ML20136F952

Topical Rept Evaluation of NEDE-24011, GE Std Application for Reload Fuel,Amend 8. Rept Acceptable Based on Methods Meeting Stability Criteria,Per GDC 10 & 12

ML20136F952
Person / Time
Issue date:	03/31/1985
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20136A677	List: ML20136A670 ML20136F083 ML20136F944 ML20136F952 ML20136G046 ML20137J540 ML20137X272
References
NUDOCS 8504290473
Download: ML20136F952 (17)

Text

.e . -

i 3

e i

SAFETY EVALCATION - # t 0F THE GENERAL ELECTRIC '

TCPICAL REPORT NEDE-24011 AMENDMENT S a

f P

t GENERAL ELECTRIC STANDARD  :

APPLICATION FOR RELOAD FUEL, ,

AMENDMENT #8 l

f i

i e

i

(

March, 1985  !

CORE PERFORMANCE BRANCH l

O e -

m l

se

[

150YA%973 y l k....... =- .=. .- .._- . - /= W .. - - ........ . . . ,. . .

EWCLOSURE 1.0 INT.RODUCTION This SER evaluates the themal-hydraulic stability licensing criteria proposed , l Oy Gereral Electric in NEDE-24011 Amendment B. The GE report NEDE-22277-P-1, "Com;11ance of the General Electric Boiling Water Reactor Fuel Designs to -

Stability Licensing Criteria' (Reference 14), is the principal docyment submitted in support of Arrendment v8 to GESTAR. This evaluation has been supported by resfew and audit calculations perfomed by Oak Ridge National Laborator/ under centracts FJN B0777 (TER-reference 8) and FIN B0794 .

(TEP-reference 9). The results obtained by ORNL in their audit calculations #

ar.d comparisens to plant data and experiments have been used by the staff to {

set the uncertainty value of SE's mett.od. ology and to deterwir.e the acceptebility  :

of GE's preposed licensing criteria. I 2.0 DEstR PIION OF GE'S THERML-MDRAULIC STABJtITY METHOES AND PROP 0 ED I LICEN5IKG CRITEPIA l 2.1 Therral-Hydraulic StaM11ty Analysis Mathods To investigate the stability of the large conlinear dynamic B'E system the j stzbility of individual components is avsluated before analyzing the.ir inter-acticn with the total system. For the BWR, these individual corporents are the chanrel and rear. tor core. The hydrody.namic stability of individual (

channels is analynd and then the chanr.els are coupled hydraulically er.d .

combined with neutronics and heat transfer to study the stability of the core.

A linearized, srall-perturbation frequenc/ durain podel, FABLE (1) is used to ,

perfom these calculations. Linear, small-perturbation theory is a special case of the general theory of r.onlirear syste.ms analysis. The ir,teraction  ;

ef the reactor core with the physical control systems associated with the nuclear steam supply and, bence, the total systeil stability, is investigated  ;

with the nonlin'cir plant transient sierulator digital model, RE.DY (13). l I

Qualification of the analytical models -is demorrstrated by comparisons with operating plant tests. Cor. trol rod oscillator tests at several plants are l

1 i

p ..._. - . _ _ -.-

q

I r s 1-

! vsed to provide open loop ar.d closed loop response characteristics of ths IMt , ,

subjected to reactivity perturbations. In additien, pressure setpoint i oscillation tests provide system response characteristics for the neutron '

flux / core-exit-pressure transfer fw.ction. These test conditions are

sirr.alated using the 8EDY and FASLE models and the results are compared to .

trrst data, Qualification of the FABLE channel hydrodynamics model is performed by comparisons to electrically-heated channel experiments and data i

from aperating reactor tests.

The output fre., the CE acalysis is a limiting best estimate decay ratio.

This decay ratio is found in the low flow /high power portien of the power

) flow n:np at the intersecticn of the power flo t curve and the red block line under natural cf reclation conditions.

2.2 5tabilityTests '

I i

Tne GE metnods have been benchearked against various operating plant test data. ,

The principal cata come from the tests performed at Peach Ectto:n(3) (gg77, f 1978),Y.ermentYankee (39S1) and a recent test at an overseas BWR plant. l i

l'or an osci]1atory response, the decay ratio is defined as the ratio of two subsequent peaks which are bcth on or.? side (i.e., above or below) of the average value .of the oscillatory parameter. Decay ratio is used l as a measure of a systes's stability. For decay ratic <,1.0, the system i is damped and the oscillatory response decays, for decay ratio >1.0, the system is undarped ar.d the ost.111ations increase in segnitude. For the  ;

j special case of decay ratio = 1.0, limit cycle response is achieved, where the os'c111ations remain.at a constant magnitude. Limit cycles are >

the characteristic rssponse of nonlinear systems as they approach the i stabl*lity threshold. ,

. 2

^ -

The possibility of instability in a BWR has been investigated since the start- ,

up of early SWR 5. These early tests oscillated a control rod within one notch position (6 inches) and measured the response, of the reactor (core-exit-pressure and APRM signal). For modern higher-power density reactors, control rod oscillator tests are not desirable because of high cost and poor signal-to-noise ratios in large reactor cores. A technique using pressure perturbations was developed and stability tests were performed at the end of Cycle 2 and during Cycle 3 at Peach Bottom 2 in 1977 and 1978. These stability tests were performed at low core flows (near minimum pump speed) and at varying core powers (up to the design reference condition). During Cycle 3 the tests were performed at various cycle exposures to evaluate the effects cf fuel exposure on stability.

