanks, etc. to uniform high temperatures (~1400°F) requires a surprising degree of precision in the installation of both heaters -and insulation. Furthermore, where calrod heaters are employed, they should be installed on opposite sides of a line, and the heaters and line should be jointly wrapped with heat shielding before insulation is applied. 7. To control accurately the temperature of any system, thermocouples should be installed at any discontinuity of either the heaters (i.e., between adjacent heaters) or the system (i.e., at the junction of two lines, etc.), and, with the exception of obviously identical installations, each heater should have its own control. 8. Where gas lines are subject to plugging due to the condensation of vapor from the liquid in the system to which the lines are connected, it is necessary either to heat the lines to above the freezing point of the vapor or to employ a vapor trap. While a vapor trap proved satisfactory in preventing the fuel off-gas line from plugging, it was not possible to heat the sodium off-gas line sufficiently (because of temperature limitation of the valves) to prevent the gradual formation of a restriction. Gas valves that can be operated at higher temperatures and are compatible with sodium are needed. 9. The double-walled piping added a degree of complexity to the system that was far out of pro- portion to the benefits derived from the helium annulus. Parts of the annulus were at subatmos- pheric pressure, and no leak tests were made on the fuel system once it was filled with the fluoride mixture. The helium flow was not needed for distributing heat in the sodium system, and it was of uncertain value in the fuel system. Conse- quently, future systems should not include such an annulus. 10. The use of any type of connection other than an inert-arc-welded joint in any but the most temporary fuel or sodium line should be avoided. 11. All joints, connections, and fittings in the off-gas system should be welded, and, in general, they should be assembled with the same meticulous care that characterized the fabrication of the fuel and sodium systems in order to minimize the pos- sibility of the unintentional release of fission gases. 97 e s e g - o s . e — i 5 S+ v b ' - f: - A R 12. As a secondary defense in the event of the release of activity in the reactor cell (i.e., pit), the cell should be leaktight. The leak-tightness of the numerous bulkheads out of such a cell should be carefully checked. 13. The air intake to the control room was lo- cated on the roof of the building, and thus adverse meteorological conditions readily introduced off- gas activity into the control room. With an airtight control room equipped with its own air supply (or a remotely located filtered intake), the control room operations could continue without concern for the inhalation of gaseous activity. 14. The off-gas monitrons should be shielded from direct radiation. Furthermore, although these monitrons were nof needed during the experiment, ~ they are known to build up background activity which would eventually mask that of the gas they are to measure. The development of monitrons in ‘which activity will not aceumulate is recommended. 15. The radiation level and the airborne activity throughout the building were measured by monitrons and constant air monitors, respectively. The data from both instruments should be continuously recorded. Furthermore, the output of the air moni- tors should be modified by the addition of a dif- ferentiating circuit so that the recorded data would give better measures of the airborne activity. 16. The helium ducts leaked to the extent that it was possible to attain only a fraction of the desired helium concentration therein. They were of the conventional bolted-flange design and should have been modified to permit seal welding of all joints; they should be subjected to stringent leak tests. 17. The helium consumption for the last three and one-half months of the ARE experiment was 2/3 million standard cubic feet (over 3000 standard cylinders). The average consumption rate during the last two weeks of the experiment was about 8.5 ¢fm, and peak consumption rates of 25 c¢fm were recorded. This extraordinarily high helium con- sumption could be reduced by the use of dry air or nitrogen for much of the pneumatic instrumen- tation in which helium was used. 18. Various valves in both the gas and liquid systems leaked across the valve seats. The valve development program should be emphasized until valves are obtained that can be depended upon as reliable components of high-temperature (~1200° F) ‘sodium, fluoride, or gas systems. Furthermore, either limit switches that would operate at higher 98 temperatures to indicate valve open or closed should be developed, or the existing limit switches should be located in cooler regions. 19. The use of frangible disks to isolate the standby fuel pump did not enhance the feasibifity of continuing the experiment in the event of the failure of the main fuel pump during high-power operaticn. The frangible disks are objected to in that they require a nonreversible operation. In general, leak-tight valves should be developed that may be used rather than the frangible disks. 20. Although the ARE, as designed, was to incorporate numerous ‘‘freeze sections,’’ only two were included in the system as finally constructed, and the operability of these two was so question- able thdt their use was not contemplated during the course of the experiment. Such sections should either be eliminated from consideration in future systems, or a reliable freeze section should be developed. 21. Much useful information would be obtained if provisions could be made for sampling each liquid system throughout the operation. This was possible on the ARE only prior to the high-power experiments. 22. The variable inductance-type flow and level indicators should be converted to the null-balance type of instruments in order to eliminate the temperature dependence of the pickup coil signal. Furthermore, these coils should be located in a region at much less than 1000°F in order to in- crease coil life. 23. Although numerous spark plug probes were satisfactorily employed to measure levels in the various tanks, the intermittent shorts that were experienced with several sodium probes might have been avoided if clearances between the probe wire and the stand pipe in which the probes were located had been greater. 24. The use of mercury alarm switches on vibratory equipment where there is little leeway between the operating and alarm condition will give frequent false alarms and should therefore be avoided. 25. Weight instruments are of questionable accuracy in a system in which the tank being weighed is connected through numerous pipes to a fixed system, especially when these pipes are covered with heaters and insulation and are subject to thermal expansion, 26. The flame photometer is an extremely sensi- tive instrument for the detection of sodium and vk g EE "o TC T T oo s Mo mew T rE o il e S mewrrr ¥ e e e st g W e PR e © i g e e — 7T ——ryy e -+ e ~ b ki G LA k. e R, o R L - e o LR bW el Pt v, . - M NaK. However, in employing this instrument to detect the presence of sodium (or NaK) in gas, the sampling line should be heated to about 300°F, 27. The magnet faces in the shim rods should be designed so that dirt particles cannot become trcpped thereon and thus require higher holding currents. ' 28. The two control points, i.e., the upstairs control room and the basement heater and instrument panels, should have been more convenient to one another in order to effect the greatest efficiency of operation. 29. The communications system in the building was inadequate. Except for the auxiliary FM system, which was frequently inoperative, there was only one phone in the control room, which was on the main station of the PA system. How- ever, the PA system was not capable of audibly supporting two conversations at the same time. Accordingly, the capacity of the PA system should be increased so that as many as 4 or 5 conver- sations may be simultaneously effected. Also, more outlets should be provided both in the control room and at other work areas in the building. 30. A megawatt-hour meter should be instalied (possibly on the log N or the AT recorder) in order to provide a continuous measure of the integrated power as the experiment progresses. 31. In order to analyze various related time- dependent data (as required, for example, in the determination of the various temperature coef- ficients), it is necessary that the recorder charts be marked at the start of the experiment. While the charts can be marked by hand, this was a source of error which became very important in the analysis of data from fast transients. Accordingly, all such charts should be periodically and simul- taneously marked by an automatic stamper. Further- more, each chart should be stamped with a dis- tinctive mark that would positively identify it. 32. To analyze the kinetic behavior of a reactor system it is necessary to have a recorder chart of the time behavior of all equipment which can introduce transients in the reactor, as well as charts of the process data and conventional nuclear data. Therefore the helium blower speeds and shim rod positions should have been recorded continually. 33. The recorded data and the data sheets comprise a fairly comprehensive picture of the experiment. However, even if all the dota are recorded they are of value only to the extent that meaningful interpretations can be made. While automatically marking the data charts will be helpful, it will still be necessary to rely on log book information which should be recorded in great detail, possibly to the extent of making this the only responsibility of a shift ‘‘historian.”’ 99-106 T T TR T T T e g TR e o T T o a4 dahd A v iAW G . oW ndl iy Ll bl i ot GGG G g g e . s < {v i ok -;‘m._w;.%s. S ey, @he ek il . v b . ..., e b Ak, | Rk Appendix B SUMMARY OF DESIGN AND OPERATIONAL DATA This appendix tabulates both the design and operational data pertinent to the aircraft reactor experi- ment. Most of the design data were extracted from two design memoranda,'+2 although some values had to be revised because of subsequent modifications in the design. In addition to the design data, the experi- mental values of the various system parameters are included. All values obtained experimentally are shown in italics and, for comparative purposes, are tabulated together with the corresponding design values. There are, of course, numerous design numbers for which it was not possible to obtain experi- mental numbers. The flow diagram of the experiment is presented in Fig. B.1. The values of temperature, pressure, and flow given on this drawing are design values — not experimental values. w. B. Cottrell, ARE Design Data, ORNL CF-53-12-9 (Dec. 1, 1953). 2y, B. Cottrell, ARE Design Data Supplement, ORNL CF-54-3-65 (March 2, 1954). DESCRIPTION 1. The Reactor Experiment Type of reactor Neutron energy Circulating fuel, solid moderator Thermal and epithermal Power (maximum} 2.5 Mw Purpose Experimental Design lifetime 1000 hr Fuel NaF-ZrF ,-UF, (53.09-40.73-6.18 mole %) Moderator BeO Reflector BeQ Primary coolant The circulating fuel Reflector coolant Sodium Structural material Inconel Test stand Concrete pits in Building 7503 Shield 7"/2 ft of concrete Heat flow Fuel to helium to water 107 2. Physical Dimensions of Reactor {in.) Cold Hot (70°F) (1300°F) Core height : 35.60 35.80 Core diameter 32.94 33.30 i Side reflector height 35.60 35.80 .k Side reflector inside diameter 32,94 33.30 X Side reflector outside diameter 47.50 48.03 i Top reflector thickness 4,00 4.25 ¥ Top reflector diameter 32.94 33.30 T Bottom reflector thickness 4.93 4.98 E Bottom reflector diometer 32.94 33.30 ; Pressure shell inside diameter 48.00 | 48.55 i Pressure shell wall thickness 2.00 2.02 E Pressure shell inside height 44,50 45.00 N Pressure shell head thickness 4.00 4.04 | Fuel elements 66 paralle| Inconel tubes containing £ 3 the circulating fuel. The tubes : ! were connected in six parallel cir- ‘ cuits each having 11 tubes in series. Each tube was 1,235 in. 0.D., with a 60-mil wall, 3. Volumes of Reactor Constituents (Cold) a. Inside Pressure Shell Volume (ft3) N BeD Fuel Na Inconel Rods? Total Core 14.55 1.33 0.80 0.37° 0.50 17.55 Side reflector 17.68 0 0.91 0.10¢ 0.267 18.95 Annulus outside reflector 0 0 0.58 0.19 0 0.77 Top reflector 0 0.21 1.62 0.08% 0.06 1.97 Bottom reflector 0 0.24 1.67 0.44° 0.07 2.42 Space above reflector and annulus 0 0 2.14 0.04¢ 0.034 2.21 Space below reflector and annulus 0 0 2.17 0.52¢ 0.04¢ 2.73 Total 32.23 1.78 9.8%9 1.74 0.96 46.60 Total volume inside inner rod sleeve. blncludes the |Inconel-¢lad stainless steel rod sleeves. €Includes all three sleeves around fission chambers TP T e R gy BT P TR T TS roghgn g 4 ncludes volume of insulating material around inner fission chamber sleeve. b. Operating Fuel System Volume (ft3) Cold Hot A Core 1.33 1137 External system (to minimum pump level) 3.48 3.60 Pump (available above minimum level) ~ 1.65 ~ 1,70 - Total 6.46 m 108 =N Gidundies ., kil oty o i ] i i e oo AT ¢ i SR, Gtk , MM i i . e i . - ki ki el e e o el e - G et ] d i el o b it i Ry il . i MR, 3 et o . il i il s e aian dbaeiamk e ndeibai s o DWG. 220478 FILTER FROM DYESEL TO STACK —e 836 QUTSIDE FILTE! AIR COMPRESSER BULK HEAD | OVER HILL GAS LINE AIR COMPRESSOR FROM I wsipEPIT |1 u4s 828 {i]__ 581 TO DRAIN ///7/?1'// b 7 I! PRESSURE 4-in. GATES _ J| P155 2-in. GATE SWITCH o ™ T - P156 B e o T T ’JT‘: Ly ! i|c, » o S 234 1 Y4-in. LINE 5 ;7 g4 &lLE g2 €12 e e 2 v, Bl |SEs (B 5 e O CaTCH venT _______ _. o |lwx 2 |& Nak VAPGR r+{ RESERVOIR '} Azin LINE | {__THE&E_J%::«q—fix«-q‘ ol |j§ 8 £§ § v § = TP BASIN | \‘-T—'—'———/ ‘;76-in.GATE - LINE : { } Ii?o w Eg W # W B | ! A —_—— | 2 S @ o Pi57 i PIT: L] < g |5 g - I AL v T V\ i S0 5| &2 ] 3-in. VALVES ™ ~ - 1 A : ____________ | ———————————— 51 | DRYER—_ L B8T Ti L AR INTAKE FROM PIT\\ - “ |! Ny-COOLED I &5 NO. NO.3 I NO. § [5160 AN AND 4 l&-in. LINE A Nz MANIFOLD CATCH BASIN 2 _!' vtae { se 10 1912144 ] LT_I CHARCOAL ADSORBERS - ) 505 505 ] ‘o —) | W He ) v - He U U Na DOLLIES s 7 prse &% RETA ’ ‘{m—(:)uao 4 801 /\"\ 801 P187 N ug4 & 879 812 © 0 3 STANDBY PUMP NQ, 2 \ 309 307 / 21 ‘Pump NO.2 ~ PUMPNQ.1 " MAIN STANDBY > & 2 3 aiZc ¥ ap= v - g Ya 5 o . =0 w Az 3 g §lza g 3z e (fe TO SAMPLE LINE (Fe) T & ~ a = 2 FUEL CARRIER TANK NC. 4 TANK NO.5 869 P213 N P21 ‘.-% A »e i Ao DUMPUNE*——J(— CARRIER FILL LINE B R B58 857 P B74 BLEED LINE T, »le O FLOW IN gpm (F) rFueL ok RELIEF VALVE 70 REFLECTOR COOLANT —==F FREEZE VALVE rgvm o e !m i ’ PRESSURE IN psig : 5 | FROM b SUPPLY TRAILER ] HELIUM () PUSH-BUTTON CONTROL B72 87 { .1, TEMPERATURE IN °F @ WATER 873 RESERVE MANIFOLD AN B PRESSURE IN INCHES OF WATER, GAGE (©) cooLant FLOW CONDITIONS ARE BASED ON THE ASSUMPTION . @ VENT SYSTEM THAT THE COOLANT HAS THE FOLLOWING PROPERTIES: 12 He BOTTLES © FLOW N ctm £ =210 Ib/#? “ - O - REMOTE MANUAL VALVE > VALVE NORMALLY OPEN Zp_ ?OT:;?afplfaFoPERAnNG CONDITIONS = Q. U (g»—@ REMOTE MANUAL ANGLE VALVE ) —»4— VALVE NORMALLY CLOSED APPROXIMATE VOLUME OF MAIN SYSTEM: - ¥ MANUAL VALVE INTERNAL 4,3 £1° —»<— THROTTLING VALVE INITIAL~EXTERNAL 3.5 ft° ¥+ SOLENOID VALVE ENRICHING FLUID _ £.7 1> MAX. —Jd— CHECK VALVE TOTAL 6.5 13 MAX. Fig. B.1. ARE Flow Di . ig ow Diagram 109 N T e T T T Ty T e T YRR = IR YR % T T T T T T T T I YO T T I . ey T R T T T TR T T T e - T T Y ey T T PR R I T T o T e O T T AT s B i ""!‘W"’: ™~ s i T YT T ; et Ak Y T e, R i ik bk, i ey ook bl u,q L B e i ° "' o ¢. Miscellaneous Fuel System Volume v (f+3) Maximum capacity ffi"'afid flush tanks 14.5 Bottom (waste) in fill and flush tanks 0.3 Dump line to fill and flush tank Ne. 2 0.8 Dump line to fill and flush tank No. 1 Farpag - 075 By-pass leg - 0.4 Heat exchanger leg from pump to fill line to Tank No. 2 1.6 Reactor leg from fill line to pump (minimum operating level) 3.0 Pump capacity, minimum to maximum level ’ 1.7 Carrier available 15.5 Concentrate available 1.3 d. Sodium System Volume (f3) Inside pressure shell 10 Qutside pressure shell ~10 Total 20 MATERIALS Amounts of Critical Materials BeO blocks (assuming p = 2.75 g/cm?) 5490 ib BeO slabs (assuming p = 2.75 a/cm3) 48 Ib Amount of uranium requested 253 b of U235 Uranium enrichment 93.4% Y233 Uranium in core : 301040 Ib Uranium inventory in experiment 126 to 177 b 111 i i E‘fi“i‘ 2. Composition of Reactor Constituents a. Fluoride Fuel Mixture Fuel? Fuel Carrier W Fuel Concentrate NaF mole 53.09 50 66.7 N wt % . 9:@ 34 20.1 21.3 ; ZrF4 mole % 40.73 50 0 ~ wt % 62.12 79.9 0 ) UF 4 mole % 6.18 0 33.3 wt % 17.54 0 78.7 fmpurities (ppm) Ni <56 25¢ 474 Fe 56 35¢ 844 Cr 4455 10¢ ~20% “Fuel composition for high-power operation. bEinal analysis before high-power operation. “Average of all 14 batches of carrier. 4E stimate based on a few preliminary analyses. b. Inconel? . Amount i Amount i Amount : Constituent (wt %) Constituent (wt %) Constituent (wi %) Ni 78.5 Cob 0.2 Zrb 0.1 . Cr 14.0 Alb 0.2 C 0.08 Fe 6.5 Ti® 0.2 Mo Trace Mn 0.25 Ta® 0.5 Ag, B, Ba Trace Si 0.25 Wwe 0.5 Be, Ca, Cd Trace Cu 0.2 Zn® 0.2 Y, Sn, Mg Trace “B. B. Betty and W. A. Mudge, Mechanical Engineering, February 1945, bAccuracy, $£100%. Y-12 Isotope Analysis Methods L aboratory (spectrographic analysis). Y-12 Area Report Y- F20-14, ¢, Beryllium Oxide* Impurities Amount Impurities Amount Impurities Amount P {(ppm) P (ppm) P (ppm) Si 1050 Ca 780 Na 330 Al 213 Fe 114 Mg 50 Pb 45 Zn <30 K <25 Ni <20 Cr 10 Li <5 s Mn <5 B 2.8 Ag <1 Co <1 *W. K. Ergen, Activation of Impurities in BeO, Y-12 Area Report Y-F20-14 (Moy 1, 1951). 12 TR rwmTEney g e TE oo cper eer e W T B Y T e T e e { - ..xj..n.\' Gk, L Sl b i, T T e e e i i ,Jfl_ L ek h R R L S i, | L e i e L B et e G oo G Wik, o e, L G e BB ia L e © e et 5 o o Data fro‘m ANP Physical Properties Group. bPreIiminary values for the liquid 600 te 800°C. “Data from The Properties of Beryllium Oxide, BMIT-18 (Dec. 15, 1949). dPorosify of BeQ, from ANP Ceramics Group, 23% at p=2.27, 0% at p = 2.83, ®Data from Metals Handbook, 1948 Ed., The American Society for Metals. I Data from Liquid Metals Handbook, NAVEXOS P-733 (June 1952). 8B. 0. Newman, Physical Properties of Heat Transfer Fluids, GI-401 (Nov. 10, 1947). d. Helium Oxygen contamination <10 ppm e. Sodium Oxygen contamination <0.025 wt % 3. Physical Properties of Reactor Materials Melting Thermal Vi i Heat Densi Point Conductivity |sco§|ty Capacity e/n5|1:;y (°C) (Btu/hr-f1.F) (cp (Btu/1b-°F) (9/cm”) Fuel Carrier:* NaF-ZrF, (NaZrF) 510 2.5 +5% 8.0 at 600°C 0.30% p =3.79 - 0.00093T 50-50 mole % 5.3 at 700°C 600 < T < 800°C 3.7 at 800°C Fuel Concentrate:“ NaF-UF , (Na,UF ) 635 0.5 (estimated) 10.25 at 700°C 0.21% p =5.598 - 0.00119T 66.7-33.3 mole % 7.0 ot 800°C 600 < T < 800°C 5.1 at 900°C Fuel:4 NaF-ZrF ,-UF, 530 1.3 (estimated) 8.5 at 600°C 0.24% p = 3.98 - 0.00093T 53.09-40.73-6.18 5.7 at 700°C 600 < T < 800°C mole % 4.2 at 800°C BeQ° 2570 16.7 at 1500°F 0.46 at 1100°F from 2.27¢ to 2.83 19.1 at 1300°F 0.48 ot 1300°F at 20°C 22.5 ot 1100°F 0.50 at 1500°F Inconel® 1395 8.7 from 20 to 0.101 8.51 at 20°C 100°C 77 < T <212°F 10.8 at 400°C 13.1 at 800°C Sodium/ 98 43.8 ot 300°C 0,38 at 250°C 0.30 0.85 at 400°C 38.6 at 500°C 0.27 at 400°C 0.82 at 500°C 0.18 at 700°C 0.78 at 700°C Heliumé <-272.2 0.100 at 200°F 0.0267 at 200°C 1.248 1.79 x 1074 at 0°C 0.119 ot 400°F 0.0323 at 400°C 0 < T <300°C 1.30 x 107 at 100°C 0.136 at 600°F 0.0382 at 600°C 1.03 x 1074 at 200°C Insulation Superex (Si02) 0.18 0.38 Sponge felt (MgSiOs) 0.13 0.48 113 T REACTOR PHYSICS - 1. General Neutron energy Thermal fissions, % Neutron flux, n/em?.sec Gamma flux, y/cm?.sec 2. Power Maximum power, Mw Design power, Mw Power density (core av at max power) w/cm® Maximum specific power (av kw/kg of U233 at 1.5 Mw) . Power ratio (max/av) Axially Radially in core Radially in fuel tube “Maximum power density in fuel, w/cm3 ‘Maximum power density in moderator, w/cm? 3. Neutron Flux in Core (n/cm2Z.sec) Thermal (max) Thermal (av) Fast (max) Fast (av) , Intermediate 4, Leakage Flux (per fission) | Reflector Fast Intermediate Thermal Ends Fast Intermediate Thermal 5. Fuel Enrichment, % U235 Critical mass, Ib of U235 in clean core U235 in core (ive., critical mass plus excess reactivity) Ib U235 in system, Ib U235 consumption at maximum power, g/day 6. Neutron Flux in Reflector Maximum flux at fission chamber holes, nv/w Counting rate of fission chambers, counts/sec.w 114; Thermal and epithermal 60 - . ~1014 ? ~100] s | 10! ? » i: 1.5 (2.5) - 1.0 : 3 (5) 94 (153) 1.5: 1 ; 1.2:1 . 1.7:1 P 110 (180) | 1.3 (2.2) “"r*'-m*"ww'fl P oo MT G S peRc COE TV mE e WM T o g T T 1.5 x 1013 . _ 0.7 x 1013 : 3.5 x 1013 1.5 x 1013 P 2.0 x 1013 0.0011 0.066 0.191 0.013 0.102 0.0002 93.4 (32.75) : 30t040 (36) | b 126 to 177 (138) E 1.5 (2.5) : 1.5 x 106 ’ 2 x 10° R i B !i 7. Reactivity Coefficients , Effect 4 Thermal base i d R Uranium mass (at & = 1) i . 4 , : Sodium density 4 Moderator density ’f Inconel density : Fuel temperature . 3 Reactor temperature vg-! . . Sodium temperature 4 Moderator temperature 4 . i . 1. Material 4 T Barytes concrete block in i the reactor pit only s Poured Portland concrete i Barytes concrete block 2. Composition i 4 Cement ! Concrete 4 Barytes 3 ‘ 4 3. Relaxation Lengths (em) ; , , _. 1-Mev gamma rays i 4 2.5-Mev gamma rays b 7.0-Mev gamma rays 1 Fast neutrons (1 to 5 Mev) 4 j. * i 4 4 5 *‘,’ Symbol Value Ak/k —-0.011 from 68 to 1283°F —0.009 from 1283 to 1672°F Ak/k | | 0.25 (0.236) (AM/M) AkR/E — ~0.05 from 90 to 100% p (Ap/p) Ak/E — 0.5 from 95 to 100% p (Ap/p) Ak/k _ e - =0.17 from 100 to 140% p (Ap/p) (Ak/R)/°F (~9.8 x 1077) (Ak/k)/°F ~5x1073 (-6.1x1077) (AR/R)/°F (-5.88 x 1077) (AR/E)/°F (1.1 % 107%) SHIELDING Thickness (in.) Density (from source outward) (g/cm?) 12 3.3 18 2.3 Barytes Concrete 60 3.3 CaCO, +Si0, + Al 0, Cement and aggregate of gravel Cement and aggregate of chSO4 Portland Cement 4.14 4,62 6.72 6.86 9.46 14.0 8.2 11.0 115 : | q———— S e e T il i R 4, Source Intensities No. per Fission Core gammas Prompt 2.0 Fission-product decay 2.0 Capture gammas 0.94 Gamma flux outside of reactor thermal insulation per watt 2.0-Mev gammas 7.0-Mev gammas Neutron flux outside of reactor thermal insulation per watt Thermal to 250 kev Fast to 250 kev 5. Activity in External Fuel Circuit per Watt Time Qut of 0.52-Mev Gamma Reactor Activity (sec) (photons/cm3.sec) 0 8.1 x 106 5 6.6 x 108 10 5.9 x 10¢ 20 5.1 x 10¢ 30 4,7 x 106 40 4.5 x 108 REACTOR CONTROL 1. Control Elements Source Regulating rod Shim rods Temperature coefficient, (Ak/k)/°F 2. Source Location Type Strength, curies Neutron intensity, n/sec 3. Regulating Rod Location Core axis Diameter, in. : 2 Travel, in. 12 Cooling Helium Total Ak/k, % 0.40 (0.40) Maximum (Ak/k)/sec (slowspeed), % 0.010 (0.011) Speed, in./sec 116 Energy (Mev) 2.5 2,5 7.0 3 x 104 y/cm?.sec 0.8 x 104 y/cm?.sec 3 % 10% n/cm2.sec 1.5 x 104 n/cm2.sec 1-Mev Delayed-Neutron Activity (n/cm3.sec) 8.5 x 103 2.3 x 103 1.4 x 103 0.74 x 108 0.50 x 103 0.35 x 103 ) =d ~=5x 1075 (~6.1x107) Core axis Po-Be 15 (7) 3.5 x 107 (1.6 x 107) 0.3 or 3 (0.32) (fast speed not used) TR P e M o ATy bl o o o W o o ¥ i e 1 " e ™ T oae T em g © e Wt MW‘!W b e - ke Lk i P o Rdh R .G Backlash, in. Servo motor Actuation Regulation There were seven regulating rods made up whi used in the experiment was the one which mos Rod number 1 2 Calculated Ak/k 0.13 0.18 Measured AE/E Weight per inch of rod, ib 0.042 0.06 4. Shim Rods (3) Location Diameter, in. Travel, in. Cooling Material Magnet release time, sec Maximum withdrawal speed, in./sec Total Ak/k per rod, % Maximum (% Ak/k)/sec per rod Motor 5. Nuclear |nstrumentation 0.007 Diehl 1A, 115 v, 60 cycle, 3400 rpm, reversible Manual Temperature error-signal (not used) Flux signal, £ 1 2% N i S1.3°F mean T ch gave a calculated Ak/k from 0.13 to 1.35%. The rod t nearly gave a measured Ak/k of 0.4%. 3 4 5 6 7 0.27 0.39 0.56 0.90 1.35 (~0.25) (0.40) ] 0.091 0.132 0.21 0.32 0.50 One at each 120 deg on 7.5-in.-radius circle 2 36 Helium B C 4 <1072 0.036 (0.046) 5.0 (5.8) 0.005, av over rod (0.0039 at 4-in. insertion) Janette 1.2 amp, 115 v, 60 cycle, 1725 rpm, reversible a. Fission Chambers (2) Function Counting-rate signal Sensitivity 0,14 (counts/sec) per (n/cm?.sec) Range 1017, i.e., 104 in instrument, 107 in position and shielding b. Parallel-Circular-Plate lonization Chambers (3) Function (2 chambers) (2 chambers) Sensitivity Range Safety level (scram signal) Regulating rod temperature servo 50 pa at 1019 n/cm?.sec ~ 103, i.e., from 5 x 1072 t0 1.5 Ng c. Compensated lonization Chambers (2) Function {1 chamber) (1 chamber) Sensitivity Range Micromicro ammeter Regulating rod flux servo 50 pa at 1019 n/cm?.sec ~ 108, i.e., from 1076 to 3 Np 117 e e e p——r T T PR T O PR T T I ' = DTy g —— TR A 6. Scrams and Annunciators E ‘a. Automatic Rod Insertion and Annunciation & ‘ Nucl P Annunciat é p ‘ ) uclear rocess nnunciator L Cause Set Point - Reverse® ; e ~ Scram Seram No. g ; Neutron level 1.2 and 1.5 NFb X 30 . ‘i Period 1 sec X 30 b 3 Period 5 sec X 32 . 4 Reactor exit fuel temperature > 1550°F x 25 , 4 Heat exchanger exit fuel temperature < 1100°F x 25 = 4 & { Fuel flow <10 gpm X 25 A ; : - . i Power Off X 30 ¥ | Seram switch Scram position X 30 i : | : b 4 %Shim rods automatically driven in. , . . PNeutron level set so that the fast scram annunciater, No. 30, would annunciate at 1.2 Ng (normal flux) although | the safety rods would not be dropped until the neutron level reached 1.5 Ng. There was not another neutron level b annunciator at 1.5 Ng. ' & b. Nuclear Annunciators Set Point Annunciator No. Safety circuit Electronic trouble 28 # Count rate meter Off-scale 26 . Servo Off-scale 31 | Rod-cooling helium Off 29 . Neutron level 1.2 Ng 27 : ! ¢ ] i e - E . - ' £ 1 § 118 S sk, o e M 50 amp <20 gpm >160°F <1150°F Maximum pumping level Minimum pumping level >5 psi >50 psi (>41 psi) Maximum pumping level Minimum pumping level >5 psi >50 amps >235°F (> 400°F) >1500°F <1100°F > 1500°F 50 amp <100 gpm >160°F <1100°F Maximum pumping level Minimum pumping level >65 psi >50 amp <100 gpm >160°F 65 psi >60°F <60 psig <4 gpm <2 gpm <4 gpm <2 gpm (>48 psig) (<2 gpm (<2 gpm (<2 gpm (<2 gpm ) ) ) ) Annunciator No. 33 34 35 36 37 38 39 41 45 46 47 48 49 51 52 53 54 65 66 69 70 Annunciator No. 9 10 11 12 13 14 15 17 18 19 20 2] 22 23 24 42 67 68 71 72 “The helium blower was interlocked so that when the fuel temperature decreased below this temperature the biower speed was reduced to zero until the fuel temperature exceeded 1150°F. 119 T gy e vy = g s ARk 4 Sodium system helium blower Pit oxygen Pit humidity Helium supply low Space cooler water temperature Low water flow (any of 5 systems) Rod cooling water temperature Low reservoir level Stack closed Pit activity Monitrons Vent gas monitor Vent header vacuum LLow nitrogen supply Low air supply DC-to-AC motor-generator set AC-to-DC motor-generator set e. Miscellaneous Annunciators Set Point Sodium system blower on before fuel system blower <90% He (Disconnected) > 10% relative humidity <500 psi >160°F > 10% below design > 160°F <16 psi <5 mph wind velocity or high ion chamber reading 0.8 pc/em® >12 me/he (>7.5 mr/br) 0.8 uc/cm® >29 in. Hg vacuum <300 psi <40 psi Off Off SYSTEM OPERATING CONDITIONS3 1. Reactor Annunciator No. W NN WA — 55 HE o e o el T " * T T " e ey k. + & ¥ P . - o YR e # *After 462 hr of circulating time, the last 221 hr were attained after the reactor first bacame critical, and 74 of these after the reactor first operated above 1 Mw. CRPERE S, VR e wemE poocte 3The design values were taken from drawing A-3-0, ‘'Primary Heat Disposal System,’* Flow Sheet, in which the physical properties of the fuel were assumed to be p = 3.27 g/em3, =7 to 13 cp, and cp =0.23 Btu/I1b:°F. The operating values were taken from the 25-hr Xenon run (Exp. H-8). : Fuel inlet temperature, °F 1315 (1209) Fuel outlet temperature, °F 1480 (1522) 4 Mean fuel temperature, °F 1400 (1365) . . Sodium inlet temperature, °F 1105 (122¢6) Sodium outlet temperature, °F 1235 (1335) ._ Fuel flow through reactor (total), gpm 68 (46) Fuel flow rate in fuel tubes (11.3 gpm), fps 4 (3) ; Sodium flow through reactor, gpm 224 (150) E Heat removed from fuel, kw 1270 (1520) k '7 Heat removed from sodium, kw 650 (577) & Fuel dwell time in reactor, sec 8.3 : Fuel cycle time . For 50% of fuel, sec 33.6 (47) £ 4 For rest of fuel, sec 46.4 (47) . i Maximum fuel tube temperature, °F 1493 | E Maximum moderator temperature, °F 1530 E Sodium circulating time, hr 1000 (635) Fuel circulating time,* hr 1000 (462) 1 120 Lol | b ey B . - B BoRg l ed ggh ik 2, Fuel System b el o e Bl M ol s e B e e i R “The water entered each heat exchanger at ambient temperature and was then dumped. 2There were two helium-to-water heat exchangers in parallel; total flow, 130 gpm. a. Fuel Loop F Temperature Pressure Flow (°F) (psig) (gpm) Reactor outlet 1450 (1522) 46 40 Heat exchanger inlet 1450 (1522) 34 20¢ E Heat exchanger outlet 1150 (1209} 12 20 f Pump inlet 1150 (1209) 2 (0.3) 40 (4¢6) Pump outlet 1150 (1209) 54 40 ‘ Reactor inlet 1150 (1209) 50 (39) 40 4 E. %There were two fuel-to-helium heat exchangers in parallel. b, Helium Loop Temperature® Pressure Flow (°F) (in. H,0) (cfm) Fuel-to-helium heat exchanger No. 1 inlet 180 2.2 7,300 Fuel-to-helium heat exchanger No. 1 outlet 620 1.5 12,300 Helium-to-water heat exchanger No. 1 inlet 620 1.5 12,300 Helium-to-water heat exchanger No. 1 outlet 180 1.2 7,300 Fuel-to-helium heat exchanger No. 2 inlet 180 1.1 7,300 Fuel-to-helium heat exchanger No. 2 outlet 620 0.4 12,300 Helium-to-water heat exchanger No. 2 inlet 620 0.4 12,300 Helium-to-water heat exchanger No. 2 outlet 180 0.1 7,300 Blower outlet 180 2.3 7,300 | “The helium passed through two temperature cycles in each loop. ; c. Water Loop _,_ Temperature® Pressure Flow? t:' (°F) (psig) (gpm) ; b Helium-to-water heat exchanger inlet 70 (61) 10 65 (103) é_ Helium-to-water heat exchanger outlet 135 (124) 8 65 1 E ! 3. Sodium System a. Sodium Loop Temperature Pressure Flow (°F) (psig) (gpm) Reactor outlet 1235 (1335) 58 224 (152) 1 Sodium-to-helium heat exchanger inlet 1235 (1335) 52 112 fi Sodium-to-helium heat exchanger outlet 1105 (1226) 51 112 Pump inlet 1105 (1226) 48 (36) 224 Pump outlet 1105 (1226) 65 224 Reactor inlet 1105 (1226) 60 (49) 224 121 G Ly b B b. Helium Loop Temperature (°F) Sodium-to-helium heat exchanger inlet 170 Sodium-to-helium heat exchanger outlet 1020 Helium-to-water heat exchanger inlet 1020 Helium-to-water heat exchanger outlet 170 Blower outlet 170 ¢, Water Loop Temperature (°F) Helium-to-water heat exchanger inlet 70 (s61) Helium-to-water heat exchanger outlet 100 (114) 4, Rod Cooling System a. Helium Loop Temperature (°F) Rod assembly outlet 240 Helium-to-water heat exchanger inlet 240 Helium-to-water heat exchanger outlet 110 Blower 110 Rod assembly inlet 110 4There were three heat exchangers in parallel. Pressure (in. HZO) P ONN A Pressure (psig) 10 8 Pressure (in. H,0) Flow (cfm) 2000 4700 4700 2000 2000 Flow (gpm) 77 (38.3) 77 Flow (cfm) 1270 4234 333 1000% 1000 bCopocity of each of the two parallel blowers; however, the second blower was in standby condition, b. Water Loop Temperature (°F) Water-to-helium heat exchanger inlet 70 (61) Water-to-helium heat exchanger outlet 100 (63) Pressure (psig) 10 8 “The water entered edch heat exchanger at ambient temperature and was then dumped. bTotal flow for three parallel heat exchangers. 5. Water System Equipment Flow per Unit (gpm) Space coolers 7 Reflector coolant system 77 Rod cooling system 6 Fuel coolant system 65 Pump cooling systems 3 Total 122 No. of Units W N 56 154 18 130 12 370 Flow? (gpm) 18 (17.6) 18 Flow (gpm) (56) (77.6) (17.6) (206) (13) (370.2) R T RN B we P T T M E TR T - TSR T R e i ey e o T TR T EM ST A W mmemr - f i memer v ™ B e i b L £ £ E s . g e g s Gk 1. Welding Specifications® Process Base metal Position Filler metal Preparation of base metal Cleaning fluid Clearance in butt joints Clearance in lap joints Welding current Electrode Shielding gas blanket Gas blanket beneath weld Welding passes Welders qualifications MISCELLANEOQUS Inert-gas, shielded-arc, d-c weld Inconel Horizontal rolled, fixed vertical, or horizontal ‘/16 to %. in. Inconel rod (Inco No. 62 satisfactory) Machined, cleaned, and unstressed Trichloroethylene %, to % in. with 100 deg bevel Flush at weld d-c, electrode negative, 38 to 80 amp ]/16 to 3}32 in. tungsten, 90 deg point Argon, 99.8% pure; 35 cth Helium, 99.5% pure; 4 to 12 cfh Ttoh Welded or passed QB No, 1% within 45 days e o e — . b 2For details, see Procedure Specifications, PS-1, ORNL Metallurgy Division, September 1952, bOperation Qualification Test Specifications, QTS-1, ORNL Metallurgy Division, September 1952. 