IR 05000261/1990024
| ML20058H982 | |
| Person / Time | |
|---|---|
| Site: | Robinson  |
| Issue date: | 11/16/1990 |
| From: | Blake J, Coley J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20058H970 | List: |
| References | |
| 50-261-90-24, NUDOCS 9011260243 | |
| Download: ML20058H982 (10) | |
Text
.,:
'
.
'
e
- [ptfo
'
UNITED STATES NUCLEAR REGULATORY COMMisslON
'
^
[
REGION ll
-
n g
j 101 MARIETTA STREET. N.W.
^*
ATLANTA,0EORGI A 30323
.
- ..+
i Report No.:
50-261/90-24 Licensee:
Carolina Power and Light Company
'
t P. O. Box 1551 Raleigh, NC 27602
. Docket No.: 50'-261 License No.:
DPR-23 Facility Name:
H. B. Robinson
,
Inspection Conducted: October 29 - November 2, 1990
~ Inspector:
k 5.-
1/-/f-90 ru J. L.
tg Date Signed
' Approved by:
W
//
-
_
J.
./ Slake, Chief D' ate Signed M dials and Processes Section E gineering Branch
' Division of Reactor Safety SUMMARY Scope:
,
Thisl routine,' unannounced inspection was conducted in.the areas of observation'
,
~
- of inservice inspection work activities, review of~ radiographs, observation of
-
codeL repair activities for A and C accumulators, implementation of Generic Letter.-90-05, and-review of licensee - actions relating to NRC Compliance
<
Bulletin 87-02(Technical-Instruction 2500/27).
Results:
-i Inservice inspection efforts at H. B. Robinson have effectively identified two
'
significant areas of ' material degradation-caused.by integranular stress corrosion cracking (IGSCC). :The two areas were (1) the control rod guide tube
. support pins and, (2) the upper level transmitter nozzles for A and C accumulators.
Automated ultrasonic examinations of the outlet nozzle to reactor: vessel weld (RPV-33) reveale< an indication which required extensive-investigation priorlto the final accliptance of the indication.
In addition,
,
. preliminary ultrasonic data from examinations being conducted on Steam
?
Generator A, Weld 5, (upper girth weid). has also revealed indications that may
"
be 4dicative of cracks.
CP&L's,rianagement, engineering and inspection-personnel have responded-very effectively to insure that technical issues are resolved in a manner that will inrure plant safety.
In the areas inspected, one violation was identified 50-261/90-24-01, " Welds not Identified on LInservice Inspection Isometric Sketches" paragraph 4.. No deviations were identified.
[$h130$$b $bbIf.f2
o
-
.
-
. -
-
- -
,
-
.,
,,
,
l REPORT DETAILS l
1.
Persons-Contacted Licensee Employees l
- R. L. Barnett, Manager, Outage-and Modifications
- W. M. Biggs, Manager, Site Engineering-Unit
- R. D. Crook, Senior Specialist, Regulatory Complience
.
- C, R. Dietz, Site Project Manager, Robinson Nuclear Power Division
- C, H. Griffin, Senior Engineer, Nuclear Engineering Departr>cnt (NED)
- E. M.- Harris, Manager, Onsite Nuclear Sofety
- J. D. Kloosterman, Director, Regulatory Compliance
>*G. V. Nuckols, Engineer, Nuclear Engineering Department
- C, R. Osman,- Principle NDE Specialist, Technical Services Department
- M. F. Page, Manager, Technical Support-
- M. F. Powell, Nuclear; Engineering Department
- R. M. Smith, Manager, Maintenance
!
- D. C. -Stadler, Onsite Licensing Engineer
-*R. B. Weber, Senior. Specialist, inservice inspection
- H. J. Young, Manager,-QC/QA Other licensee employees contacted during this inspection included engineers, technicians, and office personnel.
NRC Resident Inspectors
. L. W. Garner,_ Senior Resident inspector
.
- K. R. Jury, Resident Inspector
'
- Attended exit interview on Noverrber 2,1990-2.:
InserviceInspection(73753) Unit 2
,
.
-The. inspector observed inservice inspection (ISI) work and work activities to ascertain whether examination, repair, and replacement activities
"
t associated with - ASME Class 1, 2, and 3 components were performed in
,
accordance with Technical ' Specifications, the --applicable section and
-
revision of the ASME Code, correspondence between NRR and; the licensee
,
,
'concerning relief requests, and requirements imposed by NRC/ industry initiatives.
