ML20045C488

Responds to NRC 930517 Request for Addl Info Re Util Request for Rev to NUREG-0619, Routine Insp Criteria for Feedwater & Control Rod Drive Return Line Nozzles

ML20045C488
Person / Time
Site:	Oyster Creek
Issue date:	06/16/1993
From:	Keaten R GENERAL PUBLIC UTILITIES CORP.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
RTR-NUREG-0619, RTR-NUREG-619 5000-93-0046, 5000-93-46, C321-93-2176, TAC-M85751, NUDOCS 9306230182
Download: ML20045C488 (12)

Text

_

e o

GPU Nuclear Corporation V U Nuclear

= = n h s.

201-316-7000 TELEX 13&482 Writers Direct Dial Numbec June 16, 1993 C321-93-2176 5000-93-0046 U. S. Nuclear Regulatory Commission Att:

Document Control Desk Washington, DC 20555 Gentlemen:

Subject:

Gyster Creek Nuclear Generating Station (0CNGS)

Docket No. 50-219 Request for Additional Information - Oyster Creek Revision to NUREG-0619 Routine Inspection Criteria for Feedwater and Control Rod Drive Return Line Nozzles (TAC No. M85751)

?

By letter dated May 17, 1993 the Nuclear Regulatory Commission Staff indicated that they had performed the preliminary review of GPU Nuclear's request to revise the inspection intervals required by NUREG 0619 as they apply to Oyster Creek. To complete the review the Staff requested additional information in three (3) areas.

Those requests and the associated responses follow:

NRC REQUEST FOR ADDITIONAL INFORMATION REGARDING REVISION TO NUREG-0619 INSPECTION CRITERIA OUESTION 1:

Provide the definition of the "GE Generic Duty Cycle" labeled in Figure 1 of the submittal dated April 18, 1992. How does it

[]

relate to the thermal cycles defined in GE Report NEDE-21821-A, in terms of the number of the startup/ shutdown cycles, the o,

number of. scrams to low pressure hot standby, and the number of i

.fg scrams to high pressure hot standby? Provide bases showing

- o n.

that the thermal fatigue analysis using "GE Generic Duty Cycle" 88 bounds the analysis pertinent to Oyster Creek.

So ANSWER 1:

The."GE Generic Duty Cycle" shown in Figure 1 of the GPUN E6 submittal - (April 18,1992) is the same as that developed and g@

used by the General Electric Co. in the topical report that n<

addressed feedwater nozzle cracking (Ref. 1). This TR also formed the basis for the NRC staff safety evaluation (Ref. 2).

@@h The GE generic thermal duty cycle included both start-up/ shut-

)-S l

o a.

down cycles and scrams. On-Off operation of feedwater flow during low flow conditions creates nozzle region temperature-l, GPU Nucicar Corporation is a subsidiary of General Public Utilities Corporation 7-/0/25

C321-93-2176 Page 2 step changes.

For high pressure and hot standby conditions, there are 41/2 hours of cyclic temperatures (6 cycles per hour) between 430"F and 100*F.

For low pressure, hot standby conditions during shutdown (Figure 1), there are 6 cycles of lower temperature cycling for a 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> duration.

Each scram to low pressure is associated with a 1200 psig pressure spike and is divided into a high pressure and low pressure recovery phase.

Feedwater nozzle region temperature cycling is assumed to occur at 6 cycles per hour for 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> between temperature extremes of 430*F and 100 F at high pressure, hot standby and between 350 F and 100"F at 6 cycles per hour for 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> at low pressure (Figure 2). A scram to high pressure, hot standby and return to full power was also considered by GE (Figure 3).

This transient included only the thermal cycling conditions for high pressure, hot standby as described above.

The service conditions for the entire plant lifetime were represented by 130 start-up/ shut-down cycles followed by 3 scram cycles to low i

pressure, as described above, for each start-up/ shut-down cycle.

The scram to high pressure, hot standby and return to full power was deliberately not included in the structural analysis. The estimated number of these scrams was small in comparison to the number of scrams to low pressure.