2. Stress Anadlysis a. Moments and Stress in the Pressure Shell Ends Maximum radial stress (per psi pressure) 85 psi Minimum radial stress (per psi pressure) ~80 psi Maximum radial moment (per psi pressure) 80 in.-Ib Minimum radial moment (per psi pressure) ~55 in.-ib Maximum tangential stress (per psi pressure) 70 psi Minimum tangential stress (per psi pressure) - 18 psi Maximum tangential moment {per psi pressure) 180 in.-1b Minimum tangential moment (per psi pressure) ~10 in.-lb b, Stress in the Pressure Shell Vessel Maximum circumferential stress (per psi pressure) 14 psi Minimum circumferential stress (per psi pressure) ~23 psi Maximum longitudinal stress (per psi pressure) 13 psi Minimum longitudinal stress (per psi pressure) ~75 psi T T T T I R g S T TR RN RE T T TR T e m—" R e R e i, e o, 123 v c. Summary of Pipe Stresses® F . . ' Maximum Cold Maximum Hot 5 g i Line Pipe Size Prestress Prestress Temperature Preslsure b No. (in.) (Ib) (Ib) (°F) (psig) i ‘ Fuel Piping : z 111 2 3,850 2,170 1500 39 T 112 1, 11,230 6,450 1500 - 34 : 113 1%, 24,300 13,900 1400 ;34 : 114 1 15,400 8,850 1500 34 . 115 ] 38,000 19,600 1375 n ; 116 1%, 28,200 14,500 1325 60 - 117¢ 2 11 ‘ 118 2 28,000 12,800 1500 i ! 119 A 22,000 11,300 1325 60 : | 120 2 15,350 7,900 1325 60 - 3 Sodium Piping 1 : ? 303 2), 5,260 3,690 1050 58 304 17, and 2 4,280 3,000 1050 58 : 305 1%, and 2 4,300 3,020 1050 58 : 306 1%,2,3 13,250 9,310 1050 51 ) 307 2 4,170 2,935 1050 Es 308 1%,2,3 13,250 9,310 1050 ] 309 2 2,210 1,550 1050 65 310 2}, 4,430 3,110 1050 65 ) 313 2 28,600 16,400 1325 51 2All piping ASI Schedule 40, bNof all temperatures and pressures given were reactor design point values. This line consisted primarily of pipe connections and was not stress cnalyzed, 3. Fluoride Pretreatment Ti T Treatment Purpose (LT)G em(;:)e(;;\fure \ Hydrofluorination Remove water 2 RT to 700 Hydrogenation Reduce oxides and sulfates 2 700 to 800 Hydrofluorination Fluorinate oxides and sulfates 4 800 Hydrogenation Remove HF, NiF2 and FeF, 24 to 30 ad | diw 124 a b, kS b Wl sl ok kg e PRSP o GRERER. s " Appendix C CONTROL SYSTEM DESCRIPTION AND OPERATION' F. P. Green, Instrumentation and Controls Division CONTROL SYSTEM DESIGN The control system designed for the ARE had to be able to maintain the reactor at any given power and, when necessary, had to alsc be able to over- come quickly the excess reactivity provided for handling fuel depletion, fission-product poisons, and power level increases. The motor speeds and gear ratios were set so that safety-rod with- drawal could not add Ak/k at a rate greater than 0.015%/sec. The rods were designed to contain 15% Ak/k, and the control system was designed to limit the fast rate of regulating rod withdrawal to be used for high-power operation to an equivalent Ak/k of 0.1%/sec, with a maximum Ak/k of 0.40%, which is one doliar, for steady-state fuel circu- fation. For the critical experiment and low-power operation, the regulating rod withdrawal rate was limited to an equivalent Ak/k of 0.01%/sec. A fundamental distinction was made between insertion and withdrawal of the shim rods. Either the operator or an automatic signal could insert any number of rods at any time, whereas withdrawal was always subject to both operator and interlock permission. Withdrawal was limited, moreover, to what was needed or safe under given circumstances, Quantiiative, rather than qualitative, indication of control parameters was used where possible. Where feasible, only one procedure of manual con- trol was mechanically permitted to minimize time- consuming arbitrary decisions on the operator’s part. The control philosophy of paramount importance in the general operating dynamics was the use of only absorber rods to control the nuclear process and the use of only helium coolant to control the power generation. Most of the control and safety features of the ARE were similar to those of other reactors. Con- siderable care was exercised in the design of instrument components, control rods, and control circuits to make them as nearly fail-safe as was practical. An instrumentation block diagram of the reactor control system is shown in Fig. C.1, Yhis appendix was originally issued as ORNL CF-53- 5-238, ARE Control System Design Criteria, F. P. Green (May 18, 1953), and was revised for this report on March 7, 1955, in which the reactor is shown to be subject to effects of the control rods and to certain auxiliary facilities. The reactor, in turn, affected certain instruments which produced information that was transmitted to the operator and to the control sys- tem. The operator and the control system also received information from indicators of rod position and motion. By means of the operator’s actions and instrument signals, the control system trans- mitted appropriate signals to the motors and civtches. The actuators iocated over the reactor pit in the actuator housing, in turn, affected the reactor by corresponding rod motions which closed the control loop. In addition to the above indi- cations that were at the operator’s disposal on the console, there were many annunciators physi- cally located along the top of the vertical board in the control room. The group of eight annunci- ators labeled ""Nuclear Instruments’’ were provided to indicate that conditions were improper for raising the operating power level or that improper operating procedures had caused safe limits to be exceeded. INSTRUMENTS DESCRIPTION The reactor instruments discussed in the fol- lowing have been arbitrarily limited to those di- rectly concerned with the measurement of neutron level and with fluid temperatures and flows associ- ated with the reactor. The parallel-circular-plate and compensated ion chambers, the fission cham- bers, and BF, counter were of ORNL design, and, except for the latter two, were identical to those used in the MTR and LITR.? Since the fission chambers were located in a high-tempera- ture region within the reactor reflector, a special chamber had to be developed that could be helium cooled. The compensated ion chambers and the log N and period meters had a useful range of approximately 10%, and thus they indicated fluxes of 10'% n/em?.sec (maximum) and 10% n/cm?.sec (minimum}. The BF; counter had a maximum count- ing rate of about 10° n/sec and the fission cham- bers and count rate meters were capable of covering 25, H. Haenaver, E. R. Mann, and J. J. Stone, An Off- On Servo for the ARE, ORNL CF-52-11-228 (Nov. 25, 1952). 125 b k. *W' T RN R Yoy R e R T a range of 10'! if they were withdrawn from the reflector as the flux increased. Since the count rate meter has a long integrating time and the resultant time delays would make it an unsatis- factory instrument for automatic control, it was used as an adjunct to manual control. | The instantaneous positions of the four control rods were reported to the operator by selsyns. In addition, the positions of the rods with respect to certain fixed mechanical limits were detected by means of lever-operated microswitches. The limit- switch signals tied in with the control system and with signal lights on the control console. The upper limit switches on the three shim rods oper- ated when the rods were withdrawn to about 36 in. from the fully inserted position. Their exact location was adjusted to provide optimum sensi- tivity of shim rod action. The lower limit switches served to cut the rod drive circuits when the magnet heads reached the magnet keepers. The seat switches operated indicating lights on the console, which served only to indicate the proximity of the pneumatic shock absorber piston to the lower spring shock absorber. The lights told the operator the immediate effectiveness of a ““scrammed’’ or dropped rod and, hence, that the rod had not jammed on its fall into the reactor core. ' The regulating rod was equipped with two travel limit switches whose actions tied in intimately with the manual and automatic regimes of the ~ servo-control system, which is described in o following section. Thermocouples were located at many points on the reactor and the heat exchangers to monitor the fuel temperatures, since temperature extremes or low flow rates could have resulted in serious damage to the reactor. The temperatures were recorded, and electrical contacts in the recorders interlocked with the scram circuit to provide shut- down if the reactor inlet temperature dropped below 1100°F or the outlet temperature exceeded 1550°F. The operator was also warned by an annunciator alarm of low helium pressure in the rod-coocling system because loss of cooling would have in- creased the danger of a rod jamming due to me- chanical deformation. CONTROLS DESCRIPTION Motion of the shim rods or the regulating rod was obtained from two energy sources, gravitational and electromechanical. Rod motion by safety 126 action was obtained by gravitational force upon release of the shim rods from their holders as a result of de-energizing the electromagnetic clutches. The three shim rods, each with its magnet, were subject to a number of different possible sources of safety signal. The necessary interconnecting circuitry was centered about an ‘‘auction’’ or "“sigma’’ bus. Instrument signals were fed to this bus by sigma amplifiers. The sigma bus may be said to go along with the ‘*highest bidder’’ among the grid potentials of the sigma amplifiers. This important auction effect allowed all rods to be dropped by any one safety signal. [f all three amplifiers and the three magnets had been identi- cal in adjustment and operation, all three rods would have dropped simultaneously as the sigma bus potential rose or fell past some critical value. Usually, only one of the rods dropped initially, and the others were released by interlocked relay action. Two kinds of scrams were possible: ‘‘fast’ scrams caused by amplifier action and ‘‘slow’’ scrams caused by interruption of magnet power as the result of the action of relays. In a fast scram the shim rods were dropped when the reactor level reached or exceeded a specified flux level, as determined by the safety chambers. The signal from the safety chambers was amplified by the sigma amplifiers and subsequently caused the magnet current to become low enough to drop the rods. The slow scram was the actual interruption of power to the magnet amplifiers by relay action. This relay action could be caused by temperatures in excess of safe limits or stoppage of fuel flow. The shim rod drives were 3-wire, 115-v, instan- taneously reversible, capacitor-run motors rated at 1800 rpm. The shim-rod head, or actuator as- sembly, was a worm gear driven through a gear reducer at slightly less than 2?’/4 in./min. The sequence of shim rod scram, rod pick-up, and with- drawal required a minimum of 25 min. Both individual and group activation of shim rods were selected in preference to group control alone to permit increased (vernier) flexibility in ap- proaching criticality. Group shim rod action would have made difficult the control of flux shading throughout the reactor, particularly on rod insertion, since, upon cutting the motor power, the friction characteristics of the three drives would have caused the rods to coast to different levels. The regulating rod drive consisted of two Diehl Manufacturing Company, 200-w, instantly reversible, f B & t e B R S R ot g e s e e PRI ey T E R TR e T TREE T g W M O RPREIESTHE R RGO CWET M sreer g "’""W."?"fi?"'?"” . YW m;, rfl’- T e TR T ey & S | . sk R | e Afla-fifim‘h&“ i e i i, ik i ik s sk i e e, ek . pcliihiiid i i s el o i i i s . AL il s e T el it i, AN UG i o . o R A, ORNL-LR—DWG 6447 BF 5 COUNTER [THERMOCOUPLEl ITHERMOCOUPLEI PARALLEL CIRCULAR PARALLEL CIRCULAR PARALLEL CIRCULAR COMPENSATED | FisSION __ | sission COMPENSATED SLOW MONITRON MONITRON MONITRON MONITRON FOR START UP PLATE CHAMBER PLATE CHAMBER PLATE CHAMBER CHAMBER : CHAMBER r CHAMBER CHAMBER SCRAMS CHAMBER CHAMBER CHAMBER CHAMBER ONLY 1 DRIVE DRIVE RECORDER REGCORDER MECHANISM MECHANISM | SAFETY SAFETY JUNCTION Ath —in JUNCTION MF;EN“:'?;:N MT)EI\IN:I'I(');SN RE M;’TEN M‘:)EN“;’?;(SN PREAMPLIFIER PREAMPLIFIER BOX PREAMPLIFIER PREAMPLIFIER BOX MONITR A-fA JUNCTION L A PREAMPLIFIER BOX , HIGH YOLTAGE HIGH VOLTAGE | SUPPLY SUPPLY INLET OUTLET LOG N A / HIGH VOLTAGE TEMPERATURE TEMPERATURE TEMPERATURE AMPLIFIER SUPPLY DIFFERENTIAL SERI0D Loa N | | RECORDER RECORDER A1 LINEAR A-1 LINEAR RECORDER AMPLIFIER AMPLIFIER A~1 LINEAR SERVO FLux TROUBLE AMPLIFIER FROM MICROMICROAMMETER l BATTERY SERVO MONITOR MICROMICROAMMETER TEMPERATURE SIGMA SIGMA PERIOD DEMAND CHANNEL SERVO AMPLIFIER AMPLIFIER AMPLIFIER SIGNALS LER LEVEL EvEL TROUBLE LOG LOG ANNUNGIATOR LEV CIRCUITS COUNT RATE COUNT RATE RECORDER RECORDER RECORDER ;EL:\,)‘(O Py METER METER TACHOMETER SIGNALS AMPLIFIER RECORDER RECORDER /SELECTOR PLUGS | ON SERVO AMPLIFIER ¥ p TROUBLE c 2 MONITOR FLUX SERVO \l \I I [ scaLer | [scaLer | SIGNALS SIGMA BUS ]ANNUNC]ATOR | T0 SERVO SLOW SCRAM BUS CHANNEL SERVO MAGNET MAGNET MAGNET AMPLIFIER AMPLIFIER AMPLIFIER AMPLIFIER } | i | : | i 1 1 1 | 1 DRIVE [ DRIVE ; DRIVE I DRIVE DRIVE MECHANISM : MECHANISM i |MECHANISM : MECHANISM MECHANISM T | i i | i | | I | I 1 | t | I | TO | I ( i = MICROMICROAMMETER TACHOMETER [——A | | i | CHANNEL I | l | | [ i | | I ( ! I ! REGULATING SHIM ROD SHIM ROD SHIM ROD SOURCE ROD T TR T I T T e T - o Fig. C.1. Control System Bleck Diagram. T TR YT w7 Ty N T S T T T T ATy I 127 w - ST . ™ T e T T bl RO b ddE.flnog ke bl s two-phase servomotors that operated from reversing contactors which received their actuation from a mixer-pilot relay located in the serve amplifier. Only one of the motors operated at a time, de- pending on whether the temperature or flux servo was coupled into the control loop. For the flux servo operation, an additional speed reducer was used so that the speed of the rod was one tenth that available for temperature servo operation. Initiation of the flux servo ‘'auto’ regime was contingent on reactor power great enough to give a micromicroammeter reading of more than 20 and a servo-amplifier error signal small enough so that no control rod correction was called for. With the servo on auto the regulating rod upper or lower limit switches actuated an annunciator. Although not used in the experiment, initiation of the tem- perature servo auto regime was contingent on reactor power greater than a minimum controllable power (>2% N _) and a servo-amplifier error signal small enough that the servo would not call for rod motion. The rod-cooling system consisted of a closed loop for circulating. helium through the annuli around the shim rods, the regulating rod, and the fission chambers and then through three parallel helium-to-water heat exchangers. The helium was circulated by two 15-hp a-c motor-driven positive- displacement blowers. The motors were controlled from the reactor operating console. The fuel and reflector coolant temperatures were also controlled from the conscle. The helium coolant removed heat from fuel and sodium system heat exchangers and passed it to an open-cycle helium-to-water heat exchanger, from which the water was dumped. The helium in the fuel system was circulated by an electric-motor-driven fan that was controlled from the console. The helium in the sodium system was circulated by two hydraulic- motor-driven blowers for which only the on and off controls were located on the console. In order that the reactor operation would be as free as possible from the effects of transient disturbances or outages on the purchased-power lines, a separate electrical power supply was available for the control system. The supply consisted of a 250-v, 225-amp, storage-battery bank of 2-hr capacity. The bank was charged during normal power source operation by a 125-kw motor- generator set. This emergency d-¢ supply fed emergency lights and a 25-kw motor-generator set to supply instrumentation power (120/208 v, single phase alternating current) during purchased-power - outages. An auto-transfer switch provided an automatic switchover transient of several cycles upon loss of the purchased-power source. The auto-transfer switch automatically returned the system to normal power 5 min after resumption of purchased-power supply. The system could be returned to normal power manually at any time after resumption of the purchased-power supply. CONSOLE AND CONTROL BOARD DESCRIPTION The control console was made up of two large panels, one to the right and one to the left of the operator’s chair, and eight subpanels directly in front of the operator, Fig. C.2. Over the top of the console, the operator also had a full view of the vertical board of instruments which indicated and recorded nuclear parameters and pertinent process temperatures. The rotary General Electric Company type SB-1 switches used were of the center-idle, two- and three-position variety. They were wired so that clockwise rotation produced an ‘‘increase’’ in the controlled parameter; such operation has been described as potentially dangerous. However, the scram switch, an exception, produced a scram in either direction of rotation. The top row of the left-hand panel of the console contained the source-drive control (Fig. C.2). In the second row the first two switches were the controls for the two-speed rod-cooling helium blowers, and the third switch was the on-off con- trol for the electric motors which were the prime movers for the hydraulic-motor-driven blowers. In the bottom row were the group control for the three shim rods and the annunciator acknowledge and reset pushbuttons. The center eight subpanels were identical in size and shape and contained the six selsyns which indicated the rod and fission chamber positions and the two servo control assemblies. The temper- ature calibrated potentiometer located near the panel center was used to set the fuel mean temper- ature for automatic, servo-controlled operation. The subpanel on the extreme left contained the flux vernier (not shown) used to set flux demand into the flux servo system for controlled operation. The dial was calibrated to correspond to a setting of from 20 to 100 on the micromicroammeter Brown recorder. Associated with all controls were the necessary limits-of-travel indicating lights, along with the on-off and speed-range indicators. The right-hand panel contained the switches most 129 TR Y closely associated with the fuel system and the energy removal controls. In the upper row were the switch for the fuel-loop helium-blower prime mover, which was a 50-hp electric motor, the scram switch, and the by-pass switch (not shown on Fig. C.2). The by-pass switch was provided to permit the fuel-loop prime mover to be run, even * though the reactor power was low, to remove the afterheat that was expected to be present after a shutdown. The bottom row on the right-hand panel contained the fuel-system-helium blower speed control, the selector switch for barrier-door control, and the barrier-door drive control. The four panels in the center of the vertical board contained the 12 recorders which pertained to the reactor. They were, left to right, top row: reactor inlet temperature, reactor AT, reactor out- let temperature, reacfor mean femperafljre; center row: count rate No. 1, pile period, safety level No. 1, and micromicroammeter. In the bottom row there were: count rate No. 2, log N, safety level No. 2, and control rod position. The range of flux through which the reactor passed from the insertion of the source to full- power operation was so great, more than 10]3, that several classes of instruments were used, The normal flux (Np) was arbitrarily desig- nated as 1, ond then the first, and lowest, range encountered was the source range, from somewhat less than 10='3 to 10~'); the second was the counter range, from 10! to 10~%; the third was the period range, from 10~% to 3.3 times 1073, and’ the last was the power range, from 3.3 x 10=3 to 1. At greater than 10~3, in the period range, the servo system could be put into auto operation if desired. CONTROL OPERATIONS In the operation of the ARE, operator initiative was overridden by two categories of rod-inserting action: scram, the dropping of safety rods; and reverse, the simultaneous continuous insertion of all three shim rods, which was operative only during startup. The fast scram was in a class by itself, being the ultimate safety protection of the reactor, and could never be vetoed by the operator. The various occasions for automatic rod insertion and annunciation for nuclear trouble are listed in Table C.1. There were five interlocks between nuclear re- actor control, fuel and moderator coolant pumping, TABLE C.1. CAUSES OF AUTOMATIC ROD INSERTION AND ANNUNCIATION Fast Scram Slow Scram Reverse Annunciation Neutron level (1.5 NF) Neufr_on level (1.2 NF) leseoc period 5esec period Fuel temperature (>1550°F) Loss of fuel flow (power >10 kw) Loss of control bus voltage Loss of purchased rpower Manual screm Reac‘to'rrpower (AT > 400°F) Fuel temperature (<1100°F) Fuel helium blowers on without reflector helium blowers on Servo c;ffjrahge | | Rod coolant helium off Count rate meter off scale Safety circuit trouble 130 TR T i G i R s e iy G g e e L s ) | o s o vl s et ik it et n k. - . i o s b il i i ie st i . - i gl b i . PHOTO {5056 Fig. C.2. Control Console and Vertical Board. 131 - e r — - o " ” " a Y ™ ~ - T i T » o) TOPT— e e = e - perom: o r— T T ol s - AT s v 2 o " K T Cic - TR TTRTTYY ¢ T i i T - = . o i i ) 1 | - - . ) 1 i | | - " ‘ ‘ o 1 } | | e T = Ty TT—Y T T ' Ly I T T L i Bia oo GO L e e and helium-cooling operation. (1) When operating at power (above 10 kw), loss of fuel flow would have initiated a slow scram. (2) Loss of fuel flow would have, likewise, removed a permissive on operation of the fuel-loop helium-blower prime mover. (3) Excessive deviation of fuel temperature from set point would have produced a scram for either too high or too low a temperature. Low fuel temperature would also cause the helium pilot motor to run the magnetic clutch control for the helium blowers to the zero position and thereby stop the blowers and thus stop fuel cooling. (4) Moderator coolant flow was interlocked with the moderator coolant helium blowers to prevent helium circulation when the sodium flow was low or stopped. A slug of cold sodium would have entered the reactor when circulation was restarted if this precaution had not been taken. (5) Permissives on increasing helium flow were fuel temperature in 1. Two fission chamber drives a. Manually actuated. range, prime movers operating, reactor in power range, and power greater than 10 kw. The operator was notified by the ‘‘permit’’ light when these conditions were satisfied. A pilot decrease was not interruptible by the operator until the cause of the automatic action was alleviated. The elementary control diagram is shown in Fig. C.3. The vdrious relay and limit switches referred to in Fig. C.3 are described in Tables C.2 and C.3, respectively. These tables and Fig. C.3 show the procedural features of the control system as it applied to operator initiative. This system was divided into channels, each consisting of one or more relays actuated by the operator subsequent to properly setting up interlock permissives. The interlocks often depended upon the aspects of relays in other channels. Primarily, the channels corresponded to a mechanical unit involving a rod control, a process loop control, or an instrument system control. The channels were: b. Travel limit switches, panel-light indicated. 2. Servo control system Auto initiated manually by permission of power greater than 2% N, for temperature servo, a and power great enough to give a reading of more than 20 on the micromicrosmmeter recorder for any scale range, for flux servo, and a démand error close to zero (neither light burning) for either servo system. b. Auto regime sealed-in light indicated if initiated with proper conditions and power remained greater than that specified in 2a above. ¢. Manual regime regained by manually dropping out seal or by manipulation of control rod switch. d. Reverse interrupted rod withdrawal by any means and caused rod insertion. e. Manual contro!l rod overrode automatic regime and dropped out seal. /. Permits on manual rod control required that either the reactor to be at a power level greater than ]0"5 NF or one count rate meter be ‘*on scale.”’ 3. Slow scram a. b. A series-parallel relay system was used to initiate a seram. This system was used because a large number of series contacts would have been inducive to false drop-outs and hence false scrams. The three most urgent criteria calling for scram had to be fail-safe and were therefore placed in the series-connected section. R-16 was inactive in the operating regime; R-17 and R-18 were normally actuated and were responsive to manual scram, scram reset, fuel temperature extremes, and R-16. 4. Reverse, R-23 and R-24, plus an indicator light d. Operated ail three rods if the by-pass switch was in ‘*normal’’ and (1) the group insert switch was closed, or (2) a slow scram occurred (equivalent to a scram follower), or (3) the servo system was in the manual regime; the reactor power was less than 10 kw; and a period of less than 5 seconds occurred. On initial startup, No. a(3) initiated the reverse at any power leve! (due to jumper-shorted 133 R * TR T é i : { ] contact) until a negative temperature coefficient of reactivity was proved. : b. The b&-puss switch prevented reverse (but not slow scram) vyhen in ‘*by-pass’® position. 5. Shim rod wifhdrdwal b : d. Permissives were: ‘ ' . : (1)} no reverse in progress, ' b b (2) reading on-scale on at least one count rate meter or neutron level greater than 10-3 N | s (3) no manual insertion in progress, : (4) rod heads not resting on upper travel hmlts * 6. Shlm rod insert 2 i a. Permitted if heads not on lower travel limits, z b. Actuated individually and manually. 1 c. Actuated together by any action operating reverse. } i ] 7. Fuel-system helium-blower prime movers ¢ a. The prime movers could be started by 5-13 if ; : (1) clutch relays were actuated (rods '*hung''), or if by-pass switch $-5 was on by -pass, ; . (2) helium blbwer speed lower limits were closed, and ¥ g ‘ (3) fuel flow was not below 20 gpm. ' ¥ :: b. The prime movers were automatically turned off if a shim rod was dropped or fuel flow } dropped below 20 gpm. ! ’ c. Relay R-38 started the prime movers by activating the starter relay RF-1 located in the : basement, 8. Helium pilot controls (power-loading system) a. Permit indicators on power increase were lit if (1) the reactor level was greater than 10 kw or the by-pass switch was set on by-pass, " (2) the prime movers were running, and (3) the fuel temperature was above 1100°F. b. Power increase from the coolant system could be demanded if (1) the permit lights were on and (2) pilot switch, S-14, was thrown to *‘increase.’’ ¢. If the power decreased and the controls were not already on shut-off limits, the helium circulation would be shut off. (1) manually by using switch $5-14, or (2) automatically by whichever system suffered from low fuel temperature, from prime mover cut-out, or from low power (<10 kw), since by-pass switch contact $5-3 was closed during normal critical and power operation. Low fuel temperature interlocks automati- cally decreased the corresponding helium pilot controls if the fuel dropped below the critical temperature at either of the two heat exchangers. 9. The by-pass switch function was somewhat obscure since it occurred in the prime movers, pilot controls, and reverse circuits. Its inclusion resulted from the possibility of requiring additional fuel cooling following a shutdown and after an extensive running period at power level. A large gamma heat source, primarily in the moderator (because of its very large heat capacity), calied the a:fl'erliedt, remained in the reactor immediately following power operation. By using both hands, the operator could throw and hold S5 on by-pass while he restarted the prime movers and subsequently increased the helium pilot controls to such a point that the = _ temperature leveled off. S . i 1 The by-pass circuits permitted testing of cooling system control circuits and presetting of :i ' : cooling system control limits while the reactor was shut down and also permitted the vital . subcritical test for fuel temperature coefficient of reacfivity.a ] 3Wm. B. Cottrell and J. H. Buck, ARE Hazards Summary Report, ORNL-1407 (Nov. 20, 1952). 134 S, o iy, akide o The use of the manual reverse cut-out, 55-4, operated by throwing the by-pass switch, was divorced from the above considerations and devoted only to test operation of the shim rod motors at shutdown, at low power, or with the magnet amplifiers turned off. 10. Source drive a. Manually actuated. b. Travel limit switches, panel-light indicated. REACTOR OPERATION The preliminary operating plan for the ARE was described in ORNL-1844.% In the check operations, considerable time was spent in loading the inert fuel carrier and in the subsequent shakedown run. Critical loading, subcritical measurement of fuel temperature coefficient, regulating rod calibration, zero-power operation at 1200 to 1300°F, and power operation are described in Appendix D. In the starting of a freshly loaded reactor the operator is most concerned about a reliable neutron signal. Attention must therefore be given to as- suring that the fission chambers are in their most sensitive positions, and that they are giving a readable signal on the scalers or count rate meters. [t is imperative that the source be inserted for the earliest indication of subecritical multiplication to be seen during the first startup. Use of the source in subsequent experiments becomes less important as the history of the reactor operation builds up. For a normal, intentional shutdown the operator had the choi¢e of a variety of procedures. For every scram, however, the interlocks of the system were such that the helium flow in the fuel-to-helium heat exchangers was cut off by opening the electri- cal circuits of the motors. This action was neces- sary to prevent the fuel from freezing in the heat exchangers. The various slow scrams included the following: (1) maximum reactor outlet temper- ature, (2) minimum reactor inlet temperature, (3) a 1-sec period, (4) N greater than 3.0 Mw, (5) fuel flow below 20 gpm. A 5-sec period inserted the shim rods, which decreased Ak/k at a rate of 0.015%/sgc, and operated only during low-level startup. In addition to the process signals which initiated slow scrams, there were several other potential process malfunctions for which it might have been desirable to scram the reactor. However, the action to be taken in these situations was left up to the decision of the operator. There were two reasons for not putting these process malfunctions ADesign and Installation of the Aircraft Reactor Ex- periment, ORNL-1844 (to be published). on automatic scram. First, sensory signals usually lag the event to such an extent that a fast scram cannot provide better protection than a delayed scram. Second, reaction to the signals requires limited judgment. An annunciator alarm signal was provided to draw the operator’s attention to the possible difficulty for each case that would have required limited judgment. The cases were the following: (1) pronounced changes in differential temperatures between reactor outlet tubes, (2) lowering level in any surge tank, (3) rising level in any surge tank, (4) alarm from a pit radiation monitor or monitron, (5) high oxygen concentration or humidity in the pits. The operation of placing the ‘‘position’’-type regulating rod serve system on automatic control in any part of the power range above 2% N, for temperature servo, and a micromicroammeter re- corder reading of greater than 20 on any scale, for flux servo, was arranged to require that the reactor to be on a stable, infinite period. In the normal course of events, the rod should have been manually set at mid-range. Six d-c voitage signals were fed into the servo amplifier, three from the servo ion chamber, inlet temperature recorder, and outlet temperature recorder, one from the voltage supply adjusted by the ‘‘temperature demand’’ potentiome- ter, and two from the flux demand and micromicro- ammeter recorders. The operator therefore had a unique temperature setting for each level of power operation that produced *‘zero error’’ in the ampli- fier output, and hence no ‘‘error’’-light indications and no demand for rod movement. It was at this setting that the servo ‘‘auto’’ regime could be initiated. A small change in reactor mean temper- ature or flux level could be effected in the auto regime by resetting the temperature demand or flux demand potentiometers, but the power level could not be changed. No clear line could be drawn between normal operation and cperation under difficulty. A few of the situations in which the operator would have expected to feel more than average need for atten- tion to instrument signals and control actions were loss or interruption of power, automatic shutdown 135 e e p———_— e — - i Sl G ) i, i il A ..l ik . ke i gl otk e, e . B0 A | s . . e e i, . = it ‘ o, Sl i i i ki i = e EetE e o il i REMOVE JUMPER BEFORE CHANGING JUMPER IN PLACE FOR STARTUP-REMOVE /TO FAST SERVOMOTOR 20 /WHEN TEMPERATURE COEFFICIENT IS ESTABLISHED S04 ORNL-LR-DWG 6418 l lfls 34 Al I l AUTO MANUAL MAIN FLOW RS-24 r7<100°F ! §5-2 TR42-1 PB. P.B. A pisaca <20 gpm i1 g Rig-1% | XRap-2 XR24-4 | sg-rT| s10-1 | Sti-t: @ < T - o $8-1+ I a n o o RS-36 ~L Rs-17 L m R2a-1 | =R24-2| TR24-3 2 = 4 = 1o = TR38-2 225144 Ri3-2 3% —=83-5 Lt 130 ] 71O <« T S16-1 o=§16-2 = "8 6 TUBES - e & R20-1 = © T<1150°F 110 b — L gany ~ 5-sec ] RS-39 T © 55+ 38-3 . 