ML20002A378

Safety Evaluation Supporting Amend 43 to License DPR-43

ML20002A378
Person / Time
Site:	Yankee Rowe
Issue date:	08/25/1977
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20002A376	List: ML20002A375 ML20002A377 ML20002A378 ML20002A379
References
NUDOCS 8011140680
Download: ML20002A378 (17)

Text

(

C)

=

SAFETY EVALUATION *BY THE OFFICE OF NUCLEAR REACTOR REGULATION

'~

SUPPORTING AfiENDMENT NO. 43 FACILITY OPERATING LICENSE NO. DPR-3 YANKEE ATOMIC ELECTRIC COMPANY YANKEE NUCLEAR POWER STATION (YANKEE-ROWE)

DOCKET NO. 50-29 DATE:

AUGUST 25, 1977 8 011140 NCD

t

=

=: -

TABLE OF CONTENTS

=

PAGE I n troduc ti on.................

1 Discussion.......................................................

1 e

E v a l u a ti o n.......................................................

2 A.

Nuclear Design...............................................

2 B.

T h e rm al - Hy d r a u l i c.............................................

3 C.

EC C S flo d i fi c a ti o ns...........................................

4 1.

Changes to the Accumul ator Sub sys tems....................

4 2.

ECCS Piping..............................................

6 D.

Accident and Transi ent Analysi s..............................

6

~

1.

LOC A A n dl y s i s............................................

6 (a) E v al ua ti o n Model Cha ng e s............................

6 (1)

Revi sed Definition of End-of-Bypass Time.......

7 (2 ) Use of a Revised Pool Film Boiling Correlation.

8 4

(3 ) Changes to Achieve Flooding Rate StaDilization.

9 (D) LOC A An al ys i s Re sul ts...............................

11 2.

Control Rod Ej ec ti o n Acciden t............................ 14 3.

Co n trol Ro d D rop I nc id en t................................

15 4.

Control Rod ~ Withdrawal Incident, Boron Dilution Incident, Isolated Loop Startup Incident, Loss of Load Incident, Less of Feedwater Flow Incident, Loss of Coolant Flow Incident, Steam Line Rupture Accident, and Steam Generator Tube Rupture Accident....................................

15 5.

Other Accidents and Tran si ents...........................

15 E.

Startup Tests................................................

15 S u mma ry o f F i n d i n g s..............................................

'5 E nv i ro nmen tal Co nsi dera ti o n......................................

17 Conclusion.......................................................

17

+

• *"f %

= ~ '

UNITED STATES k

NUCLEAR REGULATORY COMMISSION

=a y"

h WASHINGTON, D. C. 20555

~5 r:

.o 7=

1

,C

=:

S*?ETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION

=r SUPPORTING AMENDMENT NO. 43

==

'g FACILITY OPERATING LICENFr NO. OPR-3 YANKEE ATOMIC ELECTRIC COMPANY

~

i:

P ~

YANKEE NUCLEAR POWER STATION (YANKEE-R0WE)

DOCKET N0. 50-29 N

Introduction By letter dated January 6,1977 (as supplemented March 11; April 13; May 2; June 30; July 7,14 and 15; August 1, 4, 5, 8, 9 and 22,1977) Yankee Atomic Electric Company (the licensee) requested changes to the Technical Specifications for the Yankee Nuclear Power Station (Yankee-Rowe). The purpose of the changes are to permit the Yankee-Rowe reactor to operate with a reloaded core ' Core XIII). The proposed changes also permit operation with certain ECCS changes, an active ECCS accumulator subsystem, and with ECCS analys's utilizing certatn modeling changes.

g The licensee also proposed to use rod groups C and A for control instead of the currently controlling rod groups A and B, and to increase the inlet temperature lir,'it from 511*F to 515*F.

The licensee also proposed to remove Technical Specifications placing restrictions on power level increases following control rod insertions if the reactor has been at zero power for more than 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> immediately precedir g the power level increase.

Discussion

..e The Yankee-Rowe reactor core consists of 76 fuel assemblies, each having d

a 16.x 16 array of fuel rods. The Core XIII reloaded core will be loaded with a two-region configuration of 36 fresh assemblies fabricated by EXXON Huclear Company (ENC) and 40 mechanically identical fuel assemblies from d

Core XII also fabricated by ENC. The licensee provided an extensive description of the replacement fuel in reports transmitted by letters dated

-JJ July 4 and November 7,1975.