The test results verified that the small pressure perturbation technique tirovides a sitrple method for determining BWR reactor core stability maigins.

In addition, stability data were obtained at decay ratio conditions higher than those schieved in earlier contro1 rod oscillator tests. Stability characteristics above the rated rod line at minimum pump speed were deer.onstrated with adequate cargin to stability at all test conditions (meximum decay ratio

- 0.5). Detailed descriptions of the Peach Bottom-2 sttbility tests during Cycles 2 and 3 can be fcund in Peferences 3 and 4.

Success of the pressure perturbation technique used at Peach Bottom 2 and the desire for data close to the stability threshold led to stability tests at Vermont Yankee Nuclear Power Station in March 1981. The tests were performed before and after the first rod sequen:e exchange of fuel cycle 8. The stability tests were conducted at natural circulation flow, single-recirculation purrp operation at sinimum pump speed, and two-pump cperation at minimum pump speed. The core power was varied to points estending above the rated rod line. ,

Limit cycle oscillations of average neutron flux as measured by the Average Power Range Monitor (APRM) Subsystem were achieved at the intersection of I

3

f

• l 1

6 1 .

(

l

! natural circulation and rated rod lina without external pressure perturbations. ,

i Visual inspection of the cortrol room APRM strip chart recordings showed that I the amplitude of the APRM limit cycle oscillation could be distinguished from the normal APRM noise level, Thus, during this test occurrence of APRM limit

cycle oscillations as the system stability approached limit cycle opGration tas observable in the control room through the regular instrumentation.

The APRMs and Local Power P.ange Monitors (LPRMs) oscillated in phase with a slight phase shift due to the time lag associated with fluid mass transport

in the axial direction, No secondary effects of the limit cycle operation were noted and the oscillations reeafned beunded. The average cperating ,

I conditions did not change, except for a slight pcwer drift resulting from l xenon burnout. The limit cycle oscillations were suppressed when a few i control rods were inserted slightly. All other test conditions were stable including"two points above the rated rod line at minimum recirculaticn'pu:np -

) speed. Reference 5 contains a detailed description of the tests and results, t

i l

Recent stability tests at an overseas SWR plant have also denon:trated the i

occurrence of limit cycle neutron flux oscillations at matural cirealation i and several percent above the rated rod line. The oscillations vsra again observable on the APRMs and 1PRMs and were suppressed by minimal control rod insertion, It was predicted that limit cycle oscillations would occur at

the operating state tested; however, the characteristics of the observed

escillations were different from those pftviously observed in other stability tests. Examination of the detailed test data of these more recent

tests showed that some LPRMs oscillated out of phsse with the APRM signal and at higher amplitudes than the core average. Although the regional i oscillations were larger than the core average (6 to 7), margins to safety limits were maintained and the oscillations were detected and suppressed by control rod inse*rtion.

• j e e

.. . 1 1

1 i

4 I

4 6 I 2.3 GE Premsed Licensir.1 Criteria The final SE Proposed Licensiog Criteria (12) were submitted in respor.se to ,

staff questions I and are as follow:

The cer:plicrce of the Seneral Electric Company's Boiling liater Reactor (Sk'R) Systees, exclusively usir.g GE BWR fuel desigrs, to the stability criteria set forth in GDC.-12 has been demonstrated. Tne bounding fuel thermel/trechanical analyses cover all licensed SE BWR fuel desist.5 ircluding those contained in GESTAA thrcugh Ar.er.dcent 10. Future SE -

BWR fuel designs will also b.e in complitoce provided that the following stability co7p11arce criteria calculated uting approved methods are satisfied. .

1. t'entron flux licit cycles, which oscillate up to the 120". APNM high neutron flux scras setecint or up to the 1FPM upscale alarm trip (without initiating scrar.) prior to operatcr sitigating action shall not result in exceeding specified acceptable fuel i design limits (Safety 1.imit Minimum Critical Power Ratio and 3%

CladdingPlasticStrain).

2. The individual channels shall be designed and operated ti be hydrodynamically stable or more stable than the reactor core for all expected operating conditions (analytically demonstrated).

These criteria will be evaluated on a generic futi type basis for future fuel designs as they are added to GESTAR.

I 8ecause the stability compliance criteria are independent of plant specific characteristics cycle-by-cycle decay ratios will not be evaluated for specific plants"."However, the operational effects of introducing new fuel designs or special operating modes, will still be evaluated on a generic basis for representative NSSS product lines and fuel designs. The new fuel designs' 5

g 7. _ .. . . . ~ . . . . . . . . . . . . . . . . _ _ . . . . - . . . . . . . . . . . . . .

O and extended operating modes will be evaluated using approved methods to deter-eine their stability characteristics relative to current fuel designs (described ,

above)whichhavedemonstratedacceptableoperationalcharacteristics.