3 8 3 & 123 3 - R21-1 B TS142 i Q © © i © THCONTACTS| 742 4% eLux $3-12 53-24 RS- o i - @ 3 — ON PLUG ON I AN B i RS-26 1588 € Rava Lrsie s - SERVOAMP RIS-3 = . . RS-18 @ @ Tr. oF ¥ VN TEMP, L 234 LR15-1 RS-41 To=1550°F MAIN FLOW 58-4 = 3 <150 S sig4 —Si9-3 TS19-2 =519-4 < — X R20-4 <20 gom Loio-z| L o g s w daw e —— & = P S N L =2 D 5 SWITCH NO. 1 < = © o . 0 P $5-3 o IN RELAY POWER o T+ RS-30 - | RI9-4% [R2t-43¢ - T S513-2 ° o - o 2 = 3 by ) ) = a>20 T« - FR16- @ - & © g 2 n CABINET I e | R23o2 a2 3 g g @ o g o B o o £ ! % - 3 - " . st s2-2 T,Si'g’% 1 NoTE 127 » Loz wR2r-4 FR29-2 XR50-1 X RA4-1 % RA7-4 XR4G-1 o - . 1 L. £ 4 S 4 @ o 3 2 - RI54-2 X X-R11-4 | RSP-1AND 2 ARE FRIGA-4 RIS WI w jRes-2 Fr30-2 ~ TSe0-1 FS20-2 FS520-3 520~ R12- - RISA- * © R7-4 R§-2 R]9-4 RID-2 CONTACTS IN SERVO 12-4 x RI5A-3 Lraq-a 4 AMPLIFIER 132 | _‘R36-5 i;R43—4 8 © - - un W0 ot o - -— - " o ~ o 2 N o o o » — @ v 1 3 < e o c e - & - @ © g 5 3 & = @ & 32 . 2;2 R R2 R3 R4 fms fmsA 'fms fns 21 R6 fmq RIGA fme RIT Tme 'fm;s R23 fnsa fR34 fRz,s fms fme fns,o fnsa {R54 ffim {RH {(:1 {cz 502 fcz, ic:z INSERT ~ WITHDRAW INSERT WITHDRAW AUTO - AUTO COWN INSERT MANUAL WITHDRAW up 1 2 1 2 3 INCR. DECR. INSERT WITHDRAW DOWN up DOWN uP MANUAL £RROR ERROR 2 RELAY PERMIT 1 FISSION FISSION S~ SERVO SYSTEM SLOW FUEL-SYSTEM WITHORAW INSERT HELIUM-BLOWER SOURCE LOWFUEL FUEL-SYSTEM HEAT EXCHANGER SBARRIERS CHAMBER NO. CHAMBER NO.2 SCRAM HELIUM BLOWER SHIM RODS SHIM RODS SPEED (PILOT} DRIVE TEMPERATURE HELIUM-LOOP NOTE: PRIME MOVER RELAY (FOR50-hp C1-4 ARE 0-C STARTERS ADJUSTABLE SPEED IN BASEMENT PRIME MOVER) 20 ‘ N 93(1) 2301) l l l l l l l J_ SWITCH NO1 IN AMPLIFIER 2 (N 7T (D) LSt TLS3 LSS =LS13 =:LS15 —=LS8 ==Ls10 TFLs2 =LS$19 TLS37 IN AMPLIFIER | POWER CABINET 3=RS13 RSt4 | POWER CABINET T T3 =52 LS4 FLS6 =LS14 =LS7 =LS9 =LS1 XRig-2 XR20-2 XR21-2 RS22- =LS20 —LS38 $6-3 ® 2 N>10 kw 94 . —=s512-2 S12-4 i i oGy [ i ~ 2 w 2 ook S$17-20= 71 S Lo —=s5ia-2 Tsi8-1 oWk M2) oi'k L Rag-2 R45-2 % Ras5-4 Egd Eg= ETod -[uw Lo E28 $12-3 ' g& - e s £30 T zi2 ™ T RI8-3T RI8-4 ofz ©° w o -Z o B 2g= p SEz ] 0y w P X1 o © <« v © ~ o @ of - o " x| n ol mp oy IR w| I sl 8 @ 5 & o > kY & (xi) - © % ® @ @ % © @ & o ® ® o © 2 & .- S @ ™ a0« & b ~ . = > > > > ® @ o «o @ ® @ @ o o o o o e o > z > a > M2 z > 2M2 ¥ = < o < 0 T T o T L 3| \ < - < - < = T - X R40-2 X RA0-4 X R46-4 MXZ) L3) L 3) © n © ¢ 5 " ~ (S 5 ; 4 “ e TO MAGNET M 2M Sl L el e e 2 Sl oL o 2 e 2 [ AenERs g g R7 )R8 JR9 JRIO RN RI12 RI9 JR20 |R21 | R25 R26 JR27 |R28 R29 R3O R45 R46 R40 R41 |R42 |R47 ]R48 |R43 | Raa o 2 oL oL (5) 3 | () () | Ga (8) | G IN AMPLIFIER i3 14 3 21 - S POWER CABINET 76 oL oL 1 1 2 2 1 2 3 1 2 3 1 2 31 2 2 3 3 v L u L U L U u u U L L L CLUTCH FIRST CLUTCH FIRST CLUTCH FIRST A>10kw L U v L | 2 FRONT BACK FISSION CHAMBER REGULATING CLUTCH SHIM ROD LIMITS CLUTCH LIGHTS oo/ SOURCE HELIUM MAGNET AMPLIFIER ROD_COOLANT BLOWERS REFLECTOR COOLANT PRIME MOVERS LIMITS ROD LIMITS DRIVE PILOT 15-hp MOTORS (4 AND 5} 5-hp MOTORS (66 AND 67} LIMITS LIMITS — Ve Omm——tm- —————C) O~ G Ot ——— —° INSERT | NEUTRAL |WITHORAW SCRAM | NORMAL | SCRAM STOP | NORMAL | START DECRE ASE | NEUTRAL | INCREASE DOWN | NEUTRAL up 1 . 1 . 1 . 1 . 1 . 2 . 2 ° . NOT 2 . . 2 . 2 . 5 . 3 . USED 3 . 3 ° . 3 . 4 ° 4 . . a . 4 * FOR SWITCHES $1,52, S3,$9, SI0 AND SH FOR SCRAM SWITCH S6& (SWITCH HAS FOR SWITCH S12- REFLECTOR COOLANT FOR SWITCH S14-HELIUM PILOT FOR SWITCHES S17 AND $18-ROD FOR SWITCH S20-HEAT EXCHANGER RED HANDLE) PRIME MOVERS COOLANT PUMPS BARRIER CONTROL Ot Qe ——— O O BY PASS | NORMAL INSERT | NEUTRAL |WITHDRAW STOP | NORMAL | START INSERT | NEUTRAL 1 BOTH 1 . 1 . . ° 1 ° . 2 . 2 . * . 2 . ° TYPICAL ENGRAVING 3 . 3 . 3 . ® NOTES: ARRANGEMENT 6 . 4 . 4 . * ALL CONTACTS ON $B SWITCHES NUMBERED OF 5B SWITCHES FOR BY PASS SWITCH 55 FOR SWITCH S8 - GROUP SHIM ROD FOR SWITCH 513 -FUEL GOOLANT FOR SWITCH 516 - SOURCE ORIVE FOR SWITGH $19 - HEAT EXCHANGER FROM HANDLE END. - OPERATION PRIME MOVER BARRIER SELECTOR ALL RELAYS SHOWN WITH COILS DE-ENERGIZED. - - Fig. C.3. Elementary Control Diagram. i P T T e T R IRI EpT TY T Tw s e 1 S B e T g T T T T e m,m - T Tttt o vme o e j— T T =T T YT iz v kil R iy e el B Adeea B, i i, i i iy B o Wk, el L ek BN e e e s, il Kb TABLE C.2. LIST OF RELAY SWITCHES Contact Instrument Set Point RS-9 Count rate recorder No, 1 Closed above 1 count/sec RS-10 Count rate recorder No. 1 Opened off scale RS-11 Count rate recorder No. 2 Ciosed above 1 count/sec RS.12 Count rate recorder No, 2 Opened off scale RS.13 Front sodium flow Closed above 60 gpm RS-14 Back sodium flow Closed above 60 gpm RS-15 Log N Closed when N > 2% N RS-16 Fuel heat exchanger No. 2 outlet Closed above 1150°F RS-17 Fuel heat exchanger No. 2 outlet Closed above 1100°F RS-18 Fuel flow Closed above 20 gpm RS-19 Not used RS-20 Log N Closed when N > 10™° N RS-21 Log N Closed when N > 107> N RS-22 Log N Closed when N > 6.67 x 1073 N , RS$-23a Safety level No. 1 Opened above 120 RS-23b6 Safety level No. 2 Opened above 120 RS-24 Fue! flow Opened above 20 gpm RS.25 Not used RS-26 Reactor outlet temperature (mixed manifold) Opened above 1550°F RS.27 Not used RS.28 Not used RS-29 Pile period Closed at less than 5 sec RS<30 Micromicroammeter Closed above 20 ppa RS-31 Count rate recorder No, 1 Closed above 1 count/sec RS-32 Count rate recorder No, 2 Closed above 1 count/sec RS-33 Fuel heat exchanger No. 1 outlet Closed above 1150°F RS-34 Fue! heat exchanger No, 1 outlet Closed above 1100°F RS-35 Pile period Opened at less than 5 sec RS-36 AT across reactor tube No. 1 Closed above 450°F RS-37 AT across reactor tube No. 2 Closed above 450°F RS-38 AT across reactor tube No. 3 Closed above 450°F RS.39 AT across reactor tube No. 4 Closed above 450°F RS-40 AT across reactor tube No. 5 Closed above 450°F RS-41 AT across reactor tube No. 6 Closed above 450°F RS.42 Not used RS-43 Not used 139 —p— T s D TABLE C.2 (continued) i + 1 Contact Instrument Set Point RS-44 Not used RS.45 ACet0.DC M-G set Closed when set was charging batteries RS-46 Not used RS-47 Not used RS-48 Not used RS-49 Not used RS-50 Fuel flow Closed above 20 gpm RS.51 Fuel flow Closed above 20 gpm RS-52 Reactor outlet temperature (mixed manifold) Opened above 1500°F RS-53 Wind velocity recorder Wind above 5 mites/hr or vent gas monitors RS-54 Fue! heat exchanger No. 1 outlet Closed above 1150°F RS-55 Fuel heat exchanger No. 1 outlet Closed above 1100°F RS«56 Fue! heat exchanger No. 2 outlet Closed above 1150°F RS.57 Fuel heat exchanger No. 2 outlet Closed above 1100°F RS-58 Main fue! pump low-water-flow alarm Closed above 4 gpm RS-59 Fuel heat exchanger No. 2 outlet Closed above 1100°F RS-60 Secondary fuel pump lowewater-flow alarm Closed above 4 gpm RS-61 Back sodium heat exchanger outlet Closed above 1100°F RS-62 Front sodium flow Closed above 100 gpm RS-63 Back sodium flow Closed above 100 gpm RS-64 Helium concentration recorder Closed at less than 90% He RS«65 Pit humidity indicator Opened at > 10% relative humidity RS-66 Helium supply header pressure Opened at <500 psig RS~67 Water reservoir level Opened at <16 psig (<18,000 gal) RS«68 Main sodium pump tank level Opened at low level RS«69 Main sodium pump tank level Opened at high level RS-70 Standby sodium pump tank level Opened at low level RS-71 Standby sodium pump tank level Opened at high level RS.72 Main fuel pump tank level Opened at low level RS-73 Main fuel pump tank level Opened at high level RS-74 Standby fuel pump tank level Opened at low level RS.75 Standby fuel pump tank level Opened at high level RS-76 Water controller for front sodium heat exchanger Opened above 160°F RS.77 Water controller for back sodium heat exchanger Opened above 160°F RS.78 Water controller for fuel heat exchanger Opened above 160°F - RS.79 Standby sodium pump mator ammeter Opened above 50 amp 140 rrd rapt i 'r T v em rERemErr e e S mnye®r e w e P Omerremye B e T IR T e g Y T TR S e T o p o ST e ey et e et e e W e By i G i iy, i o i e = b i i i | e o~ o b vl Ak wlk TABLE C.2 (continued) Contract Instrument Set Point RS-80 Water controller for space coolers Opened above 160°F RS-81 Water controller for rod-cooling heat exchanger Opened above 160°F RS5-82 Individual heat exchanger tube temperature Opened above 1500°F RS-83 Individual heat exchanger tube temperature Closed above 1150°F RS-84 Main sodium pump motor ammeter Opened above 50 amp RS-85 AT across reactor tube No, 1 Opened above 400°F RS-86 Main fuel pump motor ammeter Opened above 50 amp RS-87 AT across reactor tube No. 2 Opened above 400°F RS-88 Standby fuel pump motor ammeter Opened above 50 amp RS-89 AT across reactor tube No. 3 Opened above 400°F RS-90 Standby sodium pump lowe-watersflow alarm Closed above 4 gpm RS-91 AT across reactor tube No. 4 Opened above 400°F RS.92 Main sodium pump low-water-flow alarm Closed above 4 gpm R3-93 AT across reactor tube No. 5 Opened above 400°F RS-94 Main fuel pump loweoil-flow alarm Closed above 2 gpm RS-95 AT across reactor tube No, 6 Opened above 400°F RS-96 Standby fuel pump lowsocil-flow alarm Closed above 2 gpm RS-97 Standby sodium pump lowe-oil-flow alarm Closed above 2 gpm RS-98 Main sodium pump loweoil-flow alarm Closed above 2 gpm RS.99 Scdium outlet pressure at reactor Closed above 48 psig RS-100 Fuel inlet pressure at reactor Opened above 48 psig RS-101 Main sodium pump tank pressure Opened above 47 psig RS-102 Standby sodium pump tank pressure Opened above 47 psig R$-103 Standby fuel pump tank pressure Opened above 5 psig RS-104 Main fuel pump tank pressure Opened above 5 psig RS-105 Rod cooling helium pressure Closed above 13 psig RS.106 Fuel loop water flow Closed above 110 gpm RS-107 Water flow to fuel and sodium pumps Closed above 10 gpm RS-108 Rod cooling water flow Closed above 15 gpm RS-109 Front sodium loop water flow Closed above 65 gpm RS-110 Back sodium loop water flow Closed above 65 gpm RS-111 Space cooler water Flow Closed above 45 gpm RS-112 Vent header vacuum Opened below 29 in. Hg RS-113 Reserve nitrogen header supply pressure Opened at <300 psig RS-114 Rod cooling helium Opened at <40 psig RS-115 DCsto-AC M-G set Closed when set was running 141 S R e S e A - T = A pRm—— s TABLE C.2 (continued) Contact Instrument Set Point RS-116 AC-to-DC M:G set Closed when set was running RS-”? Heat exchanger pit radiation monitor Opened at high level RS-118 Heat exchanger pit radiation monitor Opened at high level RS-119 Vent gas monitor Opened at high level RS-120 Vent gas monitor Opened at high leve! RS-121 Sodium AT at reactor Opened at <60°F TABLE C.3. LIST OF LIMIT SWITCHES Contact Device Set Point LS.1 Fission chamber No. 1 lower limit Closed on limit L S.2 " Fission chamber No, 1 upper limit ‘Closed on limit L S-3 Fission chamber No. 2 lower limit Closed on limit LS-4 Fission chamber No. 2 upper limit Closed on limit LS-5 Regulating rod lower limit Closed on limit LS-6 Regulating rod upper limit Closed on limit LS.7 Shim rod No., 1 upper limit Closed on limit LS.8 Shim rod No. 2 upper limit Closed on limit L S-9 Shim rod No. 3 upper limit Closed on limit LS-10 Shim rod No. | lower |imit Closed on limit LS-11 Shim rod No. 2 lower limit Closed on limit LS.12 Shim rod No. 3 lower {imit Closed on limit LS-13 | Shim rod No. 1 clutch Closed when rod was attached LS-14 Shim rod No. 2 clutch Closed when rod was attached LS.15 Shim rod No. 3 clutch Closed when rod was attached LS.16 Shim rod No. 1 seat Closed on seat LS-17 Shim rod No. 2 seat Closed on seat LS-18 Shim rod No, 3 seat Closed on seat action, and buildup of xenon poison after a shutdown, As far as loss of power is concerned, the operator would have been left figuratively, and nearly literally, in the dark were the instrument bus and control bus to go dead simultaneously. In this very unlikely occurrence, all the instrument lights and all relay-indicating lights would have been out, leaving the operator without visual knowledge of whether the scram (automatic on loss of power) 142 was effective. If he had reasonable faith in the law of gravity, he would probably have remained at his post until power was restored. There would have been no normal scram indication if the relay cabinet main fuse were to have opened. However, since the amplifier power cabinet would have operated from emergency power, there was a check on release of the shim rods in the fact that the magnet currents would have been interrupted. Whenever the reactor was subject to automatic * o ek . il L L el w, o L3 e L 5 RE, ok . A.dua B shutdown action, the operator’s concern was di- vided between the causative effects and the chances of returning the reactor to normal oper- ation. The circuits prevented the operator from interfering with shutdown actions as long as the causative conditions remained. Hence the oper- ator's reaction to anything less drastic than a scram was to clear the responsible condition. When the reactor was scrammed, the chances of getting back into normal operation were of course dependent upon the time history of previous oper- ation as well as upon the need for remedying the trouble. However, any length of complete shutdown could have been tolerated after prolonged operation at normal flux, since xenon poison buildup was low. 143 e R s b } i, Bega Appendix D NUCLEAR OPERATING PROCEDURES! The importance and critical nature of the program, as well as the short time scheduled for the operation of the reactor and system, necessitated a tight experimental program and precise operating procedures. The nuclear operating procedures were, of course, prescribed in advance of the experiment, It is of considerable interest to note that the actual experimental program followed the anticipated program very closely tbrougbbut. The only significant deviations were the inclusion of some additional experiments as time permitted. The following appendix is a verbatim copy of the procedures by which the reactor was operated, Some minotr discrepancies in opetating conditions {i.e., 34-gpm fuel flow vs 46.gpm actual flow) and procedures (i.e., use of the original enrichment system, which was replaced) will be noted. It is anticipated that the reactor and its associ- ated circuits will be operated for about 250 hr with fused salts in the system prior to the intro- duction of fuel. All process instrumentation and components (non-nuclear) will be checked out during this period. The mechanical operation of the nuclear equipment will be checked out at temperature. The safety rods will be raised and dropped approximately 50 times during this period. Several tests preliminary to the fuel loading will have been run. The rated flow of the fuel carrier will be 34 gpm at a mean temperature of 1300°F, 1. The fuel loading system will be operated with carrier. The fuel storage tank will be installed and carrier forced into the transfer tank and then into the reactor system. The strain gage weighing devices will be checked. 2. The helium blowers will be operated and the resultant drop in mean temperature observed. The operating crew will practice in handling the helium flow so as to drop the mean temperature at a rate of 10°F/min and at a rate of 25°F/min. Care will be exercised not to drop the temperature of the fuel carrier to below 1150°F as it leaves the heat exchangers. As the latter temperature approaches that value, the helium flow will be reduced and the system brought back to its mean operating temperature of 1300°F. 3. With the helium blowers off and starting with fuel carrier at rated flow of 34 gpm and mean temperature of 1300°F, the flow rate will be decreased in steps to about one-third rated value. All temperatures will be observed to note any spurious changes caused by the decreased flow. ]This appendix was originally issued as ORNL CF-54-7-144, ARE Operating Procedures, Part I, Nuclear Operation, J. L. Meem (July 27, 1954), 144 ADDITION OF FUEL CONCENTRATE? The fuel concentrate will be added in batches to an ‘‘eversafe’’ container which will contain 115 kg of U235, The density of the fuel concen- trate may be represented by the equation: p (g/em3) = 5.51 - 0.0013 T (°C) At a temperature of 1300°F, the density is 4.59 g/cm?, or 287 Ib/ft>3. In each pound of the fuel concentrate there is 0.556 1b of U235, One quart of the concentrate weighing 9.59 Ib and containing 5.33 Ib of U235 will be forced into the transfer tank and weighed. With the safety rods completely withdrawn and the neutron source and fission chambers inserted, the fuel concen- trate will be forced into the fuel circuit, At this time the carrier flow will be 34 gpm and the concentrate and carrier temperatures will be 1300°F. The scram level on the safety chambers will have been set at about 10 kw and the period scram at 1 sec. A BF, counter will have been installed temporarily in place of the neutron ion chamber for the temperature servo control, and the regulating rod will be on slow-speed drive. The system volume up to the minimum operating level in the fuel pump is calculated to be 4.64 f3. Assuming a critical mass of 30 Ib in the 1.3 f13 of reactor core, there will be approximately 124 1b of U235 jn the fuel circuit when the reactor first goes critical with the shim rods complietely withdrawn. Accordingly, it is anticipated that approximately 0.78 13 or 23.4 quarts of fuel concentrate must be added for initial criticality. Twelve quarts will be added in succession with the shim rods com- pletely withdrawn and the fission chambers fully Zps previously noted, the original enrichment system was not employed, although the principles and tech- niques of enrichment outiined here were followed. wht T TEWE TR AT et BEEE T W e R T B T e err o [ g ¥y rowmeT N ey R W MET v AR D TEE ey ey s wrer BT TR o QR DO T e T YT M . NP i B N : 5 [ N i s L el g inserted. After each quart of concentrate is added, counts will be taken on both fission chambers and the BF, counter. The reciprocal counting rate will be plotted vs. the fuel concentration to ob- serve the approach to criticality, After 50% of the fuel concentrate (12 gquarts) has been added to the system, a sample of the mixed fuel and carrier will be withdrawn for chemical analysis. SUBCRITICAL EXPERIMENTS During the addition of the first 12 quarts of fuel concentrate, the safety rods have been completely withdrawn from the reactor. From here on, the shim rods will be inserted approximately 25% while a quart of fuel concentrate is being added. The shim rods will then be withdrawn and a count taken on the fission chambers as before. The shim rods will be inserted and withdrawn in this fashion for each succeeding fuel addition. The reciprocal of the counting rate vs. fuel concentration will still be plotted to indicate the increase in criticality. When approximately 90% of the critical mass has been added, the fuel addition will be stopped and several subcritical experiments performed. For these experiments it is desired that the £ of the reactor be about 0.97 to0 0.98, In Fig. D.1is shown the relationship between Ak/k and AM/M as a function of the critical mass M. Using this curve, an estimate of the point at which fuel addition must be stopped can be made. For the first experiment, the fuel temperature will be decreased at a rate of 10°F /min by starting the helium flow. Assuming a temperature coef- ficient of =5 x 10~ {(Ak/%)/°F, the resultant A%/ should be about 0.25% after 5 min. ¥ AM/M is DWG. 22502 0.3 \\ 0.2 \\ ils \-. o~ d c 04 O 20 22 24 . 26 28 30 32 34 36 38 40 CRITICAL MASS {Ib) Fig. D.1. Reactivity-Mass Ratio as a Function of Critical Mass. 145 TR T T YT approximately 4 Ak/k {Fig. D.1), the count rate should increase by an amount corresponding to the addition of 1.0% more fuel. If the change in count rate is too small to be definite, the experiment will be repeated at a rate of decrease of 25°F /min until a definite increase in count rate is observed, or until the temperature has been decreased 100°F, if, after this decrease in temperature, no increase in count rate is observed, it will be assumed that the temperature coefficient is negligibly small and the experiment will proceed. As has always been 0.008 the philosophy on the ARE, if the temperature coefficient is observed to be positive, the experi- ment will be concluded and the fuel circuit drained. For the second experiment, the reactor tempera- ture will be returned to its original value of 1300°F, and with the reactor containing 90% of the critical mass, the fuel flow rate will be gradu- ally decreased. If the available delayed neutron fraction is 0.47% Ak at full flow and 0.75% Ak when the flow is stopped (Fig. D.2), a Ak of 0.28% should appear. The count rate should in- DWG. 22354 0.007 \ 0.006 S 0.005 \ 0.004 REACTIVITY 0.003 0.002 0.00t 0 10 20 30 40 50 60 70 - 80 FUEL FLOW RATE {(gpm) Fig. D.2. Reactivity from Delayed Neutrons as a Function of Fuel Flow. 146 + BT o rre ot T e e R Y W oy B Y. o et e w merY mm o BOrERT WhgEIrw FTOEwT vt omect 0 % i £ E b e o .-.-’.g{ . dg &.;5 vk b et M, L iy - REL @ crease by an amount equivalent to the AM corre- sponding to additional Ak in the delayed neutrons. A rough experimental check on the calculated value of the delayed neutron fraction will thus be available. During the period of subcritical operation, the count rate will be carefully observed for sudden changes that could be caused by fuel segregation. No such difficulty is anticipated, but if such an effect is observed, the reactor will be shut down until the reason for such behavior is ascertained. INITIAL CRITICALITY Upon completion of the subcritical experiments, the fuel flow and reactor temperature will be returned to the initial conditions of 34 gpm® and 1300°F. Counts will be taken on the fission chambers with the shim rods 25% inserted and completely withdrawn. The shim rods will then be 50% inserted and a quart of fuel concentrate added. The rods will be withdrawn to 25% and a count taken and then completely withdrawn and a count taken. From here on, two curves of re- ciprocal counting rate vs. fuel concentration will be plotted, one at 25% rod insertion and the original curve with the rods completely withdrawn. If the reactor contained 90% of the critical mass during the subcritical experiments, about two l-quart additions of fuel concentrate will bring the reactor critical with the rods completely withdrawn. The last fuel additions will be made very slowly. When criticality has been definitely reached, the reactor will be shut down and a fuel sample taken for chemical analysis. After the sample has been taken, the reactor will again be brought barely critical (a few hundredths of a watt) and held at constant power by watching the count rate meters. The gamma-ray dosage will be measured with a ‘“‘Cutie-pie’’ at all pertinent points throughout the pits and recorded for future reference on radiation damage and shielding. ROD CALIBRATION vs FUEL ADDITION The reactor will be brought critical at a very low power and the shim rods adjusted so that the regulating rod is 5 in. above center. The regu- lating rod is estimated to be worth about 0.04% Ak/in. One quart of fuel is approximately 4% AM or 1% Ak (Fig. D.1). Therefore, ]/25 of a quart should be worth about 1 in, of regulating rod. 3Actuai, 46 gpm. Figures D.3 and D.4 show a calibration of a regulating rod taken on a mockup of the ARE at the Critical Experiment Facility. While it would be fortuitous if the regulating rod in the actual ARE gave the same calibration, it is expected that the general shape of the curves will be the same, With the shim rods in a fixed position, the regulating rod will be fully inserted and approxi- mately ]/25 of a quart of fuel concentrate will be added slowly to the system. The reactor will be brought critical on the regulating rod and its new position noted. This procedure will be repeated until the regulating rod has been cali- brated from 5 in, above to 5 in. below its mid- position, The above experiment will have been run with the shim rods almost completely withdrawn, The shims will now be inserted about 25% and ]/2 quart of fuel added (approximately 0.5% Ak), The shim rods will be withdrawn until the reactor goes critical. This procedure will be continued to give a rough calibration of shim rod position vs fuel addition over one-fourth of the rod. When the reactor is critical with the shims approximately 15% inserted, the shims will be adjusted so that the regulating rod is 5 in. above center, and a second calibration of the regulating rod vs fuel addition will be run as above, Calibration of the shim rods vs fuel addition will then be continued until the reactor is critical with the rods about 25% inserted. At this time, a third and final calibration of the regulating rod vs fuel addition will be run. The reactor will then be shut down and a fuel sample taken, LOW-POWER EXPERIMENTS The nominal power of the reactor is obtained as follows: It is estimated that in the reflector at the mid-plane of the reactor the fission pro- ducing flux is 2 x 10% nv/w, and the average flux over the length of the fission chamber when fully inserted is 1.5 x 10% nv/w. Preliminary tests on the fission chambers show a counting efficiency of approximately 0.14 counts/sec.nv. The re- lationship between counting rate and power is therefore approximately 2 x 10° counts/sec = 1 w. The nominal power of the reactor will be based on this relationship during the preceding experi- ments, Until this time the reactor has been operated on the fission chambers alone at a nominal power 147 - i - - T r— ik T - DWG. 215254 130 120 10 100 90. @ O - < ROD VALUE (cents) o o . / 40 . / 20 V4 0 2 4 6 8 10 12 14 16 i8 20 22 24 26 28 30 32 34 ROD POSITION {in.} Fig. D.3. Regulating Rod Calibration (Rod B). of about 0.01 w (about 2000 counts/sec). The reactor power will now be increased to about 1 to 10 w, at which time the neutron ionization chambers should begin to give readings on the log N recorder, the micromicroammeter, and the period recorder. The reactor will be leveled out and the flux servo turned on. One of the fission chambers will be completely withdrawn, which should reduce its counting rate by several orders of magnitude, A careful comparison of the new count rate vs the old count rate will be made. The reactor will be held at constant power for exactly 1 hr and then scrammed. A fuel sample will be drawn off and sent to the Bulk Shielding Facility for determining its activity. This activity will be compared with that of a previous sample (Fig. D.5), which was exposed in a known neutron 148 flux in the Bulk Shielding Reactor. From the comparison, the fission rate or absolute power of the ARE will be determined for this run (cf., app. H). All neutron level instruments will be calibrated accordingly. Rod Calibration vs Period. A family of curves of reactivity vs reactor period (inhour curves) with the rate of flow of the fuel as a parameter is shown in Fig. D.6. With the fuel flow rate at 34 gpm, the servo will be turned off and the reactor placed on infinite period manually with the regulating rod at mid- position. From the rod calibration vs fuel plot and the theoretical inhour curve, the regulating rod withdrawal for a 30-sec period will be esti- mated (present estimate is about 1 in,)., The rod will be withdrawn accordingly and the neutron ey »m‘!'l" mm et = e Y o T e e Tomprrwrt b s e ® re oy W "rvr CERET TN T o - oy hmyver- T WpThrRE gy wr W R Pyt gy eEE e i o gy e i, FRRPERUS N N - iy % ki . DWG. 215264 A ROD SENSITIVITY {cents/in.) o 2 4 6 B 10 12 14 i6 18 20 22 24 26 28 30 32 34 ROD POSITION {in.) Fig. D.4. Regulating Rod Sensitivity (Rod B). level allowed to rise about two orders of magni- tude, whereupon the regulating rod will be fully inserted and the induced gamma rays allowed to decay for about 20 min., The reactor will then be brought back to its original power, as de- termined by the fission chambers. If the position of the regulating rod does not return to its original value because of photoneutrons, the reactor power will again be reduced by insertion of the regulating rod until the photoneutron effect becomes negli- gible. The above procedures will be repeated for 20-sec and 10-sec periods. The 10-sec period should correspond roughly to a 2-in. withdrawal of the regulating rod. Runs will then be made at correspondingly longer periods until the period for a Y%-in. withdrawal of the rod is obtained (approximately 100 sec). At this time a repeat run on the 30-sec period will be made to ascertain whether photoneutron buildup is causing appreci- able error in the measurements. The shims will then be adjusted so that the regulating rod is 1in. below center at infinite period. The rod will be withdrawn 1 in. and the period recorded. This procedure will be repeated at successive starting positions of the regulating rod 1in. apart from 5 in. below center to 5 in. above., A check on the initial 30-sec run with the rod withdrawn from the mid-position will be made periodically to ensure that photoneutron buildup is not interfering with the measurements. From the standpoints of safety and experimental convenience, the regulating rod should be worth between 0.3% and 0.5% Ak for its full 12-in. travel. If these experiments show that the value of the rod is not in this range, a new rod will be installed ot this time and the period calibration repeated. |f convenient, a fuel sample will be taken at this time. _ Measurement of the Delayed Neutron Fraction. The delayed neutron fraction for a stationary fuel reactor is about 0.73%. For the ARE ot a design flow of 34 gpm, the fraction is calculated to be 149 —— T TR o o o e COUNTING RATE ORNL-LR-DWG 1946 (CR), ARE I[GTOT }A RE (CRhse (Glare _ e Pae = 8468 x 1070 FUEL CAPSULE, FUEL 44 CARRIER CAPSULE, FUEL 45 0 2 4 6 8 10 i2 14 1A i8 20 22 24 TIME AFTER SHUTDOWN ({ hours) Fig. D.5. Decay Curves for Fluoride Activated in the Bulk Shielding Reactor. 0.47%, and for smaller flow rates, the corre- sponding reactivity may be found from Fig, D.2. Starting with the reactor stabilized at 34 gpm and the flux servo on, the flow rate will gradually be reduced. The calibrated regulating rod should be inserted so as to indicate the same reactivity as shown in Fig. D.2. This will give experimental verification of the previously calculated inhour curves, Fig. D.6. Preliminary Measurements of Temperature Coef- ficient, With the reactor held at 10 w by the flux servo, the helium flow will be turned on so as to drop the fuel temperature at a rate of 10°F/min. As the mean reactor temperature drops, the servo will insert the regulating rod so as to maintain constant flux, and from the rod calibration, the corresponding Ak can be obtained. Before the rod has reached the limit of its travel, the helium 150 flow will be shut and the fuel allowed to return to its original temperature. From the recordings of rod position and mean temperature, a plot of Ak vs mean temperature can be obtained. The initial slope of this curve will correspond to the fuel temperature coefficient. Because of the weak signal received by the flux servo, this measure- ment will be only approximate and is to be repeated at higher power. Depending upon how the reactor responds, the procedure can be repeated by dropping the fuel temperature at rates of up to 25°F /min, Care will be taken not to drop the fuel temperature so low as to set off the low-temperature scram. APPROACH TO POWER At the conclusion of the above experiments and before the fuel storage tank is removed from the " " o T e T R e ooy wEm e - v g b T e 2o TRt TTETEY B ey g hm—5r Wgrwr € g oo r oy Y PRTYE T g e Y o R L ke il xR 4 e Rkt s . i K i u, i s . bk gy Wi b e i, e ki Cw i L Al - DWG. 22352 0.0035 0.0030 \ 0.0025 STATIONARY FUEL N 0.0020 / REACTIVITY 0.0015 / r 13.6 gpom \ ‘/;- 17 gpm \ 0.0010 L 227 gpm e - 34 gpm \ ~ 68 gpm \ "‘\_‘_—-—._.