ML20085H450

Rev 2 to Licensing Basis for Oyster Creek Long-Term Solution to Reactor Instability

ML20085H450
Person / Time
Site:	Oyster Creek
Issue date:	08/31/1991
From:	Bond G, Furia R GENERAL PUBLIC UTILITIES CORP.
To:
Shared Package
ML20085H432	List: ML20085H428 ML20085H434 ML20085H440 ML20085H450
References
TR-068, TR-068-R02, TR-68, TR-68-R2, NUDOCS 9110280279
Download: ML20085H450 (30)

Text

TR NO: 068 REV. 2 PAGE 1 OF 30 Licensing Basis for Oyster creek Long Term Solution to Reactor Instability TR #068 (REV.2)

BA No: 335425 Author: John D. Dougher August 1091 APPROVALS:

M i

I l Ronald V. Furia Manager, OC Fuel Projects i

h hb M Cordon R. Bond j _ Director, Systems Enginrering GPU Nuclear One Upper Pond Road Parsippany, New Jersey 07054 9110280279 911009 PDR ADOCK 05000219 P PDR J

TR NO: 068 REV.2 PAGE 2 0F 30 ABSTRACT This report describes the licensing basis for the oyster Creek long term solution to.the core stability issue. .The solution relies on the quadrant-based APRM system to automatically sense and suppress expected modos of thermal-hydraulic oscillations. Changes to the Technical

' Specifications are proposed to_ guarantee the availability of the

-LPRM/APRM detectors in order to ensure sufficient response to regional oscillations to prevent the violation of the MCPR safety limit. In addition, changes to the core limits section will require the operating limit'MCPR for each cycle be greater than or equal to 1 47. These two Technical Specification changes in conjunction with the existing quadrant-based APRM system will ensure the mitigation of power oscillations prior to violation of any safety-limits.

Finally, as a prudent operating strategy procedural controls will preclude normal operation in the region of potential instability..

i 1

a

TR NO: 068 REV.1 PAGE 3 OF 30 TABLE OF CONTENTS PAGE #

i 1.0 11[IP&D. llc 11 DN /

SUMMARY

4 2.0 ELETtM DESCRIPTION 7 2.1 Genore, l'escription 7 2.2 trRF 's APRM Assignments 7 2.3 APRM 2 rip ated LPRM Alarm Functions 8 2.4 Inorerable LPR!ts 9

) 2.5 Operator Inter +*sce 9 3.0 LICENSING APPR2hCH 12 3.1 overview 12 3.1.1 GDC-12 Compliance 12 3.1.2 Design / Licensing Philosophy 13 3.2 Oscillation types and HoJols 14 3.2.1 Automatic Protection Design Dasis 14 3.2.2 Higher Order Oscillation Modos 15 3.2.2 single Channel Hydrodynamic Stability 16 3.3 Design considerations 17 3.3.1 Design criteria 17 3.3.? Operational Constraints 17 4.0 SUPPORT' ANALYSIS 19 4.1 overview 19 4.2 Olobal Oscillations 21 4.3 Regional Oscillations 22 5.0 TECHNICAL SP,E.CIFICATIQl{ 29 6.0 EEFERENCES 30

f 5

7R NO: 068 l V

REV.2 PAGE 4 OF 30 l

1. 0 -1 nth 0DUCIlON/SUMMAPJ The Oyster Creek Long Term Solution to the core stability issue uses the quadrant-based APRM system to automatically sense and suppress expected 6 modes of thermal-hydraulic oscillations at magnitudes significantly less I than would be needed to challenge the MCPR safety Limit. As a prudent operational strategy and to reduce the likelihood of a reactor scram ,

.resultinq from oscillations the Oyster Creek procedures will preclude t normal operation in the region of potential instability.

t The oyster C*eek Nuclear Fw .ty, a DWR-2 plant type, has never experienced cotw thermal-hydraulic instability. Power oscillations, however, have been observed in other BWR cores both domestically and internationally. Two modus of oscillations have been observed, core-wide and regional half-cori, During core wide oscillat'.ons, the neutron flux in the core oscillates in the radial fundamental mode (in-phase). During regional half-core oscillations, higher order harmonics in the radial plane result in out-of-phase oscillations, with neutron flux in one half of the core oscillating 160' out-of-phase with neutron fivx in the other one half of the cora.

