ML17310B017

Responds to Questions Raised During 931215 Telcon Re NRC Bulletin 88-008 Actions,Results of APS Thermal Stratification Monitoring & Plans for Long Term Monitoring of Thermal Stratification of Apss Piping

ML17310B017
Person / Time
Site:	Palo Verde
Issue date:	01/26/1994
From:	Conway W ARIZONA PUBLIC SERVICE CO. (FORMERLY ARIZONA NUCLEAR
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
102-02801-WFC-R, 102-2801-WFC-R, IEB-88-008, IEB-88-8, NUDOCS 9402030224
Download: ML17310B017 (20)

Text

aCCj"j ymir usa I m.rSu Xiuis uriViujsaimliui>>X>>CIVi

~

~

REGULATORY INFORMATION DISTRIBUTION SYSTEM (RIDS)

ACCESSION NBR:9402030224 DOC.DATE: 94/01/26 NOTARIZED:,NO DOCKET' FACIL:STN-50-528 Palo Verde-Nuclear Station, Unit 1, Arizona Publi 05000528 STN-50-529 Palo.Verde Nuclear Station, Unit 2, Arizona Publi

.05000529 STN.-50-530 Palo Verde Nuclear Station, Unit 3, Arizona.Publi 05000530 AUTH.NAME AUTHOR AFFILIATION CONWAY,W.F.

Arizona Public Service Co.

(formerly Arizona Nuclear Power R

RECIP.NAME RECIPIENT AFFILIATION Document Control Branch:(Document Control Desk)

I

SUBJECT:

Responds to questions raised during 931215 telcon re NRC Bulletin 88-008 actions, results of APS thermal stratification monitoring

& plans for long, term -monitoring of thermal stratification of APSS piping.

DISTRIBUTION CODE:

IE16D COPIES RECEIVED:LTR ENCL SIZE:

TITLE: Bulletin Response 88-08 Thermal Stress in Piping to RCS.

NOTES:STANDARDIZED PLANT Standardized plant.

Standardized plant.

'S A

05000528 05000529'D 05000530 D

RECIPIENT ID CODE/NAME PDV PD TRAN, L COPIES LTTR ENCL 1

1 1

1 RECIPIENT ID CODE/NAME

POLZCH, T

COPIES LTTR ENCL 1

1.

INTERNAL AEOD/DOA NRR/DE/EMEB NRR -.DSS RGN5 FILE EXTERNAL: NRC PDR 02 01 1

1 1

1 1

1 1

1 1

1 1

1 NRR/DE/EMCB

'NRR/DORS/OGCB NRR/PDII-1 RES/DSIR/EIB NSIC 1

1 1

1 1

1 1

1 1

1 D

S NOTE TO ALL"RIDS",RECIPIENTS:

PL'EASE'ELP US TO REDUCE WASTE! CONTACTTHE DOCUMENT CONTROL DESK, ROOM.P!-37 (EXT. 20079) TO ELIMINATEYOUR NAME FROM. DISTRIBUTION L'ISTS FOR DOCUMENTS YOU DON'T NEED!

TOTAL NUMBER OF COPIES REQUIRED:

LTTR 14 ENCL 14 D

D

0~

Arizona Public Service Company

'P.O, BOX 53999

~

PHOENIX. ARIZONA85072-3999 WILLIAMF. CONWAY EXECUTIVEVICEPRESIDENT NUCLEAR 102-02801-WFC/RA'B/ZE January 26, 1994 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Mail Station P1-37 Washington, DC 20555

Subject:

Palo Verde Nuclear.Generating Station (PVNGS)

Units 1, 2, and. 3

.Bulletin 88-08 Follow-up on Telephone Discussion with the 'NRC on 12/15/93.

File: 94-055-026 94-056-026 On December 15, 1993, a telephone discussion was held between Brian Holian, NRC and Richard Bernier, Arizona Public Service Company (APS) ~

The discussion focused on NRC Bulletin.88-08 actions, results of APS.thermal stratification monitoring, and on.APS plans for long term monitoring of thermal stratification in the Auxiliary Pressurizer Spray System (APSS) piping.

APS provides the following responses to the questions asked during the telephone discussion.

Why did APS use "Plastic Pipe Element STIF20" in the bounding thermal stratifcation analysis?

