IR 05000369/1993009
| ML20046B735 | |
| Person / Time | |
|---|---|
| Site: | McGuire, Mcguire  |
| Issue date: | 07/08/1993 |
| From: | Coley J, Cooper T, Lesser M, Van Doorn P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20046B726 | List: |
| References | |
| 50-369-93-09, 50-369-93-9, 50-370-93-09, 50-370-93-9, NUDOCS 9308060128 | |
| Download: ML20046B735 (16) | |
Text
_
__ _ _ _
UNITED STATES
.
f#Sa areg%
NUCLEAR REGULATORY COMMISSION
[
~ *$
REGION 14
g 101 MARIETTA STREET, N.W., SUITE 2900
$
- j ATLANTA, GEORGIA 30323-0199
\\...+/
Report Nos. 50-369/93-09 and 50-370/93-09
,
Licensee:
Duke Power Company 422 South Church Street Charlotte, NC 28242-1007 Facility Name: McGuire Nuclear Station 1 and 2 Docket Hos. 50-369 and 50-370 License Nos. NPF-9 and NPF-17 Inspection Conducted: May 16, 1993 - June 12, 1993 Ih 7 8 13
'
Inspector:
f P. K. Van Doorn, Senior Resident inspector Date Signed Inspector:
N
'7.F.13
'
p T. A. Cooper, Resident Inspector Date Signed Inspector: \\.d i 7-P' N A
J
. C 1 y, ReEq r Inspector Date Signed
%
Approved by.
_
3 H.
S'.
Lesser, Section Chief Date Signed Division of Reactor Projects
,
SUMMARY Scope:
This rautine, resident inspection was conducted in the areas of plant operations, surveillance testing, maintenance observations, Licensee Event Report followup, followup on previous inspection findings, operating experience program and pressurizer ultrasonic
,
test indications.
Results:
In the areas inspected two non-cited violations were identified.
The first involved an inadequate test procedure (paragraph 3.b.)..
The second involved a failure to follow a procedure prescribing a response time for operating experience issues (paragraph 7.).
A general operating experience program weakness in timeliness was also identified (paragraph 7.).
9308060128 930708 PDR ADDCK 05000369 G
>
.
.
-
!
.
.
REPORT DETAILS 1.
Persons Contacted P
Licensee EmDloYees D. B a ter, Support Operations Manager A. Beaver, Operations Manager
!
J. Boyle, Work Control Superintendent i
D. Bumgardner, Unit 1 Operations Manager
,
B. Caldwell, Training Manager W. Cros, Compliance Security Specialist
'
T. Curtis, System Engineering Manager J. Foster, Station Health Physicist l
F. Hayes, Human Resources Manager G. Gilbert, Safety Assurance Manager
,
P. Guill, Compliance Engineer
!
B. Hamilton, Superintendent of Operations
'
B. Hasty, Emergency Planner
- P. Herran, Engineering Manager
!
'
L. Kunka, Cc,mpliance Engineer E. Geddie, Station Manager
- T. McMeekin, Site Vice President R. Michael, Station Chemist
- T. Pederson, Safety Review Supervisor N. Pope, Instrument & Electrical Superintendent
- R. Sharpe, Regulatory Compliance Manager
.
- B. Travis, Component Engineering Manager
!
R. White, Mechanical Maintenance Superintendent t
Other licensee employees contacted included craftsmen, technicians, operators, mechanics, security force members, and office personnel.
NRC Resident Inspectors
'
- P. Van Doorn, SRI
- T. Cooper, RI
,
- Attended exit interview
2.
Plant Operations (71707)
a.
Observations i
The inspection staff reviewed plant operations during the report l
period to verify conformance with applicable regulatory.
'
,equirements.
Control room logs, shift supervisors' logs, shift l
turnover records and equipment removal and restoration records were routinely reviewed.
Interviews were conducted with plant operations, maintenance, chemistry, hcalth physics, and performance personnel.
i
)
y
.
h
r Activities within the contal room were monitored during shifts and at shift changes. Actions and/or activities observed were conducted as prescribed in applicable station administrative directives. The number of licensed personnel on each shift met or exceeded the minimum required by Technical Specifications (TS).
!
The inspectors also reviewed Problem Investigation Reports (PIRs)
'
to determine whether the licensee was appropriately documenting problems and implementing corrective actions.
i Plant tours taken during the reporting period included, but were not limited to, the turbine buildings, the auxiliary building, electrical equipment rooms, cable spreading rooms, and the station yard zone inside the protected area.