General Design cri aria-10 (10cFR50 App. A) cequires either prevention-or ,

detection and suppe esion of power oscillations which could result in violations of fuel'de \gn limits. Analyses performed by General Electric l

l p

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

TR No: 068  ;

~

REV. 0

, PACE 5 or 30 l 4

have de.monstrated that for large magnitude cacillations the potential exists for violat*,on of the safety limit MCPR (reference 1). The ability  :

to detmet oscillations before safety limit violation is primarily dependent on the oscillation modo, plant operating lim 1.t MCPR, the initial HCPR before the oscillation, the peak magnitude of the [

oscillation, and the neutron monitoring syst6m configuration (i.e.

LPRM/APRM assignment and setpoints for neutron flux scrams). For a given ,

peak' oscillation magnitude, fuel safety limits are more likely to be protected automatically for coro-wide rather than regional oscillations r

due to a larger response from the APRM system. For regional oscillations GE's analysis demonstrated that at 10% g.eak-to-peak on the APkH's most planes have margin to the safety limit. In addition, for plants with a ,

flow biased neutron flux scram (co filtering of the neutron flux) automatic mitigation cf regional oscillations may be provided at an oscillation magnitude below that at whic.. the safety limit is challenged. Plant specific analysis for oyster creek demonstrates that the existing APRM system will sense and suppress power oscillations prior to violation of any safety limits.

Oyster Creek is a low power density plant with a relatively tight fuel inlet orifice. These plant features are stabilizing-factors for channel ,

hydrodynamic stability. The resulting lower channel decay ratio reduces i the probability of regional oscillations. Therefore, the oyster Creek core, in the unlikely event that it experiences a power oscillation, has j

.a preferred tendency to oscillate core-wide (globally). The oyster Creek

. quadrant based AFPP system is very responsive to global oscillatione and would provide a yeactor scram prior to challenging any safety limits.

1.

L -- 1

'l

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

TR Mos 068 REV. 2 PAGE 6 or 30 Notwithstanding the fact that half-core regional oscillations at Oyster Creek are very improbable, an oyster Creek specific analysis for this '

type of oscillation has been performed. The analysis evaluates the capability of the existing quadrant based APkM system to reliably detect

^

and suppress regional oscillations prior to exceeding safety limits. The [

t analvois included additional design CPR margin beyond the margin for  ;

uncertainties already incorporated into the safety limit. Various core  ;

conditions, failed / bypassed LPRM patterns and operating limit MCPRs were considered.

As a result of this analymis changes to the Technical Specification will be made to restrict the allowed number of out-of-service LPRM/APRM  ;

detectors en the A and B+1evels. This will guarantee the availability of j the LFRH/APRM detectors in order to ensure a sufficient response to regional one.11atior.s to prevent the violation of the MCPR nafety limit.

In addition, a Technical Spec'fication provision will se included in the core limits section requiring the Operating Limit MCPR for each cycle be greater thren er equal to 1.47. These two Technical Specification changes in conjunction with the existing quadrant based APRM system will ensure automatic detection and suppression of power oscillations prior to

_ violation of-any safety limita.

As a prudent and conservative operational strategy precedural controls will precluda normal operation in ths region of potential instability, thereby further reducing the possibility of any power osciliations.

Y P

, . , . ,. . . , < . . - . - . - - .- . ,w ...#, .y..s , ,, ,,,r..,,,...ye..,,,_., . -m.,., ,..- . ....,.,n..,w.r~-5 e

TP. HU 068 Rev. O PAGE 7 OF 30 t 2.0- SYSTEM DESCRIPTION  !

2.1 General Descrictl D 2 The quadrant based flow-biased APRM neutron flux system associated with a high operating MCPR limit will detect and suppress a thermal-hydraulic instability prior to axceeding safety limits.

The APRM system consiste of electronic equipment which averages the output signals from selected i.PRM emplifiers and develops an output signal related to percent of rated core thermal power. These LPRM vignals are grouped together such that the resulting 1.PR21 signale provide adequate coverage of expected oscillation modes. The trip units associated with the APRM system actuate automatic protective action when APRM signals exceed pres 9t flow-biased values.

2.2 LPRM to APRM Assianment The APRM system consists of eight independent channela - two  ;

channels per core quadrant. Channels 1 through 4 are associated with Reactor Protection System (R'S) #1 and channels 5 through 8 with RPS #2. Each core quadrant is monitored by two APRM channels, each of which are associated with a different Reactor Protector System. The two ATRM channels in a given core quadrant utilize the same four LPRM detector strings with RPS #1 APRM channel receiving inputs from A ans C level detectors and RPS #2 APRM channel

TR NO: 068 REV. O PAGE 8 OF 30 receiving inputs from B and D level detectors. Eatn APRM channel normally averages the inputa of eight LPRM detectors. Figure 2-1 provides an overview of the APRM detector arrangement.

