ML20093M378

Rev 1 to Crack Growth & Leak Rate Assessment of Oyster Creek Emergency Condenser Sys Piping Outside Containment Below 95 Ft Elevation

ML20093M378
Person / Time
Site:	Oyster Creek
Issue date:	10/13/1983
From:	Covill D GENERAL PUBLIC UTILITIES CORP.
To:
Shared Package
ML20093M374	List: ML20093M371 ML20093M378
References
TDR-467, TDR-467-R01, TDR-467-R1, NUDOCS 8410220097
Download: ML20093M378 (41)

Text

,

I I

I I

I I Nuedear TECHNICAL DATA REPORT I

I TITLE: CRACK GROWTH AND LEAK RATE ASSESSMENT OF THE OYSTER CREEK EMERGENCY CONDENSER SYSTEM PIPING OUTSIDE CONTAINMENT BELOW TIIE 95 FOOT ELEVATION.

I I

I I

g I

I I g0gy7M 6 I . __-. _ _ ___ _ _ _

I 1

TDR NO. 467 REVISI'ON NO.

BUDGET TECHNICAL DATA REPORT ACTIVITY NO. 3153n9 PAGE 1 OF 39 I PROJECT:

OYSTER CREEK NUCLEAR GENERATING DEPARTMENT /SECTION E & D/ME/FA RELEASE DATE /C//3/E3 REVISION DATE #

DOCUMENT TITLE: CRACK GROWIH AHD LEAE ItAI4 Abbndbridd Ut H16 &lutUdtG1 GUnUdtbdi

" SYSTEM PIPING OUTSIDE CONTAINMENT BELOW THE 95 FOOT ELEVATION ORIGINATOR SIGNATURE DATE APPROVAL (S) SIGNATURE DATE

- . ,, / . ,_ /A D. W. Covills b [ M/ /B -/0 9 R. T. DeMuth [ M [ Y M // D. /d. Rf F. S. Giac be h 47 dM G. E. Von Ni N d,[at M NMO APPROVAL FOR EXTER,NA,L DigTRIBUTION DATE D.K.Croneberger[ . Y8383 Does this TDR include recommendation (s)? UYes O No if yes.TFWR/TR # OO339i e DISTRIBUTION ABSTRACT: STATEMENT OF PROBLEM Justify the use of visual surveillance as the means by which I~ leak deteccion will be performed under predefined crack growth criteria on the Oyster Creek Emergency Condenser System supply (steam) and return (condensate) lines outside containment below the 95 foot elevation.

SUMMARY

One weld per pipe size (below the 95 foot elevation) was I analyzed for crack growth and leak rate assuming an initial through-wall circumferential crack length of 2 wall thicknesses (2t). Plastic instability was assumed to occur when the crack length exceeded 90' of the pipe circumference.

Under normal operating conditions, the shortest time for a crack to grow from 2t to 90* was calculated to be 18 months.

Leakage from both the supply and return lines would be readily detected by visual means.

CONCLUSIONS I 1) Visual surveillance of this piping will enable detection of leakage from a crack well before a crack would reach an unstable length.

I 2) There is sufficient time to take appropriate actions (i.e.,

shut down or isolate the affected condenser) between leak detection and the time that a crack reaches an unstable length.

I ACTIONS TO BE TAKEN The Oyster Creek Technical Specification will be revised to require a visual surveillance of the subject lines once every i

24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.

ff ACOooo3o4-83 eCOVER PAGE ONLY l

. . _ _ - _ - _. - _ - . - -. . . - . . - -. .)

I ' '

I MNuclear '"74 T

TITLE CRACLGROWTH AND LEAK RATE & ASSESSMEM Ur litt r21thrA1 CONDENSER SYSTEM OUTSIDE CONTAINMENT BELOW THE 95 FOOT ELEVATION REV

SUMMARY

OF CHANGE APPROVAL DATE I 1 Substantially revised because of recent cracking event and changes in basis and methodology. R. . emu d/I

'F .

I. hS. Gi be A Y g .

. ve .1.e.

