ST-HL-AE-5418, Forwards from Westinghouse Responding to NRC Concerns of RHR Removal Lines.Resolution to NRC Bulletin 88-008, Thermal Stresses in Piping Connected to RCS Provided

Forwards from Westinghouse Responding to NRC Concerns of RHR Removal Lines.Resolution to NRC Bulletin 88-008, Thermal Stresses in Piping Connected to RCS Provided

ML20115H495
Person / Time
Site:	South Texas
Issue date:	07/15/1996
From:	Thomas S HOUSTON LIGHTING & POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
IEB-88-008, IEB-88-8, ST-HL-AE-5418, NUDOCS 9607230153
Download: ML20115H495 (20)

Text

.

The Light company South Texas Project Electric Generating Station P.O. Box 289 Wadsworth. Texas 77483 Houston IJghting & Power July 15, 1996 ST-HL-AE-5418 File No.: G03.03 10CFR50 U. S. Nuclear Regulatory Commission Attention: Document ControlDesk Washington, DC 20555-0001 South Texas Project Units 1 and 2 Docket Nos. STN 50-498, STN 50-499 Resolution of NRC Bulletin 88-08,

Diermal Stresses infiping Connected to Reactor Coolant Systems" l

l

References:

1. Ixtter from T. W. Alexion, NRC, to W. T.Cottle, Houston Lighting i

& Power, " Resolution of NRC Bulletin 88-08, ' Thermal Stresses in Piping <

Connected to Reactor Coolant Systems", South Texas Project, Units 1 and 2 (STP) (TAC NOS. M93822 and M93823), dated Febmary 23,1996 (ST-AE-HI 94449)

2. Ixtter from T. H. Cloninger, Houston Lighting & Power, to the NRC 1 Document Control Desk, dated November 30,1993 (ST-HI-AE-4639)

3. WCAP-12598, Supplement 1, "NRC Bulletin 88-08 Evaluation of Auxthary j Piping for South Texas Project Electric Generating Station Units 1 & 2"

4. Electric Power Research Institute report TR-103581, ' Thermal Stratification, Cycling, and Striping (TASCS)", dated March 1994

5. Ixtter from G. F. Dick, Jr., NRC, to D. P. Hall, Houston Lighting & j Power, dated September 23,1992 (ST-AE-HI 93210)

6. letter for M. A. McBurnett, Houston Lighting & Power, to the NRC ,

Document Control Desk, dated January 27,1989 (ST-HI AE-2950) l Reference I discussed the conclusion by the NRC Staff that the South Texas Project has not provided I f

the requested assurance of Action 3 of NRC Bulletin 88-08 against inadvertent leakage in the unisolable segments of the charging lines and auxthary pmssuri:rr spray systems. In Reference 2, the J

g South Texas Project stated a belief that all Bulletin 88-08 actions had been satisfied based on Reference

3. The conclusions of Reference 3 are based on aspects of an analytical methodology developed by the Westinghouse Electric Corporation under a program sponsored by the Electric Power Reseamh Institute (EPRI) to investigate Thermal Stratification, Cycling, and Striping (TASCS), Reference 4.

9607230153 960715 PDR ADOCK 05000498 G PDR MISC-96\5418

p Houston Ughting & Power Cosapany -

South Texas Profeet Electric Generating Station ,

ST-HI-AE-5418  :

File No.: G03.03 Page 2 In Reference 1, the NRC discussed several concerns regarding the South Texas Project submittal to address Bulletin 88-08 and the EPRI TASCS report. Attached is a letter from Westinghouse to the

' South Texas Project responding to the NRC concerns in Reference 1.

l The South Texas Project concurs with the conclusions of the attached Westinghouse letter, and l- believes References 3 and 4 provide the basis for concluding that unisolable sections of the South Texas Pmject piping for charging lines and auxihary pressurizer spray systems will not be subjected to combined cyclic and static thermal and other stresses that could cause fatigue failure during the mmaining life of the plant. Funher programmatic controls requested by Action 3 of Bulletin are not considend necessary.

The NRC safety evaluation in Reference 1 did not specifically addmss the South Texas Project residual heat removal lines. However, the NRC expressed concern regarding these lines in Reference 5. The i South Texas Project stated in Reference 6 its belief that the Residual Heat Removal System design is adequate and no funher action was planned.

The South Texas Project requests a meeting with the NRC staff to resolve any differences regarding the technical issues supponing resolution of NRC Bulletin 88-08.  !

If you have any questions regarding this letter, please contact Mr. A. W. Hamson at (512) 972-7298 or me at (512) 972-7162. j S. E. Thomas Manager, I Design Engineering i

KJT/kjt

Attachment:

Ietter from M. A. Sinwell, Westinghouse Electric Corporation, to W. T. Cottle, Houston Lighting & Power, " Response to South Texas Safety Evaluation for NRCB 88-08", dated June 19,1996 (ST-WN-HS-960040)

MISC-96\5418

4 Houston Lighting & Power Company ST-HL-AE-5418 South Texas Project Electric Generating Station File No.: G03.03 Page 3 c:

• Leonard J. Callan Rufus S. Scott Regional Administrator, Region IV Associate General Counsel U. S. Nuclear Regulatory Commission Houston Lighting & Power Company 611 Ryan Plaza Drive, Suite 400 P. O. Box 61067 Arlington, TX 76011-8064 Houston, TX 77208

• Thomas W. Alexion Institute of Nuclear Power Project Manager , Mail Code: 13H3 Operations - Records Center U. S. Nuclear Regulatory Commission 700 Galleria Parkway Washington, DC 20555-0001 Atlanta, GA 30339-5957 David P. Loveless Dr. Bertram Wolfe Sr. Resident Inspector .

15453 Via Vaquero c/o U. S. Nuclear Regulatory Comm. Monte Sereno, CA 95030 P. O. Box 910 Bay City, TX 77404-0910 Richard A. Ratliff Bureau of Radiation Control J. R. Newman, Esquire Texas Department of Health Morgan, Lewis & Bockius 1100 West 49th Street 1800 M Street, N.W. Austin, TX 78756-3189 Washington, DC 20036-5869 M. T. Hardt/W. C. Gunst J. R. Egan, Esquire City Public Service Egan & Associates, P.C.

P. O. Box 1771 2300 N Street, N.W.

San Antonio, TX 78296 Washington, D.C.

J. C. Lanier/M. B. Lee J. W. Beck City of Austin Little Harbor Consultants, Inc.

Electric Utility Department 44 Nichols Road 721 Barton Springs Road Cohassett, MA 02025-1166 Austin, TX 78704 .