We approved the use of this fuel in the s

Yankee-Rowe reactor in License Amendment No. 21 dated December 4,1975.

j[.

~

d j

-.i uj P90 H M NAL 1

~.;.:

-u=::.

2:

h-_

i 2-

]

The licensee has.provided -and we-have reviewed, nuclear and thermal.

evaluations and transient and accident analyses with the reloaded core E

and the revised control rod designation and inlet temperature limit.

LOCA analyses-for the reloaded core are based on use of a revised defi-nition of end-of-bypass, a revised pool film boiling correlation, and model change to achieve flooding rate stabilization. _ The licensee has

._~~

supported these changes with experimental results and analysis which we

=

have found acceptable. LOCA' analysis.is also based on an active accumu-lator system'which delays accumulator water injection following a LOCA and injects at a higher pressure thereby conserving accumulator inventory

+

and shortening the refill portion of the transient. We have reviewed the design of the revised accumulator system as discussed in Evaluation Section Dl, and the results of analysis based on its use and find it acceptable.

In order to improve flow distribution from the high pressure

,t and low pressure safety injection pumps the licensee has modified the ECCS piping, and has revised his analyses to include the. effects of the modified piping. We have reviewed the design of the piping modification (see Evaluation Section C2) and the corresponding changes in the analysis

~

and find the revised piping arrangement acceptable.

Evaluation A.

Nuclear Design 4

i l

Core XIII reload configuration uses the two zone pattern used in Core XII with new fuel being loaded in the periphery of the core. Core XIII has a higher beginning of cycle (BOC) hot, full power boron content for criticality than the reference cycle, Core XI, which together with dif-ferences in fuel mechanical det4gn between Cores XI and XIII, results in Core XIII having a smaller negative SOC moderator coefficient. Other nuclear characteristics are similar between Core XIII and Core XI. The values used in the accident analysis are chosen in a conservative manner.

for each analysis, however.. Therefore the Cycle XI accident analysis bounds the Cycle XIII analysis except as noted in Evaluation Section D.

Core XIII control is accomplished with rod group C, rather than Group A which was used in Core XI ~ and Core XII.. Having a greater worth than GroSp A, Group C causes the Technical Specification multipliers used to account for-xenon redistribution effects to be slightly more restrictive.

4 1

4

=

-=

^

+>

. p 7

The licensee's calculation of control group worths for Core XIII indicates that there is a substantial excess margin over a 1%

design shutdown margin allowance throughout the cycle life. A 7.5%

uncertainty allowance has been included in the calculation of rod worth.

Startup measurements will provide additional verification

"=

that the shutdown margin will be maintained through Cycle XIII operation.

Except as noted below, the nuclear calculations for Cycle XIII were perforced by the licensee using the same calculational methods employed for Core XII which we have previously found acceptable.

The method of calculating delayed neutron fractions is the same as used for Core XII except for the added inclusion of Pu-241 effects.

The method used to calculate the Doppler defect has been changed to a method which more closely agrees with measured boron letdown curves.

~~

=

The licensee's existing Technical Specifications place restrictions on power level increases following control rod insertions to allow time for xenon maldistributions to decay off.

The licensee has proposed to remove this restriction if the reactor has been at zero power for more than 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />. After 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> at zero power the xenon from past operation has decayed sufficiently that it does not contribute significantly to power peaking. We find this change acceptable.

.S. Thereal-Hydraulic Desicn Methods " -'alysis used by the licensee in the thermal-hydraulic evalua-tion o-

, e XIII have been previously used in support of the reference Core (X1) and Core XII.

Principal features in Core XIII that might affect thermal-hydraulic performance are (1) the allowable core inlet temperature increase from 511 F to 515'F and (2) the changes in power distributions and thermal-hydraulic parameters between previous core loadincs.

The increase in core inlet temperature results in the calcu-lated Core XIII DNBR's being less than Core XI DNER's at the current design linear heat generation rate (LHGR) of 12.9 kw/ft.

To facilitate comparison Detween Core XIII and Core XI the design LHGR for Core XIII was reduced to 12.5 kw/ft. At a design LHGR of 12.5 kw/ft Core XIII design is bounded by Core XI design. The 12.5 kw/ft limit is acceptable for Core XIII considering only DNS analysis.

' ^CA analysis, as discussed in Section Dl, results '- ' more restrictive limit, wnich is controlling.

The licensee's reference thermal-hydraulic analysis for Core XI snowed that the DNB ratios predictea for a range of abnormal operating transients all exceeded 2.0 for design hot channel conditions of 12.9 kw/ft and F;g of 1.81.

