ML20085D752
Text
. .
t a S
0 :u
( a
\
' fO d W
! a
(
n e .a . m . , e . . . . . . . . ...
THE SAXTON CHEMICAL SHIM EXPERIMDiT f ^~ .
/
/
r e
\*;,
'? . . .
\- ,',;/
Editore Y J. Weisman B. Bartnoff l
Contributors i
S. Bartnoff W. D. Fletcher M. J . Ibli F. J. Frank P. Cohen A. J. Impink R. W. Colombo G. R. Taylor i T. L. Erion J. Weisman r
l
,. ---- , _3 m *
- .- N i N_ T August, 1964 l
i 00
} {, {, .,3 0110170124 91042*1 pDR FOIA POR DEKOK91-17
e .
DISTRIFJTION W. E. At'bott F. L. Langford R. J. Allio V. D. Leggett B. H. Axelson W. Lester S. Bartnoff J. S. Moore E. S. Beekjord N. R. Nelson M. J. Bell R. C. Nichols L. Chajson J. D. O'Toole
,C uC. Christensen E. U. Powell P.CoheD (V.Rajagopa R. W. Colombo J A Rengel R. J. Creagan C. Roderick W. F. Davis H. L. Russo P. G. DeHuff D . G . S a.marone W. F. Eanes T. Stern T. L. Erion R. L. Stoker G. H. Farbman G. R. Taylor H. M. Ferrari A. O. Thorp W. D. Tietcher H. J. van Hollen F. J. Frank J. Weisman R. J. French R. A. Wiesemann J. M. Gallagher R. L. Witzke H. W. Graves W. E. Wright P. B. Haga J. M. Yadon D. Hunter Marketing (15)
A. J. Impink Library (3)
A. R. Jones
CONIT.NTS E.edt
. List of Figures 111 List of Tables iv Abstract y I. Introduction 1 II. Program &.unmary 3 III. Experimental Program Description 8 IV. Reevits of the Reactivity Follow Program 21 V. Chemical Surveillance Program 37 VI. Flux Distribution Studies 69 VII. Conclusions and Future Studies 82 6
-11 l
_ . . . - ______ _ __ u
List of Figures
_ Figure Title M
. 1 Saxton Depletion 27 2 Unexplained Reactivity History at Saxton D.tring Chem Shim 28 0
3 Variation of Solution pH at 25 0 with Alkali and Boric Acid 39 4 Variation of Solution pH at 275 C with Alkali and Boric Acid 41 5 The Coolant pH Effect in Barton with ~20% Nucleate Boilin6 54 6 The Coolant pH Effect in Baxton with No Nucleate Boiling 54 7 Change of Reactivity with Coolant pH at sarton - 20 wt -
No Boron 57 8 Comparison of Thermal and Nuclear Power Distribution -
Chem Shim Operation 73 9 Nuclear Axial Peaking Factor vs. Wi.thdrawal of Rod 5 -
Chem Shim Operation 76 10 Nuclear Hot Channel Factor gF vs. Withdrasal of Rod 5 -
Chem Shim Operation 77 11 Nuclear Hot Channel Factor F00 vs. Withdrawal of Rod 5 -
Chem Shim Operation 78 I
e 4
4 l
l l
l l
t i
l t
-iii-
List of Tables
." d Title M
. 1 Principal Characteristics of the Saxton Reactor Core 2 2 Typical Analysis of Crud From Saxton Primary Coolant Corrosion Products 45 3 Typical Analysis of Crud From Saxton Primary Coolant Tramp Elements 46 4 Comparison of Saxton Crud Composition With Stainless Steel 48 5 Efficiency of Ion Exchange Purification at Saxton 50 6 Crud Deposit Surface Concentrations of Saxton Fuel Rod 65 7 Chemical Composition of Saxton Puel Rod Deposits 66 8 Nuclear Hot Channe3 Pnctors Chem Shim Operation 71 9 Smary of Root Wran Squared Differences 75 10 Core Depletion as of April 20, 1964 Bo e
e I
a e
o
-iv-s
=
j
l l
Ar s*.r a rt l An extensive program to damonstrate the feasitility of operation with i
. chemical shim control has teen conductt.d at the Saxton reactor. Borie acid remained in the core continuously from my 23, 1963, to January 20, l
1964. After a brief period of non-borated operation, chemical shim operation l
l resumed on W r:n 6 and continued .scii Aptil 20 when the reactor was shutdown
- temporerily for replascment of cxperimental fuel rods. Tbtal average core
! burnup was then 6330 ME/E, of whicn 1.!21 K4D/MI'J vas incurred during chemical shim operation. Peak b.:zt.up vos 15.600 MWD /HN.
The experimental data obtaine.! do not indicate any significant boron j secumulation on the core surface at power or any decrease in core lifetime ,I under normal opereting conditions. The data also indicate that alkali addition for pH, control is satisfattorv it. both pure water and the chemical shim solut ion; moreover , the expeco-d hot char.nel factors wv achieved.
( ,
The future program vill test tM tehavior of de.fected Zircaloy rods, 1
provide a technique for detpting heavy cr.id deposito if thef are present, and evaluate the feasibility of chemical shim operation with bulk boiling in some channels.
l l
e
=V-
- ._ _ - - . . _ - . . - _ . . _ . - . _ _ - . ~ - , _ _ _ - - . ._-
Tna Saxton Chemical Snim Experiment I. Introduction The Saxton nuclear power facility is a closed-cycle pressarized waterreactorplanthavinganom.naloutputof20 megawatts (thermal).
Tue principal characteristics of the core are summarizsd in 7bble 1. The primary purposes of the plant are to produce engineering information from planned experiments and to test new methods of reactor operation with
('
pctektial appl $ cations to power plante. The experimental program is being carried cut jointly by the Westinghouse Atomic Power Division (WAPD) sud the Saxton Nuclear Experimental Corporation (SNEC), e subsidiary of .
General public Utilities Corporation.
The first phase of the experimental program at the Saxton reactor it being devoted to investigation of " chemical shim contral" using boric l
acid. 'I?o adjustment of moderator properties by addition of a soluble neutron poison for control of reactivity with lifetime leeds to significant eccnomic advantages when applied to clcsed cycle veter reactors. Use of I
"chenical shim control" not only reduces the number of control rods and control rod drives but allows full advantage to be taken of the lover hot r! .anel factors achievable with non-uniform radial fuel enrichment.* In viriv of these advantages, the large Westinghouse reactor plants now being designed vill use chemical shim control.
.
- P. Cohen and H. Graves, Nucleonics 22, No. 5, 75 (May 1964)
'Imble 1 Principal Cnaracteristics of the Saxton Reactor Core
_ Core N ecription Average core diameter (cold) 28.07 inches Actual core length (cold) 36.6 inches Number of fuel assemblies 21 UO2 in the core 2256lbs(measured)
Uranium enrichment 5 69%
Puel R3ds Pelletdiameter(cold) 357 in Puel tube O.D. 391 in Total number of rods 1676 Control Rods Number of movable rods 6 Control rod shape Offset cruciform Themal and Ib'draulic Design Data 6
Total coolant flov 2.8 x 10 lbs/hr Nominal operating pressure 2000 psi Average coolant temperature 5300F O
-2 l .
II. Trogram Summary (J. Weisman)
A. Objectives The 1asic objectives of the Saxton chemical shim control program have been the accumulation of reactor operating experience to demonstrate the feasibility of the concept and the determination that no operating problema exist. The major concern in the overation of a chemical shim plant was the possibility that boron might accumulate in some manner on the core surface and subsequently be released rapidly. A secondary concern was that boron accululation might occur in an irreversible manner (that is, not to be reduced as the boron concentration is reduced) and hence lead to a significant reduction in core lifetime.
Extensive out-of-pile tests have disclosed only two mechanisms by which significant boron accumulation could occur. The first possibility was indicated by the study of solutions during nucleate boiling on the surface of electrically heated rods. By means of a sodium tracer, it was detcrmined that, under nucleate boiling conditions in the presence of hkevy crud deposits, e significant concentration of the sodium occurred on the surface of the rod. The accumulation disappeared rapidly when boiling on the surface of the rods was terminated. This concentration effect was observed only in the presence of crud deposits far heavier than those expected in normal reactor operation.* If such a concentration effect had been
- WCAF-37?1 "Fediotracer Studies of Hideout et High Temperature & Fressure" (June 63) L. Picone, D. Whyte, and G. E. Taylor.
E e L.
obtained in the reactor with boric acid, this would have been observed i r as a reactivity loss; in perticular as an abrupt change in power i :
i coefficient at the onset of nucleate boiling. Buch a reactivity loss i vould have been reversed once the power ves reduced below the level I i h
- at which nucleate boiling occurred. 3 i
I It has been observed that crud levels in the water of pressurized ;
veter reactors are reduced when the main coolant is maintained at an alkaline pH. Since boric acid is essentie)1y un ionized at operating i
conditions, the benefits of high pH can be obtained by addition of very scall amounts of alkali to the borated coolant. Under nucleate boiling conditions, it is conceivable that deposition of alkali borate salta could occur on the surface of the fuel rods. There id out-of-pile evidence that if this were to occur with a lithium additive, re-nolution of the salt might not occur. The problem of f metaborate precipitation'ves one of those carefully studied during l
5 the experiment.
79 1 ,
A third mechanism, one of simple " exchange absorption" of borate by the corrosion product deposits on the surfaces of the fuel, is also known. The extent of boron accumulatica on the core is dependent on the coolant boric acid concentration and the amount of crud on the_ core. However, the small quantities of borate which can be absorbed make the process insignificant from a reactivity standpoint.
4 4
4 7
_ . . . - , . _ _ . - . . . ~ . . . - . , . . . - . _ , . - . - . _ _ . . . _ . - . _ . . _ ~ , _ . _ _ . , . _ . _ _ _ . _ _
B. Ihperimental program i Three major techniques beve been used to demonstrate the l
absence of boron accumalation. The first of these has been the careful comparison of the predi:ted core reactivity with that observed. Any boron accumulation on core surfaces vould have been observed as a decrease in reactivity beyond that expected.
Farticular attention was paid to the transition from nucleate boiling to non nucleate boiling conditions.
he second technique was concerned with the demonstration that alkali metaborate precipitation did not occur. Careful follow of the alkali to boron ratio was mnintained for ped.is during the tee The absence of a change in this ratio demonstrates the absen:e of metaborate precipitation.
The third technique used was that of hot cell examination of a test subassembly after apprec;able chemical shim operation.
The surfaces of th'e fue.1 rods were scraped and the deposits removed, veighed, and analyzed. The reactivity vorth of this deposit was then estimated.
I i
Prior to chemical shim operation, an extensive series of low power physics tests was conducted. These vere followed by a period of rodded operation at power. The tests provided the information required to conduct the reactivity follow during chemical shim.
5 s
)
Chemical shi's operation at Saxton began on May 27, 1963 After operation at 15 and 20 MWt, the reactor was brought to 23 5 MWt which is 3 5 MWt above the level for which the plant is rated under rodded operation. Boric acid remained in the core continuously until January 20, 1964, although power operation terminated on November 22. During this period, the reactor operated under a varietyofconditionsandanaverageburnupof3440MWdays/ tonne was accumulated during this period. (Tbtalaverageburnupatthe endoftheperiodwas4735MWdays/ tonne.) A test subassembly was then removed for examination. A brief period of non-borated oparation began on January 30 in order to provide additional information on the effect of burnup on reactivity in an unborated core. Chemical shim operation resumed on March 6 and continued until April 20. Total average core burnup vas then 6330 MWD / tonne, of which 4521 MWD / tonne var in:urred during chemical shim operation.
C. Results The test proL.im has successfully demonstrated the feasibility of chemical shim control. Na significant operating problems have been encountered. Under normal operating conditions, the data indicate that:
- 1) There is no significant accumulation of boron-containing material on the core surfaces during plant operation. Deviations from reactivity predictions vere well within the estimated error and
, no deleterious trends could be observed.
(
- 2) There is no decrease in core lifetime because of chemical shim operatien. Examination of the central subassembly shoved no significant accumulation of high cross section materials on the fuel surfaces.
- 3) pH control of the coolant can be accomplished successfully.
Chemistry studies indicated no problem with alkali stability in the coolant.
- 4) For the Saxton reactor, the expected hot channel factors were in reasonable agreement with calet lations.
- 5) Chemical shimjunder normal conditions, causes no hazardous .
situation to arise to affect plant operation.
t l
[
t l
(
7
III. 3rperimental Program Description (T. L. Erion - J. Weisman)
A. Reactivity Follov Ground bles he safeguards report for the chemical shim experiment considered the possibility that a boron-containing material i could be dislodged from the fuel surfaces and rapidly removed from the reactor core. Since definitive operating data showing that this vould not occur vere c ot then available, it was decided that reactor operation vould be conducted in a manner which would limit any possible release to a hamless value. It was conservatively postulated that a release of material from the ccre might take place over half the coolant transit time through the core. Based on this assumption, the magnitude of the reactivity release which could be tolerated under various operation conditions was computed. A limitation was then imposed that power operation vould continue only so long as the unexpected reactivity loss l vhich could be ascociated with any poison material on the core
. surfaces allowing for experimental uncertainties was below the appropriate limit.
In order to evaluate the extent of possible accumulation of poison material on the core, the predicted core reactivity was carefully compared with that observed. To accomplish the required reactivity accounting, initial conditions were established and the control rod positions measured. he initial point chosen
. was the hot, borated, zero power condition prior to operation with boric acid at power. The reactivity changes caused by
departures from the initial conditions are calculated using previously established values of the coefficients for power, temperature, prescure, boron concentration, control rods, pH, and the time dependent reactivity worths of uranium, plutonium, samarium, xenon, and other fission products. Itc predicted reactivity changes are used to compute the predicted positions as a function of time for the single controlling rod in the core.
As experimental data vere obtained, the observed control rod positions vere corrected to account for differences between the base conditions of temperature, power level, boron concentration, pressure, and pH used in making predictions and the actual conditions at the time of the observations. The discrepancy between predicted position of the control rod and observed position as corrected was taken by the operator as a measure of " unexplained" reactivity.
Any boron accumulation vould appear as an unexplained reactivity lose. ,
In determining the allovable unexplained reactivity loss, an estimate was made of the uncertainty in the reactivity prediction, caused by uncertainty in the reactivi'v parameters used, and the uncertainty associated with the actual measurement. The statistically combined uncertainties vere then subtracted from the reactivity
\
l release value which the transient studies had previously shown to l be safe. - The difference remaining,which was about 1/2 the computed
! - allowance,was taken as the maximum allovable unexplained loss. Tbis quantityvariedbetween03and0.4%Ak/kduringthetest.
l l l
1 l
t
Upon completion of a given Tortion of the exIeriment, a core reactivity predition was made based on the actual power conditions during ind prior to the run. The experimentally observed rod position, boron concentrations, coolant temperatures,
. and pover vere then used to compute the observed reactivity.