4 Based on the operating experience with current fuel designs, operator recornendations have been developed (51L-380, Revision 1) (Reference 7) for hig5 power density plants (e.g., BWR/4/5/6) which define a region of the ograting r:ap where operation is not recomended. In addition, a second regien is defir.ed in which increased monitoring of potential neutron flux oscillations is apprcpriate. If the stability perfomance of the new design ,

is bour.ded by that of the current fuel designs then the plant perfomance is consistent vita the basis for SIL-380 Revision 1 and these recommendations still apply. If the stability perfomance of the new design is not consistent with the currer.t designs, then SIL-380 Revision I will be modified for that [esign such that the stability z3rgins calculated at the boundaries of the tonitored regicn will te maintained consistent with SIL-380 Revision 1.

3.0 STAFF E/ALUATION The staff has evaluated the GE proposed licensing criteria, This evaluation which is 63 sed on the input fron two ORNL evaluation reports (8.9) and on numerous discussions with GE staff has rusulted in the staff position stated in Section 4.0 of the SER. A summary of the ORNL TERs follows:

3.1 Review of_ General Electric Thermal-Hydraulic 5tability Methodolcoy (Decernber 31,1923)

In Reference 6, URNL presents ao evaltattoo of General Electric's methcdology fcr calculating the stability of boiling water reactors for fuel reload lirknsing purpos~es. This evaluation is prfmarily based on etaparative analysis.

cf stability tests perffrmed et Peach Bottom and Temont Yankee versus results  ;

of GE's calculations for these tests. .

6 g -- -

- ~ . . , ~ . . . . _ _ _

ORNL compares decay ratios presented in a fuel relead submittal document with decay ratios both measured and recalculated at the end of cycle for that same fuel load. They also look at the impact that fittir.g procedures used by GE have on the nunerical value determined from experimental data

• for the so-called measured decay ratio.

In this review 0FIL concludes that a criterion specifying that the decay ratio (DR) shall be less than 0.8 should be set for GE's decay ratio calculations in fuel reload licensing submittals. If the 0.8 criterion is not met, a non-confomance region in the power-flow operating map must be defined; the reactor operator would be reouired to take a series of precautions to control the reactor within this region.

3.2 Evaluation of the Thermal-Hydraulic Stability Methodolocy Proposed by the General Electric Co.pany, Part II (Septeeber 30,1984)

Reference 9 contains OPNL's evaluation of the thermal-hydraulic stability methodology proposed by the General Electric (GE) Company to license reload fuel. The results of this .e/aluation complement the ones contained in the Reference 8 (Section 3.1) in which the capability of the General Electric Company to predict the stability of relcad cores was evaluated.

The results of ORNL's initial review showed that Calculated decay ratios are affected by two sources of error. One is input related because of the imprecision involved in calculating the operating cor,ditions for which the stability will be a minimum during a fuel cycle. The other is related to core modelir.g, since it was shown that different decay ratios have been calculated for reactor core operating conditions which yielded equal cxperimental decay ratios. Based on thre magnitude of the errors found in that review OML proposed an acceptance criterion of decay ratic less than 0.8 for fuel reload calculations.

7

. 1 i

i In NEDE-22277-P-1 (Reference 14) GE proposes two different approaches to ,

demonstrate compliance with stability criteria for reload calculations Approach 1 Cemonstrate that the calculated core and channel' hydrodynamic decay ratio are less than 1.0 for all expected ' operating conditions. .

Approach 2 .

This approach involves two steps:

~

(a) Der.onstrate that each generic BWR fuel design satisfies ,

the following cortpliance criteria: .

(1) Neutron-flux limit cycles, which oscillate up to the 120% APRM high-neutron-flux scram setpoint t (without initiating scram) shall not exceed i

specified acceptable fuel design limits (ii) The individual channels shall be designed and operated to be hydrodynamically stable (decay l ratio 1.0) or more stable than the reactor core for all expected operating conditions.

(b) Establish operator guidelines to terminate limit cycle oscillatf or.s.

The first apprea'ch was covered in ORte.*5 initial report (3.1), where ORNL ,

reconnends the threshold of 0.8 fcr decay ratio calculations to account for calculational. uncertainties in predicting the 1.0 threshold proposed by GE. ,

Reference 9 is related to the second appro4ch.'.

i 8

L

.......--,--....._..j7....,...,,_,,,,,_. ,

• • ? * ~

, - , - - - - - s-

The main points of this new GE proposal which need to be proven are whether:

• 9 (a) Neutron-flux limit cycles due to core-wide instabilities and oscillating up to 120% of rated average core power do not exceed current fuel design limits.

(b) The effects of limit cycles on fuel integrity can be calculated for generic fuel designs. This type of calculation'is not necessary for every fuel reload.

(c) Local channel instability oscillations are not possible because the channels are designed and operated to be more

. stable than the core.

(d) If limit cycle oscillations occur the operator is capable of identifying and terminating them following the recorsnendations in SIL-380 Revision 1.