Central Power and Light Company ATTN: G. E. Vaughn./C. A, Johnson P. O. Box 289, Mail Code: N5012 Wadsworth, TX 77483

• Attachment to this addressee only

i (O w \

I ST-WN-HS-960040 Westinghouse Energy Systems h 355 Pmsburgh Pennsylvania 15230 0355 Electric Corporation June 19, 1996 Mr. W. T. Cottle Ref. 1: MSE-SMT-96-112

. Houston Lighting & Power Company South Texas Project Electric Generating Station P.O. Box 289 Wadsworth, Texas 77483 Attention: Ray Pate, Ed Halpin and Sam Patel HOUSTON LIGHTING & POWER COMPANY SOUTH TEXAS PROJECT ELECTRIC GENERATING STATION Response to South Texas Safety Evaluation for NRCB 88-08

Dear Mr. Cottle:

On February 23, 1996, the NRC staff issued a letter entitled " Resolution of Bulletin 88-08, ' Thermal Stresses in Piping Connected to Reactor Coolant Systems,' South Texas Project, Units 1 and 2 (STP) (TAC Nos. M93822 and M93823),"

T. W. Alexion to W. T. Cottle (HL&P). This letter includes a Safety Evaluation (SE) which discusses several NRC concerns regarding the HL&P South Texas Project Units 1 and 2 submittal regarding NRC Bulletin 88-08, and the EPRI Thermal Stratification, Cycling, and Striping (TASCS) report. The purpose of this letter is to address these NRC concerns.

If you have questions please contact me, P. L. Strauch, or D. H. Roarty.

Very truly yours,

%ON ry Ann Sinwell South Texas Project Manager

/slf Attachment cc: Ray Pate Ed Halpin Sam Patel-STP RMS S. D. Phillips-P. Blondo

- Dr. John H. Kim -

sT-WN HSeenoso/06twa6/1

l l- RESPONSE TO NRC SAFETY EVALUATION OF WCAP-12598, SUPPLEMENT 1 l*

AND EPRI REPORT TR-103581 l

Background

United States Nuclear Regulatory Commission (NRC) Bulletin 88-08 (Reference 1) was issued in June 1988 as a result of a pipe cracking incident at Farley Unit 2, and subsequent evaluations which confirmed that valve leakage caused the failure. The purpose of the bulletin was to request that licensees 1) review systems connected to the RCS for susceptibility to adverse stresses resulting from valve leakage,2) nondestructively examine susceptible piping to ensure that there are no existing flaws, and 3) plan and implement a program to provide continuing assurance that adverse stresses from valve leakage will not result in failure during the remaining life of the unit.

Supplements 1,2 and 3 of the bulletin were issued to emphasize nondestructive examination requirements and to alert utilities that periodic valve seat leakage through packing glands could result in unacceptable thermal stresses.

Houston Lighting and Power Company (HlAP) has been very responsive to NRC Bulletin 88-08 requests for South Texas Project Units 1 and 2. Pursuant to Action 1, a systems review performed in 1988 determined that the normal charging, alternate charging and auxiliary spray piping are susceptible to adverse stresses from valve leakage (Reference 2). Pursuant to Action 2 susceptible piping at both

! units was nondestructively examined, and no flaws were identified (References 3 and 4). Pursuant to Action 3, susceptible piping at Unit I was instrumented to monihr temperatures in the unisolable sections of piping connected to the RCS (Reference 4), and a pugram was established to also monitor l Unit 2 (Reference 3).

Subsequent to the installation of the instrumentation, an engineering evaluation performed by Westinghouse for the Units 1 and 2 normal charging, alternate charging and auxiliary spray piping I

wu submitted to the NRC (Reference 5). The conclusion of this evaluation was that the normal charging, alternate charging and auxiliary spray line piping integrity would not be jeopardized, should l

inleakage into the RCS occur over the life of the units. Subsequently, HL&P informed the NRC staff l

that the temperature monitoring on these lines would be removed (Reference 6). The NRC staff reviewed the Reference 5 evaluation, identified several technical issues, and concluded that the decision to remove the temperature monitoring instrumentation may have been premature (Reference 7).

A meeting between NRC, Electric Power Research Institute (EPRI) HlAP, and Westinghouse personnel was held in November 1993 for the purpose of discussing and resolving the technical issues identified by the NRC in Reference 7, and the utility position regarding removal of the temperature monitoring equipment. In addition, developments from the EPRI Thermal Stratification, Cycling and Striping (TASCS) Program were presented by EPRI and Westinghouse personnel. As a result of the meeting, the l NRC requested that a supplemental evaluation for the normal and alternate charging lines be performed, assuming a more limiting loading than would be predicted by the TASCS methodology, since these methods had not yet been reviewed by the NRC. The purpose of this supplemental evaluation (Reference 8) was to provide additional assurance of piping integrity for the normal and alternale charging piping for near term operation, assuming a worst case scenario of valve leakage and cycling. This evaluation was not intended to supersede the original evaluation and its conclusions. The NRC accepted this interim response, and stated that they would continue to review the issue and possibly provide a plant specific safety evaluation regarding long-term operation (Reference 9).

r l

1

l .'

in September 1994. EPHI provided the TASCS report (Reference 10) to the NRC for evaluation. The NRC has evaluated the South Tens analysis and the information provided in the TASCS report, as documented in Reference 11. On the basis of this evaluation, the NRC concluded that HlAP has not adequately justified discontinuing 9mperature monitoring, nor provided an acceptable alternative to monitoring. On this basis, the NRC tencluded that HlAP has not provided the requested assurance of f

Action 3 of the Bulletin against inadvencnt leakage in the unisolable sections cf the charging system lines and should therefore reestablish the pevious temperature monitoring program at both units, or implement other acceptable monitoring prograins that satisfy the provisions of Action 3 of the l Bulletin, for the life of the plant.

The Reference 11 safety evaluation contains several NRC concerns regarding the HlAP submittal (Reference 8), and the EPHI TASCS report (Reference 10). These concerns are addressed below.

l lt should be noted that engineering evaluations were also performed for the Units 1 and 2 auxiliary spray lines (Reference 5), and the residual heat removal lines (Reference 12) pursuant to Supplement 3 of the Bulletin. These evaluations demonstrated that the likelihood of failure due to isolation valve leakage is extremely low. The NRC position regarding these lines is not clear to HlkP, since a supplemental calculation was not requested for the auxiliary spray lines (as was requested for the normal and alternate charging lines), and the residual heat removal lines are not specifically addressed in the Reference 11 safety evaluation.

