ML20058E580

Nonproprietary Suppl 1 to NRC Bulletin 88-008 Evaluation of Auxiliary Piping for South Texas Project Units 1 & 2

ML20058E580
Person / Time
Site:	South Texas
Issue date:	11/30/1993
From:	Mel Gray, Roarty D, Strauch P WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML19311B207	List: ML20058E572 ML20058E580 ST-HL-AE-4639, Forwards Nonproprietary & Proprietary Version of Suppl 1 to TR, NRC Bulletin 88-008 Evaluation of Auxiliary Piping for South Texas Project Units 1 & 2, Per 931108-10 Meeting & in Response to NRC 930923 Rai.Proprietay Version Withheld
References
IEB-88-008, IEB-88-8, WCAP-12646-S01, WCAP-12646-S1, NUDOCS 9312070064
Download: ML20058E580 (129)

Text

{{#Wiki_filter:.. i m. Y 'I e ....g,. .,.'...f,': l ' 1 I r,

x y m..y y.w et ';.

-_.,..,u - y., q . r. j.w : o 2, ; 7~.. +. :.... -.:. :. ...e:... ..,a ^ ^ ~ ,.. l,'l}.. -l_;.l.? ' A "'? N-a ?'A 6 ' '.. ' 7 .~; l.._. ;+. x ~ l ' ~. _ h 5e ..y.. . _-... ;.- -... ;,: m

y ij -

N., f -r ,. :c..,4 - 4 - l -] - i '..'... ). . t... .e .. - :.,e. F .d i >:w ,.. ;c ,.3 c, . 3;,. Ni=,m,- c. t ..., : c.(.~ _ ;,.}:. . x... _ _. _. p...... - ,...,.r,~;. h....*b; >,;.; ;y.... < .T .+. - n. .s. k T 6 / e.. t i r.. V/ .-westinghotisernessystems: 4 y. 4. 1 3 c . :..R[.:. }_. ;.... ..z. h 5m;... [N ' ~.. I.: :.; +.~. .~ :.. 93120'/0064 931130 PDR ADOCK 05000498 l f .-Q PDR 4, 1

~ WESTINGHOUSE CLASS 3 (Non-Proprietary) l

WESTINGHOUSE CLASS 3 (NG-Proprietary) WCAP-12646 Supplement 1 i 5 s b i NRC BULLETIN 88-08 EVALUATION OF AUXILIARY PIPING FOR SOUTII TEXAS PROJECT UNITS 1 AND 2 NOVEMBER 1993 P. L. Strauch D. II. Roarty Reviewed b>hf[f/ Ay Approved by: M/h M. A. Gray # S.A.Swa Manager Structural Mechanics Technology C.M.'Pe% Approved by:. k d // */ M Reviewed by: zze R. D. Rishel, Manager Metallurgical and NDE Analysis Work performed for Houston Ughting and Power Company under Shop Order HOVP-964. WESTINGHOUSE ELECTRIC CORPORATION Nuclear and Advanced Technology Division P. O. Box 2728 Pittsburgh, Pennsylvania 15230-2728 01993 Westinghouse Electric Corporation

TABLE OF CONTENTS SECTION TITLE PAGE

1.0 INTRODUCTION AND BACKGROUND

1 2.0 NORMAL CHARGING AND ALTERNATE CIIARGING 2 LINE SUPPLEMENTAL EVALUATION

3.0 REFERENCES

4 4.0 MEETING AGENDA AND PRESENTATION MATERIAL 5 1 5

1.0 INTRODUCTION AND BACKGROUND

Westinghouse report WCAP-12598 (Reference 1) was issued in May,1990, to document the evaluation of the South Texas Project Units 1 and 2 normal charging, alternate charging and auxiliary spray piping for the effects of potential thermal stratification and cycling resulting from postulated isolation valve inleakage into the Reactor Coolant System (RCS), as described in United States Nuclear Regulatory Commission (NRC) Bulletin 88-08 (Reference 2). The conclusion of the evaluation was that the normal charging, alternate charging and auxiliary spray line piping integrity would not be jeopardized, should inleakage into the RCS occur over the life of the units. Nondestructive examination locations were identified as a prudent measure to supplement the existing inservice inspection locations, but were not intended as a means to provide continuing assurance of piping integrity. Subsequent to the issuance of WCAP-12598. the temporary temperature monitoring equipment which was installed on the Unit 2 piping wu remcved. A meeting between United States Nuclear Regulatory Commission, Electric Pow-r F.esearch Institute (EPRI), Houston Lighting and Power Company (HL&P), and Westinghouse personnel was held on November 8th and 9th,1993 to discuss the WCAP-12598 evaluation, and the utility position regarding removal of the temperature monitoring equipment. In addition, recent developments from the EPRI Thermal Stratification, Cycling and Striping (TASCS) Program were presented by EPRI and Westinghouse personnel. Application of this new technology on the normal charging, alternate charging and auxiliary spray piping was also presented, and was shown to support the conclusions of the WCAP-12598 evaluation. I I As a result of the meeting, the NRC requested that a supplemental evaluation be performed, assuming that thermal stratification cycling rezulting from RCS turbulence occurs at all locations within the unisolable piping sections of the normal charging and alternate charging lines (i.e., between the RCS coimection and the adjacent check valve). This cycling is assumed to occur [ ]* during normal power operation, while the line is isolated. The purpose of this evaluation is to ensure piping integrity for near term operation, assuming a worst case scenario of valve leakage and cycling. This evaluation is not intended to supersede the WCAP-12598 evaluation, and the results of that report remain valid. The purpose of this report is to document the supplemental evaluation requested by the NRC. This evaluation demonstrates the piping integrity of the normal and alternate charging piping for near term operation, assuming continuous isolation valve leakage at the critical leakage rate, and cycling [ ]* at the critical location within the unisolable piping. This report also documents the material which was presented at the November 8th and 9th meetings. I

2.0 NORMAL CIIARGING AND ALTERNATE CIIARGING LINE SUPPLEMENTAL EVALUATION 2.1 Overview of the 1990 Evaluation In the WCAP-12598 evaluation, leakage was postulated to flow through the normal and alternate charging isolation valves (XCV0003 and XCV0006, respectively). The source of this leakage is the outlet of the regenerative heat exchanger, which, under ordinary charging flow conditions, is 530 F. To conservatively estimate the stratification temperature difference within the unisolable piping, a regenerative heat exchanger outlet temperature of 477 F was assumed, which corresponds to maximum charging flow conditions. (A lower leakage source temperature will result in a higher stratification temperature difference). Since a significant length of piping exists between the regenerative heat exchanger outlet and the unisolable piping sections, the leakage flow can cool (depending upon the leakage flow rate and insulation properties). Accordingly, heat transfer calculations were performed to l determine the temperature of the leakage upon entering the unisolable piping sections. Using these calculated temperatures at the unisolable piping entrance, additional heat transfer calculations were performed to determine the temperature of the leakage along the length of I the unisolable piping section. The fluid stratification temperature difference u as then calculated [ ]'". Finite element analysis was used to provide the stress levels resulting from the stratified flow. 1 The critical components within the normal and alternate charging unisolable piping sections were determined to be [ ], since the maximum stratification temperature difference [ ]'" and maximum K stress index 3 [ ]'" coincide at this location. For this location and temperature difference, an alternating stress intensity of [ ]'" was calculated, which corresponds to more than [ ]'" allowable cycles. i Since the [ ]'" are greater than [ ]'" pipe inside diameters from the RCS connection, it was determined that, although this location could be hot from the combined effect of free convection and turbulent penetration, cycling could not result from i the interaction between the turbulent penetration and the stratified flow. Therefore, fatigue i cycles were estimated to be [ ]'". The incremental fatigue usage resulting from the leakage transient l was estimated to be 0.001 for the design life. At other locations within the unisolable piping [ ]'", the alternating stresses were calculated to be below the endurance limit stress based on 10" cycles. i i 2.2 Evaluation Assuming Cycling at the Critical Iecation Current techniques to determine the maximum distance from the RCS connection at which thermal cycling is possible are included in the presentation material provided in this report. Calculations for the South Texas Project normal and alternate charging lines have shown this j distance to be [ ]'". Since the [ ]'" l t f (