==

"I'.:

The licensee's LOCA analysis for Core XIII results in lower limits for the hot channel conditions than the design conditions used for the therml-hydraulic design of Core XI.

Because the operating limits for Core XIII are more restrictive (due to the restrictions imposed iM:

by the ECCS core cooling performance analysis) than those previously approved by us for Core XI, we find the thermal-hydraulic design of Core XIII to be acceptable.

=

C.

ECCS Modifications 1.

Changes to thi Accumulator Subsystems The licensee contends that the present performance capability of Yankee-Rowe is restricted due to certain operating limitations imposed by the ECCS accumulator subsystem.

To remove these limitations the licensea has proposed to make changes related to the pressurization /depressurization and inventory of the accumulator portion of the ECCS.

For the large break LOCA, the proposed modifications reduce the time required to reflood the bottom of the core by:

(1) increasing the accumulator flow rates by means of higher operating pressures; and (2) assuring adequate accumulator inventory by utilizing its full volume and by an increase in the delay time for its pressurization.

=

The increased delay period of about six seconds for pressurization of the accumulator reduces the loss of inventory through the broken loop during the time that the reactor coolant system depres-surizes to a pressure equal to the operating pressure of the accu-mulator for the large break LOCA.

After a timed period of about 7 seconds from actuation of the safety injection signal, nitrogen would be admitted to the accumulator to build its pre;sure up to t

473 psig within twc (2) seconds (compared with tne 337 psig operat-ing pressure in the present system). The increased pressure in the proposed modification would provide a greater driving force for injecting water into the core at a faster rate following the postu-lated LOCA.

When the water level in the accumulator drops to a low level, four accumulator level switches and their auxiliary relays are used to vent the accumulator to the containment and close valves isolating the fill and discharge lines which prevent nitrogen admission from the empty accumulator to the reactor coolant system.

t n

-r

-,. ~ -,

4 t

=:

4 h

_s_

r The accumulator discharge valve and the nitrogen admission valve to the accumulator close while the relief valves bleed

=~

off the accumulator pressure.

The individual components of the accumulator subsystem will be periodically tested on a g.

schedule consistent with the ECCS to ensure that they perform as designed. The licensee has made a commitment to test the four acct:mulator time delay relays on a monthly basis. The accumulator time delay relay circuitry design changes required to permit testing of each time delay relay during power opera-tion will be submitted for NRC review.

The acceptability of permitting power operation without monthly verification of each accumulator time delay relay rundown-time during the next six months is based on the following:

(1) a preoperational test will be performed and will not be considered acceptable unless each

~

time delay relay operates within the limits specified in the Technical Specifications for five consecutive tests; (2) the time delay relays (two per channel) perform a redundant channel function; and (3) identical time delay relays ( Agastat) have been used for several years at Yankee-Rowe and have operated satisfac-torily during the performance of periodic (18 months) functional rundown-time tests.

In addition, a complete preoperational functional test will be 7erformed to ensure proper operation of the modified accumulator subsystem. The Office of Inspecti.on and Enforcement will confirm that test results obtained by the i.

licensee prior to reactor startup for Cycle XIII operation are adequate to demonstrate proper system functioning.

The proposed modification to the accumulator subsystem includes the addition of piping, valves, nitrogen bottles and a small expansion tank to the accumulator and regulating portion of the ECCS.

Th accumulator volume will be increased from 800 FT3 to 850 FTg and the system operating pressure will be increased from 337 psig to 473 psig.

This operating pressure increase is below the accumulator design pressure of 550 psig, and below the expansion tank and piping design pressure of 720 psig.

The licensee states that all major equipment has been designed and installed in confomance with specifications which are equal or superior to those utilized for the design and installation of the existing portion of the ECCS, and that the effects of water hammer, pipe whip, accumulator overpressurization and flooding have been considered and found not to significantly alter the hazards description in the Final Hazards Summary Report.

k'

L

.j

/

r.