Further, the number of scrams to low pressure was also knowingly increased, over the estimated number, in order to compensate for the omission. The "GE Generic Duty Cycle," used by GPUN, was in all respects the same as the GE generic thermal duty cycle as described in the foregoing.

The "GE Generic Duty Cycle" provides an upper bound limit for

+

the transient thermal stress impact of the feedwater nozzles.

While actual thermal cycle duty is not known conclusively, available information strongly suggests that the actual situation is less severe.

A review of the control room logs and control room instrument I

strip charts conducted when Figure 1 in the submittal was

prepared, i.e., for the period 1978-1983, indicated that:

i 1.

Nine feedwater cycles could occur during the duration of the plant start-up.

2.

There are no indications that Oyster Creek normally i

proceeds or remains in a low pressure hot standby condition i

for any extended period of time.

It appears that the i

reactor is subcritical if the correction of the scram requires a long period cf time.

3.

The presence of a 1200 psig pressure spike during each and every scram is not justified for Oyster Creek.

I The approximate actual Oyster Creek thermal duty cycle is shown in Figures 4 and 5.

I' s

C321-93-2176 Page 3 More recently, the low flow regulating valves were replaced and upgraded during the 12R refueling outage (1988) and the-feedwater block valves were replaced and upgraded during the 13R. refueling outage (1991).

These plant modifications enable the operators to maintain level during low feedwater flow conditions without on/off flow cycling.

OVESTION 2:

Provide the crack growth rate equation used in the calculation.

It is not clear when the "best fit" or the upper bound one was used in the analysis.

ANSWER 2:

Crack growth rate is determined using the "best fit" equations shown in ASME, BPVC,Section XI, 1983 Edition, Non-mandatory Appendix "A", Figure A-4300-1, entitled " Reference Fatigue Crack Growth Curves for Carbon and Low Alloy Ferritic Steels,"

(please see Figure 6).

The curve is in bi-linear form.

It can be seen that for R=0 and:

AK < 19KSI{iii, then, d#

- - = (1. 0 2X10 -12 ) A K SS, where S

bK is applied stress intensity zange, and d

- -- is crack growth thrcugh the pressure boundary in lin/ cycle]

and for AK > 19KSI{in, then,

= (1. 01X10 -7 ) A K

'S 1

OVESTION 3:

Provide the value of the heat transfer coefficient U, used in the analysis. Also provide a brief discussion about the applicability of using this value to Oyster Creek's single piston ring type feedwater nozzle with an attached baffle.

Provide, if there is any, experimental data on U for Oyster Creek's feedwater nozzle.

Notice that the U value of 100 z

BTU /hr-ft

• F was approved by the staff for nozzles with triple sleeves or a single sleeve with zero leakage only.

ANSWER 3:

The value of the heat transfer coefficient for the annulus region between the thermal sleeve and the feedwater nozzle inside surface that was used in this analysis was 150 BTU hr ft'"F

?

4 I

C321-93-2176 Page 4 Laminar flow conditions exist in the annulus up to about 10 gpm leakage.

Leakage should be less than 1.5 gpm at full power with operational piston rings. Considering forced convection in annulus flow, as modelled in Attachment 1 (Ref. 3), and assuming that there is no heat flux between the thermal sleeve 1

and the annulus flow, being equal in temperature, it is possible to show that for 5 gpm leakage, the Nusselt number, Nu equals 5.356. Therefore, the heat tran3.fer coefficient, U,gg,can be computed:

N,**K U=

D n

where K - thermal conductivity of water 9100 F - 0.3641 BTU hr ft*F D - Hydraulic diameter - 0.392" ' 'th clading and 1.328 unclad n

Substituting, U - 60 BTU hr ft' F i

for the unclad portion near the nozzle blend radius, and U - 17 BTU hr ft' F for the clad portion toward the nozzle.

A bore rrt on heat transfer coefficient equal to 50 BTU i

hr. f t.'*F consistent.i1 e the higher value of the two above was chosen for analysis purposes and added to the heat transfer coefficient for zero leakage, i.e., conduction and natural convection only, equal 100 BTU i

hr. ft.'*F j

so that an actual value, U = 150 BTU was used.

hr. f t.'

