IR 05000269/1986011
| ML16161A735 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 05/30/1986 |
| From: | Blake J, Economos N NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML16161A734 | List: |
| References | |
| RTR-REGGD-01.150, RTR-REGGD-1.150 50-269-86-11, NUDOCS 8606190544 | |
| Download: ML16161A735 (14) | |
Text
c oiREG(,
UNITED STATES o
NUCLEAR REGULATORY COMMISSION REGION 11 101 MARIETTA STREET, ATLANTA, GEORGIA 30323 Report Nos.:
50-269/86-11 Licensee: Duke Power Company 422 South Church Street Charlotte, NC 28242 Docket Nos.: 50-269 License Nos.:
DPR-38 Facility Name: Oconee 1, 2, and 3 Inspection Co
arch20 and 21, 1986 Inspect r:
/
4a N. Ech s
Date'Signe Approv d b o 6 J. /JB ake, Section Chief Date 'Signed Eg neering Branch 1v sion of Reactor Safety SUMMARY Scope:
Special, announced inspection in the area of Unit 1 reactor pressure vessel (RPV) Shell-to-Flange weld inservice inspection (ISI) results; the followup actions taken by Region II and NRR to resolve concerns over the adequacy of the examination of this weld; and the analytical method(s) taken to assure weld and vessel integrity before resuming operatio Following the closing of this inspection, additional telephone discussions between the licensee, Region II and NRR staff covered several concerns including:
adequacy of the ultrasonic examination, characterization of the indications, evaluation and description of the method(s) used to analyze the indications as required by the Code. Also, the licensee disclosed that as a result of the above effort, they believed the indications were service induced rather than fabrica tion related as reported on March 21, 198 Because of the above concerns and disclosure, the licensee met with the staff on April 8 and 21, 1986, in the Region II office to present details on these matter During these presenta tions, the licensee attempted to show that: (1) the majority of the indications were located in the base material above the weld, (2) that, in all likelihood, most of the indications were attributable to flange geometry, reactor vessel stud hole location and cladding, and (3) that if the indications were assumed to be cracks, calculations based on code approved Linear Elastic Fracture Mechanics (LEFM) approach, showed the weld was acceptable and therefore, the vessel was fit for servic PDR ADOCK 05000269 G
REPORT DETAILS 1. Persons Contacted Licensee Employees
- M. S. Tuckman, Station Manager G. W. Grier, Corporate Quality Assurance (QA) Manager Neal Rutherford, Systems Engineer Licensing J. T. McIntosh, Superintendent, Station Services, Oconee J. 0. Barbour, QA Manager, Operations Paul Guill, Licensing Engineer
- R. J. Brackett, Senior QA Engineer
W. R. Hunt, ISI Coordinator, Oconee Other Organizations Babcock and Wilcox (B&W), Special Products and Integrated Field Services H. E. Stopplemann, ISI Coordinator, Oconee R. W. Penoza, Manager, Plant Service Tooling T. E. Smith, Contract Management R. A. Michalski, ARIS Supervisor Louis Redd, Structural Engineer M. G. Hacker, Level III Examiner, Inspection Engineering Oak Ridge National Laboratory R. W. McClung, Consultant to NRR John Merkle, Consultant to NRR Electrical Power Research Institute (EPRI)
A. J. Willets, Senior Engineer, Inspection Applications Services NRC Staff Attending Region II Office Meeting on April 8 and 21, 1986 A. Gibson, Director, Division of Reactor Safety R. D. Walker, Acting Deputy Administrator V. W. Panciera, Deputy Director, Division of Reactor Safety (DRS)
L. A. Reyes, Acting Director, Division of Reactor Projects (DRP)
A. R. Herdt, Chief, Engineering Branch, DRS V. L. Brownlee, Chief, Reactor Projects Branch, DRP J. J. Blake, Chief, Materials and Processes Section, DRS Nick Economos, Metallurgical Engineer, Materials and Processes Section, DRS Helen Pastis, Project Manager, NRR Dennis Crutchfield, NRR/PWR-B
The staff reviewed the licensee's submittals along with additional information requested during the April 21, 1986, meeting and concluded that Oconee Unit 1 may restart and operate for eight (8) weeks on the condition that the licensee submit, before June 16, 1986, results on the following work effort:
-
Perform a comparison of Units 1 and 3 vessel flange geometry, material(s) and cladding and determine why Unit 3 ISI examination of the same weld did not produce similar results
-
Conduct an ultrasonic examination study on the Mt. Vernon mock-up vessel
-
Review of original weld design and fabrication history including nondestructive examination records of the weld in question In addition, the staff stated that the NRC reserves the right to reevaluate its position on the above, in the event that a significant transient occurs on the Unit 1 vesse On May 2, 1986, NRR issued a safety evaluation report (SER) which concluded that Oconee Unit 1 can be safely returned to full power and operated with actual flaws of the size and circumferential locations described in Attachments 1 and 3 (Table 1 and Figure 1) of this report. This conclusion was based on the staff's review of the fracture mechanics analysis evaluation performed by the licensee, the pressure-temperature limits, and Pressurizer Code Safety Valve Set Point contained in the Oconee Unit 1 Technical Specifications (TSs). This conclusion has been supplemented with the following conditions:
-
The licensee will submit prior to June 16, 1986, a technical report summarizing their on-going ultrasonic testing program on the Mt. Vernon mock-up vesse The staff will review and determine whether the conclusion that the subject flaw indications are enveloped by the dimensions measured by the licensee is still conservativ Because the staff considers several of the flaws as conditionally acceptable per IWB-3122.4, augmented inservice inspections based on 10 CFR 50.55.a(g)(4) will be require At least six months before the next scheduled refueling outage, the licensee will provide a report describing detailed plans for the above ISI inspection(s).