>
During the plant tours ongoing activities, housekeeping, fire protection, security, equipment status and radiation control r
practices were observed.
b.
Unit 1 Operations I
The unit began the inspection period in Mode 6 with the core loaded. Mode 5 was entered on May 21, 1993, and Mode 4 on June 7, i
1993.
Mode 2 was entered and plant start-up began on June 12, 1993.
During the pulling of the control rod banks, a rod control general
warning light and a rod bottom light for shutdown control rod L-13 were received.
~he operators determined that a loss of indication
,
'
of the control rod (L-13) had occurred and they manually tripped the reactor. All systems responded as required following the trip.
The licensee initially thought that a bad cable between the stack coil and the bulkhead caused the rod indication problems. They replaced the cable but that did not correct the problems. The licensee then determined that the cable from the bulkhead to the data cabinet was bad. They replaced it, thereby correcting the rod indication problem. Unit start-up resumed on June 12, 1993.
c.
Unit 2 Operations j
The unit began the inspection period at 100 percent rated power
.
and operated at that level until May 30, 1993, when power was reduced to 65 percent for core extension to scheduled shutdown for
]
refueling on July 1, 1993.
On Jur.e 8,1993, due to system load demand, the unit began to increase in power. Full power was reached on June 9,1993. The inspection period ended with the unit at 100 percent rated power.
'
No violations or deviations were identified.
,
d
'
I
,
3.
Surveillance Testing (61726)
i a.
Observed Surveillance Tests Selected surveillance tests were reviewed and/or witnessed by the resident inspectors to assess the adequacy of procedures and
performance as well as conformance with the applicable TS.
l l
Selected tests were witnessed to verify that (1) approved procedures were available and in use, (2) test equipment in use was calibrated, (3) test prerequisites were met, (4) system restoration was completed, and (5) acceptance criteria were met.
.
The selected tests listed below were reviewed or witnessed in detail:
'
PROCEDURE E0VIPMENT/ TEST
,
PT/2/A/4208/10A Containment Spray 2A Heat Exchanger l
Balance Test PT/2/A/4208/01A Containment Spray Pump 2A Performance Test
,
PT/1/A/4200/09A Engineered Safety Features Actuation
,
Periodic Test PT/2/A/4252/03B Auxiliary Feedwater Train B Valve Stroke Timing - Quarterly Turbine Driven Pump Flowpath PT/2/A/4252/01 Auxiliary Feedwater Pump #2
'
Performance Test b.
Engineered Safety Features Testing
!
On May 25, 1993, and May 26, 1993, the inspectors witnessed the performance of licensee procedure PT/1/A/4200/09A, Engineered Safety fertures Actuation Periodic Test. Collowing the completicr.
'
of the test on May 26, 1993, the SR0 on s51ft ouestioned whether or not EMF-43 B had been restored to openat4.on a4 thin the TS required time of I hour.
Investigation :2tetunnad that twice during the testing of the IB ESF on May c6,1953. rid twice during the testing of the 1A ESF on May 25, 1990, FMT O li and EMF-43 B had been inoperable in excess of the I hour
~cr*
ty TS 3.3-6 and the applicable action statement had not t,. t r. u t.
EMFs 43 A and 43 8 were rendered inoperable by the loss-of-offsite power portion of the ESF test. The loss of power trips the EMF sample blower, which does not restart upon restoration of power.
This occurs twice during each train of testing. Restoration steps do not reference the 1-hour time-limit per the TS requirement; in
,
.
2 fact, during normal restoration it is commonly one and one-half to two hours before these sample blowers are restored.
If the 1-hour limit is exceeded, the licensee is required to isolate the outside
<
air supply dampers per TS.
l The procedure has never included a step to restore these sample blowers, so it is likely that the TS action statement has not been
met in the past, during this test.
The questioning attitude of the SRO on shift revealed this
,
procedure weakness. Other procedures were reviewed, notably the
'
Emergency Procedure (EP) and the Abnormal Procedure (AP) dealing with the loss of power scenario. The EP was found to be acceptable, but similar weaknesses were found in the APs.
'
Revisions were made to correct the weaknesses in all of the i
procedures.
