ML17276A180

Presentation on Number of OBE Fatigue Cycles for BWR NSSS Design (Except Piping).

ML17276A180
Person / Time
Site:	Columbia
Issue date:	09/15/1981
From:	Henrie D GENERAL ELECTRIC CO.
To:
Shared Package
ML17276A177	List: ML17276A176 ML17276A179 ML17276A180 ML17276A181
References
NUDOCS 8111090523
Download: ML17276A180 (110)

Text

PRESENTATION ON THE NUMBER OF OBE FATIGUE CYCLES FOR BHR NSSS DESIGN (EXCEPT PIPING)

SEPTEMBER 15, 1981 D.K. HENb,lE

,

! SEISMIC 8, DYNAMIC ANAl YSIS GENERAL ELECTRIC DVH-1 Bfii090523 Sii102 PDR ADOCK 05000397 A PDR

NUMBER OF OBE FATIGUE I YCLES NSSS EQUIPMENT SRP RECCflMENDATION- O'OBE WITH 10 CYCLES GE RECOMYiENDATION- 10 PEAK OBE CYCLES GENERICALLY.

GE STUDY SHOWS 10 PEAK OBE CYCL'ES OYER PLANT LIFE CONSCRVATIVE

a 'RC NUREG.'CR-1151 RECOYf',ENDED REVIS IOf'!S TO Nt C SEISNIC DESIGN CRITERIA o THE NUREG STATES THAT NPC

.REQUIREYi~NT "OF FIVE QdE CYCLES IS EXCESS IVELY CONSEF,'VATIVE" o ALSO INDICATES THAT ON THE AVERAGE, THE OBE DESIGN ACCELERATION HAS A NEP OF 90/

IN A 50 YEAR LIF.E o MASH-1000, OCTOBER. 1975 - PROBABILITY OF OBE -IS ONE IN 100.TO 125 YEARS AND ltlJT FIVE IN 40 YFARS.

CONC!US ON .

- 5 OBE EXCESSIVELY CONSERVATIVE DKH-3 9/15/81

GE STUDY ')N PROP)ABILITY OF ORE (1973) o BASIS - A STUDY OF 26 PSAR AND FSAR PLANTS o 'OUR 90 YEAR PERIODS 1810 - 1849 1850 - 1889 1890 - 1929 1930 - 1969 o MAXINUN SITE I,!TENSITY EARTVQUAK; CHOSEll FROr~ EACH PERIOD FOR EACH SITE

'ATIO OF

',:

o f'tA It "iUt GROUND ACCELERA I ION TO SSE DES IGN HAS IS GROUND ACCELERATION CALCULATH3 FOP, EACH 40 YEAk PERIOI!

NXI%,l 0,16 NININUN = INS IGNIF I CAI'/T NEAe = 0.051 STANDARD DFVIATSQN= 0,039 DK'<- o 9/15/81

e TABL'L.

GENERIC SUHHARY'L CENTRO TAFT 'OLDEN GATE CLINTON HANFORD PERRY SUS UEHANNA DURATION . 29,4 30. 0 '3.2 10.0 16.0 10.0 15.0 (SEC.)

HAX. SIT~ ACCEL.(g) 0.33 0.18 0,13 .015 .015 .007 .007 (RECORDED/ESTIHATED)

SSE DESIGN BASIS .25 .25 .15, .10 HAX. ACCEL. (g)

<0.04 <0.10 0.07 <0.04 HAX. HORIZ. ACCEL (g}/

90K hr.P IN 50 YEARS

~ ~

9/15/81

N o NUMBER OI- FATIGUE CYCLES PER EARTHQUAKE o OBTAINED BY TIME HISTORY ANALYSIS' RANDOM VS, PERIODIC EXCITATION o TABLE 2 ENVELOPED AND AVERAGED CYCLES FRON THREE EARTHQUAKES AND SIX MAJOR NSSS COMPONENTS o TABLE 5 -  % OF CYCLES ~ 50% OF PEAK o, TABLE 0 -  % OF CYCLES ~ 25% OF PEAK o INDEPENDENT OF EARTHQUAKE OR COMPONENT Ff';EQUENCY 99,5% OF STRESS REVERSALS OCCUR BELON 75% OF MAXIMUM STRESS 95% BELOH 50%

85% BELOW 25%

'o TABLE 5

SUMMARY

OF EQUIVALENT STRESS CYCLES OF ALL MAGNITUDES CONCLUSION '0 PEAK OBE CYCLES ARE CONSERVATjVE

.DKH-6 9/15/81

>> TABL AYERAGE NUHBER OF STRESS CYCLES OF ALL HAGNITUDES NUHBER OF CYCLES NORt NL I ZED FRE UENCY BANDS LONG DURATION DURATION PEAK ACCEL. PEAK ACCEL.

EM THIMAK S C 0-10 Hz 10 - 20 Hz 20-50 Hz EL CENTROID 29.4 0.33 0.25 ~168 337 425 TAFT ~

30. 0 0.18 0.25 163 ~368 ~643 GOLDEN GA[Ei ) 13.2 0.13 0.25 94 171 316 NOTES: ~i) Hay 18, 1940, El Centro, N/S Component, 29.4 sec (2) July 21, 1952, Taft, S69 E Component, 30.C sec.

(3) Hatch 22; 1952, Golden Gate, SBO E Component, 13.2 sec.

DKN-7 9/15/81

PERCENTAGE gF STRESS CYCLES

~ MITH STRESS ANPLITUDES BELOH 50K OF TllE MAXIt1UN VALUE Frequency Range 0 -.10 HZ 10- 20Hz 20 -:50 HL El . Golden El . Golden El -

Golden Earthquake Centro Ta ft Gate Centra Taft Gate Centra laft Gate

. ~

Duration sec '9.4 30.0 l3e2 29.4 30.0 l3.2 29.4 30.0 l3.2 Component PERCENTAGES B

c (5TAGItl tert) tGTatp 7,Maa)

(m~agoe)"

99.2 b9.2 99.0 98.0 97.1 95.8 99.9 99.9

99. 9 94.7 97.9 95.4 95.6 97.9 96.4

~

90.8 96.8

'2.8 94.7 97 6 99.6 99.2.

99 99.8

'7.

96.8 99,3 7

0 (3}daub havoc) ,99.1 96..9 99.5 96.2 97.2 94,1 99.2 99.6 98. 9 E (ra.) 99.1 95.8 99.9 9>.3 96.8 91.3 95.7 99.8 '9.1 F (~p Neo~) 98.5'5.7 '; 99.4 96.e 96.5 94.4 95.b 98.6 97.9

'9.8 a

Average 99.0 96.6 96.0 96.7 93.4 97.1 "9.4 98.3 (Over .All Average 97.,4)

Time History Input Cycles Below 50K of Peak El Centro '3%

Taft '. . 90K Golden Gate 95X DKH-8

TABLE 4

.,PER(EHTAGE OF STRESS CYCLES WITH STRESS NlPLITUDES BELOW 25'X OF THE HAXIHUH YALUE Frequency Range 0-10 Hz l0- 20 llz 20 - 50 Hz El . Golden El Golden El Golden

~

Earthquake Centro Taft Gate Centro taft Gate Centro Taft Gate Duration, sec 29.4 30.0 13.2 , 29.4 30.0 13.2 29.4 30.0 13.2 Caoponent P E R C E tl T A G E-85.5 85.4 96.2 80.6 77.6 81.3 81.9 89.4 87.5 85.8 84.9 96;2 80.4 78.0 80.6 82.6 80.3 89.3

'

C 93. 1 8l.2 99.4 82.3 77.3 81.9 92.'1 92.8 94.0 86.0 81.0 98.0 84.3 19.6 83.4 90.0 93.0 ~1. l 89.4 79.4 97.6 . 82.0 79.1 80.6 80.9 9l.2 iu.4 86.6 79.5 92.2 84.4 78.9 83.3 79.5 89.0 89.0 Average 87.7 81.9 . 96.6 82.3 '8.4 81.9 84.5 89.3 90.3 (Over-all average 85;9)

Tir;. History Inpu~ Cycles Below 2~4 of Peak El Centro 78K Taft 70K Golden Gate 90K DKH-9 9/15/81

tiUMBER OF STRESS CYCLES OF ALL MAGNITUDES DURING A LONG DURATION EARTHQUAKE FREQUENCY BANDS

.(CORRESPOtlDS TO COMPONENT FUNDAMENTAL FREQUENCIES) 0-10Hz 10- 20 Hz 20 - 50 Hz TOTAL NUl'lBER OF 16& 359 643 STRESS CYCLES

. NUMBER OF CYCLES BETHEEk 1 (1) 2 (2) . (3) 75K. AND 100K OF PEP.'ALUE (0.5X OF TOTAL)

NUMBER OF CYCLES BETWEEN 8 (1) 16 (2) 29 (4).