______ \ S ——— 0.0005 QRQ\Q —— e m 0 0 10 20 30 40 50 60 70 80 90 100 10 120 REACTOR PERIOD (sec} Fig. D.6. Reactivity as a Function of Reactor Period for Several Fuel Flow Rates. pits, sufficient fuel concentrate will be added to the system so that 4% excess reactivity is avail- able., This excess reactivity will be absorbed with the shim rods. The shim rods will have a rough calibration by this time, and it is expected that they will be worth about 12% in excess reactivity, Therefore, they will need to be in- serted about one-third of their length. After addition of the final amount of fuel concentrate the liquid level in the pump will be checked to ensure that at least 0.3 ft3 of volume remains for expansion of the fuel. This expansion volume will allow the fuel to expand isothermally from 1300 to 1600°F, which is well above the high- temperature scram level. The final sample of fuel will be taken for chemical analysis. - The reactor will now be shut down so that the concrete block shields can be put over the pits and the pits flooded with helium. The fuel storage tank will first be removed and a final check on all process equipment made. The BF, counter will be removed and the temperature servo chamber installed. The regulating rod will be put on fast drive. A check list is being prepared of all items to be reviewed before the pits are sealed. [f the temperature coefficient is of sufficient magnitude to control the reactor, the flux servo will be left in, However, if the magnitude of the temperature coefficient is marginal, the flux servo will be removed at this time and the temperature servo connected. After the pits have been sealed, the reactor will be brought to a power of 1 kw and allowed to stabilize. Up until this time all neutron ion chambers have been fully inserted. The ion chambers will be withdrawn slowly, one by one, while the power level is being constantly moni- tored with the fission chambers. The ion chambers will be set for a maximum reading of around 5 Mw, 151 The safety chambers will be set to scram at the same level, - The reactor is now ready for full power oper- ation. The rod will be withdrawn and the reactor put on about a 50-sec period, Somewhere in the region from 10 to 100 kw a noticeable increase in reactor period and an increase in reactor temper- atures should be observed. The helium flow will be started slowly and gradually increased until a nominal power of 100 to 200 kw is reached (AT from 25 to 50°F). At this time, the reactor will be allowed to stabilize for about ¥ hr and all readings recorded. If the temperature coefficient is of insufficient magnitude to stabilize the reactor, as determined by previous measurements, the control of the reactor will be turned over to the temperature servo. ' ‘ The power as determined from the heat exfraction by the helium will now be measured. The heat capacity of the fuel is 0.23 Btu/Ib.-°F and the density of the fuel is represented by ' (g/cm?) = 4.04 — 0.0011 T (°C) . At a mean temperature of 1300°F, the average density is 3.27 g/cm3. The power can be ex- pressed as p (kw) = 0.11 x AT (°F) x flow (gpm) . Therefore at 34 gpm the power extracted by the fuel is 3.74 kw/°F. Having obtained o heat balance, the extraction of heat from the fuel in the heat exchanger will be increased until a power of about 500 kw (AT = 134°F) is reached. Again the reactor will be allowed to stabilize and the extracted power measured. A third and final heat balance will be made at a power of about 1 Mw (AT = 268°F). This is close to the maximum power obtainable without changing the initial conditions of 1300°F mean reactor temperature and 34-gpm fuel flow. During the preceding discussion, no mention has been made of heat extraction other than in the fuel circuit. An appreciable quantity of heat will be removed by the sodium in the reflector coolant circuit. Before the reactor goes to high power, the sodium inlet and outlet temperatures will be near the isothermal temperature of 1300°F. Since there is some gamma heating in the reflector region, it will be necessary to lower the inlet temperature of the sodium by extracting heat with the reflector coolant heat exchangers. The sodium flow rate will be held at 224 gpm and the sodium 152 mean temperature at 1300°F. Since the heat capacity of the sodium is 0.30 Btu/lb.°F and its specific gravity is 0.78, the power extracted by the sodium will be 7.7 kw/°F. No more than 10% of the power generated is expected to go into the sodium, EXPERIMENTS AT POWER Measurement of the Temperature Coefficients, After the reactor power has been calibrated, the power will be reduced to about 500 kw and allowed to stabilize. The flux servo will be turned on, and by means of the helium demand the mean temperature will be dropped at an initial rate of 10°F /min. (lf the temperature servo is connected, it will have to be replaced for this experiment.) As discussed previously the initial slope of the plot of Ak vs mean temperature will represent the fuel temperature coefficient, Contrary to the procedure at low power, however, the helium flow will not be decreased after the temperature has dropped. The reactor will be allowed to stabitize, and, after about 30 min, the regulating rod will have leveled out at a new position, and the reactor will have assumed a new mean temperature. These readings will represent the over-all reactor temper- ature coefficient (fuel plus moderator). If, during the above experiment, the servo inserts the rod to its limit, it will be left at that position since the temperature should stabilize the reactor. [f the fuel temperature coefficient is quite small, the experiment can be repeated with an initial rate of temperature decrease of 25°F/min. Care will be taken that the reactor does not go on too fast a period and that the low-temperature scram fimit is not exceeded. Maximum Power Extraction. Except for specifi- cally stated instances, the reactor has been operated continuously up to this time at a mean temperature of 1300°F and a fuel flow rate of 34 gpm. Under these conditions, the maximum obtainable power is 1.1 Mw. At this power, the reactor outlet temperature and the heat exchanger inlet temperature are both 1450°F. The heat exchanger outlet and the reactor inlet are both at 1150°F. It is to be noted that when the heat exchanger outlet temperature drops below 1150°F, an alarm is sounded. The mean reactor temperature will be elevated from 1300 to 1325°F by a slight withdrawal of the shim rods. The mean temperature of both fuel and sodium will be held at this temperature. The helium blower speed will now be increased until e . w EETCFOTS - e e "R, »rm e T mmwree Y SoEEE WYR wmmAR WETEWT. mETwT . i L s vy s s b Sk o bR, B bRy g a AT of nearly 350°F appears across the reactor. With this AT, the reactor outlet and heat exchanger inlet are at the upper limit of 1500°F, and the heat exchanger outlet and reactor inlet are at the lower limit of 1150°F. The power from the fuel will be 1.3 Mw, At this time the pump speed can be increased from 34 gpm to about 40 gpm, caution being taken that the reactor inlet pressure does not exceed 50 psig. The maximum AT across the reactor will still be 350°F, and the power ex- tracted in the fuel circuit will be 1.5 Mw. This is the maximum power at which the reactor can be operated. Power Transients. With the reactor on manual control and the regulating rod at mid-position, the helium blower speed will be regulated until the power in the fuel system is 1 Mw, and the reactor will be allowed to stabilize. The helium flow will be suddenly increased to its maximum and the transient response of all temperature and nuclear recordings noted. The experiment can be repeated from successively decreasing initial powers of 500, 200, 100, 50, 20, and 10 kw. At some level below 100 kw, the power cannot be reduced further because the helium blowers will be completely shut off. When this lower limit of initial power has been reached, the series of experiments will be concluded. !f at any time the reactor period gets too fast or the upper or fower temperature limits are approached, the ex- periments will be concluded. Sudden Changes in Reactivity, The reactor will be brought to a power of 1 Mw and allowed to stabilize, From the previous calibration of the regulating rod, the rod will be placed at a position such that a complete withdrawal will give 10 cents of excess reactivity. The rod will suddenly be withdrawn and the transient response of the inlet pressure and all nuclear and temperature recorders noted. The experiment can be repeated with sudden changes of 25, 50, 75, and 100 cents of reactivity. If the fuel temperature coefficient has a value of approximately -5 x 103, as expected, a sudden change in reactivity of 100 cents should be safe. However, if the experiments with smaller reactivity changes indicate that such a large step will be unsafe, this series of experiments will be concluded. Otherwise, the experiment will be repeated from initial powers of 100 kw, 10 kw, and successively lower power levels, Reactor Startup Using the Temperature Coef- ticient., The reactor will be brought to 1 Mw of power and the shims adjusted so that the regu- lating rod is completely withdrawn, The helium flow will be cut off and the reactor power allowed to drop to its normal power with no heat extraction (estimated to be between 10 and 100 kw). The regulating rod will now be inserted by 0.1% of reactivity and the reactor allowed to go subcritical for about 10 min. After this time, helium flow will gradually be increased. If the temperature coefficient is =5 x 10=3, a drop in the mean temperature of the reactor of 20°F will bring the reactor critical again. As soon as a positive period is noted, the helium cooling flow will be held fixed and the reactor allowed to come up to power and level out of its own accord. The experiment can be repeated by driving the reactor subcritical by 0.2, 0.3, and 0.4% with the regulating rod, Care will be exercised not to approach the upper or lower temperature limits, Effect of Xenon Buildup. The change in re- activity calculated from xenon buildup in the ARE is shown in Fig. D.7. The reactivity as plotted is nominal because of uncertainties in the xenon cross sections and is probably a maximum. The shape of the curves represents the change in reactivity if no xenon is lost by off-gassing. The reactor will be operated for 5 hr at full power, and then reduced to 10% of full power. If no xenon is lost, the reactivity should change as shown in the lower curve of Fig. D.7. The reactor will then be operated for 25 hr at full power, and subsequently reduced to 10% power. Again if no xenon is lost, the reactivity change should be as shown in the upper curve of Fig. D.7. If some xenon does off-gas, the shape of the curves will be changed, and by proper analysis, an estimate of the amount of off-gassing can be obtained. At full power, the reactivity change will be calculated from the change in reactor mean temper- ature and the temperature coefficient. At 10% power, the reactor will be put on flux servo and the movement of the regulating rod will measure the reactivity change. Since the moderator will contain considerable heat when the power is reduced, the mean reactor temperature will slowly decrease during about the first % hr after the power is reduced to 10%. This will cause an insertion of the regulating rod by the flux servo. Since the xenon buildup will cause a withdrawal of the regulating rod, a correction must be applied for the effect of the drop in moderator temperature., 153 i &; b i i i ol e T vsi 1072 REACTIVITY (NOMINAL) S 1 [¥] FULL 1O % POWER | POWER —— e 1074 To 2 4 . ,,v e T e oS ¥ Vv i !!*"W ‘pr&.\ '”“M Ry =m mm TIME (sec) Fig. D.7. Effect of Xenon Buildup, oot R e TE E e pmercgpepg 8 T m T N P o s DWG. 22354 . i, L& Ry el To obtain this correction experimentally, a control experiment will be performed. The reactor will be operated at full power for ‘/2 to 1 hr and the moderator allowed to come up to temperature. During this short time, no appreciable xenon will be formed. The reactor will then be reduced to 10% power and the flux servo turned on, The reactivity change from the drop in moderator temperature will be observed and used as a cor- rection for the xenon buildup experiments. 155 . T N ST - e e e Appendix E MATHEMATICAL ANALYSIS OF APPROACH TO CRITICALITY . W. E. Kinney When the ARE was brought to critical by suc- cessive fuel additions, it was observed that the usual plot of [1 — (1/multiplication constant)] vs uranium concentrafion increased, at first, very rapidly, but, when the curve got close to 1, the rise was very slow. Qualitatively, such behavior is observed in many reactors, but the ARE ex- hibited the effect to an unusual degree., In order to explain this, the ORACLE three-group, three-region code was modified so that flux shapes at succes- sive fuel additions could be calculated. Group constants were obtained by flux weighting with fluxes from an Eyewash calculation on the ARE, Figure E.1 shows the space distribution of the thermal flux for no fuel and for runs 2 through 6. The effect of the reflector as fission neutrons be- come more nhumerous can be seen, Figure E,2 compares experimental and calculated startup curves for the fission chambers which were located in the reflector as indicated in Fig. E.1. In the calculation, where CR is the counting rate of the fission chamber, o7 is the fission cross section for group i, and ¢, is the group i flux. For the ARE startup CR]. ORNL—~LR—-DWG 4450 5 ¢ | | | | 2 | | | . | © | | 4q | | | | /\ | | 3 \\W/ ! N ! l § I | PRESSURE | | | 365 | & CORE i—a REFLECTOR i | E \ I 2 I | ! i ] e | | | : 4 \ | l | i [s - | I 0 \o\o\o__o___?,—c[-/"fi\ | 2 xl | 0 0 {0 20 30 40 50 60 70 RADIUS {cm) Fig. E.1. Thermal Flux vs Radius. 156 % t i Y g g g e B g g A R T @mg‘m'vmw?“?\ TR T e T R ey Brnemmrmr LA el L ] ol RN - b i S MEdel g s e, ey ool B M iy, | e, B i, o e e e G, s i ., ORNL—LR—DWG 4451 1.0 0.8 0.6 E ® CHAMBER | = A CHAMBER 2 I © COMPUTED 0.4 0.2 0 0 2 4 6 8 10 12 14 16 18 U235 CONCENTRATION (Ib/f3) Fig. E.2. 1 - (1/m) vs U235 Concentration. where m is the multiplication constant, CR . is the counting rate on run j, and CR_ is the counting rate with no fuel. It seems, then, that the un- expected, rapid initial rise in the counting rate of the fission chamber and the [1 ~ (1/m)] curve is due not only to the general rise in flux level but also to the formation of the thermal-flux maximum near the fission chambers. Once the shape of spatial distribution of the thermal-neutron flux is set up, the fission chambers register only the general increase in flux level and the count rate increases slowly, 157 ' g a ¢ i T ol R FROTTYCI A e 1 Appendix F COLD, CLEAN CRITICAL MASS The value of the cold, clean critical mass is of interest in connection with any reactor, since it is the calculation of this elusive number that takes so much time during the design of the reactor and system. The cold, clean critical mass may be found by extrapolating the curve of uranium (Ib of U23%) in the core vs Ak/k (%) in the rods to the point where there is no rod poisoning. On the other hand, if the mass equivalent of the rod poisoning is determined and deducted from that then in the core, another curve may be obtained. This curve gives corrected mass vs Ak/k (%) in the rod and may also be extrapolated to 0% Ak/k in the rods. Both these curves, as shown in Fig. F.1, extrapolate to the same mass value at 0% Ak/k in the rods, which indicates that the rod calibration was quite accurate and the data were reliable. The value of the cold, clean critical mass extrapolated from Fig. F.1 was 32.75 |b of U235. The data from which the curves in Fig. F.1 were plotted are tabulated in Table F.1. The data used to determine the cold, clean critical mass were those obtained during the rod calibration from fuel addition (Exp. L-2). This experiment and Exp. L-5, rod calibration from reactor period, then provided the information on the value of the regulating rod, The values of the shim rods were taken from Appendix J, ‘“Calibration of the Shim Rods.*” The Ak/k values in both regulating and shim rods were then converted to their mass equivalents by using the (Ak/k)/(AM/M) ratio of 0.236, as determined experimentally, Three points listed in the table and shown in the figure were derived by using the first regulating rod, which was subsequently re- placed because it was too ‘‘light.”” The calibra- 158 tion for that rod was not accurate, and, hence, the three points were disregarded in extrapolating the curves, ORNL—LR—DWG 6419 36 T I CURVE J ‘ 35 A: U255 N CORE vs. Ak/k IN RODS B: Ak/k IN RODS vs. U2% IN CORE CORRECTED TO NO ROD POISCNING BY USING (A4/k) / (AM/M) =0.236 | | | O OBTAINED WITH FINAL REGULATING ROD 34 [~ e OBTAINED WITH INITIAL REGULATING ROD, WHICH WAS ™ REPLACED BEFORE BEING WELL CALIBRATED i 50T A 33 /j = —_ e | - 2 Rgers ¢ o N E N 32 \‘\\ . }\ . 30 ] 29 O Q2 04 Q86 08 1.0 1.2 Ak/k IN ALL RODS () Fig. F.1. Extrapolation to Cold, Clean Critical Mass. T e . g > M rprapety B Tt g W o vy g gt © B B el - " e g e eSS — pdig, W RS k. TABLE F.1. CRITICAL URANIUM CONCENTRATIONS FOR VARIOUS ROD INSERTIONS A W aaie . Lo S A e, P o e an e, dn i, . - I‘ (Exp. L-2) Regulating Shim Rod Insertion AL/E of Ak/F of Shim Rods y2ss U235 Total Ak Total U235 i, Run Ne, Rod Insertion (in.) Reguiating (%) Concentration in Core in Rods AM/M in Clean Core (in.) Ne.T No.2 No.3 Rod(% No.1 No.2 No.3 (Ib/#3) (ib) (%) Rods (Ib) 0 0.92 3.3 1.7 0 0.016 0.19 Q.08 O 23.94 32.80 0.286 1.212 3159 A 1 3.35 3.3 1.7 ¢ 0.058 0.19 008 0 23.98 32.85 0.328 1.390 31.46 k 2 0.93 3.3 2.5 ¢ 0.016 0.19 0.13 0 24,12 33.04 0.326 1.381 31.66 k 3 4,25 0 0 3.5 0.140 0 0 0.20 24,17 33.11 0.340 1.441 31.67 4 6.35 0 0 3.5 0.210 0 0 0.20 24.22 33.18 0.410 1.737 31.44 ; 5 7.5 0 0 3.5 0.248 0 0 0.20 24.25 33.22 0.448 1.898 31.32 : 6 8.5 0 0 3.5 0.281 0 0 0.20 24,27 33.25 0.481 2,038 31.21 Level 7.4 2.2 2.5 0 0.244 011 013 O 24,27 33.25 0.484 2,051 31.20 i Trimmed 7 8.0 2,2 2.5 0 0.264 0.1 013 0 24.29 33.28 0.504 2135 31.14 k. 8 8.52 2.2 2.5 0 0.281 ¢.1t 03 ¢ 24,31 33.30 0.521 2.207 31.09 lr 9 10,92 2.7 2.8 0 0.360 0.14 015 0 24.41 33.44 0.650 2,754 30.69 10 1.25 5 5 ¢ 0.041 033 033 0 24.43 33.47 0701 2,970 30.50 E 11 11.56 4 3 o 0.382 023 016 0 24.44 33.48 0.772 3.271 30.21 :” b 159 T T IR R i oo ¥ i, s, i Appendix G FLUX AND POWER DISTRIBUTIONS A. D, Callihan There were, of course, no measurements of the flux or power distribution in the reactor during the actual experiment, These distributions were measured, however, on a zero power, critical mock- up of the experimental reactor over a year earlier. A detailed description of the critical mockup, in- cluding the reactor parameters obtained therefrom, may be found in ORNL-1634,! The neutron flux distributions obtained from indium and cadmium- covered indium foil measurements, the fission neufron flux distributions obtained by the catcher- 1A. D. Callihan and D. Scott, Preliminary Critical Assembly for the Aircraft Reactor Experiment, ORNL- 1634 (Nov. 28, 1953). D. Scott foil method, and the power distributions obtained from aluminum catcher foils, which are of particular interest, are presented here, NEUTRON FLUX DISTRIBUTIONS The neutron flux distributions were measured with indium and cadmium-covered indium foils in a number of runs in which a remotely placed uranium disk and an aluminum catcher foil were used to normalize the power from run to run, The results of the bare indium and cadmium-covered indium traverses made at a point 12,06 in. from the center of the reactor are shown in Fig. G.1. The zero of the abscissa is the bottom of the beryllium oxide DWG, 24529A 70 AT 12.06 in. RADIUS INDIUM TRAVERSE, LONGITUDINAL 60 /1’" 0 / | —BARE INDIUM ACTIVATION A 0 = % /—_Q\ x 40 / 0.8 & ) = & CADMIUM -COVERED < INDIUM ACTIVATION > E 30 \ 0.6 = l—- g g 5 /CADMIUM FRAGTION < 20 \ 0.4 & / — N\ —" TOP OF BeO = e O e 5{) o 10 e N 0.2 BARE INDIUM ACTIVATION MINUS | CADMIUM-COVERED INDIUM AGTIVATION INCONEL SUPPORT PLATE 0 =~ | | 0 0 6 12 18 24 30 36 DISTANCE FROM BOTTOM OF REACTOR (inl) Fig. G.1. Longitudinal Neutron Flux Distribution. 160 PR . S AR SRS e YT wgmE PT T oW eerTe ¢ s W T T YTEECETE b Her Ty FATT T v BPEDE R R R W T o R e e e . -ty ¥ e ke ™ e s ulk, i i, Seedeni, b i e e i column where it rests on the l-in.-thick Inconel support plate, The scattering by the Inconel probably accounted for the reduction of the cad- mium fraction at this point, ’ The radial flux traverse at the mid-plane of the reactor with the regulating rod inserted is given in Fig. G.,2, The dashed lines on the figure are activities extrapolated from the data obtained in the fine-structure measurements made near the 11-in. position. The wide gap in this traverse was unexplored because of the importance which was attached to the study of a unit core cell in the time available., This emphasis has been at least partially contradicted by the fission flux traverse described in the following section. The strong effect of the center regulating rod assembly is indicated in these curves. The comparatively small flux and cadmium fraction depressions at the center of the regulating rod accentuated the relative inefficiency of this type of rod and guide tube arrangement for reactor control. In another measurement of the radial flux, this time with the regulating rod withdrawn, the traverses were very similar to that shown in Fig. G.2, but the neutron flux was slightly greater (~8%). FISSION-NEUTRON FLUX DISTRIBUTION A measure of the distribution of neutrons that caused fissions was obtained by using the catcher- foil method in which the aluminum foil, together with some vuranium, was both bare and cadmium covered. A radial traverse taken at the mid-plane of the reactor is shown in Fig. G.3. The flux of low-energy neutrons produced fission peaks near the reflector and was depressed by the Inconel tube (regulating rod sleeve) at the center. The DWG 215304 90 I 1 I | I n | 80 l/’o—. \\ INCONE L GUIDE }\ \ . i TUBES/ } \(‘,?\ OON S _ ! i /./'J) ! \ \"l\ 1 C.7 ¥ 1 i . 8] \ / \ \¢\~ A - S 40 . = 04 2 = ~~{ [® '}n\y/’/_ ' = e B CADMIUM FRACTION| ,* '\ | e W Jj / N7 RADIAL INDIUM TRAVERSE N = —— 7 f oD -~ TN = | / AT MID-PLANE WITH REGUL ATING 03 = 30 v 7 \ 7 2 3 ; - i [ ROD IN S ' {7770 M NS B © \ \ i 1 ) - 20 e 1 \\ \“JI l r"\ £ 0.2 /\;‘\f I‘;‘ \ /’ \ ’\ T N v/ “ V| L-BARE INDIUM ACTIVATION| ™7 XN > MINUS CADMIUM-COVERED 0.4 INDIUM ACTIVATION & FUEL ‘ (\Cfl TUBES 4 0\ _ 1 L1 | | | | | 1 ! | o 4 6 8 10 12 14 16 18 20 22 24 DISTANCE FROM AX!S OF REACTOR (in) Fig. G.2. Radial Neutron Flux Distribution. 161 B RO TR T TR 1 b L k L I cadmium fraction followed roughly the same pattern. Similar half-length longitudinal traverses are shown in Fig., G.4 for a position 10.09 in. from the center, POWER DISTRIBUTION Three longitudinal power distributions in this reactor assembly were measured by using aluminum catcher foils placed against the end surfaces of the short, cast, fuel slugs contained in an Inconel tube., The size of these foils was such that the counting rate from each was higher than necessary for the desired statistics. In the arbitrary units reported, an activity of 50 represents approximately 105 counts. Had time permitted, it would have been desirable to repeat the experiment with smaller foils to improve the resolution, particularly since the results were very sensitive to foil loca- tion because of fuel self-shielding. Each catcher foil was nominally located centrally on the axis of a fuel slug. One traverse extended from below the Inconel support plate to above the top of the beryllium oxide; the other two covered the upper half of the core. The data obtained in the fuel tubes at the indicated radii are shown in Fig. G.5. The data of Fig. G.5 are replotted in Fig. G.6 to show the radial power distribution at several ele- vations in the reactor, all normalized at the center. It is to be noted thatthe 37.86-in. elevation traverse is 2 in. above the top of the beryllium oxide column. DWG. 24533A 14 l CATCHER-FOIL FUEL DISK RADIAL AT MID-PLANE f2 ? i t\ i : | . | v / | \ \3\/ e 2 d } N | = | 1 ‘ |_BARE ACTIVATION = ' i & - "~ CADMIUM FRACTION o« | £ 1 = ‘ % BARE ACTIVATION MINUS = CADMIIM-COVERED ACTIVATION = ~ 6 0.6 © > ‘ S 5 = b= | ! | uw |- INCONEL GUIDE TUBES i : s i ! i 2 | | ? 2 4 x ‘ 0.4 2 P i : QD ! 1 i : i : : | i 5 ; CADMIUM - COVERED ACTIVATION e L 2 i ‘ 1 \ f 0 L l ‘ ‘ 0 0 3 6 9 24 27 DISTANCE FROM AXIS OF REACTOR (in) Fig. G.3. Radial Fission Neutron Flux Distribution, 162 . & & EEEEE Y WEEE W e T e R e b & R "*”’WM TTRE Tt e W T T A PRI T A Rt I BB o g By PERTIRNT T Wippr e et oty i i ACTIVITY (arbitrary units) DWG 215344 | T | CATCHER - FOIL FUEL DISK LONGITUDINAL AT 40.09 in. RADIUS T ll 1.2 A~ ,BARE ACTIVATION ] 1.0 '\ = \ /CAOMIUM FRACTION 5 — N r\ -~ 08 g - Lo = BARE ACTIVATION MINUS 5 CADMIUM - COVERED ACTIVATION \ 06 2 2 ) \ 0.4 N CADMIUM - COVERED ACTIVATION\ * \ 0.2 \A @ 0 8 12 16 20 24 28 32 36 40 DISTANCE FROM BOTTOM OF REACTOR (in.) . Fig. G.4, Longitudinal Fission Neutron Flux Distribution, 163 : | e YT ACTIVITY (ARBITRARY UNITS) 164 DWG 2(535A 90 CAST FUEL IN INCONEL 80 TUBE LONGITUDINAL AT INDICATED RADIUS 0O \ & | L) [0 CAST FUEL IN INCONEL TUBE RADIAL AT 5 ELEVATIONS O 2 4 6 8 10 12 14 186 DISTANCE FROM AXIS OF REACTOR {in.} Fig. G.6. Radial Power Traverses. b Y E. rrpee T * pgr o e CmerermE € v CEETEET YR W ommogreer rrmerrar o g € i e e E “rp PO TEme et e e ERTEeT IRy 07 ¥ Y ekt ey e © i e B i, i Ry, e B B i, Lo, bR, PR 4 o ey : Appendix H POWER DETERMINATION FROM FUEL ACTIVATION' E. B. Johnson Perhaps the most basic value obtained from the operation of the ARE was the power level at which it operated. This quantity was determined both from activation of the fuel and from a heat balance. The determination of reactor power from activation of the fuel was based on the measurement of the relative activity of samples of the fuel exposed in the ARE and that exposed in a known flux in the Bulk Shielding Reactor (BSR). The activity of the aggregate of fission products is not easy to predict because of the large number of isotopes for which calculations would be re- quired. For identical conditions of exposure and decay, that is, for identical flux, exposure time, and waiting period after exposure, the specific activity should be the same, however. Further- more, at low fluxes the activity should be pro- portional to the flux. The ARE was operated at a nominal power level of approximately 1 w for 1 hr, was then shut down, and a sample of the irradiated fuel was withdrawn for counting and analysis. Since the activity in this fuel sample was too low to measure accurately, the test was repeated at an estimated power of 10 w. The ARE_experimental program then pro- ceeded as scheduled. Decay curves were obtained from the fuel samples from the ARE and were then compared with the decay curve of a similar fuel sample irradiated in the BSR in a known neutron flux of about the same magnitude. From the re- lative activity of the samples, o determination of the power of the ARE was made. THEORY 2 It has been shown® that the power (in watts) produced in an enriched-fuel thermal reactor is P = 4264 x 107" x nv, x G where G is the number of grams of U235 in the volume over which the average thermal-neutron flux is nv,,, and the constant, 4.264 x 10~ 1! contains 1This appendix was originally issued as ORNL CF-54-7-11, Fuel Activation Method for Power Determi- nation of the ARE, E. B. Johnson (July 31, 1954), and was revised for this report on April 22, 1955. 2), 1, Meem, L, B, Holland, and G. M, McCammen, Determination of the Power of the Bulk Shielding Re- actor, Part lIl. Measurement of the Energy Released per Fission, ORNL-1537 (Feb. 15, 1954). fore (P/G) the fission cross section, energy release per fission, and the necessary conversion factors. Obviously, if the thermal-neutron flux and the amount of fuel present are known, it is quite simple to obtain the power. The problem was to obtain the power without measuring the flux in the ARE. The quantity P/G obtained from the measurement in the BSR will be called (P/G)BSR The decay curve for the two fuel samples irradiated in the BSR is shown in Fig. D.5 of Appendix D. The power production in a given sample is proportional to both the counting rate and the amount of fission- able material in the sample. Thus, (P/G)ARE (CR/G)ARE (P/G)BSR (CR/G)BSR where CR = counting rate at time ¢ after shutdown, G = weight of uranium in the sample. This equation can be rewritten as CRARE GBSR = (P/G) X e— X BSR CRBSR GARE (P/G) ARE The total power of the ARE is then the product of the power per gram (in the sample)} and the total amount of uranium in the fuel circuit (Gtot)ARE‘ or P arE = (P/G)ape * {Gio)are (P/G) CRARE = X o—_—_— BSR CRgsr Gesr X G x «%OJARE ARE The unperturbed thermal-neutron flux in which the samples were exposed in the BSR was 1.895 x 107 neutrons/cm?.sec. However, the self-depression® of the flux by the uranium sample was 0.80. There- BSR become.s (P/G) = 4.264 x 10~ 11 % 1.895 x 107 x 0.80 BSR = 6.464 x 107% w/g 3w, K. Ergen, private communication. 