ML072750013

From kanterella
Revision as of 18:32, 17 April 2019 by StriderTol (talk | contribs) (Created page by program invented by StriderTol)
Jump to navigation Jump to search
9/25/2007 Licensee Handouts for Meeting Re San Onofre Nuclear Generating Station, Unit 3 - Reclassification of Control Rod Drive Mechanism 56 NDE (Non Destructive Examination) Indication
ML072750013
Person / Time
Site: San Onofre Southern California Edison icon.png
Issue date: 09/25/2007
From:
Southern California Edison Co
To:
Office of Nuclear Reactor Regulation
Kalyanam N, NRR/DORL/LPL4, 415-1480
Shared Package
ML072750008 List:
References
Download: ML072750013 (30)


Text

1 September 25, 2007 Reclassification of Control Rod Drive Mechanism (CEDM) 56 NDE Indication Southern California Edison (SCE)San OnofreNuclear Generating Station (SONGS) Unit 3 2 Purpose*Based on all information available to date, SCE has reclassified the

indication in CEDM 56 as No Detectable Defect (NDD)-Discuss the technical basis for the reclassification-Discuss the licensing issues associated with the reclassification 3 Non Destructive Examination Results for San Onofre Unit-3 Control Element Drive Penetration 56 Southern California Edison 4 Key Points*Determination-The indication in CEDM 56 is not a defect -Code Relief for Flaw Evaluation is not required*Basis-Repair of CEDM 56 was based on refuted data-All supplemental exams were negative for PWSCC-Indication in CEDM 56 lacks PWSCC features-Indication is consistent with grain noise -Current evaluation methodology would conclude no defect is present in CEDM 56 5 Technical Presentation Overview-Background-Evaluation-Conclusion-Future plans regarding inspections and head replacement-Discussion 6*CEDM 56 identified as "s pecial interest" during reactor vessel head inspection in RFO 12 *Weld overlay was applied to J-groove weld and penetration OD surfaces to ensure that CEDM 56 was returned to service in a conservative configuration*Code Relief was needed to temporarily allow evaluation of a flaw having depth greater than 0.75 t pending final NDE evaluation and repair planning *Supplemental examination techniques were developed to investigate whether indication in CEDM 56 is PWSCC Background

7Background

Inspection & Repair History

  • Cycle 12 refueling outage(RFO-12)Jan2003-SCE Commitment to NRC Bulletin 2002-01-Westinghouse demonstrated inspection technology

-Ultrasonic time of flight diffraction (TOFD) indications observed in 4 penetrations-Supplemental surface eddy current (ET) ruled out PWSCC

  • Cycle 13 refueling outage(RFO-13) Oct2004-NRC Order EA-03-009 using Westinghouse demonstrated NDE-Indications in 4 CEDMs appeared to have grown -Supplemental ET data was not consistent with apparent growth-"Embedded Flaw" repairs implemented to ensure conservatism

-Indication in CEDM 56 exceeded Code limit for evaluation (figure1)

  • Cycle 14 refueling outage (RFO-14) Oct2006-NRC Order EA-03-009 using Westinghouse demonstrated NDE-Supplemental NDE designed for CEDM 56-No discernable changes in CEDM 56 indication

8 Background

Figure 1 CEDM 56 Indication Profile 9>75%High Resolution TOFD NDDUltrasonic shear Wave (axial & circumferential)

NDD NDDOverlay PT (Surface)

NDDPenetration OD TOFD (informational)

NDD NDDPenetration OD ET (Surface)

NDD NDDJ-weld ET (Surface)

NDD NDD NDDID ET (Surface)

NDD NDD NDDLeak Path Assessment NDD NDD NDD Bare Metal (Visual)

SI / CBH SI SITOFD (Volumetric)

RFO-14 October 2006EA-03-009 RFO-13 October 2004EA-03-009 RFO-12January 2003Bulletin 02-01SI, Special InterestNDD, No Defect DetectedCBH, Cleared By HistoryBackground CEDM 56 Inspection DetailStandard ExaminationSupplemental Examination 10 Evaluation*Approach*Determination of No Growth*Inconsistent with Primary Water Stress Corrosion Cracking (PWSCC)*Review of Industry Experience 11 Evaluation Approach*WesDyne (Westinghouse)-Perform a consistent evaluation of base scope NDE data for indications of growth in CEDM 56 (and CEDMs 32, 57, and 64) from RFO 12, 13, and 14 -Assess supplemental NDE data from CEDM 56 to determine possibility of a PWSCC flaw-Evaluation and assessment performed by*Level III with extensive experience in RVH TOFD*Chief Engineer, recognized authority in NDE*ANATEC -Third party review of NDE data and Westinghouse evaluation*Level III with extensive experience in RVH TOFD*Concluded not attributable to PWSCC*SONGS Engineering -Analysis, review, oversight, and ownership of results 12 Evaluation Determination of No Growth*Reanalysis of all inspection data on a consistent basis concluded no discer nable change since first detected in RFO12.