2.3 fi}'M Trip and LPRM Alarm Functippa The APRM high power scram setpoint is flow-biased. Consequently, the APRM system requires recirculation flow signals to compare with the APRM power level signal. Flow converters develop recirculaticin flow signal outputs for use by individual APRM channels. When an individual APRM channel signal exceeds a preset flov 5iased value a reactor scram signal is sent to its associated RPS system. A full reactor scram signal is generated when a minimum of one APRM channel per RPS system has a high flux or inoperative trip conditior.

Figure 2.2 provides a plot of the Oyster Creek ilow-biased scram line.

LPRM hl/ low a.arres provide the operators with indications of possible thermal-hydraulic oscillations. Procedural controls based on APRM signals ensure operator response to power necillations too small to initiate an automatic reactor scran end too small to violate safety limits.

\

l l

TH NCis 060 f EEV. O PAGE 9 OF 30  !

2.4 -Ing,pgrable LPRMs a

Edisting tec.nicwi specifications -limit the number and distributic.n '

of bypassed or out-of-servico LPRH detectors and APRM channeas.

Based'on detailed stability analysis additional restrictions on the number and dist*1bution of out-of-service LPRHs will be added [

(section 5.0).

i t

2.5 Doerator. Inter (ggg i

No direct control interface is required for the safety related APRM trip function, However, the LPRM/APRH system has a wide range of i alarm, display, computational and data logging capabilities. With the exception of the alarm function described above, no credit will be taken for operator interface in achieving the syster functional performance requirements.

~

L b

I-l -

t l

I' l-e  ;

L I _ _ _ _ _

TA NO: 068 -;

REV 0 '

PAGE 100F 30  !

J 4g I o n 7 i

_F 7 '

41 -n -n i F 7

• l 33 -- ~o' h n n .

1 25 -

n ---6 9 -- p  :

h 1

11 o v --

t 09 a P --

L J 00 12 20 til 36 4n LPRM STR1NGS. APRtiCHANNELS o 1 5 o 2 6 A 3 7 Y 4 8 LPRM LEVELS A&C B&D RPS 1 2 Figure 2.1 LPRM Assignments to APAM Channels -Quadrant APRM 4p.mc- , , , y.., * .- ,rw. -iy-s,.4 v , , wm-. , ,,-.e.s_ -,w--,..- , . - - - . . . . - . , e

TR NO: 068 REY, O PAGE 110F 30 t

I i

l- .

' RATED POWER = 1930WYT utto rtos - el mainn ,/

8-

~

/ '

1. 5 )f /y 4 ,$

1. t 5 v#v9 o

/

/  ;

}*~ "# p $,p f,/ '

j_ <.uu .

y f narro toAn ux -

/,-

,[/

NATUPAL CDICVIATION 2-i l .I I I I I I I I u 10 M 30 40 SO 60 to 60 90 100 Core Flow (2) l Figure 2.2 Oyster Creek Operah Power / Flow Map l-

TR NO: 068 REV.0  :

PAGE 12 OF 30 ,

=i 3.0 LICENSI.fl0 APPROACH t I

~

' 3.1 g'gr11gy 4

i 3.1.1 gac-12 como11angg i i

L ,

The quadrant based APRM flow-bi: sed neutron flux system is capable of automatically detecting and supprassing-powet t

' oscillations wtaich could result in conditions exceeding _r f

the MCPR's*.foty ilmit.

Reliable protection is provided by using redundant APRM channels and a safety-grade trip -I system. These design features will ensure cc.npliance with

.)

CDC-12.

t i

McPR. safety limit compliance has been demonstrated for all l

expected modes of thermal-hydraulic. power oscillations for the Oyster Creek Nuclear facility. Although the APRM system is expected to be capable of responding to other  !

postulated modJs of power oscillations, demonstrating McPR ,

complianer-for such modes is not necessary, for the reasons discussed in reference 4.

y i

s

--.-.-.m., -,c.'_,. ., , , . ,

,_g.,, ,,;.,-_, ,, g, ,c-,,,, , _.,9 .-..y,.,,-,,7,,,.y',,%y,.,-,

,,pm,yy,,,,,,, . y.., .,,yyyp-mge,,,,,,w,,w.yw .._,7we ,.