50+ ANO 75K OF PEAK'VALUE g4.5X OF TOTAL) tIUMBER OF CYCLES BETWEEN 17 (1) 64 (1) 25i'nd 50Ã OF PEAK VALUE (10K OF TOTAL)

NUMBER OF CYCL":5 LESS THAN 143 (1) 305 (1) 547 (1)

OR EQUAL TO 25K OF PEAK VALUE

(&5X OF TOTAL)

TOTAL NUMBER OF EQUIVALENT ~

(4) (6)

PEAK STRESS CYCLES DKH-10 9/15/81

WP-2 OSER UESTION NO. 10

~ ~

No seismic transients are specified for the majority of the'omponents and the components for which they are specified require only one QBE cycle. Justification is'equired.

RESPONSE

See the text revision attached.

Summation - This item.is closed.

PCY: r f/45E4 8/18/81

0 RG? 2 Q(o 3.9 HECHANXCAL SYS~ AND COMPONENTS 3 9 1 SPECXAL TOPXCS FOR MECHAHXCAL COMPONENTS 1.1 Desi,gn Transients

'.9 This section shows the transients which are used in the de>>

sign of the ASME Code'Class 1, control, rod drive components, reactor assembly including core supports and reactor in-ternals, main steam and recirculation systems The number of cycles or events for each transient are included. The design transients shown in this. section're included in the design specifications for the components. Transients or .

combinations of transients ar'e classified with respect to the component operating condition categories, identified as "normal,," "upset," "emergency," "faulted," or "testing"'n the ASHE Boiler and Pressure Vessel Code if applicable. The cycles due to SSE and OBE used in the fatigue analysis are shown. in Table 3.7-4.

3.9.1.1.1 Control Rod Drive (CRD) Transients The normal and. test service load cycles used for design pur-poses for the 40 year life of the control rod drives are as follows: 1 Transient Cater Cvc:les

a. Reactor startup/ normal/upset 120 shutdown b Vessel pressure normal/upset 130 tests c Vessel overp assure normal/upset 10 Scram test plus normal/upset 300 startup scrams

e. Operational scrams n'oxmal/upset 300 f; Zog cycles normal/upset 30,000 ge Shim/dry.ve cycles normal/upset 1000

3. 9-1

V Zn addition to the above cycles, the following have been considered in the design of the CRD.

Transient Cate<~o ~ ~ches

'.

h.. Sc-am with inopera- normal/upset. 10 tive buffer Scram with stuck normal/upset contzol blade

.

faulted io l

All ASME ass Fcomponents of the ave been analyzed according to ASME Section XXZ Boiler and Pressure Vessel Code The capability of the CRD's to withstand emergency and faulted conditions is verified by test rather than analysis.

3.9.1.1.2 CRD Housing and Zncore Housing Transients The number of transients, their cyc3.es, and. c3.assification as considered in the 'design and fatigue analysis of the CRD housing and incore housing aie as follows:

Transient. Cater cvcles

a. Nozmal startup a. nozmal/upset 120 shutdown

b. Vessel pzessure- normal/upset 130 tests c., Vesse3. overpzessaxe normal/upset 10 tests.

d. Xnterzuption of nozmal/upset. SO feedwater flow e., Scram nozmal/upset 200 gou~~

U gg~

Pwr V

~y ~~ Z4.~ ~4

Pro Transient Catec~o ~Ccles OB~fk nomsal/upset @11 /g ge SSE ~

CRD Housin Onl h.

0 1e Stuck Rod Scram Scram no Buf'fer normal/upset nozmal/upset ~

X

/D

• • SSR is a 2a e6 conci'on; .howeve~in the

~ ~ ~

Ms-ess analy-sis r~ort

<limits X,

~

it s treate as emergen with lowe stress T. ecgxeucy of this%cycle wou indicate ergency cat os. wever, fo conservatis this OB ondition was an zed a upset bu without fat e consid ations.

3. 9-3

AMENDMENT NO. 9 April 1980 3.9.1.1.3 Hydraulic Control Unit Transients The normal and test service load cycles used for the design and fatigue analysis for the 40-year. life and the Hydrau3.ic Control Unit are as follows:

Transient CatecLoO ~Ccles

a. Normal startup & normal/upset 120 shutdown

b. Vessel pressure normal/upset 130

.tests

c. Vessel overpressure normal/upset 10 tests

d. Scram tests (cold) - normal/upset 300 e., Operational scrams normal/upset 300 (hot)

Jog cycles normal/upset 30,000

g. Drive cycles normal/upset 1g 000 1

h. Sc am with stuck normaL/upset scram discharge valve

i. OBE',

normal/upset SSF fau3.ted The requen of occurrence of this eve'nt would indi te category~~ However, fomconservatism, s event emergenc was a co yzed abnormal and dered u~

for fatigue evaluation.

conditio 10 c~s 3 9-4

3.9.1.1..4 Core Support and Reactor Internals Transients The normal and test service 3.oad cycles used for the design and fatigue analysis for th 40 year life of the CS and WMI .f-/ .

ransients Catecaorv ~cles Startup normal/upset. 120 b-. ower cycles normal/upset . 12,40

c. Los of feedwater normal/upset 80 heate

d. Scram normal/upset 198

e. Reduction to power, hot st dby 0% normal/ups t 111 shutdown, 5 ves 1 flooding

f. Unbolting n 3./upset 123 k

g. Sczam (Auto. blow- n a3./upset down & reactor ove<ressure) h Improper star of emergen cold recirc. oop Sudden s in old rt of 'mergency pump zecir . loop

j. Im oper startup 'mergency k.. ipe rupture 6 faulted blowdown 3.9 5

WNP 2 3.9.1.1.5 Main St~am System Transients The following transients are considered. in the stress analy-sis of the main steam piping:

Main Steam Transients

• Transient Cate~os Cycles

a. Startup normal '121

b. Loss of P;W. pumps upset 10 isolation valves closed

c. Scram upset 180

d. Shutdown normal

e. Reactor overpressure emergency delayed scram Single S/RV blow- upset 8 down

g. Automatic blowdown emergency

h. Hydrotes< test, 130 OHE so 3.9.1.1.6 Recwculation System Txmnsxents The following transients are considered in the st ess analy-sis of the recirculation piping:

Recirculation T ansients Transient Cate~ ~Celes 4,

a. Startup normal 121

b. Tuxbine roll and norma1 120 inc ease to power

3. 9-6

/

0

Transient Catec~o ~Cales

c. Loss of feedwate upset 10 heater

d. Partial feedwater upset 70 heater bypass

e. Scrams. upset 180 Shutdown normal 111

g. Loss of F.H. pumps upset 10 isolation valves.

closed

~ h. Reactor overpressure emergency with delayed scram

i. Single S/RV blow-down upset

j. Automatic blowdown emergency

k. Hydrotest. test 130 Reactor Assembly Transxpnts 05K'.9.1.1.7 The reactor assembly includes the reactor pressure 'vessel, support skirt, shroud suppor , and shroud plate.'he cycles listed in Table 3.9-1 were specified in the eactor assembly design and fatigue analysis.,

3.9.1.1.& Hain Steam Xsolation Valve Transients The main steam isolation valves are designed for the follow-ing service conditions and thermal cycles:

Transient ~cats as ~Calas

a. Pre~p 9100 F/hx normal/upset 150

b. Startup (heating normal/upset 120.

100oF/hr)

3. 9-7'

NNP 2 AMENDNENT NO. 16 June 1981 li TABL E 3.9-1 PENT EVENTS No. of

~Ccles Normal, U set, and Testincr Conditions

a. Bolt Up~/Unbolt 123

b. Design Hydrostatic Test 130

c. Startup (100 F/hr Heatup Rate)** 120

d. Daily Reduction to 75% Power* 10i000

e. Meekly Reduction to 50% Power* 2i000

f. Control Rod Pattern Change+ 400

g. Loss of Feedwater Heaters (80 Cycles Total): 80

h. Operating Base Earthquake Event at Rated (OSV Operating Conditions

i. Scram:

.1) Turbine Generator Trip, Feedwater on, Lsolation Valves Stay, Open 40

2) Other Scrams 140

3) Loss of Feedwater Pumps, Isolation Valves Closed 10

4) Single Safety or Relief VaLve Blowdown 8 Reduction to 0% Power,'ot Standby, Shutdown (100 F/hr Cooldown Rate) <<+

k. HPCS Operation (10), SLC Operation (10) 20 3-9-92

AMENDMENT NO. 16 June 1981 Page 2 of 2 TABLE 3.9-1 (Continued)

No. of

~Cele s Emeraenc Conditions

1. Scram:

1) Reactor Overpressure with Delayed Scram, Feedwater Stays on, Isolation Valves Stay Open 1*<<w

2) Automatic Blowdown 1 **<<

m Improper St'art of Cold Recirculation Loop 1 ***

n. Sudden Start of Pump in Cold Recirculation Loop o.. Improper Startup with Reactor Drain Shut Off Followed by Turbine Roll,. and Increase to Rated Power n Faulted Condition

p. 'ipe Rupture and Blowdown 1 <<*<<

Safe Shutdown Earthquake at Rated Operating Conditions ASME Hydrostatic Test t.25 x Design Pressure Hydrostatic Test 10 (per. NB 6222 and NB 3114)

<<Applies to reactor pressure vess'el only.