The results of the OFNL evaluation are: >

(a) Core-wide limit cycles with the average power oscillating at frequencies greater than 0.25 Hz and up to 120% rated power are not likely to produce boiling transition and, thus, fuel integrity is likely to be maintained. ,

(b) The above result is applicable to generic fuel designs because these calculations depend mainly on the fuel geometry, and not on its neutronic characteristics.

(c) Local'i'nstabilities due to ficw escillaticns have been observed in reicent experiments, and therefore, they are a possible phenomenon in SWR operation. In those experiments, the ratio of h

p-

I e

local to average power oscillations was a factor of five (i.e., ,

i the local power oscillated 60. while the average power oscillated only 12%) and the frequency of oscillation was close to 0.4 Hz. ,

Assuming that this ratio and frequency remain approximately constant, our calculations show that boiling transition is not likely to occur even if the average power oscillates up to 120%

ofrated(i.e.,thelocalpoweroscillatesupto600%ofrated).

Therefore, local instabilities can be considered by the same ,

standard as the reactivity instability [ result (a)].

(d) The operator recommendations contained in SIL-380, if properly implemented, are considered to be sufficient to identify and terminate limit cycle oscillations.

Based on these results, the following recomendations were propos'ed:

(a) Stability calculations must be performed for each fuel reload.

(b) If the calculations show that the decay ratio is less than 0.8 '

for all ~ expected operating conditions during that cycle, the stability licensing criterion is met.

1 >

t Staff Coment-There is r.o proof or certahty that local / avg ratio is not higher than 6 to 1 - in fact it has,been observed to be as high as 7 to 1 in recent tests. Therefore, monitoring of local esci11ations is a very important ingredient in proper stability monitoring procedures.

e

~

~

e f 9 y  % e 10 ,

• • 7*- .++-- * == w w ~ * - - .

= # ,. . - - ,-s.....d,,_, f A . _. .., h ,, . , _ , , _ _ _ , , _ , , , . _ ,

~.

1 2

(c) If for some expected operating condition the decay ratio is greater than 0.8. then-(1) A nonconformance region should be detemined in the, power-flow operating map.

(ii) A procedure should be established to make the operator swsre of the possibility of oscillations in that operating region. -

l (iii) Special operator instructions should be established ,

to identify and teminate abnormal power oscillations should th,ey occur.

(iv) Calculations should be perfomed showing that limit cycle oscillations up to the 120% APRM-high-neutron-fluxscrampointplusanticipatedtransients(suchas generator load rejection with bypass failure) do not l reduce the critical power ratio (CPR) below the safety i limit CpR for the particuiar fuel design. (Note: this . l calculation might be performed for a generic fuel type andplantdesign).

4.0 STAFF POSITION - ACCEPTANCE CRITERIA FOR GE BWR FUEL DESIGNS FOR THERMAL-HYDRAULIC STABILITY ,

The staff finds the GE fuel mloads bounded by the conditions in Table 1 met the stability criteria set forth in General Design Criteria 10 and 12 provided that the BWR being reloaded has in place operating procedures and Ttchnical Specifications which assure detection and suppression of global and local instabilities. Such detection and 50p' p ression should . cover all modes of operation with particular te-phasis on natural circulation and single loop operation. Fuel reloads meeting these requiraments nee'd not perfom cycle specific stability celculations. Technical Specifications which enforce the recossnetidations of GE SIL-380 would meet these j requirearents.

~

l 11.

Q---- . . ~: . . , - _ .:  ; - .c:'~r '~":: - . :: ~~; ~'" :2 =r * :- =+'. 1

2 4.1 Exception to Acceptance Cri+eria for Plants Which Have Not Yet Implemented Improved Stability Technical Specifications

. o For GE reloads using Table 1 fuels in plants which have not yet implemented improved stability monitoring Technical Specifications the current practice of using the methods of NEDE-22277-P-1 to calculate a cycle specific decay ratio must be continued. This reload will be considered acceptable if the decay ratio is shown to be less than 0.80 for all possible operating conditions. BWR 2/3 type reactors using only the approved GE fuel types described in Table I have been shown to have adequate stability margins and therefore are acceptable and their reload cycles are exempted from the current requirement to submit a cycle specific stability analysis to the NRC.

4.2 New Fuel Designs Should GE develop fuel designs in the future which exceed the bounds of Table I the prementioned acceptance criteria and exceptions may still be applied to such fuel if any of the following procedures are followed.

1. Show that the generic calculations presented in NEDE-22277-P-1 are applicable to the new fuel.

OR

2. Hedo the generic calculations presented in NEDE-22277-P-1 in order to expand the approved bounds of Table 1 to include new fuel.

. OR

3. Perfoh cycle, specific cciculaticns using the methods of NEDE-22277.-P-1 and show the decay ratio to be less than 0.8. ^

S s

I

~

12 . ,

m. _

=

...... - ... - ~ ----. .

g_ . . _ . _ . . , , . . . . - _. .......c...___,__

. 3

e TABLE 1 ACCEPTABLE FUEL TYPES & OPERATING CONDITIONS Acceptable Fuel Types All licensed GE BWR fuel designs contained in GESTAR (NEDE-24011-P-A-6 throughAmendment10). -

e.g. .