H. B. Robinson is presently in the fifth and last outage _ of the second inspection interval.
The applicable Code for this' interval is the ASME B&PV Code,.Section XI, 1977 Edition, with addenda through Summer
'
'1978..This inspection was conducted to supplement surveillance activities
,o reported in Inspection Report 50-261/90-2 a-4
.-
.
t
'
- a.
Volumetric examination of Vessel / Nozzle Welds using the automatic Ultrasonic Technique The> inspector returned to the H. B. Robinson facility on October 29, 1990 and discovered that all automated in-vessel ultrasonic examinations had been completed on October 28, 1990.
Thc vessel / nozzle examinations were perforned by Southwest Research Institute (SWRI) exeminers utilizino Solic Mark 11 ultr6 sonic instruments.
The ultrasonic data 'uas thtn processed through an enhanced date acquisition syue (EDAS) which allcked the data to be
'
recorded and displayed with computer graphics.
H. B. Robinson Ultresonic Procedure "AUT-15" was used to conduct the vessel / nozzle xaminations.
The -ultrasonic examinations had revealed recordable e
indic.ations in several vessel and nozzle weldt.
The inspector
'
,
selected the three welds with indications to evaluate the SWRI examination and evaluation results.
The three welds selected are listed below:
Weld No.
Examination No.
}IeldDescription RPV-29:
Outlet Nozzle to Shell at ten. degrees RPV-33
Outlet Nozzle tc Shell at 250 degrees
,
,
,
RPV-4 P9 Intermediate shell to
'
Upper Shell longitudinal
-Weld.
~
.In addition: to the above. the inspector reviewed certification j
records for all'SWRI examiners and equipment on site.
'
,
f Review of the data for welds No. RPV-29 and RPV-4 revealed that the indications recorded in these welds were acceptable and indicative of
small slag inclusions which resulted during fabrication of the
!
vessel.
However, an indication was detected during the examination of nozzle to shell weld, RPV-33, which was conducted with 0 degree t
=1ongitudinal and.45 degree. shear wave search units located in the
{
nozzle bore.
The indication was detected with the' 45 degree search
?
unit and appeared to be located in the weld material at an azimuth of
.!
237 degrees.
A subsequent re-look. of the area containing the reflector was - performed using the' 45 degree scarch unit that
!
initially detected the indication.
This.re-look confirn4d the existence of the indication.
.a This initial re-look shewed at least two individual reflectors which when combined, constituted an indication that appeared to be
,
rejectable relative to the Section XI,1977 with Addenda through the Summer of 1978, Table IWB-3512, acceptance standards.
l-
'
i
.
-
- - - - -..
.
. _,
-
,
.
,-.
.,
Subsequent re-examinations were performed by conducting three separate but identical calibrations using the same 45 degree search unit used to initially detect the indication.
Two different instruments were used in order to assure that an instrumentation problem did not exist.
Three separate sets of data were collected and correlated to provide the final sizing date peckage. These data
sets all showed the indicatior to be significantly lower in amplitude than the original re-lock ant accepteble in accordance with Section XI criteria. Additionally a focused 45 degree, 2.25 mHz, shear-wave, search unit, a
35 degret, longitudinal-wave, 2,25 mHz, tip-diffraction, search unit and a O degree, 2.25 mHz, tip-diffraction, search unit were used to scan the indication area.
The indication could not te detected with either of the tip diffraction units or with ths 45 degree, 2.25 mHz, focused unit.
- The conclusion drawn from this data package is that the indication is of an acceptable size in accordance with the criteria of Section XI, 1977 Edition with Addenda through the Sunener of 78, Table IWB-3512.
-The apparent cause (supported by calibration data) of the amplitude difference between the initial re-look examination and the subsequent re-examinations was due to.a poor cable connection.
Fabrication radiographs were also reviewed by the inspector for weld RPV-33.
- These radiographs revealed several small (acceptable) slag inclusions in this weld.-. However, the orientation of these radiographic indications with the' indication detected during the examination of the nozzle could not'be confirmed,
-
b.
Ultrasonic examination of-Steam Generator A Weld 5 (Upper Girth Weld)
The inspector also observed Westinghouse examiners conducting manual ultrasonic examinations on= Steam Generator A. Weld 5.
These were licensee augniented examinations, since the ASME Code required examinations for this inspection interval had been satisfied with the ultrasonic examination of weld 5 for Steam Generator B (see Re9hn II Inspection Report 50-261/90-21 fordetails).