165 ST T T R TSP peT T T g G e, s ot ol bt Wik s b de BB T s e e R e b S e E S S s e e el e, AL < : Each of the fuel samples irradiated in the BSR contained 0.131 g of U233, Thus CR ARE 6.464 x 1074 w/g x — ——— CRBSR 0.131 i PARE It 8.468 x 107> x — (Gfor)AR E CARE X This equation was then used to determine the power of the ARE as indicated by fuel activation during the previously mentioned operation of T hr at 1- and 10-w nominal power. The instruments which recorded the neutron level (log N, etc.) were then calibrated in terms of reactor power. EXPERIMENTAL PROCEDURE The BSR is immersed in a pool of water which serves as moderator, coolant, reflector, and shield. Therefore, any material which is to be activated in the BSR must be placed in a watertight container before immersion in the pool. Capsules to contain the ARE fuel were made of 25 aluminum with screw- type caps and were sealed with a rubber gasket. Each capsule contained approximately 1 g of ma- terial; one of the capsules is shown in Fig. H.1. Since it was desirable to irradiate the capsules in the reactor core, the reactor was loaded with a partial element, containing only half the usual number of plates, in the interior of the lattice, as /\ ORNL-LR-DWG 1947 Fig. H.1. Apparatus for Fuel Exposure in the Bulk Shielding Reactor, 166 > T p e ¥ Y e TR T L b TN P rErmEr T e | BTNt o m— " ————— 'F’V@"T”'W B — W e mer ol i, PN Sl . ek S SRRk B i d. " shown in Fig. H.2. A lucite holder was designed which would fit inside this partial element and support the capsule to be irradiated at approxi- mately the vertical centerline of the element. Small gold foils for measuring the thermal-neutron flux were mounted in ‘‘drawers’’ adjacent to the cap- sule, as shown in Fig. H.1. A monitoring foil was placed on the bottom of the lucite block. ORNL-LR-DWG 12483 PARTIAL ELEMENT FOR OO ) )O6 ® @) hehe) o)) ®® @)= ) =)0 ) =)@ @ @O @)=)=)f)=))® @ © @]« ]=])® @ E@@@MB%@D@@ PARTIAL ELEMENTS / \ CONTROL ROD Fig. H.2. Bulk Shielding Reactor Loading (No. 27). Two samples of each of two ARE-type fuels of different composition were irradiated. One (No. 44) contained 53.5-40.0-6.5 mole % of NaF-ZrF - UF, and was the anticipated fuel of the ARE. The other was No, 45, which was the carrier to which the uranium-containing fuel concentrate was added during the critical and low-power experiments; it contained 50-50 mole % of NuF—ZrF4. [t was necessary in the BSR to expose the samples in the aluminum capsules; at the ARE the fuel was, of course, first activated and then poured into the capsules. Therefore it was necessary to determine the contribution of the aluminum capsuie to the activity observed. Two empty capsules were exposed (in separate runs) for 1 hr at a nominal power level of 1 w. At the same time, gold foils were exposed in the positions adjacent to the capsule to determine the thermal flux. Similarly, two carrier-filled capsules and two fuel-filled cap- sules were irradiated, each separately but each for 1 hr at 1 w. The capsules each contained 1 g of material. The decay curves taken on each capsule were made with two counters because of the difference in disintegration rates between the empty and the fluoride-filled capsules. A scintillation counter was used to obtain the curves for the empty cap- sules and for the carrier-filled capsules, and the high-pressure ion chamber was used to measure the activity in the fuel-filled capsules. The data for the carrier-filled capsules were converted, on the assumption that all the activity was attributable to sodium, to equivalent countingrates in the high- pressure ion chamber. The assumption that the activity was due to sodium was apparently valid, since the decay curve indicated a half-life of about 15 hr at approximately 6 hr after shutdown. [t was found from the decay curves that the aluminum activity was a factor of approximately 10 less than that of the carrier-filled capsules 3 hr after shutdown and a factor of 450 less than that of the fuel-filled capsules at the corresponding time, and was therefore negligible. The activity in the carrier-filled capsule was about 4% of that in the fuel-filled capsule 6 hr after shutdown. There was essentially no difference between the two different fuel-filled capsules and the two dif- ferent carrier-filled capsules. For this reason, only one curve is shown for each in Fig. D.5. The nuclear power of the Aircraft Reactor Experi- ment was then set on the basis of the activation in fuel samples irradiated for 1 hr at 10 w. This power calibration was the basis of all power levels indicated in the nuclear log up to the last day of the experiment, by which time it had become obvi- ous from the various heat balances that the actual reactor power was considerably higher (app. L). A comparison of the two methods of power cali- bration is given in Appendix N. 167 v e e b F - Appendix | INHOUR FORMULA FOR A CIRCULATING-FUEL REACTOR WITH SLUG FLOW! W. K. Ergen As béin’red out in a previous paper,? the circu- circulating-fuel reactor has been discussed in some lating-fuel reactor differs in its dynamic behavior from a reactor with stationary fuel, because fuel circulation sweeps some of the delayed-neutron precursors out of the reacting zone, ond some delayed neutrons are given off in locations where they do not contribute to the chain reactions. One of the consequences of these circumstances is the fact that the inhour formula usually derived for stationary-fuel reactors,? requires some modifica- tion before it becomes applicable to the circulating- fuel reactor, The inhour formula gives the relation between an excess muitiplication factor, introduced into the reactor, and the time constant T of the resulting rise in reactor power. If the inhour formula is known, then the easily measured time constant can be used to determine the excess multiplication factor, a procedure frequently used in the quantitative evaluation of the various arrangements causing excess reactivity, Further- more, the proper design of control rods and their drive mechanisms depends on the inhour formula, Frequently, the experiments evaluating small excess multiplication factors are carried out at low reactor power, and the reactor power will then not cause an increase in the reactor temperature, This case will be considered here. In this case, the time dependence of the reactor power P can be described by the following equation:4 (1) dP/dt = (/)k,, - PP +‘Bfo°°o(s) P(t - s)ds] . 7 is the average lifetime of the prompt neutrons and k__ the excess multiplication factor (or excess reactivity), The meaning of 8 and D(s) for a TThis appendix was issued earlier as ORNL CF-53- 12-108, The Inbour Formula for a Circulating-Fuel 1’;Iucle_ar Reactor with Slug Flow, W, K, Ergen (Dec. 22, 953). 2william Krasny Ergen, J. Appl. Phys., Yol. 25, No. 6, 702-711 (1954). SSee, for instance, S. Glasstone and M. C, Edlund, The Elements of Nuclear Reactor Theory, D. Yan Nostrand Co., lnc., 1952, p. 294 ff. 4Some authors, for instance Glasstone and Edlund, loc, cit., write the equations corresponding to (1) in a slightly different form., The difference consists in terms of the order k_ (T/T) or k__f, which are negligibly small. ex ex 168 detail in ref. 2, We approximate in the following the actual arrangement by a reactor for which the power distribution and the importance of a neutron are constant and for which all fuel elements have the same transit time 6, - ¢ through the outside loop. In this approximation, 8 D(s) is simply the probability that a fission neutron, caused by a power burst at time zero, is a delayed neutron, given off inside the reactor at a time between s and s + ds. D(s) is normalized so that (2) JDls)ds =1 . If the fuel is stationary, 8D(s) is the familiar curve obtained by the superposition of 5 exponentials: 2 “As (3) BD(s) = LBrge 7 . i=1 The A, are the decay constants of the 5 groups of delayed neutrons, and the (3, are the probabilities that a given fission neutron is a delayed neutron of the 7th group. For the circulating-fuel reactor we first consider the fuel which was present in the reactor at time zero. At any time s, only a fraction of this fuel will be found in the reactor. This fraction is denoted by F(s), and by multiplying the right side of (3) by F(s), we obtain the function 8D(s) for the circulating-fuel reactor, Since @, is the total time required by the fuel to pass through a complete cycle, consisting of the reactor and the outside loop, it is clear that ats = nfll (n =0,1,2,..), F(s) is equal to 1; at s m9]+6 (n = 0,1,2...), F(s})is equal to zero. (We assume 91 2 20 so that the fuel under con- sideration has not started to re-enter the reactor when the last of its elements leaves the reacting zone.) Between s = nf, and s = n0, + 6, F(s) decreases linearly, and hence has the value (n0, + 0 ~ s)/6. Ats =nl, - 6(n = 1,2, 3...), W T RTT AR MW g BT R M S TEEET N Yol Sy LT T s e T R R PWET TR § WEPER M Fwy-cEr wemoTRRT e e - - T T e ki R N Ak - R X ek B, ! F(s) is zero, but since the fuel under consideration re-enters the reactor between this moment and s = n@,, F(s) increases linearly: F(s) = (s ~ n@; + 8)/8. For BD(s) we thus obtain: n@ + 0 —- s BD(s) = ZBZ is forn@lgsgnel-}-@, n=012..., i=1 4 S—n6]+6 5 A ) BD(S):”“‘_"E_——EB forn@l-—egsénfil,n=],2,3,..., i=1 BD(s) = 0 forn6]+9§s§(n+])61—9,72=0,],2,.... Equation (4) is now substituted into Eq. (1), and for P we set P = Poet/T. Then n6+6n6 + 0 - s As - 5 /T T i - .Fpoel = (kex - B) Poet/ + ZIB!AZ Z f e I Poe(t S)/T dS L= @ n@ s —nf, + 6 + E ! "*——“*“——'—‘] e*Ais P elt=s¥VT gg 0 0 n=1 nBI—G t/T The common factor Poe cancels out. The substitution o = n@l + @ — s transforms Jio "% w0, + 0 = ) expl-Dn, + (1/T)s} ds info exp{—n@l}h + I/T)]} exp{ oA, + ]/T)] foor exptlA; + (1/T)lol do , and the substitutiono = s — n@ + 0 tronsforms Js, o _g s —nty +0) expt—[A; + (1/T)] s} ds into expi—n0 A, + (1/T)I} expiblr; + (1/T)} foeoexp{—[)\i + (V/T)l ol do The geometrlc series exp {—n[)\ + (1/T)10,} can now be summed, and the integrals over o evaluated by elementary methods. After performlng all 'rhese operations, one obtains the following inhour formula: = ;0 .8 - (8,=0 T ]5 pt —= 1 + e - e z](LLZ-8+|)+eM’(] ) (5) by =t B - E - , . | = R - e N (6) p, = A+ (1/7) B is obtained by integrating Eq. (3) from 0 to . Expressions are obtained which are of the same type as the ones just dlscussed and whnch can be evaluated by the same methods. The result is ' ~A.8 ~-A.8 ~X.(By =6 | s AB-Tlie o l]()ti0+])+e 179 7 p =Y A Y =1 (1 —e 1 169 T T —— - - EREoT T TR TR - — Y T In spite of the formidable appearance of Egs. (5), (6), and (7), it is easy to find, for any given 0 and 6,, the value of £_ which produces a given time constant T, Furthermore, the following reasoning describes the general features of the equations. Consider first the dependence of k__on T. I T =, .= A, and the sum on the right of (5) is equal to (3, k__ is equal to zero, This corresponds to the state in which the reactor is just critical. |f T becomes very small, the u. become very large and in the fraction on the right of (5) the numerator is domi- nated by ;.0 and the bracket in the denominator by 1. Hence the fraction tends to zero like 0/p;, as T goes to zero, If 7 is very small, as it is in practice, T will be small as soon as % , exceeds 3 by a small amount. Then the complicated sum on the right of (5) is of little importance, and T is determined by 7/T = k__ - 3, that is, the reactor period is inversely proportional to the excess of the reactivity over the reactivity corresponding to the ‘‘prompt critical’’ condition, In the stationary-fuel reactor, the k__ which makes the reactor prompt critical is given by Zf8 .. With circulating fuel, the reactor is prompt critical if & = B, which is less than 23, that is, it takes less excess reactivity to make the circulating- fuel reactor prompt critical than to do the same thing to a stationary-fuel reactor, This is physically evident because the fuel circulation renders some of the delayed neutrons ineffective. That £ is less than 3. can also be verified mathematically. For T intermediate between very small positive values and +wx, we consider again the analogy to the stationary-fuel reactor, Here the inhour formula reads:> T 5 B k, =— —_—, (8) & x T " ig] T+ AT For every positive & there is one, and only one, positive time constant T, and vice versa. This is a consequence of the fact that k__ is a monotonic decreasing function of T, for if this monotony did not exist, there could be several positive T values corresponding to a given value of &__, or vice versa. In the circulating-fuel reactor the situation is qualitatively the same; the fractions under the 5See Glasstone and Edlund, loc. cite, p» 301, Eq. 10.29.1. See also preceding footnote. 170 sum in Eq. (5) are essentially of the form x =~ | + 7% w g (x + ]) + e—(a—l)x x2“ _ e-—ax] (x = “ie]la = 9]/6) 1 and an expression of this form can be shown to be a monotonic decreasing function of positive x;6 the expression is thus a monotonic increasing function of T [see Eq. (6)], and as T increases k__ decreases monotonically, according to Eq. (5); for every positive k__ there is one, and only one, positive T, and vice versa, This, of course, does not preclude that there exists for a given k__ several negative T values, in addition to the one positive value. Negative T correspond, however, to decaying exponentials, which are of no im- portance if the rise in power is observed for a sufficiently long time. Consider now the behavior of Eq. (5) with varia- tion of 6, the transit time of the fuel through the reactor, and of 6., the transit time of the fuel through the whole loop. If 6 (and hence also 6,) is very large compared to all 1/A. and 1/p, Eaq. (5) reduces to Eq. (8), the inhour formula for the stationary-fuel reactor. The circulation is so slow that the reactor behaves as if the fuel were sta- tionary, inasmuch as cll delayed neutrons, even the ones with the long-lived precursors, are given off inside the reacting zone, before much fuel reaches the outside. On the other hand, if 6, and, hence, also 6, is small compared to all 1/A, that is, if the transit time of the fuel through the com- plete loop is small compared to the mean life of even the short-lived delayed-neutron precursors, then for T >> # B 1 + /\I.T g ©) k=4 ox T 6] 5 i=1 This is the same as the inhour formula for the stationary-fuel reactor, except that all the fission yields 3. are decreased by the factor 6/6,. This is physically easy to understand, since 6/0, is just the probability that a given delayed neutron is born inside the reactor, | Of interest is the intermediate case, in which @ bw. K. Ergen, The Behavior of Certain Functions Re- lated to the Inbour Formula of Circulating Fuel Reactors, ORNL CF-54-1-1 (Jan. 15, 1954). PYEETTT OE LM A0 g RO T L owmemesetenen. 9 T ME e S emw oy Ew T T e ey e e T - TEEIETTT PN eEETERE EPSERIPET wrEyrr W T mET o e e - “ ¥ bl e R L - Al kL. o, s vk o e, B A £, b g B S s i Ak wba i is smaller than the mean life of the long-lived delayed-neutron precursors and larger than the mean life of the short-lived precursors. In that case, the long-delayed neutrons act approximately according to Eq. (9) and are reduced by the factor 6/6,. On the other hand, the neutrons with the short-lived precursors behave approximately like Eq. (8) and are not appreciably reduced. Hence, a small excess reactivity enables the reactor to increase its power without ‘‘waiting’’ for the not very abundant long-deiayed neutrons., The reactor goes to fairly short time constants with surprisingly small excess reactivities. However, to make the reactor prompt critical, that is, to enable it to exponentiate without even the little-delayed neu- - trons, takes a substantial excess reactivity be- cause of the almost undiminished amount of the latter neutrons, 171 Appendix J CALIBRATION OF THE SHIM RODS Three essentially independent methods were used to find the value of the shim rods in terms of reactivity. The first method involved an analysis of the counting rate data taken during the critical experiment, and the second was a cali- bration of shim rod No. 3 in terms of the regulating rod., The agreement between these two methods was very good. As a check on the general shape of the reactivity curves as a function of rod position, the rods were also calibrated by using the fission chambers. The various methods are discussed in detail below. CALIBRATION FROM CRITICAL EXPERIMENT ' DATA After about one-half the critical mass of uranium had been added to the system, the counting rates of the two fission chambers and the BF, counter were taken as a function of shim rod position for each subsequent fuel addition until criticality was reached (as described in the main body of this report, chap. 4). Because the fission chamber counting rates were subject to the phenomenon noted in Appendix H, only the BF3 counter data were used in the rod calibration. From the BF3 counting rates for each rod po- sition token after o given fuel injection, the multiplication M was determined from the relation- ship M N NO ’ where N = counting rate for a given fuel concen- tration and shim rod setting, N, = counting rate before start of enrichment, For each value of M the value of the multiplication factor & was determined: ] E=1-—. M Figure 4.9 of Chapter 4 shows k plotted as a function of the uranium in the system for rod positions of 20, 25, 30, and 35 in. The shim rod calibration was obtained from Fig. 4.9 by first making a cross plot of & against rod position at the critical mass, as shown in 172 Fig. J.1. Then, with this plot, a value of (Ak/k)/in. was obtained for every 1-in. movement of the shim rods from no insertion to 16 in. of insertion of the rods. A plot of (Ak/k)/in. as a function of rod position is given in Fig. J.2, and the data are tabulated in Table J.1. In order to find the integrated reactivity, or total worth of the rods, in terms of (Ak/k) as a function of the number of inches of insertion, the curve of Fig. J.2 was divided into three sections. Over each section the curves were fitted to a formula of the form o = Ad?2 + Bd + C , where o = (Ak/k)/in. The formula and its integral, which were applied to the three sections of the curve, are given in Table J.2. For each section, then, the integrated reactivity was found by inte- grating over the a curves. The integrated curves were of the form Ak 3 , 12 i p=‘k———~A’d + Bd + C'd The three sections into which the curve of Fig. J.2 ORNL—LR—DWG 6420 s————INSERTION OF SHIM ROD (in.} i5 10 5 i 0 1.0f | : : 1.00 ‘ = | // 099 // 0.98 / = 097 / 0.96 0.95 / s L/ / 18 20 22 24 26 28 30 32 34 36 SHIM ROD POSITION (in.) 093 Fig. J.1. Shim Rod Position as a Function of k at Criticality. e w b wd o ald - srarer v W AT e Bt »*W Lm_,p e oy v -k - e O ke Al g, i S, Rl i, e, | e R . Bk TABLE J.1. CALIBRATION OF SHIM RODS FROM FUEL ADDITION DURING CRITICAL EXPERIMENT Rod Position Average Insertion Movement of Rods ‘ Interval Taken AV;;Zg:iiod of Shi.m Rods Ave]:clge Ak Shim f"?ods, [(Azk/k)/in.] Av{:'Z}jjk;J;rnfiOd (in.) {in.) Ad (in.) (in.) 36 to 35 35.5 0.5 1.0 0.001 1 0.001 0.00033 35t 34 34.5 1.5 0.9992 0.0015 1 0.00150 0.00050 34 to 33 33.5 2.5 0.9975 0.0020 1 £.00200 0.00067 32 to 33 32.5 3.5 0.9953 0.0025 1 0.00251 0.00083 31 to 32 31.5 4.5 0.9926 0.0028 1 0.00282 0.00094 30 te 31 30.5 5.5 0.9895 0.0032 1 0.00323 0.00108 29 te 30 29.5 6.5 0.9862 0.0035 1 0.00355 0.00118 28 to 29 28.5 7.5 0.9825 0.0038 1 0.00387 0.00129 27 to 28 27.5 8.5 0.9785 0.0041 1 0.00419 0.00140 26 to 27 26.5 9.5 0.9743 0.0045 1 0.00462 0.00154 25 to 26 25.5 10.5 0.9695 0.0050 1 0.00516 0.00172 24 to 25 24.5 11.5 0.9643 0.0055 1 0.00570 0.00190 23 to 24 23.5 12.5 0.9585 0.0060 1 0.00626 0.00209 22 to 23 22.5 13.5 0.9523 0.0064 1 0.00672 0.00224 21 to 22 21.5 14.5 0.9458 0.0049 1 0.00730 0.00243 20 to 21 20.5 15.5 0.9386 0.0075 1 0.00799 0.00266 TABLE J.2. FORMULA (AND INTEGRAND) FITTED TO SHIM ROD SENSITIVITY CURVE Range of Use Formula No. i) Type Formola* 1 0to 8 Differential = ~0.000642 + 0.01874 + 0.0239 2 8 to 16 Differential o = 0.000542 + 0.0055d + 0.059 3 16 to 18 Di fferential o = —0.004542 + 0.1655d — 1.221 4 0to8 Integral p = ~—0.00024° + 0.009354% + 0.0239d 5 8 to 16 Integral p = 0.000167d° + 0.002754° + 0.0594 6 16 to Integral p = —0.00154> + 0.08754% — 1.221d 18 xa = (Ak/E)/in. p = De/k was divided and the correspending formulas which were fitted to the curve are given in Table J.2. In ‘the case of the integrated curves, the Ak/k found by integration in the range of the curve was added to the total Ak/k of the previous curve to give the total Ak/k to the point of integration. The resulting worth, (Ak/E), of the shim rods over the first 18 in. of their movement is shown in Fig. J.3 and the data are tabulated in Table J.3. It should be noted that the last 2 in. of the curve of Fig. J.2 (16 to 18 in.) was extrapolated. The basis for the extrapolation is the curve shown in Fig. D.4 of Appendix D, which is a calibration of an ARE regulating rod made during some pre- liminary experiments done in the Critical Experi- ments Facility., The shim rods were assumed to give the same type of curve, If it were assumed that the last 18 in. of the shim rods gave a shape of Ak/k vs rod insertion which was essentially the image of the first 173 S e e e g W e g - T T 0.32 0.30 0.28 0.26 0.24 0.22 .20 016 044 042 040 REACTIVITY (Ak /&) /in. of SHIM ROD (% /in.) 0.08 0.06 0.04 0.02 0 { 2 3 4 5 6 7 8 9 ORNL-LR—DWG 6424 SHIM ROD NO.3 CALIBRATION AGAINST REGULATING ROD AT LOW POWER (EXP.L-6) 7/ ’ ® CALIBRATION FROM FUEL ADDITION DURING CRITICAL EXPERIMENT {AVERAGE OF ALL RODS) ® CRITICAL EXPERIMENT DATA O LOW-POWER EXPERIMENT DATA 10 i 12 13 14 15 16 i7 18 19 20 INSERTION OF SHIM RODS {in.) Fig. J.2. Differential Shim Rod Sensitivity. 18 in. (see Appendix D, Fig. D.3), then the total worth of a shim rod was 5.8% (Ak/k) and the total worth of all three rods was about 17% (Ak/E). CALIBRATION AGAINST THE REGULATING ROD In Exp. L-6, with the reactor at a power of 1 watt (nominal), shim reds Nos. 1 and 2 were set at predetermined positions. Then shim rod No. 3 was moved over various small increments of its travel and thus calibrated against the requ- lating rod. With the reactor on servo the regulating rod then moved automatically a compensating distance, this distance being roughly 10 in. of its travel. From the previous measurement of the worth of the regulating rod of (Ak/k)/in. of 174 0.033%/in. and the ratio of the movement of shim rod No. 3 and the regulating rod, a calibration of No. 3 shim rod was obtained over the first 14 inches of its travel (starting from the out positicn). Figure J.2 shows the results of this method of calibration (dashed curve), and the data are listed in Table J.4. The agreement between the two methods of calibration was surprisingly good. Since the increments of shim rod movement were larger in this method thdn in the preceding method and since there were other sources of error, such as the error in the movement of the shim rod position indicators on the reactor console, no attempt was made to use this curve (dashed curve, Fig. J.2) to find the integrated value of Ak/k over the red, w ! ST O et b7 T W Fpimsrp RO T P -~ Fop e s e ey e — - T T wgmm—— P pt' e w R e we “ralf e QL1 Formula 3: (Ak/E)/in. = —0.00454% + 0.1655d — 1.221 b|ntegrc[ formulas were of the form Ak/k =A’d° + B4 + C’d Formula 4: Ak/k =~0.00024° + 0.0093542 + 0.0239d Formula 5: Ak/k =0.0001674° + 0.0027542 + 0.059d T TS W T P S Ranges: Formulas 1 and 4, d=0to 8 in. Formulas 2 and 5, d = 8 to 16 in, Formulas 3and 6, 4 =16 to 18 in. 4T otal Ak/k, adding area from each section c ) ke k. A ey Sl la G we AR R MR, e+ Kk i W e e o e e s o R MR L - ek e weiok E e & il Ges L b e ok WG AR 3 . + ¥ » TABLE J.3. CALCULATION OF SHIM ROD REACTIVITY FROM FUEL ADDITION Integral Calibration C Insertion of Differential Calibration O:fer}:frq;;glz Tofclj Shim Red, d . (in.) d? 43 Formula Ad? Bd (Ak/k.)/m. Formula? A’d3 B’d? C’d Ak} “ .é.k.;} (%/in.) — . b 0 % %) (%) 1 ] 1 1 —0.0006 0.0187 0.0420 4 —-0.0002 0.00935 0.0239 £.0331 0.0331 2 4 8 1 ~0.0024 0.0374 0.0589 4 -0.0016 0.0374 0.0478 0.0836 0.083% 3 9 27 1 ~0.0054 0.0561 0.0746 4 —-0.0054 0.0842 0.0717 0.151 0.151 4 16 64 1 —-0.0096 0.0748 0.0891 4 —0.0128 0.1496 0.0956 0.232 0.232 5 25 125 1 ~0.0150 0.0935 0.102 4 —0.0250 0.2238 0.1195 0.328 0.328 6 36 216 1 -0.0216 0.112 0.114 4 -0.0432 0.33466 0.1434 0.437 0.437 7 49 343 1 —0.0294 0.131 0.125 4 —-0.0684 0.4581 0.1673 0.557 0.557 8 64 512 1 -0.0384 0.150 0.135 4 —0.1024 0.5984 0.1912 0.687 0.687 2 +0.032 0.0055 0.135 5 +0.0855 0.1760 0.4720 0 0.687 9 81 729 2 0.0405 0.0495 0.149 5 +0.1217 0.2228 0.5310 0.142 0.829 10 100 1000 2 0.0500 0.0550 0.164 5 0.1670 0.2750 0.5900 0.299 0.986 11 121 1331 2 0.0605 0.0605 0.180 5 0.2223 0.3328 0.6490 0.470 1.158 12 144 1728 2 0.0720 0.0660 0.197 5 0.2886 0.3960 0.7080 0.659 1.346 13 169 2197 2 0.0845 0.0715 0.215 5 0.3669 0.4648 0.7670 0.865 1.552 14 196 2744 2 0.0980 0.0720 0.234 5 0.4583 0.539%90 0.8260 1.117 1.804 15 225 3375 2 0.1125 0.0825 0.254 5 0.5636 0.6187 0.8850 1.334 2.021 16 256 4096 2 0.1280 0.0880 0.275 5 0.6840 0.7040 0.9440 1.599 2.286 3 -1.152 2.648 0.275 6 —6.144 21.184 ~19.536 0 2.286 17 289 4913 3 -1.300 2.8135 0.292 6 -7.3695 23.9147 -20.757 0.3112 2.597 18 324 5832 3 ~1.458 2.979 0.300 6 —8.7480 26.811 -21.978 0.612 2.898 Differential formulas were of the form (Ak/k) in. = Ad? + Bd + ¢ Formula 6: Ak/k = ~0.00154% + 0.082754% - 1.221d Formula 1: (Ak/k)/in. = -0.0006d42 + 0.0187d + 0.0239 cd] = lower limit of formula Formula 2: (Ak/k)/in. = 0.000542 + 0.00554 + 0.059 d, = position integrated o T T T O T e e coel Somiimes L b S TABLE J.4. CALIBRATION OF SHIM ROD NO. 3 AGAINST REGULATING ROD Experiment L-4 Shim Rod Shim Rod Shim Rod No. 3 Regulating Rod Sh::: R;d - 5“:1"; R3°d Ron Mool o Nee2 s o Step Avewgs Movemant St o Stop | Mowmert (Mfiy/in, Aversse e iy s Gm Gmd Gy Gny ey OO R/ neerer (% /in.) (in.) 1A-1B 35.0 36.0 30.0 33.3 3165 3.3 13.0 3.0 10.0 0.100 4.35 1C-1D 35.0 36.0 30.0 18.1 29.05 1.9 6.5 134 6.9 0.120 6.95 3-4 27.0 266 35.5 30.6 33.05 4.9 3.1 13.2 10.1 0.068 2.95 5-6 28.0 274 30.6 28.0 29.3 2.6 3.0 13.1 10.1 0.128 6.7 7-8 29.1 28.9 28.0 26.0 27.0 2.0 3.0 13.2 10.2 0.168 9.0 9-10 30.0 30.0 26.0 24.2 25.1 1.8 2.95 13.0 10.05 0.183 10.9 11-12 32.8 31.9 24.2 22.7 23.45 1.5 2.9 13.1 10.2 0.224 12.55 13~14 35.0 36.0 22,7 214 22.05 1.3 3.2 13.2 10.0 0.254 13.95 wr e e v wmer wemens s e wp v v e rree oyt BIXC RO PETCONET EEE S v ommeTs e ECREUE T OTUFTC . "o N b Sl o vk il ORN[-LR-DWG 6422 3.4 3.2 30 SHIM ROD REACTIVITY (A.A’//() OF SHIM ROD (%} o - - - - - N N N no o © o ™ S o @ o N s o @ o o 0.4 0.2 0 2 4 6 8 10 12 14 16 18 20 TOTAL INSERTION OF SHIM ROD (in.) Fig. J.3. Integral Calibration of Shim Rods as a Function of Position. CALIBRATION BY USING THE FISSION CHAMBERS A third calibration of the shim rods was at- tempted during the critical experiments in which the counting rate of the neutron detectors was taken as a function of shim rod position for two different uranium concentrations. The reactivity was then obtained from the counting rates by using the relationship 1 =1 - —, M and a plot of & as a function of rod position then gave a check on the general shape of the curve. Figure J.4 shows the £ vs rod position for fission chambers 1 and 2. Because the uranium concen- tration in the system was low for both the runs the fission chambers were showing a subcritical ORNL~-LR-DWG 6423 0.7 ! J- P } i U235 N SYSTEM=197 1b ; M> = | I ] I U23d |y SYSTEM =164 1b e 06 T{/i * ; r FISSION ; 23‘5 I CHAMBER #1235 |\ SYSTEM =166 Ib = i ' 2 0.54 : 5 “FISSION CHAMBER NO.2 i | A ‘ I ‘ s | | U235 N SYSTEM=19.7 1b 04 | : | 0/+ ! 03 0 5 10 15 20 25 30 35 40 SHIM ROD POSITION (in.) Fig. J.4. Calibration of Shim Rods from Fission Chamber Data, ORNL-LR-OWG 6424 0.2 Q.15 04 3(ak/kY/in(%/in) 0.05 FISSION CHAMBER NO. 1 DATA | U235 |y SYSTEM: 197 1b 1@t p\' o 5 10 15 20 25 30 35 40 SHIM ROD POSITION {in.) Fig. J.5. Reactivity as o Function of Shim Rod Position, multiplication M greater than the actual value, as discussed in Appendix E. Therefore, the absolute values of &, as calculated from the counting rates, are in error. Nevertheless, the general shape of the curve is experimental verification of the curve shown in Fig. D.3 of Appendix D, Figure J.5 shows the value of (Ak/k)/in. vs rod position, 177 TR T MR v k. b as obtained from one of the curves of Fig. 4.10. but its absolute magnitude is not meaningful, Again, the general shape of this curve is about The data from which Figs. 4.10 and 4.11 were the same as shown in Fig. D.3 of Appendix D, plotted are given in Tables J.5 and J.6. i TABLE J.5. REACTIVITY OF THE SHIM RODS vs ROD POSITION g ;, y23s Rod Fission Chamber No, 1} Fission Chamber No, 2 : BF; Counter E 4 ° i é in System Position Counting Rate Counting Rate Counting Rate i o (1b) (in.) N (counts/sec) NO/N k N (countgs/sec) NO/N £ N (counts/sec) NO/N k - : 16.6 0 17.52 0.5457 0.4543 25.00 0.6512 0.3488 6.00 0.7483 0.2517 g f 5 17.60 0.5375 0.4625 25.52 0.6379 0.3621 6.05 0742 0,258 ; ] 10 18.85 0.5019 0.4981 26.69 0.6100 0.3900 6.51 0.690 0.310 ‘ : 15 20.93 0.4520 0.5480 31.33 0.5196 0.4804 6.40 0.702 0.298 E | 4 20 23.44 0.4036 0.5964 35.63 0.4569 0.5431 6.67 0.673 0.327 25 25.23 0.3750 0.6250 38.91 0.4184 0.5816 7.01 0.640 0,360 E | 30 26.00 0.3638 0.6362 40.93 0.3978 0.6022 E | 30 25.87 0.3657 0.6343 41.12 0.3952 0.6048 6.72 0.668 0.332 35 26.83 0.3526 0.6474 41.76 0.3898 0.6011 6.80 0.660 0.340 % ‘ 35 26.85 0.3523 0.6477 41.65 0.3909 0.6091 6.77 0.663 0.337 i 19.7 0 18.99 0.4982 0.5018 26.83 0.6068 0.3932 6.27 0.716 0.284 5 19.31 0.4899 0.5101 26.88 0.6057 0.3943 6.43 0.698 0.302 E 10 21.07 0.4490 0.5510 29.28 0.5560 0.444 6.56 0.684 0.316 B 15 23.63 0.4003 0.5997 34.43 0.4728 0.5272 6.91 0.650 0.350 &' ' 20 26.21 0.3609 0.63N 38.99 C.4175 0.5825 7.04 0.638 0.362 E 25 28.08 0.3369 0.6631 42.67 0.3815 0.6185 7.28 0.617 0.383 - f 27.5 29.23 0.3236 0.6764 44.59 0.3651 0.6349 7.15 0.628 0,372 i 30 29.55 0.3201 0.6793 45.97 0.3541 0.6459 7.01 0.640 0.360 i 32.5 29.76 0.3179 0.6821 46.19 0.3525 0.6475 7.04 0.638 0.362 . 35 29.76 0.3179 0.6821 46.83 0.3476 0.6524 7.09 ©0.633 0.367 . : 36 29.90 0.3164 0.6836 46.81 0.3478 0.6522 7.16 0.627 0.373 ; 178 e g G i = iy bbbk ke el k. TABLE J.6. REACTIVITY AS A FUNCTION OF SHIM ROD POSITION (FROM FISSION CHAMBER DATA) y235 in System (Ib) Limits of d d_, k_, Ak Ad (in.) 3(Ak/R)/in, 19.7 0 to 2 1 0.502 0.04 2 0.0398 2 to 4 3 0.5065 0.05 2 0.0494 4106 5 0.5125 0.08 2 0.0780 61to8 7 0.523 0.12 2 0.1147 8 to 10 9 0.532 0.16 2 0.1503 10 to 12 " 0.555 0.20 2 0.1801 12 to 14 13 0.574 0.21 2 0.1829 14 o 16 15 0.596 0.22 2 0.1853 16 to 18 17 0.6155 0.18 2 0.1462 18 10 20 19 0.632 0.16 2 0.1266 20 to 22 21 0.6465 0.13 2 0.1005 22 to 24 23 0.6575 0.09 2 0.0684 24 to0 26 25 0.6655 0.07 2 0.0526 26 to 28 27 0.6715 0.05 2 0.0372 28 to 30 29 0.676 0.04 2 0.0296 30 1o 32 31 0.6795 0.03 2 0.0221 32tc M4 33 0.6815 0.01 2 0.0073 34 to 36 35 0.6825 0.01 2 0.0073 179 T b G Appendix K CORRELATION OF REACTOR AND LINE TEMPERATURES An effect which was noted late in the operation of the ARE was that fuel and sodium line tempera- tures indicated in the basement disagreed quite radically with those recorded in the control room. It so happened that all the basement indicators gave line temperatures measured by thermocouples outside the reactor thermal shield, while the temperatures recorded in the control room were measured by thermocouples all located within the thermal shield. When this was first discovered it was thought that helium from the rod-cooling system blowing on the thermocouples inside the thermal shield was making them read low, Turning the rod cooling blowers on and off, however, was demon- strated to have no effect on the thermocouples, and therefore at the end of the experiment no positive explanation for these temperature discrepancies had been found. . This situation has created serious difficulties in trying to analyze the data for this report. There was evidence that the discrepancies were intensi- fied during operation at high power. Nevertheless, when nearly isothermal conditions existed, it was found that the absolute magnitude of the measure- ments made by the thermocouple within the thermal shield were incorrect, but the rates of change obtained from the data were correct, The tempera- tures read on the basement instruments were considered to be correct for several reasons, First, the temperature indications available in the base- ment were much more numerous than those in the control room, and the thermocouple sensing elements for these indicators were all located along lines outside the thermal shield; the temperatures agreed with each other to 110 deg from the average. Equi- fibrium conditions prevailed across the pipes be- cause of the insulation, Also, there were many reasons for the thermocouples to indicate higher than actual temperatures, but none for lower than actual indications. In order to use the temperature data from the experiment it was necessary fo correlate the tem- perature data obtained in the control room with those obtained in the basement, The correlations were then used to correct temperature data ob- tained in the control room. The results of the temperature correlations for the fuel system are shown in Figs. K.1, K.2, and K.3. The agreement between the low-power line 180 temperatures (Fig. K.1) reflects the attention which was given to calibrating these thermocouples in the isothermal condition. The curves in Figs. K.2 ORNL—LR—DWG 6563 1350 REACTOR INLET LINES ——— 1300 é{ A - — REACTOR OUTLET LINES FUEL LINE TEMPERATURES INSIDE THERMAL SHIELD (°F) 1250 1 /4 O OUTLET TEMPERATURé ® NLET TEMPERATURE 1200 fl 1200 1250 1300 1350 FUEL LINE TEMPERATURES QUTSIDE TRERMAL SHIELD (°F) Fig. K.1. Correlation Between Fuel Line Tem- peratures Measured Inside and Outside the Reactor Thermal Shield During Subcritical and Low-Power Experiments, ORNL—-LR~DWG 6564 1500 . /./ |, o 1450 _ L y L d 5o REACTOR OUTLET LINE —u| ~* % o 1400 t 5= | / w ™ / b S 1300} . 1 52 < ; L2 / REACTOR INLET LINE ? - {250 7 - - - ! 