<<*Bu3.)t: average vessel coolant temperature change in any 1-hour period.

• <<*The'annual encounter probability of the one cycle events is <10 2 for emergency and <10 + for faulted events. =-

<<**<<Includes 10 maximum load cycles per event

~y '-9' ~ i)sos Q ~~~ g+j~ io o6~

xQ, Q 9~93 b~

rt i

WP"2 DSER UESTION NO. 11

,9 I Paragraph 3.9.1.1, Design Transients, referring to Table 3.7"4, "Reactor

. Suilding"Seismic Analysis Natural Frequency and Natural Period," appears to be in error.. Clarification is required.

RESPOHSE The statement is deleted. See'the text revision attached.

.. Summati;on This item is cIosed.

PCY: rf/45E5 8/18/81

MNP-2

"

3 "9 MECHANICAL SYSTEMS AND COK?ONENTS 3 9.1 SPECIAL TOPICS POR MECHANICAL'OMPONENTS 3.9.1;3 Design Transients J This section shows the transients which are used in the de-sign of the ASMZ Code 'Class 1, contxol rod drive components, reactor assembly including, core suppoxts and reactox in-ternals, main steam and recirculation systems. 'The numbe of cycles or events for each transient are included. The design transients shown in this section axe included in the

~

design specifications for the components.. Transients or

.. combinations of transients are classified with respect to the component operating condition categories identified as "normal," "upset," "emergency," "faulted," ox "testing" i the ASME Boiler and Rressure Vessel Code

'cycles due to SSE and.OBE used in the--Eati e"anal isa alicable ~p eboen in- zanle 3 7-'4 .

3.9.1.1.1 Control Rod Drive (CRD) Transients

.The normal'. and test service load cycles used'or design pur-poses'fox'he 40 'year life of the. control rod drives are as follows:

~

. Transient C~ata<ra Cvelee II

a. Reactor staztup/ normal/upset 120 shutdown

b. Vessel pressure noxmal/upset 130 tests

c. 'essel overpressure normal/upset 10

d. Scram test. plus noxmal/upset 300 startup scxams

~ e. Operational. scrams n'oxmal/upset 300 Jog cycles noxmal/upset '0,000 g." Shim/drive cycles normal/upset 1000

'I 3.9-1

e WNP-2 DSER QUESTION NO. 12 (3.9.1)

Table 3; 9-15, Applicable Seismic Cycle Loading, is indicated as "Later". Provide a schedule for its inclusion in the FSAR.

RESPONSE

Table 3.9-15 is deleted.

The statement in Section 3. 9. l. 1. 13 that references Table 3. 9-15 has also been deleted.

Summation This item is closed.

I g.

h.

Pressure 110'4 Transient design pressure at. 575 1300 psi at 100 F installed hydrostatic test F

Circlet 130

i. 1670 psi at 100 F installed hydrostatic test 3.9.l.l.l3 Balance of Plant Transients The transients used in design and fatigue analysis of thy' ~j naianc ot p a t gpmponenta are lipted in T hie 3.9-1p gP'PP'~

r ~

3.9.1.2 Computer Programs Used in Analysis The following sections list the computer programs used in the analysis of specific components.. These programs are de-scribed in 3.12.

3.9.1.2.1

~ ~ ~ ~

"

Reactor Vessel The following programs are used in the analysis of the Reactor Vessel:

a. CB&I Program 711 "GENOZK"

b. CB&I Program 948 "NAPALM"

c. CB&I Program 1027

d. CB&I Program 846

e. CB&I Program 781 "KALNINS"

f. CB&I Program 979 "ASFAST"

g. CB&I Program 766 "TEMAPR"

h. CB&I Program 767 "PRINCESS" 3.9-14

WNP'-2.

TABLE 3.9-15 APPLICABLE SEISMIC CYCLIC LOADING 3.9-207

4l MNP-2 DSER OUESTION NO. U

~ ~

Methods of verification are required for all NSSS computer codes used in the analysis.'ESPONSE The NSSS, programs can be divided into two categories.

S~E The verification of the following GE programs has been performed in accordance with the requirements of 10CFR50, Appendix B. Evidence of the verification of input, output, and methodology is documented in GE Design Record Files.

(a) MASS (i) FAP-71' (b') SNAP (MULTISHELL) (j) CREEP-PLAST (c) GASP (k) PISYS (d) NOHEAT, (1) ANSI7 (e) FjNITE (m) SAP4G (f) DYSEA (n) FTFLQ01 (g) SHELLS (o) ANSYS (h) HEATER (p) BSTIF01 Vendor Pro rams The. verification. of the following two groups of vendor programs is assured by contractual requirements between GE and the vendors. Per the requirements, the quality assurance procedure of these proprietary programs used in the design of N-s~ed equipment is in full compliance with 10CFR50, Appendix B.

B ron Jackson Program RTRMEC CBM Pro rams (a) 711 GENOZZ (i) 928 TGRV (b) 948 NAPALM (j) 962 E0962A (c) 1027 (k)'84 (d) 846 (1) 992 GASP (e) 781 KALNINS (m) 1037 DUNHAM'S (f) 979 ASFAST (n) 1335 (g) 766 TEMAPR (o) 1606 8c 1657 HAP

'(h) 767 PRINCESS (p) 1635 (q) 953 Accordingly, the FSAR text is revised as attached.

Summation This item is closed pending NRC audit..

• To be audited by NRC.

PCY: ggt: rf/45L2 9/23/81

i Pressure Transient CvclM

g. Ll0% design pressure at 575 P 1300 psi at 100 P installed 130 hydrostatic test 1670 psi at 100 P installed hydrostatic test )

3. 9.1.1.13 Balance of Plant Transients The transients used in design and fatigue analysis of the balance of plant components are listed in Table 3.9-1.

A .complete list of applicable seismic cyclic loading for operating basis earthquake is. shown in Table 3,9-15.

9.3..2 Computer. P ograms Used in Analysis The 11owing sections list the sis of specific components.

computer programs u in the ana These progr are de-scribed x, 3.12.

3.9.1.2.1 Re or Vessel.

The following prog s are used in analysis of the Reac"or Vessel:

a. CS&Z Program 1 GWQZZ"

b. CS&I P ogr 48 . ~ALE"

c. CS&I'.Pr ram 102T

d. CS Program 846

e. CS&I P ogram 781 "KALNXNS"'79

f. CS&I Program "ASFAST"

g. CS&I Prog am 766 "TEHAPR"

h. CS&I Prog am 76T "PRINCESS" 3 9-14

WNP-2 DSER 3.9.1.2 Computer Programs Used in Analysis The. following sections discuss computer programs used in the analysis of the major safety-related components. (Computer programs were not used in all components, hence not 'all components are listed. ) The NSSS programs can be divided into two categories.

GE Proarams The verification of the following GE programs has been performed in accordance with the requirements of 10CFR50, Appendix B> Evidence of the verification of input, output, and methodology is documented in GE Design Record Files.

(a) MASS (i) FAP-7I (b) SNAP (MULTISHEL'L), (j) CREEP-PLAST (c) GASP (k) PISYS (d) 'NOHEAT (1) ANSI7 (e) FINITE (m) SAP4G (f) DYSEA (n) FTFLG01 (g) SHELL5 (0) ANSYS (h) HEATER (p) BSTIFOl Vendor Pro rams The veri&'cation of the following two groups of vendor programs is assured by contractual requirements between GE and the vendors. Per the requirements, the quality assurance procedure of these proprietary programs used in the design of N-stamped eqaipment is in full compliance with 10CFR50, Appendix B.

