ML20057E301
| ML20057E301 | |
| Person / Time | |
|---|---|
| Site: | McGuire, Mcguire  |
| Issue date: | 09/27/1993 |
| From: | Blake J, Economos N NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20057E296 | List: |
| References | |
| 50-369-93-19, 50-370-93-19, NUDOCS 9310080300 | |
| Download: ML20057E301 (7) | |
See also: IR 05000369/1993019
Text
- n neg
UTUTED STATES
S
f*
4,
NUCLEAR REGULATORY COMMISSION
- , *
4
' REGloN 11
[
S
101 MARIETTA STREET, N.W., SUITE 2900
ATLANTA, GEoRGtA 30323 0199
%....+/
Report Nos.: 50-369/93-19'and 50-370/93-19
Licensee: Duke Power Company
422 South Church Street
Charlotte, NC 28242
Docket Nos.: 50-369 and 50-370
License Nos.: NPF-9'and NPF-17-
Facility Name: -McGuire 1 and 2
.
Inspection Conducted: August 30 - 31, and September 1-3, 1993
N mmN
[M
3
i
InspectoI:
.
.
y
N. Economo~s
'W
pateSigned
i
Approved by:
'[
) 27 f3
-
J.
ake, Chief
D1te Signed
!
.
Ma r als and Processes Section
g eering Branch-
iv sion of Reactor Safety
l
SUMMARY
l
'
,
Scope:
.
This announced inspection was conducted in response to Unit I shutdown on
!
August 22, 1993, because of an indicated leak in "A" steam generator (S/G).
l
The leak rate prior to shutdown was approximately 185 gallons per day.
Results:
Following plant shutdown, the licensee determined that the primary source of
the leak was a sleeved tube, (R39-C72) in S/G "A".
The licensee's
investigation determined that the aforementioned tube exhibited a crack just
above the kinetic weld near the top of the sleeve. This crack provided the
leak path and was the primary source of the leak. Additional minor leaks
associated with sleeved tubes were identified in S/Gs "A", "B" and "D".
One
nonsleeved tube (R41-C43) was identified as a leaker in S/G "B".
The Eddy
Current (ET) indication in this tube was characterized as a single
circumferential crack that was located about 8.7-inches above the 8th tube
support plate (TSP) in the U-bend region. This indication had been miscalled
by the analysts, in the ET examination performed during the April 1993
refueling outage, as an indication with no dectable degradation.
The licensee is presently re-reviewing and re-examining tubes with suspect
indications.
$j
9310080300 930929
ADOCK 05000369
.t
Q
11
'
- .
.
.
.-- ..
. .
.
.
2
Two sleeved tubes (R39-C72 and R7-C78), are being removed for metallurgical
examination to determine the root cause and possible generic implication of
the failure. Through discussions with licensee technical and management
personnel, and a review of newly written analysis guidelines, the inspector
has determined that the licensee has dedicated adequate resources in this
effort. Procedure and guidelines have sufficient conservatism in terms of
acceptance criteria to provide a reasonable assurance that suspect indications
will be identified and properly dispositioned.
Within the areas inspected, no violations or deviations were identified.
,
1
I
.
.
'
!
l
-
.
~
REPORT DETAILS
l
1.
Persons Contacted
!
!
J. Baumann, Supervisor Eddy Current
K. Davis, Lead Analyst Level III Eddy Current
M. Geddie, Station Manager
- R. Hall, Engineering Manager
- L. Kunka, Regulatory Compliance Engineer
- B. Lowery, Nuclear Service Engineer
j
D. Mayes, Nuclear Service Engineer
!
- T. McMeekin, Vice President McGuire Nuclear Station
R. Sharpe, Regulatory Compliance Manager
M. Thompson, Component Engineer
Other licensee employees contacted during this inspection included
i
technical support, Quality Assurance, and administrative personnel.
i
!
'
NRC Resident Inspectors
- G. Maxwell, Senior Resident Inspector
T. Cooper, Resident Inspector
j
4
- Attended exit interview
2.
Plant Shutdown Due To Steam Generator Tube Leakage, Unit 1 (IP73753)
,
'
On August 22, 1993, Unit I was shutdown in response to an indicated tube
leak in S/G "A".
On August 30, 1993, the inspector arrived on site to
ascertain conditions relative to this leakage, to review previous
examination records of tubes found leaking at this time, to monitor the
l
licensee's inspection plans, and to evaluate activities and the
correctness of actions in the four S/Gs prior to plant startup.
l
l
Discussions with cognizant personnel disclosed that all but one of the
leaks were associated with sleeves in S/Gs '"A", "B" and "D".
The
~
leaking sleeves were associated with the following tubes:
Tube
Sleeve Installation Date/TYDe
Leakaae Rate
S/G "A"
l
R2-C114
4/90 (kinetic / roll)
1 drip /2 min.
I
l
R7-C77
4/90 (kinetic / roll)
1 drip /5 min.
!
R7-C111
4/90 (kinetic / roll)
1 drip /5 min.
R39-C72
9/91 (double kinetic)
Steady Stream
S/G "B"
l
R21-C61
4/90 (kinetic / roll)
1 drip /1.25 min.
R21-C62
4/90 (kinetic / roll)
1 drip /2.25 min.
- R41-C43
not sleeved
I drip /10 sec.