. q n.;

Based on our review of the thermal-hydraulic, mechanical, elec-

=]

trical, instrumentation and control aspects of ~the design, we

=;

find the proposed modifications to the ECCS accumulator subsystem lj acceptable. The bases for this determination are that:

(1) the 9

proposed design permits lower peak clad temperature fcilowing the

.y postulated LOCA; (2) the proposed modification provides conformance j

to the NRC's rules and regulations relating to the ECCS perfonnance requirements (General Design Criterion 35 in Append x A to 10 CFR Part 50 and Acceptance Criteria for ECCS in 10 CFR 50.46); (3) the 1

proposed design satisfies the single failure criterion; (4) the

]

individual ECCS/ accumulator components involved in this modifica-tion are testable; and (5) no operator action is required. This amendient revises the plant's Technical Specifications to include for the accumulator system appropriate requirements for periodic testing O

and for limiting conditions for operation.

We conclude that the proposed modifications of th: ECCS accumulator subsystem is acceptable.

2.

ECCS Piping a

In order to improve flow distribution from the high pressure and lcw pressure safety injection pumps when a rupture occurs in one of the safety injection branch lines, the licensee has modified the piping to separate the high pressure and low pressure safety injection distribution headers.

A new check valve is installed in each 4 inch safety injection branch line upstream of the Junction from the high pressure line.

The check valve prevents the high pressure injection from backflowing toward the broken branch line. Gloce valves installed 'n each of the high pressure injection branches are throttled to maintain a specified back pressure (determined by flew testing) in the high pressure distrioution header. The back pressure is sufficient to overcome reactor coolant system pressure which assures injection through the intact safety injection branch lines. The gicbe valves are fitted with locking cevices to maintain the valves at the specified se tting s.

Technical Specification surveillance requirements nave been added for confirmation of ficw distributions, and in addition the licensee has been required by Technical Specifications to verify throttle valve settings the first cold shutdown following Cycle XIII startup.

b

L We have reviewed the mechanical and thermal-hydraulic aspects of the proposed changes.

The licensee states that all major equip-ment has been designed and installed in conformance with specifi-cations which are equal or superior to those utilized for the design and installation of the existing portion of the.CCS, and c

that the effects of water hammer, pipe whip, and flooding have been considered and found not to significantly alter the hazards description in the Final Hazards Summary Report.

Based upon our review, we conclude that the proposed modificaticn to the ECCS conforms to the codes and standards utilized for the original ECCS design, and provides sufficient safety margins.

Therefore, this modification is ac:eptable.

D.

Accident and Transiant Analysis 1.

LOCA Analysis (a) Evalustion Model Changes Changes made its the WRD4 - based evaluation model that was used for Core XIII LOCA analysis were reviewed to assure that the revised model conforms to the requirements of 10 CFR 50.46 and Appendix K.

In addition to the noding changes required to describe the piping modifications the model was revised with respect to: a revised definition of end-of-bypass time; use of a revised pool film boiling correlation; and modeling changes to achieve flooding rate stabilization.

(1) Revised Definition of End-of-Bypass Time Appendix K to 10 CFR Part 50 requires tnat "For postulated cold leg breaks, all emergency cooling water injected into the inlet lines or the reactor vessel during the bypass period shall in the calculations be suotracted from the reactor vessel calculated inventory. This may be executed in the calculation during the bypass period, or as an alternative the amount of emergency core cooling water calculated to be injected during the bypass period may be subtracted later in the calculation from the water remaining in the inlet lines, down-comer, and reactor vessel lower plenum after the

~

=

~"

g.u;;p

.J 1

.J

• =:..

er bypass period".. Yankee-Rowe utilizes the lattar ffE E:

approach to calculate the amount of ECC water to

.=='==

discaro. The old definition of end-of-bypass re-quired discarding that water which was in the cold

(( j([

leg piping at the. time of sustained downflow.

