ML20199F662

Forwards Response to 971107 RAI Re Indications in Jet Pump Thermal Sleeve to Elbow Riser Welds.Eccs/Loca Analysis Is Provided as Requested

ML20199F662
Person / Time
Site:	Peach Bottom
Issue date:	11/17/1997
From:	Hunger G PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 9711240295
Download: ML20199F662 (8)

Text

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

Met 6en seppert Department g

l

. PECO NUCLEAR gggga,, i A Unit of PECO Energy *#*J* "'"

l November 17,1997 l Docket No. 50-278 -l License No. DPR-56 l i

i

[

U. S. Nuclear Regulatory Commission

'l ATTN: Document Control Desk Washington, DC 20555

SUBJECT:

Peach Bottom Atomic Power Station, Unit 3  !

Response to Request for Additional Information Related to indications in Jet Pump Thermal Sleeve to Elbow Riser Welds 1

Dear NRC Officials:

I By letter dated, October 30,19Fv7, PECO Energy Company (PECO Energy) submitted a 10CFR50.59 analysis concerning jet pump weld indications on 1 Peach Bottom Atomic Power Station, Unit 3. By letter dated November 7,1997,  ;

- the NRC requested additional information to complete their review. This letter '

provides the requested additional information. In Attachment 1 to this letter, each NRC question is restated, followed by the PECO Energy response. ,

Attachment 2 to this letter provides the results of the Emergency Core Cooling Systems / Loss of Coolant Accident analyses as you requested.

if you have any questions, please do not hesitate to contact us. l Very truly yours, j W$tht!f(b fik .

G. A. Hunter, JrJ, h '

Director .icensing -

l l

Enclosure:

Attachments

- oc: . H. J. Miller, Administrator, Region I, USNRC R. S. Barkley, Senior Resident inspector, PBAPS, USNRC R. Rl Janati, Commonwealth of Pennsylvania i

97112 u.iG 9*G  !

59

?" g " %.7 u lpppR(qll ,

l' .,

A-._ ,,.s_ _ 4 - _.

e i

I ATTACHMENT 1 P

I l

r l

l i

t

. Docket No. 50-278 License No. DPR-56 2

l l

l Question - i (1) Provide the test data establishing tne AK threshold. For each subgroup of the i test data corresponding to a specific testing program, provide the associated AK j and K values, and vert Y whether the test was conducted in the boiling water. i i

1 values for the longes (t crack found in the M pump) subjected to flow induc i vibrations under normal operating conditions, and justify the use of the 4K  ;

threshold to the current crack growth evaluation.

l 4 I Response  ;

(1). The AK threshold value of 5 kslVin is based on the test data remrted in  ;

. Reference 1 (Figure 1) and on GE test dets (Figure 2). All of t w tests were .

i conducted in air. The flow induced vibration loading is at approximately 32 '

Hertz at which the water environment effects are expected to be insignificant. _

The calculated value of the R (Kmin/Kmax) is approximately 0.5. Thus, only the '

test data close to this R value were considered. The t i AK values in the ,

fatigue crack growth evaluation were 6 or 7 kalV in. lbcacrack growth fatigue  ;

rate relationship used in the evaluation is besod on the ASME Section XI for austenitic stainless steels in air Appendix environments. C curveThe specific (Figure curveC-3210-1)lected se was for 550*F and an R of 0.5.

. i i'

No specific K calculation was necessary since a bounding stress corrosion  !

cracking (SCC) growth rate of 5x10'in/hr was used in the evaluation. The crack  :

growth contribut ons from SCC and fatigue were added together for a small l assumed time interval (100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br />) and the crack length was then updated. The  !

same calculatiors was then repeated to obtain the flaw length for a specific time interval.  ;

Reference

-[1]. Barsom,1 M. and Rolfe, S. T.,

• Fracture & Fatigue Control in Structures -

Applications of Fracture Mechanics, ' Prentice F all, Inc., Second Edition (1987).

L QuantiGD

~

_(2)_ Explain the simplified finite element model (FEM) which simulates the cracked jet pump. Also, provide the criteria that were used to establish the equivalence of

the simplified model to the full FEM for the cracked jet pump. ' Confirm that the - 3 simplified model has considered the 13-inch long crack.  ;

-w_ . _ _ _ _.= .=. _ _ _ = _ _ . . _ .__u_. - ua

Renoonse (2) Compliance due to the presence of a circumferential crack of approximately half the circumference length was first calculated using the EPRI fracture mechanics flew handbook of Zahoor, et al. This compliance was introduced in the beam model of the jet pump as a spring at the location of the observed cracking. The modal frequencies of the uncracked jet pump model and the simulated crack jet pump model were then determined. A comparison of the two sets of modal "requencies showed le-s than 3% difference indicating that the presence of a circumferential crack of up to half the circumference o the thermal sleeve pipe does not significantly change the overall response of the jet pump assembly.

Question (3) Provide information, preferably from test, regarding the relationship among the stresses used in your stress intensity factor calculations, jet pump flow rate, and the percent of rated power.

Response

(3) The cyclic stress magnitudes are in theory proportional to the square of the drive flow. Thus, at 80% drive flow, the stress magnitudes are expected to be 64% of those at 100% drive flow. The flow squared relationship is confirmed in the strain data from Browns Ferry 1.