Because the EMFs were inoperable for short periods of time and no i
transients occurred during these periods, coupled with the aggressive identification of the problem by the Shift Supervisor
'
and the prompt corrective actions taken by the licensee, the inadequate test procedure is a non-cited violation (NCV) 369,370/
93-09-01: Inadequate Test Procedure Leading to Failure to Meet TS Action Statement. This licensee-identified violation is not being
.
cited because the criteria specified in Section VII.B of the
+
Enforcement Policy were satisfied.
One non-cited violation was identified for an inadequate test procedure
!
resulting in the failure to meet a Technical Specification Action statement.
,
c.
Post-Outage Leak Rate Testing
'
The inspectors monitored problems with post-outage leak testing.
During the McGuire Unit I refueling outage, a type A Integrated Leak Rate Test (ILRT) was conducted as required by TS 4.6.1.2.a.
On June 3,1993, after successful completion of this ILRT, testing was performed on all dual-ply bellows assemblies as specified by TS 4.6.1.2.h.
The bellows assembly on penetration M-441 (Steam Generator ID Main Steam Line) experienced significant leakage.
,
Testing was conducted by pressurizing from inside containment.
Initially, a half pressure test of 7.5 psig was used and the leakage rate was extrapolated to 15 psig with a resulting leak rate of approximately 11,000 SCCH. A full pressure test was performed and leakage exceeded 14,000 SCCM.
TS 3.6.1.2.c requires a combined bypass leakage rate of less than 0.07 times the maximum allowable leakage rate (La) for all penetrations identified as secondary containment bypass leakage paths when containment is pressurized to peak accident pressure (Pa). This limit is approximately 9400 SCCM.
,
-
_-
,. - - _
-
_.
.
.
.
Further testing determined that the bulk of this leakage was returning to the ant:ulus through the open test connections between the plies. This returned flow is not considered bypass flow.
Final calculations determined that the number three bellows on
.
this penetration had a bypass leakage of approximately 4300 SCCM.
Combined with the bypass leakage from other penetrations, a total bypass leakage of approximately 6000 SCCM was measured.
i Since the leakage through this penetration represented a significant step change in bellows performance, the licensee monitored the bellows leakage on-line. The licensee determined that the most reasonable monitoring was the between-ply 4 psig pressurization test for both cold and hot conditions to establish a baseline leak assessment. Monthly testing would then be conducted to ensure that leakage was not significantly increasing over time.
'
Leakage during hot baseline tests increased from that of the cold baseline tests, thereby decreasing the margin to the limit from approximately 3000 SCCM to approximately 60 SCCH.
-
The licensee performed a limited inspection of the penetration to identify the cause of the degradation.
For the outer two bellows sections, visual inspection of the inner ply is impossible.
,
Visual inspection of the outer ply is difficult because of space limitations. Using the inspection techniques available (compact video camera equipment and a boroscope), potentially damaged areas were inspected. The licensee noted discoloration on the outermost-convolution of the number three bellows. They speculated that the discoloration was caused by a welding arc strike. They visually inspected the other three main steam line penetration assemblies and found no significant damage.
Repair options are being assessed by the licensee, but due to the scope of work involved repairs will not be made until the next refueling outage. A
,
request for a Technical Specification revision to increase the margin for allowable bypass leakage during the upcoming operating cycle is being prepared.
No violations or deviations were identified.
.
4.
Maintenance Observations (62703)
i a.
Observation
'
Resident Inspectors reviewed and/o: witnessed routine maintenance activities to assess procedural and performance adequacy and conformance with the applicable TS.
The selected activities witnessed were examined to verify that, where applicable, approved procedures were available and in use, prerequisites were met, equipment restoration was completed, and maintenance results were adequate.
_-
-.
.-
~
.~
.
-
._. -
.
.
.
.
,
t
.
.
'
>
,
s
7
The selected maintenance activities listed below were reviewed or
witnessed in detail:
"
WORK ORDER ACTIVITY 93032003 Perform PM/PT on Channel
.;
Source Check on' All Process-i and Area Radiation Monitor i
Channels
!
92091871 PM/PT on IENBLP9440, Power
Range N44 Calibration
,
93027856 Move and Install IB Reactor
'
.
Coolant Pump Motor l
93041519 Shorten the Tubing from the
!
Upper Alarm Pot to the Oil Pot i
on IB NC Pump Motor j
i No violations or deviations were identified.
j-b.