. n d-ts- ad

. . Crone erger

.I I

I

~

'I

, )

i lI l

l lI

,g

__ ,2 g>

g TDR 467 I- Rev. 1 Page 2 I TABLE OF CONTENIS Page

1.0 INTRODUCTION

5 2.0 METHODS 10 3.0 .ESet2s 1e g

4.0 CONCLUSION

S 20 5.0 ACTIONS 10 BE TAKEN 21

6.0 REFERENCES

22 7.0 TABLES 24 8.0 FIGURES 29 9.0 APPENDICES 33 TOTAL EFFECTIVE PAGES 39 I

I I

I i

l1 I

I t

z-IDR 467 Rev. 1

~

Page 3 Executive Sunmary I In 1979, Jersey Central Power & Light Canpany (JCP&L) perfomed a high en-ergy line break (HELB) evaluation of the Oyster Creek Emergency Condenser System (ECS) outside containment. The conclusion was that a pipe break could result in damage to the ECS isolation valves and controls.

I In 1982, the NRC-SEP issued criteria, canmonly referred to as the " Palisades Criteria", which permitted licensees to perform a safety assessment as an alternative to system modifications or alterations. Based upon these cri-teria, GPUN perfonned an analysis of the HELB locations identified in the 1979 evaluation and concluded that no system changes were required provided that a periodic visual inspection of the area was perfonned. The NRC-SEP accepted the conclusion reached on system modifications; however, the NRC-SEP considered that an automated leak detection method was necessary to detect low level leakage in the order of 0.1 gallon per minute (gpm) to protect piping integrity. GPUN stated that it would analyze crack growth and resultant leak rates to justify the use of visual monitoring.

I This report describes the methods used to estimate the crack growth and leak rates fran each of three locations in the ECS (one location per pipe size) below the 95 foot elevation outside containnent. TPa piping is Type 316 austentic stainless steel 8 ,10 , and 16-inch diameter. Cracks in this material will develop and grow primarily due to intergranular stress cor-

, rosion cracking (IGSCC) in the heat affected zones of the girth welds.

I I IDR 467 Rev. 1 Page 4 I Only enviromentally (ICSCC) controlled growth was evaluated; the contribu-tion of fatigue to crack growth is negligible in these lines. Both supply (steam) and return (condensate) lines were evaluated. For all cases the calculations were perfgmed for one month intervals, assming an initial through-wall crack length of 2t, where t is the aminal wall thickness, until the crack length exceeded 90* of the pipe ciremference (assmed in-stability) .

The results of the calculations show that the leak rates frm the cracks are sufficiently high to be detectable by visual means. Additionally, suf fi-cient time exists to take appropriate actions (i.e. , shut down or isolate the affected condenser) between the time of leak detection and the time that a crack would grow frm that point to an unstable length.

I The results support the use of visual monitoring as an acceptable method of leak detection.

I The Oyster Creek Technical Specification will be revised to require visual surveillance to be perfomed once every twenty-four (24) hours. Addi ti onal-ly, selected welds will be lef t exposed (i.e. , without insulation) to facil-itate leak detection.

I I

E I TDR 467 Rev. 1 Page 5

1.0 INTRODUCTION

1.1 Purpose The purpose of this report is to detemine crack growth rate and leak rate to justify the use of visual surveillance to detect leakage fran the Oyster Creek Emergency Condenser System (ECS) piping outside containnent below the 95 foot elevation and that adequate time between leak detection and the onset of crack instability exists to take appropriate actions. A system de-scription is provided in Appendix A.

1.2 Background

In 1979, Jersey Central Power and Light Canpany (JCP&L) per-fomed a high energy line break (HELB) analysis of the Oyster Creek ECS piping outside containnent below the 95 foot elevation and concluded that a pipe break could cause damage to the ECS isolation valves and controls [1] . JCP&L provided this con-clusion to the NRC. The NRC (SEP Branch) perfonned an on-site inspection, confinned JCP&L's findings, and requested modifica-tions to the ECS to provide adequate protection against the effects of a postulated HELB. JCP&L perfonned engineering studies of various modifications and concluded that none could be made on a retrofit basis that would effectively resolve all the potential problems and not impose significant limitations on access for inspection and maintenance. JCP&L notified the NRC of these conclusions and stated that they would perfona an ansi-ysis to demonstrate that the ECS piping would leak before a significant break could occur.

I

, r .-

I I IDR 467 Rev. 1 Page 6 The NPC had developed criteria [2], of ten called the " Palisades Criteria", which permitted licensees to perfonn a safety assess-ment based upon fracture mechani s or an augmented inservice inspection (ISI) program as alternatives to system modifica-ti ons. GPUN used these criteria to perform the safety analysis.