Results:
No violations or deviations were identifie )
C. Cheng, NRR/PWR-A B. J. Elliot, NRR/PWR-A W. Hazelton, NRR, Division of BWR Licensing W. J. Collins, Inspection and Enforcement, Engineering and Generic M. R. Hum, NRR, Division of Engineering Communications NRC Resident Inspectors
- J. C. Bryant, Senior Resident Inspector
- K. Sasser, Resident Inspector
- Attended exit interview 2. Exit Interview The inspection scope and findings were summarized on March 21, 1986, with the persons indicated in paragraph 1 above. The inspector indicated that he had reviewed the results of the ultrasonic examination of the reactor vessel shell-to-flange weld. Because of the apparent size and number (22) of code rejectable indications identified, the inspector determined that additional ultrasonic examinations should be performed to ensure that the full extent of the indications, e.g., size, shape, location, etc., were established before attempting to evaluate the acceptability of the subject wel In response, the licensee's representative stated that the subject weld had been examined in accordance with code requirements, that the indications were believed to be related to weld fabrication, that B&W had evaluated/
analyzed the indications per IWB-3500/3600 and concluded the weld was acceptable, and therefore, no further examinations were being planned at this time. The inspector informed the licensee that he was not in agreement with this position and would communicate the information to the Region II for further evaluatio. Licensee Action on Previous Enforcement Matters This subject was not addressed in the inspectio. Unresolved Items No unresolved items were identified during this inspectio. Ultrasonic Examination of Reactor Vessel Shell-to-Flange Weld (73755)
(Unit 1)
On March 18, 1986, the Oconee senior resident NRC inspector informed the Region that results of the ISI ultrasonic examination (UT)
on the reactor vessel shell-to-flange weld indicate the presence of code rejectable indications that had not been seen in previous UT examination On March 20, 1986, the inspector arrived at the site to discuss the examination results with cognizant licensee personnel and to review the dat The weld in question was identified as #1RPV-WR19, shell-to-flange
wel The applicable code governing ISI examinations is ASME Section XI, 1980 Edition and addenda through winter 198 Inspection requirements for this weld are delineated in Table IWB-2500-1, Category B-A item 81.30. The examination was conducted by B&W in accordance with procedure ISI-130, Re, Ultrasonic Examination of Vessel Welds and Nozzle Inside Radius Section. The weld was examined from the cladding side of the flange surface with a 00 angle, 2.25 MHz one (1) inch transducer. The licensee's ISI plan required that only 50% or 1800 of the weld circumference be examined during this outage but upon finding code rejectable indications, this examination was expanded to include the entire weld circumferenc The centerline of the weld is approximately 32" from the examination surfac Scanning was performed on the clad flange sealing surface, or the region between the 0-ring grooves and the unclad ligamen The flaw indications were sized during the detection phase using standard ASME Code Section XI pulse echo techniques as summarized in Table 1. Flaw indication with the maximum ultrasonic amplitude is shown in Attachment 3 (Figures 2 and 3).
The flaw indications have several significant features:
their apparent size, the relatively large number, and the high ultrasonic amplitude. The two longest flaw indications have relatively high ultrasonic amplitude when considering that the reflection from the back of the basic calibration block is approximately 1000% DAC (Distance Amplitude Correction).
The majority of the indications were located in the base material of the flange above the weld. The 1-inch diameter transducer has approximately an 8-inch beam diameter at a depth of 32" (weld centerline).