The difference between the obser.od and calculated reactivity is that which is reported in oubsequent sections of this report as the final value of the unexplained reactivity et the Eiven time in life. These numbers differ slightly from those used by the operator in that they take into account a) the reactivity variation of xenon and other isotopes induced by the power oscillation about the nominal power level, b) the variation in effective flux seen by the various nuclides with changes in rod position and burnup, and c) the small difference between i unexplained reactivity expressed in terms of predicted and observed reactivity changes and that in terms of predicted and observed rod positions.
I l
l
- 3. Test P*ogram
- 1. Pre-Chemical Shim Operation In order to pravi6s a firm basis for subsequent chemical shim reactivity fo11cv, a careful program of tests was carried out prior to chemical shtm operation. The j first phase of the program consisted of low power physics measurements. These included measurements of control l
rod worth, borno, and flow worth at ambient and operating
temperatures. hmperature and pressure coefficients were measured over a vide rang- of boron concentrations. The measurements covered a variety of conditions and vere far more comprehensive than vould have been required had an experimental program not been planned. ,.
After completion of the low power physics program, the reactor was brought to full power in a step-wise manner.
Transiant responses and power distributions were measured at
\ each successive power level. After conducting the transient ,
tests at full power of 20 Wb, the reactor was run at this full power for two 21-day periods. This provided the infor-mation N.nired to check the computations used for predicting reacti'^.4 changes caused by fuel burnup during chemical shim operat: .
At tne conclusion of the burnup studies, the reactor was shut down' cold in order to insert three test subassemblies.
A nine rod subassembly containing five thin-vslled stainless' steel -
clad fuel rods and four Zircaloy clad reds vas-inserted in the central test position. Four of the stainless fuel rods were ,
removable, i.e., they could be removed from the subassembly with-C out destmetion of the cen.. All rods contained fuel which produced power at the normal reactor power density-level. -The
.)
7 purpose of this subassembly was to allow crud buildup measure--
V ments to be made readily, furnish-a comparison of Zircaloy and
- h. -
stainless Steel clad behavior in a chemical shim environment, and to test the performance of elastically ,,'. lapsed cladding.
A second subassembly containing five thin-valled stainites-steel rods and four Zircaloy elements was inserted in a peripheral position. This subassembly was identical to the first subassembly except that the stainless steel rods were enriched to a slightly greater extent. It was planned to move this assembly subsequently to the central position. At thet time, the " spiked" rode vould produce if KW/ft.
A third subassembly containing a 1/2 in. I. D. thimble was inserted in another peripheral test position. This thimble was used for a small oscillating absorber. It was expected that comparison of the absorber position oscillations with the flux oscillations so produced could be ured for the determination of reactivity transfer coefficients. Unfortunately, the absorber jammed thortly after the reactor vent to power and the desired information was not obtained.
- 2. Step-Wise Approach to power with Onemical Shim l
l With the reactor at 5200F, 2100 pai, and zero-power, the l coolant was borated to a level of approxim.tely 1200 ppm B.
I The boron level was such that, when the critical control rod 1
i posit 10n was est;blished, all rods except one were withdrawn 1
fnun the core. The critical rod position was carefully mear. ired.
operation at 20 MWt, the plant was brought to 22 HWt. After re-maining at this power level for a day, during which time core flux maps vere taken to demonstrate that predicted hot channel factors were not exceeded, the plant was taken to 23 5 MWt on June 22.
Operation at the elevated power level of 23 5 MWt was
(
( especially significant. Under these conditions, calculations l predicted that a substantial fraction, approximately 16% of-the core surface area,was in the nucleate boiling region. This
. ras verified by the results of core noise analysis.* It vill be recalled that it had been thought that significant reversible boron accumulation might be possible under such nucleate boiling l conditions. The reactivity follow observation shoved no signi-ficant chsuge in unexplained reactivity during the continued f
operation at 23 5 MWt.
- 4. _ Examination of pH Effects.
It has been observed in other closed-cycle water reactors, Yankee and BR-3, that changes in the main coolant pH caused-
-observable changes in core reactivity. Although this phenomenon-is not vell understood, theories have been advanced that the reactivity changes may be related.to possible pH induced changes.
inthequantity, nature,thermalconductivitypropertiesand/or 9
- This work was partially supported under AEC Contract AT(30-1)-3E9 L . . _ _ _ _ _. _- _ _ - .
distribution of the crud on the fuel surfaces. In order to investigate this clfect under chemical shim conditions, several experiments were conducted where the alkali in the coolant was removed by demineralization and subsequently replaced.
The first of these tests was conducted under chemical shim J
conditions at a power level of 23 5 W. In later parts of the program, tests were conducted under non-nucleate boiling conditions and_, during a later phase of the program, under unborated condi-tions. The conditions under which tests were conducted were:
- 1) chemical shim - 23 5 Wt-530 F, 2000 psi -
substantial nucleate boiling
- 2) chemical shim - 15 Wt, 520 0F, 2100 psi -
no nucleate boiling
- 3) chemical shim - 0 Wt, 5200F, 2000 psi -
no power
. generation
- 4) unborated coolant - 20 Wt, 5200F, 2000 psi -
~
some nucleate boiling 5 Extended operation Under chemical shim Upon completion of the pH test at 23 5 wt, operation continued at this power level until Juy 10, when it was reduced to 20 Wt. The reactor was then shut down on July 15 for scheduled maintenance. After completion of the maintenance, the reactor was operated at 20 and 15 Wt through August and early September. During the 15 Wt operation, the previously mentioned pH test under non-nucleate boiling cooditicas.vas
performed. Operation at 2?,f MWt resumed September 12. Vith the exception of a brief sbutdown fror Scytember 30 to 0:tober 3, the reactor remained at tbst level until November 19 Daring the extended operation under nu:leate boiling con-ditions (at 23 5 MWt} careful atter_ tion was paid to the change in ,
reactivity with lifetime.
l No significent change 2:. the unexplained reactivity was observed under these conditions of bciling. This is considered strong evidence that buildtp of poisen on the core surface did not occur as a result of boilir>g.
- 6. Test Assembly Insertion and Lov Fover Ihystes Measurements In the shutdovr. following tre extended operation, the test l
subassemblier inserted prior to chami:al shim were removed. The I
l rod oscillator assembly co .taiting the jammed Saa 12atcr vae
- replaced by a unit
- onta.tr.ing c redesigned osciliator. The new subascembly eleo conteinaa 6 tellow tubes irto which direct reading l
! nuclear detectors could be incerted.
l The subassembly in the core periphery which contained Zircaloy and dain clad statnieer fuel rods var removed from the l core, exsmined with tre peric. ope, and placed in the fuci storage rack. The peris opic exaninstior. of thic assembly indicated it to be essentially free of crud deposits. This unit vas replaced by one of the normal eubassemblies which had been in tre core at the startup.
The central subassembly containing Ziraaloy and thin clad l -
stainless rods, was s.lso exer.ined periscopically. These fuel rods l
- - .. 4. w- . - , - , - _.. -- , ,
showed what appeared to be a thin deposit of reddish-brown crud over a portion of the rod. The deposit covered the greatest area in the high flux region. Itwasestimatedthatonlyabout10-15%
of the rod surface area had any deposit. After this visual examination, the assembly was placed in a shipping container and brought to the Westinghouse Waltz Mill site. After appropriate cooling, the unit was moved from the shipping cask to the hot cells, There the subassembly was dismantled and the fuel rods carefully.
examined. The examinatior. included measurement of the area of the deposits, scraping selected portions of the rod to determine the total weight of the deposit, and chemical analysis of the deposit.
The central subassembly was replaced by a new test unit. This unit contained five Zircaloy clad fuel rods and two stainless, steel clad rode. Ir. addition, it had two rods which simulated absorber rods of the rod cluster control (RCO) scheme. These ROC rods were contained within a guide thimble separated from the rod by a narrow annulus. The rode contained depleted uranium so that the heat generation of absorber material could be simulated. The purpose of this assembly was to' determine vnether, under static conditions, crud buildup in a reactor vould cause a significant increase in the force required : vithdraw the rod from the guide tilbe. A.
, special fixture was provided so that the withdrawal forces could be measured after the assembly was removed from the reactor.
Critical; positions were carefully _ determined before Lxi after the subassembly insertion. ~Following the critical measurements, the previously mentioned pH test at zero power was performed.
17
7 Non-Borrated Operation On completion of the shutdown work, the reactor was brought to temperature and deborated. Careful reactivity follov was maintained during the final stages of the debora-
. tion, which was accompl'ished by ion exchange. After the boric acid had been essentially all removed, 75 ppm of boron were added to the system. Core reactivity was again followed. If there had been a reversible deposit of boron on the core surfaces, the change in reactivity during that dilution, caused both by removal of boron from the main coolant and by the release of boron from the core surface, vould have been greater than that during the second dilution, when only removal of boron from the main coolant was involved.
No such effect ves observed.
. The reactor was then brought to power and rod oscillator experiments were conducted at both 15 and 20 MWt. The rer.ults of these tests were subsequently compared with the results of similar tests under chemical shim conditions.
The transfer functions determined by these testa shoved no evidence of any power reversible boron hideout at the test conditions.*
4
- WCAP-2627 "Saxton Kinetic Experiments" V. Rajagopal (June 1%4).
ThisworkwaspartiallysupportedunderAECContractAT(30-1)-3269 k ,
s .
s At the conclusion of the oscillator tests, the pH test with an unborated coolant was conducted. When the alkali vao restored, lithium was added instead of potassitn. Operation at 20 W t continued for approximately 10 additional days. 'lhis period was used to provide a recheck of the burnup rates under conditions where they could not be obscured by chemical shin effects.
- 8. Renewed Chemical Shim Operation - Lithium Stability Tests Prior to these tests, the hot cell examination of the subassembly which had been in the central core position was completed. From the information so obtained, it was possible to estimate the reactivity worth of tne deposit on the core surfaces. Conservative calculations indicated this to have a reactivity worth of no more than .Oly,4 k/k.
It was therefore concluded that there was no significant neutron absorbing deposit on the core at the end of the non-
, borated operation. Accordingly, the hot borated critical measured at the beginning of the renewed chemical operation '
was taken as a new base point. To this-point was assigned a mini ==1 value of unexplained reactivity lose corresponding to a conservative calculation of the vorth of the crud deposit.
Subsequent reactivity computations are all referred to this new base point.
i 9 !
The reactor was brought to 23 5 MA vith ir.termediate operntion 15 and 20 MWt for rod oscillator measurements. No change in the -
unexplained reactivity was otserved as the power was increased from 15 to 23 5 KA.
This period, March 6 through 30 of 1964, while the reactor operated with 560 - 660 ppm of boron in the coolant, was elso used to demonstrate the stability of LiOH as a pH control agent.
During this period, the lithium level was of the order of 0.4 - 0 5 ppm. The lithium concentration gradually increased because of the B10 (n, c) Li7 reaction. The lithium growth in the coolant followed predictions based on data obtained during May and June of 1963 If metaborate precipitation had occurred, a change in the Browth rate should have been observed.
The lithium concentratica in the coolant was then increased to approximate (v 1 7 ppm. This level was maintained from April 1 to April 7 vhile careful chemical surveillance was maintained.
Again the lithium growth in the coolant followed predictions.
Plant operation continued until April 20 at which time an extensive shutdown for the replacement of a number of experimental subassemblies was begun.
3
.[
\ . _ - - _ - _ _ _ _
IV. Results of the Reactivity Follev Program A. J . Impink , R . W. C olomb o , F . J . - Fr ank A. Calculation Procedures
- 1. Reactivity Parameters Throughout the chemical shim prograa at Saxton, an attempt has been made to predict and to follow reactivity changes in the reactor as closely as possible. To this end n rather comprehensive set of reacter physics measure-I ments was carried out, both at zero power and at full design power, before operations at power with boron in the main coolant were undertaken. Analysis of data obtained during the preliminary experimental program yielded values of the reactivity kenetics coefficients of temperature, pressure, and power. The reactivity worths of xenon and of control rods and boron in solution also were obtained. A later series of measurements, carried out during chemical shim operation and subsequently verified during rod-controlled,_
zero-boron operation, established the value of the reactivity coefficient of moderator alkalinity level.
In constructing the reactivity balance during chemical l shim operation, primary emphasis was placed on these experimen-tally detemined reactivity coefficient values , as opposed to values obtained by analytical predictions, for several reasons.
l The general premise was established that measured values more
, faithfully represent the net result, in terms of reactivity,
of changes in the physical characteristics of the system that do values extracted from studies based on idealized analytical models of the system. Of greater influence in favoring the experimental values was the recognition that
. deviations from predicted behavior could be detected only through measurements made during (,perations. If systematic errors are to be avoided in comparing predictions with measure-ments, the predictions vould need to be based on data obtained in the same manner as' the data in which subsequent evaluations .
of behavior were to be based.
Wherever practical, the experimental data vere compared with the results of previour theoretical studies to insure that the reactivity ccefficients derived were realistic. It was s observed, in general, that the experimental best values agreed
. reasonably well with theoretical predictions.
The one $ea:tivity effect that cculd not be measured ade-quately and therefore- had te be based on the:retical calcula-tions was that caused by burnup. Although experimental data vere used during the earliest part of the program, as soon as
-burnup contributed a bignificant component to the reactivity balance, use was made et an analytical, malti-dimensional burnup model. The model depended on experimentally measured flux distributions to ' predict reactivity changes resulting from -
long-term burnup and from spatial transient effects. Burnup 22 i . .. _ . - . _ . . _ _ _ _ . _ _ . _ _ _ . _ _ _ . _ _ . _ _ _ . _ _ _ _ . _ _ . . . _ _ . _ _ . _ . - - - - - . . . . - - . . _ _ .
rate meaburemente were made periodically both prior to and during chemical shim operaticns. Comparison of the theoretical predictions with the results of these meas'irements under speci-fled conditions permitted verification of the model.
- 2. Computational Techniqueb t
The Saxton reactivity baltnce was based on the so-called
" differential" model. The method allows detection of anomolous i
! reactivity behavior by comparing measured and predicted reac-tivity changes over a series of successive short time intervals.
By way of contrast, the alternate " integral" model reintes all reactivity changes directly to a base point which may be rather remote in time. Although the tvo methods may seem to differ significantly, the only fundamental difference is in the method of establishing the reactivity equivalent of control rod motion at various power levels.