Steam Generator Tube Inspection The inspectors reviewed the results of the licensee Steam
'
Generator (S/G) tube inspection during the refueling outage. ~The licensee performed 100 percent bobbin coil eddy current inspection
,
of the cold leg tubes,100 percent rotating pancake coil (MRPC)
.l eddy current of.the hot leg tubesheet,100 percent MRPC of the row l
2 and 3 U-bends,100 percent MRPC of the Inconel 600 rolled plugs,
!
100 pere:ent MRPC of the kinetic sleeves, and MRPC in areas of j
special interest.
A total of 41 plugs were repaired (35 in the hot leg and six in
the cold leg).
Eddy current testing resulted in the plugging of l
an additional 214 tubes (35 in S/G "A," 61 in S/G "B," 44 in S/G i
"C," and 74 in S/G "D").
No free span axial cracking was noted.
The licensee submitted the inspection results per TS 4.4.5.5 a)
,
and c) to the USNRC in a letter dated May 25,1993. A conference
,
call consisting of licensee personnel, the resident inspectors,
regional personnel, and NRR personnel was conducted on May 27, l
1993. The results were discussed and no concerns were identified.
,
No violations or deviations were identified.
c.
Nuclear Service Water Pump Erosion
'
On May 20, 1993, the licensee identified evidence of erosion on
the 2A Nuclear Service Water pump wear ring. Further inspections q
revealed no signs of erosion on the pump impeller.
Previously,
.
- - -.,. _ - -
_
-
.,
_
,
~
.
l
because of erosion on the pump impeller, the impeller had been
,
changed from a bronze impeller to a stainless steel one. However, J
the wear rings used in the pump were still bronze.
The licensee identified wear in excess of what was normally expected for this pump, but no degradation of pump performance was noted. The licensee reinstalled the wear ring and satisfied the TS LC0 that was in effect. The pump is being monitored for
'
degradation and the wear ring will be replaced during the 2E0C8 refueling outage, scheduled to begin July 1,1993.
The inspector reviewed the maintenance activities and observed work on the pump assembly.
No violations or deviations were identified.
5.
Licensee Event Report (LER) Followup (90712,92700)
'
The inspectors reviewed the LERs listed in this section to determine if the information provided met NRC requirements.
In reviewing the LERs inspectors considered the adequacy of description, verification of
,
compliance with Technical Specifications and regulatory requirements, corrective action taken, existence of potential generic problems, reporting requirements satisfied, and the relative safety-significance
,
of each event. Additional inplant reviews and discussion with plant personnel were conducted for the following reports:
'
369/92-12: (Closed)
Refueling Water Storage Tank Level Transmitters
'
Were Technically Inoperable 370,93-01: (Closed) A Unit 2 Manual Reactor Trip was Initiated as a Result of Equipment Failure
,
,
370/93-02: (Closed) A Unit 2 Manual Reactor Trip was Initiated Due to an Equipment Failure 370/92-09: (Closed) A Unit 2 Reactor Trip Occurred as a Result of an Equipment Failure No violations or deviations were identified.
6.
Followup on Previous Inspection Findings (92701,92702)
The following previously identified item was reviewed to verify that (1)
,
the licensee's responses (where applicable) and actions were in compliance with regulatory requirements, and (2) corrective actions have been implemented.
Selective verification included a review of the records, observations, and discussions with licensee personnel.
l (Closed) Unresolved Item 369,370/93-03-03:
Evaluation of OEP Timeliness. A special inspection was conducted to evaluate the licensee's Operating Experience Program. This item is closed and a Non-i
.
i l
,
Cited Violation is being issued (see paragraph 7).
7.
Operating Experience Program Review (40500)
The licensee's Operating Experience Program (0EP) is currently described in Nuclear System Directive (NSD) 204. The OEP is managed by the licensee's corporate Operational Event Analysis section (0EA). The OEA i
receives, screens and distributes various documents to the licensee's
three nuclear stations for review and development of appropriate corrective actions. Documents included in the OEP are events from the
,
licensee's corrective action program, licensee event reports (LERs),
significant event reports (SERs), significant operating experience reports, operating plant experience reports (OEs), operations and maintenance reminders, vendor information letters (VILs), NRC
'
information notices (ins), and INP0 information. The OEA defines
'
problems as " Problem Avoidance" or " Problem Awareness" issues.
Avoidance problems require a response within 90 days if they are assigned a " normal" evaluation period and 30 days if they are assigned an "immediate" period.