.,,.7.,yi-,-.m,

_. . . _ . ._m... . . . . _ - -.- ..-_...___._. ~ .. . _ _ - .-. . _. . _ _._._ _ ._ _

-)

TR NO: 068 ,

=REV. O  !

PAGE 13 OF 30 l

l t

3.1.2 g,esion/Licensino Philosophy By using a majority of the LPRMs, the APRM syetem uses the l boot available instrun.or.tation for detecting an oscillation. As a result of the uniform distribution of the fission-chamber LPRHs throughout the reactor and the >

grouping of these LPRMs to form APRM channels the system l r

in capable of immediately responding.to any neutron flux  ;

oscillation which'could pose a HCPR concern.

'The APRM system design is capable of generating a trip

, signal at a sufficiently low amplitude such that margin to the MCPR safety itmit is provided. Operating. experience 3 with core wide and regional oscillations shows that LPRHg readily respond to osci11ations, and the APRM system, being a simple summation of 8 regionally adjacent-LPRM -

detectors, will reaoily respond )s well. f I

1 :-

t..

TR !Jos 068 REV.0 PAGE 14 OF 30 i

l 3.2 poeillation Tyreo and ModmA I

3.2.1 Automatic Proteerlon_ Design Basis l

The quadrant based APRM system must be capable *f responding to any reasonably postulated mode of stability related oscillations. The system-will generate a trip signal during oscillations of sufficiently low amplitude

- to' provide margin to the MCPR safety limit for all expected modes of oscillations.

These expected modes are as follows:

t

• core Wide - Where all feel assemblies participate h

in-phase with each other in tha oscillation.

• First order Side-by-side - Also called a regional l

l i -oscillation, where neutron flux in one side of the ,

[

reactor oscillates out of phase with neutrop flux in t l

l the other side. The axis of zero oscillation L

magnitude may be at any angle relative to the x-y fuel bundle plane. i o

. - , - ,. , .e, e

TR NO: 060 REV. O PAGE 15 OF 30 First Order Processing - This is a-regional oscillatioi, whors c.he axis of zero oscillation amplitude rotates in the x-y plane, or the two regions of the reactor of peak oscillations amplitude shift from one location to another at a frequency i

slower than the ose111ation frequency.

Oyster Creek has relatively tight (sn.all diameter) inlet  !

orifices for the fuel. The tighter inlet orifico results in a higher single phase frictional oressure drep which is a stabilizing f actor f or chant.el -hydrodynaniic stability.

.The resulting lower channel dacay ratio reduces the probability of regional oscillations, i

3.2.2 Hicher Order _Qttillatien Nodtg '

I The expected modes of oscillation in a BWR are the core wide (in-phase modo) and the first order side-by-side and L pr6 cessing modes.. Higher order modes and the first order

,. ' radial modos (inside-outside) are not'expectad to occur p

l- because of1their large cigonvalue separation and the i' capability of the APRM to detect the first order mode i

s 6

l l

• e v u's. n+-< e--.,w ,- e -n e s e -w or:,m.--ww- e4 n--em ' n a.4 m e -

w rew, mm,- ,w ,- -s v a, --s we, e n m a ,,mn-, m rv m e m- ,--m--m--m-w,,e,-r~-4 --g-g,- wp-rn,--,-e- y--rw-rr'p-=,-

-- m.m.--. __--- _ . _ _ _ _ . _ . . - _ ~ _ . _ _ _ . _ _ . - - . _ _ _ . _

'l TR NO 96b I

REV. O PAGE 16 Of 30  !

-i' prior to the development of sufficient reactivity feedback t

to oxcite the higher order modes. This conelAsien is >

consistont with stability experience to date, even for tho  ;

i largo amplitude oscillations o, ,rvad St several planta.