The stress intensity factor range (AK) values are directly proportional to the peak-to-peak stress range.

The drive flow is the key factor in influencing the vibration stresses, and is controlled to assure riser integrity. Reactor power is an outcome of the drive flow selected.

..,,-,-4 -. -

r.,, --. . -,,-. ,e - , - . . v.--.,.,.-..-n.n -

_, - - . - -~ , ,c

2030 ILZDS ON 1H/11) 80191 NON 40/41/11  ;

e e

~ * * ~ * " ~ ~

i i i iiiiI I I I et,, e s.e (i.e. teal P884*91 ,

y,g , , ,

1 As. e 8 88.VGIE

• . POS 98 9.i g . .ep a . ... , . .i.

e 8.0 = 0 of -

n a Co .e i .. - . -

a ... . k. ,* .

i ..

g El

,. e m # MAD STIEL #

• • O Law. ALLOT SittL h2.0 = 4 lese Austemiff t STltLM -

o aner.r str:6

+ #888 STitL e 8

- . ASOS CL.SO.E f SilEL aus..ee 8 i.e - coa 8:istret -

5 O - I f I I I i i i i f f

-e.e ' o ei e.: o.s he o.e o , aus o.s 1.o

, . - i. .u rn.. . ... .. .rn.. 535 . .

9 4

Figure 1 Fatigue Threshold Data for Various Materials.

~.

3 C/2 'd - WdS1:20 46, di NN

3 O

. *S

('

m. l  ;

e. . .

DELTA K CfFA SQRCM)) e 7.

y E at E e2 - .

E i I I I I I I III I I I I I I i1L E e E -

2

- * - 384-1, R=e . l . 24 C C76 F) t l l

! w D - 304-3,R=8.6,177 C C3Ee F) 3 E-M ,

c - X - 304-4, R=e.7,177 C C368 F) _ 3

+ - 384-5, R=e.86,177 C C3Ee F) - .;; j

M - ~ '

.H E-SE g 4 - 304-C, R=0.85,284 C CEEe F) -

5 i a .= = i 0 -

- = E-87 D:

E A -

A n

n a p = =

0 p gl E N -

N y 3

= _

E-86-5

=

[

0 ,= - - ,

- ^ ~

I -

-= E-ee n "R N - X

• ;;! / .

- 5-

/

C p

C y e-e7 = ,+; - y:

L 2- 3 + X -

'C: '

a C - / y, X

P, h  : g E-te E d _ O - l

'd 0 C i E-M e 5

2 X .

m = E-te  !

g

- E- I

.h

, .h

" I I I I I I III I I I il i I IF i

i

!c} m E-Se g E se E al E S2 i Pn DELTA K (KSI SORCIN))

p 1e t

-n 4, s--

O e

O O

O t-s 1

6 1-ATTACHMENT 2

ttN t@ C1:3.uti GC tixtta DERGY P.2/2 e

GE Nuclear Energy o.r.,4.nw cm.v GENE.D13018691215 f r5 Ca* AWu. se.kuc CA nt/s ll/J4f)1 Oscar Lirapias Peach Bottom Atomic Power Stadon FAX (717) 456-4845 Deas Oscar.

This letter documcots the results of the FCCS/LOCA analysis pctformed in support of the Peach Dottom 3 Cracked Jct Pump Riser ($ sue.

The overall methodology and assumptions of the analysis were consistent with the generic analyses performed in suppoil of BWRVIP 2R. Many potential LOCA secnarios were evaluated and the limidng case described below was selected for further anal) sis.1his case features:

. Design Basis Accident Suction break with a single falhare . battery failure whidi disables I of 2 core spray loops and I of 4 LPCI pumps

. I c' ors spray and 3 LPCI available

. One risce auumed to crack to separation, resulting in the disauembly of 2 jet pump at the diffbser slipjoint (just below core support plate elevation)

= Conscreatively assumes riser separation is in loop with suction break (maximizes break flow out of vessel)

. Conscrvatively assumes riser separatJon is in loop with 2 LPCI (about 60% of LPCI flow lost through separated riser)

= Mnte than 24,000 gpm olECCS flow injected inside shroud (out of 36,000 rpm normally avrulable)

The SAFER analysis for this case was performed with the following auumpdons:

. Full sirr4 double-ended guillotine break of the recirculation line

. Nominal input and model assumptions used in rcrate analysis

. Current Technical Specification flow rates for ECCS The results of the SAFF1 ;atuation demonstrate that:

• Core will reflood for short term

. Peak Clad Temperature < 1200 'F e Long term level approximately 2/3 core hetght (top of intactjet pumps)

. About 14,500 gpm of ECCS flow lost through disassembled jet pump diDbsets (more tium 24,000 gpm of ECCS flow available inside shtoud)

. 2/3 core height maintained assuming all available ECCS flow continues long term

. Long term temperature and oddation results consistent with licensing basis These results are consistent with the previous evaluation for cracked jet pump risers performed in suppon of BWRVIP-28. A detailed report doctunenting this analysis will follow under separate cover, If you lute funher questions, please do not besitate to contact me or Tom Cains.

Smccrely,

_ / /

Kelly R. letcher, Manager Nuclear & Safety Analysts (403) 923-6535 11/17/97 MON 15:31 (TI/HX NO 5020] @ 002

/