Often the document is sent to a corporate technical group for coordination of the response; it is sometimes sent to individual sites as well. The person assigned to respond must do so within the 30 or 90 day period or request an extension. OEA is required to review outstanding OEP items periodically and inform management of the overdue items. The designated response group is responsible for notifying OEA if another group would be a more appropriate response group. The
,
response group is also required to indicate whether or not additional t
training should be conducted.
'
The inspectors reviewed the OEP, discussed the OEP with licensee management and selectively reviewed documentation for Problem Investigation Reports (PIRs), ins, VIls (including 10 CFR Part 21 notices), LERs and OEs to determine if (1) program requirements were
!
being met, and (2) the licensee was developing appropriate corrective
,
actions in a timely manner. The inspectors reviewed a total of 76 items. This inspection also was conducted at the Catawba Nuclear Station; for findings see NRC Report No. 413,414/93-17.
The corrective actions developed by the licensee appeared to be appropriate for the circumstances. However, a number of examples of
untimeliness were ncted as follows:
1.
The majority of the corrective action program issues reviewed by
!
the inspectors were entered under an earlier program involving PIRs as opposed to the recent program involving Problem Investigation Process forms (PIPS). The OEP process did not even begin until the PIR was fully processed and substanially evaluated, which often took a year or longer. The licensee indicated that the new OEP program utilizing the PIP process will inherently be niore timely since the PIP program requires more timely response.
For example, the "more significant event"
.
.
category requires a 30 day turnaround for development of root cause and corrective actions and the OEP process will begin at this stage. On-line computer access to PIPS will be available to
,
OEA as well.
'
2.
Although the OEP does not specify how long the " receipt and screening" time period should be, the OEP staff manager indicated that ten days were expected. The typical time to forward issues was approximately three weeks. Three outlyers were noted: PIR 1-M92-139 (also LER 369/92-10), which took seven months to distribute; PIR 0-M92-515, which took approximately 10 weeks to distribute; and VIL 92-12 (also a 10 CFR Part 21 item), which took two months to distribute.
3.
Occasionally the OEP group would send a document to an inappropriate person to respond. Typically, when this occurred the majority of the 90 day response time had passed before OEA was informed that another party should respond. The 90 day period over would then begin again.
4.
The inspectors identified examples of untimeliness involving the use of extensions. The inspectors noted frequent and liberal use
of extensions by almost all the technical groups, and for some of these items it was questionable whether an extension was requested due to the complexity of the issue or due to a lack of prompt attention to the item.
Specific examples were PIR 2-M92-]I6, PIR 2-M92-130, IN 92-43, IN 92-59, IN 92-52, Westinghouse VIL 92-29, OE 548 and OE 5190.
5.
The inspectors identified multiple instances whereby a response time had been exceeded although an extension had not been granted.
These included PIRs 0-M92-118, 2-M92-125, 0-M92-136 and 1-M92-459; ins 92-31, 92-33 and 92-43; Westinghouse VILs 92-30, 92-31, 92-33 and 92-34; VIL 92-12; and OEs 5348, 5253 and 5251. A previous example was identified and documented as Unresolved Item 369,370/93-03-03.
The licensee indicated that timeliness had been generally recognized as a problem with the OEP. A quality improvement project team had been recently initiated to develop improvements. The licensee has also recently begun to assign each OEP item as a PIP that will require timely response. Also, the licensee indicated that the OEA group was recently reorganized with more specific assignments given to individuals.
Generally the examples above indicated a weakness in timeliness of the OEP process. The items identified in example 5 are considered a violation of the OEP vocedures and 10 CFR 50, Appendix B, Criterion V, which requires activities affecting quality to be conducted in accordance with prescribed procedures. This NRC-identified violation is not being cited because the criteria specified in Section VII.B of the NRC Enforcement Policy were satisfied. Specifically, the violation was not willful; nor was it similar to a prior violation for which
.
l
,
.
.
c
corrective actions have not been sufficient to prevent recurrence.
Furthermore, corrective action has been initiated to correct weaknesses
<
'
in the OEP program. This issue is identified as Non-Cited Violation 369,370/93-09-02: Failure to Follow OEP Procedure Requirements.
One non-cited violation was identified.
8.
Review of Recorded Data and Evaluations for the Unit 1 Pressurizer Lower Head to Shell Weld Ultrasonic Indications (73755)
,
Background As reported in Region II Inspection Report No. 50-369,370/93-04, recordable ultrasonic indications were detected in Weld No. IPZR-1 by B&W Nuclear Technologies, a vendor used by the licensee during examinations conducted in 1986. This weld is the lower head to shell i
weld on the Unit 1 pressurizer. At that time several of the indications exceeded ASME Section XI, IWB-3500 acceptance criteria.