1200 1200 {300 1400 1500 1600 FUEL LINE TEMPERATURES OUTSIDE THERMAL SHIELD {°F) Fig. K.2. Correlation Between Fuel Line Tem- peratures Measured Inside and Outside the Reactor Thermal Shield During High-Power Experiments. e e ger ot e 1 R " e oo e e e e s—— wererry 7-m~”r * e S € - grEA S B R e ey Py err W per T REowerw Ce oy d & k . [ . Bk LA, G B g e e, ke SR e, ookl Bo ko, o Wk i, G R e, B, ol S, @i ORNL-LLR-DWG 6565 ./ 250 & 7 150 / 100 /+/ v 300 50 /’ 0 50 10C 150 200 250 300 350 400 FUEL LINE A7 OQUTSIDE THE THERMAL SHIELD {°F) FUEL LINE A7 INSIDE THE THERMAL SHIELD (°F) Fig. K.3. Correlation Between Fuel Temperature Differentials Measured Inside and Outside the Thermal Shield. ORNL-LR-DWG 6566 1350 REACTOR INLET LINE 1300 b ZREACTOR CUTLET LINE 1250 / / SODIUM LINE TEMPERATURES INSIDE THERMAL SHIELD (°F) 1200 1250 1300 1350 1400 SODIUM LINE TEMPERATURES QUTSIDE THERMAL SHIELD (°F) Fig. K.4. Correlation Between Sodium Line Temperatures Measured Inside and Outside the Thermal Shield During Low-Power Experiments. and K.3, both of which were obtained from data taken during operation at high (>200 kw) power, show the anomaly in question, The analogous data for the sodium system are shown in Figs. K.4, K.5, and K.6, It is evident from Fig. K.4 that an isothermal condition was never attained in the sodium system, since both o 0 o ORNL-LR-DWG 6567 o 1350 : | | N 2 REACTOR OUTLET LINE g 9 Léj / /-’ 4300 o o / o = / $ * g / — O I 41250 # w o.. = w = REACTOR INLET LINE = - | | s 1200 g 1200 1250 1300 1350 1400 R SODIUM LINE TEMPERATURES OUTSIDE THERMAL SHIELD (°F) Fig. K.5. Correlation Between Sodium Line Temperatures Measured Inside and Ovutside the Thermal Shield During High-Power Experiments, ORNL-LR—-DWG 6568 150 {100 50 e ® SODIUM LINE A7 INSIDE THE THERMAL SHIELD(®F) 0 0 50 100 {50 SODIUM LINE A7 QUTSIDE THE THERMAL SHIELD (°F) Fig. X.6. Correlation Between Sodium Temper- ature Differentials Measured Inside and Outside the Thermal Shield. the inlet and the outlet line temperature correla- tions have slopes of 1 but intercepts of ~25 and —~40°F, respectively. This was probably the re- sult of the long sodium inlet and outlet lines being used as a convenient means of adding heat to the system to maintain a thermal equilibrium in the reactor of ~1300°F. Good data for the high-power 181 T P Y TR O T R e T TR e ¢ f b 3 T correlations, Figs. K.5 and K.6, were rather scarce, but the curves shown are believed to be reasonable approximations to the actual correlation, The data in Fig. K.6 have been corrected for known error in the temperature differential at zero (or low) power. The correction, as applied, was to reduce the value of the temperature differential determined from the thermocouples outside the thermal shield by 10°F, All the correlations were based upon data which were obtained during runs long enough for the establishment of equilibrium conditions in the system or at times when isothermal conditions prevailed. The most important conclusion that can ‘be made from these data is that the temperature differentials across the reactor in both the fuel and sodium systems were about a factor of 2 {ow. In the fuel system the outlet line temperature changes read in the control room (inside pressure shell temperatures) were only one-half as great as the corresponding temperature changes read in the basement (outside pressure shell temperatures)., In the sodium system both outlet and inlet line tem- perature curves had different intercepts but the same slopes (as during the low-power runs). All temperature data used in this report that pertained to operation at high power were corrected to agree with line temperatures obtained outside the thermal shield (basement readings) by using curves K.l through K.6, Data from runs made under isothermal or equilibrium conditions where equal temperature differences were obtained on instruments either in the control room or in the basement needed no temperature corrections. The temperature correlation data from which the curves were obtained are presented in Tables K.1 and K.2. Because of the lack of coordination between the control room and basement operations during the experiment, much potentially useful data had to be discarded because the exact times of the readings were not known or the data were taken before equilibrium conditions were established, TABLE K.1. FUEL LINE TEMPERATURES MEASURED BY THERMOCOUPLES INSIDE AND ~OUTSIDE THE REACTOR THERMAL SHIELD Fuel Inlet Line Experiment Temperatures (°F) Date Time No. (Line 120) Fuel Outlet Line Temperatures (°F) (Line 111) Temperafure D ifferential (o Fy© Basement? Control Room b Basement® Control Room? Basement Control Room 10/26 2115 Before 1290 1285 operation 10/27 0630 Before 1293 1293 operation 11/4 1100 L-1 1300 1299 11/6 0340 L-4 1305 1310 1111 0300 H-11 1207 1212 1215 H-11 1210 1214 11/12 0001 H-13 1316 1308 0630 H-13 1315 1307 1003 H-14 12464 1243 11/13 0915 After 1264 1260 operation 1290 1287 0 2 1296 1293 3 0 1297 1295 -3 -4 1304 1300 ~1 -10 1522 1418 315 206 1524 1418 314 204 1323 1308 7 0 1320 1306 5 =1 1587 1475 341 232 1260 1250 -4 ~10 2Fuel line temperatures outside thermal shield. bFuei line temperatures inside thermal shield. € Thermocouple differences for no-power runs indicate extent of thermocouple errors (highest is £10 deg) and therefore the accuracy of the readings. dEsfimcfed. 182 L : IR F “ETm e S ey Ny O B TETT 0 b tep T wt e emerrr gy *ymm'w o W ¥ e TP v ey s’ ¥ gl e Ve e e i b s MGG R e uma i, Sk ok lelee s il ca e e S 4 o W ot Fdtle TABLE K.2, SODIUM LINE TEMPERATURES MEASURED BY THERMOCOUPLES INSIDE AND OUTSIDE THE REACTOR THERMAL SHIELD Sodium Inlet Line Sodium Outlet Line Temperature Date Time Experiment Temperatures (°F) Temperatures (°F) Differential (°F) No. Basement Control Room Basement Control Room Basement Control Room 10/26 2115 Before 1305 1278 1320 1278 15 0 operation 10/27 0630 Before 1312 1292 1325 1290 13 ' 2 operation 11/4 1100 L-1 1320 1295 1330 1288 10 7 11/6 0400 L-4 1325 1300 1335 1295 10 5 11/11 0900 H-11 1223 1224 1332 1271 109 47 17/12 1003 H-14(1) 1246 1258 1369 | 1307 123 49 11/9 1907 H-3(7) 1313 1290 1373 1313 60 23 11713 0915 After 1282 1258 1292 1252 10 -6 operation This points out the need in future operations of the scale of the ARE for some sort of timing device that would automatically stamp the date and time every few minutes on the charts of all recording instruments, and a more systematic method of manually recording data for each experiment per- formed in which conditions are held constant long enough for all pertinent data to be recorded. These and other similar recommendations will be found in Chapter 7 of this report. 183 Appendix L POWER DETERMINATION FROM HEAT EXTRACTION The power level of the reactor was determined from the total heat extracted by the fuel and the sodium. Data for the heat extraction determination were obtained from continuous records of the re- actor fuel inlet, outlet, and mean temperatures, the temperature differential across the reactor, and the flow rates of the fuel and the sodium. The flow rates could also be calculated from the pump speed, and there was independent experi- mental evidence from which a plot of speed vs flow rate was obtained, The power level of the reactor was calculated by use of heat capacities, flow rates, and temperature differentials for both sodium and fuel, it was found that the fuel and the sodium accounted for 99% of the extraction of power generated in the reactor. The data referred to above were all obtainable in the control room. The controls for preheating and maintaining heat on the system were located in the basement, and along with these controls were temperature re- corders and indicators for the whole system, ex- clusive of the reactor. The temperatures of both fuel and sodium lines to and from the reactor were also recorded in the basement. These basement data were used for a separate power extraction determination, An independent source of power level information was also available, since the fuel and sodium were cooled by a helium stream flowing over a heat ex- changer and the helium in turn was cooled by a helium-to-water heat exchanger., No flow or tem- perature measurements were made of the helium, but since the water exchangers were very close to the fuel or the sodium exchangers, all the heat that was taken out of the fuel and sodium should have appeared in the water, The water flow through these exchangers was metered by orifice-type flowmeters, and the outlet temperatures were meas- The fuel loop exchangers and the two sodium loop exchangers were metered separately. Up to the time of the 25-hr xenon run the reactor was operated at high power for very short periods of time, and the extracted power was determined from the control room data. During the 25-hr xenon run a comparison was made of the extracted power determined from the control room datq, that de- ured by thermocouples in wells. termined from the basement data, and that from the water data (Table L.1). Large discrepancies 184 were found, and the indications were that the control room data were low. The disagreement of the various data led to the examination of the inlet and outlet line tempera- tures, as described in Appendix K, During the xenon run the fuel outlet line temperatures were about 100°F higher than the fuel outlet manifold temperatures (control room data), and the fuel inlet line temperatures (basement data) were essentially the same as the fuel inlet manifold temperatures (control room data). The sodium outlet line temperature {(basement data) was 60°F higher than the sodium outlet temperature at the reactor (control room data), (The only check on power extraction by the rod cooling system, which was only 1% of total power, was the water dota for the rod cooling helium-fo-water heat exchanger. In most discussions this 1% is neglected.) The power extracted in both the fuel and the sodium systems has been calculated by using temperature data from several experiments, and the results are presented in Table L.2. There were three separate measurements of the fuel temperature differential available in the control room which were usually in fair agreement. The largest dis- crepancy was, as mentioned, between the control room data and the basement data. The temperature differential obtained from the basement data was an average result from a number of line thermo- couples, while the control room data came from thermocouples located on the fuel headers inside the reactor thermal shield. The so-called secondary heat balance was ob- tained by determining the heat dumped by the water from the various heat exchangers. These data are tabulated in Table L.3 for the high-power experi- ments. The various estimates of reactor power from both primary and secondary heat extraction are then listed in Table L.4, together with the power estimated from the calibration of the nuclear in- struments, that is, the log N recorder and the micro- microammeter, These latter instruments were normalized to agree with the primary extracted power as determined by the line temperature (basement data) during the xenon experiment (H-11). Equilibrium conditions were certainly attained during this 25-hr experiment and such error as there may be in using these data as the criteria for n £ » & > B . o T NPT W TR T pM g Wear s YT P > meEw o oo gy o * x TR T DET (R ¢ BT e T P L MY T g S e I T Wk R Gk m;‘\ TABLE L.1. COMPARISON OF POWER EXTRACTION DETERMINATIONS MADE FROM DATA OBTAINED DURING 25-hr XENON RUN . sl da- vy, Ll il w Temperatures (°F) Flow Power Extracted Total Power? Extracted L Inlet Qutlet AT (gpm) (Mw} (Mw) Fue! System ‘ Control room data 1212 1418 206 44 1.02 | Basement data 1209 1522 313 44 1.52 i Sodium System E Control room data 1225 1271 46 153 0.244 ‘ Basement data 1226 1335 109 153 0.577 Total Power Control room data 1.28 Basement data 2,12 Water Data In fuel loop 61 117 56 205 1.68 : In sodium loop 61 114° 53 38.3% 0.588° ‘ Total Power from 2.28 Water Data S okl il nl R AR AL ¢ s %|ncludes ~ 1% for rod cooling. bAveroge for both loops. €Sum of both loops. TABLE L.2, PRIMARY POWER EXTRACTION B BB e i Contre] Room Data Basement Data Average Experiment Run Fuel System Sodium System TT[:‘::UZ‘E‘WGT Fuel System 2;:;:: T?’:::UZ?:;r k N o AT,Outlet AT, AT, Tube o b . Rod-Cooling AT b AT P Rdfifximg o Minsinler Recorder Aveiese (b iy (o o) CF) 0N e o ok em o e ] H.3 14l 40 33.5 0.196 0.163 15 0 0.199 ¥ 2 44l 66 52 0323 0.254 0 0.308 1 3 44, 15 116 0.562 0.567 0 0.584 ; 4 4y 120 14 124 0588 0.560 0.606 153 14 00734 0678 198 0970 45 0.236 123 : 5 4l 2% 231 29 1128 1129 147 153 105 00548 1.21 : 6 4l 2y 234 23 106 1045 119 12 10 00523 124 138 0677 60 0315 1012 7wy 13 19 20 0.0653 0.0946 00975 152 22 0115 0221 H-6 a4l 26 240 245 1158 1075 LIsg 152 3.5 00182 1.2 320 1.572 H8 45 266 225 131 L 153 21 0110 134 293 1452 75 03%4 187 E K11 44 206 213 21 0984 1017 1018 153 464 0244 127 313 1520 10 0.577 2,12 b H-13 44 7 0 0.0484 153 1.4 00073 00757 312 1.530 19 0099 1.45 1 H-14 45 182 245 0.900 1.212 153 50 0262 145 355 1760 127 0.667 245 ' L | { 185 TABLE L.3. SECONDARY POWER EXTRACTION E No. 1 Sedium No. 2 Sodium Rod-Ceoling U?I Heat Exchanger Heat Exchanger Heot Exchanger Heat Exchanger Total Experiment Run Extracted Power No., No, Water AT Pioerr Water AT Py Water AT Pyo Water AT Pr.c. Secondary System : Flow (°F) ower Flow (°F) Power Flow ©F) Power Flow (°F) Power (Mw) (gpm) (Mw) (gpm) (Mw) (gpm) (Mw) (gpm) (Mw) H-3' i 204 13 0.389 38.4 0 0 38.4 0 0 17.3 3 0.0076 0.397 2 204 13 0.389 38.4 0 0 38.4 0 0 17.3 3 0.0076 0.397 3 204 30 0.897 38.4 0 0 38.4 0 0 17.3 3 0,0076 0.905 4 204 29 0.866 38.4 26 0.146 38.4 21 0.1188 173 3 0.0076 1.14 5 204 59 1.765 38.4 26 0.146 384 25 0.141 17.3 3 0.0076 2.06 6 204 59 1.765 38.4 27 0.152 38.4 24 0.135 17.3 3 0.0076 2.06 7 204 3 0.0903 38.4 26 0.146 38.4 25 0.141 17.3 2 0.0051 0.382 H-6 204 H-8 206 63 1.901 38.0 37 0.207 38.4 40 0.226 17.3 3 0.0076 234 H-11 205 56 1.685 38.2 55 0.300 38.4 51 0.228 17.6 2 0.0052 2,28 H-13 205 3 0.0902 38 10 0.0561 38.4 1 0.0056 17.1 H-14 204 63 1.883 38 57 0.318 38.4 56 0.316 16.9 6 0.0150 2.53 TABLE L.4. REACTOR POWER SUMMARY Primary Power (Mw) Secondary Nuclear Power Power (Mw); Experiment Run (Mw) Control Room Data Basement Water No. No. Data, Heat Exchanger, Log N Micromicroammeter PAT Pp Pav AT L Pp Py H-3 1 0.449 0.425 0.216 0,183 0.199 0.3966 2 0.449 0.463 0.343 0.274 0.308 0.3966 3 0.848 0.910 0.582 0.587 0.584 ¢.9046 4 0.923 0.988 0.681 0,653 0.699 0.678 1.23 1.1384 5 2.47 2.05 1.20 1.20 1.24 1.21 2.0596 6 2.12 2.34 1.23 1.22 1.26 1.24 1.012 2,0596 7 0.281 0.288 0.200 0,230 0.232 0.221 0.3824 H-6 2.21 2,16 1,20 1,21 1.25 1.22 1.93 H-8 2.12 2.38 1.44 1.24 1.34 1.87 2.342 H-11 2.12 2.12 1.25 1.28 1.28 1.27 2.12 2.278 H-13 0.125 0.151 0.0757 0.0757 0.1562 H-14 2.41 2,53 1.42 1.45 1.43 2.45 2.53 amount of extracted power is conservative, since the water heat balance at the same time showed the power to be 7% higher. The sources of error in the various measurements of extracted power were associated with the tem- 186 perature measurements on the fuel and sodium lines at the reactor, which gave the reactor AT, These thermocouples were unique in several respects; that is, they were located within the reactor thermal shield; they were exposed to the reactor pressure T b — T g VYT PR g BT e MRS e e b TR IR W T ey gt Sy TR T WU T W e e I (B N MW T i e THRRRT NMOE g s Bk e, i el sdin LR wadl. o, i S shell; they were exposed to high nuclear radiation fluxes, etc. These unique aspects immediately suggest several possible explanations for the erroneous temperature measurements. However, upon further examination, each of these aspects, with the dubious exception of nuclear radiation, has been shown to be incapable of producing the observed anomaly. The control and shim rods and the fission chambers were cooled by forced helium circulation in the rod-cooling system, Part of the helium that was blown down through the rod tubes was de- flected back up across the outlet manifold and between the pressure shell and the thermal shield. While it was thought that perhaps this was cooling the fuel outlet manifold thermocouples and making them read low, this was disproved when changing the speed of the blower or stopping it had no effect on the fuel outlet temperature. It was thought that possibly the heat radiation from the fuel outlet manifold to the colder bottom of the pressure shell was great enough to actually lower the manifold wall temperatures 100°F below the fluid temperature. This has been disproved by heat transfer calculations. 1t was also thought that since the outlet fuel lines passed through the 2-in, plenum chamber that the resulting surface cooling of the fuel might account for the lower wall temperatures, although the mixed mean fuel temperature was considerably higher. This was also disproved when no increase in wall tempera- ture was observed for thermocouples within the thermal shield but located progressively farther away from the pressure shell bottom. It is of interest that after the final shutdown of the reactor, the fuel outlet line and fuel outlet manifold temperatures agreed; in other words, with no power generation the basement data and control room data agreed. 187 T —————r ~ Appendix M | THERMODYNAMIC ANALYSES Some two years before the Aircraft Reactor Ex- periment was placed in operation a comprehensive report was written on the ‘‘Thermodynamic and Heat Transfer Analysis of the Aircraft Reactor Experiment.”'! Not only was the design of the reactor system based, in part, on the studies culmi- nating in that report, but the material therein served as a guide during the acceptance fest of the experiment, Therefore a comparison of the calculated and the actual thermodynamic per- formance of the reactor system was attempted, [t soon became apparent that the experimental data of the type needed for thermodynamic analyses were inadequate to permit a meaningful comparison with any of the calculated situations. Perhaps the most valid comparison may be made between the calculated insulation losses, the heater power input at equilibrium, and the heat removed by the space coolers. Even here the agreement is less than 50%. An illustration of the inefficacy of attempting to calculate such thermodynamic con- stants as the heat transfer coefficients of the fuel and the sodium is presented below. INSULATION LOSSES, HEATER POWER INPUT, AND SPACE COOLER PERFORMANCE At equilibrium the electrical heat required to maintain the system at o mean temperature of 1325°F should have been equivalent to the heat removed in the eight pit space coolers. The maxi- mum amount of heat ever removed by the space coolers was the 250 kw attained with 66.5-gpm-total water flow with a temperature rise of 30°F. Al- though this value agrees very well with calculated! heat loss of 220 kw for the entire system (the reported value of 240 kw was calculated for an earlier design system and was reduced by about 20 kw for the actual system), the actual electrical power input to the heaters averaged around 375 kw after the system reached equilibrium. There are several possible explanations for the discrepancies betwaen the calculated heat loss, the electrical power input, and the heat removed by the space coolers. First, the calculated loss was low because it assumed idealized conditions 1B, Lubarsky and B. L. Greenstreet, Thermodynamic and Heat Transfer Analysis of the Aircraft Reactor Ex- periment, ORNL-1535 (Aug. 10, 1953), 188 and not the octual insulation which had cracks, clips, etc. Second, the electrical heat load was high, since it did not allow for transformer, variac, and line losses. Also, the space cooler heat load may not have represented an equilibrium condition if the pit walls were still heating up and/or radi- ating heat, EXPERIMENTAL VALUES OF HEAT TRANSFER COEFFICIENTS? An attempt was made tocalculate both the sodium and the fuel heat transfer coefficients from the measured values of temperatures and flows in the various heat transfer media, that is, fuel, sodium, helium, water, These data are given in Table M.} for a typical operating condition. The desired heat transfer coefficient should be determinable from the calculated values of the various thermal resistances in the heat exchangers. To obtain the thermal resistance on the fuel side or the sodium side of the respective heat ex- changers, the helium side and metal resistances had to be subtracted from the over-all resistance. The expression for the over-all resistance is 1 q . — = ! (UgA) A2 AT ’ where U = over-all heat transfer coefficient, A area across which heat is transferred, q rate of heat transfer, AT = over-all temperature difference, H I A knowledge of the inlet and outlet temperatures on the helium side of the heat exchanger is re- quired by the above expression, and lack of this information immediately introduces a large un- certainty in the results, Of greater importance, however, is the relative magnitude of the resist- ances involved, The calculated ratio for the fuel- to-helium exchanger is RHe — =103, / where RHe is the resistance on the helium side 2H. H. Hoffman, Physical Properties Group, Reactor Experimental Engineering Division. Yt e T oY T "—r flwr e ¥ o g e b e e ¥ o T ey ¥ o e Tl Ty w ¥ ey S e B G i e, G B, b i TABLE M.1. EXPERIMENTAL HEAT TRANSFER DATA IN FUEL AND SODIUM LOOPS Primary Coolant Helium Water Temperature Temperature Temperature * Flow (°F) Flow (°F) Flow (°F) (gpm) (cfm) —_— {gpm} —/—— Inlet Qutlet Inlet Outlet Inlet Qutlet Fuel loop 44 (¥5%) 1209210 1522 £10 8000 (£30%) No data 194 (+5%) 61+t2 11712 Sodium loop 152 (£1%) 1225 %10 1335110 1700 (£20%) No data 766 (£1%) 6112 11412 *Estimated. and R, is the resistance on the fue! side, and for the sodium-to-helium exchanger is where R_ is the wall resistance, Hence, the liquid-side thermal resistance is the small dif- ference between two much larger numbers, A small error in R.. or R, is greatly magnified in R, Unfortunately, the error in R, is not small, being perhaps as great as 60%., Thus, except for over-all heat balances, little useful heat transfer informa- tion can be extracted from the available data on the ARE heat exchangers. 189 TR e - e T o T e PR WL TR T W e G o o g Bl . Appendix N . COMPARISON OF REACTOR POWER DETERMINATIONS W, K. Ergen In order to get an estimate of the reactor power and to be able to calibrate the instruments, the ARE was run for 1 hr ot a power which was esti- mated, at the time, to be 10 watts (exp. L-4). A sample of the fuel was then withdrawn and the gamma activity was compared with that of a sample which had been irradiated at a known power level in the Bulk Shielding Reactor (BSR). This method of power determination is discussed in Appendix H. A curve showing the gamma counting rate of a BSR-irradiated sample as a function of time after shutdown had been obtained. The irradiation time was | hr at a constant power of 1 w, From this curve a decay curve corresponding to the actual power history of the ARE was synthesized by taking into account not only the *‘10-watt'’ run, but also the previous lower power operation. When this synthetic curve was compared with the one obtained from measurements on the ARE sam- ple, the shapes of the curves did not agree, the ARE curve having a smaller slope than the syn- thetic curve, Also, the power determined by this method was lower than the power ultimately de- termined by-the heat balance. Both these effects can be explained in a qualitative manner by the loss of some of the radioactive fission fragments from the ARE sample. This would reduce the total radioactivity of the sample and hence the apparent power level. Furthermore, the loss of radioactive fission fragments would reduce the counting rate at short times after irradiation more than at long times, because among the most volatile fission fragments are the strong gamma emitters that have relatively short half lives (notably, 2.77-hr Kr38), This would flatten the slope of the decay curve of the ARE sample. | it has not been possible so far to treat this matter in a quantitative way, but in order to elimi- nate some computational complications a sample was irradiated in the BSR under conditions exactly duplicating the power history of the ARE, except for a proportionality factor. A comparison of the decay curve of this sample and that of the ARE sample is shown in Fig. N.1. The BSR sample contained 0.1166 g of U235: the fission cross section at the temperature of the BSR is 509 barns, ! The BSR power at the final T-hr irradiation was 190 10 w, corresponding to a flux? of 1.7 x 108 n/cm?:sec, and a self-shielding factor, estimated to be 0.8, had to be applied to this sample. Hence, during the 1-hr irradiation there were 0.1166 g x 0.6 x 1024 atoms/g-atom x 509 x 10~ 24 fissions/(atoms:n/cm?) x 1.7 x 108 n/cm?-sec x 0.8/235 g/g-atom n 20.6 x 10° fissions/sec The ARE sample contained 0.1177 g of U233 and the total U?33 in the ARE at the time of the run was 59.1 kg. The reactor power, as determined fater by the heat balance, was actually 27 w, instead of the expected 10 w, Hence there were 27 w x 3.1 x 100 fissions/wesec x 0.1177 g 59,100 g = 1.66 x 10¢ fissions/sec in the ARE sample. The ratio of the radioactivities of the BSR sample and the ARE sample should thus have been 20.6/1.66 = 12.4. As may be seen from Fig. N.1, the measured ratio is 27 at 1 hr 40 min, and 17.5 at 38]/2 hr. As pointed out above, the discrepancy can be explained by the loss of radioactive fission fragments from the ARE sample. A small amount of the radioactivity of either sample was contributed by the capsule and the fuel carrier, especially the sodium content of the carrier. This was measured by irradiating a non- uranium-bearing capsule with carrier, and counting its radioactivity. However, since this correction proved to be small and since both samples con- tained about the same amount of uranium in the sagme amount of carrier, the results would not be appreciably affected. 13, L. Meem, L. B. Holland, and G. M, McCommon, Determination of the Power of the Bulk Shielding Re- actor, Part Ill. Measurement of the Energy Released per Fission, ORNL-1537 (Feb. 15, 1954). 2E, B, Johnson, private communication. P v £ ot et e R megpry o+ P ‘W" e g g b e T e e ?_!m W "o r R w TP s CmerTi P g g pEET T NTEME R TmeEe OO rgmer 161 ACTIVITY OF SAMPLE (arbitrary units ) - . (R R k. e L Ea Fe i guuiok i e R S, AL - T, Ve . Siafiiiin ik AR L A wke L kel .ok RAGR Sein Mol aeo e GG e ™ ORNL—-LR—DWG 6569 SAMPLE SAMPLE 10 15 20 25 30 35 40 45 TIME AFTER SHUTDOWN (hr) Fig. N.1. Decay of Irradiated Fuel Samples. T WIS e T My S ST e g e e e - e R Cw T o : T Ot i Kb i s i o b ase e Aale i3 Appendix O ANALYSIS OF TEMPERATURE COEFFICIENT MEASUREMENTS 'IMPORT‘ANCE OF THE FUEL TEMPERATURE - COEFFICIENT One of the most desirable features of a circu- lating-fuel reactor is its inherent stability because of the strong negative temperature coefficient of reactivity. This was conclusively demonstrated by the ARE. It had been predicted, on the basis of the temperature dependence of the density, which was p (g/cm®) = 3.98 — 0.00093T (°F) for the final fuel concentration in the reactor, that the ARE fuel would have a negative temperature coefficient of sizable magnitude. At the operating temperature of 1300°F the fuel density was 3.33 g/em®. The resulting mass reactivity coefficient was ' o ~0.00052 (AM/M)/°F = — ———= ~1.56 x 1074/°F , 3.33 and, for Ak/k = 0.236 AM/M, the predicted temper- ature coefficient that would result from the changing density of the fuel alone was (Ak/R)/°F = ~3.68 x 10~5/°F . Actually, as was stated in the body of this report (ef., Fig. 6.4), the fuel temperature coefficient was ~9.8 x 10-%/°F, As long as the over-all temperature coefficient of a reactor is negative, the reactor will be a slave to the load demand. However, the fuel temperature coefficient was the important factor for reactor control in the ARE because it took a significant time for the bulk of the material of the reactor to change temperature and, therefore, for the over-all coefficient to be felt. In experiment H-5 (cf., Fig. 6.4), it was found that the fuel temperature coef- ficient predominated for 6 min. EFFECT OF GEOMETRY IN THE ARE When the reactor was allowed to heat up by nuclear power, as in experiment H-5, the measure- ments of the temperature coefficient were quite definitive. The experiments in which the heat extraction by the helium blower was suddenly increased, i.e., experiments E-2, L-8, and H-4, did not give clear-cut results. In fact, it was noticed throughout the operation that whenever the fuel 192 loop helium blower speed was increased there was always a sudden marked increase in reactivity. In attempting to determine this instantaneous temper- ature coefficient by observing the rod movement by the servo as a function of the fuel temperature, values of the fuel temperature coefficient of a magnitude much larger than ~9.8 x 10™% could be obtained. The change in the apparent temperature coef- ficient as a function of time during experiment H-4 is shown in Fig. O.1. The time intervals chosen were of 15-sec duration. The apparent peak tem- perature coefficient of reactivity was in excess of ~3.5 x 1074 (Ak/k)/°F, and after 5 min the coef- ficient leveled out to a value of about ~6 x 1073, A more complete discussion of this experiment is given below in the section on ‘‘High-Power Meas- urements of Temperature Coefficients of Reac- tivity,"" in which the time lag of the thermocouples is taken into account. Two coefficients were ob- tained: an initial fuel temperature coefficient and an over-all coefficient, which was the asymptotic value (=6 x 10~3) given by Fig. 0.1, The effect shown in Fig. O.1 was due partly to geometrical considerations. By reference to Figs. 2.2 and 2.3, it can be seen that as the fuel entered the reactor it passed through the tubes closest to R - -] -40 ORNL-LR-DWG 6570 {(ak/mr/ar]x o™ i | ! | | ! o O 5 & 8 9 1 o 2422 2123 2124 2125 ' 2126 “TIME OF DAY, NOV. 9, 1954 Fig. O0.1. Variation in Apparent Temperature Coefficient of Reactivity with Time, Experiment H"4o o T ® e REAE WERT R mpsEo v wr PR N W g SRR PN B Wr TG A e TR W CYEC R OWEEEUT WSS EE D pmovmms T MOSEWER- cwm U kT - ke k- e o . R e, R, ik, e oL - i e [ the center where the reactivity effect was greatest. With o sudden increase in heat extraction, a slug of cooled fuel initially entered the center of the core and caused a large reactivity change. Since the mean reactor temperature was the average temperature of all the fuel tubes, a comparison of the initial rate of rod insertion by the servo with the initial rate of change of the mean reactor temperature could give an apparent temperature coefficient much larger than its actual value. For example, in experiment L.