7x7 8x8 .

P8x8R -

BP8x8R GE8x8E GE8x8EB Acceptable Operating Conditions I

AlllicensedmodesofoperationinGESTAR(NEDE-24011-P-A-6through Amendment 10).

e.g.

1. Standard power / flow up in F3AR

2. Operating Flexibility Options in GESTAR

~

a. LoadLineLimit(LLLA)

b. Extended' Load Line (ELLLA) {

c. increased Core Flow (ICF) -

d. Single Loop Operation (SLO) ,

)

e. Feedwater Temperature Reduction (FWTR) ,

~

13 -

1. . . , +r= + . de.w+ aw- e4 =1 . = - * " < + + - ~ < - ~ * * -

2 Acceptable Exposure Range Initial cycle to equilibrium cycle exposure for limits approved in ,

GESTAR (NEDE-24011-P-A-6 through Amendment 10). .

W k

O 4 4

- ~

14 a ._ _

l' . . .

5.0 REFERENCES

1. " Stability and Dynamic Performance of the Gene,ral Electric Boiling Water ,

Reactor," General Electric Company, Licensing Topical Report, January 1977 (NED0-21506).

2. " General Electric Standard Application for Reactor Fuel," General Electric Company Proprietary, April 1983 (NEDE-24011-P-A-6 and NEDE-24011-P-A CountrySupplements).

3. L. A. Carmichael and R. O. Niemi, " Transient and Stability Tests at Peach Bottom Atomic Power Station Unit 2 End of Cycle 2 " Electric Power Research Institute, 1978 (EPRI NP-564).

4. F. B. Woffinden and R. D. Niemi, " Low Flow Stability Tests at Peach Bottom Atomic Power Station Unit 2 During Cycle 3," Electric Power Research Inistitute,1981(EPRINP-972).

5. S. F. Chen and R. O. Niemi, " Vermont Yankee Cycle 8 Stability and Recir-culation Pump Trip Test Report," General Electric Company, March 1982 (NEDE-25445).

6. R. B. Linford, " Analytical Methods of Plant Transient Evaluations for the General Electric Boiling Water Reactor " General Electric Company. -

Licensing Topical Report, February 1973 (NED0-10802).

7. "BWR Core Thermal-Hydraulic Stability," General Electric Company.

February 1984 (Service Information Letter 380, Revision 1). -

8. J. M. Leuba & P. J. Otaduy, " Review of General Electric Thermal-Hydraulic '

Stability Methodoldgy", ORNL, December 31, 1983.

I ..

I l

15

~~-

Tc - --

4- .

9. J. M. Leuba & P. J. Otaduy, " Evaluation of the Thermal-Hydraulic Stability Methodology Proposed by the General Electric Company", ORNL, September 30, 1984. . ,

10. Letter, J. F. Quirk to C. O. Thomas, " Submittal of Proprietary Report on Compliance of GE GWR Fuel Designs to Stability Licensing Criteria (NEDE-22277-P-1),datedNovember6,1984.

11. Letter, C. O. Thomas to H. C. Pfefferlen, " Request Number One for Additional Information on NEDE-24011, Rev. 6. Amendment 8, December 26, 1984.

12. Letter, H. C. Pfefferlen to C. O. Thomas, " Response to Request Number One for Additional Information on NEDE-24011, Rev. 6. Amendment 8, dated January 14, 1985. -

t

13. R. B. Linford, " Analytical Methods of Plant Transient Evaluations for the General Electric Boiling Water Reactor," February 1973 (NED0-10802).

14. G. A. WATFORD, " Compliance of the General Electric Boiling Wate'r Reactor Fuel Designs to Stability Licensing Criteria", October 1984.

(NEDE-22277-P-1) l ..

l

~

O 16

_ ... . ...,.J _ ,_ _ _ . _ ._ _ . _ , ,

l

,5 .

. . ATTACHMENT 3

./pa ase UNITED STATES 8 o NUCLEAR REGULATORY COMMISSION

& I wash m orow,o.c. zones

• g l o....

[ - -

NAY 101984

+

Mr. J. C. Chandler Exxon Nuclear Company, Inc.

P. O. Box 130 Richland, Washington 99352

Subject:

Acceptance for Referencing of Licensing Topical Report XN-NF-691(P), " Stability Evaluation of Boiling Water Reactor Cores" We have completed our review of the subject topical report submitted February 15, 1983 by Exxon Nuclear Company, Inc. letter JCC:030:83.

We find the report to be acceptable for referencing to describe the techniques and procedures used to establish the limits for norinal core e operation.- This report is acceptable for referencing in license appli-cations to the extent specified and under the limitations delineated in the report and the associated NRC evaluation, which is enclosed. The

' evaluation defines the basis for acceptance of the report. The staff approves the Exxon Stability Methodology for use in licensing reload fuel under either of the following conditions:

1. the calculated decay ratio for the proposed cycle is less than or equal to 0.75 and acceptable Technical Specification restrictions are placed on natural circulation operation; or

~

2. the calculated decay ratio for the proposed cycle is less 3 than or equal to 0.90 and acceptable Technical Specification requirements are placed on natural circulation and single loop operation including proper surveillance of both LRPMs

~

and ARPMs.