Reference 4 provides a more detailed discussion of higher ordet oscillation modes. f 3.2.3 Sinds Channel.,1tydrodynamle Stability

-Pura chanr.el hydrodynamic oscillations are thermal-hydraulic instabilities (no nucinar feedback).

i Instability events, tests and analyses have derao..strated the importance of neutrenic coupling and the tendency for t the neutronics to force the fundamental mode of oscillations. For a single channel to independently. -3 oscillate, significant local repetivity (Sedback must be present to sustain the oscillation and overcome the .i i1 eigenvalue separation between the fundamental and the ll i

single channel oscillation conditions. For the cases j

analyzed with TRACG (reference 1), it was shown that h unrealistie modi cations _to the channul's hydraulie l

E response were required to provide this local reactivity conditieno. Fer cores with hydraulically compatible fuel

= .de s ig n s , it is virtually _ impossible for thene conditions i to occur. Reference 4 provides a more detailed discussion nf.singlo_ channel hydredynamic stability.

l l

r-

. . . a. ;_,- ;_ , _ _ . _ ;___ --..-_--._.._._...-._a_..=_._ . _ _ _ _ _ . _ --- _-_ _ -

. . . _ _ . ~ . . _ _ _ _ _ . . _ . - _ _ . _ . _ . _ .--.._____.m_-_-. ..

n \: i b

TR HO 068  !

" ~

REV.} l PAGE 17 or 30  !

3.3 teeion cone.idpy,,3j;i n g l

. i i

It must be demonstrated t>nt the quadrant based APRM system is ,

t capable of responding te any reasor. ably postulated mode of th'rmal-hydraulic oscillations. The system must generate a trip ,

. signal during oscillations ot' cuf ficiently low ainplitude to provide 1 margin to the McPR Safet,y Limit for all expected modes of l

- oscillation.-

. 3.3.1 pag.ign critEE.1.a

c. The method used in determining oscillatiora W'PR trontains unquantified parameter and calculation uncertainties. As

}

a result,-an additional delta CPR margin was includsd to

-addrene these uncertainties. The additional margin ancluded a 0.10 delta CPR to addreou the calculation ,

uncertainty and 0.03 delta CPR -for paranetar uncertainties. The 0.13 margin vu added to the 0.07 cargin already incorporated-into the safety. limit.

Therefore, 2 design limit Mean of 1.20 was used to 1

ova Aaate, the acceptability of the APRM system tesponse.  !

3.3,2 0D!.blX12LOGEtraint a  !

Thy atability' interim correctiw .attions (ICA's)

- identiflod~1n t % NRU Bt110 tin Nc. 88-07, Supplement 1 i

(rtfereaca !)) have beer, incorporM sed into tha appropriate ,

, oyster Creek plart operating procedures. Au part of the P

d

__2_ - ,

s - d w% e ,- , --e, .-[r+ ~ ~ . . ..r,,, , -e-,-- , ,, re- + 4e tw w,, r. n. p-.,,45-3 .,y _.s-,.me.,, ,. -m,, ,myv .w,-. ,.m. ,e 9,,_e.. .,-e, g....n.r.,.w,,5mv,e e -r.,g'

.. ___ . _. . __._ ._. . _ . _.-._,__..m.._= . . . . . .._._ . _ _ _ _ _ _ _ _ _ _ _ _ _ . . . _ __. ____

TR NO: 068 i P.EV q .

PAGE 10 0F 30 long term solution certain procedural controls vill remain in place as a prudent and conservative operational strategy against the potential for any power oscillations. These oportienal restrictions do not irrpact norrnal startup or power operation at Oynter Croek.

The region of potential instability is deil,ed as the area on the power-flow trip below 40% flow and above the 80s rod line. Procedurally the operators are directed to avoid this region during normal operation. If this region is l

unintentionally entered, the operators are procedurally inrtructed to exit the region. If at any tirne during i

operation within the region a core thermal-hydraul.c instability does occur, tho operators are directed to

.n.anually scram the reactor.

2 I

i

{, -

t

.1 3

T'5'. es' 'gMr==-P'%rw -eeas s-W P=*+wp- -rre"-e 9m"Nrer-' 'P't-'s'+'TT-*+- r"-1

• remsi'-+J- ir

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

i

i TR MO: 068 REV.O  !

PAGE 19 OF 30  :

1 4.0- $UProkTING ANALYSIS i

4.1 overview For the purpose of confirming system response, the capability to' i

simulate a wide variet) of possible oscillations is required.

Development-of this simulation capability in based on both 4 l theoretical understanding of the phenomenon and on existing plant

- test".ata.- - The simulation involves the detinition of the peak bundle neutron flux oscillation and average oscillation magnitudes

- representative of areas monitored by LPRMs. The capability to simulate phase lags in both axial and radial directions ir also provided.

The APRM response simulation begins with a calculated 3-dimensional initial power distribution-(reference 5) and corresponding steady .

