ML20217K008

Rev 1 to Evaluation of SONGS Unit 3 SG Eggcrate Support Condition-Cycle 9

ML20217K008
Person / Time
Site:	San Onofre
Issue date:	10/16/1997
From:	SOUTHERN CALIFORNIA EDISON CO.
To:
Shared Package
ML20217J936	List: ML20217J931 ML20217K008 ML20217K025
References
NUDOCS 9710220352
Download: ML20217K008 (45)

Text

-.. -. _.. -. -

l i

l 4

e Evaluation of the I

San Onofre Unit 3 Steam Generator Eggerate Support Condition - Cycle 9 Revision 1 October 16,1997 1

Southern California Edison Company JE kri;!!!!'h solmitsNcAurorNIA EDISO N Angi. YIXE?.NA170XAL Coeyary 9730220352 971017 DR ADOCK 0500 2

1 Evaluation of the San Onofre Unit 3 Steam Generator Eggerate Support Condition - Cycle 9 Revision 1 EXECUTIVE

SUMMARY

i l

1.0 INTRODUCTION

2.0 METIIODOLOGY 2.1 Evaluation Plan i

2.2 Cause Assessment Plan 2.3 Analysis Program i

2.4 Video Inspection Program 2.5 Repair Options 2.6 Integrated Schedule 3.0 CAUSE ASSESSMENT, ANALYSIS, AND INSPECTION PROGRAMS 3.1 Cause Assessment 3.2 Analysis 3.2.1 Summary and Conclusions 3.2.2 Single Tube Analysis 3.2.2.1 Normal Operating Conditions 3.2.2.2 Loss of Coolant Accident (LOCA) and Safe Shutdown Earthquake (SSE) 3.2.2.3 Main Steam Line Break (MSLB) and SSE 3.2.2.4 Flow Induced Vibration (FIV) Normal Operation 3.2.2.5 Flow Induced Vibration - Main Steam Line Break 3.2.3 Eggerate Evaluation 3.2.3.1 Main Steam Line Break (MSLB) with Safe Shutdown Earthquake (SSE) 3.2.3.2 Loss of Coolant Accident (LOCA) with Safe Shutdown Earthquake (SSE) 3.3 Steam Generator Field Inspe::tions 3.3.1 Methods / Equipment 3.3.2 Steam Generator General Inspection (Internals) 3.3.3 Eggerate -Inner Bundle 3.3.4 Eggerate - Batwing / Vertical Strips 3.3.5 Eggerate - Periphery October 17,1997 (10:33am) i

a

,+4.-a W

_-,++ -

2 y

_em 4,

.e u

4.

A e

3.3.6 Eggerate - Blowdown Lane

- 3.3.7 Eggerate/ Ring - Stay cylinder 3.3.8 Steam Generator Chemical Cleaning (SGCC) Gauging - First Eggerate 3.4 Repairs and Repair Criteria 3.4.1 Tube Plugging Criteria for Eggerate Degradation 3.4.1.1 Implementation of Eggerate Grading 3.4.1.2 Specific Plugging Criteria 3.4.2 Plugging / Stabilizing 3.4.3 Plugging / Stabilizing Locations / Recommendations 3.5 Tube Support Wear and Eggerate Degradation 3.6 References 4.0

SUMMARY

OF CONCLUSIONS 4.1 Cause Assessment Conclusions 4.2 Analysis Conclusion 4.3 InspectionResults 4.4 Continued Operation 4.5 Effect and Benefit of SGCC 4.6 Loose Parts Evaluation 4.6.1 Existing Loose Parts 4.6.2 Potential Loose Parts 4.7 10CFR50.59 Safety Evaluation 5.0 RETURN TO SERVICE RECOMMENDATIONS / CORRECTIVE ACTIONS 6.0 POST CYCLE 9 RECOMMENDATIONS / CORRECTIVE ACTIONS

-7.0 CHARTS, GRAPIIS, AND VISUAL AIDS (Figures) 8.0 APPENDICES A.

Description of the San Onofre, Units 2 & 3 Steam Generator Eggerate Design i

and Condition B. Eggerate Evaluation Task Force (EETF)

C.

Steam Generator Tube Eggcrate Support Evaluation Operability Assessment, San Onofre Unit 2 ii

D., Evaluation of Southern California Edison SONGS Unit 3 Steam Generators With Degraded Eggerates, A SONGS-9416-1168, latest revision (on file in l

CDM SO23 915 208) l E. - Cause Assessment and Corrective Action Report F.

Response to Specific NRC Questions Regarding San Onofre Unit 3 Steam Generator Eggerate Degradation G.

10CFR50.59 Safety Evaluation, Rev 1 l

iii

EXFLUTIVE

SUMMARY

During the Cycle 9 refueling outage for San Onofre Unit 3, degradation of periphery portions of several eggerate supports was identified in both steam generators. The degradation was noted during pre-chemical cleaning assessments of the condition of the eggerate supports and was unexpected. This report provides Edison's assessment of the cause of the degradation, its impact on continued operation, the actions needed to arrest it, and the plans for periodic monitoring.

Edison performed a comprehensive inspection ofinternals of both Unit 3 steam generators. This inspection confirmed the degradation was primarily limited to the upper eggerates and was confined to their peripheral ponions. To a lesser extent the stay cylinder, blowdown lane areas, and cold leg portions of the eggerate periphery were also affected.

The degradation was caused by a form of Flow Accelerated Corrosion (FAC), a general term describing processes which use assistance from fluid flow to remove the protective oxide layer from base material. Removal of the protective oxide layer exposes the base material to the fluid environment allowing further material removal via cmrosion and/or erosion processes. Relatively rapid material removal rates may occur it ' le presence of certain combinations of fluid and material properties. The carbon steel eggerate material can be susceptible to FAC in the presence of sufficiently high fluid velocities in l

combination with high steam quality.

l The FAC apparently occurred during recent operation of Unit 3 as a result of steam generator secondary side fluid parameter changes caused by the buildup of deposits on the steam generator tubes. This buildup led to increased fluid velocities and steam quality particularly in the tube bundle periphery areas. After sufficient buildup, secondary fluid properties reached levels where FAC occurs.

The chemical cleaning of all steam generators restores fluid properties to nominal I

conditions by removing the deposit buildup Under nominal conditions, fluid parameters are not conducive to FAC and the condition is arrested. Due to impovements in controlling the sources of deposits, the rate of deposit buildup will be much less in the future, consequently fluid properties conducive to FAC are not expected to recur.

An extensive re analysis of the eggerates in their degraded condition was conducted.

Since the upper most (10th) eggerate was found to be in good condition and the other tube support members (batwings and vertical strips) were unaffected by the FAC process, the analysis results were acceptable for all accident conditions. Although excessive flow induced vibration has not occurred, Edison has conservatively plugged those tubes with two or more consecutive uncredited eggerate supports. About 115 tubes in the two steam generators were plugged as a result. Internal stabilizers were added to each affected tube prior to plugging to ensure tube vibration would not affect adjacent in-service tubes.

1

The analysis showed that substantial margin remains despite the eggerate degradation as -

can be seen in this table.

Key Analytical Margins Evaluation Margin Flow induced vibration normal operation Over 35% margin in stability ratio e-e Tubes plugged with no wear indications LOCA & SSE - accident condition -

e Meets ASME Code allowables with Reg. Guide 1.121 tube thinning of 64 %

Stress for maximum expected wear is e

40% of ASME Code allowables.

Edison performed a written safety evaluation to verify that an unreviewed safety question would not result from having degraded eggerate supports in the Unit 3 steam generators.

Given these measures have arrested further FAC and stabilized the eggerate degradation, operation over the normal refueling inspection interval would be acceptable. However, for conservatism, Edison has determined a special mid-cycie inspection of the eggerate

~

condition will be performed to confirm this assessment. This mid-cycle will be scheduled to occur at approximately the mid-point of the operating cycle (10-12 effective full power months). If effectiveness is confirmed during this mid-cycle, sufficient basis will exist to monitor any progression of the degredation on a refueling interval bash.

2

1.0 INTRODUCTION

From early in the San Onofre Nuclear Generating Station Units 2 and 3 operating history, a significant downward trend in main steam pressure at full power has been observed.

This downward trend was observed in both San Onofre units, but has exhibited a steeper rate ofdegradation in Unit 3. By 1991, this loss in secondary pressure had limited full power capability. In an effort to recover secondary pressure and to mitigate newly recognized Steam Generator (SG) tube failure mechanisms, Steam Generator Chemical Cleaning (SGCC) was planned for the Cycle 9 refueling outages.

As part of the SGCC process, inspection of the SG intemals was performed to establish conditions prior to Chemical Cleaning. These inspections specifically included remote visual examination of the tube bundle "eggerates". Due to limited accessability, the bulk of eggerate inspection was performed around the outer annulus between the tube bundle and the feedwater baffle. Figure 7.1-1 is a cut-away depicting steam generator internals.

No unusual indications were observed during the Unit 2 inspections. However, inspection of Unit 3 eggerates found unexpected degradation predominately in the inspection areas of the outer annulus. Following SGCC, an extensive inspection of the eggerates and tubing supports was performed. These inspections confirmed the results of the pre-SGCC effort. Of the tubing support structures, only the eggerates have been affected.

A special task force was established by Edison to effectively evaluate the as found condition. Appendix B describes the composition of the Eggerate Evaluation Task Force (EETF), its mission, goals and consulting support.

1-1 i

2.0

- METHODOLOGY Edison developed a plan and evaluation methodology to provide a roadmap for a complete and thorough assessment of the Unit 3 SGs and the implications to the Unit 2 SGs. -

2.1 Evaluation Plan Edison developed a three prong approach to the evaluation: analysis of the SG condition, cause and corrective action assessment, and identification of possible repair options.