P ( [ ]'" are located greater than [ ]'" pipe inside diameters from the RCS connection, i cycling is not possible. However, an evaluation was carried out assuming that cycling does .l occur, resulting in stress cycles which are conservatively assumed to occur [ ]'". Over a forty year plant life, assuming an availability factor of 80%, continuous isolation valve l leakage and cycling [ ]'", there would be [ ]'" cycles of the leakage stratification transient. Since the alternating stress intensity remains unchanged at [ j'", the allowable is [ j'" cycles. The design fatigue usage for the normal and alternate charging [ ]'" l (Reference 3). Therefore, fatigue usage accumulation from design transients plus leakage transients is estimated to 1e no greater than [ ]'". This assumes that one i charging flow path (either normal charging or attemate charging) is always isolated, and therefore experiences all leakage transients. l ? He design analysis (Reference 3) assumes that each charging flow path will experience 60% i of the charging design transients, i.e., that switching operation between the normal and l alternate charging lines will occur. Since the leakage transient can only occur in the line l which is isolated from charging pressure, it is conservative to assume that each line will i experience a maximum of 60% of the leakage transients [ i ]'". On this basis, fatigue accumulation from design transients and leakage transients is estimated to be no greater than [ ]'". At the request of the United States Nuclear Regulatory Commission, a review was made to j determine the effect on the cumulative fatigue usage at the [ ]'" for combining the isolation valve leakage induced transients with the design transients. In the evaluation presented, the leakage transients were combined with themselves, and the resulting fatigue usage was added to the design fatigue usage. In an ASME design fatigue calculation, the leakage transients would be considered with the design transients, and transient combinations would proceed according to maximum alternating stress level. To determine if this more rigorous approach would affect the conclusions, the design transients (Reference 4) were reviewed. It was determined that the conclusions presented in this report would not change, based on the following reasons: - The design transients which significantly contribute to fatigue usage result from the shutdown and reinitiation of charging flow. These temperature excursions result in large positive and negative stresses (from shutdown and reinitiation, respectively). - Since the design transients are comprised of high temperature ranges [ ]'", the design transient stresses would combine with each other (positive with negative), and not with the leakage transients, which would result in a lower stress range. Therefore, it is concluded that the incorporation of the isolation valve leakage induced transients in the design fatigue calculation would not result in a significantly different result than obtained using the simplified method presented in this report. 3

2.3 Supplemental Evaluation Conclusions The value of fatigue accumulation at the normal and alternate charging line [ ] (critical locations within the unisolable piping) is estimated to be no greater than [ ]. This assumes that leakage occurs continuously during power operation in one of the charging flow paths, and that cycling occurs [ ]. Based on this calculation, which assumes a worst case scenario for valve leakage and cycling, the piping integrity of the normal and alternate charging piping is ensured for near term operation. l 1

3.0 REFERENCES

1. WCAP-12598, "NRC Bulletin 88-08 Evaluation of Auxiliary Piping for South Texas Project Units 1 and 2", May 1990, Westinghouse Proprietary Class 2. (Westinghouse Class 3, Non-Proprietary version is WCAP-12646).

2. United States Nuclear Regulatory Commission Bulletin 88-08, " Thermal Stresses in Piping Connected to Reactor Coolant Systems",6/22/88; Supplement 1, 6/24/88; Supplement 2, 8/4/88; and Supplement 3, 4/11/89.

3. Bechtel Calculation IC159RC5307, "ASME Class 1 Stress Report for Normal and Alternate Charging Line", Rev. 2, 11/2/88, South Texas Project Job No.14926.

4. Westinghouse Systems Standard Design Criteria 1.3, Rev. 2, " Nuclear Steam Supply System Design Transients", and Appendix A, Rev. O, " Fluid System Design Transients 3XL and 4XL (South Texas) Plants".

l l l 1 4 )

I 4.0 MEETING AGENDA AND PRESENTATION MATERIAL NOTE: PAGES WHICH CONTAIN A

• IN THE UPPER RIGHT HAND CORNER WERE NOT INCLUDED IN THE PRESENTATION, BUT ARE INCLUDED IN THIS REPORT TO CLARIFY THE TASCS METHODOLOGY.

5

AGENDA FOR MEETING BETWEEN 11L&P, WESTINGliOUSE, AND USNRC IN REGARD TO BULLETIN 88-08 "TIIERMAL STRESS" NOVEMBER 8 THROUGH 10.1993, IN PITTSBURGH, PA Monday. November R.1993 i 1:30 pm Start of Meeting and Introductions - USNRC & Contractual support personnel (BNL and subcontractor) i - Houston Lighting and Power Company personnel - Westinghouse personnel - Others (e.g, EPRJ) 1.45 pm ~ Meeting Objectives i 2.00 pm ~ Discussion of Phenomena (PROPRIETARY) - TASCS testing program - MHI testing program - Videotape l 3 00 pm ~ TASCS Evaluation Methe ' ology (PROPRIETARY) 1 - Note: Primary focus on South Texas Project, not generic - lleight of Stratified flows - Heat transfer of stratified flows - Thermal cycling zone - Turbulent penetration length - Heat transfer of uniform flows 4.30 pm ~ Discussion and Question / Answer Session (PROPIUETARY) 6-30 pm ~ Adjourn 1 l l

i AGENDA FOR MEETING BETWEEN IIL&P, WESTINGHOUSE, AND USNRC IN REGARD TO HULLETIN 88-08 "TIIERMAL STRESS" f NOVEMBER S THROUGH 10.1993. IN PITTSBURGil, PA 0 Tuesday. November _9, _L993 ) t 8 30 am Reconvene Meeting (PROPRIETARY) - Continuation of discussion from previous day, if necessary j i 9.00 am ~ Review of HL&P Monitoring and Analysis (PROPRIETARY) i - Monitoring program l f - Residual heat removal analysis 1 - Charging and alternate charging analysis i - Auxiliary spray line analysis t 11:30 am ~ Lunch t 12:30 pm ~ Discussion and Question / Answer Session (PROPRIETARY) [ . t. - Continuation of morning session, if necessary - Time available for independent study, if necessary 6:30 pm ~ Adjourn j i l t Wednesday. November 10.1993 8:30 am Reconvene Meeting (PROPRIETARY) - Continuation of discussion from previous day, if necessary l 9-00 am ~ Tour of Westinghouse Test Facilities i 10:30 am ~ Discussion and Question / Answer Session (PROPRIETARY) '{ t 11:30 am - Summary and Conclusion i - Follow-up actions and documentation 1 - Final remarks i 12:00 Noon Adjourn Meetieg l i ?

I i t p EPRl/NPD I EPRI PROGRAM ON THERMAL STRATIFICATION, CYCLING AND STRIPING (TASCS) l l Presented by Jong Kim EPRI 1 HL&P/NRC/EPRI Meeting l Westinghouse Energy Center 1 November 8,1993 SAFETY ASSESSMENT 003VIJH8Unos 1 1 l l EPRl/NPD EPRI TASCS Program r Initiated in mid-1989, involving all four i Owners Groups Objectives are to: - Develop screening enteria - Develop methods to evaluate the potentialimpact on the structuralintegrity of the component - Document TASCS guidelines in a handbook i SAFETY ASSESSMENT / 003VIJHavnec 2 ) l l

.~e [EPRUNPDTASCS Program Content ) l Phase 1 Plant data collection / analysis Literature survey Review of foreign and domestic data Categorization of plant geometries and l TASCS mechanisms Thermal striping evaluation Preliminary turbulence penetration evaluations Develop detailed Phase 2 program f SAFETY ASSESSMENT oezvuusu.= s F i [ EPRVNPD TASCS Program Content (Continued) Phase 2 Detailed analysis, testing and data correlation t l Workshop TASCS handbook and other documentation ) I SAFETY ASSESSMENT / l oesvu l L [

~ i-i I EPRl/NPD I Future Plans l r Tailored-collaboration Utilities workshop / seminar Updato handbook i ? SAFETY ASSESSMENT / l comJHw s i i i i r J I 1 l i

I 4.1 LOW TEMPERATURE TURBULENT PENETRATION TEST PROGRAM Introduction and Objectives in work completed in TASCS Phase I and subsequent efforts in this carrent work, a model for the empirically measured decay of the turbulent velocity in a branch line has been developed. The model, based upon a turbulent energy dissipation expression formulated by G.I. Taylor, has been accepted for almost 60 years. Further consideration of the turbulent energy convected into the branch line and its dissipation, in conjunction with some elementary concepts of turbulences, led to a law of turbulent penetration which may be described by two variables: a critical Reynolds number and the turbulent velocity at the inlet to the branch line. The purpose of the low temperature turbulent penetration testing, therefore, was to address two basic areas of concem: how the inlet turbulence varies with flow and geometric conditions, and the effect of "non-straight" piping geometries on the decay law developed from earlier work in Phase 1. j t Test Faciliti and Operation A sketch of the experimental arrangement is shown in Figure 4.1-1. It consisted l primarily of a simple recirculation loop of 10-in. diameter PVC piping with a 75 hp, 6,000 gpm pump. After leaving the pump, the flow moved up and passed through an elbow to turn horizontally through two 10-ft sections of 10-in. diameter pipe. It then entered a 4-ft-long convergent section that reduced the pipe cross-section to 6.375 in. l After passing the convergent section, the flow passed through a 10-ft section of 6.375 in. ID transparent pipe that led to a standard 6 in, by 3 in. Schedule 40 flanged steel tee. The tee had a flow-through diameter of 6.0 in., and it was felt that the sudden contraction entering the tee would alter the results, so the tee was bored out to 6.375 in. Coupled to this tee were various configurations of the test section. 1 m n io p i a n 4.1-1 i

