ML13331A251

Forwards Status of Actions in Response NRC Bulletin 88-002, Rapidly Propagating Fatigue Cracks in Steam Generator Tubes. Primary-to-secondary Leak Rate Monitoring Program to Be Implemented Due to Discovery of Evidence of Denting

ML13331A251
Person / Time
Site:	San Onofre
Issue date:	04/06/1988
From:	Medford M SOUTHERN CALIFORNIA EDISON CO.
To:	Martin J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
References
IEB-88-002, IEB-88-2, NUDOCS 8804210004
Download: ML13331A251 (11)

Text

Southern California Edison Company P. 0. BOX 800 2244 WALNUT GROVE AVENUE ROSEMEAD, CALIFORNIA 91770 M.O. MEDFORD TELEPHONE MANAGER OF NUCLEAR ENGINEERING April 6, 1988 (818) 302-1749 AND LICENSING U. S. Nuclear Regulatory Commission Office of Inspection and Enforcement Region V 1450 Maria Lane, Suite 210 Walnut Creek, California 94596-5368 Attention:

Mr. John B. Martin, Regional Administrator

Subject:

Docket No. 50-206 San Onofre Nuclear Generation Station Unit 1 The status of Southern California Edison (SCE) actions in response to the requests given in NRC Bulletin No. 88-02, "Rapidly Propagating Fatigue Cracks In Steam Generator Tubes," is contained in the enclosure.

In summary, review of the past steam generator tube inspection data obtained at the uppermost tube support plate indicate evidence of denting and the associated tube support corrosion at the tube-to-tube support intersections evaluated. Since denting has been confirmed, an enhanced primary-to-secondary leak rate monitoring program will be implemented as described in the enclosure.

Subscribed on the _lo day of April, 1988 Respectfully submitted, SOUTHERN CALIFORNIA EDISON COMPANY By:

M. 0. Medford Manager of Nuclear Engineering and Licensing Subscribed and swo to before me this

______day of idi

/_f_

OFFICIAL SEAL AGNES CRABTREE Notary PUbhc-C8liffnia SLOS ANGELES COUNT' No y Public in and for the County of

,MyComm.Ep.

Sep.14.1990 Los Angeles, State of California Enclosure 8804210004 :380406 PDR ADOCK 05000206 X

Q DCD cc:

U. S. Nuclear Regulatory Commission, Document Control Desk F. R. Huey, NRC Senior Resident Inspector, SONGS 1, 2 and 3

RESPONSE TO NRC BULLETIN NO. 88-02, RAPIDLY PROPAGATING FATIGUE CRACKS IN STEAM GENERATOR TUBES March 28, 1988 The following information is provided in response to the actions requested in NRC Bulletin 88-02 as it relates to San Onofre Unit 1 for minimizing the potential of a steam generator tube rupture event caused by a rapidly propagating fatigue crack such as occurred at North Anna 1.

A. Inspections (Paragraph A of NRC Bulletin 88-02)

"The most recent steam generator inspection data should be reviewed for evidence of denting at the uppermost tube support plate. Inspection records may be considered adequate for this purpose if at least 3% of the total steam generator tube population was inspected at the uppermost support plate elevation during the last 40 calendar months.

"Denting" should be considered to include evidence of upper support plate corrosion and the presence of magnetite in the tube-to-support plate crevices, regardless of whether there is detectable distortion of the tubes.

The results of this review shall be included as part of the 45-day report.

Where inspection records are not adequate for this purpose, inspections of at least 3% of the total steam generator tube population at the uppermost support plate elevation should be performed at the next refueling outage. The schedule for these inspections shall be included as part of the 45-day report and the results of the inspections shall be submitted within 45 days of their completion. Pending completion of these inspections, an enhanced primary-to-secondary leak rate monitoring program should be implemented in accordance with paragraph C.1 below."

San Onofre Unit 1 Steam Generator Inspection Data Review The most recent steam generator inspection data has been reviewed for denting at the uppermost tube support plate (the fourth tube support plate in San Onofre Unit 1).

Evidence of denting as defined above was found at many of the tube-to-tube support plate intersections evaluated.

Since denting has been confirmed, an enhanced primary-to-secondary leak rate monitoring program will be implemented as described in paragraph C.l.

NRC BULLETIN 88-02 March 28, 1988 Page 2 of 10 SCE ACTIONS COMPLETION DATES

1) Review prior steam generator 3-11-88 inspection data for denting.

(review complete)

B. Future Inspections (Paragraph B of NRC Bulletin 88-02)

"For plants where no denting is found at the uppermost support plate, the results of future steam generator tube inspections should be reviewed for evidence of denting at the uppermost support plate. If denting is found in the future, the provisions of item C below should be implemented. Commitments to implement these actions shall be submitted when the results of A above are submitted."