The cause and couective actions includes a fact finding and cause assessment plan similar to that used to resolve plant non-conformances, an operability assessment germane to Unit 2, review of other industry data (including European) with regard to SG internals degradation, and input and review from recognized experts in the fields of flow induced vibration and flow accelerated corrosion.

The analysis approach includes: specific SG performance analysis, flow and stress calculations, verification of analysis assumptions (inspections), and short and long term actions necessary to ensure the SG condition remains bounded by the analysis and its assumptions (including substantive inspection).

The plan for potential repairs included a review of all available short and long term repair l

options and an evaluation of the feasibiFty of repair options and implementation methodologies for anticipated degradation possibilities and internals locations.

The followiag sections provide additional detail for the three plan activities identified 4

above.

2.2 Cause Assessment Plan The cause and corrective action determination was developed with engineering assistance from Chemistry, Mechanical, Inservice Inspectica and Independent Safety Engineering divisions. Edison staffin these disciplines was augmented and independently checked using support from Dominion Engineering, APTECH Engineering and ABB Combustion Engineering services.

The cause assessment effort was initiated from detailed evaluation of the degradation surfaces, distribution, and other unique features of the Unit 3 condition. Based on these features, a number of scenarios were postulated and evaluated to identify candidate explanations. Analytical evaluation of selected candidate scenarios was performed to assess consistency with observed degradation features. This process resulted in identification of a primary cause and potential contributing factors. Roots of the primary cause and contributing factors were identified for corrective actions and forect.sts of 2-1

future eggerate performance were made based on both expected and worst case success of the corrective actions.

2.3 Analysis Program An analytical a3sessment of the eggerate was necessary in order to develop a basis for evaluating the degraded eggerates. This consisted of confirming the design t. asis requirements of the steam generato s are met, determining acceptance criteria for the inspections, and determining repair criteria such as plugging / stabilizing if required. The methodology that was used to accomplish this was to update the analyses with revised geometry to reflect the thinned eggerate condition, determine the impact on all existing analyses, determine the need for new analyses and update or generate new analyses as required.

Discussion of design basis requirements, acceptance criteria for the inspections and repair (plugging / stabilizing) criteria is provided in this report. Details of updating the eggerate geometry to address thinned lattice bars and the impact on the analytical results is provided in Appendix D. The existing analyses were updated and as required additional assessments were made within the existing analyses to ensure all aspects of eggerate degradation were evaluated. No new analyses were deemed necessary as part of this work. The original analyses which covered evaluation of design basis accidents and normal operation to ensure that steam generator tube integrity is maintained included all the necessary conditions to address the eggerate degradation.

2.4 Video Inspection Program After assessing preliminary information from the pre-chemical cleaning inspection, a plan to perform a thorough and comprehensive inspection of the eggerates after the chemical cleaning was developed. The desired outcome of the inspection was to have enough visual information available to allow a complete assessment of the eggerate material conditions, provide a baseline to support future inspection and to validate the assumptions of the analysis.

The methodology used to conduct the comprehensive inspection consisted of three phases.

Acquisition Analysis Documentation The first component was the video acquisition portion. This was the actual field work associated with obtaining the video documentation of the steam generator eggerates. The following six areas were examined in some detail:

2-2 l

l

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

'o General Inspection 0-Inner Bundle

[

Batwings and Vertical Strips-t Eggerate Periphery j-

- Blowdown Lane i

Stay Cylinder

!=

The second element was the analysis of the tapes made during the acquisition phase. The assessment team's responsibility was to review each of the tapes and document the status of the eggerates on " maps" prepared for the periphery inspection of each eggerate. Each

~

video was reviewed with a primary viewer and then rechecked by a secondary viewer.

2.5 Repair Options Edison utilized in house resources in both engineering and the construction groups who f

have performed SG intemals rnodifications to the feedring and feedring distribution box in past outages including the on site construction contractor (Bechtel). To provide a i

broader base of repair options, Edison also obtained the services of the SG replacement i

organization from the Bechtel Gaithersberg offices.

I l~

Followng preliminary review ofinspection data, review of cause and corrective actions j

and analysis results, it became clear that none of the major construction repairs would be e essary. In fact tube plugging and stabilizing, supplemented with inspections (all ble of being accomplished with existing and proven technologies) were considered to

ae probable selected repair options. As such, further pursuit of detailed naplementation plans and evaluations for the major repair techniques was terminated.

2.6 Integrated Schedule Figure 7.2-1 shows the updated integrated inspections, schedule and logic developed and utilized by Edison to ensure all Edison analysis, cause assessment, corrective action, and repair activities were identified completed, and integrated into the overall outage effort to effectively support the evaluation and issue the final report.

2-3

3.0 CAUSE ASSESSMENT, ANALYSIS, AND INSPECTION PROGRAMS -

3.1 Cause Assessment Details of the assessment of the cause and determination of corrective actions designed to stabilize eggerate degradation at San Onofre is discussed in detail in Appendix E. The following provides a sununary of the inputs, process and results developed in that evaluation.

Background

Examination of the damaged eggerates from Unit 3 identified substantial thinning and a scalloped surface indicative of Flow Accelerated Corrosion (FAC). Flow Accelerated Corrosion involves a material loss which is perpetuated by mechanical removal of the protective corrosion products or base metal by fluid flow. The rate of material loss is a function of variables including temperature, pH, fluid velocity, material properties and steam quality. As evidenced by acceptable operation of Unit 2, the combined effect of these variables is not expected to result in significant material loss under normal operating conditions. This resistance to FAC during operation is due to the increased toughness of the protective oxide magnetite above 450 F.

As predicted from integrated iron transport and steam pressure loss analyses, the extent of fouling in Unit 3 was observed to be significantly greater than that of Unit 2. However, the degraded eggerate surfaces appeared to be free of fouling deposits which suggests that FAC was active during recent operation. In addition, areas of substantial thinning were found in close proximity to undamaged material on the same lattice bar suggesting that -

neither material properties nor secondary chemistry played the dominant role in the Unit 3 degradation process.

Review of related industry events identified only one facility, the EdF plants at Gravelines, which have experienced a similar degradation process. At Gravelines, two areas of the upper Tube Support Plate (TSP) periphery have undergone a visually similar degradation process. These areas are approximately 180 degrees apart. The failure mode at Gravelines was concluded to be from Flow Accelerated Corrosion which was initiated by the use of ammonia for pH control and possibly enhanced by the Hydrazine used for O control. (Similar EdF plants using morpholine pH control do not have this TSP 2

degradation). As a corrective action, the Gravelines plants have implemented morpholine pH control. Use of morpholine appears to have reduced, but not eliminated the TSP degradation. EdF is in the process of replacing the Gravelines steam generators for other reasons. The Gravelines plants have not undergone chemical cleaning.

Cause Determination Important characteristics of the San Onofre eggerate degradation include eroded surface features, the distribution of degradation and the lack of degradation in Unit 2. These 3-1

characteristics define the correct cause. In the cause determination process, a number of postulated scenarios were evaluated for consistency with these charact.:ristics. Of the scenarios postulated, the primary cau:;e was determined to be that the elevated fouling of Unit 3 steam generators significantly reduced the available flow area within the tube

-bundle which caused flow diversion to the periphery of the eggerate. The resulting large velocity gradients at the periphery initiated vortices which further elevated local velocities that are capable of dislodging the protective magnetite layer and initiated erosive FAC.

This cause conclusion supports a mechanism for creating the necessary fluid velocities to overcome resistance to FAC at elevated temperatures. The resulting velocities are consistent with the erosive features of degraded surfaces. The scenario provides an -

explanation why Unit 2 is unaffected which is supported by significant differences in total deposit removal from the two units. The scenario is consistent with the observed l

distribution of damage around the tube bundle periphery and the observr. tion that the l

degradation process has been recently active.

ATHOS Comouter Model The thermal hydraulic code ATHOS was used in the cause evaluation process to assess the relative response of steam generators to various thicknesses of uniform eggerate fouling. Bulk velocities and steam qualities from ATHOS were manually adjusted for local geometric effects at the eggerate to determine average conditions at the tube bundle interface. In the fouled condition, there calculated average velocities are barely sufficient to initiate FAC, however the predicted velocity gradients at the perirhery would give rise to flow vortices which could easily initiate erosive FAC. Contributing to this cause are the potential for entrained particulates from Icose deposit material and potential effects of hydrazine on magnetite.

Details of the nature and distribution of deposits in Unit 2 and Unit 3 would significantly affect steam generator dynamics. However, these details are unknown and, therefore, could not be modeled in ATHOS. Consequently, ATHOS could identify trends and relative areas of concem, but could not be used to sharply contrast the Unit 2 and Unit 3 conditions relative to FAC. For this reason, it is emphasized that the cause is based largely on qualitative assessments and consistency with unique features of the Unit 3 condition rather than from fundamental analysis.

Corrective Actions Although San Onofre has used ammonia fcr pH control as was the case at Gravelines, use of ammonia and hydrazine in chemistry control are not considered to be the primary

- causal factors because of the acceptable performance of Unit 2 and FAC evaluations which indicate that normal flow velocities in the San Onofre design are well below the threshold where aggressive FAC could be expected.

3-2 i

Based on the identified primary cause, steam generator chemical cleaning is expected to arrest the degradation process in Unit 3. Chemical cleaning can be expected to restore flow margins such that initiation of FAC will not be a near term concem. In addition to chemical cleaning and consistent with the EdF corrective measures, transition from ammonia to ETA for pH control is also being implemented along with an evaluation for optimizing the use of hydrazine.