t Low Temperature Turbulent Penetraton Test Program I I t f I v i i ? Standard 6X3 Tee 6 in. Pipe h s' Flow 6 in. Pipe I Probes v Probe Holder } i 10 ft Section of 3 in. Clear Pipe ( 6,000 gpm j Pump I Position of Probe Axial and " Leak" Injection Radial Motion Device I F i j l l Figure 4.1-1 Low Temperature Turbulent Penetration in Test Loop LC8-10 314 It> 102603 S 4.1-2 4 e.ss., e- .,- +. ---,.+r -,, - ~, e-n n .. - +,

l Low Temoerature Turbulent Penetration Test Program i 1 l Hot Film Anemometer Probe Thermocouple i 3.6 in. 1 " 0.0625 in. / / i w ) ( I l 1.0in. s' { \\ ) 's / Figure 4.1-5 Low Temperature Turbulent Penetration Test Fixture with Thermocouple and Hot Film Anemometer twsao m wunm 4.1-9

r Low Temperature Turbulent Penetration Test Program t i ii i i i LW PFESSLRE TEST: COWIGl. RATION I l [ 6* ( IN ECTION ) [ [3 l 2 I WADER) BRAPCH PIPE i PIPEd (wA) ) I E + LW PfESEURE TEST: COWIGURATION 2 i 6" r (1.5" N..,A N E i .. s.. v.t.... i VR= Vlevet Record, e, PIPE WA Het ritm Anemometer ]

• Locetten (w) d I

i 1 Figure 4.1-2 Low Ternperature Turbulent Penetration Test, Configurations 1 and 2 4 i l tE78-10 wpf Ib<1CC691 -4.1-4 i i

0 Low Temperature Turbulant Penetration Test Program I ? i Lw P!w35.NE ma COWIOLMATION 3 ( twmcTzow ) r ""T /' f t r, 2D i = e i Mucuri PIPE + b i w ) LN PFESOLN! TESTS Co WIGURATION 4 A i e g

1) INITIALLY FILL F PIPE VITH P

gggg gBS1TY M.AJID STAPmP M.AEWL PIPE ADC TAME

2) M M vExcus

,-r Man m _r L5c Wi W h * " t nonumewrarnos may E M:t X:1. P" ] MA= W1 F4 to Anoammet.r Q . t. tn 1 l Figure 4.1-3 Low Temperature Turbulent Penetration Test, Configurations 3 and 4 4 J t YC7813 m-/151M41 4.1-0

p i } Low Temperature Turbulent Penetraton Test Program l I L f LOW PRESSrJE TESTS COWIGLRATION 5 5 i M) i / w HEADER PIPE j i i Luer PRESSURE TESTS: COWIGURATION 6 fr ~ cen-a m-w ( YR ) INKrFUENTATION MtY ) q tw= H. der ve.locity vn. vi.u.t n..ed WAs He t PI ts An.meme t e r j = L...ti n I s- ) i I -) Figure 4.1-4 e Low Temperature Turbulent Penetration Test. Configurations 5 and 6 nei.w.pr mm 4.1-6 -l

4. Low Temperature Turbulent Penetration Test Program Table 4.1-1 Low Pressure Test Matrix Configuration Header injection Data Velocity Flow Measurements (ft/sec) (gpm) VR HFA j 1 Honzental - 3 in. 10 0 X 25 0 X 39 0 X 10 .2 X 10 .5 X 39 .2 X 39 .5 X 39 1.0 X e 2. Honzontal-1.5 in. 10 0 X. 25 0 X 39 0 X 3. 90 Degree Elbow - Short 10 0 X 25 0 X 39 0 X 10 .2 X 39 .2 X 39 .5 X i 39 1.0 X 4 Vertcal Down - Long 6 0 X l f 10 0 X 15 0 X 5. 45 Degree Elbow - Short 10 0 X l 25 0 X 39 0 X 10 .2 X j 25 .2 X 39 .2 X J 39 .5 X -{ 39 1.0 X { 39 2.0 X 6. 90 Degree Elbow - Long 39 0 X i I 39 .2 X i HFA - Hot Film Anemometer VR-Taped Video Record i w s.10 wpf15teu93 4.1-7 y

= - 4.2 LOW TEMPERATURE STRATIFICATION TEST PROGRAM i The purpose of the low temperature stratification testing was to provide a basis for interpreting the results of the high temperature-high pressure experiment and to provide an independent basis for checking the stratification theory. Three experiments were performed, each at several different flow rates for the stratified flow, to determine the stratified flow depth along the pipe. The three cases considered " normal" water injected into the pipe full of dense water to produce the stratified layer at the top of the pipe, denser water injected into the pipe full of normal water to produce the stratified layer at the bottom of the pipe, and finally the second scenario with the pipe sloped. Test Facility and Operation The low temperature stratification test facility consisted of a pump for the injection leak flow, holding and receiving tanks for the injection fluid, a variable area flow meter for t measuring the injection flow, and the test section with instrumentation. SchemGucs of the test section are shown in Figures 4.2-1 and 4.2-2. Test Section and instrumentation The test section for the low temperature stratification test used a 6-in. diameter pipe, which falls in the middle range of sizes for typical plant applications, was typical of lines where stratification is known to occur, and was of sufficient size for instrumentation and visualization. The low temperature stratification tests were performed at room temperature and atmospheric pressure, with injection flow rates ranging from 0.25 gpm to 4.5 gpm. The heavier water was doped with calcium chloride untilit was 20 percent heavier than regular tap water. This density difference corresponds to values which might be j encountered due to temperature differences ranging from 200 F to 300 F in a typical nuclear plant. LCL-11 wpf INU93 4.2-1

~ - Low Temperature Stratdication Test Program f i f l Duttot Intet h Condwetivity Pr 6 w Water W Water g YerIt Dertb beelving h1 _ t i 1 1l1 riev M.ter Tonk scevity I; I; j a / l 6' Plexistee DepedFluid(HighDonelty) i t i I i i Figure 4.2-1 Low Temperature Stratification Test Facility-Low Density injection Mode Simulating Hot Water injection into Cold Water l tc:?s.t p i m 3 4.2-2

Low Temperature Stratification Test Program [ t i 1 High Donelt i 8* Plexigles flulal ( ) [ H M *r l 1. 1 x T T T1.T T Y' Reeelving Depth rios

• f *r Tak Cen k tIvity Prob e t

f Intet 7 i l l I Figure 4.2-2 Low Temperature Stratification Test Facility-High Density injection Mode Simulating Cold Water injection into Hot Water ? tr.?B-11 wpf.1b.102693 4.2-3

Low Temperature Stratticaton Test Program i n.e. i i I e I h i t i Figure 4.2-3 Low Temperature Stratification Test Data - Low Density injection Along Top of the Pipe, Depth of Stratified Flow Layer Versus Distance from inlet j l tes.: -pt it io:es) 4.2-6

J 4.3 HIGH TEMPERATURE STRATIFICATION TEST I PROGRAM t i I introduction and Objectives l The main objective of the high temperature-high pressure tests was to support the development of evaluation methodology by obtaining thermal stratification data under heat transfer conditions in the ranges of the applicable leak fic,w rates. In particular, these tests were to verify the relationships among the flow rate, layer thickness, and f Richardson number, which in turn would verify the scaling rationale of the low temperature testing. Two experimental modes were planned: cold water flow under hot water, and hot water flow over cold water. In the first mode, the pipe is oriented with the dense j instrumentation at the bottom of the pipe, and water at ambient temperature (typically 50 F) is introduced into the lower part of a horizontal pipe which is filled with hot water at approximately 550*F. In the second mode, the orientation of the pipe is reversed, resulting in the dense instrumentation at the top of the pipe, and hot water at i approximately 550*F is introduced into the upper part of a horizontal pipe which is filled with cold water at ambient temperature. The hot water injection test has not been performed at this time. The temperature difference between the hot and cold water in both of these modes is approximately 500 F, and the flow rate of the flowing l layer varies between 0.2 gpm and 5.0 gpm. An additional transient flow test with a j peak flow of 8.5 gpm was also performed. i These conditions simulate the plant conditions in which a leaking valve allows cold or hot leakage flow into horizontal unisolable piping, but do not simulate the influence of f the main reactor coolant flow on the branch pipe (turbulent penetration). Turbulent penetration is addressed in the low temperature testing. l Test Facility and Operation i The high temperature-high pressure experiments performed under this program were carried out in a modified high pressure test facility. This test facility was designed and 4 11027&9 wpt14102793 4.3-1 l 1

1 High Terrperature Stratification Test Proatom i ..c.. i i l l l l l l J i \\ i l l l l l l 1 I Figure 4.3-5 Pipe Wall Thermocouple Locations t W7&G wptitF102593 l 4.3-10 1