B ron Jackson Program RTRMEC CBEI Proarams (a) 711 GENOZZ (i) 928 TGRV (b) 948 NAPALM (J) 962 E0962A (c) 1027 (k) 984 (d) 846 (1) 992 GASP (e) 781 KALNINS (m) 1037 DUNHAM'S (f) 979 ASFAST'g) (n) 1335 766 TEMAPR (o) 1606 Ec 1657 HAP (h) 767 PRINCESS (p) 1635 (q) 953 PCY: ggt: r f/45L3 9/23781

MNP"2 DSER 3.9;1.2 1 Reactor Vessel and Internals 3.9.1.2.1.1 Reactor Vessel CBM Programs (a) through (q) listed above are used to analyze the reactor pressure vessel. Detailed descriptions are provided in Sec-tion 3.12.

-

3.9.1.2.1.2 Reactor- Internals

'\

The following'computer programs are used in the analysis of the core support structures and other safety-related reactor internals: MASS, SNAP (MULTISHELL) GASP, NOHEAT, FINITE, DYSEA, SHELL5, HEATER, FAP-71, and CREEP-PLAST. Detailed descriptions of these programs are provided in Section 4.1. I:

J PCY: ggt: rf/45L4 9/23/81.

n

i. CB&X Program 92S TGRV"

j. CB&I Program 962 "E0962A" CB&X Program 984 1 CB& rogram 992 "

m. CB&Z Progr 7 DUNHAM'"

n. CB&I Pra am 133 oo Programs 1606 and 1657 CB&X Program 1635

q. CB&Z Program 953

~ ~ o .o Piping The allowing programs are used in. the analysis of pip'.

ADLPIPE

b. DYNAMIC ANALYSIS'P PIPING'YS c . P PANELS SPACE STRU ANALYSZS (MASS)

SHELL ALYSXS'ROGRAM TISHELL)

e. TIME DEP PM~'RCE PIPE DYNAMIC YSXS PROGHAM (PDA)

3. 9. 1.2. 3 Recirculati Pump No computer program were used in the sign of the reci cu-latian pumps.

3.9.1.2..4 E S Pumps and Motors An ecpxiv ent static camputer analysis was performed the, ECCS motar rotor shafts. The model consisted of 1 ed, mas s simulating the distribution of mass in. the system, 3.9-15

MNP"2 0SER 3.9.1.2.2. Piping 3.9.1.2.2.1 Piping Analysis Program/PISYS PISYS is a computer code specialized for piping load 'calculations. It utilizes selected stiffness matrices representing standard piping com-ponents, which are assembled to form a finite element model of a piping system. The technique relies on dividing the pipe model into several discrete substructures, called pipe elements, which are connected to each other via nodes called pipe joints. It is through these joints

- that the model interacts with the environment, and loading of the structure becomes possible. PISYS is based on the linear classical elasticity in which the resultant deformation and stresses are pro-portional to the loadihg, and the superposition of loading is valid.

PISYS has a full range of static and. dynamic analysis options which include distributed weight, thermal expansion, differential support motion modal extraction, response spectra, and time history analysis by modal or direct integration. The PISYS program has been benchmarked against five Nuclear Regulatory Commission piping models for the option-of-response-spectrum analysis and the results are documented in a report to the Commission, "PISYS Analysis of NRC Problem," NEDQ-24210, August, 1979.

3.9.1.2.2.2 Component Analysis/ANSI7 The ANSI 7 computer program determines stress and accumulative usage factors in accordance with HB-3600 of the ASME Code, Section III. The program was written to perform stress analysis in accordance with the ASME Code sample problem, and has been verified by reproducing the results of the sample problem analysis.

3. 9. 1.2. 3 ECCS, Pumps and Motors 3.9.1.2.3.1 Rotor Assembly Analysis Program/RTRHEC RTRMEC is a computer program which calculates and displays results analysis of motor rotor assembly when acted upon by external of'echanical forces at any point along shaft (rotating parts only). The shaft deflection educ to magnetic and centrifugal forces was analyzed. The calculation for the seismic condition assumes that the motor is operating and that the seismic, magnetic, and centrifugal forces all act simultaneously and in phase on the rotor-shaft, assembly. Note that the distributed rotor assembly weight is lumped at the various stations, with the shaft weight at a station being the sum of one-half the weight of the incremental shaft length just before the station, plus one"half the weight of the adjacent incremental shaft length just after the station. Bending and shear effects are accounted for in the calculations.

PCY: ggt: rf/45LS 9/23/81

WNP-2 OSER 3.9.1.2.3.2; Structural Analysis Program/SAP4G SAP4G is used to analyze the structural and functional integrity of the ECCS pump/motor systems. This is a general structural analysis program for static and dynamic analysis of linear elastic complex structures.

The finite element displacement method is used to solve the displace-ments and stresses of each element of the structure. .The structure can be composed of unlimited number of three-dimensional truss, beam, plate, shell, solid,. plate strain-plane stress and spring elements that are axisymmetric.'he program can treat thermal and various forms of mech-anical loading. The dynamic analysis includes mode superposition, time history, and response spectrum analysis. Seismic loading and time-dependent pressure can be treated. The program is versatile and efficient in analyzing large and complex structural systems. The output contains displacements of each nodal point as weIl as stresses at the surface of each element.

3. 9.1.2.3.3 Effects of Flange Joint Connections/FTFLG01 The flange joints connecting the pump bowl castings are analyzed using FTFLG01. This p~ogram uses the local forces and moments determined by SAP4G to perform flat flange calculations in accordance with the rules set forth in Appendix II and Section III of the ASME Boiler and Pressure Vessel Code; 3.9.1.2.3.4 Structural Analysis of Oischarge Head/ANSYS ANSYS is used to analyze the pump discharge. head flange and bolting taking into account of the prying action developed by the flat face contact surface. The program is described in detail in 3.12.

3. 9.1.2.4 RHR Heat Exchangers'.

9. 1.2.4.1 Structural Analysis Program/SAP4G SAP4G is used to analyze the structural and functional integrity of the RHR heat exchangers. The description of this program is provided in Suhsec ion 3. 9.1.2. 3.2.

3.9.1.2.4. 2 Local Stiffness Calculations/BSTIF01 BSTIF01 is used to estimate the local stiffness of the heat exchanger shell at the attachment point of the supports. The method used in this program is based on the shell'tiffness calculations by P. P. Bijlaard as groundwork for WeIding Research Council Bulletin 107. The results of BSTIF01 are used to determine equivalent beam properties of the lower and upper heat exchanger support bracket to shell attachments included in the finite element model of the heat exchanger.

3.g-ill PCY:ggt:rf/45L6 9/23/81

WP 2 ected by massless elastic members, simula~g the dis buaer of shaft stiffness. The analysis was performed iterativ to obtain campatibiMty between the ro dis-placements e magnetic. and centrifuga3. fo s acting on the rotor..

All other'nalysis of sp 'fic motor amponents and pump components. cansisted of han t

al tions 3.9.1 2.5 RHR Heat ers Pollawing are the c ute programs used dynamic and static ana1ysis eteznCine st~taxraI, and ctional integ-rity of the eat exchangers:

uppart Load Seismic Analysis 'ED-6)

Stress Analysis- of Supports (ZD-8)

Other computer programs us ana3.yses Catec " of stru stems, and

'mic Other Computer Program Used in Anal functional camponents, int and static

= 'mic equipment and supports a" s ed in 4.1 and4 3.9.1.3 ExperimentaL Stress Analysi.s.

When experimenta3 stress analysis is used in Li.eu of ana3.yti-

~ cal methods for Seismic Category I ASHE Cade items., the re-qui-ements far experimenta3. testing enumerated in the which a e applicable for the specific components under AS'ade-test sha3.1 be applied. Rien testing is. requi ed for Seismic Category 3: non-ASME Code parts account. shaLL be taken af size effects and dimensi.ona3 tole~ces which exist between the actual part and the test pard ar paW as well as Mfe-ences wh'ch may exist in the u3.~te strength or other governing materia3. propexti.es of the actual part: and the tested parts, to assure that the loads obtained from the test are a realistic or conse~tive representation of the load c~ing capability of the actual stature under the postu<<

lated loading.

3 . 9-3.6

Il RP 2 Results of both ZSOFXNXT and iMSTBAN a e given in Table 3.12-2. As can be seen, theze is close car elation, between the deflections, with.NASTEAH giving Lazger values th=oughout the flued, head, than ZSOFZNXM. This is due to the lack of rotational freedom at nodes with HASTKQf ove the more" flexible shear elements in ZSQFTKZZ. This leads ta pre-diction af. hi.gher st esses usinq ZSQFXNXTZ(as can be seen by comparing pages 2 and 3 of Table 3.12-2). The compute proqpua XSOFXNXT is therefore a canservat've methad, for determining st=esses in fined head. fittinqs.