Post Cvele 9 Oneration Based on significant reductions in feedwater iron transport it is projected that the deposit loading which might accumulate by 2023 will not reach 50% of the loading which existed in Unit 2 prior to chemical cleaning. On the basis that Unit 2 did not experience significant FAC, it is anticipated that a subsequent chemical cleaning for the purpose of avoiding FAC will not be required.

3.2 Analysis 3.2.1 Summary and Concludons An analytical assessment of the degraded eggerates was performed in order to develop a basis for accepting the condition of the degraded eggerates. The results demonstrated that with plugging and stabilizing of selected tubes, the eggerate degradation is within the existing acceptance criteria.

To account for the eggerate degradation during normal operating conditions, some tubes have been conservatively plugged and stabilized to prevent potential leaks from occurring due to wear from flow induced vibration (FIV). The majority of the tubes plugged were shown to be analytically acceptable but plugged to increase the margin. Ninety-two percent (92%) of the tubes plugged for eggerate degradation d.id not show any signs of wear from the Cycle 9 eddy current testing results. The remaining 8% of the tubes that were plugged for eggerate considerations did not show wear in the locations of the degraded eggerates. The wear for these tubes was found in the vertical supports, batwings and cold leg eggerates as discussed in Section 3.5. A discussion of the tube plugging criteria is provided in Section 3.4. No further modifications are required to ensure that the design basis of the steam generators is maintained.

For accident conditions, the results of the degraded eggerate analyses show that eggerate degradation is an acceptable condition for the steam generator. That is, eggerate degradation will not compromise steam generator tube integrity during an accident, and will maintain structural integrity so as not to damage any tubes during accident conditions.

Updated analyses were performed by the original equipment manufacturer, ABB-CE.

Independent verification of the analyses was performed by SCE and consultants AEA Technology & Engineering Service, MPR Associates, and Dr. Robert Blevins. The 3-3

- ~

consultants independently generated assumptions, methodologies and acceptance criteria and in some cases performed independent analyses to confirm the results. The ABB-CE analyses were reviewed by the consultants and comments resolved prior to issuing the analyses.

Margins and conservatisms associated with each type of analysis are provided in the following sections. These margins are associated with the current condition of the degraded eggerates. The limiting nargin for the accident condition for maximt'm tube wall thinning of 64% is bounded by existing Regulatory Guide 1.121 burst pressure limitations, and is not impacted by the eggerate degradation. Assuming a tube thinning of 39% results in a margin of 40% in the accident stress between Cycle 9 and the mid-cycle outage. The margin for the normal operation FIV is over 35%. These margins are intended to quantitatively demonstrate the conservatism used in this analysis but do not represent all the potential conservatism that have been included.

The analytical tools and methodologies used to perform the dc;;raded eggerate calculations are the same or similar to the tools used to perform the original design basis.

Detailed results are provided in Table 3.2-1 which describes the original design basis of the eggerates. The table was sent to the NRC in Reference 6 of Section 3.6. It has been updated to include the results of the degraded eggerate evaluations. The remainder of the discussions in this section are organized in the order presented in the table, i

3-4

Tchin 3.2-1 -S=n Oxfre Unit 3 Stecm Gescrctrr Eggerntz-Desig2 Brsis A= clysis Summ ry Report Section Design Basis Analysis of Record Acceptance Analysis of Record Degraded

'l Criteria

. Quantitative Eggerate Results Quantitative Results 3.2.2.1 Normal Operating SingleTube Analysis ASME Section III, Conditions and Upset Class I (1971 Edition Condition with Addenda through Summer 1971):

a) Primary local a) 12 ksi a) 22 ksi, including membrane plus bending Operating Basis

<l.5S. = 35 ksi Earthquake (OBE) b) Primary plus b) 173 ksi b)273 ksiincluding secondary SI range OBE

<3S. =69.9 ksi c) Fatigue usage factor c) U-0 c) No Change U<l.0 3.2.2.2 Loss ofCoolant Single Tube Analysis ASME Section III, Limiting tube row 147, Limiting tube row 120 Accident (LOCA) with Class I (1971 Edition w/64% thickness w/64% thickness Safe Shutdown with Addenda through reduction:

reductkm 80.4 ksi(5)

Earthquake (SSE)

Summer 1971):

f,(0.7 S.) = 80.6 ksi 69.9 ksi(1) 3.2.23 Main Steam Line Break Single Tube Analysis ASME Section III, Limiting tube row 25:

No change (Bounded

}

(MSLB) with SSE Class 1 (1971 Edition 21.9 ksi(1) by i

with Addenda through LOCA & SSE)

Summer 1971):

f,(0.7S.) = 80.6 ksi f

I 3-5 i

?

i

Report Section Design Basis Asclysis af Record Acceptnee A:alysis of Record Degraded Criteria Quantitative Eggerate Results '

Quantitative

' Results 3.2.2.4 Normal Operation - -

SingleTube Analysis' a) Free stream velocity a) VNa <=0.45 -

a) VNem <=0.64 for l'

Flow Induced Vibration (V) < critical velocity one eggerate uncredded

.l (V,)

a) Cross flow VNa <=0.66 for -

l' b) Parallel flow alternate eggerates l

c) RCP impeller uncredited l'

i d) RCP pressure pulse due to RCP impeller vane b) Midspan b) Displacement <=

b) Displacement '

interaction displacement <0.0625" 0.0022"

<=0.0027" c) SG tubes meet code c) fn.= 49.2 cps (t 0) c) 12.7 < fn < 40 cps l

endurance limit, (6)

S<Sa< fatigue l

l t

endurance limit l

i l

f d) Alternating stress d) Alternating stress =

d) Resulting stress l

[

due 0.958 ksi(limiting tube tw-pble. U=0 l

[

~

to pressure pulse good row 25; U=0) g

{

for infinite number of l

~

cycles (U=0) l 4

3.2.2.5 MSLB Flow Induced SingleTube Analysis Free stream velocity V/V, <=0.47 No ' change (7)

Vibration Loads -

(V)

[

<= critical velocity

(

(V )

f i

3.2.3.1 Main Steam Line Break Eggerate Evaluation Maintain tube integrity Tube integrity Tube integrity j

(MSLB) with Safe (2) maintained maintained Shutdown Earthquake (2)

(8)

(SSE) i 3-6

i Report Section Design Basis

. An lysis cf Record Acceptance Analysis of Record Degraded i

Criteria Quantitative Eggerate Results '

Quantitative Results 3.2.3.2 Loss of Coolant Eggerate Evaluation Maintain tube integrity Tube integrity Tube integrity Accident (LOCA) with (Whole Bundle)

(4) maintained maintained Safe Shutdown (3)

(4)

(9)

Earthquake (SSE)

NOTES: (1) Stress value shown varies from the value shown in the UFSAR and represents the latest analysis of record.

(2) Manufacturer met this criteria by limiting stresses in the eggerate lattice bars to 1.05 S, = 44.1 ksi. Calc slated stress = 10.7 ksi.

(3) Design basis analyses performed to support Operating License issuance evaluated only single tube response to design basis events; this design analysis is not part of the original design basis.

(4) There is no specific design basis acceptance criteria associated with this consideration. An analysis was performed to assess alternative methods to control secondary chemistry, and included a uniform corrosion allowance for all lattice bars. Analysis demonstrated that the eggerates, and therefore the tubes, retained integrity with no credit taken for some of the lattice bars.

(5) Bounded by existing Regulatory Guide 1.121 burst pressure limitations v.hich are not afTected by eggerate degradation.

(6) Revised analysis results indicated some tubes with a natural frequency in the range of RCP natural frequency. For these tubes, the analysis determined that be maximum stress was still well below the code endurance limit.

(7) No support degradation and therefore no change to VN,=0.47 in the bend region of the bundle. VN, at degraded eggerates is not a concern due to the short duration of the event.

(8) Bounded by eggerate evaluation for LOCA and SSE (9) Analysis demonstrates that the eggerates, and therefore, the tubes, retained integrity with no credit taken for some of the lattice bars.

(10) Original analysis showed fnw>2x pump frequency (=40 cps) or <2/3x pump frequency (=12.7 cps). With fry outside of this range, l

tube stress was below code endurance limit.

l 3-7

[_-

3.2.2 Fingle Tube Analysis 4

y

[

The single _ tube analyses are ASME Code evaluations that are intended to ensure that i

_ steam generator tube integrity is maintained during normal and accident considerations 2

coincident with an earthquake. The single, tube analyses demonstrate that the tubes are within ASME Code allowables with degraded eggerate conditions. Structural integrity 1

evaluation of the degraded eggerate is provided in Section 3.2.3.

i

- 3.2.2.1 Normal Operating Conditions j

a. Methodology 4

The loads included in the normal condition analysis are flow loads, thermal expansion,

[

and internal pressure. As described below, there is no change to these loads as a result of i

eggerate degradation. Flow induced vibration as a result of normal operation is evaluated

)

separately in Section 3.2.2.4.

In the top portion of the tube bundle where the tubes are horizontal, the steam water i

mixture flows across the tubes and applies a flow load. The stress evaluation of the tubes for flow loads remains unchanged because none of the vertical supports which support the flow load are degraded. The thermal expansion stresses in the tubes could possibly be i

slightly reduced due to any additional movement permitted at degraded supports, but no i

credit is assumed for this effect. Tube intemal pressure is unaffected by eggerate degradation.

j

b. Results i

Since the loads included in the normal condition stress analysis are not changed, there is

{

no change to the stress summaries provided in the calculation of record :

Primary local membrane plus bending stress = 12 ksi, which is less than the ASME Code

[

allowable stress of 1.5 S. (= 35 ksi).