High Terrperature Stratification Test Program s c.e i t t t 9 i t 3 l f Figure 4.3-6 Test Section Schematic Showing Thermocouple Locations t tC278-9 wpt15102593 i 4.3-11

4 High Temperature Stratstication Test Program Table 4.3-2 Summary of High Temperature Test Results Test Number 4 6 7 10 9 Approximate Flow Rate (gpm) 0.17 0.42 0.82 - 0.62 3.4 - 4.6 8.5 Hot Water Temperature ( F) 540 550 490 550 560 Cold Water Temperature ( F) 45 75 75 45 70 i The basic conditions include the approximate flow rate of the cold injection fluid for the test, and the initial hot and cold water temperatures. Figures 4.3-7 through 4.3-66 show the following for each test: Injection Flow Rate Versus Time - Figures 4.3-7,19,31,43, and 55 show the actual flow rate as measured by the orifice plate. Fluid Temperature Versus Time - Figures 4.3-8, 9, 20, 21, 32, 33, 44, 45, 56, and 57 show the fluid temperatures along the pipe diameter measured by the thermocouple probe, at Stations I and 11. These figures illustrate the heat and mass transfer effects from the pipe wall and the fluids on the thickness of the cold layer and the sharpness of the stratification interface with respect to time and distance traveled by the leak flow along the pipe. Fluid Temperature Versus Height - Figures 4.3-12,13, 24, 25, 36, 37, 48, 49, i 60, and 61 show the fluid temperatures along the pipe diametet at Stations I i and ll at a given time. Clearly illustrated is the difference in the sharpness of the interface as the flow progresses along the pipe. Inside Wall Temperature Versus Time - Figures 4.3-10, 11, 22, 23, 34, 35, 46, 47, 58, and 59 show selected pipe inside wall temperatures along the circumference of the pipe at Stations I and ll. These figures illustrate the heat transfer effects from the fluid on the pipe wall as the flow ptogresses. Inside Wall Temperature Versus Height - Figures.4.3-14,15, 26, 27, 38, 39, 50,51, 62, and 63 show selected pipe inside wall temperatures at various diametral heights along the circumference at Stations I and ll at a given time. These figures illustrate the heat transfer effects of the fluid temperature interface sharpness on the pipe wall. IM7&9,rpt it>102693 4.3-14 i

m m h i e' h I 8 t a s k TASCS EVALUATION METHODOLOGY I t f i I i 1 l - i %f I i i 4' I i m i

P .f C .g I i e-i i k l 1 I t i 8 i E f N 6 HEIGHT OF t b STRAT FIED FLOWS l .i E i r a 9 k ( J i 44 i I i L t P h i a 4 l

5.1 t HEIGHT - BACKGROUND AND VERIFICATION 1 Objective This section provides a method to calculate the height of the interface between hot and cold fluids of a thermally stratified flow in a pipe. r

Background

i When thermal stratification occurs, the pipe is partially filled with hot water and partially filled with cold water. In some cases (see Figure 5.1-1, top), the interface is very small and the gradient is very large. In the other extreme (see Figure 5.1-1, bottom), the transition between the hot and cold fluid can occur over the entire pipe cross-s,ection. The causes of each of these two cases are related in a complex fashion to the flow rate, temperature difference, length of flow, pipe slope, pipe material and temperature (insulation characteristics), entrance conditions, and exit f conditions. The case in which the interface is very small is generally more severe in terms of its effect on pipe stresses. To evaluate this case, it is necessary to obtain an estimate of l the height of the interface. Once this is identified, the velocity and other fluid parameters can be calculated for given volumetric flow rates. This is important l because calculation of the heat transfer and stability of a stratified flow is dependent on the flow velocity. Thermally stratified flow is similar to open channel flow when the interface is small. Since much work has been done to determine the height of open channel flow, this will be used as a starting point to develop a method for determining the height of a j stratified flow stream. Reference 3 contains a method to calculate the flow height based on integrating the energy equation from the critical depth (at the exit) back i toward the source of the flow. This formulation determines the height of the interface versus axial location. This formulation will not be used here because it is cumbersome to apply (a numerical integration is required) and it is generally not l necessary or practical to evaluate the depth of a stratified flow as a function of length. i t 10278-13 wotib-102693 3 o.1-1

Height - Background and Vinfication j i St Inttriec. ( ] j g Thicknow // n // Y/ L d li /._Wrnte K / \\ p_____4 Interfwe nickn y ____j + _s -1 + j / i c..ler umanamn s, i Figure 5.1-1 Comparison of Different Stratification Interface Thicknesses 1 Top - Small thickness (with respect to diameter) results in nonlinear temperature profile Bottom - Lange interface thickness results in a more linear temperature profile i t W7613wpf:1tF1020it3 5.1-2 ~ t I

i Height - Background and Venfication i i l 0 I e Y Interfee. N ight {/ x_ f, Ay i-1 i Y y A ^ s J 7 nt.cr.. a.i,st 1 g h A i P j nammuunnr 'l l l Figure 5.1-2 Terms Describing Thermal Stratification interface and Flow Area t C278-13 wpf 1610?$93 5.1-4

A Critical Depth Calculation i D3 Using the Froude number as the starting point, the equation for critical depth y, may i be determined as follows (see Figure 5.12). l The critical flow condition is defined by-j s.c.e l ) i l f I i b I I

A g. To account for the buoyancy effect in a stratified flow, g' can be substituted for g yielding: s.c.. The following formulas define the relationships that define the geometry of a stratified flow as illustrated in Figure 5.1-2. a.c.e ~ i ) i i l 1

Height - Background and Verification i a,c.e 5 l 8 I E h Figure 5.1-4 Correlation Between Critical Height and Measured Interface Height for All Data, Each Test Type Identified 1027&13 wot:1tFtC2693 5.1-15

_~ ~ t Height - Background and Verification E Using this data, a simple correlation is provided: l n,c.e i a k i { i P I l i t L i I i ? l, h Figure 5.1-5 Correia, tion Betwee,n Critical Height and Measured Interface Height for Horizontal Pipe l

Data, l

• c*

I t 027& t3 wpt 1tF102793 5.1-17 i f

1 E i t t e ) d 4 i HEAT TRANSFER OF STRATIFIED FLOWS i i t f i l ] i i W 4 u 4 ,,-r-w

Stratifk:ation Heat Transfer J a.c.e i i I l f 3 i Figure 5.2-1 Nomenclature for Properties Associated with Heat Transfer Calculations in Pipe with Thermal Stratification Flow j i t 02?&14 wpt 1D-102793 5.2-3 l

o I Given a leek flow into a horizontal pipe, the following equation describes the heat { balance on the system: n.c.e (eg. 5.2-1) i k l i i Rewriting eq. (5.2-1): ..c.. (eq.5.2-2) where: (eq. 5.2-2a) and (eg. 5.2-2b) i

Writing the energy balance on the portion of the wall in the vicinity of the leak flow (see FigureJ.2-1): n.c.e G (eq. 5.2-3) or, (eq. 5.2-3a) (eg. 5.2-3b) where: ..c.e t lt will be assumed _ ] This assumption has been verified for the range of a.c.e l conditions benchmarked in this program. Solving eg. (5.2-3) for T, and substituting in eq. (5.2-2) yields: a.c.e (eg.5.2-4) which has the solution: (eg. 5.2-5) Wiiere: D ,,g,, 9 d I .h

5 Stratication Heat Transfer 4 e' 1 1 } [ .. c.. t i i i t i i l I J Li Figure 5.2-2 Comparison of Predicted and Actual Heatup Curves, Test 4 I I tW7814 wpt 1b 102793 5.2-8

Stratification Heat Transfer s.e.. f i l i f i uI B Figure 5.2-5 Comparison of Predicted and Actual Heatup Curves, Test 10 tC27&14 wpt.1tp-102793 5.2-11

Stratification Haat Transfer 3 l 8,C,8 t Figure 5.2-7 Comparison of Predicted and Actual Heatup Curves, Types 1 and 2 Tw Case, O = 10 liters / hour tW78-14 wpt16102793 5.2-13 i )

~ l Stratifb:etion Heat Transfer 1 1 4 i a,c.e ) i L i a Figure 5.2-8 Comparison of Predicted and Actual Heatup Curves, Types 1 and 2 Tw Case, ) O = 30 liters / hour 10278-14 wptt>102793 5.2-14 i

{ l i i i h b THERMAL CYCLING ZONE t [ t I I 1 41 l 1 i

5.3 THERMAL CYCLING - BACKGROUND AND VERIFICATION Objective This section provides a method to estimate the location in a pipe where thermal cycling could occur as a result of the interaction of turbulent penetration and a thermal stratification flow. Recommendations on the temperature distribution and number of load cycles will be provided for fatigue analysis.