However the cold leg piping is full at the time of

~

==

sustained downflow so that even though.the required eg.= -

amount of water is still removed from the system,

~~~

the t'.me required to start water flowing from the cold leg into the downcomer is reduced which reduces

~~

the time required to refill the lower plenum. The e1=

proposed definition of end-of-bypass removes.some E= =E -

conservatism in the previous definition of end-of-

-,_;=

bypass; however, the proposal is still conservative and in conformance with Appendix K.

The assumption that the cold legs are filled with a

"?S plug of ECC water is substantiated by tests at CE,

' 53 Westinghouse, BMI and CREARE.

The ECC water injected during bypass for Yankee-Rowe is much greater than that calculated to be required to fully quench steam in the cold legs. Therefore, the assumptions in the Yankee-Rowe analysis are deemed to be reasonable and j us ti fi ed.

We conclude that the proposed method for calculating end-of-bypass time for Yankee-Rowe is an appropriate change to be incorporated in the Yankee-Rowe Evalua-

=

tion Model. We-find the proposed definition of end-of-bypass to be acceptable.

(2). Use of a Revised Pool Film Boiling Correlation During initial Core XIII Yankee-Rowe LOCA analysis of a 2-1/4 inch ID break (small break), the peak cladding temperature predicted obtained from the T00DEE2 heatup calculati]ns appeared to be unrealistic.

In that analysis, it was discovered that the temperature of the pin at a location which is-immersed in water for a long period of time was predicted to be higher than that of a location producing the same power but which

..m; had uncovered sooner.

In the absence of a pool boiling

=.

h

=

s.

.e e

i e

,,... ~,

~

~

=:;

_g.

correlation, T000EE2 calculates heat transfer coefficients using the Dougall-Rohsenow correlation.

Use of the Dougall-Rohsenow correlation at low flow rates results in the use of a prescribed minimum 2

heat transfer coefficient of 5 BTU HR-FT

• F, which results in 'a calculated T of 1500*F from the cladding to the coolant with a coolant temperature of 500*F.

It was concluded that the use of the Dougall-Rohsenow for near stagnant conditions is too conservative. The mass flux less than 20,000 lb /HR-FTg correlation for licensee proposed a pool film boilin m

1 Yankee Atomic proposed to use the Ellion pool boiling correlation, using as the boiling length term in the

=,.

corralation, the total heated length of the fuel. The licensee compared the Ellion correlation, used in this manner to available experimental data and showed that it conservatively bounded the data. The data used in the comparison includes test conditions and geometries similar to those under whi-h the correlation will be applied.

The Ellion correlation, as applied, was compared with the Bromley correlation, Berensons correlation, the Pomerantz correlation and the modified Bromley corre-lation; the Ellion correlation resulted in the minimum heat transfer coefficient prediction.

We find that the Ellion correlation as modified by Yankee Atomic; i.e., using the total pin heated length term in the correlation, conservatively bounds cvailable data for pool film boiling. The conservatism in this modified Ellion correlation is larger than that in an already approved application of the Ellion correlation which is also considered to be conservative. Therefore, the Ellion correlation, as modified by Yankee Atomic in YAEC-ll31 (submitted June 30,1977), is an acceptable pool film boiling correlation for use in the Yankee-Rowe Evaluation Model.

(3) Changes to Achieve Flooding Rate Stabilization The licensee computed unrealistic reflood rate oscill-ations following the initial oscillatory interval when water first enters the core. These oscillations occur R

4 k

i

!~

z--

% =

=.

=; w..

approximt,ely 25 seconds after the octtom-of-core-

..l.?l.".

recover; (80CREC), and follow an interval of stable core s.?Fi; reficoJing.

The oscillations result from the quenching x=r of discreta fuel segments, and are more the result of

2.,

fuel modeling than of heat transfer modeling.

The discon-tinuity in the fuel quenching model arises from the re-quirement for modeling the core fuel in a limited number of axial segments, with each segment responding to heat addition or removal uniformly over its length. As a

result, a process occurring around a moving plane in the m.....

core, such as a moving quench front, which occurs over

~

a shorter length than the axial fuel segment will respond in a discontinuous manner.

The heat tranfer changes from 30 to 1000 BTU /HR-FT _op ig 2

when a fuel segment is comrated to quench. Thus, steam generation transients occur which result in upper plenum pressure oscillations, and in turn produce the reflood

. =

rate oscillations computed by the model. The introduc-tion of these upper plenum pressure oscillations can be eliminated oy use of a fixed core outlet enthalpy as an input condition for the upper plenum balance equations.

To obtain conservative upper plenum pressure conditions during reflood the licensee proposed use of the steam generator secondary fluid temperature as the reference temperature for the fixed core outlet enthalpy for the reflood calculation.