Response to NRC Evaluation items Evaluation item 1:

The South Texas thermalload rakuhlkns kr the unhobbleporthns of the chartmg andallenale chartmghhes are based on the temperature difference cakuhledat the marknum turbulent penetrabon dhiance. The South Texas check Iake oulkis are healedala dhiance from the EG greater than the turbukn/penetralkn dk/ance, and nH/ therekte not e>perknee thermalc)r/wg At otherhcolkns nhere cychhgkducedbyEG /urbulence hpossible the thermalstresses were determhed to be bekw (he endurance hmH, because the temperature diT/erence decreases as the dhiance /o the EG hreduced Thh doesnot correspond to the kHures alParley and7thange. In thesephnis the check ake and the Msi e/bor were both heated nHhm the cakukled turbulent penetrathn length, but the faHures occurred at the &st e/bor velds andm the elbow base metal

/e., at heahons nhere the temperature diT/erence approaches zero, basedon steadystate heal transfer cakuhtions of the unhobble segments as desenbedh Secthn M 7kirbuknlPenetra/hn ThermalC)thhg" ofthe RSCSreport. The RSCSmethodohg): basedon steadystate heal transkt andstress anal)sh cakuhlkm; therefore does not appear to idenhTy the correct heathns and the

/mie g>an nhere a h/gue kHure h most hkely to occut; andapparently does not account hr other thermo-hydraulkphenomena uhkh exh/ al these heathns andmay also cause s&m7kant thermal ejr/hg; Response to Evaluation item 1:

This item presents an argument that the TASCS methods would not predict the failures at Farley and Tihange and therefore would not be acceptable for evaluating the South Texas charging lines.

j The TASCS methodology conservatively predicts high thermal loads for the Parley and Tihange cases.

2

i  !

l . .

l . A summary of the application of the methodology folloos.

I ApplicatioILotthe_TASCSlethodologLlo_thelarley/Tihange Safety Iniection Lines a) The nominal fluid temperaiure of the unisolable piping is estimated to be equal to the RCS cold leg temperature of approximately 550*F, since the pipes are routed upward from the cold leg and terminate in a short horizontal run at the check valve. This estimate is a result of the conservative i approach presented in the screening criteria for a " Leakage Case", on page 2.0-12, item c-2 of Reference 10.

b) For the estimated leakage rate of 0.5 to 1.0 gpm (at Farley, based on a flow diversion test), the leakage entering the unisolable pipe is at the containment ambient temperature (approximately 100*F). Preheating of the leakage flow upstream of the check valve is negligible, considering the higher leakage flow rate, and the fact that convective currents from the cold leg can not extend into the upstream piping due to the check valve. This estimate is a result of the conservative approach

. presented in the screening criteria for a " Leakage Case", on page 2.0-12 item c-1 of Reference 10.

c) The potential for stratification (Richardson number, Section 3.1 of Reference 10) is evaluated and found to occur for the farley/Tihange case. This case is therefore considered as having a potential for a TASCS issue.

d) The location where thermal cycling may occur is then determined. Section 3.3 of Reference 10 discusses thermal cycling resulting from turbulent penetration The length of piping susceptible to turbulent penetration thermal cycling (L.) for the Farley/Tihange case, as determined using equation 3.3-1 of Reference 10, is about 12 pipe inside diameters (5 feet from the cold leg inside diameter). 1 This indicates that all of the unisolable piping (from the cold leg to the check valve) has sufficient turbulent energy to cause thermal cycling. Leakage therefore enters the unisolable piping within the turbulent penetration thermal cycling zone, i.e., the check valve outlet is located within the distance 1,from the cold leg.

I The farley monitoring data clearly indicated a cycling mechanism at about 6 pipe inside diameters l from the cold leg (Figure 5.3-2 of Reference 10). Cracking occurred at approximately 5 pipe inside i diameters from the cold leg, which is within the zone of thermal cycling. Very little cycling probably i occurred at 8 pipe inside diameters from the cold leg, the location of the check valve to pipe weld, where no crack indications were found. The calculated 1%therefore bounds the observed data.

e) The leakage heats up a negligible amount (<5'F) over the length of horizontal pipe from the check valve to the cold leg. Therefore, the temperature difference between the ambient pipe fluid and the leak flow at Farley/Tihange is estimated to be 450'F for the parameters discussed above. This ,

temperature difference is independent of leakage flow rate. The temperature difference between the ambient pipe fluid and the leak flow at Parley was estimated to be 335'F. Therefore, Reference 10 predicts a conservative stratification temperature difference.

f) The number of thermal cycles (or frequency of cycling) to apply in a fatigue analysis is discussed I

in Section 3.3, Step 3 of Reference 10. This approach is conservative and results in a frequency of me cycle per minute (or 525,600 cycles / year). A case specific heat transfer analysis could also be l

r. red to justify a lower frequency. The Parley monitoring indicated that a cycle occurred about every

! 4 minutes. (Cycles with larger temperature fluctuations occurred about every 13 minutes, and cycles with smaller temperature fluctuations occurred about every 6 minutes).

g) llsing the monitoring data results in a fatigue and fatigue crack growth analysis for Farley resulted 3

i

L l.

.~ in a prediction of failure of 3 to 6 calendar years, as documented in Reference 13. The plant had been in operation for about 6.5 years, therefore the prediction of failure was conservative with

!' respect to the actual plant age. Using the methods of Reference 10 would have resulted in an even l

more limiting time for failure, since a higher stratification temperature difference and a higher l

frequency of thermal cycling would have been used.

Applicationdthe IASCSJethodnlogy to the South Texas Charpine Lines There are significant differences between the Farley/Tihange cases and the South Texas charging system lines that justify that thermal stratification, cycling and striping are not a significant fatigue concern for the South Texas Units 1 and 2 charging system lines. These are discussed below.  ;

a) For the normal and alternate charging lines, the unisolable piping extends from the cold legs to the outlet nozzles of check valves XCV0001 and XCV0004, respectively. The total length of unisolable i piping is 82 inches (24 pipe inside diameters) for the normal charging line and 73 inches (21 pipe 1 insider diameters) for the alternate charging line. The pipe inside diameter is 3.438 inches. These unisolable portions of these lines are routed horizontally from the cold leg, although the normal charging line has an intermediate riser extending downward about 15 inches. The nominal fluid temperature of the unisolable piping was estimated to be equal to the RCS cold leg temperature of approximately 560'F. This estimate is a result of the conservative approach presented in the

screening criteria for a " Leakage Case", on page 2.0-12, item c-2 of Reference 10.

b) The charging lines have a hot leakage source, since leakap must pass through the regenerative heat exchanger before reaching the unisolable piping. (A mimmum regenerative heat exchanger i outlet temperature of 477'F was used in the South Texas analysis, based on maximum charging flow 4 conditions. This is conservative since under ordinary charging flow conditions, this temperature is

! 530*F). The leakage flow temperature at the entrance to the normal and alternate charging l unisolable piping (check valve XCV0001 and XCV0004' outlets, respectively) is dependant on the leakage  !'

flow rate, since the leakage source is hot. Lower leakage flow rates approach the limit of ambient containment temperature of approximately 100'F. Higher leakage flow rates approach the  ;