APS modeled the APSS stratified section by using the plastic pipe element in ANSYS 4.4A in order to allow for the input of the variable pipe wall temperatures.

experienced during stratification. The plastic capability of this plastic element was not utilized in the calculation (13-MC-.ZZ-643), because the stresses remained within the elastic region of the. stress-strain diagram for the APSS piping material.

Thus, the accuracy of the calculation has not been degraded by the use of the plastic element.

OX0088 9402030224 94012b PDR ADQCK 0500052B 9

PDR

II 4l Qr

U. S. Nuclear Regulatory Commission Document Control Desk NRC Bulletin 88-08 (Thermal Stratification Monitoring in APSS)

Page 2 2.

Was the stratified temperature data from the APSS piping cyclic?

The temperature stratification data measured by APS on the APSS piping in Unit

,3 during and after the second refueling outage did not show high cycle fluctuations.

Once a stratification profile was established, it remained constant in relation with a constant valve in-leakage rate. Therefore, in the absence of cycling, APS concluded that the temperature stratification recorded on the APSS line does not represent the potential for high cycle thermal fatigue as described in NRC Bulletin 88-08, and consequently, is not within the scope of the Bulletin. APS has determined that the observed temperature stratification in the APSS line was the result of valve in-leakage as demonstrated by starting and securing the charging pumps which produced a marked change in the recorded temperature distribution.

Since the static temperature stratification was not considered in the original design basis calculation, APS prepared a bounding thermal analysis calculation and is preparing for additional temperature measurements in the APSS line. The required instrumentation will be installed in Unit 3 during the Spring 1994 refueling outage.

This APS long term monitoring is designed to establish the limiting conditions to help revise the associated calculations.

The enclosure to this letter is a summary

'fAPS actions and plans. regarding NRC Bulletin 88-08.

Finally, in accordance with the 12/15/93 telephone discussion, APS willcontinue its long-term monitoring plan for temperature stratification in the APSS line as described above.

Also, APS considers all action items of NRC Bulletin 88-08 closed.

Ifyou have additional questions, please contact Richard Bernier at (602) 393-5882.

Sincerely, WFC/RAB/ZE/did Enclosure cc:

K. E.

Perkins'.

E. Johnston B. E. Holian

Ol 4I

ENCLOSURE DEVELOPMENT AND BASIS FOR THE TECHNICALPOSITION AT PVNGS WITH RESPECT TO NRC BULLETIN88-08

0

DEVELOPMENTAND BASIS FOR THE TECHNICAL',POSITION AT PVNGS'ITH RESPECT TO NRC BULLETIN88<<08 BACKGROUND NRC Information and Evaluation Bulletin 88-08 The circumstances that resulted in the original IE Bulletin 88-08 involve a small LOCA resulting from a through wall crack on an unisolable section of ECCS piping at the Farley 2 nuclear power reactor. It was postulated in the bulletin that a globe valve located upstream of the crack that isolates the RCS from the charging pumps had been leaking, allowing cold water to flow into the relatively hot ECCS piping connected to the RCS.

Consequently, a fluid stratification-profile developed in the piping resulting in thermal loadings not accounted for in the original plant design. Interaction of the stratified layers with secondary turbulence within the unisolable section of piping produced cyclic condi-tions which ultimately resulted in high cycle thermal fatigue.

As a result of the incidence at Farley 2, the NRC requested that all nuclear power plants review their design for any unisolable RCS branch piping that could be subject to temper-ature differences that produced unacceptable thermal stresses and to take action where such piping was identified. Specifically, in the event that such unisolable sections are found, the licensee was requested to perform nondestructive examination of the welds, heat-affected zones, and high stress locations,to ensure that, there were no existing flaws. In addition, the licensee was required to implement a:program to provide contin-ued assurance that unisolable sections of RCS branch piping would not be subject to combined cyclic and static. thermal or other stresses. that could cause failure during the remaining life of the unit. Each licensee was.required to report their findings and confir-mation of completion of those items specified as Actions Requested.

Supplements to /E Bulletin 88-08 Subsequent.to the event at Farley 2, Supplement 1 of IE Bulletin 88-08 reported that another incidence of piping failure occurred at the Tihange 1 PWR at Tihange, Belgium.

Similar to the failure at Farley-2,,the crack developed in an unisolable section of piping as a result of thermal stratification induced high cycle thermal fatigue. Supplement 2 emphasized the need for sufficient examination for the detection of cracks and flaw indi-cations.