The calculated core outlet enthalpy during the quasi-steady state intervals of the reflood transient is below the steam generator secondary reference temperature enthal py. Assuming the steam generator secondary enthalpy for the core outlet provides conservative input conditions for ccmputing upper plenum pressure during reflood, and results in a somewhat conservative calculation of core reflood rate relative to the case which calculates the 1

exit enthalpy.

A sensitivity study performed on the fixed core outlet enthalpy selected shows that the reflood rate oscillations are avoided by this assumption, that peak clad temperature (PCT) sensitivity to this parameter is negligible for enthalpies above saturation for the computed core outlet pressure, and that the PCT increases with increasing core outlet enthalpy up to saturation en thal py.

f g

x....

se L' ;;..-.

Experimental results of reflood quenching have shown this raj??

process to be highly oscillatory in small scale tests

..?!

designed to simulate reactor core and downcomer resoonse

• =9]

in the reflood process.

Si:ch oscillations were observed-

..]

in the FLECHT-SET *. experimental program. A prediction of

s this experiment was performed using the time-averaged

.s (non-oscillating) reflood rate together with the heat

=?

transfer correlation derived from the FLECHT experiments F

for ccnstant flooding rates.

The calculation of the heat transfer in this manner was conservative with regard to the actual heat transfer derived fr >m the FLECHT-SET experiments (see Figure 3-30, WCAP-8238).

From these considerations, and the sensitivity study perf ormed in the

.s reference YAEC report, it is concluded that use of the fixed core outlet enthalpy to eliminate the unrtalistic i

reflood oscillations computed by the model is a easonable

=

and justified solution to this computational difficulty.

__f We conclude tha*. the proposed use of a fixed core outlet enthalpy assumption for reflood calculatior.s with the YAEC-EM model is an acceptable assumption when using the steam generator secondary temperature evaluated at containment pressure as the reference for selecting the enthalpy value to be used for the

=

cal cul ation.

(b) LOCA Analysis Results The calculational model used by the licensee to evaluate the ECCS cooling performance was previously used for Yankee-Rowe Core XII analysis, with the exct-? tion of the

=:

mecel changes described in the previous section. Because the licensee's LOCA analysis addressed only operation in the 4 loop mode, we have modified the Technical Specifica-tiens to forbid 3 loop operation.

The licensee provided analyses with break discharge coefficients (C ) of 0.6, D

0.8 and 1.0 for both the double ended cold leg guillotine (DECLG) and double ended cold leg split (DECL5) breaks.

The analysis showed the most limiting break to be the 1.0 DECLG, for which the peak clad temperature was calculated to be 1983*F at 97.7 seconds, well below the acceptable limit of 2200*F as specified in 10 CFR 50.46(b).

In

,=e

'wCAP-a238, "?WR FLECHT-SET Phase A Report", J. A. Blaisdell, L.E. Hochreiter J.P. Waring, Decemoer 1973.

S

l W

L i l 21 -

[

1 addition, the maximum local metal / water reactor of less-than 2.0% and the total core wide metal / water 1 reaction of

'5E@

j

'less than 1% were within the allowable limits of 17% and 1%,-

E..

respectively.

Based on this ~ analysis, the. licensee proposed-5.]

j to limit the peak linear Neat generation rate (LHGR) to 9.7

_.5M i

kw/ft for implementing the analysis results.

=

=

'The licensee.provided sma'1 break analysis for a break i

postulated to occur. in a small length of ECCS piping (1 -

Tin. -

j

-to'2 feet) immediately downstream of the check valve which

_ 5 is nearest one of the reactor coolant system (RCS) cold leg

~

1 injection points. This break location results in RCS blowdown f

' through a 2.25 in.- I.D. themal sleeve and ECCS spillage e

1 through a 3.438 in. I.D. ECCS line to containment. With i

'the modified ECCS system the PCT was calculated to be 1124*F for this break.

This break location was considered the a

worst small break for Core XII analysis. The substantial

=E =.

2 margin between ll24*F and the acceptable limit of 2200*F.

_ " =

1 provides assurance that other small breaks will be

~J.. =..

I acceptaM e.

However, the licensee has commit 9d to provide

~~

j confirmatory analysis demonstrating the acc4 ability of other small breaks within one month. These analyses will reflect j

the flow characteristics of ECCS pumps, piping and throttle j

valves as measured during Core XIII preoperational testing.

~

i Because the allowable peak LHGR for fresh fuel of 9.7 kw/ft j

corresponds to approximately 90% of full core power at BOC for i

Core XIII,. the licen;ee intends to.perfom a burnup sensitivity

~T j

study as soon as possible. Until this burnup study. is approved i

the. licensee will limit the allowable total LHGR to 9.7 kw/~ t

. ~~-

f for bo'.1 fresh and recycled fuel.

Operation by this limit for j

. a maxmm of ~225 effective full power days (EFPD)- is conservative since:

}

(1) the fresh fuel is limiting at BOC conditions due to the

=

l significant' pellet to clad gap and the resultant high j

stored energy in the fuel pellets, and; 4