This praqram is. referred ta in 3.8.6.4.4.

3 12.10 ADLPZPE LPME is a digitaL camp'uter program developed by the A~M~

m Me Co'. and used for static and dynamic analyses of c Lee, piping systems. Znput data preparation uses pip~g Lan e and output information is presented for eas~iate-made for 'ut preta" n. The input data may be pze-processed an~plots and madel evalua~n.

prepazatio the e are many input ~r To aid in pid data diagnos cs. The output au I.cally includes a st=ess anal~ s as reaui=ed by ANSX.831.2. 967); 331.1 (1973); ASME e Section ZXX, CLass 1, Cl s 2 and Class 3 (1971 d 1974) . The ASME. Code, Section ~M, Class 1 analy incLudes cal-culation of fat'gue aqe factor an simlified elastic-plastic analysis. ALL ozces, m ts, deflecmons, tions and. a summary st=e zepa am included in the output or~@ aphic, iscmeC 'c, and, Addi.tianally, the program ste eoscaaic intezpre~g plating computed cap ze

'ty ts.

to aid checkinq input and The pipinq system is adeled as a eries of sect'ons. that L'e between network po's. A seam~ n x campased. of st=aight and curved memb ~ , and each member ma have common ar differ-ent laads. and ysicaL propert.es. The etwork points may be free, Lly or fuLly resMained an have specified displacem s that. represent thermal anchar splacements or seismic chor motion. Zntezmediate springs ground or jo' othe members may be placed within the section ta re sent spring hangers, pipe bellows, skew and . ided re-aints, support and eouipment stiffness. T ans e t=ix technicpxes aze used to.. reduce-'the si"e af the sti"Ress mat=w 3.12-15

ANSYS is a. general~uzpose finite element'computer program designed to solve- a variety of problems in engineering analysis.

The.AHSYS program features the following capabiU.ties:

1. Structural analysis including static elastic, plastic aad creep, dy aaa.c, I ~

seismic aad d~aic plastic, aud lazge deflection and stability analysis.

2. One-dimensional fluid flow analyses.

3. Traasieat heat tzansfer anaLysis including coaductioa, convection, aad radiatioa with direct input to thermal-stress anaLyses.

I 4 Aa. extensive finite. element 1ibrazy, including gaps, fzictioa intezfaces, springs, cables (tensioa oaly), direct interfaces (compression only),

curved elbows, etc. Many of the elements coataia complete plastic,

, creep, aad sweLLing capabilities.

Plotting - Geometry plotting is available for all library, including isometmc

..

aad perspective views of elements three-dimensional'ou:Sties in the AHSTS

6. 'ty Restarc Ca p abili - The ANSYS program has restart ca bili aualyseses types." 'Aa optioa is also available for savin the 5 tiffne zness g

mat~ once it is calculated for the structure., aad using it for other 1oading, conditions.

The program is maintained cuzzent by Swaason Analys' ysis ystemss Inc. of Pittsburgh gh, Penasylvania and is supplied. to General Elec ectmc f for use oa the Honeywell 6000.

The AHSYS program has beea used for productive analyses since early 1970.

Users now include the nuclear, pressure vessels and piping, miaiag, stzuctures, bridge, chemicaL, and automotive .industries, as well as many consulting firms.

~riNP 2 The static loads on the piping system may be th~1, dead ight, static "g seismic loads, exte nally applied for es moments, and wind 3.oads. The dynast.c loads are comput us g nozma3.'ode theory and seismic response spect-a or his zy forcing functions in one or more directions.

/ ~

,

The. a roach used in ADLPIPE to compute the respansegf piping ystem to graund shack inputs is, based upon w normal mode or odal supezpositian method- The fozmtz1ation in terms of normal modes, which is particularly advantageous',for ient repon e prablems, follows generally the form'iscussed

"~-

by Young (R e"ence 3.12-X5) .

f st ste. in the application. of this method is the

'he deteaninatian the natural f equencies.and mode shapes of the f=ee vMrati ns of the system. Pcu=i'a cansezvative linear5.y~lastic umped maps system, e governing mat ix ecuation is:

is the (dia incr~

'f M, )

E K is the. smdfness t ix, and.

u is the/column mat the displacement.

'oordinates.

The stifmess mat=ix KR utilized in e dynamics formula.ation differs miSe stiffness mat ix KR dgve3.aped by ADLP~~-

for the network points. The latter ma~, developed by w~sf ex.-,m'at=ix techniques, includes mass~ ants and interior network. paints. The stif&ess matrix, for. e dmmucs farm-Xatiani "equi es stiffness values at mass pa's only. Thus, K A~ a reduced fozm of K, and. can be shown be equal. to:

K > A.B.E '33 where t; K ~AS g

A ~ Mass points sub-matrix g B,D ~ Coupling suh~atrices B, ~ Branch points mxhaatM Deters~ation of the naturaL modes can proceed. by one of several'ethods. The two eigemra1ue routines ed in ADLPXPE are ~~e Jacahi (Reference 3.I.2-16) rota an scheme and the Given+ Householder (Ref erence 3.12-16 scheme; the latter has bees@ mod'~ied to incorporate a gestian made by Silk ~on (ReSpx'ence 3.12-17) . Xn the acabi, rau>>~, the operation's are ~ied out in care memo and the. number of degrees af f eedom is Limited by avails&le core. The Givens-Househalder rou~e i.s unlimited by ~re utili ation of secondary storage and yroduces the Xowes>> eigenvalues and associated eigenvec o s.

K Par a system having N deg= of freed', the eigenvalue routines will produce up ta&,eigenva3.ues (natu=al frequen-cies) ~i (i~1 ...N) and up to% sets of eigenvectars, pij (i~1, ...N, j~l, army )ij is ca11ed 2<... Thejj~ column of the (N x N) eigenve~r or the jth mode whU.e the a~y itsel is alled the. mo mat=ix The normal made ~ rmulation of the ws onse of a, lmped sys-tem. ta a shock placement can be 'ed out by considerin the kinetic potential ene~ies of th'e Loaded system.

Assuming the elastic displacement: of +e ith caa dinate is, u;, the total displacement ecgm1 ta. ui + si, where s; is <<& shock displacement of the ith coarse "normal" coo 'es qn (t) and p( ) (t) are defined bathe linea=

armation:

ui(t) ~ E. pi q (t) n~L N

s. (t) Z y~p~(t) n~L

3. 12-LT

'8NP-2 where pi is the ith element of the nth eigenvector. These t=ansformations are useful because the zesul~g equations f motion in terms of. the normal coo dinates a e completel coup3.ed f~ one another.

The kinetic energy T and the potential enmity V of th sys-tem, in'erms of the normal coordinates, are gven (ref ence 3.12-15):

)

N N H N 2 ~

1 i~1, n~1 r'~1 i~1 where m> ar the individual mass clem, and.

8< ~ pen ali"ed mass or the

\

~ mode', defined by the rela on:

2 MZ Substitution o f these en cxpzessions TIlto Lagzang& s, equation leads to the. e t" ns of motion:

~a 2

+ Ol <>L>>

n nJP-L wh c4 <4~~ I ~ the identity ma ~.

(t) l Ql s ~ (T) 't-.'2) dT Defining R (t) l s<(T) s (tW) dt the modal response can be ~~ simply as:

q (t) ~ E)<

nX.

~

The. expression -L represents Z4 n e poz~ of the maximum

<

modal response d eloped by. each zmal caor~te, and may be thought f as a measure of " e ex"ent of which the nth no~ mode p Acipates in the syn sponse af th s~~~a3 system. As, ~,sis of the total re-the s~e ar=ay 4 <,which the inverse of the modal " is termed the medal p wcipation mat=ix, with each e caz=es ndinp to the "pa~>>icipation in the overall, response t of the mat=ix synth is of mode n, and mass point E..

Th s

term Rn(t), expressed by the convolution ts the response of mode n as a func~n of at made n is uncoupled fram the other modes of

',

in~~;assuminq e

repre-systxn; i.e., R (t) is. the response of a single defame. o eedom systsn to Ae t=ansisnt loading given hy s (t) .

3 &-L9-,

.

The th ee dimensional shock inpu displacement, Dgn, is given in terms of a maxirann>>valued sperm (such as the Housnez earthquake input spectra) fo each principal ax@.s.

F instance) 'the vertical response may be dif erent~f~

the izontal response. Thus i the prese ibeX input

. foz h mode is the max~xm-value of .the respond Ru{t) devel dazing the overall duration of the. rp(ponse.