.The licensee's purpose for expanding their ultrasonic examination of
'
weld 5 to-include Steam Generators A and C was to determine whether any of H. B. Robinson steam generators were cracked.
Significant Cracking adjacent-to this weld has been detected -and repaired at a number of sites including Indian Point, Surry, and Zion.
Although,
~
the licenstie ultrasonic examination of Steam Generator B recorded somet isolated indications, these indications were not considered
-
indicative of cracks. However, preliminary in-process examination of
'
Stecm Generazor A, weld 5, has ~ recorded indications which have been tentatively plotted.
These plots indicate that the ultrasonic reflectors are at the vessel ID and that they run in a circumferential direction ranging from 3e outer edge of the weld to one inch beyond
.,
- - _ _ - - - _ _ _ - _ _ - - - _ - _ - - - _ - _ _ - __ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _
-
.
.
.
,
-the weld edge in the base material.
All of the indications were of relatively low amplitude thus far.
However, these indications are located where cracks have been confirmed at other utilities.
Examination, sizing and evaluation efforts are severely restricted for.the H. B. Robinson steam generators due to a shield wall constructed around each steam generator.
Therefore, during the inspector's exit meeting these restricted ccnditions were presented to the plant management, along with NRC's recommendation that the
'
secondary side of the steam generator be opened, end weld 5 be enemined using the magnetic particle examination method.
No commitment to open the steam generator was made by plant management during the exit. meeting, nor was one expected at'that time, c.
sRepair and Replacement Activities
~(1) Repair Activities While performing a hydrostatic test on the Safety Injection (SI)
Accumulator "C", a leak was detected in the stainless steel nozzle coupling for one of the upper level transmitter lines (2-inch diameter, 2000# half-coupling, pipe line number 2-SI-601R).
There are three such S1 Accumulators at the RobinsonNuclear' Plant (RNP).
The RNP SI accumulators were manuf actured for Westinghouse by the Delta Southern Corporation, formerly located in Baton Rouge, i.
. Louisiana. Delta Southern Corporation is no longer is business.
. A search by the licensee of the Nuclear Plant Reliability Data
,
System (NPRDS) revealed that the. Prairie Island Plant (Northern States Utilities) had previously experienced a similar leak in one of their Delta Southern Corporation accumulators.
Their-nozzle failure was also in a two-inch diameter level transmitter nozzle.
Westinghouse: had-performed a failure analysis of the Prairie Island nozzle and determined the cause of failure to have been intergranular. stress corrosion cracking (IGSCC). The stresses responsible for. initiating the IGSCC in the Prairie-Island nozzle were believed to have been the result of an-improperly fit-up and welded socket weld (pipe apparently bottomed into the socket prior to welding).-
Failure analysis testing on the H. B. Robinson-(HBR) failed nozzle was performed by the CP&L Metallurgical Services Section-
'
located at-the Harris Engineering and Evaluation (E&E) Center (New Hill, N.C.).
The failure analysis results for the HBR nozzle inuicated the cause of the leak to be IGSCC.
It should-be noted that for IGSCC to initiate, the following three conditions must exist simultaneously:
tensile stresses, sensitized base material, and a corrosive environmen *
.
'
..
.
The axially oriented crack had initiated on the ID of the coupling.
Although the crack appeared to extend along most of
<
the coupling ID length which was contained within the tank shell, that portion of tb crack that was open to the coupling
- OD was cnly approxinetely 3/16 inch long and did not appear to extend into the nozzle attachment fillet weld reinforcement.
Unlike the Prairie Island nozzle, socket joint fit-up and welding wa:; not a contributina factor in the failure or the HBR level transmitter nozzle coupling.
CP&L believes that the full
'
penetrationweld(double-Vgroovedesign),conneri.ingthenozzle to the shell, is the source of residual stresses which initiated the cracking. The axial orientation of the cracking seems to provide. son:e credence of this hypothesis.
- r The failed nozzle coupling was sensitized. The. sensitization of the coupling was ettributed to a ccrbination of the welding to the coupling and the post weld heat treatrent (PWHT) perforacd by the manuf acturer af ter the completion of-the welding to tne accumulator.
Typically, when a manufacturer plans.to perform PWHT on a component which contains stainless steel materials such as the 304/316 grades, a low carbon grade of the material containing.035 percent maximum carbon content-(e.g., Types 304L or 316L). would be recommended.
These low carbon grades of stainless steel are much less susceptible to sensitization and-are,therefore typically considered immune to IGSCC. The carbon content for, the= failed coupling was determined to be.062 -
percent carbon during the failure analysis.