i (2) the burnup dependent allowable LHGR for Core XIII is l'

expected to behave as it did for Core XII.

Examination of Figure 3.2-1 of the Cycle XII Technical Specifications i

reveals that the allowable LHGR did not become more 3

limiting than the fresh BOC limit until. approximately 340

~

]

EFPD (or 87% of full power core life).

a

=

j-We have added a Technical. Specification requiring that reactor

=

operation may proceed after 225 EFPD's only after appropriate

~

LOCA analysis has been approved by the NRC staff.

.'~

j.

3

.=

a :.

n.,

.,-rw.,

,~~.-w

,n

t

' ::.22.

Recently the staff has noted that in LOCA calculations for some

""==

PWRs, a decrease in primary coolant inlet temperature has re-sulted in a predicted increase in PCT.

In discussions with the g

PWR vendors we have learned tPat they have all observed this

='

~

trend while performing LOCA calculations with their individual approved evaluation models.

In the past, it has been widely accepted that it was conservative to assume the highest possible

= = = -

initial coolant temperature for LOCA calculations (typically maximum full power operating temperature plus 4*F for measure-ment uncertainty).

The apparent cause of this behavior stems from the fact that a reduction in coolant inlet temperature results in a reduction in the coolant saturation pressure.

This decreases the flow rate from the vessel side of the break after the short period of subcooled blowdown. This reduced flow, for the postulated cold

~"

leg break, decreases the magnitude of the downward flow rate

"~

through the core that exists for a large portion of the blow-down period.

This decreases the he transfer coefficient and consequently less stored energy is r,,noved during blowdown.

Reducing the coolant inlet temperature also changes the flow rate from the top of the vessel to the hot leg and out of the break through the steam generator and reactor coolant pump.

The change in hot leg flow caused by a reduction in inlet temperature tend to decrease the core flow rate during the period of positive core flow. This also leads to the removal of less stored energy during bl owdown. Thus, the fuel temperature is higher at the end of bypass. Most PWRs exhibit PCT during reflood, and entering the reflood period with a greater fraction of stored heat remaining after blowdown may cause an increase in the PCT.

It has also been observed that the decreased negative core flow may extend the time to end of bypass. Then in the evaluation model more accumulator water is assumed to spill out of the break.

If, as a result, there is insufficient accumulator water remaining to fill the downcomer, reflood will be delayed.

This will also contribute to the increase in PCT.

However, a reduction in coolant inlet temperature may not 5

always result in an increase in PCT.

It has been observed that if the clad rupture location changes to a different elevation where the core power is less, PCT may decrease.

m.we e m y

.g.

.c L

,:~~.

14 -

=

- = =

In addition to the predicted changes in blowdown core flow and heat transfer,' reducing coolant inlet temperature also causes a slight reduction in containment back pressure during reflood.

Reducing this pressure is known to result FI in lower reflood rates with correspondingly higher clad temperatures. However, the effect due to containment back pressure is minor compared to blowdown core flow and heat transfer effects.

At this time the staff believes that nominal value! of inlet temperatures and steam generator shell side steam e onditions should be used in all LOCA calculations since the e ffects of variations in inlet temperatures and steam conditi.ns on PCT are not consistent and are at best second order effects.

The maximum sensitivity of PCT to inlet temperature that the staff has seen to date shows that for a 1*F decrease in inlet temperature the PCT following a large break LOCA would increase 4 *F.

However, a reduction in inlet temperature results in a corresponding reduction in core average temperature and steam generator shell side steam pressure.

A reduction in steam generator secondary pressure results in lower PCT.

If we now assume a decrease in coolant inlet temperature and adjust the steam generator shell side steam conditions, the PCT for

'ankee-Rowe would increase less than 20 F.

Thus we consider this a minor second order effect. We conclude that the cur-rent ECCS analysis on file for Yankee-Rowe meets the criteria of 10 CFR 50.46 and is therefore acceptable. However, our review of the matter has enabled us to identify areas in which there may be additional improvements in accuracy of modeling coolant inlet temperature and steam generator steam conditions.

We are currently seeking additional information to be used in a generic evaluation of the effect of coolant inlet temperature and steam conditions on ECCS performance.

t The licensee has agreed to submit an ECCS analysis of the worst break LOCA assuming coolant inlet temperature and steam condi-tions equal to their nominal values.

Nominal inlet temperature and steam conditions as used hers refer to the most probable 4

4 s.

't m

\\

=::

~;i-g, l '