(These ues are obtained, for example, by sur ament wiA displac record.

t meters such as cantilever gages which peak value of the cU.splacement uzing the response period. )

(D<) R't) maximum The=efoze, the modal pli es beccme

'q (t) < E)~ )n For each norma3. mod, the

.given by am Mes at each coonLmte i aze h

L where ~

N

~ elas~~ amp1itude, a coordinate. i.

n~ixi u< ~ l. summed over modes T 's- then provides a set of displacements, x', for each of the modes. These individual sets of displacemen can then be applied to the system as equivalent static defle 'ons on a mode-by-mode basis. The corresponding network fo s aze oh~ed by ADLPME 3 3,2-20

L AHENDNENT NO ~ 9 Aaril 1980 ADLP each mo formed is g individual used As seen previously, the network mern ta compute rated by the transfer mat fore~ament

'f computes the non-mass network, fare~am s sets for offness matrix a series of many

s. This same accumu ted ~nsfer matrix is ts at interior points of the piping system (in ing e mass points).

The cumulative effect all mades is estimated by taking the square root of e sum of s es af the farce~ament acts at each pasitio the piping syst Por closely saaced frequencies n option exists which ena the addition of the abso e va3.ue of those moda3. maments then farming the squa of that sum in the square root of "squar'e tian This program is refe~ed to in 3.9.).2.2.

3 ~ 12 '11 RELAP3 This program describes the behavior af wate~oled nuc3.ear react ors du incr postulated accidents such as loss-of-caolant< num@ failure< or power ~sients. The behavior of the nrimarv cooling system and the reactor is emphasized. The program calculates flaws, mass inventories energy inven-tories, pressures< temperatures< and qualities along with variables. associated with reactor power, reactor heat transfer,- or control systems.

REEAP3 is an NRC accented computer program and is in the public. domain. For a camplete discussian of this program see Refe ence 3.12-18.

This program is efer ed ta in 3.6.2 2 Th and 3.6.2.3.1.

3. 12 ~ 11 1 RELAP4/NOD5 RELAP4 is a computer program- written in FORTRAN IV for the digital, computer analysis of nuclear reactors and related systems. Zt is primarily applied in the study of system t an-sient response to postulated perturbatians such as caolant loan rupture, circulation pump failure, power excursians, et .

The arogram was written ta be used for wate~oled. (PNR and BWR) reactors and can be used for scale;models such as LOFT and SEMXSCALE. Additiona3. versati3.ity extends its usefulness to related applications, such as ice. candense and contain-ment subcampartment analysis. Specific aations are available for ref load (FLOOD) analysis and far the HRC Evaluation Madel.

WNP-2 QUESTION NO. 14 All computer programs used in the design and analysis of systems and components within the BOP scope must be listed.

Methods of verification are required for all BOP programs.

RESPONSE

See revised FSAR pages (attached).

Summation This item is closed.

3.9. 1.2.7 BOP Commuter Pro rams A list, of &e principal computer programs used in dynamic and static analyses in the BOP scope is given in Table 3.9-18. With the exception of the Burns and Roe developed program, these programs are recognized and widely used in the industry with a history of successful applications. The Burns and Roe developed program listed in Table 3.9-18 is documented, verified and maintained by Burns and Roe as described in SRP 3.9.1 II2.b.

3 9.1.2-7-1 SRVDAM SRVDAMA (Safety Relief Valve Discharge Analyses Model 4) is a computer model which simulates the transient flow of steam, air and water in a safety relief valve discharge line (S/RVDL) for a time period of approximately 0.5 seconds after S/RV opening.

The model calculates transient fluid properties, forces and

, thermal distributions in the S/RVDL.

The piping system is initially filled with air and a water at the exit submerged in the suppression pool. Upon S/RV slug

-

actuation, steam enters the line and compresses the air which expels the water slug. The piping system is represented by two models: (1) a gas (steam and air) and (2) a water slug, which are coupled by common pressure and velocity at the air-water interface. The gas flow equations are expressed in finite difference form solved with the method of characteristics.

Provision for axial variation in flow area is included. Motion of the water slug is solved with a one-dimensional ordinary differential form of the momentum equation which is integrated axially to determine flowrate and displacement.

SRVDAM4 is based on the analytical model described in the G.E.

Report NEDE-23749-P (ref. 1) and G.E. computer code RVFOR04 described in NEDE-24695 (ref. 2).

Pro ram Version and Computer Currently SRVDAM version 4 is being used by Burns and Roe, Inc.

in conjunction with a CDC Computer.

Extent of Aa lication SRVDAM4 is a transient pipin'g fluid analysis program which began development in 1975 and is supported by Burns and Roe. It has been'sed on several in-house projects.

Test Problems SRVDAM4 has been benchmarked against problems provided in references 1 and 2 which have been compared with in-plant test data from Quad Cities, Monticello and CAORSO BWR plants.

References

1) "Analytical Model fo Computing Transient Pressures and Forces in the S/RVDL", NEDE-23749-P, General Electric Co., February 1978.

2) "RVFORO4 User's Manual, SRVDL Clearing Transient For Z-Quencher Devices", NEDE-24695, General Electric Co., December 1979.

TABLE 3. 9-18 Computer Programs Used -For Dynamic and Static Anal ses in the BOP Scone Document Compute Pro ram Traceabili Svstem Used ADLP IPE Arthur D. Little, Inc. (1) CDC 7600 Acorn Park (2) CDC 172 Cambridge, Ma. 02140 (3) National Advanced Systems AS/5000 ANSYS Swanson Analysis (1) CDC 176 Systems, Inc. (2) National. Advanced Elizabeth, Pa. 15037 Systems AS/5000 Argonne National (1) CDC 176 Laboratory (2) CDC 7600 SRVDAM 4 Burns and Roe CDC 175

t WNP-2 OSER OUESTION NO. 15 (3. 9. 1)

The computer code utilized in the analysis of the ECCS Pump Motor Rotor Shafts addressed in Paragraph 3.9. 1.2.4, ECCS Pumps and Motors, is not identified. This code should be identified and data 'presented for the validity and applicability for use of the code.

RESPONSE

Referring to the response to guestion No. 13, RTRMEC was used by the motor vendor to estimate the ECCS motor shaft displacement. The results of the calculation have been verified by (1) comparison with motor rotor bend test data and (2) comparison with the SAP46 results obtained by GE.

The comparison demonstrated the conservatism of RTRMEC.

Summation This item is closed.

h PCY: ggt/45L7 9/22/83. ~

WNP-2 DSER UESTION NO. 16 Provide additional details concerning the test program performed on the

'orificed fuel support to establish the validity of the program. In addition, provide justification for using the allowable limits by applying a 0.65 quality factor to the ASME Code allowables of 1.5 Sm for upset.

condition.

' ( I

RESPONSE

1: Test Program The following is a detailed description of the test program.

(Note: WNP-2's orificed fuel support is not required to conform with the ASME Code; however, the test program is designed to conform to the code ia order to verify the design adequacy.)

Two separate tests were conducted, and each test was designed to be in conformance with Appendix II of th ASME code Section III. The first test series verified the structural capability of the fuel support casting to sustain vertical design loads. A production fuel support was stresscoated and subjected to .an extremely high vertical load to identify the location and principal stress directions.

of the highest stressed regions. A second fuel support was instrumented with strain gauges: 12 uniaxial gauges were used where the principal stress directions were known from the previous stresscoat test.

Six rosettes were used where the principal stress axes could not be determined. (All the gauges used in the experimental 'eadily stress analysis were put in the regions of highest stress as determined by the previous stresscoat, test. ) The fuel support was mounted in a fixture simulating the geometric characteristics of both the load and support in the reactor. Vertical loads only were simulating the wei'ght load of the fuel assemblies. 'pplied, It was found that the fuel support could sustain a vertical load of 104,000 pounds before the onset of yielding in the highest stressed region. This 104,000 pound load represents a safety factor in excess of 35 based on yielding over the normal applied vertical load.

A second series of tests were conducted to investigate the resulting stresses induced in the fuel support by a horizontal (or lateral) load applied by the fuel assemblies during. a seismic event. A fuel support was instrumented with 15 three"element rosette strain gauges. The location of these gauges were determined from an initial computer analysss, and. represented the areas of highest, stress plus a few key locations of minimal material thickness.

PCY: ggt: rf/45LS 9/23/81 .

0 WHP"2 DSER The test fixtures used were designed to apply equal loads on all four pods. This was achieved by using two hydraulic cylinders to load two spreader bars. The load was transmitted into each spreader bar through balls which prevented moment build-up. Each spreader bar then loaded two arms, which in turn loaded dummy fuel lower tie plates. At the interface of the tie plates in the fuel support, the dimensions of these dummy tie plates were identical to those used in the production components. During loading, weight was placed at the top of the load arms approximately in the center of the fuel support. This loading simulated a vertical load which would be present due to the fuel assembly weight.