1.3 NRC Alternative Safety Assessnent Criteria (Palisades Criteria)

- 1.3.1 Detectability Requirements A leak detection system is to be provided to detect through wall cracks, both longitudinal and circunferen-tial, of a length of twice the wall thickness for minimun flow rates associated with nomal (Level A) ASME B & PV Code operating conditions.

1.3.2 Integrity Requirements 1.3.2.1 Level D Loads Show that circunferential or longitudinal through-wall cracks of four wall thicknesses in length subjected to Level D loading conditions exhibit stable crack growth and ensure that local or general plastic instability does not occur from Level D loads and the specified crack

' leng th s .

I I

I . . . . -- _

I I

TDR 467 Rev. 1 Page 7 I 1.3.2.2 Extreme Conditions ,

Denonstrate the stability of a circunferential through-wall crack of a length equal to the 1

greater of 4 vall thicknesses or 90* circunfer-ential length under fully plastic bending loads; hanger effects are to be neglected; snubbers are to be assuned as ineffective.

1.3.1.3 Material Properties Lower-bound material properties are to be used and justified.

1.3.3 Sub-Critical Crack Growth Consideration shall be given to the types of sub-critical cracks which may exist in the piping.

1.3.4 Augnented ISI Piping systens shall be volunetrically inspected to the ASME Code Section XI for Class 1 Systems regardless of the actual classification if corrective measures are not prac ticable.

GPUN adopted the Palisades Criteria, with the exception of auto-mated leak detection devices, and used them for a leak-before-break analysis.

I I

I _ _ _ .

I IDR 467 I Rev. 1 Page 8 1.4 Leak Before Break Analyd s It was concluded that: 1) the ECS A and B supply and return lines all exhibited crack stability (i.e. would leak before break) and, therefore, no system modifications were necessary and 2) visual surveillance may be used for leak detection since all the lines exhibited excellent stability for 90* circunferen-tial length through-wall cracks under both Level D and extreme conditions [3] .

The NRC-SEP accepted this position on stability; however, the NRC-SEP considered an autanated leak detection system was still necessary. GPUN responded by stating that it would perfonn a crack growth analysis to justify that the resultant leak rates ,

would be sufficiently high so that visual monitoring is an acceptable method of leak detection.

1.5 Leak Rate Analysis GPUN has perfonned a leak rate analysis on each of three loca-tions (one location per pipe size) identified in (4] as the most highly stressed weld. Each line is fabricated fran Type 316 austentic stainless steel. Welding of these lines resulted in circunferentially oriented, sensitized heat affected zones which will promote ICSCC. The analysis was performed assuning that a I

I -

I I IDR 467 Rev. 1 Page 9 I 2t-long circunferential through-wall crack was present at each loca tion. The crack growth rate and resultant leak rate were I calculated for one month intervals until the crack length ex-

- ceeded 90* of the pipe circunference. The effects of fatigue on the crack growth rate were not included since their contribution to crack growth on these lines is negligible; therefore, only envirotunentally (IGSCC) controlled growth under steady-state conditions was evaluated.

I I

i I

I  !

TDR 467 Rev. 1 Page 10 I 2.0 MEDIODS 2.1 Introduction To determine leak rates as a function of time, one requires the following information for each location analyzed:

a) Operating Stresses

1) Tensile (pressure)

2) Bending a) Gravitational (deadweight) b) Themal b) Crack Orientation c) Crack Gemetry d) S tress Intensity at the Crack Tip e) Crack Growth Rate f) Leak Rate Calculation Methodology.

Each of these items is discussed in this section.

2.2 Operating Stresses GPUN reviewed the stress analysis [4] and selected the most highly stressed point for each pipe size outside containnent and below the 95 foot elevation. Only steady-state stresses were evaluated since a seismic event is of such short duration that it will not significantly contribute to IGSCC propagation. The stresses for each pipe size evaluated are shown in Table I.

I l

1 l

I I TDR 467 Rev. 1 Page 11 The effects of weld induced through-wall residual stresses were not included for this analysis. For purposes of analysis of crack growth, the residual stress pattern for welds under one inch thick, which is appropriate for the ECS piping, has been idealized as being asymmetrical around the mid point of the wall thickness (i.e.(d"~ d * *#*

od is the residual stress at the outside of the pipe wall and (d is the re-sidual stress at the inside of the pipe wall.