Background

Thermal stratification can cause high material stresses, particularly when the temperature difference involved is large. One of the mechanisms for material damage i from the stress loading is fatigue. The severity of the damage is related to the number of stress cycles, the stress range, the existence of stress risers, and the material. When leakage flows are significant enough to cause significant stratification in the vicinity of the main coolant pipe, there is the possibility of stress cycling in the i branch line. The cycling occurs if the turbulence in the branch line is sufficient to mix the stratified flow. Because the level of turbulence in the branch can fluctuate, the j stratification will periodically occur and then be mixed, resulting in temperature ] fluctuations in the fluid. It is important to understand turbulent-penetration-induced cycling to determine effective solutions and to evaluate the potential for a particular piping system to be susceptible to this mechanism. This is believed to be the mechanism which caused the pipe crack in the Farley and Tihange safety injection piping, prompting the issuance of U.S. NRC Bulletin 88-08. Figures 5.3-1 and 5.3-2 illustrate the configuration and outside wall reading for the cycling that affected the Farley safety injection piping. 1.027815 wpt 10102693 5.3-1

t Thermal Cycling - Background and Venfication a,c,e s i i l i Figure 5.3-1 Location of Temperature Measurements in Farley Safety injection Piping, i 1 t 0276-16 wpt itF102603 5.3-2 1 l l

i Thermal Cycling - Background and Venhcation g a,c.e j t l Figure 5.3-2 Temperature Data from Farley Safety injection Piping lilustrating Thermal Cycling 1:027& 15 wp':1 4 107693 5.3-3

Thermal Cychng - Background and VeYication N Summary of Experimental and Plant Data Two main bodies of data are available to evaluate the turbulent penetration cycling mechanism. One is the low temperature simulation tests performed to measure the turbulence using hot film anemometers and to obtain visual records of cycling (see Section 4.1). The second is the high temperature simulation tests performed by MHI to duplicate interaction of cold leakage and hot turbulent penetration, and to obtain M data through thermocouple readings Low Temperature Simulation Tests These tests were conducted to obtain the following information: 1. Characterize the turbulent penetration by measuring the velocity in the branch pipe using a hot film anemometer, which was capable of obtaining accurate velocity readings to less than 0.1 ft/sec. Readings were made as a function of distance from the main header pipe until the readings fell below the sensitivity of the equipment. Readings were made for five different geometric configurations to determine the effect of pipe size and layout. Based on the results of these tests, the following general conclusions can be drawn: a,c e ep e t 0278-15 wpt ib-102693 5.3-4

Thermal Cycling - Background and iterification s,c.e u 9 Figure 5.3-3 Turbulent Penetration Data 3-in. Test Pipe (Maximum turbulent penetration velocities normalized to header velocity versus distance from the header pipe, best fit and upper bound exponential decay indicated) t e$is -pt+ioren 5.3-5

Thermal Cycling - Background and Venfication a,c e 2. The interaction of turbulent penetration and stratification was studied by introducing a small flow of water dyed and doped with calcium chloride to a specific gravity of 1.20. The interaction of the leak flow and turbulent penetration was recorded on video tape and observations were made of the 1 location in the branch pipe where the stratification was mixed by the turbulent i penetration. The critical observation was the greatest distance into the branch pipe where cycling occurred. The purpose for obtaining this data is to support the development of a model and correlation for predicting where cycling could occur in the branch line. Based on this test the following preliminary a conclusions ;an be drawn: a,c.e l 1 l t027& 15 wpf;1b-102693 5.3-7

Thermal Cycling - Background and Venf."atbn N ~ ~ a,c,e I 1 i i Figure 5.3-5 Time Versus Turbulent Penetration Velocity Trace Configuration 1, at 4 Diameters From Header tW7815 wpt.itr 10?693 5.3-8

4 Thermal Cycling - BacCtground and blenfication t a,c,e - j l i ? i 1 i i I .l l i Figure 5.3-6 Time Versus Turbulent Penetration Velocity Trace, Configuration 1, at 10 Diameters From Header 1027&15 wpt 1b 1M693 5.3-9

O Thermal Cycling - Bacttground and Venfication MHI High Temperature Simulation Test A large test program was undertaken by MHI to study this mechanism under l conditions similar to those in typical PWR plant designs. To study cycling, the temperature at the bottom of the pipe was reviewed to determine where cycling occurred with respect to distance from the header (simulated main coolant pipe), since ~ all cases involved a cold leak into a hot pipe. Typk al time traces are shown in Figures 5.3-7 and 5.3-8. Figures 5.3-9 and 5.3-10 illustrate the location of the thermocouples for the two layout configurations tested. The data from configuration 1 illustrate significant fluid cycling at locations 5. 6 and 7 and none at location 9. The configuration 2 data indicate little cycling except in the fluid at location 12. It is interesting to note that the inside wall reading at that same location indicates no significant fluctuations. This is because the fluctuations have too short a cycle time for the metal to respond. Farley Plant Data a,c,e b e 5 Model to Predict the Upper Limit of Cycling (L) Based on the low temperature tests, two correlations have been developed for characterizing the maximum turbulent penetration velocity in the 3-in. and 1.5-in. branch lines (see Figures 5.3-3 and 5.3-4). I t W7& 15 wpt 1b 107793 5.3-10

Thermal Cycling - Background and Venfication a.c.e Figure 5.3-8 Temperature Versus Time Traces Illustrating Thermal Cycling, Type 1 Configuration 1027& 16 wptib-102693 5.3-12

t i Thermal Cycling - Background and Venfication l a.c.e i f i i i l kf Figure 5.3-9 Pipe Configuration Type 1 With Thermocouple Locations Indicated tc24is wpt 1>1026a3 5.3-13

b Q f Thenvl Cycling - Background and Venfication l P r i _, a,C,C l 4 / i I b I l~ l-t I \\ t t I 9 E i P l 'I b i 1 l i P i i tW78-16 wpt tb-102693 5.3-15 1 l l I ..m

Th:nnal Cycling - B2ckground and Venfication t i t The turbulent-kinetic energy per unit volume in the branch line is then given by: .c.. i (eg.5.3-4) l where: h I Thus, the specific maximum kinetic energy is obtained (for the 3-in. case) by substituting (5.3-3a) into (5.3-4): ..c, t i. (eg.5.3-5) it is reasonable to assume The total i specific energy of a leak flow is given by: ..c.. j (eq. 5.3-6) i I where: t i I L 1:Wl7&15 wpf 1b 102693 5.3-16

i Thermal Cycling - B:ckground and Verification n.e.e j For all cases of interest, ,,c,, .l J (eq. 5.3-7) l l l l 1 (eq. 5.3-8) Comparison of Model with Experimental Data and Application Recommendations Figure 5.3-11 is a comparison between the experimentally determined ( and the model L. All predicted data were conservatively calculated using the header pipe temperature and initial leak temperature. This was done to be consistent with a typical plant application. This correlation is excellent considering the differences between the low temperature and high temperature test parameters and the large range of parameters investigated. It would be conservative to predict a larger value of l x/d than actually occurred, since this would extend the length of pipe potentia!!y exposed to cycling. Frora the graph, it can be seen that all points yield a conservative prediction except those above 12 diameters. IVE7& 16 wpt 1&102ft93 5.3-17

i Thermal Cycling - B ckground end Venfication I k j i a.c.e t i i 1 F L i i 6 li i i t o i Figure 5.3-11 Comparison of Predicted and Measured Locations of the Maximum Extent of Thermal Cycling t02741& wpt1b 102693 5.3-18 ) )

i Thermal Cycling - Background and Venfication id A review of tfie test data clearly explains this apparent discrepancy. All test data in that regime involved very small leak rates, with the maximum being 0.13 gpm. All i these cases heated up to within about 90 percent of the ambient fluid at the location of cycling. Therefore, the actual lApl was only a fraction of the value used in the prediction. A check was made which confirmed that a conservative prediction of 1, would be made with the correct inputs. ( l For design purposes, a 20 percent margin was added to the predicted length. l Including this 1.2 factor, the resulting correlation is: ..c.. l 1 I i (eg.5.3-8) Where: l t 1, location of cycling from turbulent penetration with a design margin l = of 20 percent on the length Figure 5.3-12 is a comparison of the predicted location with the 20 percent factor f versus the measured location. Criteria for Pipes Under 3-in. Diameter l Equation (5.3-1b) characterizes the decay of the turbulent penetration peak velocities I in the 1.5-in. model. [ i _ e.c.e 1 j lt should be recognized that we would expect the cycling location for the smaller pipes ) [ ] * *'* This approach is conservative and shall be recommended until additional data are available for the smaller pipes. t W/F15 wot.1tF102693 5.3-19

Thermal Cycling - BacCcground and Verification l ..c.. .l l ~, Figure 5.3-12 Comparison of Predicted and Measured Locations of the Maximum Extent of Thermal l Cycling Using Design Equation for Prediction (includes 20 percent on length) tt1278-15 wpt.1D-1026W3 5.3-20

en A - -,s.. a n-M P l I t i t h e TURBULENT PENETRATION LENGTH s i ? b t b -*-namp. .w.% e e f I I 4, i l i

l. l i l-l l l 3.8 TURBULENT PENETRATION LENGTH Objective This section provides recommendations to estimate the effect of turbulent penetration l as a mechanism to heat up a pipe section.