Supplement 3 of IE Bulletin 88-08'reported another case of piping failure at a foreign reactor similar in circumstance to those reported previously. However, in contrast,to Far-ley 2, where in-leakage of relatively cooler water past the isolation globe valve is sus-pected of causing thermal fatigue, the failure at the foreign reactor appeared to be the result of hot RCS water leaking past the packing gland, of the isolation valve. The leak at the packing gland allowed the hot RCS water to flow into the unisolable section of branch pipe above the stagnant cooler water. Consequently, a stratified profile developed which presumably resulted in the failure of'the pipe.

The significance of Supplement 3, as discussed therein, is that the third incidence of pip-ing failure apparently occurred as a result of out-leakage of relatively hot RCS water into

~I 4i 1

f

a cooler stagnant unisolable section of pipe whereas the failure at Farley 2 occurred as a result of in-leakage of relatively cooler. water into a warmer stagnant unisolable section of pipe. Consequently, the criteria by which many utilities evaluated their systems, in accor-dance with the original bulletin, had changed to include branch piping identification based on both in-leakage to and out-leakage from the RCS.

PVNGS InitialResponse

- Study 13-MS-A 12 To address the issues of IE Bulletin 88-08 and Supplements 1 and 2, Engineering Action Request (EAR) 88-1088 was initiated to assess the impact on PVNGS. The disposition of 88-1088 resulted in the study 13-MS-A12 which tabulated all RCS branch piping and applied the criteria as perceived from the original IE.Bulletin to evaluate which system interface piping may be susceptible to high cycle thermal fatigue. Consequently, the study determined that only systems that were at higher relative pressure to the RCS were susceptible to,thermal fatigue as described in the original bulletin. The review indi-cated that only the auxiliary pressurizer spray line was susceptible to the thermal load-ings suspected of the failures at Farley 2 and Tihange. The safety injection lines, although cross-tied to the charging pumps at higher relative pressure, were not subject to the thermal loadings as these lines are isolated by two normally closed isolation valves. Leakage past, both valves was not assumed credible. Thus, the safety injection lines were dismissed from further review.

Study 13-MS-A12 and the actions initiated as a result of EAR 88-1088 formed the basis of the initial response to the NRC. In ANPP correspondence 161-01356-DBK/BJA, dated October 03, 1988, PVNGS had Identified all RCS branch piping susceptible to stratifica-tion and the associated thermal stresses on the basis described above. In addition, to comply with Requested Action 2 of the original bulletin, non-destructive examinations of the AuxiliarySpray Lines for each unit were planned during the next respective sched-uled refueling outages. No cracking was detected in any of the units. Regarding Action 3 of the original bulletin, no long term plan was specified within the PVNGS response to assure that the AuxiliarySpray Line would not be subject to fatigue failure resulting from stratified conditions. However, to fulfillPVNGS's commitment to Requested Action 3, Procedure 73TI-9RC03 was prepared to obtain temperature data for the AuxiliarySpray Line. Subsequent fatigue evaluation based on the temperature data during the full cycle of Unit 3 was prepared to assess integrity of the AuxiliarySpray Line. Therefore, PVNGS was in compliance with,the requirements of IE Bulletin 88-08.

Revised Response to IE Bulletin'88-08 Supplement 3 to IE Bulletin 88-08 was issued after the initial PVNGS response to the NRC. As noted,. Supplement 3 contained evidence of RCS out-leakage that resulted in stratified conditions and ultimately thermal fatigue in an unisolable section of branch-pip-ing. Consequently, all licensees were required to evaluate the potential of similar circum-stances at their facilities in addition to those previously reported in the original bulletin.

Concurrent with this revision in evaluation criteria, the NRC had reviewed the original PVNGS response. According to NRC's communication, dated October 18, 1991, the PVNGS response did not adequately address Action 3 of the bulletin..As these details were not included in the original response,.the NRC was not informed that PVNGS had

~I 4l I

l

proceeded with installation of temperature. monitoring equipment and collection of tem-perature data with which to assess the AuxiliarySpray Line and'its potential, for thermal fatigue. Instead, the NRC had concluded that the NDE. actions proposed in response to Action 2 were devised to:concurrently fulfillthe requirements of'Action 3. The NRC also

're-emphasized. the need to address the specific concerns related to Supplement 3.and that, although no specific response was required, audit of the implementation of IE Bulle-tin 88-08 and all supplements,was possible. The NRC did however request additional information concerning planned, actions regarding Action 3 assurance plans.