SCE Response Concerning Future Inspections This action item is not applicable to San Onofre Unit 1 as denting has occurred in this unit.

C.1. Enhanced Monitoring Program (Paragraph Cl of NRC Bulletin 88-02)

"For plants where denting is found, the NRC staff requests that the following actions be taken:

1. Pending completion of the NRC staff review and approval of the program described in C.2 below or completion of inspections specified in item A above to confirm that denting does not exist, an enhanced primary-to-secondary leak rate monitoring program should be implemented as an interim compensatory measure within 45 days of the date of receipt of this bulletin.*

Implementation of this program shall be documented as part of the 45-day report. The enhanced monitoring program is intended to ensure that if a rapidly propagating fatigue crack occurs under flow-induced vibration, the plant power level would be reduced to 50% power or less at least 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> before a tube rupture was predicted to occur. The effectiveness of this program should be evaluated against the assumed time-dependent leakage curve given in Figure 1.

This program should consider and provide the necessary leakage measurement and trending methods, time intervals between measurements, alarms and alarm setpoints,intermediate actions based on leak rates or receipt of alarms, administrative limits for commencing

NRC BULLETIN 88-02 March 28, 1988 Page 3 of 10 plant shutdown, and time limitations for (1) reducing power to less than 50% and (2) shutting down to cold shutdown. Appropriate allowances for instrument errors should be considered.

Finally, the program should make provision for out of service radiation monitors, including action statements and compensatory measures.

• While this bulletin was being prepared, licensees for a few plants committed to an enhanced primary-to-secondary leak rate monitoring program at the staff's request. These plants have been identified on a preliminary basis by Westinghouse as being potentially susceptible to rapidly propagating fatigue cracks. These enhanced programs should be upgraded as necessary to comply with this paragraph.

However, no relaxation from current commitments should be made without prior approval by the NRC staff."

San Onofre Unit 1 Program For Enhanced Primary-To-Secondary Leak Rate Monitoring SCE will formalize and implement an enhanced primary-to-secondary leak rate monitoring program upon return to service from the present mid-cycle outage (tentatively scheduled for April 29, 1988).

The enhanced monitoring program will meet the criteria for being able to monitor a tube failure with leakage characteristics similar to that which occurred at North Anna and take appropriate actions prior to rupture of the tube. The following provides details of the San Onofre Unit 1 enhanced monitoring program.

(a) Initial Monitor Setpoints The initial alarm setpoint for the air ejector monitor (R-1215) will be established prior to returning the unit to criticality. This alarm setpoint will correspond to the activity expected from a primary-to-secondary leak rate of 15 gallons per day (gpd).

Due to design and operational difficulties with R-1215, it has been out-of-service since July, 1986.

Presently, the air ejector monitor is undergoing extensive maintenance based on a review of its outstanding deficiencies and is expected to be operable prior to return to service from the present outage. If the air ejector monitor can not be declared operable, or it becomes inoperable after return to service, the steam generator blowdown monitor (R-1216) will be utilized to establish trending and the enhanced monitoring program as specified in section (e),

"Actions For Air Ejector Monitor Out-Of-Service," will be implemented.

NRC BULLETIN 88-02 March 28, 1988 Page 4 of 10 (b) Monitoring Leak Rate Data During Power Operations During full power operation and with primary-to-secondary leak rates less than 15 gpd, the San Onofre Unit 1 leak rate monitoring program will consist of the following multi-faceted approach: 1) continuous monitoring utilizing R-1215 and R-1216, 2) radiochemistry analyses, 3) leak rate calculations, and

4) trending of data. The activity indications from the continuous monitors will be trended and reviewed once-a-day. The air ejector off gas flow rate will be logged and reviewed each day to assess changes in condenser air inleakage. Primary-to-secondary leak rates will be calculated daily as a function of the air ejector activity readings, air ejector flow rate, and the most recent primary coolant gaseous activity. The primary coolant total gaseous activity is normally measured three times per week for use in determining primary-to-secondary leak rate.

(c) Actions In Response To An Increasing Leak Rate If an alarm is received from the air ejector monitor or calculations of primary-to-secondary leak rate show that the leak rate is increasing, confirmation of the change will be obtained by other methods. These will include, but are not limited to, reviewing the trends of the continuous blowdown monitor, or performance of radiochemistry analyses to allow independent calculations of the primary-to-secondary leak rate.

If evaluation shows that the primary-to-secondary leak rate is greater than 40 gpd or the leak rate has increased more than 15 within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and continues to increase, the frequency of evaluation of the leak rate will be increased to at least once per shift. The evaluation will consider the possibility that a tube plug or sleeve may have caused the increased leak rate. Unlike the continuous increasing leak rate that occurred at North Anna, the leak rate from a defective plug or sleeve should stabilize and may even decrease.