.-~

We do not intend to repeat our review of the matters described in the report and found acceptable when the report appears as a reference ~1n -

license applicatiens, except to assure that the material presented is applicable to the specific plant involved. Our acceptance applies only to the matters described in the report.

In accordance with procedures established in NUREG-0390, it is requested '.

that. Exxon' publish accepted versions of this report, proprietary and non-

~

proprietary, within three months of receipt of this letter. The accepted versions shall incorporate this letter and the enclosed evaluation between

, the title page and the abstract. The accepted versions shall include an -A (designating accepted) following the report identification syrbol.

I .

4_

g. .

I e j Mr. J. C. Chandler MY 10 g i ,

should our criteria or regulations change such that our conclusions as to the acceptability of the report are invalidated, Exxon and/or the applicants referencing the topical report will be expected to revise and resubmit their ,

respective documentation, or submit justification for the continued effective  ;

applicability of the topical report without revision of their respective -

documentation.

Sincerely.

O. 6

@ Cecil 0. Thomas, Chief Standardization and Special 1 Projects Branch Division of Licensing

Enclosure:

As stated

- i N

i

__ ___.______m--

-?.

.3,. .. .

,e .

s

.O

. ,d ,

t, . '- . . .

flk t;

'?

i- SAFETY EVALUATION 4

[' 0F THE .

EXXON NUCLEAR COMPANY

?!

TOPICAL REPORT XN-NF-691(P) l.-

?-i

? *

.i

\.

i e i I Y: !

af p-i i-)

STABILITY EVALUATION OF BOILING WATER REACTOR CORES 44 e, .

-: i S .t.

i

- - 'i APRIL 1984 4

CORE PERFORNANCE BR/SCH t

e k

• ep ,

. ji - .

I r'/ L ! +,A A,f f4T, , . gio8

-g 4

't W -

t '--#-* - --+ - - - - -

j ,* . ', '

1.0 INTRODUCTION

, ~

This SER evaluates the themal hydraulic stability methodology proposed *

,'E by Exxon in XN-NF-691(P) and as supplemented by XN-NF-691(P), Supple-f ment 1. This review has been supported by review and audit calculations

.t.

performed by the Oak Ridge National Laboratory under contract FIN B0777. *

[. The results attained by ORNL in their audit calculations have been used by the staff to set the uncertainty value for Exxon's methodology and to

[, detennine the acceptability of Exxon's approach.

p.

t

2.0 DESCRIPTION

OF EXXON'S THERMAL HYDRAULIC STABILITY METHODOLOGY 4

Stability is defin'ed by Exxon for an operating system as follows: a t! system is stable if, following an input perturbation, the transient h response returns to a steady, non-cyclic state. For a time domain y' analysis of reactor core stability, the degree of stability is defined by that decay ratio (the magnitude ratio, X M , of successive transient 2 O j maxima or minima). The decay ratio is determined from the core

, average power response to a rapid perturbation in system pressure or control rod position. When the decay ratio is less than 1.0, the I;

reactor core is stable.

i4 Exxon Nuclear Company's reactor core stability analysis methodology l

{ utilizes the COTRAN computer code. The one-groep cross sections used in g the iterative flux solution are determined from input two-group values

and modified at each time step for thermal hydraulic feedoack. The two-

. group input cross sections for COTRAN are obtained from the XTGBWR core 1 A7 simulator model.

~

p .. i i

- I i XTGBWR calculations are performed along the rated power-flow line and '

oe

,l along therpower-flow ifne corresponding to natural circulation to obtain the appropriate COTRAN input. '

. The COTRAN model utilized for" reactor core stability analysis simulates ,

the core average fuel design. The hydraulic flow channel is modeled

.. with the spatial detail of the neutronic calculation and extends from -

4

9 n

i

- ._ m _. _ . - .2 _ _ - - - - _ - - __

~

F . '

i; p

l [.  :.4 the inlet orifice to the upper tie plate. The modeling methodology is , ,

consis, tent with that utilized for the reactor core stability verifi-

!-]Q) cation with integral plant data from Peach Bottom Unit 2 and Dresden

j. Unit 2. , ,

Stability transients are initiated by perturbing the steady-state f

operating conditions and applying the steady-state core average pressure

[. drop as a boundary condition. A typical perturbation is a ramp decreased

,ei in pressure of s 4 psi in 0.10 seconds. At the end of the ramp pressure

- } l- change, the system pressure is fixed for the remainder of the transient.

The resultant transient power response is analyzed to detennine the j, operating state decay ratio.

z.