Fracture mechanics analysis was performed to show that the vessel was qualified
'
for continued operation. Subsequently, the areas containing the unacceptable indications were scheduled for re-examination during three consecutive inspection periods as required by ASME Section XI, Paragraph IWB-2420 (b.).
In 1990 the indication areas were re-examined by B&W Nuclear Technologies, who determined that there was no evidence of change in flaw dimensions that would indicate growth.
In 1993 the areas were re-examined by Duke Power personnel. However, the results of their preliminary manual ultrasonic analysis indicated an
.
increase in the number of flaws, as well as an increase in flaw length
'
and through-wall dimensions from what had been previously reported. The increase in the number of indications was thought to be partly due to Duke's lower recording threshold, 20% Distance Amplitude Curve (DAC) vs 50% DAC. However, significant differences were noted between the flaw indication amplitudes recorded by B&W and those recorded by the licensee. Most indication cmplitudes recorded during the 1993 examinations were below 100% DAC, whale the majority of indications recorded in 1986 and 1990 were greater than 100% DAC and many were greater than 200% DAC.
In addition, many of the indications recorded during the 1993 examinations were thought to have initiated from the inside surface of the pressurizer, which might have indicated degradation caused from thermal fatigue cracking.
To resolve the above discrepancies, Duke Power contacted the Electric Power Research Institute (EPRI) Nondestructive Examination (NDE) Center and B&W Nuclear Technologies to provide support and to help characterize the previously reported indications using EPRI's automated ultrasonic ZIPSCAN instrument and B&W's automated ultrasonic data acquisition and analysis system ACCUSONE i
.
.
'
.
S
,
On May 19, 1993, the inspector arrived at the McGuire facility to review l
the inspection results and the flaw evaluations performed by Duke, B&W,
,
and EPRI. The inspector held discussions with the cognizant Level III examiners and reviewed the recorded ultrasonic system data. This review revealed the following:
,
a.
Beam spread comparisons of the ultrasonic transducers used in the 1986,1990, and 1993 manual examinations revealed that the
'
transducer beam profiles for the 1986 and 1990 examinations were very similar. Since the same edition of the Code also was used c
for these inspections the recorded results are nearly identical.
!
However, the transducer beam profiles for the 1993 manual
,
examination revealed significantly larger dimensions than those of the transducers used during the 1986 and 1990 examinations. The result of using the larger beam dimension in 1993 would account for the apparent increase in reported flaw dimension, even though the physical target size remained unchanged. This gave the appearance of flaw growth and was misleading. This coupled with the lower recording criteria explained why the preliminary analysis of the 1993 manual examination data concluded that the flaw dimension were increasing.
b.
Another notable difference in the 1993 manual ultrasonic data is the amplitude of the ir.dications. During the41986 and 1990 i
examinations if the indications were located in the inner 10%
thickness of the wall of the pressurizer the indication was recorded relative to the amplitude of the calibration notch and designated as "SPF" on the data sheets. However, on this specific calibration block the notch response provides a much lower response than the DAC curve from the side drilled holes because of the cladding applied to the block's inside surface.
B&W procedures required the examiner to place the lower amplitude notch signal at 80% full screen height in lieu of the DAC when
'
evaluating indications located in this area of the weld.
Therefore, an adjustment is required, but had not been performed, to directly compare the 1993 manual data amplitude values to the 1986/1990 amplitude values since the 1993 data were recorded to the DAC curve only from the side drilled holes. When these adjustments were made to the 1986/1990 data the results were in general agreement with the 1993 data.
c.
The third major difference in the 1993 manual ultrasonic data was
,
the fact that many of the flaw indications appeared to be
'
connected to the inner diameter (ID) surface. To analyze this B&W re-examined all accessible areas of this weld using their automated ultrasonic system. A recording threshold of 5% full screen height was used for data collection (the ASME Code only requires recording indications of 50% DAC and greater). All indications of 50% DAC and greater were documented noting the maximum amplitude, length and end points at 50% DAC, the through-wall dimension, depth of the indication and the weld wall thickness at the indication location.
,
--
.
,
l l
,
.
'
The ASME Code reqaires that the through-wall dimension be measured at the 50% DAC level for indications of 50% DAC through 100% DAC and at the half-maximum amplitude (HMA) level for indications
,
greater than 100% DAC. The method of determining through-wall i
measurements based on amplitude levels has been found to be a less accurate measurement of the through-wall dimension of an indication.