! regenerative heat exchanger temperature of approximately 477'F. (This is a significant difference i from the Farley/Tihange case, since the leakage flow temperature at the entrance to the unisolable piping is at containment ambient temperature, and is independent oi leakage flow rate). An example i application of the methods used in the charging line analysis is presented in Reference 10, Section 3.9 )

(Example 1).

c) The potential for stratification was evaluated and found to occur for the South Texas charging system lines. Thercfore, additional analysis was performed.

l d) The length of unisolable piping susceptible to turbulent penetration thermal cycling (6) for the South Texas charging lines is 11 pipe inside diameters, or about 38 inches. (This was conservatively rounded to 40 inches in the analysis). The total length of unisolable piping is 62 inches for the normal charging line and 73 inches for the alternate charging line. These unisolable piping lengths are longer than (, therefore 42 inches of unisolable piping extending from the normal charging line L check valve and 33 inches of unisolable piping extending from the alternate charging line check valve are not susceptible to thermal cycling. Monitoring data from the South Texas Unit 2 alternate charging line confirmed that thermal cycling does not occur at 55 inches (16 pipe inside diameters) from the cold leg connection. This is discussed in more detail later, e) Postulated leakage would enter the unisolable piping well beyond the thermal cycling zone defined 4

. ' by 1,% therefore heating of the leakage floa could occur uithin the unisolable piping, as discussed in

. Reference 10. Section 3.2. Thermal cycling was conservatively assumed to occur at%1. Thermal cyclinc is also possible at locations closer to the cold leg connection, but the stratification temperature differential would be lower than at 1. % due to additional heating of the leakage flow from the unisolable piping.

As discussed in item b above, lower leakage flow rates enter the unisolable piping at essentially the  :

containment ambient temperature, but heat up significantly in the unisolable piping due to their  !

- smaller cross sectional flow area and lower flow velocity. Higher leakage flow rates enter the unisolable piping at essentially the regenerative heat exchanger temperature, but do not significantly j heat up in the unisolable piping, due to their larger cross sectional flow area and higher flow velocity. 1 Therefore, a worst case leakbge flow rate, yielding the highest possible stratification temperature i differential at 1,% was determined for the charging lines. i f)In the Reference 5 evaluation, the number of fatigue cycles applied at the check valve outlet welds was 1200 (resulting from 200 heatup and cooldown cycks, plus 1000 miscellaneous events). At other locations within the unisolable piping, where turbulent penetration thermal cycling is possible, the alternating stresses were determined to be below the endurance limit stress based on 10" cycles, g) The Reference 5 evaluation resulted in negligible fatigue usage for isolation valve leakage into the charging system lines.

Remarks on South Texas Temperature Monitoring The. Unit 2 normal charging line monitoring location was about 9 pipe inside diameters from the cold leg. This line was in operation during the period reviewed, except for a maintenance period when 'il .

was secured. The monitoring data did not indicate any significant thermal activity other than design transients.

The Unit 2 alternate charging line monitoring location was approximately 16 pipe inside diameters from the cold leg. This line was not in operation during the period reviewed. However, it was discovered that a lift check valve was installed instead of a spring loaded check valve (CV0007)in a bypass line around isolation valve XCV0006. (The spring loaded check valve was designed to open at a set differential pressure to relieve excessive pressure that could potentially occur when the normal and alternate charging line isolation valves XCV0003 and XCV0006, respectively, are closed). The lift check valve admitted a constant bypass flow around isolation valve XCV0006 and into the cold leg.  ;

16nitoring data from the alternate charging line indicated a noneyclic stratified load during normal I power operation due to this bypass flow, as shown in the alternate charging line monitoring data plots in WCAP-12598. Supplement 1. (Note that the monitoring location was beyond the thermal ,

- cycling zone of 11 pipe inside diameters defined by 1,. The noneyclic monitoring data therefore  !

substantiates the turbulent penetration thermal cycling methodology, and also confirms that cyclmg  ;

will not occur at the check valve outlet welds, which are located much further from the cold leg connection).' The stratification AT was about 40'F. except during a maintenance period when it reached about 140*f. The lift check valve has since been replaced with the correct spring loaded check valve.

I The total length of auxiliary spray piping from the main' spray line to check valve CV0009 outlet

- nozzle is approximately 18 inches for Unit 1 and 21 inches for Unit 2. The Unit 2 monitoring location i 5

l

l

. cas at the lee connection to the main spray line, Monitoring data from the auxiliary spray line l indicated constant stratification ATs of approximately 20*F to 40'F. which is less than the acceptable value of 50*F.

l 1

1 1

i The residual heat removal suction lines at South Texas have a short length of piping (approximately 4.5 pipe inside diameters) extending downward from the hot leg to a horizontal section of piping which includes the isolation valve. An evaluation of the South Texas residual heat removal lines concluded that there are significant differences (Reference 12) between the South Texas lines and the cracked residual heat removal line (Reference 1) that would preclude stratification and cycling from ,

occurring at South Texas. An evaluation of stratification was performed, nonetheless, which resulted 1 in negligible fatigue usage for isolation valve leakage in these lines. Using current TASCS methods,  !

turbulent penetration (im,, Section 3.8 of Reference 10) and free convection heat transfer (Section 3.6 l of Reference 10) will heat all of the unisolable piping. (Temperature measurements taken near the l isolation valves confirmed this). Therefore, outleakage through the isolation valve will not result in adverse stresses in the unisolable pipe, since the temperature difference between the leakage flow  :

and the fluid in the unisolable section of the residual heat removal suction piping would be small.  ;

I Evaluation item 1 Conclusions The TASCS methodology, when correctly applied to the Parley or Tihange cases, predicted fatigue cracking.

4 There are significant differences between the Farley/Tihange cases and the South Texas charging system lines that justify that thermal stratification, cycling and striping are not a significant fatigue concern for the South Texas Units 1 and 2 charging system lines. Temperature monitoring data from South Texas Unit 2 and the absence of cracks in the check valve outlet weld at Farley Unit 2 i substantiate the turbulent penetration thermal cycling methodology used in the analysis of the South l Texas charging system lines. The South Texas monitoring data also confirms that cycling will not  !

occur at the normal and alternate charging line check valve outlet welds.  ;

Evaluation item 2:

a)The Low Temperature Turbulent Penetrabbn Testprogram, describeda Sechbn // of the TASCS report, nas conducted pHb constanth/eakage, and therefore assumedthhh/eakage to be the same osleakage through a sehr check va/re. No verfkahon has beenprovidedthat thh assumpfhn h mhdundera//How condthbns l . b)Furthermore, these fests wereperformedof almospherkpressure androom temperature.

c}The relucHy component uhkh wasmeasured& un.goecKkd andthe corre/abbn between the relocHy and the temperature m the branch hne h unclear d)In addihbn the data h thhsechbn doesnot correspond to lhalpresentedb Reference Il 6 \

i

}

e) C)thhg ves noted to occur beloven the kner and the upper hieraclkn reghns The nature of thh c)thng h unckar, and the frequene)' of thh c)thhg appears not to hare been measured or recorded '

r i i Response to Evaluation item 21 This concern addresses selected portions of the test program without pointing to the specific TASCS method being questioned. Each TASCS method, as summarized in Section 3 and explained in detail in Section 5 of Reference 10,' uses a number of tests (both low and high temperature, where applicable), :

as well as theoretical analysis in its development. The " method"is then developed using a conservative interpretation of the data, with added margin. The specific methodology related to the reviewers questions is " Turbulent Penetration Thermal Cycling" Section 3.3. This methodology was  ;

developed starting from a theoretical analysis and bench-marked using low temperature testing and j high temperature testing. The method resulted in predictions which bounded all test data and which  !

supports the data obtained for the Farley case.