I Primary plus secondary stress intensity range = 17.3 ksi, which is less than the ASME Code allowable stress of 3 S. (=69.9 ksi),

j-The fatigue usage factor = 0, compared to the ASME Code allowable value of 1.0 For the case where up to two consecutive eggerates are uncredited, there would be an i

increase in the maximum operating basis earthquake (OBE) stress in the vertical tubing runs. The maximum OBE stress for this configuration is less than 10 ksi. Adding this upset condition stres ; to the primary local membrane plus bending stress listed above, the resulting stress intensity is 22 ksi, which remains well below the ASME Code allowable i

stress of 35 ksi. Adding the OBE stress to the primary plus secondary stress intensity k

y e

range, the resulting stress intensity range is approximately 27.3 ksi, which is less than the ASME Code allowable stress of 69.9 ksi.

3.2.2.2 Loss of Coolant Accident (LOCA) with Safe Shutdown Earthquake (SSE)

a. Methodology The analysis of record has been updated to evaluate the impact of degraded eggerate lattice bars. The revised analysis includes the same LOCA rarefaction loads as were used in the analysis of record. These Guid dynamic loads are based on an assumed double ended guillotine break of the reactor coolant system cold leg. Although leak-before-break technology has been accepted by the NRC to eliminate dynamic effects of high energy pipe ruptures in the reactor coolant system, postulated pipe breaks in smaller branch lines must still be assumed. Rather than recalculate the LOCA rarefaction loads based on smaller break sizes, the revised analysis conservatively assumes loads corresponding to the same break size as the calculation of record. Another factor which influences the magnitude of the LOCA load is the assumed number of tubes plugged. The revised analysis assumes that 1000 (10.7W abes are plugged per steam generator, which l

maximizes the LOCA rarefaction loed. There are currently less than 500 tubes plugged l

per steam generator.

l The analysis of record showed that tube row 147, supported by eggerates 1-10, is limiting l

when all eggerates are intact based on:

l l

This tube row has the longest horizontal span, and l

It has the highest rarefaction load.

l Although there is degradation of some eggerates below eggcrate 10, eggerate 10 is in I

good condition and remains available as a LOCA support (for tube rows 121-147 which l

pass through this eggerate). Therefore, tube row 147 can no longer be considered l

limiting. Tc determine the most critical row, in view of the degraded condition below l

eggerate 10, the analysis was re-run for a series of bounding configurations. This re-l analysis, similar to the analysis of record, is based on performing a time history analysis l

for each bounding configuration using the ANSYS finite element program to calculate l

tube stresses due to LOCA. The seismic loads were developed from the steam generator l

feedwater nozzle response spectra.

The degraded eggerates were modeled by assuming that, except for eggerate 10, the top two eggerates are not credited to provide support to the tube model for LOCA loading.

l (As discussed in Section 3.4.1, any tubes with more than one (1) consecutive eggerate support missing have been plugged, which provides additional margin.) In this I

configuration with two (2) consecutive eggerates uncredited the stresses are below code l

allowances.

l l

3-9

For LOCA loading with the top one or two eggerates uncredited, the horizontal 1

displacements at the top of the bundle tend to be larger than those in the analysis of record. The gap between the tube and the shroud was modeled, and impact stresses due to any potential contact with the ring are included in 1: stress evaluations. Details can l

be found in Attachnymt B of Appendix D.

b. Results The maximum stress intensity due to LOCA rarefaction, LOCA shaking, and SSE, and internal pressure were combined and compared to the ASME Code Class 1 allowable stress (80.6 ksi) for faulted conditions. Assuming tube wall thinning equal to a maximum of 64% per UFS AR paragraph 5.4.2.3.1.3.C, the maximum stress intensity evaluated per Regulatory Guide 1.121 is 80.4 ksi at tube row 120 for two (2) consecutive eggerates I

uncredited, which remains below the ASME Code limit. Hence the maximum tube l

thinning of 64%, which is based upon Regulatory Guide 1.121 burst pressure requirements, is not affected by eggerate degradation. For one (1) eggerate uncredited the l

maximum stress intensity is 72.9 ksi. (1,000 tubes plugged per steam generator and 107%

l flow).

l Sixty-four peicent (64%) thinning of tubes is not expected between cycle 9 and the mid-cycle outage. A conservative estimate of the maximum expected thinning is 39%. This is based on wear plugging criteria of 30% plus % the maximum wear of periphery tubes, which is 18% per cycle. The resultant stress for two (2) conseestive eggerates missing is l

approximately 48 ksi, which is 40% below the ASME Code allowables.

3.2.2.3 Main Steam Line Break (MSLB) with SSE

a. Methodology The analysis of record for this accident condition is not affected by eggerate degradation.

The loads which are included in this load combination are: MSLB impulse load, hydraulic flow loads, SSE, and internal pressure.

The MSLB impulse load, which results from the secondary fluid escaping from a postulated pipe break, is evaluated in the analysis of record as a tube sheet displacement time history relative to the tubing supports. Since the lower eggerates are not degraded, there is no change to the ability of the tubes to sustain this load.

The hydraulic flo ci load is included in the analysis of record based upon the maximum pressure difference across the tubes due to the steam flow past the horizontal tube runs at the top of the bundle. There is no change to the ability of the tubes to sustain this load, since there is no degradation of the vertical tube supports in this area.

The stresses due to SSE will not increase in the cross flow region since there are no degraded vertical supports. In the vertical tubing runs where the seismic stresses increase 3-10

due to degraded eggerates, the MSI.B impulse loads and hydraulic flow loads are negligible. Evaluation of SSE loading in this area is bounded by the LOCA plus SSE load combination.

The tube internal pressure load was conscwatively calculated in the analysis of recor i by assuming that the primary pressure remains at normal operating conditions while the secondary pressure decreases to zero as a result of the MSLB, which maximizes the tube pressure differential.

b. Results a

For the reasons discussed above, there is no change to the maximum MSLB with SSE maximum stresses. Therefore, the maximum stress intensity remains 21.9 ksi for the case of a 0.048" full thickness tube, at row 25, versus an allowable faulted condition stress of 80.6 ksi. For the case of a tube with an assumed 64% thickness reduction, the maximum MSLB plus SSE stresses are bounded by the LOCA plus SSE load combination.

3.2.2.4 Flow Induced Vibration - Normal Operation During normal operation, steam generator tubes are subjected to a fluid motion mainly in the axial direction - parallel to the axis of the steam generator. In addition to the axial flow component, horizontal components exist in the horizontal and U-bend areas of the I

tubes, at the tubesheet, and at the flow deflector plates. (See figure 7.3.40). In cross flow l

conditions, flow induced vibration (FIV) of steam generator tubes can occur as a result of l

fluid-elastic instability, turbulence, and vortex shedding. Fluid-elastic instability should be evaluated carefully since it has the highest potential for damaging the tubes. Tube vibration due to turbulence is generally small. The vortex shedding phenomenon is generally considered insignificant in two-phase flow regimes such as the flow inside a steam generatorWA. According to References 1 and 2 (Section 3.7), vortex shedding is a concem only in a single-phase flow motion.

The analysis of record addressed the effects of the reactor coolant pump (RCP) excitation on the tubes. The RCP generates mechanical excitation due to impeller imbalance at a frequency of about 20 Hz. It also generates a pressure pulse due to pump vane interaction at 95-100 Hz. The design requirement for RCP excitation is to ensure fatigue endurance l

l limits are not exceeded. RCP induced excitation is addressed in Section 6.6.1 of l

Appendix D.

a)

Methodology The susceptibility of the steam generator tubes to FIV was investigated under degraded eggerate support conditions. Absence of eggerate support is expected to reduce the tube natural frequencies, which increases the potential for tube instability in either a cross flow or a parallel flow regime. The tube stability evaluation is based mainly on the current theory of flow induced vibration, developed by Connors*. Flow velocities and fluid 3-11

density were calculated using the ATHOS3 Code. Some of the parameters appearing :'

the mathematical formulas, e.g., the instability constant and critical damping ratio, we. i determined based on laboratory testingW. Details of the FIV evaluation, and the ATHOS3 basc flow velocity calculations can be found in Attachments A and D of Appendix D, respectively.

The steam generator tubes with the largest unsupported spans, due to eggerate degradation, were identified as the most critical tubes in the bundle for the FIV evaluation based on their low natural frequencies. Susceptibility of these critical tubes to fluid-elastic instability was assessed by calculating the stability ratio, S.R., defined as the ratio S.R. = V n/V e

cr where the effective flow velocity, V,g, was calculated based on the actual flow velocity; and the critical velocity, V, represents the value beyond which the tube would be ct susceptible to FIV. It follows that a stability ratio of I corresponds to potential instability. Generally, a lower value of S.R. corresponds to a higher margin of stability.

The limiting cross flow velocity location for degraded eggerates is at the flow deflector plates. This is designated asV in Figur 7.3-40. The calculation of Veg and V, can be 2

c described briefly as follows:

(1)

Calculation of the Critical Veloqjly l

The critical flow velocity, V, was calculated for the analyzed tubes using the er well known Connors equation (see Section 6.5.1 of Attachment A, Appendix D).

The value of the criiical velocity corresponds to a midspan root mean square (RMS)Csplacement of 10 mils.