In particular, when the flaw size is smaller than the i
sound beam width, the measured dimension usually represents the beam width of the transducer and not the dimension of the flaw.
This method provides a conservative estimate of flaw size, which
'
can be much larger than the actual dimension of the flaw.
Therefore, if the indication can be characterized as volumetric in
nature, it is desirable to use as small a sound beam as possible to measure the through-wall dimension.
It is even more desirable to use a measurement technique that is independent of amplitude to
.
measure the through-wall dimension.
'
B&W's analysis of the data indicated that the weld contains flaws that are intermittent and occur randomly along the weld length.
However, only eight of the 148 flaws detected were of recordable
amplitude according to the recording criteria of the ASME Code when using B&W's automated system. The flaws are at varying
'
levels and the majority of them are located in the lower 1/3 of the weld, usually at the weld to base metal interface.
Based on
'
drawings of the weld-joint configuration, most of the flaws are at the location of the intersection of the root of the weld. The i
drawings indicate that this weld was made from the outer diameter
(0D) of the pressurizer with a closed root. Once the OD portion of the weld was complete, the root was excavated from the ID of
,
the vessel.
The weld was completed from the ID of the vessel and i
then clad.
Inclusions and lack of fusion are commonly found at i
the root of these type of welds.
When a high frequency transducer (5 MHZ) was used by B&W the true
sizes of the flaws could more clearly be seen because of the
smaller beam dimensions available with the high frequency transducer.
In addition, clear separation between the indications
and the ID of the vessel was illustrated in the data. The responses from the recorded flaws are typical of those from inclusions.
The plotted beam profiles also did not indicate ID connection in that, they were nearly symmetrical. Geometrically the flaws are generally located at the weld to base metal interface. Where flaw lengths are relatively long the flaw is
usually at the same elevation and consistent with what would be considered one weld bead layer in depth.
I Personnel from the Electric Power Research Institute (EPRI)
Nondestructive Examination (NDE) Center also applied a forward scatter time-of-flight technique to measure the time to the top and bottom of selected indications to provide a direct correlation to the flaw through-wall dimension.
This technique is independent of amplitude and has been shown to be more accurate
._
,
i
-
for flaw sizing than the techniques required by the Code. The EPRI NDE data was supportive of the data taken by B&W. In general, the through-wall dimensions measured with the time-of-flight method range from 2mm (0.079") to 5mm (0.197") and none of the flaws were found by EPRI to be connected to the ID.
In summary the examinations conducted during the 1993 outage are the most comprehensive performed to date.
Based on the automated data results, the indications are characterized as typical manufacturing fl aws. There is no evidence to suggest that the flaws contained in the head-to-shell weld are service-induced.
In addition, the licensee found tnat previous data, which was believed to be preservice data for the Unit 1 pressurizer and which indicated that there were no previous recordable fabrication flaws in this weld, were actually obtained from Unit 2.
Pre-service data for the Unit 1 pressurizer did reveal indications; however, since data were recorded below 100% DAC (the
,
recording level at that time) each indication was not separately
-
recorded or sized. Review of fabrication radiographs also revealed that this weld had experienced severe cracking during the fabrication process.
Therefore, to ensure that there is no question of the integrity of this weld, the licensee stated that the weld would be inspected during the next 40-month neriod presently scheduled with an i
automated system.
This will be done so that the data can be more easily
compared to the enhanced data taken during this outage for determining flaw growth.
No violations or deviations were identified.
9.
Follow-Up on Reactor Coolant Leakage From Chemical and Volume Control System (71707)
On May 31, 1993, the licensee identified a 1 gallon per minute reactor coolant leak. The leak was identified as coming from a socket weld on an elbow upstream of valve 2NV-458, the 75 gpm letdown orifice isolation valve.
In response, the normal letdown flowpath was isolated ar.d the alternate letdown flowpath (excess letdown) was established.
Following isolation of the leak, the weld was ground out and repaired. A liquid penetrant examination of the final pass of the weld repair was performed. The licensee attributes the failure mechanism to vibration-induced fatigue failure. Similar cracks in the system occurred in September, 1992.