A predicted etange in reactivity embodying the sum of the 1
reactivity equivalents of all changes in ccre conditions (tem-l
! perature, pressure , power level, boren and xenon concentrations ,
pH level, and burnup) is ccmpared with the reactivity equivalent of the observed control rod motion required tc compensate for the change. The difference between the predicted change in reac-tivity and the measured change is then, by definition, an incre-mental change in unexplained reactivity. The total unexplained reactivity at any time is the aggregate of all incremental che.nges l -
in unexplained reactivity which have occurred in successive time intervals since the reference starting point. In part!-
cular, borte deposit on core surfaces, which is not accounted for in the reactivity balance, venld be indicat ed by an unex-plained reactivity loss. Since unexplained reactivity is ex-pressed in terms of a readily observed physical parameter, e.g. control rod position, it is possible for reactor operations personnel to be aware at all times of the status of the reactor with respect to shutdown restrictions.
A gradual increase in the degree of sophistica. t of the reactivity balance calculations took place durinE the initial phase of chemical shtm operation. During the initial power operations the balance calculations were carried out entirely by hand at the site, both by WAPD reactor physics personnel and by SNEC operations personnel. During this period a rather simple version of the burnup calculation, based largely on earlier expertmental results , was in use. As operations pro-gressed and burnup became more significant a more detailed theoretical method was adopted for burnup reactivity evaluation.
Finally, when the basic models had been shown to be adequate for maintaining a close reactivity balance, an IBM 709h computer code (SCOOP) was prepared to expedite carrying out the prediction e.nd follow calculations. Incorporated in the code was a refine-
, ment of the burnup calculation technique which gave a better
-ek-5
i l
representation of spatial effects. Otherwise the code repro-duc ad the computations which earlier had been carried out by bend. Provision was made for direct communication between )
the Saxton site and the East Pittsburgh computer installations so that WAPD personnel at the site could obtain prediction and follow information directly and speedily.
The repeated analysis af earlier data incorporating improved computation procedures and particularly the refine-ment of the burra; calculation has resulted in a series of progressively more detailed evaluations of the behavior of unexplained reactivity during the first phase of chemical shim operation at Sarton. Minor improvement of methods and elimination of known deficiencies still continue.
The results of the reactivity follow presented below are those obtained from the most recent analysis of the experi-mental data. ' At the time that these calculations were made, several deficiencies in the SCOOP code and its library had been identified but had not yet been eliminated. The actual data plotted represent computer output corrected by hand cal-culations to compensate for systematic errors introduced by the then current version of the code.
B. Follow Results
- 1. Observed Reactivity Behavior Thc reactivity behavior during chemical shim operations
-e5-i
- - . . . - . - ._ - . . . - - . . _ ~ . . - . - - . - - . - - - . ~ - -- .- - .-
i at Ssxton may be analyzed both macroscopically and micro-scopically. Macroscopically, one may consider the long term variation of core reactivity as a function of burnup
+
under otherwise constant cond.itions. Figure 1 shows the
, excess reactivity available in the core at various stages of burnup over a period extending from beginning of life to well after the first phase of chemical shim operation.
In this case the remaining excess reactivity was calculated from the results of a series of periodic boron concentration measurements made at successive stages of burnup when a pre-scribed set of physical conditions had been established in the core. For comparison,the equivalent of the maximum allovable unexplained loss in reactivity is'also shown.
It is apparent that no significant change in core reactivity l
l vas observed either at the start' or at the termination of the first phase of chemical shim operation.
The microscopic behavior of the unexplained reactivity is shown in Figure 2. In this figure core conditions (boren concentration, pH level, power level. and control rod position) are pletted as functions of calendar time. The interval ~
begins with the start of power operations _vith boron in the-main coclant in May 1963 and continues to the shutdown and termination of the first part of the chemical shim test _in November 1963 Superimposed on the history of changes in core
.ap . e a m .&M--,% . w ka.. . _e. ._ ._s._w e. - -a _
9 4 9
. T_
00 O et m
- ~
y 5_N E m
~ - -
=, .
ED o C Sb U
= . - o 18
~
E8 om 5 m s
~
W q -
@ D S S z 5 5 w a z9
=
s s
8 8- w z '
- * - z 8 _.
E E ~Sz2~ o u.
opy < <
o m 0
q
~
^
- n <
3:
5 - <
55 ,8 4<* ap E
< 50 - m 4 u - ~
O
@ y N -
o * * ~ e m o
, ,m , m _
O 8 0 8
3p 39 h M 1
oggj J Plo)zy 3 s 2-(.
t l
l
CONTROL R00 2 =
CONTROL R0D b T. - i-l . 41 -. ;
- .l,-ltj
.s
- j- ..
4 .T_,VF --.-+ m.
). ---t-.
p=- --
4
..p
.t y-
- x-cl. *
. } ;:
- 1
'y .
i +
at p
j l-
'l~
t q
t_
t.
+
- L:
- .l1.i1-r p - J. + : a. p _.+=s? :=>+
4 ts 4 ._4 :g :L +
-:p !._ ..-..
@q ._ .ggE
[ d:p.hr.F h:{.i- 2J i' . ;lh' . .Q . E.-f -
-. .' '. gb , .
-l 4 4HmM0i@h w niiME e '
4 r@! 2.,
~te .
g" :
=
- 4. t 1 ib @hiiGWiWlif 'iiim :Hninui!- W c n# de -
min F ~ i= = ~
v A H1
~~
9 vy; ~_;.:;; g: a g mp: my 9
- uu y y g g .=.
g.
? .
g g, ,
, , , ,. 4 .. .
3
- 3....- +,,.
l ll 4
1 -
H l "! d, M
m -
j@&.y v . . -
' ii w a= e p liii M iU HH fie
=y f5
- o.
y y#
!!i !
@yu ln #
.., 3 .,
.. .r .
- k. f.
S !NN L; ' i"
~
i U hii dl "D J' !!F -iF 4!. ..-
M i-- +
- {[. .h h e s m, e wnu w w. q a w w g ; F""g. Q. .m, :
g g W ..3..l.
et 1.. .
.+ 3 M I
'O h.: fLI
^
7 : ::
- -' - i gE ... . .% ....
- .p.x 4 ;
I
._kJ % p.g'I~
, - ~ i. _ a ..
- rg
~ ~-
ra . . ; .
Q[=l c[r -
a: _
. - p_ - a
,- WP bF. h
.p .- y.=
= z
.TD 7
= z # 4 ..p 4 ,$
f_
M;; :g c-+1 q!
, - 1..r.pt -. . - p- .. . . .
m_. -
r
. j :p.
.t' [ , .
.}. _
w p m, .- .
- .x. -
~_El_ J__ l ' 7 I"~ 'b .p j_- -
A .l [ l m a-1 l ! F; e + a p 4 .-
y; p -g j + ~F , . I.; ,h.
1 -
.m 1+I _L.h % F = f .y . .y r , j 9
_4 o g_
i H + .j rp j. .>:' ! _2
- j 1
4 4 '; -4 N- -.
. ~ [_ .a j..:.
5
!- :- i -j. . p. ~~1 4-- .._m. :I
- 4+
- e$ - -
- F .1';. ..-
= g- .q- .p -
g- :q:
- t. . g 1 t .,
_ y n 2. 7
,,,,,,,,, .g . g.g ; ..,,3
.I
- i.
.j .;._
j h;;i h 7
p L ..:- ,
1
~
t
\
~
P ".3o; =_ _ . q. ;
3 pr. . -. 7, w.
. p- --
w- .3 = % g
'" " 3-
- y
~
' ~
5 9": ff: E
U
- _ un ,.. . _
p . _
q i ~!U
- r ;,;,,
( ,,
'. p t .'- ,
5 . r.q n ..
- g. n_
I, ,> ----- =. = ;;. u y =
- =". .w -
kl.Y f.
^ ' '
[ .}- 4 4 :: q.._
~
T '
E' "
-i m
m
%. #. %g--
g L2 . :
4 .
- = 4 .a
'n
.== :_
g p_ t i 1; ar :- -
y, i
. bdji 7
I i ;
p : yl egg. - t--,,y . . .
-- y.._ ;. , . _ . . =, , , .= ,
i~
j
. lE -
t ;
+
~"
,n b O [2.; d.i
~ '"
r"
'E
- [. i. 4
.y 7
p i
F
.j: m
=l q
- L
,-p.
% g 3.;; q q.
- j g ,
. .." ' ~ - =+=
j.
9 f. .
t " ;TI' #i d .. L. i .a i'f-1 ,
_. -i d.. .
, .1 - .{: . g ,. . - ,
sp. q t ' i. ~ ) .
. ci ph:_ Q.. _
- 4; . ;. . j .
- .
3 7 ,f" g
I I l '! -1 l 7 '
- p. g;3 I
~
d,
- }: CD
~
7[3 i-i ;
] - .;2
.c - -
+
-m
= ._
4-s gg
_ 1 .
y _: ,-
u 1; :
4 -4 4 .
I L
'f: I. l .Tg 2 :. r
.q.: , I -.
s .- Mm '
-- -- I 'I-
- 1 ]1 , -
, ,- -i ~ ,. -l l d c-! .i F j ;g (; a" '
%. ~ *i 4: :h j- r n ur , ,c. g
- [L ,
j j; ,
a.
' , o . _
-p :
j- ;
c i $. ..a - .
I -
.h -- t r --_ -*l'. -
i p u_;._. ; .
n ir3 b q; ; ;; 3 p 7-. .,
L '-
+*- i- . :.La M.i,* : --
N W. j.g .;;- @ ; !
n; u sini- :ibs . iae = +;s .u.: . mp. .mm z
.=. -
e- c l- :1.
N:ENE h' +
~-h hkh
'~
=n : . . . .
M -
-N N,H :i Nba:
d iii sC?i niil;r "2- ~Lin ".
@~!~Iih .F a
d:' dji
.p ["".:
.1
=.+a. . _ -mi. .5:,i; .- :+fi y-g ggE hiihr-3 a; g
~ * -... _
l_q .;g %! g
'; ==
Li,
.a .
- gj. i;j.p. -s s g;; ..
q- --n -
. a._;+g+
rw . . . .
az.
~.
z O m.L
=ge g_.- -.
j_ "'"
m 3 m
.i -
! ..? 22Fi: "
y; ;. m- :. ppd; g;ga; dj; ~.gha. _. gg :. _u,,, .~ M:.
c i- P. . ; ...u-nifi-i = -
.:d.:n .. 7 ;ts . -- -~ ~
r = = im, -E mi h- 2.1- i.;ii$r: i:1
. :{ i U 5II 5 ih
..i -
_ [.
- .;:--- : -+ =r .: n , :, .79 it tN. F'.- I h:2l"
. 2. lN m.1 . .~.;.m..
, , ,. . . , . ,,g,7 ,.,* ..
N ' ' "I .;.' -, I[
~
- :' :n- . ,;1 - *"
'^
4. } -ha ' . ' . , ' . _' , ' ..'
{~~. ~
- 2. --
-..L , , ,
=Q"* ; . .,:..
_ g . . . . . , , .
a.
~~ n ...r r-1p : i-+ :E- i -522. n : - .
2 "- -
_+_
~." =. =" , _ -- ..- S. .. p. 1,.. -
2
,]=-. . _ . . . .- - . . .
y .. . [ ._,. ge3; _g u n-a.
.Ng '.7,q .
b -i 4 y=n=f q 2
e,.
.=t= .a
= =m y = % g. w 3 ,. g .. .. ._ .g g _
__d. ___ J -. . _ , __ . . .
4 .e M -
e m'
3 .
1 i. l.c 1I J .p 1 ,: _ .
.J e e.{ j j o, . . =v =
l t y r r-
'7
., . y_
q j-r 4.{.
. p. .
. .h _
.i.
}- :l l .
-t-
.j e.
7_ .
.i
! i re . 3 l -l -l. p j ;:
.-j y. i.
i j t i i { , j
,::p.:-
+
r .,-
a;. . ,
,t l
4 ._.; ;
7, F
-l 3r 3 p .. m .. i - - -
-... ,i u -
-i'
-g . ; ; m *
.l gl _,
i'
[" ~I- !I _. b._ L_.._ N I' Id 'L_ E n = j. i -
1 ;F L_
VN: q i
~
w 1.1 L ..
,-! a _L
.. p m
.i p.t g.. i.
, =
s ..
__ -1 .l.. -
~l_- ca. ci - l> a-g; p p~ .
i'
]. r_ -
.. } .j -h L j k~'h . F .. p-
~~
., y }. y; 1.< :I Wi Vi:: O [8 *:-r-m i
. .- @ p' F t~ d
!- J pi y;4:.. 4'
~
- u. i ge .
~W ' -
~'- rt .1 .-I2:r -
.:rt.. ;-- - - ~o . . . .
a.
a g..m
- v. .
~ g*
.~
CL a Air 4 := . - 5 7 7 _
.p .
4 a t c : ,
m -
p , e p -
. = ;L -
7 m..
7 V
-,. Un , :qiT -
1 -
- b r -
- l- p -
= .
+ un .J
.1<.,
. a , .
E,_$ .
4 -: g, 2 i zu
_ -i
- :C .,h ; d[' i f j.q ' j.:_ '
. . f /. .'l::! -d- ,, d -
.j- _t .,_. i' '
_ h "- f f _
- f .'
['
'l.,
'. l i, I '"' ' "
!!'I
- j j +- + p .y a- - -
y -
+ .; .t . i . Ife 'i -
.l. :p };f l- -l . p,_ h.
l ,
.p +l q I' .! 4-1 - l- .q7. l, f
i :
_,;_4 a- -1. -
up
- i
~3 z j-j-
l i
,i .
,j.-
m a_,4 . ! ., ; q ]
g p"
_ .~ 7.
c (, =' ' '24. . .
-j -
- u. = - .. - -
- m 3;- s in =c y - i
- i
.t g q .# u n.
_. . , 4
_4 .,. . . -
- c. . . . . ... . . . . .
El- ;M -
gi 1 .
- . iii
^
.i L . N x- ... . ~ - -
- j -j !
& ~i! pl t .-._a l-- - --
- l
- _
} :.
tp _ m: 4: 1 -
- ~c + -
,m . L b
'j "
} .. . .I Nk , [.1 }
' ^
e* ~
( +_ - +-- . u. m :.G :.- _ a; % . , .
m .g :
r.i: g s :y.c -
- d. .. ,- 4 +; ;ip . r .ie & 3 .E:7 a -
sil :ia = a.iFli: ;iiipi as e 2 -
4- I" p
rp i
1 ef. '
= m : $ , h= @ -
d:P xi ! aH - .iij di *+ nl '5 h V id 9 Elp l 4 :b :"i ~ Oi: P .