$i e 7,t This transient analysis procedure is applied over the range of power / flow c'onditions to yield a range of the core stability margin at the limiting

]. ,

end-of-cycle operating state. It is expected that the least stable

• { .

operating point in the range will occur at the intersection of the rated

~

power / flow line and the natural circulation flow line.

t Y Exxon has used this methodology to compare the COTRAN calculated core decay ratios to BWR stability test data available to them. Their ej analysis showed the COTRAN calculated core decay ratios to be in general

[g "

agreement with the reported dats, and conservative at core conditions 1 which result in high decay ratios. Extrapolation of a least squares fit hf

• - h?

of the data indicate that for conditions which woul.d result in COTRAN 4.".t; calculated core decay ratio of 1.1, the core decay ratio derived from

'f measurements would Le expected to be 0.74. Their statistical analysis 9

..e of the differences between the core decay ratios calculated with COTRAN dp and 1;be reported core decay ratios indicate that a COTRAN calculated

' I core decay ratio of 0.9 will provide 95 percent probability that the ,

.I . core decay ratio is less than 1.0 with 95 percent confidence, which -

allows for uncertainties in the extension beyond the data.

[ - -

a '

l

) "

l

,; )

e

% }

Ii e g- - - _ _ . . _ . . - .. ,._ ,.. . _ , . _ - _ _ __ _ , _ _, _ , _ _ . , . _ _ _ _ . _ _ . _ _ _ . . . , , _ . _ , . _ _ , _ , _ _ _ _ _ , _ _ _ _ . _ _ . , _ _ _ _ . _ _ _

- - . . . . . - . - - . . . - - . - . . . . . . . . . . . . . . -w--x.--------

~

l l'

I 4

3.0 STAFF EVALUATION

^

The Exxon thermal-hydraulic stability methodology (Refs.1, 2, 3, and 4 has been reviewed by the staff and has been audited by ORNL using the frequencydomaincomputercodeLAPUR(Refs.5and6).

} Reference 2 contains the results of a sensitivity analysis perfomed by

) Exxon's code COTRAN (Ref.1) for end of cycle conditions in the Dresden Unit 2 boiling water reactor (BWR). Reference 3 contains a series of benchmark analyses of the COTRAN code versus the Peach Bottom Cycles 2 and 3 stability tests (Refs. 7 and 8). The LAPUR audit calculations focus on the end of cycle in Dresden Unit 2. For the Peach Bottom case, I

the results of the perturbation tests are used for comparisons to Exxon's calculations.

The staff has reviewed Exxon's methodology and used ORNL's audit calcu-lations to detemine the main sources of error in their methodology and to set an approximate uncertainty in their calculated decay ratio.

This review shows that Exxon's methodology has three main sources of

error in calculating decay ratios: (a) COTRAN does not model the j recirculation loop, instead it assumes a constant pressur'e drop across '

the core as a boundary condition which introduces a conservative error i of approximately 50 percent at limit cycle conditions (i.e., the COTRAN l .

. decay ratio is 1.0 when the reactor decay ratio is 0.5); (b) COTRAN uses f a core n.odel with only one radial zone with average power and flow which I

introduces a non-conservative error that is estimated to be approximately 10 percent (i.e., the COTRAN decay ratio is 0.45 when the reactor decay ,

ratio is 0.5 if no recirculation loop error is included); and (c) finally, .

there is an error involved in defining the operating conditions (such as power, flow, axial power shape ...) for which the stability will be minimum during a new fuel cycle. We estimate that this error introduces a 15 percent uncertainty in the calculated decay ratio.

. t e .

\

• f '

l Comparison between Exxon calculations and results of Peach Bottom tests , , k

show ttiat a 0.2 dispersion error exists in calculated decay ratios and, for most cases, the calculations were non-conservative.

Based on the above considerations the staff feels that the uncertainty in the decay ratio estimated by Exxon's present methodology could be as high as 25 percent. As such an uncertainty would require a decay ratio to be less than 0.75,Ve cannot approve a methodology for determining themal-hydraulic stability which is based solely on calculation with a

,[ 0.9 acceptance criteria as proposed by Exxon.

The staff approves the Exxon stebility methodology for use in licensing

' reload fuel under either of the following conditions:

1._ the calculated decay ratio for the proposed cycle is less than or equal to 0.75 and acceptable Technical Specification restrictions are placed on natural circulation operation; or

2. the calculated decay ratio for the prr., posed cycle is less than or equal to 0.90 and acceptable Technical Specification requirements are pieced on natural circulation and single loop operation in- ,

'.' cluding proper su veillance of both LPRMs and APRMs. 1 4.0

SUMMARY

The staff has reviewed the themal-hydr.eulic stability methedology

. proposed by Exxon in XN-NF-691(P) and as supplemented by XN-NF-691(P),

Sup;1ement 1.

The staff approves the Exxon stability methodology for use in licensing ,

reload fuel under either of the following conditions: -

1. the calculated decay ratio for the proposed cycle is less than or -

equal to 0.75 and acceptable Technical Specification restrictions

';, are placed on natural circulation operation; or m_- . -. - - - ,- , - , - - - , -

. . e 5

-~

2. the' calculated decay ratio for the proposed cycle is less than or equal'to 0.90 and acceptable Technical Specification requirements

'f i are placed on natural circulation and single loop operation in-cluding proper surveillance of both LPRMs and APRMs.

S e

e e

6 r

M t

w l .

+ -- - . .A -~~~==. m m . ; ,, m ,2 %, A .m

l.

, - - - ~

. z ,, . . , . . .