(ii)

Dlculation of the Effective Velocity The general purpose finite element pcogram ANSYS, Version 5.3, wrs used to generate.2 detailed finite element model for each of the analyzed tubes. Partially degraded eggerates were conservatively removed from the model. As an example, Figure 7.3-1 shows the model used for tube row 147 showing one of the mo&.s of vibration. Modal analysis was then performed to calculate the mode shapes and the natural frequencies of vibration were calculated, The resultant cross flow velocity was calculated as the square ront of the sum of the sqtures (SRSS) of the radial and circumferential components obtained from the ATHOS3 analysis. This analysis was performed using a model of the steam generator representing conditions after chemical cleaning (no fouling),

3-12

O The cross flow effective velocity, V, was calculated based on the mode er shape, from the finite element analysis, and the flow velocity from ATHOS3.

In addition to demonstrating stability, a calculation of tube midspan displacement due to cross-flow and parallel flow turbulence was performed. These displacements were combined to calculate the resultant tube displacement. Stress and tube wear evaluations were performed based on the tube midspan displacement.

Details of the FIV analysis can be found in Attachment A of Appendix D.

b)

Results Results of the FIV evaluation, in the form of stability ratio and tube midspan displacement, can be summarized as follows:

(I)

Limitine Stability Ratio A limiting stability ratio of 0.64 was calculated for tube row 110 with one eggerate not l

credited. The limiting SR with attemating eggerates not credited is 0.66 for row 110 with l

eggerate 7 and 9 not credited. In many locations with more 'han one eggerate not l

credited, a stability ratio less than 1.0 was calculated. Detailed results are shown in Table l

6.1-1 of Appendix D. Conservatively, however, tubes with more than one consecutive l

eggerate not credited have been plugged. Conservatism in the fluid elastic instability can i

be assessed as follows:

Since a stability ratio of 1 is interpreted as possible onset of fluid-elastic instability, S.R.=0.64 implies a margin of 0.36, i.e.,36%.

The damping ratio used in the critical velocity calculation does not include the damping due to the effect of parallel flow. To assess the effect of parallel flow damping, consider the limiting tube row, No. I10, in Table 6.1-1 in Appendix D.

The damping ratio for this tube row is 0.022, and it can be shown that the parallel flow damping ratio is about 0.005. Therefore, ignoring the parallel flow damping results in under-estimating the critical velocity by about 10%.

Combining the stability ratio margin and the damping margin amounts to about l

46% in the limiting case. This margin can also be interpreted as a 46% margin on the velocities cal.:ulated using ATHOS3.

The instability factor in the critical velocity calculation was based on tube W

W midspan RMS displacement of 10 mils only. Testing of tube bundles indicated that no instability was observed at this level of displacement.

Furthermore, a lower oound value ofinstability constant of 3.2 was used 3-13

throughcot the analysis to eliminate the critical velocity even though the tube confign ation in several regions of the bundle justifies the use of a higher value of 4.9.

No credit was taken for the eggerate in the analysis if two sides at the tube are badly degraded even when the other two bars are healthy. This approach is conservative because the remaining sides would tend to increase the natural 1

frequency of the affected tube, and hence the critical velocity, Similar conclusions can be made for the case of tube row 110 with eggerates 7 l

e and 9 not credited (V/V,, = 0.66).

l l

In addition, key FIV results in Appendix D were independently verified by MPR Associates,Inc. Good agreement exists between the results of the MPR and ABB-CE analyses.

(ii)

Tube Disolacement The limiting resultant tube midspan displacement is 0.0027 inches. The corresponding bending stress in the tube wall it. negligible (< 0.3 ksi).

This stress is significantly less than the endurance limit of the tube material (> 13.6 ksi). Similarly,it was concluded in Attachment A of Appendix D that this level of tube midspan displacement does not l

represent a tube or eggerate wear concem. Steam generator fouling, prior l

to Cycle 9 chemical cleaning, could have resulted in slightly higher flow l

velocities. The resulting tube stresses were not calculated specifically for l

this case. These stresses, however, would still be well below the l

endurance limit based on the wide margin shown above.

l (iii)

Primary Pumo Excitation Per Section 6.6.1 of Appendix D, the load due to the reactor coolant pump (RCP) mechanical excitation is equivalent to a load resulting from 0.25g acceleration only, assuming resonance and a damping ratio of 2%. The I

resultant stresses are less than the endurance limit. It was concluded, I

therefore, that the etTect of the RCP mechanical excitation is negligible.

(iv)

Pressure Pulse Due to Primary Pumo Vane Interaction A pressure pulse, at a frequency of 95-100 Hz, is generated during normal operation by the primary pump due to vane interaction. An evaluation of We worst case tube assuming resonance was performed in the steam generator design report (see Section 6.6.1, Appendix D). The resulting stress in the tube was less than 2 ksi, which is much smaller than the 3-14 i

-J

endurance limit for tube material. Therefore, this results in no change to

- the original analysis conclusions.

3.2.2.5 Flow Induced Vibration - Main Steam Line Break

_ a)

Methodology An evaluation of FIV tmder main steam line break (MSLB) was performed in Section 6.6.2 of Appendix D. It was concluded that FIV during MSLB does not represent a j

concern based on the following considerations:

(I)

MSLB is a short duration event, with end of blowdown occurring at less than two minutes from the caset of the accident.

.(ii)

The flow velocities inside the steam generator during MSLB are expected to be higher than during normal operation, which would result in increasing the amplitude of tube vibration. However, the total number of cycles would be very small based on the short duration of the event.

Therefore, fatigue and wear have no significant effect on the current evaluation.

b)

Results No reevaluation of the tubes in the horizontal region of the tube bundle was performed since no support degradation was detected in that region. Furthermore, the degradation of the eggerates has no impact on the support of the horizontal region. Therefore, previous analysis results for the horizontal region are considered applicable without change.

3.2.3 Eggerate Evaluations The eggerate evaluation or "whole bundle" analysis is intended to consider the structural integriy of the eggerate to ensure that it does not fail in a way that could damage the tubes. This is not an ASME Code evaluation; however, ASME Code techniques are used to generat'e and assess the results.

3.2.3.1 Main Steam Line Break (MSLB) with Safe Shutdown Earthquake (SSE)

a. Methodology The stresses in the eggerate lattice bars due to MSLB with SSE were evaluated in the original anahtical report. An individual lattice bar was assumed to carry the pressure

' load due to MSLB over its entire effective length within the eggerate. Simply supported end conditions at the support rings were conservatively assumed. The bending stress at lattice bar midspan, due to the MSLB loading, was combined with the seismic stress to obtain the maximum total stress within an eggerate for this accident condition.

3-15

b. Results As shown in the analysis of record, the maximum stress is 10.7 ksi, which is approximately 24% of the ASME Code allowable stress of 44.1 ksi. At the midspan, which is in the interior of the bundle, lattice bar thinning is insignificant. Therefore, there is no change to the MSLB plus SSE maximum bending stress calculated in the analysis of record.' Shear stresses in the lattice bars at the outer edge of the bundle (i.e., at the support rings) are insignificant, and therefore lattice bar stresses in the thinned areas (at the outer edge of the bundle) are bounded 'uy the LOCA plus SSE loading condition.

3.2.3.2 Loss of Coolant Accident (LOCA) with Safe Shutdown Earthquake (SSE)

a. Methodology Eggerates with der,raded lattice bars were analyzed using the "ANSYS" finite element analysis prograra. The tubes apply horizontal loads to the eggerates under the LOCA plus SS3 accident condition. The magnitude of these loads was determined using a

/

model of the tube bundle. Based on the number of tubes in eacn row, the tube rows were consolidated into 31 groups of similar configurations. A time history analysis was performed by applying the LOCA loading to each of the 31 groups.

The LOCA loads for each group were determined for postulated double ended guillotine l

breaks in the hot leg as well as the cold leg reactor coolant piping. As discussed in l

Section 3.2.2.2, no credit was taken for the fact that leak-before-break technology has been accepted by the NRC to eliminate dynamic effects of high energy pipe ruptures in the main loop.

The LOCA loads are maximum at eggerates 9 and 10. For each of those eggerates, the total reaction force is equal to the reaction force of each representative tube that passes through the eggerate, multiplied by the total number of tubes in its group, and combined algebraically to the loads obtained in a similar manner for all other groups which pass through that eggerate. The eggerate model was loaded in groups, with the loads for each group entered incividually.

The inspection results for eggerates 7 through 10 were reviewed, and it was determined that eggcrate 9 is the bounding eggerate for the following reasons: The degradation of eggerate 10 is significantly less severe than for eggerate 9. Eggerate 9 has a significant number oflattice bars which werejudged to be in category "B", which means their thickness range is 10% to 50%. He wever, for eggerate 10, all lattice bars are at least 50%

of the original thickness. Loadmg on eggerates below 9 is governed by seismic loads rather than LOCA loads, and the seismic loads are much lower. Comparing the inspection results for these eggerates, the degradation is slightly me:e severe in eggerate 8, but considering that the total applied load to eggerate 8 is estimated to be less than %

the eggerate 9 load, it isjudged that eggerate 9 is the limiting case. Because of the 3-16

diminishing LOCA loads at the lower elevations, the conclusion for eggerate 8 also applies to eggerates 7 and below.

^

Comparing the direction of LOCA loading due to a hot leg break to that of a cold leg break, the loads applied by the tubes to the eggerates are in opposite directions. He cold leg postulated pipe break puts the hot leg eggerate into compression, and a hot leg postulated pipe break puts the hot leg eggerate into tension. Both conditions were analyzed.

A detailed "ANSYS" model of eggerate 9 representing all lattice bars on the hot side was used to analyze the loadings. The eggerate 1 inch and 2 inch strips, support rings, and scallop bars were included in the model. The loads were applied to the model at the time step corresponding to the maximum total load from the time history analysic. For the hot leg break, where the tubes put the hot leg eggerate in tension, the effective area of the 2" strips was reduced to account for the 1" slots (at the intersections with the 1" lattice bars).