In correspondence 161-04571-WFC/JMQ,.dated January 15, 1992, APS supplied'the NRC with detailed information regarding existing efforts to collect,.reduce, and evaluate

, temperature data consistent with.that. required by Action 3 of the bulletin. Further, APS committed to developing a long term monitoring plan to ensure the integrity of the auxil-iary spray line for the design life of the plants, and that, the monitoring plan would be based upon the results of the U3 data assessment.

The revised response also indicated that APS was.participating with the Westinghouse / Electric Power Research Institute efforts in.the. development of screening criteria-and evaluation methodology and was par-ticipating with related activities, of the,Combustion'Engineering Owners Group. All.efforts were concerted.to bring about closure of the related'issues.

Completion of the Unit 3 cycle 3 auxiliary spray line monitoring concluded that, although stratification in,the auxiliary spray line. did,exist, the temperature profiles were not cyclic and the mechanism associated with the failures:described in the bulletin and supple-ments was absent. Further, since the maximum top to bottom temperature difference was only 115 F, as compared to 600 F from analysis, in-leakage was not suspected.

It was postulated. that the observed profiles were the result of steam/water stratification,,

natural convection, external convection effects, or.otherwise. Independent of leakage, the stratification was assumed constant and therefore provided a bounding steady state thermal load which could be used for a bounding structural analysis.

The structural analysis for the auxiliary pressurizer spray line was revised accordingly to include the thermal static load. The bounding analysis assumed that the. magnitude of the load.to be the maximum'ecorded during the Unit 3 full,cycle data collection.

Recommendations for Continued Evaluation of the AuxiliaryPressurizer Spray Line The Unit 3 cycle 3 data was again reviewed to determine if such monitoring..was neces-sary assuming that the-stratification was not due,to in-leakage. Re-evaluation and corre-lation of the data-with changes in plant operation demonstrated a relationship between charging pump operation:and the stratification profile in the auxiliary spray line. Engi-neering concluded that the stratified profile was a result of in-leakage, contrary to the, first assessment of the data.,ln evidence of the in-leakage in the auxiliary spray line, Engi-neering concluded. that continuous monitoring was required since the resultant static load was ultimately a function. of leak rate. Although no evidence of cycling was found, Engineering proceeded to evaluate the stratification as.an;IEB 88-08 related issue. In accordance with the requirements transmitted by the NRC in "Evaluation Criteria for

. Responses to NRC Bulletin 88-08, Action 3 and Supplement:3," a pilot program was pro-posed for Unit 3 in the fourth refueling outage in the spring of 1994. Based on-these efforts and the data collected, permanent monitoring for all three units would be

~I

~I

)'l

prescribed, In addition, the limitingtemperature distribution may.be evaluated analytically provided specific thermal-hydraulic parameters may. be accurately determined and veri-fied.

ASSESSMENT OF THE PVNGS SPRAY LINE DATA The auxiliary spray system provides a flow path from the CVCS charging pumps to the RCS at the main spray line. During warm-up and normal operation the auxiliary spray line is isolated from the RCS by isolation valves HV-203 and HV-205 as shown in Figure

1. Depending upon specific conditions, the valves are cycled during cooldown when inadequate main spray flow exists. It was assumed that leakage past the isolation valves HV-203 and/or HV-205 would result in stratified conditions and consequently thermal loadings not originally accounted for in the design of PVNGS. To establish the existence of such conditions, temperature measurements were made for the auxiliary spray'line and the data reduced and assessed for PVNGS by CE-ABB-Fatigue Evaluation Ser-vices. The data collection occurred in Unit 3 during heat-up after refueling outage 2. Note that the outage was extended and that the plant was placed in.cold shut down following start-up. Consequently, auxiliary spray, line data obtained from this period represents partial heat-up and cooldown cycles as well as normal operating temperature data.

HV-203 HV-205 12" CH-431 12" TC 21 TC 23 Main Spray 12" I

TC 15 T

18 TC 19 TC 17 TC 20 TC 22 TC 16 Pressorizer Fiiture l. Location ofUnit 3 thermocou plea on auxiliary pressurizer spray line.