No - certified material. test reports were found for the-2,1, and 3/4 inch nozzles although, the Manufacturers' Data ' Reports indicated that
',
the' nozzle couplings are SA-182-F-304-ELC.(extra low carbon).
Although the presence of chlorides and sulfates.was confirmed by water leachable analyses of the failed. coupling, the exact corrosive environment which contributed to the crack initiation
!
was not determined.
The licensee subsequently performed liquid penetrant. and ultrasonic examinations on all eight nozzles for each accumulator.-
These examinations revealed that accumulatcr A also had a two inch nozzle with a crack indicative of IGSCC and accumulator B had a forging defect in a one inch no::zle that was apparently caused during fabrication.
The indication extended into the coupling for approximately 1/2 the coupling through-wall and may run the entire length of the coupling but i
did not extend to the ID surface of the pipe.
The inspector reviewed drawings, material certification reports,
'
engineering evaluations of the failure, and engineering evaluations for the repair of the failed nozzles, Structural Integrity Associates repair program for the nozzles and ultrasonic and liquid penetrant examination reports.
In
..
q
,
'
.
,
'
.
.
]
ir
{
addition, cognizant personnel including metallurgist, welding
enginects, system engineers and inspection personnel were l
interviewed to assess the failure, methods of repair, expanded
)
inspection results, determine whether other components were
!
involved, and to determine if augmented inspections would be
,
performed in the future on nozzles that ere susceptible to
'
failure.
The inspector also observed the nozzle excavation areas to determine whether they met applicable code repair
!
criteria for weld repairs without PWHT.
The inspectors review of the above revealed that the licensee
!
-was performing the technical assessment and repair activities in accordance with plant procedures and the ASME Code.
'
(2)
Replacement Activities Occurrences of failed reactor control rod guide tube support i
pins- (split pins) during the early 80's caused Westinghouse to
!
issue a service letter identifying ) plants with potential for-failures (suspect-heats identified.
Robinson could not identify heats associated with their pins at the time but I
assumed that-they were satisfactory since no-problems had been n
' encountered.
During March 1989 operations, a split pin failed i
at H. B. Robinson due to cracking. A part of the failed: pin was l
Linjected into the "C" Steam Generator and its retrieval required a forced shutdown on April 3,1989.
As a ' result, Robinson committed to ultrasonic examination of all split pins during the current outage.
Ultrasonic examinations were. conducted by Westinghouse with results as follows:
38 pins with. indications indicative of crack (represents 36 percent of the 106 total pins)
'
37 pins.of the above revealed cracks in the pin collar
's to sh6nk change of section area 2 pins (including one of the 37) have cracks in the
<
leaf area As a result 'of the high failure rate delineated' above,, the-licensee has' decided-to replace all 106 spilt pins this outage with pins that are -less susceptible to crack.
The inspector reviewed the Westinghouse ultrasonic procedure used to detect the cracks and discussed the inspection findings with cognizant engineers.
(3) Modifications (57090)
In addition to the repair and replacement activities described
. above the inspector audited welding activities by reviewing completed radiographs.
The radiographs reviewed were for helds
m
.
.
.
__
- _ _ _ _ _ _ _ _ _ _ -
-
.
-
,:
,
on an electrical penetration modification.
ASME,Section III, Subsection NE, Class MC was the applicable Code for this review.
Radiographs for the following wolds were reviewed:
Weld ID Diameter Drawing No.
C-5-1 (Cut 1, Repair 2)
10 inches Conax 7EEG-6000 E-1-1 10 inches Conax 7EEG-6000 E-10-1(Cut 2, Repair 3)
10 inches Conax 7EEG-6000
,
d..
Review of Licensee's Implementation of NRC Generic Letter (GL) 90-05 On June 15,1990 GL 90-05 was issued to all holders of operating s
licenses for nucleer power plants.. This GL provided guidance for performing temporary non-code repairs of ASME Class 1, 2 and 3 piping. A summary of the guidance is as follows:
NRC: approved relief requests are required for all temporary-
-
non-code repairs in Classes 1,
2 and 3 piping that is impractical to repair during power operations.
'
-
Flaws detected during scheduled shutdowns must be Code repaired
-
prior to-restart.
Temporary repairs can remain until the next scheduled outage
-
. exceeding 30 -' days, but no later than the next scheduled refueling outage.
-Temporary-repairs-in Class 1 and 2 piping and high energy Class
-
3-piping must meet certain load-baaring requirements similar to that provided by _ engineered weld overlays or engineered -
mechanical clamps.