For the cold leg break case where the tubes put the hot leg eggerate in compression, potential buckling of thinned lattice bars was considered. In both cases, inelastic analysis was performed to obtain load redistributions after stresses in some of the strips exceeded the elastic range. The tubing loads applied to the hot leg eggerate were assumed to remain constant, even though it would be expected that the cold leg eggerate would carry an increasing percentage of the total tube bundle load as a result of the increased hot leg eggerate flexibility,

b. Results For the hot leg break, the finite element model determined that the maximum strain within the lattice bars needed to maintain structure integrity of the eggerate remained below the ultimate strain, thereby ensuring that the eggerate remains attached to the wrapper. In the tension case, the tubes move away from the support ring, so there is no concern about impact of the bundle on the ring.

Similar to the hot leg break case, the cold leg break case also yielded results wa' ich indicated that the eggerate retained structural integrity with the support ting. The maximum deflection of the edge of the bundle toward the support ring was found to be approximately.058" at the most critical location, which is less than the 0.174" minimum gap at that location. Therefore, it was confirmed that eggcrate will provide sufficient support to ensure that the tube bundle will not impact the support ring.

3.3 San Onofre Unit 3 Eggerate Video Inspection To support analysis assumptions and further define the condition of the eggerates, expanded SG intemals inspections were conducted. Figures 7.3 7.3-37 show the inspection scope for the inner btmdle inspections, maps for individual tube support conditions, peripheral tube support conditions, and the tube plugging maps.

3-17

- ~-

3.3.1 Backgrcund '

The first indication of eggerate degradation to cause concern was discovered during the l-L pre chemical cleaning inspection of SG E088 and was documented in NCR 970400975.

Expanded inspections were deemed necessary and were conducted after chemical cleaning. These inspections included a top-to-bottom look at the bundle at the periphery.

The video probe is " dropped" down in the tube bundle via a guide tube to its lower most location and then withdrawn. During the withdrawal process the probe is halted at the various eggerates for the inspections.

The method used to gather the data was tailored to meet the inspection objectives. The team was required to provide video documentation of all areas where indications of support bar degradation was suspected and to verify that other areas did not exhibit these same characteristics.

The video inspection was divided into six areas: a) general inspection.. b) inner bundle, c) batwings and vertical strips, d) eggerate periphery, e) blowdown lane and f) stay cylinder.

The inspection required state of the art (including underwater) small lens camera technology with manual remote manipulation as indicated below.

Area General Inner Batwings/

Periphery Blow.

Stay Cylinder Inspection bundle Vertical down strips lane Equipment 6 mm Welsh 1.5 mm 1.5 mm fiber 6mm Welsh 6mm 10 mm Allen and -

fiber optic lens Allen Welsh CCD video CCD video optic lens Allen camera camera 3.3.2 SG GeneralInspection (Interuals)

General area, U-bend and annulus region inspections were performed in both steam generators from the top of the moisture separator can deck. The inspections were

_ performed utilizing an underwater CCD camera with high intensity lighting suspended above the tube bundle. The camera was capable of pan tilt zoom and remote focus. A 6 mm video probe was utilized in areas inaccessible by the pan and tilt camera. The areas inspected included I-beams, I-beam to shroud attachments, drains, vertical supports, batwings and the batwing hoop, and baffle anti-rotational keys. These inspections identified no areas of concern or degradation. Figures 7.3-2 to 7.3-3 are examples of the inspected areas.

3.3.3 Eggerate -Inner Bundle Inner bundle consists of that area between the outer or peripheral tubes to the inner tubes of the stay cylinder. This is the largest section of the tube bundle and the most difficult to 3-18

~

u

inspect. The inner bundle inspections were performed using a 1.5 millimeter fiber optics lens " dropped" from the upper elevations of the tube bundle. The inner bundle inspections were performed in both steam generators from the can deck.

The purpose of the inner bundle inspection was to assess the general ma'erial n ndition of the eggerates away from the periphery area. On 50 E088 25 drops were performed,6 on the cold leg and 19 on the hat leg. On SG E089 25 drops were perfonned 18 on the hot leg and 7 on the cold leg. The areas inspected in each steam generator are shown in figures 7.3-4 and 7.3 5. A review of the inspection tapes indicated the inner bundle did not exhibit the degraded characteristics of the periphery eggerates. Figures 7.3 6 and 7.3 7 show the typical conditions in the inner bundle of each steam generator.

3.3.4 Eggerate - Hatwing/ Vertical Strips During the inner bundle inspection, interior batwing and vertical strips were assessed. No indications of thinning were detected as shown in Figures 7.3 8 and 7.3 9.

l 3.3.5 Eggerate - Periphery Comprehensive peripheral eggerate inspections were performed in both steam generators l

from the can deck. A 6 mm x 35 foot video probe was inserted into a guide tube and

{

directed between the batwings to access the " drop tube" location at the upper most eggerate. The camera was then lowered from the can deck down through the eggerates until the 2nd eggerate was reached. Once the 2nd eggerate was reached, the inspection started by scanning all visible eggerates/ tubes (typically 4-6 tube:) from the " drop tube" location as the camera was withdrawn. This process was repeated for all desired locations. Areas ofinterest includeo the lattice Lars and tube to lattice bar interfaces at each eggerate. Locations were verified by eddy current detection of the probe or by visual confirmation from tube bundle geometry. Figures 7.310 to 7.313 are examples of the types oflattice bar degradation found. The results of these inspections were then mapped and assessed. A grade was given to each lattice bar as follows:

A = greater than 50% of nominal thickness B = 10% to 50% of nominal thickness C = less than 10% of nominal thickness D = severed lattice bar E = appears thinner on opposite edge than edge being viewed. i.e. not assessable X = lattice bar obscured /not viewed N = no visible thinning Figures 7.3 14 to 7.3 35 show the results of the periphery inspection at those elevations exhibiting degradation.

3-19 l

3.3.6 Eggercle. Hlowd:wn Lane Inspections of the blowdown lane eggerates were performed in both steam generators through the 6" handhole at the secondary face of the tubesheet from the handhole to th-ray cylinder. Areas ofinterest included the lattice bars, eggerate rings, and tie rods. A guide assembly was directed upward through each eggerate with a video probe / camera.

The inspection scope was to sample the eggerate area nearest the tubes on both the hot and cold leg sides of tne blowdown lane. No tubes were plugged as a result of this inspection.

3.3.7 Eggerate/ Ring - Stay Cylinder For this inspection a Brooks S.I.D. (support plate inspection device) was used The inspection was intended as a " general area" inspection to assess the overall condition of the eggerates and ring in the stay cylinder. No tubes were plugged as a result of this inspection.

3.3.8 Steam Generator Chemical Cleaning Gauging - First Eggerate la accordance with the plan for steam generatur chemical cleaning, physical gauging was performed of the first eggerate (near the bundle). This gauging confirmed the assumptions of the chemical cleaning program. All measurements were well with:a the allowances established for operational and chemical cleaning corrosion. Maximum thinning measured was approximately 7 mils compared to an allowance of I 8 mils, 3.4 Repairs and Repair Criteria 3.4.1 Tube Plugging Criteria for Eggerate Degradation The analytical results were utilized to determine the extent of repairs required to ensure that the design basis allowables are met with degraded eggerates. To accomplish this, a comparison of the analytical results and inspection results was made end a tube plugging list was generated. In all cases, conservative application of the inspections and analysis were made. Standard tube plugging associated with eddy current results for the outage was performed independently of tube plugging for eggerates. However, a comparison of the plugging for eddy current results and eggerates results was made. A discussion of the comparison is made in Section 3.5.

For flow induced vibration, the results of the analyses showed that the tube stability I

criteria are met in all cases when one support or attemate supports was uncredited. With l

two consecutive supports uncredited the stability criteria was met, with few exceptions.

I Furthermore, results of the accident analyses showed that two, or more, supports l

uncredited were acceptable. Therefore, the plugging criteria was based on FIV bounded I

accident condition requirements.

l l

3-20

A more consenttive criterin than the analytical acceptance criteria has been used for generating the plugging list for eggerate degradation. Specifically, for each tube that has more than one (1) consecutive eggerate considered uncredited, the tube was plugged. This is approximately a 50% margin because the majority of the tubes plugged were analytically acceptable with two (2) consecutive eggerates not credited.

Maps showing the locations of the tubes plugged for eggerate degradation are given in l'igures 7.3 36 and 7.3 37 3.4.1.1 Implementation of Eggerate Grading A grading system of A, B, C, D, E, X and N (no detectable thinning) was utilized in the visual inspections. The analyses conservatively considered "A's" to be 50%, "B's" to be 10% and no credit was taken for "C's", "D's" and "E's" or "X's". For example a lattice bar with a rating of "B" would be considered credited. A lattice bar with a rating of "C" would not be credited (see Figure 7.3 38). These grades were assigned to individual lattice bars which in turn were used to assign a rating to each eggerate on a tube by tube basis.

1 It is very conservative to consider the grading in each category as the minimum value l

because the lattice bar thickness within each category varies considerably. The grading system was chosen to facilitate the capabilities of the video assessors in accurately distinguishing between the condition of the lattice bars, therefore, the "B" category is an average of much higher than the minimum thickness of 10%. In a survey of one of the eggerate videos to confirm this conservatism, grades of"B+" were given to a significant number of the B's. This indicates a higher average than 10% for the "B" rated bars.