Background

Branch pipes, attached to header pipes containing fluid with a large velocity, are i heated up by eddies which travel from the main header-tee into the branch line (see j Figure 3.8-1). This significantly influences the temperature distribution in the branch line which is needed to evaluate TASCS mechanisms in many cases. For branch pipes connected to typical PWR (pressurized water reactor) coolant i systems, testing and plant measurements have found that this mechanism can cause a line to be at temperatures near the header pipe temperature for[ la,c,e Recommendations Since analytical methods that can predict the plant behavior based on current knowledge have not been developed, a predictive tool will not be recommended. Table 3.8-1 summarizes currently available data on turbulent penetration length. t C'84 W 'b.100t93 3.8-1 i \\

_q i I l Turbulent Penetraton Length Q \\ t f l i i I f i l Weder Pipe / / / Turbulent oddlee -i NCO CD C) C) p \\ mw CD c 1 i e J j t s t ( N l M O r i 5 h - ), i Figure 3.8-1 Turbulence in the header pipe causes secondary turbulence in the branch pipe. -j which influences the branch pipe temperature. [ t i tens-s # wc:m 3.8-2 i i

Turbulent Penetraten Length Table 3.8-1 ~ Current Date on Turbulent Penetration Length .,c.. l l Based on this data, it will be necessary to assume a range of turbulent penetration I lengths or detemline the length by plant measurement. 1. For sines 3 in, and greater, in typical PWR designs, turbulent penetration length should be assumed to be [ ] * from the header. g Typical PWR conditions would include: Velocity of header between 45 and 55 ft/sec Temperature of the header between 540 F and 620 F Diameter of the header between 25 in. and 35 in. ID t 2. If a more accurate determination of turbulent penetration length is required, the branch line(s) should be monitored with outside wall thermocouples for sufficient time to encompass a typical range of operating parameters. Caution: Plant data have also indicated that the turbulent penetration length can j change during operation when activities such as power reductions take place. The magnitude of the change has been less than a few diameters; but this data is not sufficiently complete to determine any correlations that can be applied to other applications. See the example in Figure 3.8-3 for more details. m u -m 3.8-3 i

4 Twbulent Penetratcon Length Once a turbulent penetration length has been assumed, it is generally conservative to assume that the pipe will be at the header pipe temperature for the entire length. The temperature distribution of the pipe after that length should be estimate based on conduction or free convection (see Sections 3.5 and 3.6, as applicable). If it is critical to quantify the temperature distribution in a line due to turbulent penetration, and these recommendations result in a significant range of possible temperatures, it is recommended that plant measurements be performed. Range of Applicability 1. Branch pipe inside diameters from about 2.5 in. to 16 in. (approximately). 1 2. Header pipe diameters from 25 in. to 32 in. (approximately). 3. Pipe layout is not known to have an effect on Lrp. 4. Header pipe fluid temperatures (saturated water) between 500 F and 650 F. Appilcation Example The following examples illustrate the application of turbulent penetration length along with other effects discussed in previous sections. Figure 3.8 2 illustrates a simple case of a branch line off a main coolant pipe, with velocity U, = 50 ft/sec at T, = 560 F, and the isolation valve in the branch line closed. The various temperature distributions that would be postulated based on the turbulent penetration length, L p. are shown in the figure. The sections of the temperature plots 7 labeled "A" would be defined by the free convection methods described in Section 3.6. The sections iabeled 'B' would be defined by conduction, using the methods described in Section 3.5. Figure 3.8-3 illustrates a situation where a " cold trap" exists, which may or may not be within Lrp. If 1 rp < L, or L, > L, then this situation is reduced to that shown in Figure 3 1 2 3.8-1 with respect to turbulent penetration temperature. However, if L < L, < L, then i 7 2 the cold trap will affect the distribution beyond L. This case represents a potentially rp significant issue with rer,pect to TASCS. If Lrp is almost equal to L, then a small 2 t i tF** 1.. Mc:293 3.8-4

Tutoulent Penetration Lengtb k increase in L, could result in a significant change in L - L, as its distribution is 1 3 2 changed from case 2 to case 3. Also significant is that when this has occurred, and based on plant measurements, the section between L, and L has a significant 3 stratification loading for at least a few hours, until conduction heats up the section to nearly a uniform temperature. The magnitude of this stratification has approached the temperature difference between the header pipe and the ambient fluid. The possible differences in temperature distribution are illustrated in Figure 3.8-3. Another typical layout case is illustrated in Figure 3.8-4. This also illustrates the various temperature distributions that may be obtained depending on the relative lengths with respect to Lrp. I a Y i 4 1 tes-s wit 1Mme 3.8-5 j

Turbulent Penett. ten Length a.c.o I 3 r L I t i i i Figure 3.8 2 Turbulent Penetration and Free Convection Temperature Distribution in a Horizontal Pipe 002784 wpf l>102893 3.8-6 l l

Tumulent Penetration Length s.c.e P I l i Figure 3.8-3 Turbulent Penetration, Conduction, and Free Convection Temperature Distribution in a Pipe Section with a " Cold' Trap 002754 wpf.IblC893 3.8-7

t Turbulent Penettotion Length a.c.. k l i I t I h t i s 1 -l Figure 3.8-4 Turbulent Penetration, Conduction, and Free Convection Temperature Distribution in a Pipe Section Which Supports Free Convection. t C34 wpf It>102793 3.8-8

TASCS Screening Cntena 4 l e s i SaW Los LAvan* 20' ao* I ./ tim 2 LoaATrm 3 myg di 4' W AX AL LB STM F,* Pr % 3 a xu TrG4 1

• ALL DDelerGS M 9084 AM9pcDatELY m mamurrvic -

LocATra ,9j y gu numt omwueTran n.us comueTran i e ~ ~ uM:ATrGt 3 ux= Trow ex - rao (v) uxaTrow l =m ~ aM.Y = t / mens:T t i e as ao so o AKrAL LD STM (PT)' l . Au. Tecumnus a wascrms --me - I 1 %I 1 Figure 2-2 Plant Monitoring Data of Pipe Section Heated by Free Convection tC'8-19 wpf 1>1(E793 6 2.0-14

e h an. A .a e e a 1 M' 4, 1 J l i 4 I i l 1 P J' THERMAL STR P NG I i i 4 i I j 2 t 1 i 6 ? i j t 'd I e 9 \\l i 8 i e 4 9 1 1 1

I Step 1: Determine Pipe Richardson Number The pipe Richardson number shall be calculated for each thermal load being j considered: t I where: .. c.. k r i i Step 2: Determine the Bulk Fluid Temperature and Time For each thermal load case, determine the bulk fluid temperature difference (AT) and total tirra duration for the temperature case.

Step 3: Determine the Wall Temperature and Frequency For each$herrnal load case, determine the wall temperature fluctuations (AT, )and the frequency distribution. a,c.e where: ..c.. 1 I t 'h i -e 9 t

.i = Step 4: Determine Stress for Fatigue Analysis The procedure up to this point has defined the wall temperature fluctuation and the appropriate number of load cycles. The following equation shall be used to calculate wall stress for fatigue analysis. ..c.. where: .. c.. Step 5: Determine Load Cycles ..e.. Limits of Applicability This procedure is applicable to typical power plant pipes containing saturated water with the following limitation: Pipe material is stainless steel

b I i REVIEW OF HL&P MO NITORING \\ AND j A\\lALYSES II P i i l 1 i ) i

l REFERENCE DOCUMENTS REPORT MT-SME-522(88), OCT 1988 IDENTIFICATION OF UNISOLABLE PIPING INSPECTION & MONITORING LOCATIONS e WCAP-12067, DEC 1988 RHR LINE EVALUATION RESULTS WCAP-12108, JAN 1989 e 4 RHR LINE EVALUATION DETAILS

• WCAP-12598, MAY 1990 CHG, ALT CHG AUX SPRAY EVALUATION MONITORING DATA EVALUATION ISI RECOMMENDATIONS i

k

E M.L t'r f 9 9 s MONITORING PROGRAM

• MONITORING PERFORMED AT UNIT 2:

i NORMAL CHARGING ALTERNATE CHARGING AUXILIARY SPRAY

• RHR LINES WERE NOT MONITORED e MONITORING DATA FROM 2/12/89 - 3/27/89 WAS EVALUATED (WCAP-12598) e IN ADDITION TO RTD'S, P'LANT OPERATIONAL DATA INCLUDED:

HOT & COLD LEG TEMPS RCS PRESSURE PZR SPRAY FLOW STATUS RCP STATUS RHR PUMP STATUS NORMAL & ALT CHG FLOW

~ NORMAL CHARGING LINE j UNISOLABLE PIPING t RTD'S LP 1 15" XCV0001 N 560 F N 33" 4 LENGTH OF UNISOLABLE PIPING = 82"(24 L/D)