?-

- 6

.)

i

5.0 REFERENCES

1. "COTRAN: A Two Dimensional Reactor Kinetics Program With Rea'ctivity Feedback for Boiling' Water Reactor Core Analysis," Exxon Nuclear

Company, XN-NF-661, December 1982. . ,

2. " Stability Evaluation of BWR Cores: Sensitivity Analyses," Exxon NuclearCorporation,XN-NF-691(P), February 1983.

L 3. " Stability Evaluation of BWR Cores: Benchmark Analyses." Exxon Nuclear Corporation, XN-NF-691(P) Supplement 1. June 1983.

I 4. Letter to C. O. Tt.omas from Exxon Nuclear Corporation, Contents:

,j. " Description of Stability Evaluation Program Framework," September 30, 1983.

$- 5. "Modeling the Dynamic Behavior of Large BWR Cores," P. J. Otaduy, lV) Ph.D.' Dissertation, University of Florida, 1979.

2: .

'. i 6.  !'A Comparison of BWR Stability Measurements with Calculations Using

, the Code LAPUR-IV," J. March-Leuba and P. J. Otaduy, IRJREG/CR-2998 and ORNL/TM-8546, January 1983. .

3

. 7. " Transient and Stacility Tests at Peach Bottom Atomic Power Station

'i - Unit 2 at End of Cycle 2." EPRI-NP-564, June 1978.

s;

$' 8. " Low Flow Stability Tests at Peach Bottom Atomic Power Station

Unit 2 at During Cycle 3." EPRI-NP-972, April 1981.

.. 3 .

lAj ti 4

a ,

O ,

• .j

, .p h

iX .

.ib. ~A -

_y f .h

/l -

m- -

r i*..

a

9. . ,

_s- ~ , 1 - -e---- em8he4E M M ag. M Eif*A".*," .

O

-. - - - . _ .. = .. ~~ _. - . - - .

I Attachment 4 - Generic Letter TO: ALL LICENSEES OF OPERATING BWRs ,

SUBJECT:

TECHNICAL RESOLUTION OF GENERIC ISSUE fB THERMAL-HYDRAULIC j STABILITY The staff has been studying BWR thermal-hydraulic stability characteristics for

, several years under generic issue #B Thermal-Hydraulic Stability. We have

, recently completed our review of this issue and the purpose of this letter is I to inform you of our findings on the resolution of generic issue fB-19.

, Specifically, we have recently completed our technical evaluation of topical reports1,2 submitted by General Electric and Exxon which describe their analysis methods and have concluded the following:

4 GE/ Exxon methods for calculation of core stability decay ratio are un-

! certain by 20%/25% in predicting the onset of limit cycle oscillations (decay ratio = 1.0). Thus a core having a calculated decay ratio of 0.80/0.75 may, in fact, be on the verge of limit cycle oscillations with-in permissible operating space. The result of this conclusion is that i BWR 4, 5, 6s (BWR 1, 2, 3s have sufficient margin when utilizing current fuel designs) may not be able to show compliance with Genaral Design ,

Criteria 10 and 12 solely using analysis procedures toiprove that thermal j hydraulic instabilities are prevented by design. However, we have con-l cluded that operating limitations which provide for the detection and I suppression of flux escillations in operating regions of potential in-stability consistent with the recommendations of General Electric SIL-380, are acceptable to demonstrate compliance with GDC 10 and GDC-12 for

! cores loaded with approved fuel designs, i

l It is our understanding that most, if not all owners of BWR 4, 5, and 6s, have or will be submitting revised technical specifications which enforce GE .

~

l SIL 380 recommendations. However, owners who choose to show that their plants i ~a 're stable by design must include approved uncertainties for the calculation 1

i i

" T " "f*_7 "" E ,., _ _. ,.C'* 7 ' ~ % . .- P** C~T*~M'* 2 " C ' I ' T b L ; -- - -, * * "

~ < - - - - - - - -

a 2

of core decay ratio in their safety evaluations of,all core reloads and any ,

licensing actions which affect core thermal-hydraulic stability.

This generic letter does not involve any reporting requirements so that no OMB clearance is necessary.

Sincerely, Hugh L. Thompson, Jr. , Director '

Division of Licensing Office of Nuclear Reactor Regulation gOm m_

d ATTACHMENT 5 OPERATING PLANT & NT0L BWR 4, 5, 6 STATUS

• A PLANT STATUS Browns Ferry 1, 2, 3 I I = Intend to implement Brunswich 1/2 I revised Tech Spec.

Duane Arnold R R = Have revised Tech Spec. .

Hatch 1,2 R.  ? = Owner unsure at time of Nine Mile Point 2 I survey.

La Salle 1 I Peach Bottom 2 I Peach Bottom 3 R Susquehanna 1,2 R Grand Gulf R Vermont Yankee  ?

Cooper I ,

Fitzpatrick I ,

Hope Creek 1/2 R Perry 1/2 R .

Fermi 2 R Shoreham R h Limerick 1/2 R a

h

+ l M *

~

l l

l l

.,.,.,...n.. . , , . ,, _ . . , -- - , ,