- 5 state LPRM readings. For global and regional oscillations LPRM p readings are simulated by superimposing the oscillation contour l_ (defined as the distribution of occtllation' magnitudes) or, top of l

the initial steady-state readirga.- Due to the propagation of a pressure density wave up the fuel channel a ninety degree axial phase' lag over the active fuel length'is applied to the A/B/C/D'LPRM detector response. The specified APRM channel reading is thon I determined by averaging the ausigned LPRM signals that femd that h channel. The APRM channel signal is evaluated by comparing it against the-flow-biased scram setpoint and the oscillation magnitude can be adjusted until the APRM channel signal reaches the setpoint.

i

-__. .m-O P y mWW- *T Pfr r ymyr y <y- - gew q er y'ggyp*".r--tc.gg p-4N"W'M- -'rWWWt='rt*3-= = * -"p'"'7TT P-M'FWPN* 7*'f*-W'9F=--**Cp**WFY7TS f*tt."T6*- M'

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

TR NO: 068 REV. }

PAGE 20 OF 30 Ldsed on TRACG analysis CE has estimated the thermal margin response during oscillations in terms of delta CPR divided by initial CPR (6 CPR/ICPR). These resulte have been correlated to peak bundle neutron power as shown in Figure 4.1.

j- The resulting CPR/ICPR characteristics can be applied to various initial conditions to estimate the minimum CPR for a given magnitude of oscillation. Ti.e initial (pre-oscillation) CPR, however, depends c.~ a nun.her of otheir rarameters uh! h may in lude plent opera' 'ng P

etere (core power and flow), margin to c = rating limite and operating limit MCPR. For conditions follow.ng a runback f the recirculation pumps, initial CPR w'll also depend on McPR change as  !

a function of L' low and on the magnitude of the flow char.gr. A wide range of ICPRs is therefore_possible and has been taken into consideration for this analysis.

The cycle 12 loading pattern and core conditions were used to perform this vialysis. The thermal-hydraulic performance ~of future fuel designs is expected to be similar to those designs used in cycle 12. Reasonable variations would net be expected to have a

-significatnt impact on the results of this analysis. However, each new fuel design introduced into the Oyster Creek core will be evaluated to determine its impact on t he conclusions of this report.

i l

v L .- -- . . - _, ._ . _ . _ . _ , , _ _._;

E TR NO: 068 l Rev.1 PAGE 21 OF 30 l 4.2 global Otti 11ations I The APRM response simulation begins with an initial power i

distribution represented by the steady state LPRM reading). For the core wide oscillation, each LPRM is then given the same normalized i oseillation megaitude. As the oscillation magnitude increases, the parameters of the simulation are varied to enoure that the ,

non-linear behavior of the signal is properly modeled (i.e. as the oscillation magnitude increases the peak of the signal is further i

above the average then the minimum is below the average, since the .

signal can never be zero). The specified APRM channel readings tre p- then determined by summing the assigned L' AM signals that feed the respective APRM channel. This APRM channw. .,1gnal can then be  ;

evaluated by comparing it against the flow-biased scram setpoint.

Various initial power distributions, oscillation magnitudes, and f ailei hypassed LPRM patterns can then be evaluated. -

'x as ,

c Core-wide oscillations do not present a challenge to the MCPR safety 74

~

limit. The quadrant averaged APRMs are very responsive to core-wide 3e

}p;'

oscillations and initiate reactor' protection with adequate margin to (E' J

the safety limit. - Under the most restrictive detector $

-out-of-service configuration allowed by the Technical specifications a reactor scram is initiated at a peak bundle neutron power of approximately 150s. Applying figure 4.1, this corresponds to an MCPR of approximately 1.45, well above the design limit of 1.20.

e- er e-m.er-,, -.er.+ee m--..=-1-e se sw- c'- --c.e-,,.ve.'.:c. m, v .,- c - w we E w-e- r -e-< <=.e-v=- w- a -d ma v' w ir -v - d v e w *i- % **wy --mr e- t* w w- -*gr'-- 9

• N W m- -

""t

TR NO: 068  !

REV. ] ,

PAGE 22 0F 30 l 4.3 Pecional Oscillation -f l

For regional oscillations, the simulation is basically the same as described fot the core wide oscillation with the difference being I 1

the distetbution of oscillation magnitude and relative phase angle.

Estimates of-the contour were determined:from data measured during oscillation events at otner BWR's (reference 1). Figure 4.2 shows several first order side-by-sido contours that were used in the .

bundle and nuclear instrumentation response esiculation. The additional information provided by the contour is the line of ,

E symmetry about which the signal of the harmonic component of the neutron-flux changes.