( d l' F:
2.Eh::- . +: "
i f-N : dli :-h = h' : Hiit r Ni# - Mi "
i J. g, .r 4 ., .
m.e u y .m. = .g .g ,
e
, 4,. - 3+ .=m 4
.' 3 ... . .
n .. -
5tr ihU i' I .
h 5 fE -.4 4 .n. l d i[:5h,,7. '
e ~'
56 _ p 1 ,hb s a = g yir ;=i .=. =i
![,i :
.. e w we m -
=u :
- n- : 'ilp p.
-il p ;+ . q .ij. iF Ji jii g. -
.n .n- "-
_ 3_
8
. az
.A
- g. .mj'i? :i% *i n= ; *W .g +: e n .
e * @ ii e;; 4 ad
.a ,. O.a -
q in "F'U-i !-
- '~ ' :.4: 'I. Ihd:d ,N ' *h" % *. i55. . 5; *
'7 p.' [] . L 'E3 j."e[n: -d' h . J: Ig h : M' -
- L, '
- i i* L!'} ,
if 'b'k '
20- .}* , .h i .h. .
.ki {
k h.} ' : j. b
- / ,-4 !
W_ .I e -
- = 9;; w-@ cp .
=1.7 :si+ :
u; ;
un gi rih nqs: 5. = w riT gi. a= w, ai- ..
'n :
Q E E IO Q I
~
% Ji: IE ;
- E' 23 + ii;tU 5 E L .M ' MI
~'
'i"j -Hfri!; .j Ell := is H!i :3-en ii? e # ::9.i i
- i- ' '
hi 2Hi ;j ; q i !in :Ei 'iii. . ci::i; p: -. --i J -
- Hi ?? .ti! -
~
'ai ~
- ' ":2Edi
~
'ih@ih in' i O [Fi'in.@i "- 7 -iiid.!s id IE !!E n id Msf : fi;i '
ri: * :'i: ig
(
! [ ; i i di u-
?? .ii-lili j' W i hr n: di :ni.ii ;ni iiE;- -i!! l i_ ii) d p; j:h i j 4*. ,, : ri .
~_Mi eu 24iy gp ;i :
f.'
"Ei :. qg i:+ _
iin dig : zum s ngp s i y :iii is iij[ii, eh = :u. -
s_ .a X
?.3 h_
sh @E E lh;f 5 f];
=- 2d E
- : :.".. 'c.7?
f h3: Mit jg I [fof F .? hp-33 j k. d -
I LT :
di Il h +.i t
- -i Sh : ii : 4 iin ti] c p _ 39 -@g ;ipi.jd spii p:p+ iis ipi h"jjjm j 3:. dp -tj:f .r -iii
- hi un .
eli :m pq ' ih ";i ; up e agi lii; :g;pi; , . dii% -
lg -
Hi -
- FH in i nu w -
~
- cu- nii ; .Ib 9 '
dijg i;.: y . Ei - ih
. g pli "-
llii pii :E'jih i i.2i Si .: f :" . .i .:
Ri
- :m.7;29 .Ei :[a- .di 7p ui. du if H 5:n: i Fjiu.' HE iEi .jyi. d!i jg ;p ; igg;; .!yit! ::i;iy g:.7 g [m idj
=. ! "n . : 8 g]M:
p a -2; i --d-@ 5
$- p 1- i
_-$. Sg W-i'- - - - i
+. a= .-
vm. -
a nun =. ww w .m , ,-e sp= s a;_._x . .
wm -
m -
- ~
... 'INir i7 :
LnF -/.i 'ii: SIE: 3 5:lii4pa!Ni i iM5'E d' E -
55M2 :iii 5 4Ni tie' b i- -
'iii
- Mi . ..
- m. -
"i.JjdpL -
iini ilh;n ;.n:( Ehg ph 151 -i hi if liii in :=. ,;. iiii :ji r - jiEhp -dij :E-- g:; ;. E. ; Hi!
iqi. an ..:n $ips 9i; .di iiii
- c: :F. i;;; ;;3.sfi!W :g:-i dd t.; g ii9 : djg
- {2 ? ; 3 g Mii- -
h ::h.:h: p =. E. Sii ji$Qji[i iu; _
!!i j ;. dii ,n@p@.g ; "" ; un"" 3j? ;. gp +
Cn 'lii.i 4;p i i -ij a.; iiijiii: ihiji@
-+w.g wg as :
sy wg "=,
m4; : ap E is :p[l 3!p f {; 9. ..- :iii !;=
'.Hp= M'.iE g h= g j@j g 3.i E ;iii ijpi 5 _ i Eii 2;- 11:4 E : i-i ji 5 i 'r' y # a~ a i"i
- -4 ==9 .
'iE i-ESM U= #iiMFijjjm !=; s Ee E n= . = M =
ifdi s = .
- ~" =
,4 ep= d.w .
"ir P"f.h, -m *
- E eE i
- F inF=i mi1 EP -
R " =:~ -
M h=iE = =r 5-= 4 =
t.p. . .n. . 2p=;.=mp + ri u.ti= v,= . p. .i e u . 4. _= p.
4
== g.. =_ <. =..
.= 3 : = iA 9 .i.e .=p.+ .a = . .& , =.a
- 4. 9
....__..._.m . - . m. . _ . . . , . . _ . . _ - - . . m _.._m ..m . . . ~ . . ...m__.. _ _ _ . _ - . . . . . ~ . . . - .
I i _
t i 0, r.- - ;
- 1 e 4 J-t- i
' ~
...--.;~.._.j 71[
'*{
t ..
g- -i .
,,.g_. ..). .. . .
.y -
- l' '
l .. . . .b._ ,',g_
I 'j.. .., h 4. 1.h . ~ I. -
. ' .h. .
7 _+ __ _; . g, ,,
m.
p ! ~h !6
,_i --!' t.~;ln j - " ' '
~ T l-d r p. 4 2.q:1 l &P -
+944 -
t+p! %- -
.g? d-3
-.--a. i ;. . .. -
i '! , !- !
_ iq !- _ l-
.). .j_ ..q g E 1 g i j- ;..p .4 L;.L.- ;4 :1_. _L. _Q_
I j
r t- d I
- } '
. $j. l8
, , g. -t W b
- . . _ . . _ . . _ . . . . , , . . . . . ,. _ , {, _.,; .. , , , , , . _
f,' , __, __. _,
Q, .
-- H_ .
=
e e g
,E ._h. . - J. ...f.. ._l ~ _ . _. J _.1 n le J !
' i /, H' 4-. t -i i
- 1. :I k r ~- ! - '- 1
,j
},__ .7]- 1 1
. 4"q>
.U
} J- -
.ci.. _ j__{ 1 g {~E i
2,_4 j.
.h __
p L..__ p_ _j_1.p ;J. _L. j_1
- - g> .
i 4 a ,- .-
g . _ __.
_ ;.r t py .ps-J1.
hh h J 1, n
~
> l---- _2 h-"-- g'* "-
t---
-t I
-.-4-. -
[ ;-.[ ; -
- c y-- i-
.)= .- "-I^
_, m 3
. r m'1.y g 1
I~! (. -j-
!~ h sjn *? h.ld_.. :
}
l.
l 3
.ik] _
g .:. i-E I: .'> 6g..
' h:__b 9
h ';.1
- v.- -
ar e-s , 'y -- .
- h
- 23
, l' ,f9 I j. f.. ! !- d ., . i.
~
f :
.j t - 6""~ ~ " - '"i _ E" -
- j. .In..
g .-.}~ . J.-..T...-~"
. -"-'.)... ']. . .,
- j. ..a..-
-g ,
.- -t 77 u.-,-.,,
.j- V -
- ., c
'.f..
7
.(. ,
La u....
(. . 4 .
i-
.n L t-1 i 1: .
-3 i .l:
.q..:_ a 4- -e -
- 7 ._N 3.. 4
--6p...e.
~~;- e -> _e .ia .z - p- .t 4-
,- -m-- :
i 4
er--j- z L
E.
J
--I g .-
i
-. j
- a. _ .y
. l z.jyM 3 .. j ,
. i ;4 -
_; .. l
.g na+
'I. .. 4- -
eg .: , , 4,_ jg;;-- =;= =j, p' n~__,-v.. * .f~
~
-tJV - ; .T' TWpq q- ,.
? q= Y. : ~
J "'-k'.N. .... . .
=
+. ;
j- .':
.i, . k. .: . h.,. , , . . ~, b k,-hJ. .i.t. N :
n
. . , .-_. . y .
.J .
i 4 _p i g
- Q .
[, pp+ ,
yL m y- t-g.h.
,i i .
- i. ...... . z....
S+ M.M+ nim.h-
'TTi.n - d. =:L %. 5 if.
mi.;L
['n
,;#.. ; j
[i --
, -- e
!: 1.ki,T,DM. .; g_ MM@i -
-m
. . p!. 5.-F .
- - rrh 1 p: ..
2;j _L F. .: c .. 19. g% :;p . . , . . . . i O. :i. 1. "C~
..- t F, -i: -M-d---
..l.-: sngeM4. .
e- --e. : , , . . . . . . ..
p ..;:-n; p. p. , :- u n}' .g' .-
u;q ----t---
t -
[l gp ..
]. .b ...g -..nd a!. 4 q; l
f -CONTROL R0D 2 i
CONTROL R00 5 l l l I I I l l l l l l l l l l l l l l l l
=
~ ~ ~ 00m QmN w - _ _ - _ m m m ~ m m m m h m h s@= m . m Q N & M Os c N & m C0 C N &
m CONTROL R00 POSITION (INCHES AB0VE BOTTOM 0F CORE) l l l l l
l l l
l I l l l l l
.n m g= . . _ _ _
y a en a w w - o - N o o o anen o - m= -
cm es o ca UNEXPLAINED REACTIVITY BORON CONCENTRATION (IN UNITS OF 0.10/0.so) (PARTS PER MILLION) l I I I I e I I I I I -
1
\ , - - m m m m m m m m m x
& m ao o m & y u y , y a - -
THERMAL POWER LEVEL MODERATOR pH LEVEL (MEG AWATTS)
- - . - _ .. - = - - - - . _ .. .- . - _ . - - .. - . . .. - . .- - . - - . -.
conditions is the corresponding history of the variations in unexplained reactivity, again defined as the accumulated difference between predicted and measured reactivity changes.
s Included also are the maximu.s allowable and shutdown limits imposed on unexplained reactivity loss. The shutdown limit, which is the controlling fac. tor during power operatione, is defined as the maximum allowable unexplained loss in reactivity reduced by the estimated maximum combined uncertainties in the reactivity equivalents of the core parameters and represents the maximum reactivity loss pe mitted during operation.
Examination of the detailed behavior of unexplained reac-tivity as plotted in Figure 2 shove tha^ the unexplained reac-tivity tends to fluctuate within a band width of about 0.2%
and that superimposed on the fluctuations are two periods of relatively rapid loss of about 0.1% in reactivity in late June and early October, respectively. Although it has not been positively demonstrated, it is strongly suspected that the-small fluctuations are, in general, systematic in nature and arise as a consequence of the method used to synthesize spatial.ef-fects in calculating burnup. It has been noted that there is
- a strong correlation- between the fluctuations and the corres-ponding central rod position with respect to a series of pre-scribed reference-positions. -
The two relatively rapid losses of reactivity occurred '
t .
I I
s
, . . , - - -w- me-. ,,:~, - . , . , , . - -, w., ..-.n ,,-n.. , , .,,n..._..v. .-,.,s..., ,-,,,,,-nn~ ~,e_ . , _ . -
1 g I during operations at about 23 MWt under generally stable conditions. Cicse chemistry follev both prior to and bemediately after each cf these periods showed no evidence of abnormal conditions in tne system. Nor does any model
. Ea yet hypothesized for a possible mechanism for accumulation of boron on the cere surface efter an explanation for the reactivity loss in terms of boron deposition. Since similar behavior of reactivity was observed in the Yankee reactor when operating at povsr without bcrcn in the main coolant',
it is believed that the abnormal behavior during these two periods may be the result of changes in the physical charac-1 teristics of the fuel (pcesibly pellet cracking, for example) coupled with attendant cnanges in the flux and xenon distri-butions. It is noteverthy that the reactivity lost in both cases appears to have been fully recovered during succeeding Weeks. Although neither the fluctuations nor the relatively
, rapid 1 cases in reactivity appear to bear any relation to boron hideout , these pertarbaticns did tend to blux the details of the reactivity follev.
- 2. Analysis of the Observed Reactivity Behavior Several different hypotheses have been advanced to de' scribe how bcren deposition might occur. The hypotheses differ both
'WCAP 6050, " Analysis of the Reactivity Cnaracteristic of Yankee Core I" C. G. Poncelet, pp 11-lk, 17
_ _ _ _ _ _ _ - - _ - _ - _a
. . . - - - .-. -- = - _ _ . - . . . . - -
o i in their internal physio-chemical mechanism and in the ex-ternal physical conditions urder unich deposition could be expected to occur. The respective models predict that if there is an unerplained change in reactivity caused by boron accumu-lation, it should be evident during, a, the initial rise to power at the inception of chemical shim power operation.
- b. long term steady state operation at power,
- c. the transition f rom nucleate tailing conditions in the core tc fully subcooled conditions.
Comments and obar'v ons on operations under each of these three conditions fcllow.
The Initial Power Rise Initial cperation of power with a significant quantity of boron in the main coolant began on May 27,1953. Erpected reactivity changes durinE the rise to power ard the eneu--J transient period were those caased by the power defect and by changes in number densities- of fission product nuclides , parti-l cularly xenon-135, in the core. During this initial period l
externe 11y controlled core conditions (pressure, temperature, boron concentration, pH level, and, after the power rise, power
, level) were held constant. Reactivity changes were compensated for by control rod motion.
e 5
t i The relevant model for boron deposition indicated that boron could be expected to deposit either concurrently with, or soon after, the initial rise to power. Indication that deposition had taken place would be an unexplained loss
, of reactivity. Examination of the results obtained from the reactivity follow reveals that a slight gain in reactivity occurred. More detailed calculations than those carried out by the SCOOP code indicates that this gain is associated with spatial redistributions of fission products and of plutonium nuclide cor.centrations.