I values.for the plant when it is operating at 102% power. This AE; will confirm the applicability of the PCT sensitivity, stated above, to the Yankee-Rowe plant and the effect of the use of most probable values for these nominal conditions.

l Based on our review of the Yankee-Rowe large break and small l

break LOCA' analysis, and considering the effect of coclant 3

inlet temperature variations, we conclude that with the Tech-nical Specification limits acccmpanying this license amencment operation of Yankee-Rowe reactor with Core XIII will be in compliance with 10 CFR 50.46.

y; 2.

Control Rod Ejection Accident

=e By application dated August 1,1976, the licensee submitted analysis l

for the rod ejection accident for Core XII in support of license Amendment No. 31 issued November 23, 1976.

This analysis bounds the

~32 i

full power control rod ejection accident for Cycle XIII.

An analysis for the zero power control rod ejection accident was performed for i

Core XIII, and it was found that no clad damage nr fuel melting would i

occur as a result of the accident.

We find that the licensee's analysis and results are acceptable.

3.

Control Rod Drop Incident The licensee's bounding analysis of the control rod drop incident indicates that damage would not result from this incident even if no drop in core power were assumed.

We find that the licensee's analysis' and results are. acceptable.

4.

Control Rod Withdrawal Incident, Boron Dilution Incident, Isolated Loop'Startup Incident, Loss of Load Incident, Loss of Feedwater Flow Incident, Loss of Coolant Flow Incident, Steam Line Rupture Accident, and Steam Generator Tube Rupture Accident Transient and accident analyses were performed for the reference cycle (Core XI) using the most limiting parameters during the core life. For the reload cycle, the input parameters are more favoraole than for the reference cycle. Therefore, the results for the reload cycle will be bounded by those for the reference cycle.

We find this acceptable.

5.

Other Accidents and Transients The remaining accidents and transients in-the licensee's FSAR are not affected by the proposed core design changes and therefore the l

previous acceptable results still apply.

l

%=

E

~ ~

l l

E.

Startuo Tests

g..

The startup program for Yankee-Rowe Core XIII has been reviewed for

=

completeness and clarity. The program as orginally submitted was discussed with the licensee and additional tests were added.

The startup tests will provide a check on the fuel loading and verify the calculational methods used in determining control rod worth, shutdown margin and power distributions.

To verify control rod worth and shutdown margin the worth of roo groups A, B and C will be measured.

These measurements represent one half of the total rod worth of the core.

To verify the power distribution, measurements will be made at various power levels.

The results of this startup program will submitted to NRC.

Based on our review of the physics startup program we find it acceptable.

~

Summary of Findincs s

From our review of the material submitted by the licensee on the Core XIII

~

reload, including the ECCS cooling performance evaluation, we find:

'

• L :,'

A.

The mechanical design of the fuel, the nuclear and thermal-hydraulic analyses, and the analyses of accidents and transients are acceptable.

. B.

The ECCS cooling performance for Core XIII has been calculated with an approved evaluation model in conformity with Appendix K and meets i

the acceptance criteria in 10 CFR 50.46.

e.-

C.

The acdifications to the ECCS to enhance system performance are acceptacle.

These modifications, as described in this Safety Evaluation, are changes in the accumulator subsystem to acconplish celayed injection, and changes in the hign pressure and low pressure injection piping to improve flow distrsibution.

1 D.

The proposed Technical Specifications, implementing the ECCS cooling performance evaluation results, changing the control rod groups, and j

modifying restrictions on reactor pcwer level increases required due

'~

to xenon maldistributions, provide acceptable limits.

4 6

4 i '

3-.

i

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

n.

.n,

• mi- -

=

l':..

==

r.: -

Environmental Consideration

=2

~

s.

We have determined that the amendment does not authorize a change in effluent types or total amounts nor an increase in power level and will

~~f j

not result in any significant environmental impact. Having made this cetermination,.we have further concluded that the amendment involves an action which is insignificant from the standpont of environmental impact and pursuant to 10 CFR Sl.5(d)(4) that an environmental impact statement er negative declaration and environmental impact appraisal need not be preparec in connection with the issuance of this amendment.

Conclusion We have concluded, based on the considerations discussed above, that: (1)

....~~

there is reasonfole assurance that the health and safety of the public will not be endangtred by operation in the proposed manner, and (2) such activities

=E will be concucted in compliance with the Commission's regulations and the issuance of this amendment will not be inimical to the common defense and -

security or to the health and safety of the public.

~

Date:

August 25, 1977

'W e--.

-+

.p----