ASME Section XI, 1980 Edition, Subsection IWA-5214 specifies that a system hydrostatic test must be performed for the repaired component
'
prior to resumption of service. Subsequent to the repair of the weld a hydrostatic test was attempted. Since the weld was a class 2 weld, IWC-5222 requires that the test pressure be at least 1.25 times the design pressure for the system. The design pressure specified for this portion of the system is 2485 psig; therefore, pressure for the hydrostatic test wauld be 3100 psig.
!
.
-
~
l
When the system hydrostatic test was performed, a pressure of 2150 psig was the maximum nressure that could be achieved. The licensee suspected I
'
that the requireu test pressure could not be reached because of packing leakoff and/or leakage past one or more of the valves used to form the boundary for conducting the test. Test pressure of 2150 psig is well above the normal operating pressure that the weld would be subjected to.
.
The repaired weld is located downstream of the letdown orifice. When the reactor coolant passes through this letdown orifice a large pressure reduction occurs, resulting in an operating pressure of approximately 335 psig. The test pressure that was achieved is significantly greater than 1.25 times the expected normal operating pressure.
On June 4, 1993, the licensee requested relief from the requirement to test at 1.25 times the rated pressure.
The following alternate tests were proposed:
1)
a VT-2 exam at a reduced test pressure of 2150 psig, 2)
a liquid penetrate test of the final pass of the weld repair, 3)
a VT-2 exam at normal operating temperature and pressure, and 4)
observation for possible leakage by monitoring inputs to the containment sump once every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
The licensee has also committed to modifying and replacing the orifice
arrangement to substantially reduce the vibration in the system. The inspectors met with the licensee and reviewed the relief request.
Relief was granted on June 9, 1993. Normal letdown was established and the alternate testing will continue until the upcoming refueling outage, scheduled for July 1,1993.
No violations or deviations were identified.
10.
Follow-Up of NRC Bulletin 93-02: Debris Plugging of ECCS Strainers (92701)
The inspectors reviewed the licensee response to NRC Bulletin Number 93-02, Debris Plugging of Emergency Core Cooling Suction Strainers.
The licensee reviewed all containment ventilation systems to determine the amount and location of cny fibrous filter media either permanently or temporarily stored inside containment.
Three containment ventilation systems were identified which utilize a fibrous filter medium: the upper containment ventilation system (VU),
the incore instrumentation room ventilation system (VT), and the containment purge ventilation system (VP).
The filter on the VU system is above the flood level for a LOCA and there are no LOCA line breaks post ilated for the filter location. There i
.
i i
is no plausible filter migration route from this area to the lower sump area.
The Incore Instrumentation Room Ventilation System units are also above the LOCA flood level, but there are LOCA line breaks postulated for that area. The licensee has determined that there is no plausible filter migration route from this area to the lower sump area. The VP filters had been previously evaluated for sump plugging due to flooding. This evaluation revealed that there was no significant impact on the ECCS
'
sumps. Various line breaks associated with a LOCA could occur, but each l
potential break location is constructed of heavy duty industrial construction with a metal support structure.
In accordance with maintenance procedures, each filter is wired in place.
In the event of
jet impingement some of the filter material may be blown from the filter housing.
In the event that the entire 32 square feet of fibrous filter material from one filter became lodged in the containment sump, the ECCS system would still function.
It has been demonstrated during pre-operational testing that the ECCS system will still function with a 50% blockage (67.5 square feet of filter material). The plant design does not use an in-line ECCS suction strainer; the entire 135 square feet of the sump has a screen. The licensee does not plan to remove these filters or take any compensatory measures.
There is no fibrous filter media stored temporarily inside of containment. The ECCS sump screens are inspected each outage for any damage or debris.
Per licensee requirements, any material taken into containment during an outage is removed prior to unit startup. The inspectors reviewed licensee procedure PT/1/A/4600,,03F, Containment Cleanliness Inspection, which is performed prior to Mode 4 to verify that there was no material left inside containment that could block the ECCS sump.
No violations or deviations were identified.
'
11.
Exit Interview (30703)
The inspection scope and findings identified below were summarized on June 15, 1993, and discussed with those persons indicated in paragraph I
above. The following items were discussed in detail:
Non-Cited Violation 369,370/93-09-01:
Inadequate Test Procedure Leading to Failure to Meet TS Action Statement (paragraph 3.b.)
Non-Cited Violation 369,370/93-09-02:
Failure to Follow OEP Procedure Requirements (paragraph 7.)
The licensee representatives present offered no dissenting comments, nor did they identify as proprietary any of the information reviewed by the inspectors during the course of their inspection.
--- -