The analysis also conservatively interpreted that two (2) or more lattice bars uncredited at a location would constitute the eggerat: at that location not being credited. Therefore, a condition around a tube that shows two (2) lattice bars uncredited is considered to be equally as degraded as four (4) lattice bars uncredited. This is conservative because any contact between the tube and support will signincantly increase the natural frequency which reduces the potential for FIV. Also, the inspection results showed that there are no cases of four (4) lattice bars uncredited around a tube, in addition,if a single lattice bar cannot be credited at an eggerate location, but the other three (3)1attice bars in that location can be credited, then that eggerate can be credited as a supported location for se tube. This is based on the failure mechanism associated with FIV. Specifically, FIV ', rom Huid elastic instability is caused by the onset of orbital motion which leads to wear (see Figure 7.3-39). If a single lattice bar cannot be credited (3 lattice bars credited) then the only potential motion would be on a single-axis which will not cause fluid clastic instability. Any contact between the tube and the support would lead to higher natural frequencies and less likelihood of FIV, 3 21 w

3.4.1.2 Specific Plugging Criteria Tube plugging was specifically based on the following:

1) At a given eggerate elevation, grade each lattice bar surrounding a given tube, a)

If two (2) or more of the lattice bars cannot be credited (C, D, E or X) then the eggerate at that elevation may not be credited.

b)

If three (3) or more lattice bars can be credited (N, A, B and sometimes X), then the eggerate location can be credited.

2) Repeat the grading oflattice bars for each tube at each eggerate elevation.

3) Evaluate each eggerate on a tube by-tube basis. Plug the tubes with two (2) or more consecutive eggerates that cannot be credited.

Le process for tube plugging was perfonned independently by a team of 3 engineers.

Their plugging list was then independently reviewed and verified using the same process j.

by another team of engineers.

In many cases where video data was not available for a lattice bar (graded X), the lattice bar was not credited. However, in some cases in regions where video data was not available, and the surrounding lattice bars were in better condition (N's, A's or B's), the 2

lattice bar with the missing information may be credited. This was conservatively applied, based on the condition of the surrounding lattice bars.

4 3.4.2 Plugging / Stabilizing A plugging / stabilizing repair technique was assessed for use in those areas where eggerate degradation exceeded the repair criteria. The technique consists ofinserting a j

standard 40 foot long, stainless steel cable into the tube to span the areas of eggerate l

degradation and extend through to good supports on either side of the degradation. The j

tube ends are then plugged with a standard mechanical tube plug.

The plugging / stabilizing repair technique was evaluated to determine its effectiveness if various support configurations were assumed to be missing. The presence of the cable adds sufficient stiffness and damping to avoid fluid clastic instability of the tube / stabilizer combination and prevent future tube to tube interaction e

f 3.4.3 Plugging / Stabilizing Locations / Recommendations For tubes with eggerate intersections that have damage exceeding the repair criteria expressed in Section 3.4.1, the plugging and stabilizing repair technique was used to stabilize those tubes and prevent future tube-to-tube interactien. For SG88, thirty-eight i

]

3-22 4

l-

(38) tubes were repaired for cssociated eggerate damage exceeding the repair criteria. An cdditional eleven (11) tubes were plugged and stabilized to surround previously plugged.

and unstabilized tubes that now exceed the eggerate repair criteria. This brings the total repairs for eggerate damage in SG88 to forty nine (49) tubes. For SG89, sixty (60) tubes were repaired for associated eggerate damage exceeding the repair criteria. An additional four(4) tubes were plugged and stabilized to surround previously plugged and-unstabilized tubes that now exceed the eggerate repair criteria. This brings the total repairs for eggerate damage in SG89 to sixty four (64) tubes.

3.5 Tube Support Wear and Eggerate Degradation An increase in tube to tube support mechanical wear was noted during the current, Cycle 9, refueling outage eddy current inspection of Unit 3 steam generator tubing. After reviewing the locations where the increase was noted, Edison concluded the increase in i

wear likely stems from the same cause, secondary tube deposits, as the eggerate degradation but was not a direct result of thinned eggerate supports.

Some peripheral tubes have indications of tube to tube support mechanical wear.

flowever, this wear is independent of peripheral eggerate degradation, in SG 88, there t

are no cases where a tube is being plugged for eggerate degradation where that tube also has indications of wear. Although 25 other tubes in the periphery show wear, the wear locations are not systematic. Instead they are distributed throughout the bundle on both hot and cold leg sides.

In SG 89,8 tubes with minor wear indications are being plugged for eggerate degradation. Of the 8, two are on cold leg eggerates, two at batwings, and the remaining r

f four are associated with vertical strips. I1 other tubes in the periphery have indications of wear. As in SG 88, they are also not systematic and are distributed throughout the tube bundle. Thus, these wear indications are not directly related to eggerate degradation, increased wear was identified within the tube bundle interior, particularly at vertical strips and the upper most eggerate supports. The inner bundle eggerate and vertical strips were visually inspected in the areas with the highest concentration of tube support wear indications. All supports were found to be in good condition. Consequently, the increased level of wear resulted from changing flow field parameters (particularly fluid density and viscosity) within the tube bundle and not changing clearances at tube supports. The changes to flow field parameters is one result of the deposit condition which Edison believes is responsible for the eggerate thinning discussed throughout this report.

By examining the frequency of tube to tube support wear, a probable initiation period for the on-set of FAC can be derived. Figures 7.3-41 and 7.3-42 contain the history of cuihulative wear indications at tube support intersections for Unit 2 and 3 respectively.

These figures show a clear increase in the trend ofindications in Unit 3 occurred between the 1995 and 1997 refueling interval inspections (approximately 8.6 and 10.0 EFPY).

3-23

Tube wear at support locations is the result of energy being dissipated from a vibrating tube to its supports. Fluid flow continually imparts energy to the tubes causing their vibration. Wear can increase if the energy imparted to the tube increases or the ability to dissipate to the Dow field decreases. As deposits increase, steam quality in the upper bundle region increases, lowcring density and fluid Held viscosity. For the tube, this means less energy can be dissipated to the Dow Deld and more is dissipated to the support intersecticns.

3.6 References 1.

Welding Research Council Bulletin 372, Guidelines for Flow induced Vibration Prevention in lleat Exchangers, hiay 1992, 2.

American Society of hiechanical Engineers (AShiE) Boiler and Pressure Vessel Code, Section 111, Division 1. Appendix N, Code Year 1995.

3.

Connors,11. J., Jr., " Fluid Elastic Vibration Of11 eat Exchanger Tube Arrays," AShiE Publication 77 DET-90,1977.

4.

lleilker, W. J., Beard, N. L. and Park, J. Y.," Flow Induced Vibration Analysis in Support of the Design of the Yonggwang Units 3 and 4 Steam Generators," Proceedings of the International Symposium on Pressure Vessels Technology, Nuclear Codes and Standards, April,1989 (REF 96 017).

S.

Evaluation of Southern California Edison SONGS Unit 3 Steam Generators With Degraded Eggerates, A SONGS 9416 ll68. (S023 915-l 208) l 6.

Letter, J. L. Rainsberry to USNRC dated hiay 26,1997:

Subject:

Docket No. 50 362," Steam Generator Tube Eggerate Supports" 7.

SO23-XXVil 4.3 Steam Generator Secondary Side Visual Inspection, Unit 3 Cycle 9 8.

SO23 XXVil 4.4 Steam Generator inspection Assessment 9,

SO23-XXVII-4.5 Steam Generator Stay Cylinder Visual laspection 10.

SO23 XVII 9 Steam Generator Secondary Side VisualInspection 11.

CE Report, CE-NPSD 543, Eggerate Support Stress! Corrosion Evaluation 12.

hiemo for file, hieasurements on vertical strip and EC-1 3 24

i l

13-M:morandum for File.. san Onofre Unit 2 Chemic:1 Cleaning Video Inspection 3 25

4.0

SUMMARY

OF CONCLUSIONS 4.1 Cause Assessment Conclusions Postulated causes of eggerate degradation are consistent in their conclusion that chemical cleaning by itselfis suflicient to arrest accelerated eggerate material loss in addition to chemical cleaning, the beneficial effect of changing from an ammonia based pil control to the ETA based control will further reduce FAC susceptibility. With implementation of ETA for pil control, it is projected that the integrated fouling will not exceed 50% of the deposit level that was present in Unit 2 at the end of Cycle 8, until the year 2023. Thus, recurrence of the degradation process is not expected.

4.2 A nalysis Conclusion An analytical assessment of the degraded eggerates was performed and the results demonstrated that with plugging and stabilizing of selected tubes, the eggerate degradation that was found is acceptable. The results of the degraded eggerate evaluations were compared with the results of the existing design basis analyses and were found to be within the existing acceptance criteria. The analytical tools and methodologies used to perform the degraded eggerate calculations are the same or similar to the tools used to perform the original analyses and no new an' lyses were required.

4.3 Inspection Results Edison performed a comprehensive inspection ofinternah of both Unit 3 steam generators. This inspection confirmed the degradation was limited to the upper eggerates and was confined to their peripheral portions. To a lesser extent the stay cylinder and blowdown land areas were also affected. The cold leg portions of the eggerate periphery were also affected to a lesser extent.

4.4 Continued Operation The analysis has concluded the existing degraded eggerate condition is acceptable. These results further indicate additional degradation can be absorbed during the operational period following this outage. By conservatively scheduling the special mid-cycle inspection, this uncertainty in this assessment is well bounded. Although it is probable the FAC degradation began recently and thus proceeded rapidly, i' is improbable those pre cleaning rates will continue even if the deposit removal is not completely effective.

The FAC process is non linear, once the initiation threshold is reached, rates can be quite large. Even a moderately successful cleaning will certainly red' ce the process below the initiation threshold as evidenced by the absence of FAC degradation in Unit 2 in the presence of substantial deposits (approximately 80% of Unit 3) last cycle. If FAC does continue at a reduced rate, a possibility considered remote, that margin is ava!!able to absorb the resultant degradation.