During the initial phases of the testing, it was discovered that the stresses induced by a horizontal load were,a maximum when the applied vertical load was a minimum. Hecause the fuel support is not attached to the guide tube and sits on a chamfered seat on the guide tube provided for that purpose, it was found that an increased downward vertical load actually enhanced the fuel support's ability to sustain a horizontal load. (With increased vertical load, additional rigidity was provided to the fuel support casting by the guide tube. )

A load cell was calibrated and installed on the lower hydraulic cylinder. Load data was recorded on a continuous recorder, and strain gauge data was recorded on a multi"channel recorder.'he total applied load was twice the load cell readings.

The first; horizontal loading applied simulated the ASHE code upset condition. For this condition the total vertical load was calculated to be just under 1,000 pounds with a. horizontal load of 2,600 pounds being applied. The calculated vertical load-applied to the fuel casting included, its weight, the upward component of a 1/2g seismic load, and the differential pressure across the fuel and the fue'1 support. The 2,600 pound load was taken from the fuel support design specification for the upset. event. A horizontal test load oi 3,000 pounds was applied to compensate for possible increased hydraulic piston friction, changes in friction due to a small amount of misalignment and/or cocking of the load arm in relation to the piston travel direction.

The test results simulating the upset horizontal loading conditions produced a maximum stress of 10,833 psi. The differential pressure stresses across the castings were computed. The 1,580 psi value obtained from the computation was then added to the test results.

(Differential pressures across the fuel support were not simulated in the test program.) The total resultant stress was U,413 psi for the upset condition. The total stress resultant was less than the ASME code allowable of 15;580 psi for the upset condition.

PCY: ggt: r f/45L9 9/23/81

WP-2 OSER A second series of test loadings were applied to the support casting and were designed to simulate the faulted conditions. No vertical

~

load was applied during this phase of the testing because of the net result of lg downward force due to gravity and the lg upward

~ component of force due to the safe shutdown seismic faulted event.

The horizontal test load was applied to simulate 5.,200 pounds of force for the faulted event.

Testing simulating the faulted horizontal loading produced a maximum stress intensity of 21,225 psi. A computed stress value of 1,580 psi for the internal pressure was added to the test result similar to that of the upset event described above. The addition of these two

'stresses resulted in a maximum stress intensity of 23,505 psi, which is significantly less than the 35,400 psi allowed by ASHE code for the faulted conditions.

2. equal ity Factor The 0.65 quality factor accounts for the fact that not all castings are fully volumetrically examined. It is specified in the ASHE Code, 1976 Edition, Summer 1976 Addenda, Paragraph NG-2571.2(a).

Summation This item is closed.

PCY: ggt: rf/45L10 9/23/81

NNP-2 Expand Paragraph 3.9.1.4.1.2 (page 3.9-18) to describe the actual mounting of the hydraulic control units and to just-ify the validity of the assumption utilized in the FSAR.

~Res onse Please refer to revised 3.9.1.4.1.2 (page 3.9-18) for the information requested.

Summation This item is closed.

)

~ ~

No analysis has been made fox the non-code components of the CRD for the abnormal, condition.

The. design adecuacy 'of non-code components of the CRD has been 'veri 'ed by extensive testing programs= on compo-

-nents parts, specially instrumented prototype d ives and production d ives. The testing included postulated abnormal events as well as the service life, cycle listed in. 3.9.3..1.1.

3.9.3..4.1.2 Hydraulic Control Unit.

The Hydraulic Control Unit (HCU) was analyzed for the SSZ faulted conditions, thxough the implementation of the com-

~

- vere obtained back to back ".."

))CUX ~~uC< ~eAeV'!i'>Y27H p Y

't puter code SAliXS (See 3.12) .. Using the method of "Sum of Absolute Values of the Nod'al Loads," the maximum stress on the HCU f=ame was .calculated to be 54,310 psi,. The maximum allowable for SSZ is 60,000 psi xor.the HCU. Theme stresses by as"---'ag ~<two HCUs -

The fundamental frequency ox the HCU is close to the fxe-cpxency at wWwa.'ch peak seismic shock will occu . This results

~ ~

togetne the top and bottom of the

~

)

~+) I~ '. ) ~

in overstressed conditions 'n the piping connected to the

~

HCU during the saxe shu down earthquake (SSE) . 3y the.

~

application-

~

of boltwn st"'fening struts

" '. and diagonal braces along the rows o+ installed to ~we HCU frames HCU's, the.

fundamental frecuency is raised sufficiently o avoid peak

~

) seismic response. The stresses in the .connected piping are thus reduced to acceptable values The analysis of the HCU unde faulted condition loads estab-lishes the structu=al integ ity of the system.

.3 9.1.4.3..3 CRD Housing The SSZ is classi.fied, as a faulted condition; however, in the CRD housing analysis the SSZ. event has been t eated as an emergency ccndition. The maximum membrane stress inten-sity occux's at the tube to tube weld near the centexof the housing. The stresses are within elastic limits and'are

'hown in Table 3.9-.2(v) .

~ ~

3 9-18

~ ~

Y

~ ~

NNP-2 Response to MB SER Question 517 Insert A to FSAR Page 3.9-18 each pair of tied HCUs is supported in each of three mutually perpendicular directions by means of struts and diagonal bracing connected from the HCUs to a three dimensional seismic support frame enclosing rows of HCUs and anchored to the concrete foundation. See attached Figure Q 17-1.

I ~ ~ ~ ~ ~

s

<<o I

o, F 1 ~ I~ '

~ ~

~

~ J' ' ~

~

~ ~

'

~ ~

~

f I~

A~ SSS'11 nr r<<wf

~ ~ F ~

%%sr A%Oooo jr >>sf <<IA'sf%1 roof>> I <<AS%

rlr rf<<k

<<%%II/'s iAO PN<<i Fo %$ % Ostl

~ IA11 1

t. toto Iss l.'I I ~

Aktwt t4 riess 4 4k.

A

'fsssf I V' 11114 r<< Ks ~ Io fsH hof% V4%se 4 i sf

~ I'lof

~o lf Aso V

~ sskA<<4 %%or toff ~ o%oo1 fo 4

~o Sss ls1

~ IO

~ 11% M Is

!

)I,C ~

I ~

I r%

~ o% ~

<<<<N

~ ssrs>> r

~ ~ N

~ Ooksro

~ ~

Nk>> ssf I ~

I

~ ~

~ Iso ~

r>>N s<<NI K sf N<<<<o%%

rs ~ ff r<<k s% FV

~>>Iffy~ ~ ISIS ~

4 wsl ss ro<<soo<<r sr

~aIKiaaa 41 %I ~ ~

4N I~

1%%% rsr ~ It A. A IIAIIrottf Ukst 1 1 C II 5TIIF tssts Wtl'au<<lf<<I y 5 P. PFI AI% Ot X tfaC *15 Pg<<f g ( <<'t III 4 Pj Ii I 0'I

~ ks11 Grill fAA1 g% rlECTBtC PAEt. IMWARY Na T PoP. PFILII ISSI4sSUttLIISIT I s. ~ ~ I l~

I.OrISTpACTitssi Illsltl 4;ri) lf ~ ..I'Sl~

X a Ptt. CCrssstD<<%

F,<.

question

18. Provide a commitment in the FSAR stating that all required piping restraints, components and component supports have been installed in the piping systems prior to testing.

~Res onse: Paragraph 14.2.4.1.2 indicates that certain test prerequisites must be satisfied prior to the initiation of any preoperational test.

System lineup-tests (SLTs) which require, as part b of paragraph 14.2.4.1.2 states, that pipe support inspections and adjustments be completed are examples of these preoperational test prerequisites.

In addition a separate, distinct SLT governing verification of proper installation and adjustment of component supports has been generated.

Execution of applicable'portions of this SLT on each piping system provides formal documentation of required support operability.

r The administrative frame work imposed upon the preoperational test program as described in FSAR Chapter 14 provides a commitment which requires that all required piping restraints, components and com-ponent supports have been installed prior to testing. In summary, sufficient discussion presently exists in the FSAR to'ddress concerns in this area.

Summation This item is closed.

~oestion j9. The applicant's preoperational test program covers the vibration and dynamic effects. However, the thermal expansion effects required f

in SRO 3.9.2.II-l.d, e and are not adequately addressed. The ther-mal motion monitoring program should deal specifically with verifica-tion of snubber movement,.adequate clearances and gaps to allow free movement of the pipe during heatup and cooldown and should include acceptance criteria and test procedures. Additional information on this program is required.