Review of the control room logs corresponding in,time to the temperature variations recorded during warm-up indicated a relationship between the auxiliary spray line tem-perature distribution and charging pump operation. Starting and securing the charging pumps produced a marked change in the auxiliary spray line temperature distribution as shown in Figure 2. Securing all charging pumps produces an instantaneous temperature rise in the auxiliary spray line up to and consistent with pressurizer temperatures.

Con-versely, starting a charging pump produced an immediate uniform decrease in.exterior pipe wall temperature. Such rate of change'in the exterior pipe wall temperature can only be explained by forced convection within the pipe (i.e. full pipe flow).

II 0

7 I

II

Although this does not demonstrate stratification, it does demonstrate leakage past the auxiliary.spray isolation valves. Thermal stratification is a function of the leakage flow rate which in turn is a function of the relative pressure difference between the RCS and the charging system. Consequently, at normal operating pressure, the relative pressure difference between the RCS and charging system is decreased and the leakage flow rate is correspondingly smaller. Leakage flow rates, insufficient to maintain full pipe flow, result in a stratified fluid profile.

400 Top Therruocouple (t8)

Midde Thenaocouple (20)

Betters Them> ocoupte (t 9) 200 E

I-100 + Ig ~ Cha ing puraps s red tt - Charging unrp E started 20 10 0

5 55 25 30 Time (houre) 8gure 2. Temperature distribution downstream ofthe auxiliary spray inlection check valve. Refer to Hgure 1. Distribution demonstrates the effect ofsecuring all charging pumps at to and the effect ofstarting onc charging pump at tt. Charging pump flow, leaking past thc isolation valves, cools thc pipe uniformly prior to and after to and tt rcspecnvely. Cessation ofcharging pump flowallows pressurizer steam to enter the pipe resulting in a sharp increase in pipe wall temperature. The relatively uniform temperature between to and tt indicates that tbe water layer height is small and thc pipe is fullofsteam.

Data collected during normal operation substantiates this conclusion. Specifically, the normal operating data shown in Figure 3 demonstrates stratification, the magnitude of which is proportional to system temperature and pressure. The data does not reflect high cycle fluctuations in the spray; once established, the stratification is constant corre-sponding to a constant leak rate.

Consequently, the assumptions regarding the bounding thermal stratification, loads used in the class 1 structural evaluation are not correct. The magnitude of the thermal load is directly proportional to the leak rate which may vary from cycle to cycle and with valve maintenance. Therefore, the leak rate or the actual temperature distribution must be.

monitored to ensure that the maximum top to bottom temperature difference does not exceed that assumed in APS calculation 13-MC-ZZ-643. Alternatively, if a correlation can be made relating the leak rate and the resultant temperature distribution, the

~l 11

maximum stress can be determined and factored into the fatigue evaluation thereby eliminating the need for continuous monitoring.

Top Thenaocouple (18)

Middie Thermocoupte (20)

Betters Thenuoccuple (I9) 450 8

400 EI-Tcts 0

100 200 300 400 500 600 700 800 Time (min)

FIGURE 3. Normal operating temperature distribution on the auxiliary pressurizer spray line. &e recorded data does not demonstrate cycling during normal operation. Data is representative ofmaximum normal operating top to bottom temperature difference.

MONITORING PROGRAM AND CLOSURE ACTIONS In the absence of cycling, Engineering has concluded that.the. stratification recorded on the auxiliary spray line does not represent the potential for high cycle thermal fatigue and therefore is beyond the scope of NRC Bulletin 88-08. The static load itself, however, was not considered in the original design basis and must be properly incorporated. The exist-ing calculation accounts forthe recorded temperature load but there is no assurance that these are the limiting conditions. Engineering, therefore, has prepared an installation plan for the Unit 3 fourth refueling outage to re-instrument the auxiliary spray line.

The auxiliary spray line willbe re-instrumented with thermocouples at the approximate locations used in the previous data collection effort. Additional thermocouples will be used to further define the temperature distribution upstream of the auxiliary spray isola-tion valves. The equipment willbe installed during Unit 3 refueling outage 4 for data col-lection during cycle 5 and willprovide one complete cycle-of data. The effort will conclude at the end of the cycle and data analysis willbe completed thereafter.

Il