41

An imponant input into the margin assessment is the use of unifonnly thinned tubing to the design limit (64%). The cssumption is that a tubing degradation mechanism (most likely tube to tube support wear) will occur in the tube locations most affected by the l

eggerate thinning. Close examination of the tube to tube support wear experience at Unit 3 shows wear is not the result of degraded eggerates, Coupled with this is Edison's conservative use of a wear plugging criteria of 30%, which is substantially less than the technical specification limit of 44% and the design limit of 64%. As a result, substantial uncredited margin exists between eopected operational conditions and design requirements.

Given these measures have arrested further FAC and stabilized the eggerate degradation, operation over the normal refueling inspection interval would be acceptable. Ilowever, for conservatism, Edison has detennlued a special mid cycle inspection of the eggerate condition will be performed to confinn this assessment. The mid cycle will be scheduled to occur at approximately the mid point of the operating cycle (approximately 1012 effective full power months). Subsequent refueling interval inspections will be perfomied until it is apparent the degradation mechanism has been arrested.

To establish the duration of the operating period until the next inspection, Edison I

conserva1vely assumed the degradation process was not arrested by chemical cleaning l

and only the tolerance in the inspection results would be used. Analytical margins were l

not used to support the derivation of the operating period. This tolerance is based on the I

conservative assessment of the visual inspections for the as found eggerate condition.

l The most bounding tolerance for all applicable cases is the "whole bundle" analysis.

l 4.5 Affect and Benefit of SGCC Chemical cleaning was conducted as planned during the Unit 3 Cycle 9 outage.

Approximately 19,000 lbs of corrosion products were removed from each SG. This l

reflects approximately 25% more corrosion product in the Unit 3 SGs than in the Unit 2 l

SGs. In addition, physical measurements were taken of a sampling of upper vertical strips and eggerate number 1 using go no go gauges. All materials indicated that corrosion was within allowances. Although a small amount ofinsoluble deposits were seen in some tube eggerate crevices, the general cleanliness of the tubing is very good.

Consequently, the SGs will be returned to service with nominal design thermal hydraulic conditions.

4.6 Loose Parts Evaluation The existence of unrestrained foreign objects (loose parts) inside the secondary side of a steam generator creates the potential for damage to the tubes which may result in leakage of primary coolant into the secondary fluid. Such loose parts may be introduced from outside the steam generator during maintenance activities; they may be entrained in the feedwater flow; or they may be the result of the failure of steam generator internals. All three types ofloose parts have been found in San Onofre steam generators. ABB CE has 42

coniicted tests and an lyses of various loose parts in steam generators, including an cnalysis specifically for S:n Onofre in 1993, and has shown that tube damage from most postulated loose parts is unlikely.

4.6.1 Existing Loose Parts Eddy current examinations and routine visual examinations of the steam generators have confirmed the presence of several foreign objects in Unit 3 steam generators.

Examination of parts which have been removed from the steam generators suggests the likely source is the feedwater distribution ring which suffered damage in the late 1980s.

The behavior of these objects was investigated and analyzed in 1993 and it was predicted that the objects would remain in stable positions. Given their small size (the largest was estimated to weigh less than 4 ounces), it was considered improbable that severe tube damage could occur as a result ofleaving the pieces in the steam generator. A small tube leak was considered the most likely " worst case" damage scenario, in fact, Unit 3 operated for the latter part of Cycle 6 with a very small primary to secondary leak which was subsequently determined to be caused by loose part wear on a steam generator tube.

l 4.6.2 Potential Loose Parts The degradation of the eggerate lattice has lef1 some strips completely detached from the eggerate ring and thinned to less than 0.010". It can be postulated that pieces of this degraded strip could break away from the remaining bar and become a loose part.

However, the geometry of the strips limits the maximum size of the postulated loose part to approximately 2" x 1" x 0.090". A loose part of this size would weigh less than 1 ounce. As shown in the ABB CE analysis, this is too small to cause any tube damage.

Another postulated loose part scenario involves strip thinning to the point of complete detachment of a partial eggerate (level 8 10) from the ring. For eggerates 9 and 8, the eggerate would be held in its proper vertical orientation by the tie rods which pass from the tube sheet through these eggerates. Eggerate 10 has no tie rods, but has not exhibited significant degradation. Thus, the entire eggerate structure could not become a loose part.

4.7 10CFR50.59 Safety Evaluation Edison performed a safety evaluation, consistent with 10CFR50.59 requirements, to determine whether an unreviewed safety question would result from having degraded eggcrate tube supports in their present condition in the Unit 3 SGs. The complete evaluation is provided in Appendix G.

To confirm the eggerate design / licensing bases and applicable design basis events, Edison conducted a review of the UFSAR, Technical Specifications, applicable licensing submittals, and the NRC's Safety Evaluation Report (NUREG 0712). Edison identified, reviewed, and evaluated the impact of degraded eggerates on the applicable design basis analyses. Where warranted, re-analysis of the eggerates in their present condition was 43

performed. Where original design margins permitted, and when justified by maintoining compliance with acceptance criteria and analytical assumptions were refined.

Edison conducted a cause assessment investigation. Inspection data was utilized to corroborate proposed hypotheses. Recommendations for mitigating actions have now been implemented (i.e., SO chemical cleaning) to arrest eggerate degradation from flow accelerated corrosion and to limit future eggerate degradation to nominal values.

Considering eggerate supports in their present condition and the potential for loose parts from eggerates, Edison's safety evaluation concluded that:

(1) the ASME Code allowables for SO tubes for normal operation and design basis accidents were met, (2) the entire SO tube bundle is adequately supported for liiniting design basis accidents, (3) the SG tube plugging acceptance criteria was not impacted, and (4) the SG tubes' ability to function as the reactor coolant pressure boundary, to remove heat from the reactor coolant system to achieve and maintain safe shutdown, and to function as part of the containment boundary were I

not impacted.

Therefore, having degraded eggerate tube supports in their present condition in the Unit 3 SGs does not represent an unreviewed safety question under 10CFR50.59.

4-4 4

5.0 RETURN TO SERVICE ItEC05151ENDATIONS/ColtRECTIVE ACTIONS As result of the evaluations and inspections described herein, Edison implemented the following actions before to returning Unt: 3 to service for Cycle 9 operatica:

1.

Completed the planned Steam Generator Chemical Cleaning Project and post cleaning inspections.

2.

Performed a comprehensive review and analysis of the SG design bases including support and input from various industry consultants.

3.

Completed an expanded inspection scope of both Unit 3 SGs to detennine the eggerate condition and validate analysis assumptions.

4.

Completed a cause assessment of the eggerate degradation..

5.

Developed repair / plugging criteria to determine where repairs are necessary.

6.

Implemented the necessary repairs (plugging / stabilizing) per the established repair criteria.

7.

Submitted a compt bnsive report to the NRC on June 5,1997.

8.

Completed a 10CFR50.59 Safety Evaluation.

l 5-1 r

~

6.0 POST CYCLE 9 RECOMMENDATIONS / CORRECTIVE ACTIONS Aller the retum of Unit 3 to service, the following actions will be completed:

1, Complete and submit a final report to the NRC (90 days from close of breakers)

[ Completed Rev 1, October 16,1997]

l 2.

Prepare for and complete S0 inspection program during a mid-cycle outage. For Unit 3, Edison has committed to a mid cycle inspection of the steam generator eggerates. The purpose of this inspection will be verification of the cause determination and corrective actions for the eggerate damage, in that further accelerated erosion has abated.

Similar methodologies to those employed in the original inspection will be used.

Ten hot leg locations will be sampled for evidence of continued thinning. The sampling will include established areas of high thinning including locations where l

tubes were plugged. Should the assessment reveal further degradation, beyond operational norms, the scope will be increased such that the extent of the degradation can be bounded and characterized.

l Edison is evaluating several methods to monitor for continued degradation.

I Although a quantitative technique is desired, it may not prove practical. If so, a qualitative method will be utilized.

A similar inspection will be performed during the next refueling outage. Longer term inspection intervals will be based on the results from these two folbw up inspections.

Unit 2 has also committed to a mid-cycle inspection to monitor the steam generator tubing for freespan and eggerate cracking. During this outage, approximately ten locations on the hot leg periphery of both steam generators will be monitored for eggerate degradation. The scope of this inspection will be increased to fully bound and characterize any degraded conditions obsened beyond operational norms.

3.

Document these commitments / actions on the San Onofre Regulatory Commitment Tracilag System (RCTS).

6-1

7.0 CIIAllTS GRAl'lIS AND VISUAI, AIDS (FIGUltES)

Section 7.0 is a compilation of all the referenced figures in this report to facilitate their location.

I i

I s

7-1 g

8;0 APPENDICES A.

Description of the San Onofre Unit 2&3 Steam Generator Eggerate Design and Condition B.

Eggerate Evaluation Task Force (EETF)

C.

Unit 2 Operability Assessment D.

Evaluation of Southern Califomia Edison San Onofre Unit 3 Steam Generators With Degraded Eggerates, A SONGS-94161168 Revision 0, 6/4/97 E.

Cause Assessment and Corrective Action Report F.

Response to NRC Questions O.

10CFR50.59 Safety Evaluation, Rev i 1

8-1 P

-- ^

i i

i ENCLOSURE l

3 Thermal-Hydraulic Analysis of the Southern California EdL9n San Onofre Nuclear Generating Station Unit 3 Steam Generators with Degraded Eggerates j

A-SONGS ~9416-1168 1

l Cetober 16,1997 l

i a

G

j

\\

ENCLOSURE 1

.