All indications except one were detected from the clad surface out-board of the 0-ring groove These indications were not observed during examinations from the clad region between the 0-ring grooves or from the unclad surface. Only one of the indications (#20)
was recordable from the unclad side of the flang This indication is located between two stud holes and is at a depth of 24" from the flange fac This is outside (above) the area required to be examined by ASME Section XI and appears to correspond to geometric discon tinuties on the inside diameter (ID) and outside diameter (OD)
of the vesse This indication was observed between stud holes 3600 around the vessel although not at a recordable leve This examination identified a total of 22 code rejectable indications which are listed in Attachment 1 to this report. Examination results of this weld from previous ultrasonic examinations are summarized in Attachment During the 1970 preservice (PSI)
examination, manual contact examinations were performed from both the ID and the OD of the reactor vessel using 00 and 450 angle transducer No indications were recorded from either the ID or the OD of the vessel during the PS During the 1977 and 1979 inservice inspections, an automated examination was performed on the weld areas which lie beneath the unclad areas of the flange using the immersion technique with 12" water path distance. No indications were recorded from the unclad area of the flange in either the 1977 outage (00 angle) or the 1979 outage (00 and two 50 angles).
In 1981, the 10-year automated vessel inspection was performed from the vessel ID only. The immersion technique was again
used with a 7" water path distance and 0, 450, 600, and 700 examination angle The 700 angles were used for the examination of the near surface region in accordance with NRC Regulatory Guide 1.15 No indications were recorde On March 21, 1986, the inspector reviewed these data and discussed the findings and U/T technique details with cognizant B&W/licensee personnel who stated that their evaluation of the U/T data showed the indications were related to weld fabrication. However, the inspector concluded that because physical constraints on the flange surface, e.g., 0-ring grooves, cladding and stud holes limited the inspection effort, the credibility of the in spection results was in questio Therefore, the inspector could not determine, with any degree of confidence, whether the reported indications had been fully investigated and accurately sized as required by the Cod Lacking this degree of confidence in the examination and data, the inspector stated that he could not see how the licensee could declare the indications acceptable per IWB-3500/3600 criteria and therefore, determine the vessel fit to return to powe Instead, the inspector suggested that additional U/T examinations be conducted using transducers of different sizes and angles in order to verify the credibility of the test dat The licensee disagreed and the inspector stated that he would discuss the matter with the Region and determine the course of action that would be taken to resolve the issu The inspector discussed the licensee's position with Regional management and arranged for a telephone conference between the Region and the licensee's management which took place prior to close of business on March 21, 198 The inspector did not participate in this conferenc Following this inspection, Region II discussed with NRR details of the Oconee Unit 1 reactor vessel U/T examination results and requested assistance in evaluating the adequacy of the examination and the fracture mechanics analysi Subsequently, the staff telephoned the licensee to discuss the matter and thereby obtain additional information on the examination and evaluation of result During these conferences, the licensee disclosed that because of further evaluation of the data, they now believed the indications were service-induced rather than weld fabrication related as believed earlie During these discussions, it was agreed to hold a meeting in the Offices of Region II on April 8, 1986, when the licensee would be provided the opportunity to present the U/T examination results and the analysis used to determine the acceptability of the indication On April 8, 1986, the licensee presented the staff details of the examination including the U/T technique used, results, typical sketches showing approximate locations of indications with respect to the weld and stud hole, flaw size to aspect ratio graphs used to evaluate acceptance/
rejection of indications allowed by ASME Section XI, Appendix A and a historical account of examinations performed on this weld as shown in Attachment Through this presentation, the licensee attempted to demonstrate that the indications could be geometric reflectors resulting from interactions with
the inner and outer surface of the vesse To support this proposition, the licensee stated that he conducted experimental U/T examinations on a similar flange-to-shell weld on a mock-up vessel at the Mount Vernon facility of B& During these tests, geometrical reflectors were confirmed that provided signals with. time-of-arrival similar to those observed in the Oconee vesse However, the experiments were not performed with a calibration block to establish appropriate inspection sensitivity, nor was any further evaluation done on the geometrical indications to determine the nature or size of flaw that could have been inferre Despite the recognition that the indications could be spurious geometric reflectors, the licensee's inability to verify flaw size accuracy, forced him to assume that the indications were from real flaws/crack This premise resulted in an analysis per ASME Section XI based on the apparent sizes from the 50% amplitude-drop metho On this basis, less than half of the indications had sizes and aspect ratios that were acceptabl The majority of the indications required further analysis (according to Section XI of the ASME Code) by fracture mechanics to determine whether they were acceptabl Even though the indication signals indicated that the alleged flaws were well below the surface, as an act of conservatism, they were analyzed as though they were on the surface. With this premise, the analysis apparently determined the indications to be acceptabl However, at the close of this meeting, the staff still had some concerns which included the licensee's uncertainty about the actuality of such flaws and concern over the accuracy with which the indications were sized for fracture mechanics analysi The latter concern arose in part from the limitations on scanning to establish flaw boundarie These limitations were imposed by annular grooves in the cladding (and the edge of the cladding itself) that limited the inspection surface in the radial direction on the flange. This, coupled with the variable size of the signal through the austenitic stainless steel cladding, reduced the confidence that the full extent of the flaw in the through vessel thickness direction was being measure The staff felt that assurances were.needed to show that the apparent lateral size of the indication had not been influenced by scanning limitations (e.g., that the loss of amplitude to 50% was not biased by the search unit partially overlapping a groove or the edge of the cladding).