-8 {cf., Figs. 5.13 and 5.14) a comparison of the slopes of the red position curve and the mean temperature curve near the beginning of the run was made. If it is assumed that the temperature response of the thermocouples lagged behind the response of the regulating rod movement, it is interesting to compare the points of steepest slope for the two curves. By using the slope from Fig. 5.13 at time 0218:30, and the slope from the mean temperature curve (Fig. 5.14) at 0220, a temperature coefficient of —1.6 x 10~ * (Ak/k)/°F is obtained, as opposed to the actudl fuel temperature coefficient of ~9.8 x 1073, A detailed discussion of this experiment is given in the section on ‘‘Low-Power Measurements of Temperature Coefficients of Reactivity.” This effect was characteristic of the geometry of the ARE but not necessarily of circulating-fuel re- actors in general. TIME LAG CONSIDERATIONS As mentioned above and in the “*Reactor Kinetics”’ section of Chapter 6, one of the most consistently noted phenomena of the ARE operation was the time lag in reactor témperature response during every phase of the experiment. These time lag effects can be partly explained by the geometry effect just described; and, in this sense, the time fag is not a true time lag but only an apparent lag due to the differences in location and response of the thermocouples and the neutron detectors (and, hence, regulating rod movements ). Other possible causes for the time lags were the fuel transit time around the system, the heat transfer phenomenon within the reactor, the design and location of thermocouples, and a mass-temperature inertia effect. These effects were all discussed briefly in the section on ‘‘Reactor Kinetics,”” Chapter 6. MThe regulating rod was controlled by o flux servo mechanism which received its error signal from o neutron detector (cf., App. C). Whether or not the time lag was all or partly real is academic. The fact that it did give an observed effect during many phases of the ARE operation made it mandatory to take the lag into account in interpreting much of the data. The manner in which this was actually done was to assume that the. response of the thermocouples lagged behind the nuclear response of the reactor (by as much as 2% min for low-power operation and by about 1 min during the high-power regime) in those experiments in which the equilibrium between the reactor and its load was upset (i.e., rapid fuel cooling rates). The thermocouple readings were then ‘“moved up"’ by that amount, and the new readings were com- pared with the appropriate nuclear instrument observation. For those experiments in which equi- {ibrium prevailed, but in which cooling was taking place, it was only necessary to correct for temper- ature readings (cf., app. K). The temperature coefficient measurements were probably more affected by the temperature-time lags than any other single type of measurement made on the ARE, mainly because of the short duration of the experiments and their great de- pendence on time correlations (for example, corre- lations between regulating rod motion and mean temperature changes). The results of temperature coefficient measurements which contained time lag corrections were not included in the main body of the report because such corrections needed to be discussed in detail inappropriate to the context of the report. These experiments are described in the following sections of this appendix. SUBCRITICAL MEASUREMENT OF TEMPERATURE COEFFICIENT The suberitical measurement of the temperature coefficient {exp. E-2) was described in Chapter 4. Briefly, the procedure followed was to cool the fuel by raising the heat barriers on the fuel heat exchangers, turning on the fuel helium blowers, and then observing the increase in multiplication with the two fission chambers and the BF; counter. The BF, counting rate was so low that the sta- tistics were poor; therefore, the fission chamber data had to be used. The subcritical multiplication of the fission chambers was then subject to the phenomenon discussed in Appendix E. In this experiment the cooling was rapid and equilibrium conditions v_vé;e not attained. | Consequently, in order to find a value of the temperature coefficient from these measurements, it was necessary to 193 TR TRETTTSE T T W ¥ o (170 T apply three corrections: a correction for the fission chamber multiplication error, a time lag correction, and a temperature correction of control room ob- served temperatures {app. K). Throughout the subcritical experiments ample data were taken simultaneously on the BF counter and the two fission chambers for a correlation plot between the counting rates of the BF, counter and the fission chamber to be easily obtained, as shown in Fig. 0.2. A plot of the raw data obtained from the fission chambers before corrections were applied is shown in Fig. 0.3,2 which shows the counting rates of the chambers plotted as a function of the reactor mean temperature. The hysteresis effect is the result of the time lag. The progress of the experiment can be read from the curves by starting on the right side at 1004 and proceeding counterclockwise around the loops. The fuel blower was turned on at 1004 and allowed to cool the fuel for 5 min, after which the blower was turned off and the system then slowly returned to its initial condition. The fission chamber counting rates increased while the blower was on and decreased after the blower was turned off again in immediate response to the cooling. The reactor mean temper- ature change, on the other hand, iagged behind both when the blower was turned on and when it was turned off. The blower was turned off shortly after 1009, but even though the counting rate started to decrease immediately, the reactor mean temperature continued to fall for approximately 2% min before it began to show a warming trend. Undoubtedly the fuel temperature did actually follow closely the changes introduced by the blower (otherwise the counting rate changes would not have been observed as promptly as they were), but because of the various effects noted in this appendix and in the “Reactor Kinetics'' section of Chapter 4 the thermocouples were slow in responding. |f the thermocouples had shown instant response there would have been no hysteresis effect observed, and the plot of £ vs mean temper- ature would have been a straight line with a nega- tive slope proportional to the temperature coef- ficient. The plot of & vs mean temperature was obtained by applying the three corrections noted above in 2This plot is actually a cross plot of the curves of Fig. 4.5, which show both the counting rate and reactor mean temperature plotted as a function of time. 194 the following way. The first correction was applied by changing the fission chamber data to BF, counter data by using the curves of Fig. 0.2. The resulting points obtained from each fission chamber were averaged and then plotted on a time scale along with the reactor mean temperature to produce a plot similar to the curves of Fig. 4.5. The maxi- mum of the counting rate curve and the minimum of the temperature curve were then matched up (the temperature curve was effectively moved up 2]/2 min in time), and new temperatures were read from the temperature curve corresponding to the time that the counts were taken. From the counting rates the multiplication factor £ = 1 — (1/M) was determined for each point, and then a plot of & as a function of mean temperature was drawn up, as shown in Fig. 0.4. A straight line could reasonably be drawn through the points. The third and final correction to be applied to the mean temperature was obtained from Appendix K. Since this experiment was one in which the fuel was cooled rapidly and equilibrium conditions were not met, the curves of Fig. K.2 are applicable. From Fig. K.2 it can be shown that a change of 1°F in the true mean temperature corresponds to a change of 1.40°F in the mean temperature read from the control room instruments. Thus the mean temperature change observed had to be increased by a factor of 1.4. A measurement of the slope of the curve of Fig. 0.4 gives a (Ak/k)/AT of 1.65 x 10~4. The average & over the plot is 0.922. By applying the factor 1.4 to the observed mean tem- perature change, the fuel temperature coefficient was calculated to be ~1.65 x 104 4 =~1.28 x 10~ — (ARR/AT = 7 o= (Ak/k) 0.922 x 1.4 This value is about 30% higher than that given by the results of experiment H-5, but it is in fair agreement in consideration of all the necessary corrections. A consideration of the errors involved showed that the maximum error was of the order of magnitude of 2.4 x 10~3. Therefore, a = ~(1.28 + 0.24) x 10~4 [f the lower limit of this value is taken, the agreement between this value and the accepted value is fairly good. This experiment did not yield an over-all temperature coefficient. £ i b £ Tron - e i E meE o T EREerE TR T R Thw WA e WETTIE: Ay e e HTERE R rEnre EETEE oW RPN " e el TET P ETRE T BT Repe R b, Sl il ek ke, e ok iRy Ahka AL e ORNL—-LR—DWG 6571 BF, COUNTER VS FISSION CHAMBER NO.2 COUNTER VS FISSION CHAMBER NO.4 FISSION CHAMBER COUNTING RATE (counts/sec) 4 2 5 10 2 5 10° 2 5 10 BF, CHAMBER COUNTING RATE (counts/sec) Fig. 0.2. Correlation Between the Counting Rates of the Bl'-'3 Counter and the Two Fission Chambers. 195 | Py prer R T R T T T E. ;i ;: ORNL~LR—- DWG 3884A 800 550 7 | TIME féo/_- ‘? ', / b 790 _ 540 - i A7 k 3 / “ / -~ & 1007 . ?805)\ toH QO 530 b 1042 : >\ b \ \ - 7 v -~ -~ - ¥ e 520 770 \O \\ , \ 1005/ 740 1319 7O \ i " N \ \ .S 14220 1004 \ 480 T END 490 Ji —_ . \'\ \ E f 1 § \\\\ "“;\ ; = | S [~ O 006 € T 760 1015 W1l ™o \ 510 3 : 3 N ~ \ & : =L \\ \ \ - - S N \ o 3 @ \‘ \ m & L “54Y'Q \ o . @ . 500 I o D 750 < i g : 3 | N \ © .k G \\ \‘ S ; O \ a2 . a N \ i £ w E @ | i i | I e rkw’#’ "y g v 1 720 \ 470 740 460 ‘f f. 700 450 - 1240 1250 1260 1270 1280 1290 1300 1310 REACTOR MEAN TEMPERATURE (°F) Fig. 0.3. Subcritical Measurement of Temperature Coefficient of Reactivity (Uncorrected Fission Chamber Data). 196 e -.-5#%“ i, e ki b win, AL - i e i e e e ORNL-LR-DWG 6572 0.927 ! - 0.926 ‘ O BLOWER ON ® BLOWER OFF 0.925 0.924 tommm b 0.923 kK 0.922 0.924 0.920 0.949 \ 0.218 \ Q.917 1230 1240 1250 1260 1270 1280 1290 1300 A REACTOR MEAN TEMPERATURE {°F) Fig. 0.4. Reactivity as a Function of Reactor Mean Temperature as Determined from the Sub- critical Temperature Coefficient Measurement. Ex- periment £-2. LOW-POWER MEASUREMENTS OF TEMPERATURE COEFFICIENTS OF REACTIVITY The low-power measurements of the temperature coefficients of reactivity were similar to the sub- critical measurements, except that with the reactor critical and on servo at 1-w power, reactivity introduced by cooling the fuel was observed by a change in the regulating rod position. The experi- ment is described in Chapter 5. As shown in Fig. 0.5, which is a plot of the regulating rod position vs the observed mean temperature during the experiment, a hysteresis phenomenon was obtained. 1t is significant that no time lags needed to be taken into account in the interpretation of this data, because thie experi- ment proceeded slowly enough for equilibrium conditions to prevail. However, since this was a cooling experiment, a temperature correction had to be applied. Cooling took place along the {ower half of the figure and heating occurred along the upper portion. Each of the curves had an initial steep slope corresponding to an initial fuel temper- ature coefficient and a less steep slope from which an over-all reactivity coefficient was found. After correction for the mean temperature read- ings, the average of the two initial slopes gave a fuel temperature coefficient of reactivity of ~9.9 x 10~2, and the other slopes gave an average over-all reactivity coefficient of about —5.8 x 102, These values agree well with the accepted values of ~9.8 x 107> and —6.1 x 10> for the fuel and over-all temperature coefficients of reactivity, respectively. 197 bkl o " T TR T : ORNL—LR -~ DWG 6573 25 ’O/ @ , TIME~LAG EFFECT 20 f © REACTOR BEING COOLED 3 ® REACTOR HEATING 2 / 2 /L z v/ = v/ 2 J ! a / / 0 2 @ / / / o = /@ / H /| e / (&) 2 {4/ / © ) o / & & & TEMPERATURE COEFFICIENT S -5 5 FOR CURVE 1, —9.46 x40 & FOR CURVE 2, =5.70x107° FOR CURVE 3, —f0.4 x {073 O/ FOR CURVE 4, —5.84 x107> 5 ® AVERAGE CURVES 1 AND 3,-9.9 x40~5 / AVERAGE CURVES 2 AND 4, —5.75 x10™5 ¢ 0 1230 1240 1250 1260 1270 1280 1290 1300 1310 Experiment L.5. 198 REACTOR MEAN FUEL TEMPERATURE (°F) ! i Fig. 0.5. Regulating Rod Position as a’ Function of Observed Reactor Mean Temperature During 1320 b i S e e e gy b oo Fopepe T R v e WYY s gt rorame yErT T oEmmene T owpmpmeve il ot il T TR R i . A A Wi . . i Ak ok b odRe il ol HIGH-POWER MEASUREMENTS OF TEMPERATURE COEFFICIENTS OF REACTIVITY During the high-power operations an experiment was conducted at a power of 100 kw which was similar to the low-power experiment just described. The fuel helium blower was turned on with the reactor on servo, and the fuel was cooled. How- ever, the action took place so rapidly that a 1-min time lag® of the fuel mean temperature had to be accounted for in plotting the data in addition to the temperature correction. Figure 0.6, in which the regulating rod movement is plotted as a function of the mean fuel temperature, shows the experi- mental measurements., Two distinct slopes were observed that corresponded to an initial fuel temperature coefficient and to an over-all temper- ature coefficient of reactivity. From the steep slope (curve No. 1) a fuel temper- ature coefficient of =1.17 x 10~% (Ak/k)/°F was obtained, and from the other slope an over-all coef- ficient of ~5.9 x 103 (Ak/k)/°F was found. These two values are in fair agreement with the accepted values of =9.8 x 107> and =6.1 x 10~ for these coefficients. 3Time lags at high power operation were observed to be shorter than those at low or no power. For a dis- cussion, see Chapter 4, ORNL.-LR-DWG-6574 20 . 15— S - . p £ = S S E 1 o F O g \ X g 10 - \\ - O = E < J = o Lf o A I \\ TEMPERATURE COEFFICIENT FOR CURVE I, -1.17 x 10 % \ FOR CURVE 2, -5.9 x {0 > 0 .oe 1240 1250 1260 1270 128C 1290 1300 1310 1320 REACTOR MEAN FUEL TEMPERATURE (°F) Fig. 0.6. Regulating Rod Position as a Fl.'mc- tion of Reactor Mean Fuel Temperatyre for Experi- ment H-4, 199 T AN R TR o, o o o e, o R s b Gieh sl Appendix P THEORETICAL XENON POISONING The xenon poisoning which existed in the ARE was determined experimentally to be an almost negligible amount. It was therefore of interest to compute the amount of such poisoning that would have been present if no xenon had been lost due to off-gassing of the fuel so that a measure of the effectiveness of the off-gassing process could be obtained. The poisoning by Xe has reached equilibrium, is given by! 135 when the xenon content o lyy + ¥9) B¢ 2/ - (Ay + 0py) Ez: where P, = the ratio of the number of thermal neutrons adsorbed in the xenon to those adsorbed in the fuel, o, = microscopic xenon cross section for thermal-neutron absorption, (v, +y,) = the total fractional yield of Xel35 from fission, both from iodine decay and direct Xe formation = 0,059, 1\2 = the decay constant for Xel3% - 2.1 x ]0-5/sec, E/Eu = the ratio of the macroscopic thermal- neutron cross section fo;sfission to that for absorption, for U 5. /5 = f Tu 0.84, $o = the average thermal-neutron flux in the fuel (since the entire fuel vol- ume, 5.33 ft3, is equally exposed to this flux, although there is only 1.37 #1% of fuel in the core at one time, the average flux in the fuel is 1.37/5.33 or 0.26 times the average flux in the reactor, which is 0.7 x 1013), The Xel35 absorption cross section, Oy is smaller in the ARE than at room temperature, be- cause the average neutron energy in the ARE exceeds the average neutron energy corresponding to room temperature and because the Xe!33 ab- sorption cross section drops off rapidly with ]S. Glasstone and M. C. Edlund, The Elements of Nuclear Reactor Theory, D. Van Nostrand Co., Inc., New York (1952), p 333, 11.57.2. 200 increasing neutron energy. R. R. Bate, R. R. Coveyou, and R. W, Osborn investigated the neutron energy distribution in an absorbing infinite moder- ator by using the Monte Carlo method and the Oracle. By assuming a constant scattering cross section and a constant 1/v cbsorpfion cross section, they found that the neutron energy distri- bution is well represented by a Maxwellian distri- bution corresponding to an effective temperature, T, provided v 2 < 0,06 K = Y- — 06 , 3 =, where the macroscopic absorption cross section 2 is measured at the moderator temperature and ES is the macroscopic scattering cross section. The effective temperature T, = T (1 + aAn) , where T is the moderator temperature, a is a constant approximately equal to 0.9, and A is the atomic weight of the moderator. The present state of the theory does not permit consideration of the inhomogeneous distribution of the various constituents of the ARE core, and therefore the main constituents were considered to be evenly distributed over the core. The nuclei per cubic centimeter were thus Oxygen 5.5 x 1022 Beryllium 5.5 x 1022 u23s 7.8 x 1017 Other elements made only a negligible contribution to the cross section of the core, The following cross sections were used; Oxygen, scattering 4 barns Beryllium, scattering 7 barns Uranium, absorption 360 barns? 20btained by converting the room temperature value of the cross section to the value at the reactor operating temperature by multiplying by the square root of the ratio of the temperatures: \ / 293 (°K) 630 (bams) X —— = 360 (barns) . 1033 (°K) T W W MR m T re. o CoTrw Rt SN TR YT T g s CER W W my v oy R [ » . ’ F'W SRR m., "M" » - bR L, i, s Rk [RF TR S M, ki, b G il iy Thus 2 360 x 7.8 x 1077 K = - = 0.038 . 3 (4+7) x 555 x 1022 For the atomic weight, A, the average of the values for beryllium and oxygen, 12.5, was as- sumed. Thus T = 10331 + (0.9 x 12.5 x 0.038)] e = 1474°K = 2200°F . The Xe!33 absorption cross section in the reactor was then determined, as shown in Table P.1, which gives the energy intervals of the neutrons, E., the fraction of neutrons in these energy inter- vals according to the Maxwell-Boltzman distri- bution, n(E.)/n, and the total Xel33 cross section, o, for each energy interval,® from which the average Xe'3% cross section is obtained. In the energy range in question the xenon adsorption cross section is approximately equal to the total xenon cross section, o,. As shown in Table P.1, t the value of this cross section is 1.335 x 10° barns or 1,335 x 10~ 18 cm?2, The anticipated xenon poisoning during the 25-hr xenon run (exp. H-11) may then be computed from Eq. 1 by using the value determined above for the Xe'35 absorption cross section: TABLE P.1. Xe!3% ABSORPTION CROSS SECTION IN THE REACTOR E (o) /= 2o Ap 0 E) fx 0 (E) : " (barns) (barns) 0.02 0.060 2.50 x 10% 0.150 x 108 0.04 0.073 2,75 x 10% 0,200 x 108 0.06 0.076 3.25 x 10% 0.248 x 108 0.08 0.075 3,30 x 106 0,248 x 10° 0.10 0.072 2.82 x 105 0,201 x 10° 0.12 0.067 1.92 x 10% 0,129 x 10° 0.14 0,062 1.27 x 10% 0,079 x 108 0.16 0.057 0.75 x 10% 0.042 x 10° 0.18 0.051 0.45 x 10 0.023 x 108 0.20 0.046 0.32 x 10 0.015 x 10° Average Utxe 1.335 X 108 (such as Inconel, etc.), and during the 25 hr of operation, if the xenon had all stayed in the fuel it would have reached 69% of its equilibrium concentration, ? By applying these various cor- rections, it is found that the xenon poisoning in the ARE at the end of the 25-hr run should have 1.3 x 10=18 x 0.059 x 0.7 x 10'3 x 0.26 PO == X 0.84 (2.1 x 1075) + (1.3 x 1018 x 0.7 x 10'3 x 0.26) 0.0117 x 10=3 0.0117 4 = = 0.005 . (2.1 + 0.24) x 10~5 2.34 This value has to be corrected because about one-~third of the fissions occur at energies above thermal and therefore have only little competition from xenon absorption. Furthermore, the reactivity loss due to poison was about 89% of that computed above because of the absorption in other poisons been 0.2% in Ak/k if no xenon had been off- gassed. 3BNL-170. 4Glasstone and Edlund, op. cit., p 333, 201 O R all R Appendix Q OPERATIONAL DIFFICULTIES The operational difficulties described here are only those which occurred during the nuclear phase of the operation, that is, from October 30 to No- vember 12, the period of time covered by this report. With a system as large, as complex, and as unique as that which constituted the ARE, it is amazing that so few difficulties developed during the crucial stages of the operation. Furthermore, such troubles were, without exception, not of a serious nature, This is in large measure attribut- able to the long period of installation and testing' which preceded the nuclear operation, the safety features inherent in the system, and the quality of workmanship which went into its construction. All major difficulties and impediments which arose are discussed below and are grouped by systems, with special regard for chronology of occurrence. ENRICHMENT SYSTEM As mentioned previously, the fuel enrichment system was changed (shortly before the critical experiment was to begin) from a remotely operated two-stage system, in which the transfer was to start with all the fuel concentrate in a single con- tainer, to a manually operated two-stage system, in which small batches of the available concentrate were transferred, one at a time. A portion of the equipment used is shown in Fig. Q.1. Although this change resuited in an improvement both in safety and control, the temperature control of the manually operated system was persistently diffi- cult. Furthermore, in order to avoid plugged lines because of the concentrate freezing at cold spots, the lines had to be continuously purged with gas and the exit gas lines then plugged as a result of concentrate-vapor condensation. Both these diffi- culties could have been avoided with proper design. The temperature control was a greater problem here than anywhere else in the system because of the small (5/16 and 3/8 in.) tubing used in the trans- fer lines and an inferior technique of heater instal- lation. This problem was aggravated by the virtual inaccessibility of the connection between the transfer line and the pump at the time the system was revised. The final heater arrangement used, which proved to be satisfactory, consisted of IDesz’gn and Installation of the Aircraft Reactor Ex- periment, ORNL-1844 (to be issued). 202 double tracing of the line with calrod heaters staggered so that successive pairs of calrods did not meet at the same point. Even with this arrange- ment, thermocouples were necessary at every heater junction and each calrod or each pair of calrods should have had a separate control. After it became an established part of the enrich- ment procedure to continuously bleed gas through the transfer line into the pump (in order to keep the line clear), the gas was vented through an extra line at the pump. The extra line was not properly heated and soon plugged with vapor condensate. |t then became necessary to use the ‘primary pump vent system which had a vapor trap. By reducing the bleed gas flow to a minimum, this vent system could be used without becoming plugged with vapor condensate. The transfer line to the pump served adequately throughout the critical experiment, although it had to be reworked four times either because of the formation of plugs or the development of leaks (in Swagelok connections where the line was cut and replaced). During the last injection for rod cali- bration the transfer line again became plugged at the fitting through the pump flange. This fitting was a resistance-heated concentric-tubing arrange- ment which provided an entrance for the 1300°F transfer line through the 700°F pump flange. The oxidized fuel from previous leaks had shorted the heating circuit, and the fitting had therefore cooled and plugged. Attempts to clear the fitting caused it to leak; the leak was sealed but the fitting was then inoperable. The final injection of concentrate, which was required for burnup at power, xencn poison, etc., was therefore made through the fuel sampling line. - A special batch of concentrate was prepared and pressurized into the pump through the sample line. This technique worked satisfactorily but had previ- ously been avoided because the line had not been designed to attain the high temperatures > 1200°F required by the melting point of the concentrate. Furthermore, the sample line was attached to the pump below the liquid level in the pump, and if a leak had developed in the line a sizeable spill would have resulted. et g et 0t oy - - opr oy e e . T e - e g e i g q?m,fit—- mmrg—»_p 'M'F R W MR e o Powr A el mTETT C pIeTR Mo Lok e b CoME Ly, BEGEL L easdod M e e e e e e o e e e e e il b B e M ey b L elkikie . e R el Fig. Q.1. Enrichment System Transfer Pot and Transfer Lines, €0t - R T T " 2 T NI Y T T WA Ty T T e ST T [T T TR e W T e T b Aot e AN - PROCESS INSTRUMENTATION For an appreciation of the generally excellent performance of the instrumentation, knowledge of the number and complexity of the instruments is required. There were at least 27 strip recorders (mostly multipoint), 5 circular recorders, 7 indi- cating controllers, 9 temperature indicators (with from 48 to 96 points apiece), about 50 spark plugs, 20 ammeters, 40 pressure gages, 16 pressure regu- fators, 20 pressure transmitters, and numerous flow recorder indicators and alarms, voltmeters, tachome- ters, and assorted miscellaneous instruments, Furthermore, many of these were employed in sys- tems circulating fuel and sodium at temperatures up to 1600°F. Since all instruments were subject to routine inspection and service, petty difficulties were kept to a minimum. In the course of nuclear operation, oniy the following instruments gave cause for particular concern: fuel flowmeter, main fuel pump level indicator, several sodium system spark plugs, fuel pressure transmitter, and several pump tachometer generators. Of these only the tachometer generator ““failures’’ were not caused by the matericls and temperatures being instrumented. The tachometers, as installed, were belt driven rather than direct- coupled and were not designed to withstand the side bearing loads to which they were subjected. The tachometers were replaced, however, before the pits were sealed, and they performed satis- factorily during the high-power operation. The fuel flowmeter and the fuel pump level indi- cator were similar instruments in which a float or bob was attached to a long tapered iron core sus- pended in a *‘dead leg.”” Coils were mounted out- side the dead leg which located the position of the core. The position of the core could be interpreted as a measure either of fuel level or of fuel flow up past the bob. The coil current, however, was very sensitive to the fuel temperature, which had to be maintained above the fuel melting point. In ad- dition to the temperature sensitivity, several coils (spare coils were provided on each instrument) opened up during the experiment, presumably due to oxidation of the coil-to-lead wire connection. The fuel flowmeter oscillated rapidly over a 10- gpm range throughout the later stages of the experi- ment, although the electronics of the instrument appeared to be in order. On the other hand, oper- ation of the main fuel pump level indicator was satisfactory up to the last day of operation, at which time the spare coil opened up (the main coil 204 had previously opened). It is felt that these instru- ments would have performed satisfactorily if the iron core were designed to move in a trapped gas leg above the float rather than in a dead leg below the float so that the operating temperature could be reduced. Furthermore, the coil reading should be ““balanced out'’ to eliminate the temperature sensitivity. Of the numerous spark plugs which were employed in the various fuel and sodium tanks as the measure or check on the liquid level, only three of those in the sodium system showed a persistent tendency to short. These shorts could not always be cleared by *‘short-burning,’”’ but they frequently cleared themselves as the liquid level dropped. Although the probes were located in a riser above the tank top to minimize shorts, the shorts could probably have been eliminated by using larger clearances than were afforded by the use of ]/8—in.—OD probes in a ]/2-in.-IPS pipe riser. The high-temperature fuel system pressure trans- mitters suffered a zero shift during the course of the experiment. These transmitters employed bellows through which the liquid pressure was transmitted to gas. 1t is probable that the bellows were distorted at times when the gas and liquid pressure were not balanced as a result of oper- ational errors or plugged gas supply lines. In any case, the gas ports in the transmitter occasionally plugged, and a zero shift in the instrument was observed. In view of the large number of thermocouples in use throughout the experiment (in the neighborhood of 1000), it is not surprising that a small number were in error. However, those which gave incorrect indications included some of the most important ones associated with the entire experiment. As discussed in Appendix K there was serious dis- agreement between line thermocouples inside and outside the reactor thermal shield, although it would appear that they both should have given the same indication. Other misleading temperature readings were obtained from the thermocouples on the tubes in the fuel-to-helium heat exchangers. These thermocouples were not properly shielded from the helium flow and read low. Although most of the thermocouple installations were designed to measure equilibrium temperatures and did so satisfactorily for a number of experi- ments involving fast transients, it was important that the response of the thermocouples be > 10°F per sec in order to correlate the changes in fluid £ B Cmer o T T b TR e T R RETT T P TR i g R TN grRRr s R T Pomemwr mwemr ¢ wmwwe s Do eyt e T e e e —w R W e Y . © o g A e e i iy, o d Res bRl i e L vl o f EBodd e = o B gy e N e g e temperatures with nuclear changes. Unfortunately it is not certain that this was the case, and, in addition, in certain instances, as with one of the thermocouples on each of the reactor AT and re- actor mean temperature instruments, the thermo- couples were mounted on electrical insulators which increased the thermal lags. The above discussion covers most of the instru- mentation difficulties that arose. This is not meant to imply, for example, that all the 800-odd thermocouples lasted throughout the experiment; there were open thermocouples scattered through- out the system. Furthermore, the instrument me- chanics were kept busy; when not doing installation work, they were usually involved in routine service and maintenance work. NUCLEAR INSTRUMENTATION AND CONTROLS It is difficult in the case of the nuclear instru- ments to separate operation problems from those inherent in routine installation and debugging, since these latter operations were continued right up to the time the instruments were needed. How- ever, during actual operation, all nuclear instru- mentation performed satisfactorily. The control mechanism operated as designed, except that one shim rod had a higher hold current (it was sup- ported by an electromagnet) than the other two. This situation was improved by cleaning the magnet face and filtering the gas surrounding the magnet. ANNUNCIATORS The control system included an annunciator panel which anticipated potential troubles and indicated off-design conditions by a fight and an alarm. During the course of the experiment, cer- tain annunciators consistently gave false indi- cations, and therefore the bells (but not the lights) of these annunciators were finally disconnected. Included were the standby sodium pump lubrication system flow, the standby fuel pump cooling water flow, the fuel heat exchanger water flow, and the fuel heat exchanger low-temperature alarm. The first three annunciators had mercury switches which were either improperly mounted or set too close to the design condition, but the fuel heat exchanger low temperature alarm error was due to faulty thermocouple indication; that is, instead of indicating fuel temperature, the thermocouple, which was located in the cooling gas stream, read low. ' HEATERS AND HEATER CONTROLS During the time the system was being heated, the heater system power was over 500 kw. This heat was transferred to the piping and other components by the assorted ceramic heaters, calrods, and strip heaters that covered every square inch of fuel and sodium piping, as well as all system components which contained these liquids. Although there were numerous heater failures resulting from mechanical abuse up until the time the pits were sealed, all known failures were repaired before the pit was sealed. Only four heater circuits were known to be inoperable at the time the reactor was scrammed ~ two heaters showed open, two shorted. Except for the fuel enrichment system in which the heating situation was aggravated by the higher temperatures as well as the small lines, the available heat was adequate everywhere. The control of the various heater circuits was, however, initially very poor; it consisted of four voltage buses to which the various loads could be con- nected plus variacs for valve and instrument heaters. However, in order to obtain satisfactory heater control for all elements of the system, 13 additional regulators were installed in addition to numerous additional variacs. With the additional regulators it was possible to split up the heater load to get the proper temperatures throughout the system without overloading any distribution panel. Even with the helium annulus, one function of which was to distribute the external heat uniformly, it must be concluded as extremely desirable, if not an absolute necessity, that all components of any such complex high-temperature system be provided with independently controlled heater units in order to achieve the desired system eguilibrium temperature. SYSTEM COMPONENTS The only major components of either the sodium or fuel system which caused any concern during the nuclear operation were the sodium valves (several of which leaked) and the fuel pump (from which emanated a noise originally believed to originate in the pump bearings). In addition, there were problems associated with plugged gas valves and overloaded motor relays. The sodium valves that leaked were the two pairs that isolated the main and standby pumps and at least one of the fill valves in the lines to the three sodium fill tanks. The leakage across 205 T S TR TR T - the pump isolation valves was eliminated from concern by maintaining the pressure in the inoper- ative pump at the value required to balance with the system pressures. The leak (or leaks) in the fill valves was of the order of % to 1 1 of sodium per day, and was periodically made up by refilling the system from one of the fill tanks. In the course of these operations, two tanks eventually became empty, and it was then apparent that most of the leakage had been through the valve to the third tank. This leakage was initially abetted by the high (™~ 50 psi) pressure drop across the valve, but even though the pressure difference was re- duced to 5 to 10 psi the leckage rate appeared to increase during the run. It was of interest, as well as fortunate for the ultimate success of the project, that the fuel carrier fill valves were tight. However, in the fuel system, only two valves were opened during the filling ~operation and only one of these had to seal in order to prevent leakage. This valve did seal, and it was opened only one other time, i.e., when the system was dumped. Each sodium pump {main and standby) and each fuel pump (main and standby) was provided with four microphone pickups to detect bearing noises. Shortly after the fuel system had been filled with the fuel carrier, the noise level detected on one of the main fuel pump pickups jumped an order of magnitude, while that on the other increased substantially. At the time the noise was believed to be due to a flat spot on a bearing, and operation was therefore watched very closely. When the noise level did not increase further (in fact, it tended to decrease), it was decided to continue the experiment without replacing the pump (a very difficult job which could conceivably have resulted in contamination of the fuel system). The pump operated satisfactorily throughout the experiment. Subsequent review of the pump design and behavior of the noise level indicated that the noise probably originated at the pump discharge where a sleeve was welded inside the system to effect a slip connection between the pump discharge duct from the impeller housing and the exit pipe, which was welded to the pump casing. Vibration of the sleeve in the slip joint could account for all the noise. Although the vent header was heated from the fuel and sodium systems to the vapor trap filled with NaK (which was provided to remove certain fission products but also removed sodium vapor), 206 the temperature control of the header and the individual vent lines connected to it was not adequate to maintain the line above the sodium melting point and yet not exceed the maximum temperature limit (400°F) of the solenoid and diaphragm gas valves. Consequently, these vent lines became restricted by the condensation of sodium vapor. |t was apparent that a higher temper- ature valve would have been desirable, that the gas line connecting the tank to the header should have been instailed so that it could drain back into the tank, and, also, that good temperature control of the line and valve should have been provided. As it was, exceptionally long times were required to vent the sodium tanks through the normal vent valves. |t would have been necessary to use the emergency vent system, which was still operable at the end of the experiment, if a fast dump had been required. In addition to the above failures or shortcomings, there were numerous probiems of less serious nature in connection with auxiliary equipment, motor overloads, water pipes which froze and split, and air dryers which burned out. However, nothing occurred in such a manner or at such a time as to have any significant bearing on the conduct of the experiment. LEAKS The only sodium or fluoride leaks that occurred have already been discussed. These included one minor sodium leak in the sodium purification sys- tem (discussed in chap. 3, *‘Prenuclear Operation’’) and two fuel concentrate lecks from the enrichment system. That neither the reactor fuel system nor sodium system leaked is a tribute to the quality of the workmanship in both welding and inspection that went into the fabrication of these systems. In all, there were over 266 welded joints exposed to the fivoride mixture, and over 225 welded joints were exposed to the sodium. In contrast to the liguid systems, there were several leaks in gas systems. 