?

To evaluate different absolute magnitude regional oscillations, the 1

. base oscillation contour of Figure 4.2 (case 1) was modified by multiplying each region's normalized magnitude by.a constant value.

The oscillatien contour was superimposed over the steady-state LPRM '

response and the oscillation magnitude was increased antil the APRM ,

responor' oxceeded the trip setpoint. The peak bundle :wci11ation magnitude was cal'eulated assuming a radial peaking factor of 1.45.

This value-is considered conservative for the core conditions under-which oscillations are expected to occur and is' supported by plant operating data. From this information and the application of Figure L4.1 the oscillation HC*R was determined. Various initial power -;

distributions, core average exposures, oscillation magnitudes, and failed / bypassed LPRM patterne have been evaluated for half-core f

. .-:--.-._-_.=__-.....-.--.-.- - . . - . - - . -

TR NO: 068 REV. 1 PAGE 23 OF 30 regional oscillations. The evaluations considered fuel thermal performat.ce as it relates to the APRM and peak bundle oscillation magnitude responses.

9 Table 4.1 provides a summary of the regional analysis which assumed all LPRM d^tectors which feed the APRM system were operable. The results show the transient MCPR at the APRM reactor -Jeram setpoint for Cycle 12. The MCPR results decrease with increasing cycle exposure and power level with the most limiting condition occurring at end-of-cycle, 50% power and 30% flow.

Table 4.2 summarites the results of the detector out-of-service study. The analysis was performed at end-of-cycle, 50% power and 30% flow conditions. APRM channel response is more sensitive to the availability of the A and B level detectors than to the C and D level detectors. APRM channel rasponse significantly improved if it can be assumed thr.t at least one channel-(per RPS system) responding to the oscillation has no more than one A level or B level detector out-of-service (detector configuration (3) in Table 4.2).

An evaluation was performed assuming vnrious operating MCPR limits to determine the required APRM system response in order to ensure the etability design limit MCPR of 1.20 is not exceeded. Table 4.3

.provi des the results. The Oyster Creek Technical Specification scram setpoint at 20% flow is 781. Table 4.3 shows that an operating limit MCPR below 1.47 would require a enange to the Technical Specification scram line to avoid the possibility of a safety limit violation if regional oscillations occur.

t

4 TR NO: 068 REV. O PAGE 24 0F 30 m en V' ,d I TRACG, HOT CHANNELS 0 TRACG OTHER CHANNELS 0 46' - ]

+

NEDE-22277-P-1 g l

+

i e .

a e

' 5 0.30 - i E

e a a ,

U

.t

.- a g

d 0 2* -

o 8 t 0

0 0 s

0.10 - 0 0

+

0.00 , , , , ,-

0 100 200 300 400 50C

' PEAK OSClLLATION MAGNITUDE L(7. OF RATED)

Figure 4.1 Thermal Margin Response Ouring Oscillations.

(Re ferer.ce 6.1) r

..<--,-y=.w u.---xe+-v+,,4 -c,-~+.yem --+--r Wrr- v---m -

.,e- - rr e -

--*-+e-+ - - --

+-r*+---+-*-h-wte-.r=-w-

TR NO: 068 ,

t REY. O  ;

PAGE 25 0F 30 l.

t I

80 ~

o 70 1

• l

.~.

60W ( i CASC $

} '

$ 50-40-

6 30-5 20a _

CASC 1 h < [ t 10 s w>

; Jl 00 ~

1 7 9 7 5 - 6 8 10 8 2 CHANNCL CROUP NUW9(4 i

8 1

8 6

5 7

l 9 7 e Li ne o f Symme try 1

Figure 4.2 Input Oscillation Contours.

(Re ference 6.1) l

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

TR NO: 068 REV.0 PAGE 26 'OF 30 Ib,DLE 4.1 4

MINIMUM CPR AT SCRAM SETPol!!T DURil10 PEGIONAL OSCILLATIO!_JS PEAK POWER ICPR (ADJ) ANPLITUDE MQEB

. DOC 35% 2.47 246 1.67 DOC 50s 1.79 199 1.32 MOC 35% 2.45 260 1.63 MOC 50% 1.75 203 1.29 EOC 35s 2.46 249 1.66 EOC 451 1.96 211 1.39 EOC 50% 1.79 _215 1.26 NOTE: (1) -NO APRM/LPRM DETECTOR 9 OUT-OP-SERVICE (2) RESULTS ARE BASED Oil RESPONSE OF QUADRANT-3, APRM CHANNELS 3 AND 7