The meeting concluded with the staff requesting that the licensee meet again on April 21, 1986, in the offices of Region II to provide additional in formation concerning transducer centerline location with respect to flange surface, determination to maximum acceptable flaw by fracture mechanics analysis, and results of additional U/T examination on the mock-up at the Mt. Vernon facilit In the interim, Region II determined that the integrity of the Oconee reactor vessel would be regarded as acceptable by the NRC until additional examinations could be performed during the next refueling outage if either of the following conditions were met:
-
Tests on the Mt. Vernon mockup reactor vessel, which is similar in design to the Oconee 1 vessel, using the same test equipment and configuration that was used at Oconee, establish that indications observed in the Oconee 1 vessel were likely geometric reflections and not actual defects, or
-
Fracture mechanics analysis demonstrates that sufficient vessel strength remains even if all observed indications are actual defects and defects exist under the entire area of the stud hole On April 21, 1986, the licensee accompanied by representatives from B&W and EPRI met in the Region II offices as agreed earlie The licensee provided the information concerning transducer location and results of the fracture mechanics analysis on maximum flaw size acceptable by the code as had been agree The licensee concluded that the technique used in sizing the indications provided conservative results, that the fracture mechanics analysis showed the reactor vessel was fit for service, and that a considerable amount of margin existed prior to exceeding code limitations for acceptabilit At the conclusion of the meeting, the staff stated that because the licensee has not provided any information that proves conclusively, the indications were geometric reflectors, the NRC assumes the indications were actual flaws/cracks which developed at a linear rate between the 1981 and 1986 U/T examination Also, the staff stated that, based upon the information by the licensee during this meeting, Unit 1 can be safely restarted and can operate for eight weeks provided that the licensee submits the following additional information and the staff's review of this information supports the preliminary conclusion that Unit 1 can be safely restarte Information required by April 29, 1986 What are the thermal, pressure and boltup stresses across the flange-to-shell weld during the limiting transient conditions?
2. What are the limiting transient conditions?
Why are emergency faulted conditions (i.e., Pressurized Thermal Shock) not limiting conditions for the flaw evaluation?
3. What are realistic flaw sizes when considering beam width magnification and the loss of signal amplitude from clad geometry? Assuming that flaws grew to their present size (Item 3) in the last five years (no flaw observed in 1981 inspection), what is the growth rate?
Considering this accelerated growth rate, what will be the maximum predicted flaw size at the next refueling outage?
5. Based on the flaw size in Item 4, perform flaw evaluation per IWB-3600 of ASME Section XI, and submit results to the NR Information required by June 16, 1986 1. In the flange to shell region, compare geometry, material and cladding for Oconee 1 with that of Unit 3 (previously inspected within UT geometry indications).
Why does Unit 1 have U/T geometry indications and Unit 3 does not?
2. For the flange-to-shell weld in the reactor vessel of Oconee Unit 1, what was weld geometry, weld process and inprocess inspections per formed on this joint and were there any significant weld repairs?
Confirm weld geometry through review of shop weld information and processin The staff also requested the licensee to notify Region II inspection personnel of pending U/T examinations on the mock-up vessel at the Mt. Vernon facility and to provide a report on this activity as appropriat Within the areas inspected, no violations or deviations were identifie Attachments:
1. Table 1, Summary of Indication Dimensions 2. Table 2, Comparison Table of Examination Techniques, Variables and Results 3. Figures 1, 2 and 3
)N
ATTACHMENT 1 TABLE 1 SUMMARY OF INDICATION DIMENSIONS INDICATION INDICATION LENGTH WIDTH AMPLITUDE DEPTH
.7"
.8" 65%
30"
2.6" 2.1" 159%
32" 2*
.7"
.45" 65%
29" 3/4 7.4" 2.1" 100%/1600%
30" 5*
.4"
.4" 50%
30" 6*
.5"
.5" 50%
30" 7*
.4"
.5" 50%
30"
2.0" 2.25" 501%
32" 9*
2.8" 1.9" 80%
30.5"
6.0" 1.2" 1000%
11 3.8" 2.0" 159%
12*
1.4"
.55" 100%
29.5"
2.8" 1.0" 159%
14*
1.1"
.65" 100%
29.5" 15*
1.2"
.4" 100%
29.5" 16/17*
1.8" 1.7" 65%/159%
31.5"/32" 18/19*
4.4" 2.3" 125%/100%
32"/31.5."