The alarm setpoint for the air ejector monitor will be increased to a value which corresponds to approximately a 15 gpd increase over the latest leak rate.

(d) Actions For Large Leak Rate If the leak rate approaches 100 gpd, an evaluation of the need to reduce power or commence shutdown will be performed. This evaluation will consider the leak rate

NRC BULLETIN 88-02 March 28, 1988 Page 5 of 10 data to date, the rate of change of the leak rate, and the leak rate measurement uncertainty. This evaluation will provide adequate time to reduce power or shutdown should a tube fatigue failure be in progress.

Implementation of this program will be more conservative than would be required by the leak rate-time curve generated from the North Anna event.

(e) Actions For Air Elector Monitor Out-Of-Service If the air ejector monitor is out-of-service, several methods will be utilized to ensure that primary-to-secondary leak rates continue to be monitored. These methods will include, but are not limited to, daily secondary coolant and air ejector off gas activity analyses, and review of trending data utilizing R-1216, Steam Generator Blowdown Monitor.

The first gas samples collected will be evaluated within 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> of when the air ejector monitor is declared out-of-service. Thereafter, samples will be collected daily when leak rates are less than 40 gpd and shiftly when leak rates are greater than 40 gpd.

The importance of monitoring primary-to-secondary leak rates while the air ejector monitor is out-of-service is recognized and it is intended to continue trending and review of other pertinent data to ensure any changes in leak rate are promptly evaluated.

(f) Actions To Enhance The Air Elector Monitor SCE intends to closely evaluate the performance of the air ejector monitor and, if required, implement to the extent practical design changes during the Cycle 10 refueling outage scheduled to begin July, 1988 which will enhance the reliability and operability of the air ejector monitor.

(g) Training On The Enhanced Monitoring Program And The North Anna Event Training on the basis for the San Onofre Unit 1 enhanced primary-to-secondary leak rate monitoring program will be conducted for all on-shift licensed reactor operators prior to return to full power operation from the current mid-cycle outage. In addition, Chemistry personnel will be trained on the methods utilized in the San Onofre Unit I enhanced monitoring program prior to return to full power from the current mid-cycle outage. Additional operator training on the North Anna Tube Rupture event is targeted for implementation the last half of 1988.

NRC BULLETIN 88-02 March 28, 1988 Page 6 of 10 Presently, the Unit 1 Operator Requalification Training program has included both steam generator leaks and tube ruptures while at the simulator'. In the simulator, there have been on the average two Steam Generator Tube Rupture (SGTR) scenarios used per year for each of the last 4 years. During these scenarios, the operators use the San Onofre Unit 1 specific Emergency Operating Instructions (EOI's) and the same indications are available, e.g., the air ejector monitor was removed at the simulator to train in case the air ejector monitor was taken out-of-service at the plant. The EOI's are covered in the formal classroom training on a recurring basis. Additionally, the initial operator training program consists of both classroom and simulator training on SGTR's. Therefore, SCE considers the training aspects to be sufficient for the operators to capably respond in the eventuality of such an event to the same high degree of efficiency as experienced by the operators at North Anna.

SCE ACTIONS COMPLETION DATES

1) Formalize and implement End of mid an Enhanced Monitoring cycle outage Program (4-29-88)*

2) Training of licensed operators End of mid and Chemistry technicians cycle outage on Enhanced Monitoring Program (4-29-88)*

3) Training on North Anna Tube Rupture Event 12-31-88 Tentative date C.2. Program To Minimize Probability Of Rapidly Propagating Tube Failure (Paragraph C.2 of NRC Bulletin 88-02)

"2. A program should be implemented to minimize the probability of a rapidly propagating fatigue failure such as occurred at North Anna Unit 1. The need for long-term corrective actions (e.g., preventive plugging and stabilization of potentially susceptible tubes, hardware, and/or operational changes to reduce stability ratios) and/or long-term compensatory measures (e.g., enhanced leak rate monitoring program) should be assessed and implemented as necessary. An appropriate program would include detailed analyses, as described in subparagraphs (a) and (b) below, to assess the potential for such a failure. Alternative approaches and/or compensatory measure implemented in lieu of the actions in subparagraphs (a) and (b) below should be justified.

NRC BULLETIN 88-02 March 28, 1988 Page 7 of 10 Although the 45-day report shall provide a clear indication of actions proposed by licensees, including their status and schedule, a detailed description of this program and the results of analyses shall be submitted subsequently, but early enough to permit NRC staff review and approval prior to the next scheduled restart from a refueling outage. Where the next such restart is scheduled to take place within 90 days, staff review and approval will not be necessary prior to restart from the current refueling outage. An acceptable schedule for submittal of the above information should be arranged with the NRC plant project manager by all licensees to ensure that the staff will have adequate time and resources to complete its review without adverse impact on the licensee's schedule for restart.