Extended Operations A second hypothesized boron deposition mechanism postulates that depositica might take place over extended periods of opers-tion at rates dependent both on the main coolant boron concen-l
. tration and on the extent of nucleate boiling in the core. In i the absence o,f superimposed experimental perturbations, reac-tivity prediction calculations need take into account only burnup -
effects (including temporal transients) and the compensating con-trol rod motion. Boron accumulation, were it occurring as hy-pothesized, vould appear as a gradual unexplained loss of reac-tivity in the reactivity follow calculations. In principle, a l
change in reactivity induced by boron accumulation would be
-- indistinguishable from an- apparent change .in reactivity arising from an error in the burnup calculations and, indeed, could 9
32-4 -
2 g
w w---rw, .or +- - - eayya t-+. -,e-+- .--,-.w-g-g-w yw i ,--p,ggq. ,-,,.-.-pg.- -
w.-e-.. ,,,9, y.y s- yt---en-~g*,pe-.ys+ ,,-4.-g ,--y --'
- g. -gr p ypgqe-p r, c TCv,ib>my e9 wy
e .
readily be masked by an error in the calculated burnup, provided, of course, that the loss of reactivity caused by burnup is linear with energy production.
Actually the etperimental program pursued during the first phase of chemical shim operation exposed the reactor to a variety of operating conditions. Periodic shutdowns, and startups after varying times at 2,ero power, led to temporal transients in both the observed and the calculated burnup rates caused by production and burnout of plutonium and sonarium. Operations were carried out at power levels .
both with and without nucleate boiling in the core. Boron concentration in the main coolant varied over a broad range from 1200 ppm to 600 ppm and was successively decreased and increased as dictated by experiments and operations in progress.
Each of these perturbations in burnup rate, power level, and boron concent' ration should have had a unique effect on the apparent rate of change of unexplained reactivity.
Close analysis of the detailed reactivity follow during the periods of extended operation under a variety of conditions shows no significant correlation of the rate of chan6e of unex-plained reactivity with any of the suspected parameters. As noted above, the changes that are observed show a rather good correlation with control rod position as is expected on theoreti-cal grounds.
_- i
Power Reversible Boron Deposition It has been observed in out of pile tests that nucleate boiling on a heat transfer surface tends to increase de-position on the surface of some materials from concentrated
, solutions in the coolant. The pestulated mechanism for this effect predicts that when a core in which a significant amount of boron deposition has occurred is taken from the nucleate boiling regime of operation to the fully subcooled regime, in which all boiling is suppressed, an observable gain in reactivity should take place. This gain should occur au the baron initially concentrated on the core sua sees is released and swept out of the core. The reactivity gain would appear as a change in unexplained reactivity either si-multaneous with or shortly following the power reversal.
An experiment to test this hypothesis was conducted prior to sne late November shutdown in an attempt to observe the postulated reactivity gain. Detailed follow results indicate I
that there was no detectable change in unexplained reactivity that could be associated with the power reversal itself, nor I
was there any delayed change in unexplained reactivity during the subsequent xenon transient. There is some uncertainty in g
the latter observation because of a fluctuation in unexplained reactivity which began about a day before the power reversal as a result of a change in central rod posit $on with respect
\
i 1
i l
1 l
. - . . ~ . - . - - - - - - . - - - . - . . - - . . . - .- . . . . _.-.-.-. . . . ~ . . - . . . .. . . . - . -.-
i- .
to the reference position and persisted through the remainder of the test period.
- C, Conclusions Experience extending over approximt.tely six months of operation
~
r at power with up to 1200 ppm boron in the main coolant at Saxton indicates that:
- 1. At no point in the Saxton operation did the amount of unex-plained reactivity possibly attributable to boron deporttion exceed the operational shutdown limit. Furthemore, in a
~
variety of tests no evidence was found of' boron accumulat1An ;
exceeding in reactivity equivdent the minimum amount detec-table with the methods currently in use.
- 2. The reactivity follow procedures currently available and in use at Saxton are. capable of detecting unexplained changes in reactivity of'the order of 0.1% in reactivity or less where these occur over short or intermediate periods of time, l
!' 3. Over long periods of time, during which complex burnup effects may become significant, long term unexplained changes in reac-
~
l . tivity may be less readily detected. In this case the decreased degree of confidence in the validity of the reactivity follow results-is reflected in an increased allowance lfor uncertainties.
- These uncertainties are applied to the maximum allovable unex-plained loss in-reactivity.in determining the operational shut-l-
35-l l
e.-- .se -,n.w m+-- w~e,m-t-'--w w,-w, e ve ,,,mr w - + v e+w, ---,-+wr--,er-m, . ,-,,,,m-,,-e,y ~. ,-.,r,i-y -,.vy-, ...w., ,,,-,->w.rw,-tay. g-- .c,=a,=+ww--y--- = , i-ew-.,-- er++--
down limit for unexpladned reactivity loss.
l h. It is feasible at Saxton to maintain a detailed reactivity l
l follov over extended periods of time encompassing many, if not nil, of the operational maneuvers likely to be experienced in a conventional pressurized water power reactor. .
l i
[
l i
e i
l 4
V. Chemical Surveillance Program
- W. D. Fletcher, G. B. Taylor The surveillance program at Saxton has two principal objectives.
These aret 1. ) to improve the state of knowledge regarding the effects of chemical shim on pressurized water reactor integrity and general operation, and 2.) to demonstrate by chemical means that use of chemical shim causes no hazardous situation to arise which could affect continuous plant operation. To fulfill these two objectives, an intensive chemistry program was conducted. The most significt nt contributions made from this to the overall Sarton program were found to be from the following areas of study:
- 1. Chemical shim solution stability at reactor operating conditions.
- 2. The pH effect on core reactivity.
3 The tcsts for possible hideout of poison in the reac' or core.
- k. Exam' nation of crud deposits (loosely adherent corrosion products) from the core.
5 The assessment of the effects of the chemical shim solution on plant materials of ecastruction.
A su= mary of the' results of the chemical surveillance program is given below. ,
A. Coolant Technology Numerous prior studies at WAPD and other installations formed the basis for choosing the water chemistry for the Saxton primary coolant, for both before and during chemical shim operation. The
- This work was partially supported. under AEC contract AT(30-1)-3269.
, i
\s
- _ . _ - . - - - - _ _ . . . - - . _ - - _ - . ~ _ - - . _ - - - . - - - -
selectier. of this chemistry was deemed the most satisfactory for long term plant operation, and one that vos believed te be completely stable at reactor operating conditions.
l
- l. Control Chemist ry_
Prior to chemical shim, the Saxton primar/ coolant contained additives of potassium hydroxide at a nominni
' concentration of N 10~ molal (up to 3.8 ppa potassium) and dissolved hydrogen at 25-35 ee (UIP)/kg coolant. These, respectively, vere maintained in the coolant to reduce the transport of radioactive corrosion products throughout the system, had to prevent ntt radiolytic decomposition of the veter. With chemichl shim operation, boric acid vac added to the ecolant to the extent of 1200 rpm boron, while potas-sium hydroxide and hydrogen concentrations were maintained the same. With continued operation, boric acid was progressively removed from the coolant, as required to compensate foi core de-pletion.
- 2. Chemical Properties Work at WAPD prior to chemical shim operation at Sarton showed that properties of the boric acid solution at normal temperature (70 F) are somewhat different from those of con-ventional reactor coolants. Mostly, the solution pH and con-ductivity of the chemical atim solution, as measured at room temperature, are significantly different from that of pure -
water containing alkali. This it illustrated in Figure 3
4 t
-6 C- .
i POTASSIUM CONCENTRATION,' PPM o -
to w A u o N oo e 5 26 m
t
~
Q 3 \ '
w u g >
y '$00 PN 8CROM 4
r 8 m UN 80R0!!
~
% 500 9FM 8 g y 9 m
, r. o a ,
9 F
%z5 k
J 4>
6 -A 00 - ~
k E
E
-O
? a - . - .
C
- N0 BOR05 \
3 .
)
I
=
l l
I v-e--.e4 ..w,-..%,,y~.-v--g m.-9-e ,+>wc-*-, +--r---w-e-r.,---- m we ,. awe g 43 ..,w-w,.w ww,ww-,--we.-,..wv -, i.- ,merm, . ,--v.,vmeyw,r...<.-4-w-e-,--+v-
__ _ _ . _ . _ _ ~ _ _ __._. _ . _ ____._ _ ___ _ __ -_. _ _ . _ _ _ _ _ . _ _ _
. . i which shove the variation of solution pH at room temperature with boric acid and potassium hydroxide concentrations. In contrast to the solution properties at room temperature, and probably more important, are the solution properties at reactor cos wt temperature. In this case, at a solution temperature of about 530'r, the average coolant tempernture at Baxton, it has been determined by calculation that the solution pH is ;
controlled mostly by its alkali hydroxide content, and influenced little by the presence of boric acid. This is illustrated in Figure h, which shows the variation of high temperature pH vith
.; borie acid and potassium hydroxide concentrations. The net.
result of using beric acid with alkali hydroxide, therefore, is to maintain apprcximately the same environment for primary system materials at the elevated temperature.
As vill be described later in this report , the solution pH at elevated temperatures has been a valuable tool for inter-4 pretation of the pH effect on core reactivity.
B. Operational Chemistry ,
The demonstration of successful reactor operation with chemical shin has required very close surveillance of coolant chemistry during initial power runs as well as during special physics or chemistry tests. This was accomplished by on-site analysis of the coolant by WAPD personnel at the time of interest, followed by trans-
+ fer of samples to WAPD for further analysis. The results of these
,, , . , . .. _ . . ,..,,,__,m,.-_-__..-,.._,__-__.- --m, , , . . . - _ _ _ . . - - _ .
I L
=
.a o i 3
5:
5 5 5 m oo 2 s
m 5
=
5 m
m u s
E 5 "
- R 'E u s
< hl 8 g8" "
2
=
_ = oo g '.c z s
- 9 %
B a
D o
- - - _ O e m =
N 5
o s
. o
-J a cy
( L o
I 5
wun3N 3
I m o L
E E
e e m n
~ so m e m m -
o Wdd 'N011V'1N3DN00 d WntSSV10d
-41 I
analyses are summarized belcv 1, Stability of the chemica*. Shte Goiy.lon
. Chemical shim operation began on May 27, 1963, at a power level cf 15 Mvt . followed over the next few weeks by a programmed rise in power to 23 5 Mvt. During this period, the coolant centained from 1050 to 1200 ppm boron as boric acid and about 3 ppm potassium as the hydroxide.
Very careful analysis of the coolant during this and sub-sequent periods of power operation showed no unexplained variation in either boron or pctassium. This was found to be the case regardless of whether or not local boiling existed on the cere surfaces. (About 16 percent of the core surface is estimated tc be at nucleate boiling conditions at a power level cf 23 5 Mvt.J It is recognized that one part per million change of boren in the ecolant cculd, if deposited uniformly en the reactor core, account for approximately 0.h percent change in reactivity, sc that censistent borcn analyses tc + 1 ppm, in themse.ves, dc not prove tne absence of boron accumulation. The coolant analyses, however, coupled with core physics measurements , do present a convincing argument against such accumulation. Moreover, it is reasoned that if salt de-position were to be induced by nuciente boiling on cere surfaces such that deposits of potassium borate formed, an equivalent of one part per million boron from the ecolant vould be accom-panied by four parts per million potassium. This is a very l
-hp.
I i
. ~ . . - , . - - . - - - . . , - - - -
_ - . -- - - - _ ~ _ _ _ . _ _ . - _ -_ - --
I sensitive indication of salt deposition since potassium analyses are reliable to + 0.1 ppm. With no unexplained variation in the potassium evident, within analytical pre-cision, it may be confidently assumed that either no potas-sium borates were formed at Saxton conditions or that any formed vere it: mediately re-dissolved in the coolant.
- 2. Changes in Chemistry 7,neurred with Chemical Shim Numerous samples have been collected from the Saxton primary system sirice initial power operation for the purpose of noting any significant changes in their elemental or radio-chemical content upon operating with chemical shim. Primary coolant camples have consisted moctly of solutions , crud-filtered from the coolant solution, and gases, dissolved in the coolant. Analyses have shown the followinc:
- n. Chemical Imnuritfes Coluth . Essentially no sirmificant changes in solution impurities have been evident with the change to partially neutralized, boric acid coolant chemistry. The exception to this has been, as expected, the ingrowth of lithium from the B (n.a) Li Inuclear reaction. Measure-ments on the rate of lithium increase in the Saxton coolant during chemical shin. operation can be accurately predicted by the expression'
- Weisman, J. Editor, WCAP-25h8, Saxton Quarterly Progress Report for the Period Ending Septembe,r 30, 1963, January 6,1964, p.15 s
v
]
o = 10~3 0P3g vhere d Li/dt = change in parts per billion of Li in coolant / day Cp a total boron concentration in coolant, ppm Pg = reactor power level, MWt Crud-Filtered from Coolant Bolutions - The crud level or concentration of insoluble impurities in the Saxton "
coolant has been found to remain fairly constant at 450 parts per billion during normal power operation. On shutdown, or following a thermal or hydraulic change in the system, the crud level has been noted to rise tem-porarily to 250 ppb. Analysis of typical crud samples for before and during chemical shim operation are given in Tables 2 and 3.
It can readily be noted that both chromium and maganese contents of the crud have increased with the change to chem.ical shim operation. Each of these elemente nas increased by a factor of two to five, while the other elements have changed little or not at all. This increase is best shown by comparing element ratios in typical crud i
j -44
-.-.r. - , , . .- , . - - . , . , -c., , -.,-.,,,-.,,-n. - n.,.,-.
Table 2 ,
TYP! CAL ANALYSIS OF CRUD FP,0M SAXTON PRIMARY COOLANT i CORROS!0N PRODUCTS BEFORE CHEM StilM DURING CHEM SHIM U U ELEMENT AFTER >2500 HOURS AT 530 F AFTER 3000 HOURS AT 530 F WITH KOH WITH KOH & H3 B03 PERCENT -
i Fe 17.5 26.4 Cr 2.4 13.4 !
Y Ni. 6.2 4.5 Mn 0.6 1.2 Mo 0.3 0.8 i
4 Co < 6.1 ,
0.2 t
l' !{ l!* i . li;;l:t i I,
! iti!i! l l
,[_
. F >
- 0 0
M53 30 I
T N H AT83 S
A SH _
L O
O MR&
E U 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 7
4 H O H 3 0, 0, 7, 1, 0, 0, 0, C
C HO 5 I 3 2 i 9 5 Y G 0 K 1 l 6
R A N0H R3I0 T I
M U D RW I
R t P i F
N A _
O g T .
X S F .
A T N
0 0
S E 3 a M M 5 _
e O E MT H A I
wo R L F E T SS D P RH U M A
MUO E
0 0
0 0
0 0
0 0 0 0 0 0
0 0
0 8
3 8
R R HOK 4 0, 2 5, 6, 0, 0, 8 C
F T C 0HH T 4
1 6 1 1 1 1 1 0, 5
O E R5W 0 I O2
~
S S
I E>
F _
BR Y
L t A I N l A A L '
A C
I R T P T EN DN .