Responses to specific concerns of Evaluation Item 2 follow.

a) A swing check valve will not modulate flow rate due to the low pressure drop across a check valve. ,

b) The test data from low temperature, low pressure tests were nondimensionalized to estimate -

results at RCS conditions.

c) Turbulence from header pipes has two significant effects on connected branch line piping. It will cause the branch line piping near the header connection to be at approximately the header pipe temperature. The extent of this piping is called the turbulent penetration length (1,), which is a function of the header pipe fluid velocity and kinematic viscosity, and the branch pipe inside diameter, as shown in Equation 5.4-2 in Section 5.4 of Reference 10. The second effect is that turbulence will cause stratified flows to mix, which can result in turbulent penetration thermal cycling. The extent of piping which may experience turbulent penetration thermal cycling is predicted by 1%(Equation 5.3-8 of Reference 10), its derivation was based on the energy of the turbulence and the stratified flow, as discussed in Section 5.3 of Reference 10.

In the 1.ow Temperature Turbulent Penetration Test program, velocities were measured in the axial  !

and tangential direction; only the axial velocity was significant. The velocity data is normalized to the velocity in the main header pipe. Temperature is accounted for by considering the density in the kinetic energy and total energy (Equations 5.3-4 and 5.3-0 of Reference 10), which were used in the development of %1, the upper limit of turbulent penetration thermal cycling.

d) The conclusions for Reference 14 were preliminary since they were based on preEminary test results from the EPRI TASCS program. Reference 10 uses more substantial conclusions based on final test results, which includes additional testing.

l e) The purpose of the test was to determine the location at which cycling may occur, not the frequency of the cycling. Reference 10 discusses the nature of cycling in terms of energy in Section 5.3. It also discusses frequency and why it is not quantified.

. 7

Evaluation item 3:

Theflowlemperature tests (#17)h Sechon U andthef})gh temperature lesis(HHT)m Sechon M bo/h of the TASCSnport. do nol reRect the thermaloperalms condhons shkh led to BuHelm 08-08 These condhons nere the m/eakage ofcold nalerpast a snhg check >ake blo an u/hnm/e source ofDuclualms hghpressure, hgh Row forbulent holRud nhereas these lesis assumed constant m/ealate and cul/eakage mio stagnant coldandhot Hales; respechte/r Response to Evaluation item 3:

3 The Low Temperature Stratification Tests were performed to provide a basis for interpretation of the high temperature-high pressure test results and to provide an independent basis for checking the stratification theory. The hw temperature stratification tests were performed at room temperature and atmospheric pressure, and utilized ordinary water and water doped with calcium chloride to ,

simulate stratified flow. The effect of header pipe turbulence on the stratified flow was not the goal of this testing; this effect was studied in the Low Temperature Turbulent Penetration Testing. ,

The High Temperature Stratification Tests were performed to obtain high temperature-high pressure thermal stratification data under heat transfer conditions to verify the relationships among the flow rate, layer thickness and Richardson number, which in turn was used to verify the scaling rationale of the low temperature testing. The effect of header pipe turbulence on the stratified flow was not the goal of this testing; this effect was studied in the Low Temperature Turbulent Penetration Testing.

These tests reflect the more general condition of cold leakage into a hot pipe, typical of NRC Bulletin 88-08 conditions, used as inputs for the stratification heat transfer model. i Evaluation item 4: i Test 9 of the HHTles/srepresents hilermH/ent coldm/eakage mio stagnant hol nafer Agure 43-57 of the TASCSte  :

h Fgure /3-6) port shons nasgreater that the than elStahon thermalc)chhg i ahkh a the h ahere the leakage Huid entered theat Slabbn les/ sechon. Hof the fest  :

LAenke, the mside and outside naH lemperatures at Stahon H exhibH much greater cychng than those atStationi No dhcusshn or exp/atalkn ofthh behanorhas beenf iren.

Response to Evaluation item 4:

The thermal cycles that occur at Stations I and 11 are a result of the variation of the injection flow rate through time (Figure 4.3-55) The data illustrates that when the injection flow is terminated, the fluid and pipe metal at Station 11 heat up more quickly than at Station 1. Since the depth of the stratified injection flow decreases as the flow approaches the exit point of the test section (the elbow downstream of Station 11), the injection flow and pipe metal heat up more quickly (at Station 11) when the flow is terminated, due to the thinner stratified layer. This is not a finding which was relevant to the methods developed, therefore it was not discussed in the TASCS report.

Evaluation item 5:

The HHTlesis also do not reflect the condhons b Supplement 3 ofBuHehh BB-06 smce no m/ealage  ;

lesisfrom a hot, larbulentsource wereperformed Theyalso do notreRect the condhons desenbed  ;

h BuHelm 08-11shee no hol safer h/ealage les/s wereperformed l

. 8

l Response to Evaluation item 5:

The cracking described in Supplement 3 of Bulletin 88-08 involved hot outleakage from the RCS l through an isolation valve which is located in a " cold trapped" section of piping. The isolation valve l is located approximately 29 pipe inside diameters from the RCS connection, sufficiently far from the l influence of RCS turbulence. Since the South Texas charging system has a higher operating pressure l

than the RCS, outleakage from the RCS can not occur Therefore, this item is not relevant to the South Texas charging system lines. I j The surge line stratification issue (Bulletin 88-11) involves flows into and out of the pressurizer.

l' through the pressurizer surge line, which connects the pressurizer with the RCS hot leg. This line does not contain any valves. This is a separate issue which has been thoroughly investigated and resolved.