In other words, the averags residual tensile stress driving the crack to grow is equal to the average residual ccznpressive stress that inhibits crack growth. Therefore, residual stress is not used as a factor in the circunferential growth of a through-wall crack [7,8]. This is a conservative apprcach since at discrete locations within the wall, the compressive force will, in fact, act to retard crack growth; therefore, the cal-culated crack growth rate, neglecting residual stress influen-ces, will be greater than that realized in actual cases.

For crack growth calculations, design pressure and calculated deadweight stresses were used for both the supply and return lines. Thennal stresses were used only for the supply line calculations; the return lines are at ambient temperature be-cause the return line valve just outside containnent is kept closed during nonnal operation.

I I TDR 467 Rev. 1 Page 12 I For leak rate calculations, Level A (normal operating) pressures were used sin:e this is operating condition for which leak de-tection is required [2].

Additionally, shrinkage stress resulting from the application of I weld overlays [14] to the system piping were included in the crack growth analysis. A value of 3000 psi (twice the maximun value assumed in the analysis of the overlay shrinkage stress

[14]) was assuned regardless of each analyzed location's proxim-ity to any overlay.

2.3 Crack Orientation Circunferentially oriented cracks were selected for analysis based upon a canbination of factors. For all the points select-ed except one, the axial stress exceeds the circunferencia1 (hoop) stress; therefore, the crack driving force will drive the crack tip in the circunferential direction. Also, field exper-

.I ience has been that axial ICSCC growth occurs only in furnace sensitized piping; the Oyster Creek ECS piping is girth-weld sensitized.

i I B

1. 1 4

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

I I TDR 467 Rev. 1 Page 13 I 2.4 Crack Geanetry The crack growth and leak rate analyses were perfonned assuming that a 2t circunferential ley;th through-wall crack existed at each point selected. The crack shape was assuned to be semi elliptical with an I.D. to 0.D. length ratio typical of I .

circunferential cracks caused by IGSCC in the sensitized heat affected zone of girth-welded stainless steel piping experienced in the ECS piping at Oyster Creek.

I In [3], GPUN showed that cracks in all three pipe sizes exhibit-ed excellent stability even when the crack length equalled 90' of the pipe circunference. For the purpose of this analysis, it was conservatively assuned that crack instability will occur when the crack length exceeds 90' of the pipe circunference. By selecting an initial crack length of 2t, the margin between the time of leaf detection and the time of the onset of assumed crack instability is easily identified.

2.5 Stress Intensity Factor at the Crack Tip Ihe fonnula for the stress intensity factor at the crack tip in tension and bending is given by I x- o tpra r + cr, fra F, t

tsi I

I - -

I g TDR 467 3 Rev. 1 Page 14 I where [t and b are the tension and bending stresses, re-spectively, 2a is the length of the through-wall crack on the 0.D. surface, and F and F are dimen ionless functions of b

a. This formula represents the stress intensity factor at the tip of an edge crack in an infinitely long plate subjected to remote tension and bending and is based upon linear-elastic fracture mechanics (LEEM). Values and formulas for F and F 8re provided in several docunents (e.g. , [5, 6,13]) . GPUN b

perfonned an engineering assessment of published solutions for F and F and selected appropriate values for each.

b 2.6 Crack Growth Rate 1 The formula for calculating crack growth rate is given by:

da/dt = C(K)", where K= K or K df and C and n are em-pirically derived constants. K, was used and the values used for C and n were 5.65x10 and 3.07, respectively [10].

C and n were derived from fitting a straight line on experimen-tal data points to give the best estimate of crack growth for furnace sensitized Type 304 stainless steel in low oxygenated water under constant stress.

I I

l l

I -

1 B

l TDR 467 5 Rev. 1 Page 15 I he effects of ' fatigue, primarily fra heatup and cooldown, were not included since the nunber of cycles is in the order of one to 10 per year. Extrapolation of fatigue curves to extremely low cyclic values indicates that the contribution of fatigue to crack growth in these lines is negligible; therefore, only en-virorsnentally (i.e. , IGSCC) controlled growth under steady-state I conditions is evaluated [10].

For all points the crack growth was calculated in one month intervals.

I 2.7 Leak Rate Calculations he leak rate calculations were perfonned using a research can-puter program developed under EPRI sponsorship. Given the up-stream thennodynamic conditions and the crack geonetry, the estimated leak rate through the crack can be calculated. he analytical model is a modified version of the Henry non equilib-riun two phase critical flow model. he details of the model and the assunptions are provided in [11]. he program was run for the Duane Arnold safe-end leak [12] in order to obtain a benchnark result. he rate calculated was 3.25 gallons per minute (gpm) versus an actual leak rate of approxi-mately 3 gpm. his shows good agreement between the calculated and actual leak rates. However, for conservatism, the calcula-ted results were reduced by a factor of 2 for the purpose of evaluating detectability.