Y N E P O UA T I RL M SL TL g l a u g h 1 COO E RI A A 5 C C M P S L
AM bpC .
E F pNI p
_ ! !! i. lj i1;' !i,j ; ,
l l
i l samples (Table b) with corresponding ratios in AISI30h r,r 316 stainless steel. From this comparison, it appears that l '
vith boric acid, the transport corrosion products more l
i l
nearly agree in composition with that expected for uniform 1
I corrosion of the base metal.
I I Other elements in the Oaxton crud, shown in Tahle 3, i
are mostly tramp contaminants such as silicon and aluminum, l
vhich arises from the effusion of jack-sparks from the coal-fired boilers of the old power plant on the site.
The presence of boron in the crud sample removed from
, boric acid solution during chemical shim is not unexpected.
The 'emount of boron in the crud due to absorption alone, i however, in difficult to ascertain, since the analysis shows
. the absorbed boron plus any boric actd occluded as solution in the sample. A concentration of 1000 ppm boron in the crud
! is not considered excessive.
1 l Gases - Dissolved in the Coolant - Disco 1ved hydrogen is
- maintainee. in the coolant by a hydrogen overpressure in the purification surge tank. There has been some difficulty in adjusting the hydrogen to lie between the narrow limits l
t l of 25-35 cc (STP)/kg because of the various operational se-l quences relating to pressurizer and surge tank levels. This, however, is an operations problem and is not affected by the 47-1
- - . , . - .-. . _ .- _ - - . . ~ _ - . _ _ .
Tably h COMPARISON OF SAXTON CRUD COMPOSITION WITH STAINLESS STEEL WEIGHT AISI 304 OR 316 CRUD ANALYSIS . CRUD ANALYSIS RATIO STAINLESS STEEL PRE-CHEM SHIM DURING CHEM SHIM l'
4 3 I Fe/Cr 14 i ,
f Fe/Ni '7 3 5 Fe/Mn 35 130 21
! ~
Cr/Ni 2 0.2 2 I
1.
presence cf chemical shim.
Diabolved exygen was routinely analyzed for ea:h day, and ther e was no noticeable change in the values upcn operating v2th chemical shim. Occasionally, the pressuriter liquid vculd show up to 0.1 ppm oxygen, and the prLmary coolant up tc 0.05 ppm but for the most part tne exygen was at, or below, the limit of detection cf 0.005 ppm.
- b. Purificatten by Ior. Ex:bange The purification demineralizer, shich contained K-OH cycle , mixed-bed resin, was used frequently during pcver runs pr cr to chemical shim. Since adding boric acid to the ecolant , however , there have been only brief periods of cemineralater operation, due mostly to rigid requiremen',s of varis us physics measurements on the core.
To c:nduct a sat 16f actory physich program, all operations involvir.g temperature cr chemistry changes were held to the minimum. Fct this reason, only limited data is yet available at tc decineralizer effiwiency of the K-borate form resin.
The d-ta cbtained thus f ar is reported in Table 5, in terms of the de:ontaminaticn factor , DP, the ratio cf bed inlet '
to bed outlet activities.
The DF for radio-iodine is expected to be very high
.h9 N.
i
L Table 5 EFFICIENCY OF ION EXCHANGE PURFICATION AT SAXTON i
PRE CHEM SHIM DURING CEM SHW I
K-OH FORM RESIN- K-BORATE FORM RESIN t
DECONTAMINATION FACTOR GRETIOURETF 4
I-133 9x10 I-131 4 x 104 2 1 x 10 i
Co-58 2x10 3 9xM Co-60 3 x 103 i x 103 1
Cs-1.7 10 7 l K 2 Xe-133 0.68 I Xe-135 0.73 -
3
_ J
- AT 10 GPM C00LAl4T FLOW =3 GPMIFT2 1 1
i i
r . . . , _ . . - --4 ._ - - _ . -
- - - - . _ - _ _ ~ - - - _ _ - - - - . - . - - _ - - _ . - . -- _ _____
s i vith the K-OH form resin, owing to the favorable equilibria of OH form anion resin for iodide and to
.lecay of iodide in the be'." In the borate form, the exchange for iodide is not as favorable, yet a DF of 130, shown in Table 5, is quite adequate to maintain a lov level of iodine activity in the coolant.
The fractional DF for xenon shows that more xenon is in the bed effluent, than in the influent, due to decay of iodine precursors in the bed. The lov DT for K-b2 is expected since the resin is in the potassium form. Any difference between inlet and outlet activities of K-h2 can be attributed to isotopic exchange with natural potcssium in the bed. The exchange of cesium with a potassium-form cation resin should not be too favorable, as indicated by a DF of 47 in Table h. Earlier work by l Simon and co-workers has also shown that equilibria are i
not favorable for cesium exchange with potassium form resins.** Einee cesium activity growth in the coolant is slow, and requires long operating times to attain significant amounts, this nuclide vill undoubtedly be controlled through norual coolant feed and bleed processes.
- WCAP-3716, "lon Exchange in Boric Acid Solutions with Radioactive Decay",
W. D. Fletcher , flovember ,1962.
- WAPD-CDA(AD)-528. "The Performance of Base-Forn Ion Exchangers for pH
- Control rnd Removal of Fission Product 9 from Pressurised Water Reactors", G. P. Simon et al April,1959.
l i t l
- c. Radiochemien1 Impurities l
Radio-nuclides of both fission products and l corrosion products are found in the Saxton primary coolant. These are usually analyzed for at WAPD by radiochemical separation and gamma spectrometry.
l Samples collected before and during chemical shim l
operation have shavn, in general, only minor differences. !
The except2cn to this has been apparent in the Cr-51 content of the crud filtered from the coolant. As vould be expected, the increase in elemental chromium in the crud with chemical shim has also increased the Cr-51 activity of the samples. Other than this, no significanc changes in corrosion product activities are noticeable.
l Fission products in the Saxton coolant are believed l
. to result from one or more defective fuel tubes in the l
reactor, core. It is estimated that fuel generatirg l
( apprcxistately 0.02% of the tota) power is contributing fission products to this coolant. This would constitute 0.4 average fuel tubes. From this defect, fission ptoducts I-131. I-133, Xe-133 and Xe-135 predominate. There is also f
j rome Cr-137 and Kr-88, but in minor amounts. There ves no
! noticeable increase in the amounts of fission products, or t
l the computed fuel-affected, after operating with chcAical shim for 43000 hours at 530 F.
l 1 -. - . - _ . . -. . - - _-. . .. . . - _ _ - . . _ --- -_ - - ,_ ,
i i C. Special Tests A number of tests of a chemical nature have been performed at Saxton in keeping vitn the tvc Trincipk1 objectives of the chemical surveillance program. These are described in the following
, 1. The pH Effect on Rehetivity One of the most interesting observations made at Saxton is ti.e apparent relationship between coolant pH and core reactivity Based on observations made from tests first performed at the Yankee Atomic Electric plant, ab summarized in the Yankee monthly operational reports, a series of experi-ments were designed and carried out at Saxton to learn more about the pH effect. In the Saxton tests, and whila the reactor was at full power of 23 5 Mwt with chemical shim, alkali metals potassium and lithium vere removed from the coolant by cation exchange, As the alkali r.etal concentration was lowered, core reactivity decreased accordingly, until ion exchange was stcpped, Pctassium was next added to the coolant, and within a period c f about twelve hours, the reactivity lost during k1kali removal was recovered. The data have been corre-lated in terum of solution pH at the elevated temperature of 5306 F from the curves presented as Figure h, for 1000 ppm boron solutions. The result for reactivity change with pH is given in Figure 5 It was found that the data correlated reasonably vell to give a slope for Ap/4 pH of 0.36 percent reactivity per pH unit.
l l
l 0
i i T 1 ru re ;
6.6 -
e 23.6 M RE AC16R PWER
~20% RUCLE ATT. 991LlWG 6.6 -
e
, 6.4 e
e 0*I ~
M A,H
- 9.16 w
% e J '.
E 6.0 -
e i
5 e g ,_ , ,
u 7.2 T : rura. 6 E 7. 0 - e il M Rf.ACTM P9tU S W9 RUCLEATE telLIM 3
0 g6.8 9 I 6.6 4
6.4 -
As/daH = 0.16 6.2 -
6.e -
\
l i I 1 ...i . ... I 4. ,
F -
CDRE AEACilVITY CKAM E, As, PERCENT THE COOLANT pH EFFECT IN SAXTON t.e. sr. si222i-C 1
O 8
- u. *
.; e (
i i
l A similar pH test was perfcreed at 15 Mwt to determine 1 if there vould be any difference at conditions where no local i
boiling existed on the core surfaces, as was the case at 23.5 ,
Hvt. Again, potassium end lithium concentrations were altered, ,
e either by potassium addition or by ion exchange, and, as before, j a reactivity change was encountered. The data were computed in terms of solution pH at the elevated temperature, and the result is shovn in Figure 6. The slope of this line agrees v1th data collected at 23 5 Mwt, again giving a dp/d pH-of 0.16 percent reactivity per pH unit. The coolant boron concentration r i t
during this test was 61050 ppa.
A third pH test was performed while the reactor was just critical with n: power. The coolant alkali was almost completely removed. Then, after 2k hourb, the alkali was restored by addi-
. tion of potassium hydroxide. Critical measurements taken through-i out the entire sequence of alkall adjustment showed no change in core reactivity within an estimated error cf + 0.01 percent.
This test was conducted with less than 10 ppm boron in the Coolant ,
A fourth pH test of reactivity was conducted while the plant was at 20 Myt, rnd_vith e primary coclant boron concen-tration of about 7 ppe. For this test. potassium was reduced from 3.18 to 0.1' ppm by iun exchange. Approximately 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> after the start of ion exchange, a maximum 1 css in reactivity t
f .
\ *, .
7
~ . , . - , . . - , _ , , .....,..o.,,m -#.m,,,._.._, . . , , .
4 i of 40,17 percent op was observed. Alkali vas restored in the coolant by the addition of L10H;H 0, to the extent of 2
0 7 ppa, and the reactivity previously lost was regained in about 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />. The average pH coefficient of reactivity vas measured to be NO.13 percent 4p/4 pH. The kinetics of this test are shown in Figure 7. Very careful analysis of the coolant shoved no unexplained variation in boron concen-tration within the analytical precision of + 0.1 ppm. These tests have shown the pH effett on reactivity to be a valid parameter to be considered in the operation of Sarton, as well perhaps in other nuclear reactors. Since boric acid was present in the coolant in gross quantities (41000 ppm B) during the first two tests, it could nat be stated unequivocally that boron absorption on core surfaces was not responsible for the reactivity behavior observed with a chan6e in pH. In this respect, the core surfaces would have to function as an ion exchanger, wherein berate vould exchange with hydroxyl ion, giving rise to increased or decreased reactivity. However, during the last pH test at 20 Mvt, with 7 ppa boron in the coolant , very car,eful solution analysis showed no variation l
! in boron within + 0.1 ppm. To account for the observed reac-t l tivity evings, approximately 0.4 ppm boron would either appear or disappear from the coolant if ion exchange on core surfaces were occurring. Moreover, the kinetics of the pH effect would tend to contradict the process of ion exchange of any substance 56-w- -m --- ew,, -
a w --+ v-, +,.-----:e-w--y,--
-w .,-m --rv w%:-vi '
w+ e* - 'r'i-' -*'u -- v-mW' ^ --e w -w
. i
~
7.5
- u. F - -
o g
^
7.0 K '
. e--
< 6.5 \ -
i E \,
g 0
.6.0 CATION
\ ~ ~
\
0 Lion ADDED v 5.5 -b EXCHANGE !
I 333 .-
g p = 0.11 AYG-u.
y ** ,%'N - NORMAL B7RNUP
'N, 1 E f
/./
- 533 h, , = s. sn E
$ 332 N -
8 'N
'N M, j /
o 5 31 53C I } i 1 3 iz 24 36 4 eo TIME, HOURS I
- - - - _ - _ - _ _ _ _ . _ . - --- - - _____ - _ - _ - - ~ - . . - - . _ - . - _ . - _ _
i e
on the core surface =. The time required to attain the ful3 sving in reactivity caused by a step change in pH is of the order of 10 to 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />. It would seem that ion-exchange processes, even those limited by diffusional mechaniser, would occur much more rayidly than this, since the reaction zone is at elevated temperatures. At present there is an hypothesis that the changes in reactivity are possibly caused by a change in heat transfer properties of the cere surfaces.
This vould result from a olov reorientation of the crud de-posit structure with the change in colution pH. At high pH, the crud is believed to be a dense crystallini, material with good heat transfer properties. At low pH, the crud is a volu-minous mass of material having poorer heat trannfer prcperties, causing increased fuel temperature which, through the Doppler effect, lovers the reactivity. A change in cr ud str ucture with alteration of pH is not too ur. reasonable an assumption. Obser-l vations have been made of decreasing liquid pressure drops through l
flov test sections with the addition of alkali to the high tem-l perature water.' By analogy, the voluminous crud in the tent section reorients to a dense deposit, lessening rertrictions to flow. Quantitative verificat$m of this hypothes!e is not possible because of lack of suitable heat transfer data for crud deposits as a function of pH. Calculated values for the temperaturu
1
-58 6
n
_ - - _ . . . . . , _ _m . -
, e l
char.ne in tha fuel, required to account for observed reae-tivity effects caused by Doppler, have been higher than seem pl aus ible . On the other hand, the absence of a pH effect at zero reactor power shovs the effect to be power dependent.
. Before the pH effect is satisfactorily explained by the heat-transfer consideration or any other mechanism, additional studies must be made. Tests are currently planned at Saxton t to explore this phenomenom further.
- 2. Eucleate to Non-nuclente Boiling Test The continued use of boric acid solutions for shim con-trol in a power reactor is dependent on complete solution sta-bility at all conditions of operation. This requirement die-tates that no significant amounts of boron be deposited in the system by any mechanism that could be rapidly reversed.
l -
Such mechanisms have been evaluated at Saxton considering deposition of boron on the fuel rod surfaces as boric acid or as borate salts. If boron vere to deposit on core surfaces, a thermal or hydraulic transient could conceivably release a sufficient amount of the deposit to result in an undesirable l
! reactivity insertion.
l
, *he postulated mechanisms for boron deposition are based or 1*po story studies of adsorption and of salt deposition with boiling. In the adsorption
- ctudies , it has been found that
" Fletcher, W. D. , Krieg, A. , and Cohen P. , "The Behavior of Austenitic Stain-less Steel Corrosion Products in Hi6h Temperature Boric Acid Solu-lutions", WCAP-1689 Rev. , May ,1961.