Evaluation item 6:

i 7n Chapter 53of the RSCSreport. ThermalCychng - Backroundand ientkalkn" Ahstated that

."te lesis wereperformed under condthbnsshnHar to those exhhhg alFarley Ieryhille data on thh i

testprogram has beenpresentedm the reports Figure 53-8of the RSCSreport shon the temperature-thne hblones measuredal ranbuslocabbns along the hollom ofthe msh'e surface ofa lest conhgurathn shnHar lo the safelymjechbn hhe alParley The correspondmg temperature-hme '

hhlories on the ou&ide surface are not shom figure 53-2 ofthe RSCSreport show the l lemperature-/hne hhlones measured around the chcumference of the ouhidepipe surface al i Farley hb correlabbn E thereforepossible between the lest data and the farley data. " j Response to Evaluation item 6:

l I

l The test data utilized fluid temperatures and the farley data utilized outside wall temperatures to l arrive at similar conclusions regarding where cycling does and does not occur. A correlation relating ,

l the test fluid temperature with the farley outside pipe temperature was not necessary to determine j where thermal cycling occurs with respect to distance from the main coolant pipe. Rather, locations j where temperature fluctuations did and did not occur from the test data and the farley data, were  !

used in the development of the turbulent penetration thermal cycling methodology.

The test program was specifically designed to simulate the Farley case. This included matching the j safety injection pipe dimensions, the valve type, the leakage source, and the nominal RCS flow rate. -

l pressure and temperature. The data obtained from the test shows that cycling occurs ~in the fluid at  !

7.8 pipe inside diameters from the main coolant pipe connection for the estimated Parley leak rate. i l No cycling is seen on either side of the check valve disc.

l l The Farley data (Figure 5.3-2) shows that cycling occurred on the outside surface of the pipe at l about 6 pipe inside diameters from the main coolant pipe connection. Cycling probably did not occur i' l at 8 pipe inside diameters from the main coolant pipe connection, since no crack indications were found in the check valve to pipe weld.

! The test data and the Farley data therefore provide similar results regarding where cycling does and does not occur. A correlation between the test data and the Farley data is not necessary, and would

, not change the turbulent penetration thermal cycling methodology developed in the TASCS report.

This item therefore has no impact on the South Texas Project evaluation.

i 1

9 i

Evaluation item.7t l'

. /&ure 53-2ofthe RSCSreport represents a sma// segment ofthe orathble data from Farley Erlensive temperature hhlorks nere measured on the outsMe surfaces offra safely hjechbn hnes 1

h thhphnt, both nih and vilhouth/eskoge, and upstream and downstream of the schg check  ;

rs/res; //hnot apparenthos: orK these dela were usedh the development ofthe RSCS 1

/methodobgy Response to Evaluation item 7:

The farley cracking incident was evaluated using thermal loadir,g cases reflecting the various magnitudes of outside wall temperature fluctuations as shown in Figure 5.3-P, of the TASCS report.

This figure best indicated the thermal transient activity at Farley, therefore othcr Farley data was not provided in the TASCS report. The recorded temperature profiles of the Parley Unit 2, loops B and C safety injection piping, on the upstream and downstream sides of the check valves,is documented in Apendix A of Reference 13.

The Parley data, as well as high and low temperature testing, were used to evaluate the turbulent penetration thermal cycling methodology provided in the TASCS report, as discussed in Evaluation item 6. As summarized in the response to Evaluation item 1, this was successful. Note that the

. methodology was not intended to be able to simulate an' actual plant: rather it is intended to be used to generate a conservative' set of thermal loadings.

Evaluation item 8: l

' Equation 52-5 of Chapter 52of the TASCSreport, 'Strah7kation Heat Transfer',' h based on slesdy stateBor condition.s: nhkh do notreDect actualtransknl temperature condHhnskpipes with h/eakage. Thh can be seen from thegoodcorrektion ofcakuhledresults nth the #7/Ttestresun.s:

andthepoor correhlkn ofthe cakuhtedresu//s andthe lestsshm.hths the Parley crackht '

heident. "

Response to Evaluation item 8:

The purpose of the WHT tests was to determine the heat transfer film coefficients for application to

! stratified flows. A constant flow rate was used for individual tests, but various flow rates were L studied in the test program. The method established determines how fast a small stratified flow (cold) would heat up in a hotter ambient fluid. To provide a conservative prediction of leakage

! ' heatup, the method estimates the minimal heatup rate. A minimal healup rate maximizes the temperature difference between the leakage and ambient fluids, which maximizes thermal stress. The l

WHT test conditions minimized heat transfer by minimizing the mixing between the leakage and

~

ambient fluids. The poor correlation was a result 'of the additional heat transfer resulting from mixing. (Note that this same mixing location.was used to verify the thermal cycling location).

4

~

Evaluation llem 9: -

10

F The bash for BuHelk 08-08 nas the faHure ofa safely mjechbn hne of Ferler Uml 8 due to .

madrerlent h/ealage. /Renke. /he hashforSupplement 1 of the BuHelm nas the faHure ofa t safelymjechbn hhe of Tihange. 17te ther of thh hne up to the f#st elbowh mehnedal 30 to the hoikonfalphne. Leahhg cracls were foundnear the welds andm the base metalof thh elbow No 1 delaHed analyses of these hnes have been performedlo predk/ the thermalhklorks and the locahon l nhere lhe actualkHure occurred or to eshmale the hme klena/H fookfor a crack to mHhle from i the start ofleaAage.

I l

Response to Evaluation item 9:

1

! The Parley safety injection cracking incident has been thoroughly investigated from the standpoint of l l metallurgical investigations, temperature monitoring, flow diversion testing (to estimate leakage flow i rate) and stress, fatigue and fatigue crack growth analyses. This information is documented in Reference 13. Thermal histories, evaluation of the cracked location and the time estimate for crack initiation and propagation through the wall are included in this report.

, l For Tihange, metallurgical investigations were performed which confirmed that the failure mechamsm '

was fatigue. The Tihange cracking would be predicted by the TASCS methodology,in a similar manner as for Parley, as discussed in Evaluation item 1.

Evaluation item 10:

l The equalkn for turbulentpenetra/hn nHh lesiate under operahng condRhns appears to be based on an ad-hoc assumphon andlangenth/mean rehcHy dala deleramedfrom the Low Temperature Turbulentpenetrabbn testprostam. Nojush7kathn ahythh&acceptableforactualoperalmg temperatures andpressuresm an enrkonmentsimHar to STP Umts / and2 has beenprovided ho correhlkn nHh larbulentpenetrathn data actuaHy measureda nue/earpla,71s under operathig and m/ealage or ouuealage condRhm: shnHar lo those desenbeda BuHelm 88-08 has been shown Response to Evaluation item 10:

l The equation for turbulent penetration with leakage is based on axial velocity, not tangential or mean velocity. The data was nondimensionalized for correlation to actual operating environments. The turbulent penetration data was used in the development of the thermal cycling methodology.