1t is felt that these feaks would not have occurred if the gas systems had been fabricated according to the standards used for the liquid systems. The notable gas leaks were from the fuel pumps into the pits, from the helium ducts in the heat exchangers into the pits, and from the pits into the building through the various pit bulkheads, as well as the chamber which housed the reactor controls. ik o e e e » ¥ opm wmonr Vipawowr ¥ owrmcrtE rgerops S et e e e - chR b, Lo T e S A o e, R . 1"y The combination of the leak out of the fuel pump and that out of the pits required that the pits be maintained at subatmospheric pressures in order to prevent gaseous activity from contaminating the building. Accordingly, the pit pressure was lowered by about 6 in. H20 by using portable compressors which discharged the gaseous activity some 1000 ft south of the ARE building. The activity was of such a low magnitude that, coupled with favorable meteorological conditions, it was possible to operate in this manner for the last four days of the experiment. The ledks out of the helium ducts in the fuel and sodium heat exchangers resulted in a maximum helium concentration in the ducts of the order of 50%, and to maintain even this low concentration, it was necessary to use excessive helium supply rates, i.e., 15 cfm to the ducts alone and another 10 cfm to the instruments. 207 R Appendix R INTEGRATED POWER The total integrated power obtained from the operation of the Aircraft Reactor Experiment does not have any particular significance in terms of the operational life of the system. However, a total integrated power of 100 Mw was more or less arbitrarily specified as one of the nominal ob- jectives of the experiment. While at the time the experiment was concluded it was estimated that this, as well as all other objectives of the experi- ment, had been met, the estimate was based on a crude evaluation of reactor power. Since subsequent analyses of the data have permitted a reasonably accurate determination of the reactor power, it is of interest to reappraise the estimated value of the total integrated power, The total integrated power could be determined from either the nuclear power or the extracted power (cf., section on ‘‘Reactor Kinetics'' in chap. 6). The power curves, which should have equivalent integrals, could be obtained from any of a number of continuously recording instruments, i.e., the nuclear power from either the micromicro- ammeter or log N meter, and the extracted power from any of the several temperature differential recorders in the fuel and sodium circuits., The total integrated power has been determined both from integration of a nuclear power curve (log N) and from the sum of the extracted power in both the sodium and fuel circuits, as determined by the temperature differentials in each system together with their respective flows (which were held constant). For both power determinations a calcu- lation of the associated error was made. EXTRACTED POWER The integrated extracted power was determined from the charts which continuously recorded the temperature differential (AT) in each system. As a matter of convenience the sodium system AT was taken from a 24-hr circular chart, while the fuel AT was taken from one of the six Brown strip charts which recorded, in the control room, the AT across each of the six parallel fuel circuits through the reactor. The individual circuit recorders had a much slower chart speed than that of the over-all AT recorder and were therefore much easier to read, To keep the results well within the accuracy of the whole experiment, tube No. 4 was selected for the determination because 208 the chart trace very closely corresponded to that of the over-all fuel AT across the reactor. From Fig. K.3, Appendix K, the control-room-recorded AT was converted to what was accepted as the correct AT, The power extraction was then calcu- lated and plotted against time in Fig. R.1. The sodium AT across the reactor was obtained from the circular charts of the recorders located in the control room. These AT's were also corrected by using Fig. K.6, Appendix K, and the plot of the power extracted by the sodium as a function of time is also shown in Fig. R.1 on the same abscissa as that of the fuel plot. The total integrated (extracted) power, i.e., the sum of the area under both the fuel and sodium system power curves, was then determined by using a planimeter; it was found to be 97 Mw-hr, A calculation was also made of the magnitude of the ““maximum’’ possible error in the determi- nation of the reactor power. The power equation, which was calculated for both the fuel and sodium systems, was P = kf AT , where P = reactor power, k = a constant containing heat capacity and conversion units, f = flow rate of fuel or sodium, AT = temperature difference across reactor, The consequent error equation is AP = Ef A(AT) + EAT Af + AT Ak , where Ak = maximum error in the heat capacity, Af = maximum error in the flow rate, A(AT) = maximum error in the temperature difference. Nominal average values of these factors for the fuel system were f = 46 gpm AT = 350°F k= 0.11 kw/day gpm Af = 2 gpm = 5% AMAT) = 10°F = 3% Ak = 0.011 £ 10% Therefore AP (for fuel system) = 20% T e P M W 1 TR U sty ame YO PO T R} o rep g T PR T e Ppr v ™o oewowp T e b » T T e & - o ' e »Fg s A g et v b L sk f ek s B i ol L e wi koo Ak EXTRACTED POWER {Mw) EXTRACTED POWER (Mw) EXTRACTED POWER (Mw) o 1300 0 G100 2.8 2.0 0.8 0.4 1300 1500 {700 4900 NOVEMBER 8 PRACTICE OPERATION 0300 0800 Q700 1500 {700 {1900 NOVEMBER i 2100 OPERATION PRACTICE EXP H-7 0900 IEMONSTRATION D Iro AlR FORCE 2100 FUEL SYSTEM POWER ——=—— SODIUM SYSTEM 2300 0100 0300 050C O70C 0900 400 1300 NOVEMBER 9 OPERATION TO AIR FORCE PRACTICE EXP. H-10 RUN { 1100 1300 1800 {700 1900 2100 2300 0100 NOVEMBER 40 DEMONSTRATION QOF HIGH-POWER QOPERATION EXP, H-i4 2300 0100 0300 0500 O70C 0900 00 1300 ‘ NOVEMBER 12 Fig. R.1. Power-Time Curve. ORNL-LR-DWG 63555 1500 0300 1500 1700 S 1800 0500 0700 NOVEMBER 1 {700 1200 2100 0900 2100 2300 0100 1400 2300 1300 209 e TS T e T N XN — e TR e The values for the sodium system were f = 152 gpm AT = 50°F k = 0.0343 kw/day.gpm Af = 4 gpm = 3% A(AT) = 10°F = 20% AR =0 Therefore AP (for sodium system) = 11% The weighted, over-all percentage error was therefore AP P,»XOQO PSXOO]] —_— = + =17, p P pP where P, and P_ are the power extracted in the fuel and sodium systems, The errors made in the power integration by using the AT chart are only those associated with the determination of power extraction. Since Fig. R.1 is a reproduction of the AT trace and it was integrated by using the planimeter, any integrating errors were assumed to have been averaged out. Therefore, the error that applies to the integrated exfracted power is the 17% that is applicable to the power level determination; therefore the integrated extracted power was 97 £ 16.5 Mwshr ., NUCLEAR POWER During most of the time the reactor was operating at any appreciable power level, the nuclear power level was kept fairly constant and was recorded in the log book. During the times the power level was either not constant or not recorded in the log book, the nuclear power was determined from the log N recorder chart by integrating the area under the power trace, (The log N trace was used rather than the micromicroammeter because no record was kept of the micromicroammeter shunt value as a multiplying factor.) Since the log N chart gave a log of power vs time plot, exact integration under the curve would have been an extremely difficult task; accordingly, the area under the curve was integrated graphically. The curved portions of the trace were approximated by straight lines, and an average log N valuve was determined for each line segment. |t was assumed that there were as many positive as negative errors in this method and that the errors cancelled out, The total integrated power was then obtained 210 by adding up all the incremental areas which were in terms of average log N units times time, From the 25-hr constant-power xenon run a relation between the log N reading and the extracted power was obtained, |t was found that 22.5 log N units = 2.12 Mw, or 10.6 log N units = ] Mw; therefore log N 10.6 for any segment under the curve. By using this correlation between actual power and log N reading, the total integrated nuclear power was calculated to be 96.6 Mw-hr. This value of the integrated nuclear power, O, as determined from the log N chart, was, in effect, found from the following relation x time (hr) = Mw-hr 0 = MCA: , where M = average log N recorded value over an increment of time At, At = increment of time in hours, C = a constant which converts the log N value to Mw-hr, The constant, C, in the expression is equal to the percentage error in the extracted power level determination, which was calculated in the preceding section to be 17%. Therefore the error in C is (0.17) (1/10.6) = 0.016. The 74-hr period of high-power operation was divided into two parts, one part being the 25-hr period of constant power during the xenon run and the remainder being the 49-hr period of variable power. The errors were of different magnitude for each part, since during the xenon run there was no error in time or in the determination of the average value of M, For the 25-hr xenon run the integrated power was 53 Mw-hr. Thus M = 22.5 log N units At = 25 hr C = 0.094 Mw/log N AM = 0 A(A?) = 0 AC = 0.016 = 17% Therefore AQ] = MC AAt) + M AC At + AM C At = 9 Mw-hr . For the balance of the operating time (49 hr) and integrated power (43.6 Mw-hr), the average M must Free g TRE i e T R R BT R T ONMPE Ry e g Yoty tpees T VTR T Ry e YE pRErTE R treer o AR T R T T e iy T rRr B Ty ER g e o B TR RELT R, s b Rk e gt be determined as H=-Q—= 43.6 — . = 9.4 Ct (0.094) (49) ¢ and for AQ, M = 9.46 log N units 49 hr = 0.094 Mw/log N AM =1 a2 I I AAL) = 1 hr AC = 0.016 = 17% AQ,=MC A(At) + M AC At + AM C At = 12.8 Mw-hr, The total error was therefore AQ = 12.8 + 9.0 = 21.8 Mw-hr or 22.5% . The total extracted nuclear power then was 96.6 + 21.8 Mw-
1005 Heat barriers up. 1005 Blower started. 1005:10 Blower up to 275 rpm. 1010 Blower off. Water heat exchanger rose 25°F, Manometer reading 5.45 in.; corresponds to ~ 170 gpm, YR TRt M M TR o e » PR WY amp WA Y MO omey e vy MR R gt AR VR wewr rSeesr oy TSI o Wi 1615 Sample from fuel system taken, This is sample 10, 1820 Results from sample 10 show 9.58 + 0.08% uranium, 2015 Sample 11 taken for analysis. 2020 Referring back to run 8, experiment E-1: it is estimated that 5.5 Ib from can 5 went into transfer tank. Approxi- “mately 0.2 Ib was lost in the leak and 5.3 |b went into the system, Counts vs shim rod position were taken at 5-in, intervals on shim rods, , Time of day was 1442 to 1512, 2215 Sample 11 showed 9.54 + 0.08% total uranium, = November 3, 1954 0136 Second 5 b batch from can 5 injected. Rods at 20 in, Counts taken at intervals for each 5 in. of rod withdrawal, ol 0152 Rods inserted to 20 in. ! 0205 Injection nozzle shorted. T R MR TR A A ER e “*_. ; . “ & i. i g £ fr 216 BXReE . W Run Time (8) 0226 0256 9 0459 0523 10 0836 0911 0941 1018 11 1110 1147 1230 1245 12 1453 1536 1545 1547 1603 1604 1626 0830 0840 EXPERIMENT E-2 (continued) November 3, 1954 Third 5 [b batch from can 5 injected, Rods at 20 in. Counts taken as rods withdrawn, Final batch from can 5 injected, Chemists estimate about 21/2 Ib, Counts taken as rods withdrawn. First batch consisting of 5.5 b from can 11 injected. Rods at 20 in. Counts taken as rods withdrawn, Remainder of can 11 injected. Rods at 20 in. Counts taken as rods withdrawn, 5.5 Ib from can 22 injected. Rods at 20 in. Counts taken as rods withdrawn, Second 5.5 Ib from can 22 injected. Counts taken as rods withdrawn, Completing injection from can 22, Can 22 empty. Counts taken as rods withdrawn, Reactor monitored. Reads ~§ mr/hr. 5.5 Ib from can 31 injected. Rods at 20 in. Counts taken as rods withdrawn. Second 5.5 Ib from can 31 injected. Rods at 20 in. Fission chamber 2 withdrawn 2 orders of magnitude. Balance of can 31 injected. Rods at 20 in, Counts taken as rods withdrawn, 5.5 Ib from can 20 injected, Rods at 20 in. Counts taken as rods withdrawn, Source withdrawn — reactor subcritical, Second 5.5 Ib from can 20 injected. Rods at 20 in. Counts taken as rods withdrawn, Reactor critical. Control given to servo. Radiation survey made, Reactor reads 750 mr/hr at side; ~ 10 mr/hr on grill above main fuel pump. Reactor shut down, Shim rods brought out to about 18 in, with source in core, November 4, 1954 Sample 12 taken. Sample 13 taken. EXPERIMENT L-1 Objective: One-Hour Run at 1 w! (Estimated) to Determine Power Level; 1107 1118:40 1218:40 1250 1300 1400 .Radiatien Level Check Shim rods coming out, Reactor up and leveled out. Estimated 1 watt.! On servo, Micromicroammeter 1 x 10=9; 48.6 on Brown recorder, Reactor scrammed. Sample 14 taken._' Sample 15 taken, Estimated power from sample 15, 1.6 w. repeated at 10 w. 2 Count was low. Must be YActual power subsequently determined to be 2.7 w, 250 Appendix H, **Power Determination from Fuel Activation."’ 217 T T TR T iR T e i il Run 218 Time 1525 1651 1657 1707 1710 1746 1758 1804 1814 2009 2018 2024 2030 2230 0142 0158 0210 0235 0240 0245 0250 0252 0346 0350 0354 0357 0401 - i d EXPERIMENT L-1 (continued) November 4, 1954 Four samples taken from the fuel system since going critical. Analyzed as follows: Time Sample No. Uranium (total) (wt %) 0830 12 12.11 £ 0.10 0840 13 1221 £ 0.12 1250 14 12,27 + 0.08 1300 15 12.24 + 0.12 EXPERIMENT L.2 Objective: Rod Calibration vs Fuel Addition Reactor brought critical before injecting first penguin, Rods coming out. Reactor up to ~1 w. Regulating rod at 13.1 in, Reactor subcritical, Penguin 11 injected, Reactor up to ~1 w. Regulating rod at 10.6 in. Rod moved 2.5 in,, from 13.1 to 10.6 in. Readjusted rods. Regulating rod at 13.0 in. Reactor subcritical. Penguin 19 put in furnace, Penguin 19 injected. Time between pips, 0.75 min, Reactor up to ~1 w. Red at 5.7 in. Reactor scrammed. ' Started changing from rod 4 (19.2 g/cm) to rod 5 (36 g/cm). Counts on fission chambers and BF , taken before changing rod. EXPERIMENT L-3 Fuel system characteristics were obtained, EXPERIMENT L.2 (continued) November 5, 1954 Started withdrawing rods, Upto~1lw, Reactor subcritical. Up to ~1 w again. Temperature had drifted. Regulating rod position, 12,1 in, Reactor subcritical, Penguin 4 injected. Reactor at ~1 w again. Regulating rod position, 12.1 in, Reactor subcritical, Apparently penguin didn't come over (no dip line). Reactor at ~1 w, Regulating rod position, 11.5 in. Reactor subcritical, Penguin 14 injected, Reactor at ~1 w, Regulating rod position, 9.9 in. Reactor subcritical. Regulating rod movement, 1.6 in. ~j FoRT TR mwmprme B T TTT Nt TTECT Oy £ TR nuste miadeats BT R RE A BN W WY e ITTTEEOMEY SR T T v T W TTRE S Ry o iy < i il ke b, RN - G Awm & £ ke B g b Gl RGN ok W Run Time 4 0429 0431 0439 0442 0445 5 0519 0524 0529 0536 0542 6 0607 0612 0619 0626 0631 0715 1015 1040 1105 1350 1403 1413 1414 1416 1417 1418 EXPERIMENT L-2 (continued) N ber 5. 1954 ovember 5, Reactor at ~1 w. Regulating rod position, 9.7 in. Reactor subcritical, Penguin 16 injected. Reactor at ~1 w. Regulating rod position, 7.8 in. Rod moved 1.9 in. Reactor subcritical, Reactor at ~1 w. Regulating rod position, 7.65 in. Reactor subcritical. Penguin 13 injected. Reactor critical at ~1 w. Regulating rod position, 6.80 in. Reactor suberitical. Reactor critical at ~1 w. Regulating rod position, 6.5 in. Reactor subcritical, Penguin 5 injected. Reactor critical. Regulating rod position, 5.50 in. Reactor subcritical, Shim rods 1 and 2 inserted. Reactor shut down due to increase in radiation level above fuel pump tank upon adding concentrate to pump. During fuel addition for run 9 the background picked up to ~ 50 mr/hr and, on run 5 addition, increased to 55 mr/hr. The normal background at the point of measurement being ~1 mr/hr. Trimming pump level to normal operating probe level. Trimming pump level to estimated ‘4 in. below normal operating probe, Took fuel sample 16. EXPERIMENT L-2-A Objective: Test on Activity of Yent Lines by Operating Reactor at ~1 w for ™~ 10 min and then Venting as Though Adding Fuel Preliminary data recorded. Rods 1 and 2 being withdrawn. Rods at 30 in., rod 2 being withdrawn., Fission chamber 1 at full scale, Period meter shows slight period, ~ 400 sec. Rod 2 at upper limit, Rod 1 being withdrawn, Period meter reaches 100 to 50 sec. L.og N reading, 2 x 10~4, Fission chamber 2 at 500 counts/sec, Rod 1 at upper limit Regulating rod being withdrawn. Micromicroammeter reading, 10. Fission chamber 2 at 1000 counts/sec. Log N, 3 x 10-4. Period, ~400 sec, 219 Term T I T TR g e EXPERIMENT L<21 Mw. Moved shim and regulating rods to obtain new heat balance. Sodium system blower speed up to 2000 rpm, Extracted power: from fuel, 1920 kw; from rod cooling, 20 kw; from sodium, 653 kw; total, 2.6 Mw, No. 2 rod cooling blower started. Front sodium system blower off; reactor power reduced to ~ 100 kw. Reactor mean temperature and power leveled off; power, ~ 1.5 Mw, Reduced fuel system helium blower speed to minimum, Fuel AT nearly constant. Raised fuel system helium blower speed to maximum, Fuel AT essentially restored. Fuel system blower turned on; speed reached 1500 rpm, Blower speed reduced to 1000 rpm. Reactor AT about 200°F, Regulating rod withdrawn to raise mean temperature from 1320 to 1325°F. Regulating rod inserted to lower mean temperature from 1325 to 1320°F. Regulating rod withdrawn to raise mean temperature, After demonstration, power leveled off to 1.5 Mw. Demonstration of reactor operation. End of demonstration of reactor operation, Reactor run more or less steadily at 2 Mw until time for scram, Reactor power up to ~2.5 Mw, Reactor scrammed. 231 R e p——— e i Appendix U THE ARE BUILDING The ARE building, shown in Fig. U.1, is a mill type of structure that was designed to house the ARE and the necessary facilities for its operation. The building has a full basement 80 by 105 ft, a crane bay 42 by 105 ft, and a one-story service wing 38 by 105 ft. The reactor and the necessary heat disposal systems were located in shielded pits in the part of the basement serviced by the crane. One half the main floor area was open to the reactor and heat exchanger pits in the basement below; the other one half housed the control room, office space, shops, and change rooms. In one half the basement were the shielded reactor and heat exchanger pits; the other one half of the basement was service area and miscel- laneous heater and control panels. The control room, office space, and some shops were located on the first floor over the service area. The first floor does not extend over the one half of the basement that contains the pits. The crane is a floor-operated, 10-ton, bridge crane having a maximum lift of 25 ft above the main floor level. Plan and elevation drawings of the building are shown in Figs. U.2 and U.3. The entire reactor system was contained in three interconnected pits: one for the reactor, another for the heat exchangers and pumps, and a third for the fuel dump tanks. These pits, which were sealed at the top by shielding blocks, were located in the large crane bay of the building. The crane bay was separated from the control room and offices, and the heating and air conditioning systems maintained the control room at a slightly higher pressure than that of the crane bay. Fig. U.1. The ARE Building. 232 s W R R h | eV fi Ty T AL b L R R R YD I . A YT —TEE TR T » M TR gy T T T MR VR oty R kM ek Sk e Sew s W L e~ ke A g e s Wi Mo . - S oW K - o R, B o, Ae A LS AR here it _ — — o — 4 4! A . 3§ LEGEND ORNL-LR~-DWG 6553 STANDARD MIX CONCRETE |—" c V277777 SOLID CONCRETE MASONRY-STANDARD MIX 81 ft : SOLID CONCRETE BLOCKS LAID DRY - STANDARD MIX i ; A REMOVABLE SPECIAL MIX BLOCKS ; 4zt i : C ; ‘ M e NS o i = o pemeeseecs o 1 ey — , _ LOADING AREA EQUIPMENT BATTERY CRAFT FOREMAN INSTRUMENT ROOM CRANE BAY AND SHOP :, FIELD ENGINEERS — : DOWN E ‘ CONTROL PIT : A : T 1 + § ! l | __ AMPLIFIER /< CABINET d OFFICE BASEMENT INSTRUMENT PANEL TEST |[i{PITS § D TEST PIT REMOVABLE ||| ROOF SLABS 2 A HI H W = A = < b S 8 K o o - OFFICE Z 2 /< = o = 0 P 8 ° 3 ! = | . & L H HEAT EXCHANGER PIT B8 o N g 8 B 8 . L OFFICE : t 1+t A | ] Y ‘ i RELAY CABINET I i POWER EGQUIPMENT ROOM | ; ' E T é///////////, ,_E B | N 0 : P WOMEN| | JANITOR | ROOF SLAB STORAGE "o =l 2 SRSy SLAB STORAGE S SRR OPEN TO BASEMENT 83 ft 2in. Lec FIRST FLOGOR PLAN BASEMENT PLAN Fig. U.2. Plan of the ARE Building. £eC TR TR TIRT T T e T ——" T S T T oo™ e o | " T T I T TN O 0 SRS emr Ty YT T e T P T T IOV NI T e el e R N ® LEGEND STANDARD MiX CONCRETE SOLID CONCRETE MASONRY-STANDARD MIX SOLID CONCRETE BLOCKS LAID DRY - STANDARD MIX - REMOVABLE SPECIAL MIX BLOCKS T '_ -———Ifi_,_L s \ T AT A n ' M -~ 8 L A W H Ip i — o F 10 TON BRIDGE CRANE ™™~ A M T N | 4 i PIT ORNL-LR-DWG 6554 ! CONTROL {EQUIPMENT ¥ HEAT EXCHANGER PIT Lonall SLAB STORAGE PIT Lo J-__. _'___:.__-._-r ' Ll X B b ! i il J A . T v . 1 : — rdy ~ > v ’ » ' v SECTION C-C CRANE BAY . 4 TEST PIT TEST PIT Fews SECTION A-A ¥4 e T OWET T gy cw e oppemm g WERTREC rfeanomt RTTUT TN CWEERCS e g Rre ) wmppme mereprt EwEeEy SWUTCNET YA oMyt O oMEEecl Ty ey . 3 - -I], “GAS SEAL PLATE HEAT EXCHANGER PIT 20 ft Bin: e 42 ft Qin. . I i BIBLIOGRAPHY Aircraft Nuclear Propulsion Project Quarterly Progress Reports for the Periods Ending: March 10, 1951 “June 10, 1951 September 10, 1951 December 10, 1951 March 10, 1952 June 10, 1952 September 10, 1952 December 10, 1952 March 10, 1953 June 10, 1953 September 10, 1953 December 10, 1953 March 10, 1954 June 10, 1954 September 10, 1954 December 10, 1954 ANP.60 ANP-65 ORNL-1154 ORNL-1170 ORNL-1227 ORNL-1294 ORNL-1375 ORNL-1439 ORNL-1515 ORNL-1556 ORNL-1609 ORNL-1649 ORNL.-1692 ORNL-1729 ORNL-1771 ORNL-1816 235 T T DATE 1948 12-15-49 5-21-51 4-17-51 6-5-51 7-25-51 8-13-51 8-29-51 10-15-51 1-5-52 1-8-52 1-11-52 1-14-52 1-22-52 1-29-52 1-29-52 3-10-52 3-20-52 3-24-52 3-28-52 4-16-52 4-22-52 5-8-52 6-2-52 8-8-52 236 TITLE Metals Handbook (Nicke! and Nickel Alloys, p 1025-1062) The Properties of Beryllium Oxide Activation of Impurities in BeQ Perturbation Equation for the Kinetic Re- sponse of a Liquid-Fuel Reactor The Contribution of the (n,2rn) Reaction to the Beryllium Moderated Reactor Radiation Damage and the ANP Reactor Alkali Metals Area Safety Guide Physics Calculations on the ARE Control Reods Some Results of Criticality Calculations on BeO and Be Moderated Reactors Physics of the Aircraft Reactor Experiment Statics of the ANP Reactor — A Preliminary Reporf The ARE with Circulating Fuel-Coolant A Flux Transient Due to a Positive Reac- tivity Coefficient Heat Transfer in Nuclear Reactors Safety Rods for the ARE Effect of Structure on Criticality of the ARE of January 22, 1952 A Simple Criticality Relation for Be Moder- ated Intermediate Reactors Health Physics Instruments Recommended for ARE Building Induced Activity in Cooling Water - ARE Optimization of Core Size for the Circu- lating-Fue! ARE Reactor Physics Considerations of Circulating Fuel Reactors Note on the Linear Kinetics of the ANP Circulating-Fuel Reactor Statics of the ARE Reactor Reactor Program of the Aircraft Nuclear Propulsion Project The ARE Critical Experiment AUTHOR(s) American Society for Metals M. C. Udy, F. W, Boulger W. K. Ergen N. Smith et al, C. B. Mills N. M. Smith, Jr. L. P, Smith Y-12 Alkali and Liquid Metals Safety Committee J. W. Webster R. J. Beeley J. W. Webster 0. A, Schulze C. B. Mills C. B. Mills C. B. Mills C. B. Mjlls R. N. Lyon W. B. Manly E. S. Bomar C. B. Mills C. B. Mills T. H, 1. Burnett T. H. J. Burnett C. B. Mills W. K. Ergen F. G. Prohammer C. B. Mills Wm. B. Cottrell (ed.) C. B. Mills D. Scott REPORT NO. BMI-T-18 Y-F20-14 ANP-62 Y-F10-55 ANP-67 Y-8 Y-F10-71 ANP.66 Y-F10-77 Y-F10-81 Y-F10-82 Y-F10-63 CF-52-1-76 CF-52-1-192 Y-F10-89 Y-F10-93 CF-52-3-147 CF-52-3-172 Y -F 10-96 Y-F10-98 Y-F10-99 Y-E10-103 ORNL-1234 Y-F10-108 - o} T TR TR I R T W - e o vy e b vy EEE T - ¥ vemes W R ROTBP o R R R T T i ¥ S gr-*. ¥ Y LT e t reeee " DATE 7-30-52 9-52 9-52 9-25-52 10-22-52 11-1-52 11-24-52 11-13-52 11-17-52 11-25-52 1-7-53 1-9-53 1-12-53 1-27-53 1-27-53 1-27-53 2-45 2-11-53 2-26-53 3-17-53 3-30-53 4-7-53 TITLE Loading of ARE Critical Experiment Fuel Tubes Welding Procedure Specifications Welders Qualification Test Specifications Radiation Through the Contrel Rod Pene- trations of the ARE Shield Xenon Problem — ANP Structure of Norton’s Hot-Pressed Beryllium Oxide Blocks Aircraft Reactor Experiment Hazards Summary Report Effects of lrradiation on BeO Structure of Be0 Block with the 11/8 in, Central Hole An On-0Off Serveo for the ARE Delayed Neutron Damping of Non-Linear Reactor Oscillations ARE Regulating Rod A Guide for the Safe Handling of Molten Fluorides and Hydroxides Delayed Neutron Activity in a Circulating Fuel Reactor Components of Fluoride Systems Delayed Neutron Activity in the ARE Fuel Cireuit Some Engineering Properties of Nickel and High-Nickel Alloys Heating by Fast Neutrons in a Barytes Con- crete Shield Methods of Fabrication of Control and Safety Element Components for the Air- craft and Homogeneous Reactor Experi- ments Corrosion by Molten Fluorides The Kinetics of the Circulating-Fuel Nuclear Reactor Minutes of the Final Mee*ing of the ARE Design Review Committee AUTHOR(s) D. Scott P. Patriarca P. Patriarca H, L, F, Enlund W. A, Brooksbank L. M, Doney J.H. Buck W. B, Cottrell G. W, Keilholtz L. M. Doney S. H. Hanauer E. R. Mann J. J. Stone W. K, Ergen E. R. Mann S. H, Hanauer Reactor Components Safety Committee H. L. F. Enlund Wm, B, Cottrell H. L. F. Enlund B. B, Betty W. A. Mudge F.H. Abernathy H. L. F. Enlund J. H. Coobs E. S. Bomar L.S. Richardson D. C. Vreeland W. D. Manly W. K. Ergen . W. R. Gall REPORT NO. Y-B23-9 PS-1 QTS-1 Y-F30-8 CF-52-10-187 CF+52-11-12 ORNL-1407 CF-52-11-85 CF.52-11-146 CF.52-11-228 CF-53-1-64 CF-53-1-84 Y-B31-403 CF-53-1-267 CF.53-1-276 CF-53-1-317 Mech. Eng. 73, 123 (1945) CF-53-2-99 ORNL-1463 ORNL-1491 CF-53-3-231 CF-53-4-43 237 T T T T T e T R B Ty - e rrrm—— DATE. 5-18-53 6-19-53 7-20-53 7-20-53 7-20-53 8-3-53 8-10-53 9-1-53 9-1-53 9-3-53 9-22-53 © 9-25-53 10-18-53 12-1-53 12-22-53 12-18-53 3-2-54 4-7-54 5-7-54 7-20-54 7-27-54 .7-31-54 - 8-28-54 To\"lv_ae issued TITLE ARE Control System Design Criteria The Stability of Several |ncone!-UF4 Fused Salt Fuel Systems Under Proton Bombard- ment . Current Status of the Théory of Reactor Dynamics Interpretation of Fission Distribution in ARE Critical Experiments Composition of ARE for Ciéticality Calcu- lations ARE Fuel Requirements Thermodynamic and Heat Transfer Analysis of the Aircraft Reactor Experiment ARE Fuel System Static Analysis of the ARE Criticality Ex- periment (A Preliminary Report Pending Reception of the ARE Criticality Report) Experimental Procedures on the ARE (Pre- limindry) The General Methods of Reactor Analysis Used by the ANP Physics Group Supplement to Aircraft Reactpr Experiment Hazards. Summoary Report (ORNL-1407) Preliminary Critical Assembly for the Air- craft Reactor Experiment ARE Design Data The Inhour Formula for a Circulating-Fuel Nuclear Reactor with Slug Flow Analytical and Accountability Report on ARE Concentrate ARE Design Data Supplement ARE Instrumentation List Analysis of Critical Experiments ARE Operating Procedures, Part |, Pre- Nuclear Operation ARE Operating Procedures, Part 1l, Nuclear Operations Fuel Activation Method for Power Determi- nation of the ARE Critical Mass of the ARE Reactor Stress Analysis of the ARE AUTHOR(s) F. P, Green W. J, Sturm R. J. Jones M. J. Feldman W. K Ergen Joel Bengston J. L, Meem J. L. Meem B. Lubarsky B. L. Greenstreet G. A. Cristy C. B, Milis J. L. Meem C. B. Mills E. S. Bettis W. B. Cottrell D. Callihan D. Scott W. B. Cottrell ¥W. K, Ergen G. J. Nessle W. B. Cottrell R. G. Affel C. B. Mills W. B, Cottrell J. L. Meem E. B. Johnson C. B, Mills R. L. Maxwell J. W. Walker REPORT NC CF-53-5-238 ORNL-1530 CF-53-7-137 CF-53-7-190 Secret rough dr CF-53-8-2 ORNL -1535 CF-53-9-2 CF-53-9-19 CF-53-9-15 ORNL -1493 - CF-53-9-53 ORNL-1634 CF.53-12-9 CF-53-12-108 CF-53-12-112 CF-54-3-65 CF-54-4-218 CF-54-5-51 CF-54-7-143 CF-54-7-144 CF-54-7-11 CF-54-8-171 ORNL -1650 £ ¥ 4 w i e SR e e row CTE W e g ey gy T rrer