I TDR 467 I Rev. 1 Page 16 I 3.0 RESULTS 3.1 Results The results of the calculations are tabulated in Tables II through IV. Figures 1 through 3 are plots of crack length and leak rate versus time.

3.2 Discussion 3.2.1 General The most highly stressed point is the location that will most likely exhibit the shortest time to failure (f nata-bility) once a through-wall crack has developed. This is because the crack growth rate is dependent upon stress intensity factor (K) which, in turn, is dependent upon the stresses to which the point is subjected, and the crack length. The result of these dependencies is that there will be little crack growth for a period of time and then crack growth will occur at an increasingly rapid although stable, rate until the unstable length is reached. The leak rate frcza a growing crack will in-crease accordingly.

I I

I

I TDR 467 I_ Rev. 1 Page 17 I 3.2.2 Return Lines Calculations show that a 2t-long crack will grow to 90*F of the pipe circunference in approximately 4 years, mini-m un . The leak rates fran these cracks are high enough to be readily detected at an early stage of crack growth.

- For example, the leakage fran a 2t-long crack in the 8-inch return line is approximately 3/4 of a gallon per-minute. This results in the leckage of approximately 1100 gallons of water in a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period.

This quantity of water would be readily detected. Even if surveillance were not able to detect the leakage fran the piping, the water acctnulating on the floor would be readily observed. Also, if surveillance on one day were to miss detection of leakage, it is highly unlikely that surveillance on the next day would not detect leakage.

I The results also show that crack growth is slow enough to be able to take appropriate action (shut down the plant or isolate the affected condenser) long before a crack would reach an unstable length.

I I

I I

I . . -_

B  ;

Ii IDR 467 Rev. 1 Page 18 A ~through-wall crack was detected in the "A" Return line in March 1984. A leak was visually detectea during a hydrostatic test of the " A" condenser. Destructive eval-uation led to the conclusion that the crack was a result of IGSCC. A major inspection and repair effort followed

[14]. Several welds were replaced and others were re-paired vith weld overlays.

The crack growth calculations perfonned for this evalua-tion appear to be conservative when canpared to the actual cracking experienced. Oyster Creek had been in operation for approximately 13 years before the cracks were detected. The fracture surf aces of the destructive-ly evaluated cracks were heavily oxidized indicating that they had existed for a long period of time. Crack growth calculations, both through-wall and circunferential, predict failure in a much shorter period of time.

3.2.2 Supply Line A supply line leak will be detectable by both visual and audible means. The points of interent on both the A and B supply lines are located just downstrea:n of two valves located outside containment (see Figures 1 and 2 of I .

I I

I I '

TDR 467 Rev. 1 Page 19 I Appendix A). Packing leaks in these valves have been readily detected both visually and audibly. The temper-ature of the area in which these lines are located is near ambient; therefore, condensation of the steam will form rapidly and will be readily detected by visual means.

E B

I I h I

I I

I I l I

i I

TDR 467 I Rev. 1 Page 20 I

4.0 CONCLUSION

S It has been shown that leakage fran the subject lines can be easily detected by audible and/or visual means well before unstable crack extension will occur in the subject lines. The methods and assunp-tions used yielded conservative results in that the calculated crack growth rate is higher and the calculated leak rate is lower than what I would be expected under actual plant conditions.

I I

I I

I I

I I

I

, I I

I

I I TDR 467 Rev. 1 Page 21 5 5.0 ACTIONS TD BE TAKEN The Oyster Creek Technical Specification will be revised to require a visual surveillance of the Emergency Condenser System area, both on the 95' elevation and 75' elevation of the Reactor Building once every twenty-four hours. The operator shall visually monitor the general areas around and under the Emergency Condensers, the supply and return piping including any valves or other system components, lie shall look for, listen for and report any evidence of water leak-ing fra the return lines, steam leaking fran the supply lines, or any leakage from other systen caponents. To facilitate the surveil-lance and leak detection, the sheet steel and insulation is wrapped around the piping will be removed for a distance of approximately two inches on each side of selected welds in the system. The most highly stressed weld per pipe size, per condenser on the 75 foot elevation will be exposed.