74.4 % t-w*70.2 % t-w RFO 14 74.3 % t-w 77.6 % t-w RFO 13 74.4 % t-w 55.7 % t-w RFO 12 Reanalysis Original AnalysisRefueling outage* RFO14 High resolution probe positioned indication at just over 75% t-w(>0.75 t) 13 Evaluation Inconsistent with PWSCC*RFO 12 -Automated J-weld and OD Eddy Current concluded no surface connected PWSCC*ET is a very effective exam for detection of PWSCC surface & near surface connections*A 0.7" surface or near-surface crack is readily detected (figure 2)*Lack of surface connection refutes PWSCC-Indication was dispositioned as grain noise*Large population of distributed reflectors (figure 3) 14 Evaluation Inconsistent with PWSCC*RFO 13-Manual OD scan employing TOFD & ET (informal)*ET: No crack like features*TOFD: No lateral wave interruption-J-weld ET: No crack like features-No effect from repair on ID exam 15 Evaluation Inconsistent with PWSCC*RFO 14-Confirmed no discernable change-High Resolution probe measured maximum indication depth at just over 75% from OD surface (could not withdraw 0.75t relief request)-Ultrasonic shear wave exams detected no evidence of a crack-like flaw*No shadowing of distributed reflectors*Continuous observation of overlay weld interface (figure 4)-Axial crack would obscure view of overlay interface-Transparency of indication is consistent with grain noise, not PWSCC 16 Evaluation Review of Industry Experience*Beaver Valley (BV) J-weld and OD ET exams had clear detection of PWSCC-Indication in CEDM 56 penetration OD would be larger and more easily detected if PWSCC-CEDM 56 J-weld scan found no crack-like indications*BV TOFD included typical PWSCC characteristics that are lacking in CEDM 56-No loss of backwallsignal-Crack tip diffraction response not evident-Indication has not changed since detection*Application of updated WesDyne analysis procedures would determine CEDM 56 as NDD 17 Evaluation Technical Summary*Ultrasonic TOFD (tip diffraction) technology is very sensitive, and commonly detects normal material structures in addition to real defects-Real cracks affect TOFD inspection results in a predictable way*Backwall reflections should be suppressed over certain angles*Distributed reflectors should be obscured behind crack faces *Characteristic tip diffraction signal response-CEDM 56 penetration includes TOFD features typical of grain noise, and not PWSCC *Confirmed growth in a TOFD indication over time is compelling evidence of PWSCC-Uniform analysis of RFO 12, 13, and 14 examination data has concluded that no discernable change in indication has actually occurred*Supplemental NDE is standard practice for validating TOFD indications-Eddy Current is a reliable confirmatory test for PWSCC surface connection. Supplemental ET of J-weld & Penetration OD in RFO 12 and 13 refuted PWSCC hypothesis.-There was no shear wave obscuration of backwallfeatures or distributed reflectors which would be expected behind a crack face 18 Conclusion*Summary of Facts-PWSCC disposition in RFO 13 was based on refuted data-All supplemental exams were negative for PWSCC-Indication in CEDM 56 lacks PWSCC features-Indication is consistent with grain noise -Current Westinghouse evaluation methodology would conclude no defect is present in CEDM 56*SCE'sConclusion-The indication in CEDM 56 is not a defect -Code Relief for Flaw Evaluation is not required 19 Future Plans*Continue with EA-03-009 examinations every refueling outage *Respond to any new indications or unexpected changes as appropriate*Replace the SONGS 3 Reactor Vessel Head during RFO 17 (approximately Fall of 2012) 20 Licensing Licensing correspondence associated with CEDM 56 21*Relief Request ISI-3-8: -Contingency submitted prior to RFO 13 "Embedded Flaw" Repair (< 75% t-w) --

Used for 3 CEDMs during RFO 13*Submitted December 3, 2003*Supplemented March 15, 2004*Approved May 5, 2004

  • No Commitmentsmade at this time Licensing 22 Licensing*Relief Request ISI-3-13: -Submitted during RFO 13 "Embedded Flaw" Repair (>75% t-w) Specifically for CEDM 56*Submitted October 26, 2004*Supplemented December 2, 2004**Approved December 23, 2004
    • Commitment:

SCE will identify a long-term repair method and implement that

repair in RFO 14 23 Licensing*Meeting with NRC Staff:-March 2, 2006, presented long term disposition of CEDM 56-Existing repair is adequate for one cycle-SCE will inspect to validate stability-SCE will submit 2 relief requests*If no discernable growth -Operate 1 cycle (ISI-3-21)*Contingency repair alternative (ISI-3-22)*Teleconference with NRC Staff-March 22, 2006, agreed to above approach 24 Licensing*Relief Request ISI-3-21:

Allowing 1 cycle of operation contingent upon no discernable

growth (<0.02")-Submitted May 11, 2006 (Prior to RFO 14)-Teleconference November 14, 2006 (During RFO 14)*-Supplemented November 20, 2006 (During RFO 14)**-Verbal Approval December 1, 2006 (During RFO 14)-Written Approval February 7, 2007

  • SCE informed NRC no discernable growth
    • Commitment:

SCE to either reclassify as NDD and withdraw Relief Requests ISI-3-21

and ISI-3-22 or will repair during RFO 15 25 Licensing*Relief Request ISI-3-22:Contingency repair in RFO 14 to allow operation to RFO 15 if growth was greater than or equal to 0.02" (no discernable growth) 26 Licensing*Based on all information available to date, SCE reclassified the indication in CEDM 56 as NDD-August 31,2007, SCE documented completion of NDE evaluation and withdrew Relief Request ISI-3-21 27 Questions 28Beaver Valley ODCEDM 56Figure 2: Comparison of penetration tube ET results for OD PWSCC 29Figure 3: Vertical (UT) slice of CEDM 56 (gray section) showingindication and grain noise throughout penetration tube base material 30Figure 4: Unobstructed detection of OD weld overlay interface onCEDM 56 70° /40° circumferential shear wave UT analysis of CEDM 56Overlay weld interface