(a) The analysis would include an assessment of stability ratios (including flow peaking effects) for the most limiting tube locations to assess the potential for rapidly propagating fatigue cracks.

This assessment would be conducted such that the stability ratios are directly comparable to that for the tube which ruptured at North Anna.

(b) The analysis would include an assessment of the depth of penetration of each AVB. The purpose of this assessment is twofold: (1) to establish which tubes are not effectively supported by AVBs and (2) to permit an assessment of flow peaking factors.

(Note: Most steam generators have at least two sets of AVBs. This applies only to the set that penetrates most deeply into the tube bundle.)

The methodology used to determine the depth of penetration of each individual AVB shall be described in detail in the written report. The criteria for determining whether a tube is effectively supported by an AVB shall also be identified. (Note: An AVB that penetrates far enough to produce an eddy current signal in a given tube may not penetrate far enough to provide a fully effective lateral support to that tube.)"

NRC BULLETIN 88-02 March 28, 1988 Page 8 of 10 San Onofre Unit 1 Program To Minimize Probability Of Rapidly Propagating Tube Failure The San Onofre Unit 1 program to minimize the probability of a rapidly propagating fatigue failure such as occurred at North Anna Unit 1 includes the following elements:

(a) Eddy Current Testing Eddy current testing (ECT) data has been obtained on all tubes in service in rows 11 through 17 in all three steam generators.

(The nominal antivibration bar [AVB]

position is row 14.)

The ECT data will be analyzed to identify all of the tubes in this region of interest that are dented. The ECT data will also be evaluated to determine the number of AVB's in contact with each tube and the position of the contact points/arcs relative to the uppermost tube support plate. This AVB contact information will be used to determine the position of each AVB relative to the tubes it supports (the depth of penetration of each AVB). This effort will include consistency checks and independent verification of AVB position through geometric scaling techniques as required.

(b) Thermal/Hydraulic Analysis ATHOS, a three dimensional flow analysis code, will be used to calculate flow conditions in the regions of interest. A model will be set up with geometric parameters specific to San Onofre Unit 1. Detailed velocities, densities and void fractions for current operating conditions will be calculated.

(c) Flow/Vibration Testing Hind tunnel tests will be performed to determine the effects on fluidelastic instability of columnwise variations in AVB insertion depths. The critical velocities for fluidelastic instability will be obtained to determine the.effects of AVB stagger on the instability. Peaking factors obtained will be expressed as a ratio of critical velocities among various AVB configurations, including the tube which ruptured at North Anna.

NRC BULLETIN 88-02 March 28, 1988 Page 9 of 10 (d) Tube Vibration Analysis A dynamic analysis of the tubes in the regions of interest will be performed using the FLOVIB computer code based on the results of the thermal hydraulic analysis. The.vibration analysis will determine tube stiffness, frequency and fluidelastic stability ratios for these tubes.

(e) Identification of Potentially Susceptible Tubes A stress ratio criteria will be developed based on the upper bound stress amplitude that could have been responsible for the failure of the tube which ruptured at North Anna. The AVB position information, the flow peaking factors, the results of the tube vibration analysis and the impact of past operation will be combined to determine which, if any San Onofre Unit 1 steam generator tubes exceed this stress ratio criteria.

(f) Long-Term Corrective Actions Long-term corrective actions will be developed depending on the results of the preceding analysis. If potentially susceptible tubes are identified, a variety of corrective actions (e.g., preventive plugging/stabilization of potentially susceptible tubes, hardware and/or operational changes to reduce stability ratios, etc.) and/or long-term compensatory measures (e.g., enhanced leak rate monitoring programs) will be considered and to the extent practical will be implemented during the Cycle 10 refueling outage scheduled to start in July 1988.

(g) Schedule For Implementing SCE Program To Minimize Probability Of Rapidly Propagating Tube Failure The following schedule is applicable to resolution of the issues identified for San Onofre Unit 1.

SCE ACTIONS COMPLETION DATES

1) Completion of evaluation of ECT data 4-1-88

2) Identification of potentially susceptible tubes 5-13-88

3) Identification of long-term corrective actions (if required) 6-24-88

NRC BULLETIN 88-02 March 28, 1988 Page 10 of 10 SCE ACTIONS COMPLETION DATES

4) Submittal of analysis results and planned corrective actions to the NRC 7-15-88

5) Completion of the San Onofre Unit I cycle 10 refueling outage 10-15-88 0879P