I I

I I

I I

I I -

I TDR 467 I Rev. 1 Page 22 I

6.0 REFERENCES

1) " Jersey Central Power and Light Canpany, Oyster Creek Nuclear Generating S tation, Pipe Rupture Analysis of High Energy Emer-gency Condenser Lines Outside Containnent Below the 95'-3" Ele-vation of the Reactor Building", EDS-Nuclear Inc. , Report No.

02-0370-1021, Rev. O, November 1979.

2) Alternative Safety Ass .snent for Selected High Energy Pipe Break Locations at SEP Facilities, Appendix 1 of Attachment to Enclosure 2, USNRC Letter to Consuner's Power,12/4/81.

3) " Fracture Mechanics Analysis of the Oyster Creek Nuclear Genera-ting S tation Emergency Coolant System" Revision 1, Fracture Proof Design Corporation, 6/30/82.

4) " Analysis of Faergency Condenser Piping Outside Containnent",

MPR-830, MPR Associates, Inc., July 1984

5) Estimation of Stress Intensity Factors and the Crack Opening Arer of a Circunferential and Longitudinal Through-Crack in a Pipe, Appendix 2 of Attacinent to Encloa. ore 2, USNRC Letter to Consumer's Power, 12/14/81.

6) " Elastic - Plastic Fracture Analysis of Flawed Stainless Steel Pipes", EPRI NP-2608-LD, September 1982.

7) Private conmunication with S. Ranganath (CE), 8/3/83.

8) Private conmunication with D. Norris (EPRI), 8/5/83.

I I

I - -. - -- -- - _ - - _-

I TDR 467 I Rev. 1 Page 23 5 9) " Guidelines for Flaw Evaluation and Renedial actions for S tain-less Steel Piping Susceptible to Intergranular S tress Corrosion Cracking" (Draf t), EPRI/SIA, April 1984, i

10) "The Growth and Stability of Stress Corrosion Cracks in Large-Diane cer BWR Piping", Volunes 1 and 2, EPRI NP-2472-SY, July 1982.

11) " Calculations of Leak Rates through Cracks in Pipes and Tubes",

EPRI NP-3395, December 1983.

12) " Investigation and Evaluation of Stress-Corrosion Cracking in Piping of Light Wter Reactor Plants", NUREG-0531, February 1979,

13) " Equations for Fracture Mechanics", K. E. Hoffer, Jr. , Machine Design, February 1,1968.

14) " Isolation Condenser Systen Piping Cracked Welds - Repair and Failure Analysis," GPUN TDR 580, August 1984 I

I I

I I

I I

I

I I'"

TDR 467 Rev. 1 P.ge 24 5 7.0 TABLES I I. S tre.se. used ,or cr.ek gr..th ..1y.e..

I

, 11. m e e. m .. .. . . . ,. . . .. . . .. e . . 1. .k .. . .. r .. < _ m . . .

g m. m u m. a .. .. . .. ... . ..a ....e .. 1.a .. . .. a .. - u .. .

g m. m m e. u .. .. . . . ,. . . .. . . .. m . . 1. . .. . .. , s ,, m . . .

I I

g I

I I

I I

I I

I

l l

TDR 467 Rev. 1 i Page 25 i TABLE I I STRESSES USED FOR CRACK GROWTH ANALYSES I

Pipe S tress Line Diameter (KSI)

Return 8" 11.07 Return 10" 10.24 Supply 16" 13.97  !

l I

.I l

I

I l3 lI I

I fg i

I -__-____ __ _

I TDR 467 Rev.1 Page 26 I

Table il l

8" Return Line g

TIME CRACK LEAK RATE '

(MONTHS) LENGTH (GPM) l I O 1.00 .78 l 3 1.08 .85 6 1.17 .92 I 9 1.28 1.00 Il 12 1.40 1.10 15 1.54 1.21 18 1.70 1.34 l 21 1.88 1.47 24 2.10 1.65 I

27 2.36 1.85 h 30 2.67 2.10 33 3.05 2.40 36 3.52 2.77  !