Turbulent penetratica data actually measured in nuclear plants under operating conditions with leakage is not avalim and is not practical, since instrumentation within the piping would be necessary. However, oeide wall temperature measurements have been obtained for correlation of the zone of thermal cycling, as defined by 1.,, (see Evaluation item 1 for the South Texas alternate charging line and Evaluation Item 6 for the farley safety injection line). This methodology was confirmed using prototypical conditions as explained in the response to Evaluation item 2c.

The TASCS methodology would predict the Parley cracking,i.e., the cold inleakage would be cold at the

elbow, and thermal cycling would occur between the valve and the elbow.

Based on the preceding paragraphs, the turbulent penetration thermal cycling methodology is applicab!e to South Texas. In fact, monitoring data from the South Texas alternate charging line substantiates the methodology, and also confirms that cycling will not occur at the check valve.

l 11

i

\ . .

1 -

Evaluation llem 11:  !

1he derivalkn ofthe equahon hr turbulentpenetra/hn rHb strahlicahon How appears to be over/r ,

shnph&d andshouldhave been basedon the njorous appkalkn ofIranskn/ thermo-hydrauk mechamesprhep/es(e'.g. Reference 16) appkable to the operalms condihonsfornuclearppmg allachedto the HL2 hTlectssuch asRCLpressure ranahon, possiblegas entrapmenth tnepipe sechbn between the block andcheck sa/re.s andcheck sa/re cha//erk t underlowHow condibbns i nere also not considered  ;

l l l 1

Respoi.se to Evaluation item 11: ,

A simple energy approach was initially applied and then compared to experimental data to confirm l its adequacy. This approach was sufficient because the method generated from the approach was a i simple conservative analytical tool to predict the maximum distance from the header pipe at which I thermal cycling is possible. A more rigorous academic approach was not applied because it was not l required since this problem could be addressed with the simplified approach. In particular, for the l South Texas charging line evaluation, there is no reason to believe RCL pressure variation, check valve  !

chattering or possible gas entrapment would be significant issues to conclude differently.

The reference cited by the NRC does not combine complex phenomena like turbulent penetration and i stratification. No proven technology which couples such phenomena is available.

Evaluation item 12.

I Cer/sh thermo-hydraukphenomena a stagnantpphs allached to the RCL or other hikh Row rate ppmg have been desenbedh the hierature (References 16 - 19) These researchers hare idenh&d a hekoida/ or corkscrewHowpattern. shkh hasnot been obsenedh theflests Ab reason for thhhas beenprovided Response to Evaluation item 12:

The phenomena discussed are poorly documented and cannot be evaluated for their applicability to l' l the South Texas analyses.

l A " swirling" type of flow in the branch line was noted in the Westinghouse low temperature turbulent

- penetration tests, and visualization tests. This data is included in the program.

Response to NRC Conclusions

! Conclusion item 1:

The mechankm of turbulentpenetralkn b asjet not ueB de/medand tmderslood andhasnot been.

fuHybresbjoledunder the TASCSprogram. Essigmkance h thefaHuresalParle): Tihange and Genkaihas not been clearly estab/hhed The roof cause for the faHures desenbedh BuHe/h 88-08 i remaks undelermhied 12 .

l

I . .  !

Response to Conclusion item 1:  ;

The mechanism of turbulent penetration, and its effects on heating branch lines ("lurbulent i l penetration length") and interacting with stratified flows (" turbulent penetration thermal cycling")

l has been thoroughly studied in the TASCS program. Practical engineering tools to conservatively l evaluate the effects of turbulent penetration on nuclear power plant piping have been developed and l utilized in the South Texas evaluation of the charging system piping.

l The farley safety injection cracking incident has been thoroughly investigated and is documented in Reference 13. The investigations were comprised of metallurgicalinvestigations, temperature monitoring, flow diversion testing, stress evaluation, and fatigue and fatigue crack growth analyses.

Reference 13 contains the evaluation of the cracked location and the time estimate for crack initiation and propagation through the wall. Thermal histories are included in Attachment A of Reference 13. The root cause of the cracking was determined to be high cycle fatigue resulting from stratification and thermal cycling. The Tihange cracking is similar to Farley.

Conclusion llem 2:

The RSCSmethodolog does not appear lo have the capabHHy topredk/ the observedhlfue kHures desenbedm BuHelm 88-08 sace the mos/ hkely kHure locathns appear lo be at the chsest elbow 1 to the HCL neH wHhm the turbulentpenetralkn reghn. The DSCSmethodohg doesnotpredk/

kHure althese hes/hns Response to Conclusion item 2:

i The TASCS methodology does predict failure at the observed crack locations, as discussed in the responses to NRC Evaluation items 1,9 and 10 above. A summary of the results of the TASCS methodology for the Farley/Tihange cracking is as follows:

a) The unisolable piping is estimated to be equal to the RCS cold leg temperature.

b) The leakage entering the unisolable pipe is at the containment ambient temperature.

l c) The leakage flow will be stratified, based on the Richardson number.

l l d) All of the unisolable piping (from the cold leg to the check valve)is susceptible to thermal cycling.

e) The temperature difference between the ambient pipe fluid and the leak flow is 450*F. which is j higher than the estimated value of 335'F for Farley.

f) The frequency of cycling is conservatively one cycle per minute. (The Farley monitoring indicated that a cycle occurred about every 4 minutes. Cycles with larger temperature fluctuations occurred about every 13 minutes, and cycles with smaller temperature fluctuations occurred about every 6 13

l l~ minutes).

l I

~

g) fatigue and fatigue crack growth analysis resulted in a prediction of failure of 3 to 6 calendar  !

years, which was conservative with respect to the Parley plant age. j l

)

I i

1 Conclusion item 3: l H/2Phasnot adequatelyjitsh7ted dhcon/muhr temperature momforks at STP l/mls 1 and2 nor i prosMedan acceptable alternalke to momforks. On thh basis the slafftherefore concludes thal H/APhas notprorMed the requesledassurance ofkhbn 3 of the BuHehh agahst hadvertent leakage h the unhobble segments of the charght hhes and auxHhrypressurker spray systems at South Teaas (Jm/s1and2 TheIkenseeshouldberequesledto checkforpotenth//eakageh these sistems andet/herreestab/hh theprerhus temperaturemomforkgpmgram alboth umis or knplement other acceplable momforkgprograms that couldsalbfy theprotxrkns ofkhon 3 of the l Bu//ehh. for the h7e of theplant. andprorMe a desenpthn of thek ochbns to the stall

\

Response to Conclusion llem 3: l 1

The removal of the monitoring instrumentation at South Texas Units 1 and 2 is well justified as j demonstrated through evaluations performed using the TASCS methodology. These evaluations

! concluded that piping integrity would not be jeopardized, should leakage occur over the life of the j units. i Westinghouse Conclusions Regarding the NRC SER l

In light of the NRC concerns (Reference 11) regarding the HlhP submittal (Reference 8), and the EPRI l TASCS report (Reference 10), and the responses to these concerns provided herein, Westinghouse i concludes that the analyses performed for South Texas Project Units 1 and 2 are conservative and  ;

have demonstrated the structural integrity of the susceptible piping for the full licensed life of the units. Therefore, Westinghouse maintains that temperature monitoring for adverse stresses from isolation valve leakage is not necessary at these units.