Temporary repairs in other Class 3 piping may be analyzed usisig
-
the NRC specified. " wall-thinning" or "through-wall flaw"
,
methods.
The inspector discussed GL 90-05 with CP&L's Cutage end Modificaticn Manager and reviewed CP&L's Plant Program Procedure PL-025,
"In-Service Inspection Progrco" and Technical Support Manastrent-Manual TMM-015. " Inservice Inspection Repair and Replacement Program" to ensure that necessary recuirements have been properly established to implement Generic Letter 90-05 Guidance. During discussions with the Manager of 0utage and Modifications the inspector was assured that Robinson would have no non-code repairs on-any ASME Code' Class
<
components when Unit 2 resumes power operations.
Within the areas examined, no violations or deviations were identified.
.
i-
.,.
.
.
.
.
i
<
3.
NRC' Compliance Bulletins (Temporary Instruction 2500/27)
(Closed) NRC Compliance Bulletin 87-02, Fastener Testing to Determine
Conformance with-Applicable liaterial Specifications.
The purpose of this bulletin was to request that licensee's 1) review their receipt inspection requircments and internal controls for fasteners and 2) independently j
determine, through testing, whether fasteners (studs, bolts, cup screws and nuts) in stores at their facilities met required mechanical and chemical specification requirements.
On January 22, 1988 CP&L Letter Serial No., NLS-88-013 was submitted to Region II.
This letter provided the results of CP&L's review and testing program.
These results indicated that 52 of the 54. fasteners tested met the applicable ASTM specifications.
The two fasteners not in complete compliance with their specifications were evaluated ano determined to be acccptable for their intended use.
CP&L's review concluded that their current procedure for receipt inspection and material handling met or exceeded applicable requirements.
On May 22,1989 NRC issued Temporary Instruction (TI)
2500/27.
This instruction was issued to provide inspector guidance in -
evaluating.the adequacy of certain licensee's root cause analysis and the implementation-of corrective actions in response to NRC Bulletin 87-02.
The. instruction listed CP&L's Robinson plant as having one fastener that was significantly out of specification.
The fastener listed was Sample
~
,
RNP-006, a 3/4" diameter ASME SA 193 GRB8.Capscrew.
The sample capscrew met the testing requirements with the exception of hardness.
The'small difference between the measured valve of 26 Rockwell C scale-(equivalent to 102 Rockwell B scale) and the specification maximum off100 Rockwell 8 scale was not considered by the licensee to be signifi-cant.- The. inspector reviewed the licensee's response to Bulletin 87-02 and CP&L's backup data file that supported-their response.- In addition, c
discussions were held with metallurgist working for the licensee and NRC concerning-the two points difference in Rockwell B scale hardness.
It was
.the consensus opinion-of all-the specialists.that the two points difference on= tss Rockwell D scale did not indicate a significant difference in hardness.
Therefore, the inspector concluded that the response to NRC Compliance Bulletin was adequate and no further action is required.
Action on Previous' Inspection Findings (92701, 92702)
(Closed) Unresolved Item 50-261/90-21-01, Welds not Identified on Inservice Inspection Isometric Sketches."
This item reported that isometric sketches used in the ISI program to identify the configuration of piping, weld joint locating and weld population in the ISI program were
found to be in error in that welds were missing from the sketches.
The unresolved item was open until the inspector could discuss this discrepancy with the Office of Nuclear Reactor Regulations (NRR) in order to determine the affect this finding would have on the ISI program currently approved by NRR and also on the ISI program that has been prepared by the licensec
'for the 3rd inspection interval which will start next refueling outage.
.
- '
Subsequent discussions with NRR concluded that this discrepancy could be j
handled adequately with 10 CFR 50, Apperdix b, Criterion V, violetien.
l
--u v
.
.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _.......... _.. _..
.
.
,
-. -
e
,
Therefore, -this unresolved item is considered closed and the issue upgraded to a violetion (50-201/90-24-01, " Welds not Identified on Inservice Inspection Isometric Sketches").
5.
Exit Interview The irispection scope and results were summarized on November 2,1990, with those persons indicated in paragraph 1.
The inspector described the areas inspected and discussed in detail the inspection results listed below.
Proprietary information is not contained in this report.
Dissenting connents were not received f rom the licensee.
(0 pen) Severity Level 4, Violation 50-261/90-24-01, " Welds not Identified on Inservice Inspection Isometric Sketches" Paregraph 4,
,