ICPa(adj) = The reduced power ICPR is adjusted to reflect full-power ICPR on limits L

L b

TR NO: 068 REV. 1 PAGE 27 OF 30 TABLE 4.2 PJTECTOR OUT-OF-SERVICE STVpl till!111Vli_C17 AT SCPAM SflPOf t1T OF 781 QMJtll10 PECIONAL OSClLLAI.LQJ1E OUT OF DETECTOR APRM SERVICE PEAK CQNFIGURATION CJL@l!LIEL [lETECTORS ICPHfADJ) h!iEQIgl]I tiCF2 (1) 3 A1, A2, A3 1.79 249 1.20 (1) 7 B1, B2, B3 1.79 298 1.12 (2) 3 A2, A3 1.79 220 1.28 (2) 7 B2, B3 1.79 269 1.16 (3) 3 A2, cl, C3 1.79 152 1.46 (3) 7 B2, D1, D3 1.79 239 1.22 NOTE OPERATING LIMIT MCPR = 1,51 (CYCLE 12)

-a

- - .._ __- _ . . . . _ . _ . _ - _ . ._.____..,_.m...___ ._-_._..m.__._._._______._

6-TR NO: 068 REv. 1 PAGE 28 OF 30 TABLE 4.3 ffRAM SETPOINT RESULU OPERATING-LIMIT 11018  !

1.51 (PEAK OSC) 250%

(SCRAM STPT) 82%

1.49 (PEAK OSC) 240%

(SCFAM STPT) 80s 1.47 (PEAK OSC) 230%

(SCRAH STPT) 796 1.45 (PEAK OSc) 220% .

(SCRAM STFT). 774 NOTE: (1) RESULTS. PROVIDE FLOW-BIASED SCRAM SETPOINTS REQUIREMENTS ASSOCIATED WITH 30% FLOW. .

(2) DETECTORS OUT-OF-SERVICE CONFIGURATION (3) OF TABLE 4.2 WAS USED.  ;

.(3) RESULTS ARE BASED ON APRM CHANNEL 7 RESPONSE.

i (4) DESIGN MCPR LIMIT = 1.20 4

.._-__ .== - _ - -

1vA e .w ev y--pm-stec 4. y us g m-w sg-- y *eHM pv M - WPVV T9.wIam emg

I TR NO: 068 REV. 2 i PAGE 29 of 30 [

$.0 IECHNICAL Sflgif! CATION Dased on the stability analysis results provided in this report, the following additional Technical Specification provisions are required:

fection 3.1.D.3 Except during the performance of Technical specification required LPRH/APRM surveillances,.

reactor power shall be reduced below the 80% rod-line or the corresponding RPS trip system shall be >

placed in the tripped condition, whenever all J three of the following conditions exist:

^

1) Reactor power is greater than 35%

- and -

2) More than one LPRM detector is bypassed or failed

- in the A level _or__._the D level assigned to'a single APRM channel.

- and -

-i

- 3) - The diagonally oppocite quadrant contains n'aingle '

APRM channel with more than one bypassed or failed l_ LPRM' detector on the same axial lovel as thel l

L _ bypassed or failed detectors'specified in (2) l -:

above.

L' t

l-  ;

L Section 3.10.C_ .The MCPR' limit for each cycle as identified in the i COLR shall be greater than or equal to 1.47.  ;

b L

i e - --

TR No: 068 REv. o ,

PAGE 30 0F 30 -l 6 '. 0 BIFEREHerg

1. 0.A. Watford, " Fuel Thermal Margin During Core Thermal ,

Hydraulic oscillations in a Dolling Water Roactor", NEDO-31708, i June 1989. f

2. " Power Oscillations in Boiling Water Reactors (BWRs)," NRC I

Bulletin No. 88-07, June 15, 1980.

3. " Power Oscillations in Dolling Water Reactors (DWRs)," f t

Supplement 1, December 30, 1988.

4. Letter, S. D. Floyd (BWROG) to A. Thadanni (NRC), " Licensing Basis for Long Term Solutions to DWR Stability", February 2, 1990.

5. ' Letter from A. W. Dromerick (NRC) to'P. B. Fiedler (GPUN) CPUN Topical Report TR 021, Rev. O, " Methods for the Analysis of

, Boiling Water Reactors Steady State Physics", September L7, i 1987.

t 1

l-si m

., - . - . . ,