38 4.10 3.22 42 4.85 3.81 g

45 5.82 4.58 l

47 6.64 5.22 I  ;

lI . _ _ - _ _ - -

I ron a7 Rev.1 Page 27 l, Table lli i 10" Return Line i I ,

TIME CRACK LEAK RATE  :

(MONTHS) LENGTH (GPM)

I O 1.19 .94 l 5 1.33 1.05 10 1.51 1.19 I 15 1.72 1.35 l 20 1.97 1.55 25 2.29 1.80 30 2.70 2.12 l 35 3.23 2.54 40 3.93 3.09 I

45 4.89 3.85 50 6.26 4.92 55 8.30 6.53 g

I I  :

,I .

!I l <

I TDR 467 Rev.1 l _ _ _ ___ _. -.

Table iv

.l

- 16" Supply Line EXIT TIME CRACK LEAK RATE PRESSURE (MONTHS)

LENGTH (LBM/SEC) (PSI) l 0 1.69 .13 645 I 2 1.94 .15 645 l 4 2.26 .18 645 6 2.65 .20 645 8 3.16 .25 645 l 10 3.84 .30 645

'l2 4.75 .37 645 14 6.03 .47 645 L

I 16 7.89 .61 645 18 10.77 .84 645 19 12.81 1.00 645 I

LI l

'I l

I e-- --w--- w-w

+

8.

,I

~

TDR 467 I Rev. 1 Page 29 I 8.0 FIGURES i

i

1. 8" Return Line - Crack Length and Leak Rate vs. Time.

.I 2. 10" Return Line - Crack Length and Leak Rate vs. Time.

3. 16" Return Line - Crack Length and Leak Rate vs. Time.

I -

I I

I I P I

1 l

LI l

Iw I

I I-I I

g .

l I

TDR 467 g Figure 1 - 8" Return Line "O'so 7

I 90' 6 -

5 -

CRACK 4 -

I LENGTH (IN) 3 -

I 2 -

I '

I I I l 15 30 45 60 g TIME (MONTHS)

I 7 g 6 -

! g 5 -

l 5 LEAK 4 -

l l RATE _ l (GPM) 3 -

I 2 -

1' -

l g i i i 15 30 45 60 l TIME (MdNTHS)

I

.I TDR 467 l Figure 2 - 10" Return Line %k l

8 - 90' 7 -

g 6 -

CRACK 5 -

g LENGTH (IN) 4 .-

l 3 -

2 -

1

.l 10 i i 20 i

30 40 i i 50 60 l TIME (MONTHS)

I- 8

l 7 -

! 6 -

5 -

Lg LEAK l- RATE 4 -

i

~ (WPM) l 3 -

l l l l 2 -

I 1

1 i

.g i i i i 10 20 30 40 50 60 l TIME (MONTHS)

I -

.I TDR 467 I' Rev.1 Page 32 Figure 3 - 16" Supply Line 14 -

l 12 90' 10 -

l CRACK LENGTH 8 -

g (IN)

E 6 -

4 g

2 -

I I I l 6 12 18 24

'I TIME (MONTHS)

I.

1.5 I

~

I 1.0 -

l LEAK RATE g (LBM/SEC)

.5 -

I I 6 i ,

12 i

18 24 I TIME (MONTHS)

LI

- .- - - - a ._ __

I ,

TDR 467 Rev. 1 Page 33 9.0 APPENDICES I

A. ECS System Description I ,

l l

i l

i

. I I

I 1 lI I

I

. _ ._..._,...,,._...._..,_,,_,,__.,___...____,_._.,,,.,_m._.,_, , _ _ _ _ , , _ _ _ _ - , , , , , _ . . -

I  !

I TDR 467 Rev. 1 Page 34 APPENDIX A I

EMERGENCY CONDENSER SYS7EM DESCRIPTION

' The ECS piping runs fra the reactor vessel to the two (2) isolation condensers. Portions of the piping are both in the drywell (inside contaiment) and outside the drywell (outside contaiment) . The piping material is Type 316 stainless steel.

There are two (2) return (condensate) lines frm each condenser. These lines are eight (8) inches in dianeter and run down frm the condenser through the 95 foot elevation floor to the 87 foot elevation where they run parallel to the ceiling until they join into a 10 inch diameter line. This line then joins in series with an isolation valve and the superpipe which penetrates the drywell at the 87 foot elevation.

I The supply (steiam) lines 10 inch diameter, penetrate the drywell at a 90 foot elevation, expands to 16 inch diameter, and runs parallel to this

- elevation until it turns upward and penetrates the 95 foot elevation floor.

l Above the 95 foot elevation, the 16 inch live branches into two 12 inch m lines which eventually enter the ends of the condenser tanks.

lI Piping isonetrics of all four (4) lines are shown in Figures 1 through 4 cf l' this Appendix. The break point identification is also shown with an arrow on these drawings.