1. PIPING IS 4" SCH 160 (ID = 3.438")

e MONITORING LOCATION 9 L/D FROM RCS TOP AND BOTTOM OF PIPE I S g

ALTERNATE CHARGING LINE UNISOLABLE PIPING i l i l 73" COLD LEG s XCV0004 RTD'S LP 560 F e LENGTH OF UNISOLABLE PIPING = 73" (OR 21 L/D) PIPING IS 4" SCH 160 (ID = 3.438") l

• MONITORING LOCATION 16 L/D FROM RCS TOP AND BOTTOM OF PIPE

AUXILIARY SPRAY LINE UNISOLABLE PIPING TO PRESSURIZER n 3X2 RED RTD'S ) \\' 4 ' [ CV0009 18" (UNIT 1) ~ 21" (UNIT 2) ~ FROM RCS COLD LEGS (560"F) i I e PIPING IS 2" SCH 160 AND 3" SCH 160 MONITORING LOCATION AT TEE INLET TOP AND BOTTOM OF PIPE i 1

u. 5 P I &C,0 i) 4 6 N L ) I P b i i ) I l 6 1 P i ) i j l h t, t 1 1 P P ? ih ' f f ? h i r 4 L s 't ,i e l r 9 o i h v k I ..n ? s f i ) i assuem L P k l h k i , n

S V g t a b h b t 1 s 1 i 1 4 eur-duuun i

p d I S.C.O 1 j .1 i-i t I 1 1 1 r J 6 1 i 9 a i i s I Y ii s P 1 I h 6 h i t 4 8 t t I h i 9 6 e 6 1 4 f r ) r 4 s f 4 f I 'l. t

l 9 4 1 1 i l l l k 8,C.8 i m 1 4 4 I, h h i I t I-F I l I s B 5 1 6 J l s I i l } l d I I t b I o i h M f i I t n t I I i 3 e i k i e I 1 I e

l MONITORING DATA CONCLUSIONS l i NORMAL CHARGING LINE ~ t e NO SIGNIFICANT THERMAL ACTIVITY OTHER THAN DESIGN TRANSIENTS ALTERNATE CHARGING LINE l i e LIFT CHECK VALVE, ADMITTING FLOW THROUGH BYPASS LINE, RESULTED IN ] STRATIFICATION OF ABOUT 20 TO 40F j e MAX' STRATIFICATION WAS ABOUT 140F i WHEN CHARGING FLOW WAS SECURED

• CYCLING WAS NOT OBSERVED l-AUXILIARY SPRAY LINE e STRATIFICATION APPROX. 30F o CYCLING WAS NOT OBSERVED

) I 1

h 6 METHOD OF ANALYSIS

4 ANALYSIS METHODOLOGY i TRANSIENT DEFINITION L HEAT TRANSFER CALCULATIONS STRESS ANALYSIS LOCAL THROUGH-WALL STRESS GLOBAL MOMENT STRESS i FATIGUE EVALUATION i 4 b

d A 1 e a .y u - 4 a D RESIDUAL HEAT REMOVAL J I i I E l 1 l 1 I ) i 3 v r p e

i RESIDUAL HEAT REMOVAL LOOP 1 UNISOLABLE PIPING LEG 3' 2.6' 3.7' 19' ISOLATIOt VALVE PIPE IS 12" SCH 140 (ID = 10.5") LENGTH OF UNISOLABLE PIPE = 36 L/D LENGTH TO HORIZ PIPE = 4.5 L/D

RESIDUAL HEAT REMOVAL LOOP 2 UNISOLABLE PIPING LEG ISOLATION .3' VALVE 2.6' 11.6' / 3.6' PIPE IS 12" SCH 140 (ID = 10.5") LENGTH OF UNISOLABLE PIPE = 22 L/D LENGTH TO HORIZ PIPE = 4.5 L/D

O d P RESIDUAL HEAT REMOVAL LOOP 3 UNISOLABLE PIPING 13' N 19' ISOLATION \\M VALVE i LEG 4 3.6' 1.3' 2.6' PIPE IS 12" SCH 140 (ID = 10.5") r LENGTH OF UNISOLABLE PIPE = 45 L/D LENGTH TO HORIZ PIPE = 4.5 L/D

~ t COMPARISON OF NRCB 88-08 SUPP. 3 AND SOUTH TEXAS RHR LINES SOUTH SUPP. 3 TEXAS LINE SIZE 8" SCH 140 12" SCH 140 (ID = 7") (ID = 10.5") LENGTH TO 24 L/D 4.5 L/D HORIZ PIPING (VERT. DROP) ) HORIZONTAL 5 L/D 32 L/D DISTANCE TO 17 L/D ISOLATION VALVE 41 L/D LENGTH OF VERT bD ENTIRE PIPE EXPERIENCING LENGTH TURBULENCE i ISOLATION VALVE OPEN N/A LEAKOFF LINE

4 TEMPERATURE OF STRATIFIED FLOW 4 I i i \\ \\ 4

FINITE ELEMENT LOCAL STRESS MODEL s A w _ _'2*_ _ _ Insulated hot / cold interface, I ..e. I l .. e.. 8,C,C FOR DELTA T = _,,c,, MAX AXIAL STRESS = KSI ,,c,c MAX STRESS INTENSITY = KSI STRESSES AT OTHER DELTA T'S ARE PROPORTIONAL

a FATIGUE RESULTS FOR LOOP 3 RHR NODE ALT STRESS CUF CLIF NO. (KSI) DESIGN TOTAL 494 10.2 0.04 0.04 493 16.8 0.26 0.26 490 21.5 0.01 0.057 467 7.2 0.87 0.87 498 14.8 0.00 0.00 TRANSIENT ANALYZED ALTERNATED FROM STRATIFIED TO STAGNANT CONDITIONS ENDURANCE LIMIT STRESS = 17 KSI CYCLIC PERIOD ESTIMATED AT 2.5 HRS BASED ON SUPP. 3(20 MINUTES) AND LOOP 2 RHR HORIZ LENGTH TO THE ISOLATION VALVE (15 FT) )

RHR EVALUATION ) USING CURRENT TASCS { EVALUATION METHODOLOGY

• TURBULENT PENETRATION LENGTH IS ASSUMED TO BE

[

• ^* PIPE DIAMETERS FROM HEADER PIPE a,C,C e CONSERVATIVELY ASSL)JWjlNG[ ' PIPE ID'S, THE

] TP LENGTH IS ABOUT] FEET FOR THE RHR LINES 1 . SINCE THE VERTICAL DROP FROM THE RCS IS ONLY ABOUT 4 FEET, THE PIPING WILL BE HOT FROM i THE RCS TO THE VERTICAL PIPING . THE HORIZONTAL PIPING WILL BE MAINTAINED AT APPROXIMATELY THE RCS TEMPERATURE DUE TO FREE CONVECTION . LEAKAGE FLOW THROUGH THE ISOL.ATION VALVE ~ WOULD NOT RESULT IN ADVERSE STRESSES IN THE UNISOLABLE PIPE, SINCE A TEMPERATURE GRADIENT WOULD NOT EXIST RCS i 625F 1.3' 2.6' ISOL. VALVE 15' TO 36' I POTENTIAL LEAKAGE i

p k t 4 RHR MONITORING DATA

• i

..c.. i i I t t i P 1 p 6 ? l 5 ? i i F 1 1 i 4 l l ) ~ l l l 1 } I

i I ) CHARGING, ALTERNATE CHARGING AND AUXILIARY SPRAY 1990 ANALYSIS i l l i l e < I .)

.- ~.. ~ t i i CHARGING SYSTEM LAYOUT i h 7' {, 2 ' XCV0009 7' 5 t I 2' 2' 'y 18' LV3119 3 { g REGENERATIVE o' o O. HEAT 25' 21' EXCHANGER 17 'Q AUXILIARY s-t y--. s NORMA 4' 2' SPRAY LINE p l' I ,7, I' CV0e (TO RCS COLD LEGS) i i CV0008 W f f POINT A" 3' 18 1' 6' 13' 2' COLD LEG XCV0003 XCV0002 XCV0001 1, t 3 24' W 5' 3' 1* 1' 29' NORMAL CHARGING LINE XCV0 W / 7' 10' p. 49' 3' 1' 1' 1' 5' e N RCS COLD LEG 2 XCv0005 XCV0004 LOOP 3 ALTERNATE CHARGING LINE j i REGEN HX OUTLET TEMP = 477 F (MAX CHG)

• APPROX LENGTH FROM PT "A" TO CHK VALVE CHARGING - 86 FEET ALT CHARGING - 107 FEET AUX SPRAY - 183 FEET 1

l i i i i i

DETERMINATION OF LEAKAGE TEMPERATURE AS IT ENTERS THE UNISOLABLE PIPING 1 Tamb = 100 F ISOLATION CHECK 560 F VALVE VALVE RHX X / RCS V V \\ o p LEAKAGE ENTRANCE TO UNISOLABLE PIPING LEAKAGE EXITS RHX AT 477F (MAX CHG) LEAKAGE COOLS IN ISOLABLE PIPING DEPENDENT ON: PIPE LENGTH LEAK FLOWRATE INSULATION PIPE SIZE AMBIENT TEMP