1 l
1 __ _____ _ . , _ . , . --
, e borates can be reversibly adsorbed by corrosion product oxides in anounts proportional to the solution bovic acid concentration.
In the radiotracer studies', it was shown that with sufficiently thich porous deposits on a heated'eurface, as for example, fuel elements heavily fouled with crud, salts from solution can precipitate in the deposit during nucleate boiling.
With sufficient crud deposits on the core surfaces, therefore, both of these mechanisms (if valid) could lead to cignificant quantities of boron accumulation.
To demonstrate the stability of boric acid solution at reactor operating conditions and to determine if these con-
. ditions vould lead to deposition on core surfaces, precise chemical inventory measurements vere made at the start of chemical shim operetion. During the initial phase of opera-tion at 20 Mvt with 1200 ppm of boron and 3.8 ppm of potascium l in the coolant, analyses indiented no change in either consti-tuent during two weeks of these conditions, except for slight changes because of solution make-up additions.
- 'Picone . L. F. "Radiotracer Studies of Hideout at High Temperatue and Pres-l sure", WCAP-3731, June 1963.
4.
rm.-- . - - . , - - - , - - - , . , - -, , . - . _ . , - . . - . -w. ,, , ~ - , .. . - , . . , - . _ _ _ .-,--r--_,-,,-, 3.,,,mm, , .y,
Upon increasing the power to 23.5 Mut, which resulted in nucleate boiling on about 16 percent of the core sur-faces, the boren and potassium concentration again remained unchanged with continued operation. Following about three months operation at 23 5 Myt, the reactor power was reduced to 15 Mut , changirg from conditions of nucleate to non-nucleate boiling. Analyses of boron, potassium, and lithium shoved no increase in concentration, and there was no change in core reactivity which could be attributed to the reduction in power.
It is concluded from these tests that no boron or alkali deposition, within the accuracy of the analyses, it evident during conditions of either nucleate or non-nucleate boiling.
This is fitrther substantiated by physics measurements of core
, reactivity described elsewhere in this report.
- 3. Lithium Stability Test Since the start of chemical shim operation at Saxton, it has become increasingly evident that the use of potassium as the alkali for pil control is not altogether necessary. If lithium vere selected as the alkali, it would be compatible with that lithium produced by the B10 (n,n)LiI reaction, and result in a coolant with only one, instead of two, different alkali additives. Because of the possibility that with boric acid lithium could lead to metaborate salts which have retrograde i
4 l
l
solubility, there has been, in the past , some reluctance to use lithium with boric acid. It was considered possible that these salts might deposit on heated surfaces, particu-l l larly under nu;1eate boiling conditions, and once formed the lithium metaborate might not redissolve but remain in place.
To explore the use of lithium hydroxide in the chemical shim solution, a test was performet' at Saxton in which the reactor power was increased from 15 to 23 5 Hvt while the coolant contained 560 to 660 ppm boron 10.45 ppm lithium as the hydroxide. After Nb days at 15 Hut , the pover was in-creased to 20 Mwt , and then, after 3 days , the power was fur-ther increased to 23 5 Mvt, where 16 percent of the core sur-face was at conditicns for nucleate boiling. Throughout the entire sequence, coolant chemistry was carefully monitored.
i Lithium analyses of the coolant were compared with the cal-culated rate of increase expected at each power level and boron concentration. At no time during the test was there any indication of unusual concentration changes for either boron or lithium. After three days at 23 5 Mut the very careful chemistry follow vas terminated; no lithium metaborate deposition ves evident at these conditions.
1 j -62 l
_ - .. - .-. - - .._ _ -__ - _ ._-- .- _ _ ~ _ -
During the first weer cf April, the reactor was operated at 23 5!Gt with 1.7 - 1,6 ppo of hi in the cociant Again the experimentally observed rate of Li buildup agreed closely with that expected, thus in-dicating there was nc metaborate precipitation.
D. Crud Examinations * (L. F. Picone)
One of the ecst important observations made at Saxton, in substantiation of the indicated minimal deposition of boron on core surf aces, was examinat, ion of the f uel rods of an experi-mental f uel assently Following the reactor shutdown on Novem-ber 22, 1963, the central 3 x 3 subassembly was removed from the core and transferred to the W Post-Irradiation Facility (WPIF) for detailed examination in the hot cells, This assembly had been placed in the ccre during the February,1963 shutdown and had, therefore, been in service throaghout the entire chemical chim test, It has accrued more than 3000 equivalent full power hours of :hemical shim service and, since it was located in the core-center, had been at cenditions of nucleate bolling at a l
l reactor power of 23.5 Mvt.
l i The 3 x 3 assembly contained both stainless steel clad and Zircaloy clad fuel rods that could be removed from the assembly.
l These rods were examined very carefully in the hot cell, and the
! crud deposit (that which was loosely adherent to the cladding ,
surface) was removed by scraping. This material was weighed and
'This work vas partially supported under AIC contract AT(30-1)-3269
, l s
l
analyzed for its chemical and radiochemieni content. A summary of the results of sampling and examination of the crud deposits is as follovt
- 1. The stainless steel clad fuel rod had an average crud sur-face corcentration of 70 mg/dm (3 rods sampled) on the ereas where crud appeared,
- 2. The Zircaloy clad fuel rod had a crud surface concentration of 58 mg/dm2 (1 rod sampled) on the areas where erud appeared.
3 The maximum thichners of a section of corresion product de-posit on the stainless steci fuel rod vas 1 5 mils. This value was obtained from the metallographic examination. The deposit thickness varied around the rod circumference from the 15 mil value to essentially zero,
- b. Chemical analyses indicate that the amounts of neutron poison associated with the observed amounts of deposited crud could constitute only a small reactivity loss in the Garton core.
5 The most significant deposition on the fuel rods occurred from about ik inches to 23 inches from the tops of the rods.
The surface concentrations of the corrosion products deposits as determined from the weights of material scraped from the rods are reported in Table 6. These values include only that material which is deposited from the coolant on the fuel rod surface and is more or less transient (deposit). No values are available to date 4
on the concentration of the ti &bt corrosion products which are essentially oxides grown in place ifilm).
Table 7 reports the chemier.1 analysis of the corrosion product deposits. The composition of the Saxton deposit is significantly different from that of deposits obtained from other reacter systems and in-pile loop tests; 1.e. , FW : ore I Bisnket Bundle, Yankee Core I. AFD-in-pile loop test. The Sarton deposit is significant3y enriched in chromium and manganese as compared with the other reactor deposits. No explanation can be given at present for this difference in chemical composition.
TABLE 6 Crud Deposit Surface Concentrations of Saxton Fuel Rod Surface Crud -Thickness of Total Crud on Total Rod
. Rod Number Concentrations Deposit (1)
- Area Crudded Rod Surface (2)
(3) mg/dm 2 (mils) mg dm 2
22 (SS) 83 .210 28 .3h0 13 (SS) Th .167 35 .h68 lb (SS) 53 .134 18 .333 103 (Zr) 58 .146 20 .350
- Neminal rod 2
area 13 dm
TABLE 7 Chemical Composition of Saxtot. Puel Rod Depajj+3 i Cbesical Composition (% of Oxide Sample)
Rod 22 Rcd 13 Rod 103 Rod IL Fe* 35 2h 37.88 h7 17 ,
32.94 Ni' 7 01 8.26 10.14 6.h9 cr* 11 91 11.33 8 73 9 18 Co .0625 .0975 .0950 .0950 Mn' 8.8 11.1 11.0 7.6 Zn .163 .143 .142 .09h Ag < .001 .025 .035 .002 In .1700 .1350 .1150 .6000 Cd .1500 .1425 .1700 .1170 B .0615 .0725 .0168 .0950 Al N 1.0 N 1.0 N1.0 N1.0 Ca .0675 . 3 '. 0 .0550 41 5 Cu N .10 N .1 % .10 N .1 K < .02 . 0; < .02 < .08 Li .02ho .0297 .0170 .0687 Ms N30 N 3.0 $3 0 . N3.0
( Na .01h0 .0065 N .0065 .0220 l Nb .0142 .0120 .0117 .0079 Pb N .h3 % 38 N .26 .28 Si N15 N13 ~N1.1 N1.5-Sn -N 37 N .32 N .29 N .21 Ti 4 55 % 50 N .45 % .30 V .03h0 .02h5 .0260 .0122 l
Zr .0222 .0212 .0645 .0118
- Results 1 20% relttive, all ntliers 1 30 to 50% relative.
l (1) Assumes that the deposit is essentially magnetite having a porosity )
of 0.70.
(2) Based on the assumption that all deposits on the rod are of the some concentration as the sampled area.
, (3) It is estimated that the probably error is + 10% to - 50%.
Based on the chemical composition of the crud given in Table 7, it may be shovu that a unifo.m : ora deposit of h70 mg/dm of this material would be required for 0.3 percent loss in reactivity. As given in Table 6, only a smal] . 7ction of the total area of the fuel rods was covered, and in this are 9r -. .imum deposit was N80 mg/dm . Since examination of a peripheral i
as h s .. i,y , ring the November 22 bhutdown, shoved no deposits of crud, it woula be. cv;rly conservative to assume a uni"orm core deposit of 80 mg/dm .
The actual crud deposit over the entire rod (having N3 dr area)is about 8 mg/dm . If it is assumed this is equivalent to 16 mg/dm ,because of the crad being located in the high worth area of the core, this leads to a maxi-mua of 0.01 percent ak/k vorth of the crud deposit. This, of course, is escentially the same as no less in reactivity that could be attributed to irreve- t 'e deposition of poison material in the crud.
P. Conclusions The f6regoing summary presente conclusive evidence that from the plant chemistry viewpoint, operation of a pressurized water reactor with chemical shim is completely satisfactory and that no real problems of safety or plant operations are evident. The Sax-ton program has fulfilled the objectives of the Chemical Surveillance a
Program in that,1.) the kncvledge of effects of chemical shim operation on plant integrity and general cperatf.on has been in-proved, and 2. ) that chemical shim operation un.ter normal con-ditions causes no hazardous situation to arise to affect plant operation.
G l
_ . . . - ~ _ _ . . = . . . .. - ~ - _ - - _ . - - .. . -_- -
VI. Flux Distributien Studies F. J. Frank A, Summary The three-dimensional power distribution and nuclear hot channel facters in the Sarton core under chem shtm operation were determined using the measured flux vire activations in conjunction with analytical two-dimensional (x,y) povsc distri-butions as determined by the FDQ code.
The following conclusions can be made regarding the results:
- 1. All hot channel factors were below the design limit.
- 2. There is good agreement between the power distribution de-termined by thermocouple data and that determined by flux vire data. The root mean squared difference ranged frem h.7% to 7.h% for the cases studied .
- 3. The power distribution as determined by thermal data is more uniform than that determined by flux vire data.
- h. The fuel assembly grids located at 4 pcsitions along the length of the assembly cause a decrease in the axial hot channel f actor because of flux depression, i
Three quantities are used to describe the power distribution in the core. First, the axisl peaking facter Fg vhich is defined as the ratio of the peak axial power to average axial power along
(
any channel. Second, the hot channel factor F vhich is defined AH as the maximum to average enthalpy rise of the coolant in a flow I
channel. And finally, the hot channel fa:tcr Fq vhich is defined as the ratio of maximum pcVer density at a particular point to the average power density.
A su= mary of the nu: lear het char.nel f actors T g and Fq is given in Table 8, These values include an added 10% uncertainty factor. The ho* channel factor F qis determined by the product of the axial peaking f actor F and W het channd factor F 3 33 The hot channel facter F AR represents the average of four fuel rods (plus 10%) which are located at each ecrner of the assemblies noted in Table 6.
- 3. Comparison of Expected and Observed Resu ts The design hot channel facters Liovable during 23.5 MWt chem shim operatica vere originally reported in the Nuclear Design and the Hazards heport as a nuclear F AH 2<03 (2.i.7 including engineering factcr) and a nuclear F qof 2.79 (2.92 including engineering factor). The cal ,ulated nuclear f actc2 F q did n. have the usual 10%
margin below the design value It was hcped that measurements might yield this margin. This however did not happen.
In the estimation of the design value f:r 23.5 MWt, consideration was not given to the fuel follower rods and the heat generated out-side of the fuel pellets. A re-evaluation of the design hot channel factor F qwas performed by the Thermal and Hydraulics Group. With 90 fueled follover zods fully inserted inte tne core, 2 7% heat l
TABLE 8 Nuclear Het Char.nel Factors Chem Shim Operation Height of Location Rod 5 Power B 16,H, Q Corner of Assy Run No.
18.39 23 MW 700 1.90 2.78 D3.E3.Dh Eh 308.7.8 20.12 23 19 685 1.88 2.69 308.7 10 21.22 23 25 717 1 97 2.81 308.7 9 21 72 20 1107 2.00 2,98 308.3.2 21 74 14.6 1200 1.97 2.96 308.1.8 22.34 23 783 2.03* 3.01" 308.7.5 23 17 Ik 5 1200 1.98 2 95 308.1.10 2k.36 23 1060 1 91 2.90 308.3.7 l 26.07 23 5 100a 1 98 2 77 308.3,8
, 26.72 23 873 1 93 2 71 308,7.2 l
23.5 MW Design Value 2.03 3.03 B E F
- Only b vires irradiated fD 1
. Saxton peak -2 Assembly Locations channels - ~ _,
_k 5
, ,,-r ,-
generated outside of the fuel pellets , and a peak heat rate of lb 1 kv/f t , the maximum permissible nuclear Fq can be increased to 3.03 (3.17 including engineering factor), Thus, as Table 6 indicates, the measured hot channel factors having had a 10% un-certainty added to them are still below the revised design limits.
A comparison of hot channel factors which occurred during rodded operation with those which occurred during chem shim operation shove the latter values 'to be higher. The nuclear hot channel factors observed during rodded operation vere of the order 4
of 1,76 and 2.68 for F AH and FU.C. reCPectively. The higher values observed during chem shim are due to the uniform fuel loading in the Saxton core. Froper chem shim design required non-uniform fuel loading in order to achieve a uniform power distribution.
1
! Boration of the core then permits an optimum rod configuration
- which minimizes the distortion of power and consequent higher l
l hot channel factors caused by the rods. However, control rods are beneficial in flattening the power distribution in a uniformly loaded core. Thus the programmed rod configuration in the Saxton core results in a more uniform power distribution and consequently lower hot channel factors than that experienced under the borated conditions.
l A summary of the power distributions determined by flux vire data and that determined by thermal data is shown in Figure 8.