Response: The WNP-2 Thermal Expansion Program is conducted during the Startup Test Program which is described in FSAR section 14.2. The specific thermal expansion program is described in section 14.2.12.3.17. This section prescribes test purposes, prerequisites, a test description and acceptance criteria. This program will be applied to systems which experience an operating temperature greater than 250oF and are classified in one of the following categories:

- ASb1E Code Class 1, 2 or 3 piping system

- High energy piping system inside Seismic Category 1 structur es

- High energy system whose failure could reduce the functioning of a Seismic Category 1 feature to an unacceptabl e safety 1 evel

- Seismic Category 1 portions of moderate energy piping system located outside containment

- Condensate/feedwater piping per Reg. Guide 1.68.1 c.2.fhg A combination of visual inspection and remote monitoring of certain inaccessible locations on critical piping will provide data to make evaluations which address. the defined test purposes. Specifically on selected systems, a pre-heatup visual inspection to establish baseline test conditions is performed which confirms that no potential obstiuct~an thermal movement exists, pipe hangers are at their "cold positions", snubbers are at the mid-range of travel and adequate pipe whip restraint clearance exists. At an intermediate point and

,again at rate temperature, a visual examination of the selected piping systems is performed to confirm proper thermal- movement relative to the baseline conditions. At corresponding reactor system temperatures, data is also recorded from the remote monitoring devices and compared against test acceptance criteria. Following several heatup and cool-down cycles, the thermal movement measurements are recorded a second time to determine that proper "shakedown" of the systems has occurred.

Appropriate action based upon the test results is taken which includes a review of the system performance by the responsible piping design engineering organization and issuance of their findings.

i, 4 l

During the visual inspections, special attention is directed to the following areas of piping/reactor system support components:

- Pipe whip restraint to pipe clearance at rated temperature

- Snubber expected movement and swing clearances at various temperatures including rated

- Control rod drive support structure to CRD housing gap at rated temperature

- Main steam piping penetration guide movement at rated temperature-

- Reactor vessel seismic supports operability vessel to sacrificial shield stabilizers sacrificial shield to biological shield stabilizers

- Safety related process instrument piping movement such as:

Reactor Vessel Level instrument piping Main Steam Flow instrument piping RCIC Steam Flow instrument piping

- Hot. pipe containment penetration temperature profiles The remote monitoring locations have not been finalized. at present.

Piping systems to be monitor ed have been tentatively identified that include-: the main steam,. recirculation, feedwater, r eactor: core isolation cooling and, safety relief valve discharge 1'ine piping.

The actual locations and selected piping systems will be established after. an iterative selection process which consists of an assessment of the most advantageous measurement'ocations coupled with a review of possible monitor ing locations. Both the responsible piping design organization'nd the Startup Test'rganization. will thus cooperate to achieve an effective piping thermal movement monitoring program.

The finalized, detailed test procedure which delineates selected piping systems, applicable test acceptance criteria, visual inspec-tion techniques, remote monitoring locations and required test condi-tions will be available on site for NRC inspection 60 days prior to commencement of the Startup Test Program on a schedule consistent with the preparation of other Startup Test procedures..

Summation The Supply System will provide a reference in 3.9.2 to Chapter 14 of the FSAR. This item is closed.

NOTE: Chapter 14 will be revised per the above response.

t l,

Mt(P-2 OSER (3. 9. 2. 1)

The applicant has not given a clear description of the acceptance criteria for steady-state piping vibrat. ons. The staff's position is that acceptance limi s for vibration hould be based on half the endurance limit as defined by the ASME Cnde at 10 cycles.

ReSPONSE For steady-state vibration, the piping peak stress due to vibration only (neglecting pressure) will not exceed 10,000 psi for Level 1 criterion and 5,000 psi for Level 2 criterion. These limits are below the piping material fatigue endurance limits as defined in De. ign Fatigue Curves in Appendix I of A%ME Code for 10 cycles. The defin1 '.ions of Level 1 and Level 2 criteria are clarified in the text revision attached.

Summation This item is closed.

The FSAR will be changed to quantity Level 1 and Level 2 Criteria as indicated above.

NOTE: References in Chapters 3 and 14 will be verified.

PCY:rf:rm/45E8 8/19/81 .

l amp'itude of displacements and number of cycles per trans-ient of the main steam and recirculation piping are measured and the displacements compared with acceptance crite.:ia.

The deflections are correlated with stresses to verify the pipe stresses remain within Code limits. Remote vibration and deflection measur .ments are tak n during the following transients: t'.

Recirculation pump starts;

b. Recirculation pump at 100% of rated flow;

c. Turbine stop valve closure at 100% power; Manual discharge of each S/R valv:

d.

and at planned transient tests th: tatresult 1,000 psig in

.S/R valve discharge.

3.9.2.1.5 Test +valuation and Acceptance Criteria The piping response to test conditions are con"id<red accept-able if the organization responsible for the stress report reviews the test results and determines that the tests verify that the piping responded in a manner consistent with the predictions of the stress report and/>r that the tests verify that piping stresses are within Code limits. To en-sure test data integrity and test safety, criteria"have been established to facilitate assessment, of the test while it is in progress. These criteria, designated Level 1 and 2, .re described in the following paragraphs.

.2.1.5.1 Level 1 Criteria Zf in ourse of the tests, measurement 'ndicace tl at the piping x sponding in a manner t would make test termination prude the test is t inated. Leve' cri-teria establishes boas on movement that, a test hold or terminatio if exceeded, make datory. The limits on movement are based on maximum towable stress limits.

3. 9.2. l. 5.2 el 2 Criteria Confo ce with Levee 2 criteria demonstrates that e piping i esponding in a manner consistent with the stress re t 3.9-24

I t

Q(m Level 1 establishes the maximum limi;s for the leve'i of pip~ motion which, if exceeded, makes a test hold or termination mandatory.

If the Level 1 limit is exceeded, the plant will be placed in a satisfactory hold condition, and the responsible pipirg design engineer will be advised. Fo'lowing resolution, applicable tests must be repeated to verify that the requirements of the Level 1 ljmits ar satisfied.

CZL~

-Level 2 specifies the level of pije motion which, if exceeded, requires that the responsible piping design engineer be advised.

If the Level 2 limit is not satisfied, plant operating and startup testing plans should not necessarily be-alter .d. Investigations of the measurements, criteria, and calculations used to generate the pipe motion limits ~ould be initiated. An acceptable resolution must be reached by all appropriate and involved parties, including the responsible piping design engineer. Oepending upon the nature of such resolution, the applicable tests may or (ay not have to be repeated.

predict that the pipxn

.

'he system is not re predictions and fur'~ ana necessary. L

'n Failure to meet Level 2 s onse is .uns~~sfactory; 1 2 criteria is in teria doe- not rr~an it means that accordance with theoretical

'.ased on test results is d to screen out test results are consistent with predictxo d nee~ no ical review from those that must be evaluate 3.9.2.1.6 Corrective Actions During the course of the tests, the remote measurements are regularly checked to determine compliance with Level 1 cri-teria. If trends indicate that Level 1 crit:ria may be violated, the measurements are monitored at i,:ore frequent intervals. The test is held or terminated as soon as Level 1 criteria is violated. As soon as possible after the test hold or termina ion, the following corrective actions will be taken:

a~ Installation Inspection. A walkdown of the piping and suspension is made to identify any obstruction or improperly operating suspension components. If vibration exceeds criteria, the source of the excitation must be identified to determine if it is related to equipment failure.

Action is taken to correct any discrepancies before repeating the test.

b. Instrumentation Inspection. The instrumentation installation and calibration are checked and any discrepancies corrected. Additional instrumen-tation is added, if necessary.

c. Repeat Test. If actions (a) and (5) identify discrrpancies that could account for failure to meet. Level 1 criteria, the test is repeated.

d. Resolution of Findings. If the Level 1 criteria is i'iolated on the repeat test or na relevant discrevancies are identified. in'a) and (b), the organi~ation responsible for the stress report

,shall review the test results and criteria to determine if the test cari be safely continued.

I:tern 8 21 Snubber i'estin The Supply System's response to the letter from R. Tedesco .to R. Ferguson "Preservice Inspection and Testing of Snubbers" dated March 6, 1981 is contained in the letter from J. Shannon to R. T=desco, G02-81-313, "Pre-service Inspection and Testi"g of Snubbers," ."ated September 24, 1981.

This letter states that the Supply System will comply with all of the requirements contained in NRC letterof March 6, 1981.

Summation This item is closed.