2.7" 1.55%
200%
24"
2.7"
.7" 159%
24" ALL LENGTHS DETERMINED BY 50% DAC THE BACK REFLECTION FROM THE CALIBRATION BLOCK IS APPROXIMATELY 1000% DAC
- INDICATES WIDTHS DETERMINED BY 50% DAC
O2NPARISN TABLE OF EXAMINATICt TEDCHICXES, VARIART 3;, & RESULTS EXAMINATICrN VARIABLE 1970 EXAMINATION 1977 EXAMINATION 1979 EXAMINATICtq 1981 EXAMINATICN 1986 EXAMINATICN TECDNIQUE (rNrACT-MANUAL D+tERSICN -
12" D+ERSICN -
12" IRERSICN -
7" 0(7rACT-MANUAL WATER PAH WATER PAT WATER PATH EXAMINATCT SURFACE VESSEL ID & CD UNClAD AREA C0 (JULAD AREA (N VESSEL ID CLAD AREA C L C1D AREA FLANE FLAGE FINKE Ct FLAvkE CALIBRATION BLCK
- 7 40342 40342 40390 50304 40387 EXAMIAICr ANGLES 00,450
00, 50 00,450, 600,700
00 ULTRASCNT C INSTRMIEr SPERRY 735 USIP-11 USIP-11 KB-6000 USIP-11 USIP-1I SEARCH UNIT 2.25 Htz z Al4ATION ALL 2.25 MH.25 MPz 2.25 MR.25 H1 AUTO4ATICN AUn2ORATICN &
KB-AE
~li SEARCH UNIT CABLE &
N/A TRIAX-425'
TRIAX-425'
TRIAX-475'
OIAX-20'
COAX-20'
]LEN~fli REFLu'IOR DEPTS N/A 23.75",31.75",
23.75",31.75",
3",6",9",12" 8*,18",32" 23",24.5",
39.75" 39.75" 26" REFERENCE SENSITIVITY N/A 62DB 64DB 28-43DB 46DB 30DB EXAMINERS M. A. YOUNG T. F. MCDE'R1T R. A. kICHA1SKI W. H. ZIM4ER J. C. WALDEN C. E. 7TIC4S(W H. I. NEWITN B. P. WEAVER L. M. LIBERATORE R. J. PATTERSCN C. R. MEREDITT1 R. A. KICHALSKI R. J. PAINE RESULTS N RECORDABLE 'N RECORDABLE A 3600 INDICATIO.
NO RECDRDABLE 22 INDICATICNS INDICATICNS INDICATICUS WAS SEEN DUE TO INDICATIONS REXCRDED -
ALL FLAtE GE14ETRY
.
OF WIUQa WERE RECORDED FFEN ThE CLAD AREA NOTE:
CALIBRATION BLOCKS 40342 AND 40387 ARE APPROXIMATELY OF TfE FLANGE 15DB LESS SENSITIVE THAN 5030 N INDICATTCOS REClRDED NC UtCIAD SIDE OF FLAME
ATTACHMENT 3 FIGURE 1 OCONEE UNIT I REACTOR VESSEL
- 2o
RE *24 4S"S eg
&
013 r
0*%
us27
.
Y
ATTACMIENT 3 Pg. 2 FIGURE 2 -'FLAW INDICATION P4 ri-7
1
_1I
-
4 mf'.
II I
I I
IL
.1
.
a1 Ptb F*
.
O
ATTACHMENT 3 -
pg FIGURE 3 -
CIRCUMFERENTIAL LOCATION OF FLAN INDICATION #4 ROLE laBOLE
- 13 BOLE #
BOLE
IO. SWAM
+-
-4s.-r-.
INDICATION #4 SUACZ 2 NOTE: X. Transducer centerline Location DUKE POWER COMPANY OCONEE UNIT #1 WELD NO.: _RY-19 FIGURE MO.: 801.030.001A CONT. HO.: 580-2034-12-02 Jwe
.v
£JsIA1Th
4-l 0e