-72
i l
Run 308.1.8 5-29-63 !
Rm 303.1.10 6-5-6p ,
Run 308.3 2 6-12-03 :
Run 308.3 7 6-24-63 Run 308.7 2 10-8-63
. Run 308.7.8 10-22-63 Run 308.7 9 10-31-63 B C D E F r,
1 I I
'T ,
w'
.70 .75
.7o 62
,7 _ .72 .74 I
.68 .75
.72 .79
< .75 .81 ro ,
. n_ ,1 < _qE r ,r
[ t
,r ,r g_ p g ,
1 33 1.26 . 72 .75 1 32 1.26 71 .13 i 1.22 1.21 76 .72 ,g 1.21 1.22
-i a .75 .70 1.17 L15 . ,
1.21- 1.29 . .]4 m
r ir ,t aj
( r p
,m L17 is L22 .
i L 72 L ' L 58 1,47 L72 1. A L57 1.b4
. L66 1.64 1.45 . L 40 -- *3
,1 ,- L63 1. 62 ' ~ Lk5 1.ko 9
.1.61 L 56 1.4A 1.29 w e M , 1: R_ i 1:M 1:R 2 % .], ,
. m - >< , - i.
r
.63 .73 1.08 1.06 ' , 74 ,p
.63 .68 1.og 1.09 .73 .75 75 69 1.11 1.o7 ,
a
, ,p ,p _4 76 .68 1.14 1.09 - -
,78 ,
.78 .74 '1.17 1.14 *p .
- 2
.69 .66 1.16 1.06 **
.70 g , _"? j , .18 1.09 ,, ,,
,j7s8 _J 6 =-il
ai .
, .67 .62
. 67 75 5
.71 .U $
.74 78
.76
- 83
.72 .75 < --
..n _?R .
t > < 2 i
Figure 8 Co parison of Thermal ana 6 clear Power Distribution Chem Shdx operation
, s
The nuclear distributions were determined using the flux vire activations in conjunction with analytical two-dimensional (x,y) power distributions as calculated by the PDQ code. The PDQ calculations were based on the 20 MVt,1000 ppm boron, no xenon, no depletion condition.
A measure of agreement between the nuclear power distribution and the thermal power distribution is obtair.ed by calculating the root mean squared difference, where the difference quality is de-fined as:
P -T 1/2
- o. 1 N
(
i i) 2 N I Pg 1=1 where:
Pg = the normalized nuclear power in assembly i T g = the normalized thermal power in assembly i A summary of the root mean squared difference for the various l
tests is shown in Table 9.
l
. - ~ . . . -
TABLE 9 Summary of Roet Mean Squared Diff erences Height Power Bc'on r
Dat e of Rod 5 (MVt) (ppm) c5 Run No.
308.7.8 10/22/63 18.39 23 700 5 9h 21.22 23.25 717 6,59 308.7.9 10/31/63 10/8/63 26.72 23 873 6.61 308.7.2 308.3.7 6/2k/63 2L.36 23 1060 6.0S 308.3 2 6/12/63 21,72 20 1; 07 4.69 308.1.8 5/29/63 21.74 IL 6 1200 7.37 308.1.10 6/5/63 23.17 14.5 1200 6.89 The variatler. of the axial peaking f actor F vith 3
the movener.t of rod 5 is shown for thrce thimbles in Figure 9 The variation of the nuclear hot channel fa:ters Fqand F gg is shown in Figures l
10 and 11, respectively.
The resalts sh0V a dip in tne value of the axial peaking f ac-tors fcr thimbles A6 and 03 when rad 5 is inserted beycna appr xi-l mately 21.5 inches . Tne peaking factor for thimble Il does not experience a dip. Since the thim.bles A6' and X1 vere nearest the hot channel and therefore represent the best estimate of the p;wer i
I distribution at that point , the hot channel f actore F and F AH also experience a decrease.
- These effects can be explained by considerin6 the variation
-7 5-
l.70 1.65 - ,
t
! l.60 -
^
1.55 -
N a u_
a g 1.50 -
W ' I.45 -
O A6 E I.40 '-
x 0
- w l.35 -
o_ ' o
.a y p I.30 -
- l 4
w 1.25 8
z f.20 -
A THIMBLE XI l.15 -
. 7xiuste ,,
0 THIM8LE C3 l I.10 -
1.05 -
,,og 1 l I I i I I I I L I I I I I i l l l 10 12. 14 16 18 20 22 24 26 28 30 l
. HEIGHT OF R0D 5 UNCHES) .
NUCLEAR AXlAL PEAKING FACTOR vs WITHDRAWAL OF R0D 5. CHEM SHIM OPERATION FIG. 9 E. D. SK. 322162-B
~
l 3.25 3.20 -
CORNER OF ASSEMBLIES 03, E3, D4, E4 3.15 -
, 3.10 -
otsics VAtut 3.05 -
[
$ 3.00 -
.3 o
8 l2.95 -
3
. # l2.90c-d-
2.85 -
1 2.80 - 1 2.75 -
2.70 -
2.65 10 12 14 16 18 20 22 24 26 28 30 IIEIGilT OF ROD 5 (INCllES)
NUCLEAR 110T CilANNEL FACTOR F g vs WITi1DRAWAL OF R0D 5.CFIEM StilM OPERATION E. D. 5 Vm 322159-B
i 1
2.20 !
4 --2.15 -
2.10 ecsien VAtoc
. 2.05 -
2.00 a
^ -^
aa i l
i 1.95 -
a !
-l c ^
- l.91 ' - a !
i i
b a g
z :1.85 -
i 1 % ~x !
'.3 4 - 1.80 -
- us -
C9RRtR OF ASSEMBtlES te, CS, M, c4 - 'I j '
l.70 -
! l.65 ~
j i 1.60 -
i 1.55 I I I '
i I
1.50 I I l' I I I I I I I I I I I I 26 21 % i 14 16 18 29 22 . 24 10 12 . -
i ~
HEIGHT OF ROD 5 (INCHES) .
! l l
NUCLEAR 110T CHANNEL FACTOR FAH'vs WITHDRAWALOF ROD 5, CHEM SHIM OPER rig, y :
E.D. SK. 322158-5
of the axial power distribution (F ;7for thimbles A6 and X1
- a rod 5 is inserted. In general, as rod 5 is inserted, the
. poser peaking becomes more prono aced and the position of the peak moves toward the bottcm of the core. These effects are most noticeable in those 1.himbles nearest the banked rod (as in thimble X1) as the peak of the flux distribution apprcaches the absorbing sps,eer- grids with insertien of rod 5, the flat-tening of the tcp of the flux distribution becomes more proni-nent. This effect is observable for the flux distribution of thimble A6. The peak of the flux distribution measured in thim-ble X1 is near the spa.er grid for all the measured rod heights.
Thus, while the peak is depressed, the dip in the axia.1 factor does not occur.
! The depletlen distribution of the Saxton core at the- Apiil 20, i
19616 shutdevn ar.d for subassemblies then present is shown in Table
- 10. The aittributier. van de.te:r.aned esit.g the SADAR (Sarton Data Reduction) code. Tne average wre depleticr. vis a total of 6330 MWD /MIt'(131,01.0 k#nts) of whien 1521 MWDMIU (95,736 Mvhrs) was 4
- incurred during chem shim operatic
- .. Peak burnup was 15,600 MWD /MTU.
1 I
e
-79'-
4
, . . . . . . , . - - - , . - - . .,n.,,-,..-_,, n ~ , , - . , , , - , - . , , ,,-,-e,. ..~..,-.-,.,,m. ,,,,..nn ,,,--.. ..~ . . . . . , _
. . _ . - . , .. . . - - .. . _ . _ _ .. -- .~ -.
F D L P l l I;-
-2 N2 N1 -3
~
k N6 m
5 N
T TAE*E 10 C Core Leplet icr. as et April 20, 196k A Assembly MV hre MWD /MTU C C1 4? 3' 6 4625 D D1 5:20.3 5528 E El ao;..0 L324 B B2 4615,k 5012 C C2 7;;8., 7730
.D D2 iv7d.9 81k1 E E2 1235 5 7072 F .a 4159 ; 4595 B B3 5.Sc 9 5707 l
C C3 70. 2 7832 D D3 i005.6 6979
- E E3 7773.6 8560 l F F3 520;.0 5727
! B Bk h939;; L828 l C C4 6295.k '5932 r D DL *200.0
, 80h3 1 E EL 1012.3 7680 F Fk 4596.6 5137 C C5 L375.1 kB88 D D3 55k9.5 Sk2k l I E5 L229.5 L593 i
N- N1 460.k2 k000 N h2 (No fuel) 0 0 N N3 19L ,7 3 2052
- N Nh 151 26 1537 N N5 118.26 1077 l
l 8c._
VII. Conclusions and Future Studiec E Iartnoff, J. Weieman A. Conclusions It is believed that the Saxton program has demonstrated the feasibility of chemical shim operation under normal conditions.
For normal conditions, the experimental data indicate:
- 1) No significant accumulation of boron cont.aining material on core surfaces during plant operation.
- 2) No decrease in core lifetime because of chemical shim condi-tions.
- 3) Alkali addition to enable operation at a high pH is satisfac-tory.
b) The Saxton hot channel factors meet predictions during chemical shim operation.
"'ne first conclusion, lack of boron accumulation durinE operation, has been the most difficult to demonstrate. However, it is believed that the careful reactivity follow conducted during this experiment has provided a conclusive demonstration. Durir.s the chemical shin test period the predicted reactivity has essentially remainea within 0.2% A k/k of that predicted. A variation in unexplained reactivity of this ma5nitude must be expected considering the uncertainties in the prediction.
As noted in Section IV, examination of the reactivity follow results of Figure 2 indicate that when the reactor power was first i incret. sed to 23 5 MWt, there was a sm ll increase in the unexplained
?
l
reactivity lost. Ali.ho%t the magnitude of this loss is well within the expectes errors, its occurrence at this time requires careialt consideration. This email less aces not correspona to a power reversible deposit since a nutber of such power changes were
, made and no trend vae observed. Examination of the core central subassembly after chemical shim showed that the reactivity verth of core crud remained very r M ? after chemical shim opera-tion. Hence, tae approach to 23 5 IGt after non-borated operation essentially duplicated the first rise to 23 5 IGt under chemical shim. Thie second rise to 23 5 !Gt showed no chstge in unexplained reactivity in going from 15 to 23 5 !Gt. This is believed further to justify the conclusion that the original rise from 20 to 23 5 W vae not accompanied by a boren deposition.
The fact thst chemicC shim cperation leads to no decrease in ecre lifetine vae demonntrated by the results of the het cell examination of the centre.1 sablesembly. Even with the conservative assumption that the e:.tf re ce"e surface was coated to the same degree as the assembly, the rea.tivity vorth of the deposit was negligible.
The chemical surveillance program has shown that smai.1 quantities of potassium or lithium may be used for cbtaining the benefits of high pH operation. Careful study has shown no hint of difficulty-at the alkali levels required.
l t
I Hot channel factors, computed from flux vire irradiations and l
! thermal-hydruulic data, vere highly satisfactory. Even after i allowing for the porsible uncertainties, the predicted values were
[ .
not exceeded.
- 3. Future Studies _
i l
Although it is believed that the major question of chemical shim feasibility has been ancvered, work remains to be done in a few l
aress, i.e.:
- 1. Oscillator Studies i
Because of the jamming of the first oscillator, completion of this program has been delayed. The preliminary dcta have shown i
no evidence of power reversible boron hideout,* The prograra vill be continued during the remaining portion of the experiment.
I
! 2. Artificial Crud Te:sts l
The tests which have been run to date have shown chemical shim to be satisfactory under normal reactor operating conditions.
It is desirable also to investigate operation under certain abnormal conditions; in particular, operation with h1 6h crud deposits on ccre surfaces. As previously mentioned, out-of-i l pile tests indicate the possibility of a power reversible boron accumulation. It would be very useful to have a tech-1 l nique which would allow a reactor operator to determine whethcE o *WCAp-2627 "Saxton Kinetic Experiments". V. Rajacopal, (June,1964)
such a situation existed. This can be done by simply cutting back on reactor power or changing reactor temperature and pressure to suppress nucleate boiling. This vould eliminate any reversible deposition (if any were present) and show up as reactivity gain. The absence of this gain would assure the reactor operator that this -
high crud situation does not exist.
The nucleate boiling - non-nucleate boiling transition test has bee'n-carried out at Saxton. As expected, since the crud level was lov, significant reactivity changes were not observed. Under normal operating conditions, it is not believed feasible to acctmlnte the crud levels required to achieve a significant reactivity change.
In order to demonstrate this technique, it is planned to inject synthetic crud into the main coolant system. Out-of-pile tests l have demonstrated that such material tends to deposit preferentially on heated surfaces. Hence, it should be possible to build up an appreciable deposit on the core surfaces. The nucleate to non- .
nucleate boiling transition under these circumstances should yield a significant change.
1
! 3. Tests of Defected Zircaloy clad Rods l
Incomplete information is available on the behavior of defected Zircaloy clad rods in a chemical shim environment. Since Zircaloy is proposed for use in some large chemical shim controlled plants, it is desirable to obtain additional information. Accordingly, tests of purposely defected Zircaloy rods have been scheduled for the nezt portion of the program.
- 4. Boiling Studies Tne present tests have been limited to nucleate boilin6 conditions. No steam has been present at the exit of any of the channels. Operation at higher exit temperatures with chemical shim control deserves study since it would lead to higher pressure steam and more efficient energy conversion.
The effect of bulk boilin6 on possible boron deposition mechanisrw is unknown. A test program to evaluate chemical shim and bulk boiling at the exit of the hot channels is bein6 planned.
i l
l 1
l l
l e I
4 .
Acknowledgment The success of the Saxton Chemical Shim Program vse, largely made pocsible by the efforts of the SNEC and Westinghouse personnel at the Sexton site.
- . We should like to acknowledge the cooperation of W. R. Iaytaan, Manager of SNEC, H. J. W1111 cms, Saxton Plant Superintendent, D. E. Hetrick, I. Finfrock, D. Hoverd, J. Both and R. W. Swift of the SNEC staff. The efforts of the Westinghouse engineers at Saxton, T. L. Erica er.d R. W. Colombo, ably assisted by E. A. Hooper and S. L. Pekar, are gratefully acknowledged.
h The program relied heavily on many people at a Westin6 ouse APD. The efforts of the personnel of Chemical Development, Plent Development, and Reactor Development whc participated in tnis program are also acknowledged, i
0 l
l
'e