~

LI I - - - _

?,

I TDR 467 Rev. 1 i Page 35 i I During nomal operation, the return line valve just outside contaiment is closed. All the reaining valves in the systs are open. Tne pressure and temperature of the supply lines are 1034 psi and 548'F, respectively, and the pressure and temperature of the return linee are 1048 psi and 102*F,

, respec tively. When the conlensers are needed, one or both of the return line valves are opened and the pressure and temperature of the systs will vary depended upon the length of condenser operation.

I I

I I

I-I I

I I

I I

I .. . . . -. _ - _ _ - - _ _

- , . . , . , .-g , :n n ,

Appendix A ..-

~~

Rev. 1 Fiyure1 I ij - Page 36~~ -

l,l

'A Supply Line -

~

N h!\ g-l -  !!! '!!!

bw

-s;jt E lil!

uiin ; g g lI gj ,

. . , 5 A.., '

A\

\_ \ .

'j" NX \ N: ,

l ,

2,/ ii  !

" 1:. i g . ..

/ '

F l i.

'A T e

i f . /* ,

t

[s

• g, ' 'r

\'.

/\ .

9 /

\

/s

. 4 s ,: \u ,

7. /\

n.

. h

> S  %, %, / 't

' y' l l- ,. , , rr r

l!/ '\_Q /

g; g-

//

/'

i 4 \[.!

s 9

'\h\\ s l'

\

+'.9 ,. '-

g- .

x.

. s

\

g '> 0 @x , ~ k I. g' / f j i g. yx x t

u i

!- 1

-i 1 - i-6 m 1 ,,,g_,,,__j, _ _ z_ _ { ,, , . y_ _ _,j _,,, y

. _ _ _ . } _ _. . _ l _

l , Appendix A l. ., 1 _, TDR 467 Figure 2 . '

Rev. 1 --

3 'B Supply Line Page 37 ill I

0 I-i"

\,// -

l!n,,!!!! !!!!!!lil!!]!'

ll

~

2

. s . i .

i

[Th l-

\g .,

\

's-I.

/ 8 g{

ss *l l$

I

E a

v 1.

li

.I_

O v ..

~

c e [ ., ,

'\ .

... \

! ' ~#

s  ! \ 's #

e s.  %; _k\ h s: .,i

hs.;s'sfyj q/ x s \ u \. , 5 .< -

I ~

it .

y "g ,

e t%* .-

is

/

rk 4

g

/

\

~

M,  !.

A i

ig I. g .

e s!!

e --I--e-- l - -.-- ! - =-- I --

-- I - o 1 -- - ~ TDR 467

Figure 3 -

8 Page 38

'A' Return Line -

k%  ;

{!

j.

r.

y

, g g es i i g

.y,f , %gr'

=

.o 3.x _ g .

ss c

,m3 #

s -

xxg\c .!. ,

!p,-

1 v:,,p, I'/

/\ ,i f

's/;  !-!

I : /7s 4., \' , .

!!!  :[:

s//

.. g. ..

aj ,  ;

I I

A p / / ' A \\

/

A '

. - , , /1., .

I -

/

i  : \ /

i .

s

\ "x \

i E

,o (

<+

s G..

g

, ., N

.v j

\. .

.g . ..y m a

.? n\

'i!i a49%g \\y.(./ f 1 1 u i

I

%L' V i

~

1- / /(2 N

'Ns N i

/ g\ ['!

l- e x\\+c\ .

t d' . . -. . . - a

..}. ,. .[ r i . - ...t -. .. f.

Appendix A

- -l v- TDR 467 __

1 Rev. 1 e Figure 4 -

,- Page 39

'B Return Line i t 'ji I  ;' i lii!.' r g

\ \. '

I'FI'

'N bglldlI!] 1l .

, .}" N's \

'4/s/

9l.h I' h)(N s

ji! is

\ x s'

!  % ii Is la/. .

,n !

t t

o l1

~

.g [ ,

,p/

?

,/

lI g, ey'YA" I-gi ,

t,

/

c'r \

I! T, i\ //

l m , ..

\

\,  ! k X7 nx y(h~

s. '%s c gr

/b '\\ [u .N'j@ .

\ i\

i b 3

l. / W , s' '

!!- )

i 10ky j r i ,'

i _, -_ ._

,