F

l C321-93-2176 Page 5

REFERENCES:

1. Boiling Water Reactor Feedwater Nozzle /Sparger Final Report (Supplement 2), General Electric Company Report NEDE-21821-02, August 1979, Class III (Company Proprietary).

2. BWR Feedwater Nozzle and Control Rod Drive Return Line Nozzle Cracking, NUREG-0619, Nov. 1980, Appendix C.

3. Incropera and DeWitt, fundamentals of Heat Transfer,1st Edition, 1981.

GPU Nuclear believes that the information provided in this letter along with previous submittals provides the technical basis to justify extending the Feedwater (FW) and Control Rod Drive Return Line (CRDRL) NUREG-0619 inspection intervals.

Specifically, GPU Nuclear plans to revise the inspection intervals for these nozzles as follows:

1)

Perform UT inspections of the FW and CRDRL once each service inspection interval in accordance with ASME Boiler ad Pressure Vessel Code Section XI.

2)

Eliminate future NUREG-0619 routine PT examinations.

Internal PT examinations would only be performed if flaws, which would compromise nozzle integrity, are detected.

If you have any qttestions please contact Mr. Michael W. Laggart, Corporate Nuclear Licensing Manager, at (201) 316-7968.

Sincerely, V

~

R. W. Keaten Vice President and Director l

Technical Functions RWK/SDL/plp cc:

Administrator, Region I Oyster Creek NRC Project Manager Senior Resident Inspector, OC

4 i

i a

600 FULL POWE R y-m OPE R ATION

~g 500 485 430 425 f s a x

(

E 9 4o0 4e 300

= 0 2o0 oO ONtOF F FW FLOW 100 2

3 100 too CYCLING AT Six jh I cycies PER hout 0

TIME thr) 2112 hr 2hrHP HP HOT HOT STANDgY STAND e

W h

2 hr LP MOT ST ANDBY

+

4--

1200 -

1045 j p 3

1D45 o

3 som sao q

w

@ soo ay soo a

5 400 E

2m c

TIME trwl Startup/ Shutdown Cycles (130 Per 40 Year Plant Life)

FIG.1

heLh PJA!H e

OPERA FULL POWE a T i O r.

00g n A T 10*.

w_

ggg OkiOF Fs h 5 L OW F

C vCLINO AT Si x r

S bOU

~

C ytics PE ll Neuf 4

1

=

w

'l 400 3 _-

,.. M s e.

h 2h 30v f

l I

. % ; *r: ^< y

< v

'Qwn %

W~

?

'I N

,M ggg Ou.

i tM E teit) to er LP H OT S T A*,O M e

2 pt HP MOTSTANDdV

'#UU 1200

• 5 2045 ica 93u /

_.r a

g Boo 30 sw E

5 400 V

t zw

,2o o

TtME thd j

Scram to Low Pressure Hot Standby and Return to Full Power (349 Per 40 Year Plant Life)

FIL 2 L

I I

d

.7 FULL 1

POWE rt OPE D A-r ggg pgwt gg TtON OPE R AllON 600 -4 i

(

- 500

[c[

425 430 4 2 ',,

g 3 o 400 l

<t o 5E 33g d,

\\

ON/DF F FW F t.OW l

\\-

CYCLING Al UlX j

rycles PE n nom i

l gy o o 200 I

1 l 52 L

i aR 100 l

100 0

llMI thd i

k 2b'HP m

HOT STANOBY i

l i

1200 1200 I

_1045 104$

l

- 1003 9

930 s

y 800 a

g 600 E

co 400 u(

E 200 0

TIME thri

)

Scram to'H1Eh Pressure llot Standby and iteturn to fu)) Power (62 Per 40 Year Plant Life)

F/(

3

h.

• A 4

e.

J h

O a

g I N

g OYSTER CREEK STARTUP/ SHUTDOWN CYCLES-'

700 S.

600 EC wL gg 485 e cyctEs e cyctEs 300

/

430 [

gj 430 i

42s Sw 400 N

w Ez 300 33:

w.