1 14 l

l

?-

References:

1) United States NRC Bulletin 88-08, " Thermal Stresses in Piping Connected to Reactor Coolant Systems",06/22/88; Supplement 1,06/24/88; Supplement 2,08/04/88; and Supplement 3,04/11/89. l 2)llL&P letter ST-HL-AE-2788," South Texas Project Electric Generating Station Units 1 & 2, Docket No. STN 50-498/499. Response to NRC Bulletin 88-008. Supplement 1, and Supplement 2. Thermal ,

Stresses in Piping Connected to Reactor Coolant Systems'," 09/28/88, J. H. Goldberg to United States l NRC Document Control Desk.

3) HL&P letter ST-HL-AE-2994, " South Texas Project Electric Generating Station Unit 2, Docket No.

STN 50-499, Supplemental Response to NRC Bulletin 88-008. Thermal Stresses in Piping Connected to Reactor Coolant Systems." 02/23/89 J. H. Goldberg to United States NRC Document Control Desk.

4)liL&P letter ST-HL-AE-3239. " South Texas Project Electric Generating Station Units 1 & 2 Docket Nos. STN 50-498 and 50-499, Supplemental Response to NRC Bulletin 88-008: Thermal Stresses in Piping Connected to Neactor Coolant Systems," 11/20/89, S. L Rosen to United States NRC Document  !

Control Desk.

5) HL&P letter ST-HL-AE-3615," South Texas Project Electric Generating Station Units 1 & 2 Docket Nos. STN 50-498, STN 50-499, Transmittal of WCAPs on Evaluation of Auxiliary Piping," 11/05/90, M. A.

i McBurnett to United States NRC Document Control Desk. (This reference transmitted Westinghouse l Proprietary Class 2 report WCAP-12598 "NRC Bulletin 88-08 Evaluation of Auxiliary Piping for South Texas Project Units 1 and 2", May 1990, and Westinghouse Class 3, Non-Proprietary report WCAP-12646).

l 6) HL&P letter ST-HL-AE-3579, " South Texas Project Electric Generating Station Units 1 & 2. Docket L Nos. STN 50-498 and 50-499, Deletion of Temperature Monitoring Instruments, Supplemental

! Response to NRC Bulletin 88-008: Thermal Stresses in Piping Connected to Reactor Coolant System,"

09/21/90, M. A. McBurnett to United States NRC Document Control Desk.

l

7) United States NRC letter," South Texas Project - NRC Bulletin 88-08, Thermal Stresses in Piping Connected to Reactor Coolant Systems (TAC Nos. M69689 and M69690)," 09/23/92, G. F. Dick to D. P. )

I Hall (HL&P)

8) HL&P letter ST-HL-AE-4639, " South Texas Project Units 1 and 2, Docket Nos. STN 50-498; STN 50-499. Response to Request for Additional Information Regarding Bulletin 88-08. Thermal Stresses in Piping Connected to Reactor Coolant Systems (TAC. NOS. M69689 and M69690)," 11/30/93. T. H.

l Cloninger to United States NRC Document Control Desk. (This reference transmitted Westinghouse Proprietary Class 2 report WCAP-12598, Supplement 1, "NRC Bulletin 88-08 Evaluation of Auxiliary l Piping for South Texas Project Units 1 and 2," November 1993, and Westinghouse Class 3, Non-Proprietary report WCAP-12646, Supplement 1).

9) United States NRC letter," Resolution of NRC Bulletin 88-08, Thermal Stresses in Piping Connected I to Reactor Coolant Systems, for South Texas Project, Units 1 and 2 (TAC NOS. M69689 and M69690),

04/11/94, L. E. Kokajko to W. T. Cottle (HL&P).

15 1

References (continued):

10) Electric Power Research Institute Report EPRI TR-103581 Project 3153-02, entitled " Thermal Stratification, Cycling, and Striping (TASCS)," March 1994, prepared by Westinghouse Electric Corporation, EPRI Proprietary Licensed Material.

l 11) United States NRC letter, " Resolution of NRC Bulletin 88-08, ' Thermal Stresses in Piping Connected  !

l to Reactor Coolant Systems,' South Texas Project, Units 1 and 2 (STP)(TAC Nos. M93822 and M93823)."

l 02/23/96 T. W. Alexion to W. T. Cottle (HL&P).

. 12) HLAP letter ST-HL-AE-2950, " South Texas Project Electric Generating Station Units 1 and 2, l Docket Nos. STN 50-498, STN 50-499, Evaluation of Thermal Stratification for the Residual Heat  ;

Removal Lines," 01/27/89, M. A. McBurnett to United States NRC Document Control Desk.

13) Westinghouse Report WCAP-11786, "J. M. Farley Unit 2 Engineering Evaluation of the Weld Joint Crack in the 6" Sl/RHR Piping," April 1988 Westinghouse Proprietary Class 2.

14) Kim, J. H., A. F. Deardorff and R. M. Roidt. " Thermal Stratification in Nuclear Reactor Piping Systems," presented at the International Conference on Nuclear Engineering, November 1991, Japan. J 1

15) Baron, F., M. Gabillard and C. Lacroix, " Experimental Study and Three-Dimensional Numerical  !

Prediction of Re-circulation and Stratified Pipe Flows in PWR " Fourth International Topical Meeting i on Nuclear Reactor Thermal-Hydraulics," Karlsruhe, Germany, October 10 - 13, 1989.

16) Robert, M., and J. D. Mattei. " Thermal-hydraulic Phenomena in a Zero Flowrate Pipe Connected to I a High Flowrate, High Temperature Circuit." First international Symposium on Engineering Turbulence Modelling and Measurements," September 24 - 28,1990 Elsevier, New York.

l 17) Robert, M., " Corkscrew Flow Pattern in Piping System Dead Legs " Fifth International Topical j j Meeting on Nuclear Reactor Thermal Hydraulics " Salt Lake City, September 1992.  !

l 18) Levy, L A., " Secondary flow Heat Transfer Phenomena in Branch Piping " Thesis, Department of

! Mechanical Engineering, MIT May 1993.

i

19) Van Duyne, D. A., et. al.," Thermal Stratification and fatigue of Piping in Nuclear Power Plants,"

1990 ASME Pressure Vessels and Piping Conference, Nashville, Tennessee, June 17 - 21, 1990.  :

t l

, ~

l 4

16 l