4 6 LEAKTEMP AT ENTRANCE TO UNISOLABLE PIPING VS. LEAKAGE FLOW RATE ..e.. s-- ,-.m

1 TOP-TO-BOTTOM OF PIPE TEMPERATURE DIFFERENCE VS. l DISTANCE INTO UNISOLABLE PIPING S,C,8 l I [ t 1 k h a I [ .i

4 ANALYSIS ASSUMPTIONS UNISOLABLE PIPE IS 560 F 1 MINIMUM LEAKFLOW CAPABLE OF = STRATIFYING IS GPM MAX DELTA T IS "^ MINIMUM LEAK DEPTH = AMBIENT TEMPERATURE = 100 F i Y l e

o THERMAL STRATIFICATION FINITE ELEMENT MODEL Insulated f + x B l Es..- 3 g 4g T= 560 F <e so 36 / p' 6 sW Temperature Transition Zone , s. T= 160 F 7 n a e E 9 End Insulated s 3 8,C,C MAX STRESS INTENSITY = KSI ~ FOR TEMPERATURE DIFF. OF STRESSES AT OTHER TEMPERATUPES MAY BE OBTAINED BY PROPORTION

FATIGUE EVALUATION _ a.c.e l_ MAX DELTA T OF ., WAS USED e BASED ON MIN LEAK FLOW CONSIDERED MAX PEAK STRESS OCCURS AT e [ $lNCE MkX ,c,, DELTA T .' AND MAX K3 OCCUR AT THESE LOCATIONS ~ " ~ MAX PEAK STRESS = MAX ALT STRESS = [

• INCREMENTAL FATIGUE USAGE:

CHG, ALT CHG = 0.001 ALT SPRAY = 0.03 AT OTHER LOCATIONS WITHIN THE UNISOLABLE PIPING WHERE TURBULENT PENETRATION IS POSSIBLE, ALT STRESSES ARE BELOW FATIGUE ENDURANCE LIMIT

ANALYSIS CONSERVATISMS i RHX OUTLET TEMP = 477 F (MAX CHG), WHEREAS FOR NORMAL CONDITIONS TEMP = 530 F HEAT TRANSFER IN UNISOLABLE PIPE DOES NOT CONSIDER FLUID CONVECTION OR TURBULENCE, WHICH WOULD TEND TO REDUCE DELTA T HEAT TRANSFER IN THE ISOLABLE PIPE CONSIDERS ONLY INSULATION RESISTANCE AND NOT FLUID FILM COEFF, OR FREE CONVECTION TO ENVIRONMENT ENTIRETY OF UNISOLABLE PIPE IS AT 560 F

• COMMON HEADER PIPING LENGTH INCLUDED IN UNISOLABLE PIPE HEAT TRANSFER CALCULATIONS

i l IDENTIFIED INSERVICE INSPECTION LOCATIONS ..e.. FREQUENCY - 10 YRS.

I ,i I i } CHARGING, ALTERNATE CHARGING AND AUXILIARY SPRAY i ANALYSIS USING CURRENT i i TASCS METHODOLOGY f P t b + i i I i I

4 i I r CHARGING / ALT CHARGING LEAKAGE TEMP AT ENTRANCE TO UNISOLABLE PIPING ,c.. ? f t i r f m.mmuun b t

1 .l 5 t i SOUTH TEXAS CHARGING LEAKAGE HEATUP UNISOLABLE PIPE 560F ..e.. l i h f i i f i I i t m um -i i I i

i i i } l i l SOUTH TEXAS CHARGING LEAKAGE HEATUP i UNISOLABLE PIPE 560F t ,,c, 5 l l + t t i I i f i i l l

i -l l CHARGING AND ALTERNATE CHARGING LINE DETERMINATION OF Lm (TURBULENCE CYCLING ZONE) i i THE MAXIMUM DISTANCE FROM THE RCS AT WHICH TURBULENCE MAY INDUCE THERMAL CYCLING IS GIVEN BY: a.c.. I 2 l i ) N 4 m,

r CHARGING AND ALTERNATE CHARGING LINES i MAXIMUM STRATIFICATION DELTA TEMPERATURES

• L a.c..

I r i [ t l-I t i I + l h

CHARGING AND ALTERNATE CHARGING LINES SOUTH TEXAS STRATIFICATION DELTA T AT MAX CYCUNG LENGTH FROM RCS (Lm) i i m

l CHARGING AND ALTERNATE CHARGING LINES LEAKAGE TRANSIENT FATIGUE EVALUATION ) e MAXIMUM STRATIFICATION DELTA TEMPERATURE WITHIN THE CYCLING ZONE (Lm) IS 3,C,C CONTROLLING COMPONENT WITHIN THE CYCLING " Se i ZONE IS THE } i MAXIMUM ALTERNATING STRESS WITHIN THE CYCLING ZONE IS THEREFORE [ ["* THIS ALTERNATING STRESS IS BELOW ENDURANCE LIMIT STRESS 1 o AT LOCATIONS WITHIN THE UNISOLABLE PIPING BEYOND THE CYCLING ZONE (Lm), A MAXIMUM ALTERNATING STRESS IS __ "'S* e FATIGUE USAGE IS ESTIMATED AT 0.001 FOR THIS ALTERNATING STRESS i J-

A 2 ) i SOUTH TEXAS AUX SPRAY LEAKAGE TEMP AT ENTRANCE TO UNISOLABLE PIPING a.c.e P t i t J i 1 ] l 4

4 SOUTH TEXAS AUX SPRAY LEAKAGE HEATUP UNISOLABLE PIPE 560F l s.c.e m i s B l 4 4 e h i a 9 f 4 i; 4 + enum m l l l

1 1 1 ) SOUTH TEXAS AUX SPRAY LEAKAGE HEATUP 4 UNISOLABLE PlPE 560F l i ? t P t i f e F b I i i 4 I

AUXILIARY SPRAY LINE SOUTH TEXAS STRATIFICATION DELTA T AT ENTRANCE TO UNISOLABLE PIPE

• l i

j N m

4.a w ..r.e a m m a 'k i r AUXILIARY SPRAY LINE DETERMINATION OF Lm (TURBULENCE CYCLING ZONE) e,s e t t i I a I i 4 t t 7 i l f l I f i .k

~ .l AUXILIARY SPRAY LINE LEAKAGE TRANSIENT FATIGUE EVALUATION e MAXIMUM STRATIFICATION DELTA TEMPERATURE IS [ ["* ALTERNATING STRESS AT [ IT IS ASSUMED THAT CYCLING CAN OCCUR ANYWHERE WITHIN UNISOLABLE PIPING DUE TO MAIN SPRAY TURBULENCE \\ BASED ON DESIGN MAIN SPRAY LINE TRANSIENTS, TOTAL LEAKAGE TRANSIENT CYCLES ESTIMATED TO BE ] "* FATIGUE USAGE IS ESTIMATED AT 0.33. THIS ASSUMES CONSTANT LEAKAGE AT THE CRITICAL LEAK RATE [ ],' AND THAT ALL MAIN SPRAY LINE OPERATIONS RESULT IN TURBULENCE SUFFICIENT TO INDUCE CYCLING [

i i THERMAL STRIPING CHARGING AND ALTERNATE CHARGING: e FOR A STRATIFICATION DELTA TEMPERATURE QF i [A MAXIMUM PIPE WALL DELTA T OF WAS ~ CALCULATED FOR A STRIPING FREQUENCY OF i ~ ' CYCLES / HOUR.

• THE ALTERNATING STRESS LEVEL CORRESPONDING TO THIS DELTA T IS

- ~ KSI, WHICH IS BELOW j THE ENDURANCE Llhili. l AUXILIARY SPRAY:

• FOR A STRATIFICATION DELTA TEMPERATURE OF

['A MAXIMUM PIPE WALL DELTA T OF WAS l ~~ CALCULATED FOR A STRIPING FREQUENCY OF $ _ CYCLES / HOUR. e THE ALTERNATING STRESS LEVEL CORRESPONDING TO THIS DELTA T IS [ ~ KSI, WHICH IS BELOW THE ENDURANCE LIMIT. a

e 9 J + 4 o -4% e a ( t CHARGING LINE STRATIFICATION LOADING 4 3,C,C ? 1 I i f f h t L - f a t i i [ t 5 i I l i i p p e 1 i i i r I L ? e 2-31 i b r

e 4 L h 5 / 1' L /) i, g \\, k J{ ( t '~ i R .= bi 5 a,c,e E i ) i I ) 4 i i l I

o a,c,e c FIGURE 3-1: CHARGING STRESS INTENSITY 3-7 ..}}