00 100 100 TIME (HR) p 1400

{ 1200 C

1045 1045 ff

$1000

[

'g 930 930 13 E

800 a

p 600 t

400 200 TIME (HR) 4

. FIG. 4

OYSTER CREEK PLANT SPECIFIC SCRAM CYCLES 800 700 zC Ct-600 55 12 CYCLES b 500 4

430

,f

[C 425 425 b b 400

~

"O 92

(( 300 200 100 100 TIME (HR) 1400 1200 1050

-[

f1045 1045 g 1000

\\

930

/

E IE 800 8g 600 5"

400 200 -

TIME (HR)

FIG. 5

1 iI i983 Edition APPENDIX A - NONMANDATORY Fig. A-43001

.)

1CW f

~

• Linear 'nterpolation es recom -

~

menced to secount f or rateo

[

oependence of water environment f

f

~

g cueves, for 0.25 <R <. 0 65 f o<

O' g shattoa sioor

o. W

[

i s

f 4

19S

[p

/

~

. (101 x 10 10 4#

2 07 = 3 75 4 + 0 06

., ~ p'

^[-

~

[j

[

8*K IK en,n mas 200 x

/

I g/

)

S.,bswiface Os ws

}

[

g too las environments

~

80 0267 x 103:; K 3 726 j

i e'

i 70 g

j

,o

/

3 Determ.ne tne ar at which vne

/

g,o3 3

M

~

iam changes by calculation of

_O y

sne entersection of the two g

I curves.

T s

~

$Wrface flaws f

4 (wafe' reactor 20 environment) 2

- sooncacie for j

t R E 0 25

\\

]

• O 25 < R < 0 65

/

\\

R 3 0 65 i

~

8*Kmin,' Ama s.

• I

/

s y

7 t.

\\

f 2

• L,near interootation is recommenoco

[

to account f or R rai,o oepenoer.ce o

-3 of water environment cuews. f or f

(

0.25 R - 0 65 for s'eco siope q

i e

f 5 95 O

88

• t t 02 x 104: 03 ;K ON Si Di* 26 94 5 725

/

A ' K i n 'K rr man I

/

I 1

I i i i ! ri i

1iIIi 1

2 5

7 10 20 50 70 100 i

s. @ p Stress intensity S ector Range i;r k

F]G. A-43001 REFERENCE FATIGUE CRACK GROWTH CURVES FOR CARBON AND LOW ALLOY FERRITIC STEELS t

fl6. 6 i

i l

l A TTACHWEH7.t i

r DETER.M IN ATICN or NvS$ELT NUMSETi' foft AwJucus FLov4 r., s -

,/

• ~ j, l

,/

A r-

• L-7, e Moure 8.8 The concentree tube annuius If umform heat flun conditions exist at both surfaces, the Nusselt numbers may be computed from empressions of the form Nu A,u =

1 - (q,",, grl (8.68)

N u,,

^."

" l ~ (y,"!q,"YI,*

I8 69I T he influence coefficients (Nu,,, Nu,,, #l and (/l) appearmg in these equantons may be obtained from Table 8.3. Note that y" and q," may be positive or negative, dependarig on whether heat transfer is to or from the fluid. respectively.

Table 8.3 inhvente coethcients for fuhy coveloped lammar flow m a circular supe annulus wem umform hest tips mamtamed al born surfaces D Q ku,.

kg 17 n;

O 4 364 5

0 0 05 17 81 4 792 2 18 0 0294 0 10 11 91 4 834 1 383 0 0562 0 20 8 499 4 833 0 905 0 1041 0 40 6 583 4 979 0603 01823 0 60 5 911 5 099 0 473 02455 0 80 5 58 5 24 0401 0 299 1 00 5 385 5 385 0 346 0 346 used eita permas.pn kom W M man one H c Poemens Henanoon er near Fransfer f

Chapter 7. W M MO tSenOe SAql J P 94SMnell 106 MC(af d e-Mdt New Veet 1977 l

l