,
-
_
I
.
m
2
'
S/G "D"
R6-C43
4/90 (kinetic / roll)
I drip / min.
- The only leaker found on the cold leg side of the steam
generators that was not sleeved.
Background on Leaky Tubes:
Tube R39-C72 S/G "A":
Through discussions with cognizant licensee personnel and records
review, the inspector ascertained that the subject tube was
sleeved in April of 1991. The overall length of this sleeve is
approximately 17{ inches. Approximately five (5) inches of the
sleeve length is positioned in the tube sheet. The sleeve is
secured to the parent tube by energizing a prepositioned explosive
charge near each end of the sleeve. The explosive force of the
charge expands a narrow band of the sleeve fusing the sleeve to
l
,
the tube. A subsequent thermal treatment relieves the residual
stresses generated by the process. This fused area is tested
qualitatively with ET to determine joint integrity.
Prior to the
1991 outage, sleeves were secured using the hard roll expansion
technique on the lower end of the sleeve inside the tubesheet and
the kinetic weld / roll technique on the upper end of the tube
located in the free span tube region. The double kinetic weld
along with an improved joint cleaning technique was meant to
improve sleeve integrity. A review of the log book disclosed that
the sleeve in the subject tube had been returned to the vendor for
i
repairs prior to installation. The repair involved
'
relocation / realignment of the charge and replacement of a missing
wafer. The licensee stated that physical measurements taken
during the current outage places the location of the charge
slightly above optimum sleeve height but still within tolerance.
Sleeve Examination:
Through discussions with cognizant licensee personnel, the
inspector ascertained that the subject sleeve had been examined
during the April 1993 outage and there were no detectable
indications observed. During the present outage, examination with
a rotating pancake (RPC) probe disclosed a crack like indication
in the parent tube at the upper weld, just above the apex of the
i
expansion. This location coincides with the location where
remote visual examination, performed during this inspection,
showed a small amount of water cascading over the top of the
sleeve. Based on the above mentioned ET test results, it would
appear that a crack was probably present in the aforementioned
tube location, but its magnitude was below the threshold of ET
probe detectibility. From the time the plant returned to power
until the time the leak was detected, by plant monitors, it
appears that the crack grew at an accelerated rate until
propagating through the tube wall.
Following the close of this
l
l
._
._
_
_
-
i
1
3
.
inspection, the licensee pulled the subject sleeve, from the S/G,
,
and forwarded it to B&W in Lynchburg, VA. (BWNS) for a
'
metallurgical investigation to determine the root cause of the
crack and possible generic implications.
In addition to R39-C72,
the licensee identified a total of approximately 31 sleeved tubes,
,
located in all four S/Gs, which were labeled as suspect due to the
l
presence of indications which were not crack like, identified by
Bobbin and confirmed by RPC. Tube, R7-C78 located in S/G "A",
was .
'
also pulled and sent to BWNS for metallurgical investigation, as
it exhibited a ET signal interpreted as a geometry distortion
which was not fully understood.
Tube, R41-C43 S/G "B", Cold Leg:
As stated earlier in this report, by visual inspection, the
,
licensee determined the leak rate of this tube to be approximately
!
one drop per 10 seconds. The source of the leak was established
at a height of approximately 8.7 inches above the 8th tube support
plate, on the cold leg side of the S/G, at the transition U-bend.
Examination of the tube with bobbin coil and RPC probes, during
'
the present outage, confirmed the presence of a single
'
circumferential crack which had propagated through the tube wall.
The bobbin coil signal was analyzed as a complex dent signal with
-
a voltage of approximately 5.7 volts. During the April 1993
outage, this tube had been examined with a similar bobbin coil
probe. The indication was analyzed by the primary and secondary
i
analysts and classified as an indication which had no detectable
degradation (NDD), even though the voltage of the signal was in
the range of 6-7 volts. Cognizant licensee personnel stated that
the high noise level associated with the U-bend region of the tube
bundle was in part responsible for the miscall i.e., NDD vs Dent
i
signal. However, this inspector's position on the high noise
phenomenon in the U-bend region, is that the analysts should have
been cautioned to be extremely sensitive to indications in this
area which could be masked by the high noise level.
The smaller leaks in the other tubes were believed to be related
to one or several steps in the process used to secure the sleeves
to the parent tubes. One of the most likely causes was thought to
be inadequate cleaning of the joining surfaces. Because of
concerns raised over the adequacy of the ET examinations, the
sleeving process, as well as the potential for generic
implications on the other unit and the Catawba Station, the
licensee implemented the following tube inspections / examinations.
(1)
Rewrote specific analysis guidelines, dated 9/1/93, to
review bobbin coil ET data, acquired during the April 1993
outage, for all dent-like indications with 2: 2.0 volts. The
objective was to detect any degradation with dent-like
indications throughout the entire length of the tube above
the Tubesheet.
Initial screening of data for this review
was done by computer. The 2.0 volt threshold was raised
I
.
.
,-
. - - ,
, . - - -
-.
. . .
.- -
.
l
l
4
l
-
l
later to 2.5 volts because it was determined that signals at
l
the lower level, were insignificant and beyond the scope of
the review.
(2)
RPC examine all indications with bobbin coil ET signals a
5.00 volts.
(3)
RPC examine the following sleeves in each of the four S/G/s.
S/G "A"
S/G/"B"
S/G "C"
S/G "D"
294
171
94
263
(4)
Ultrasonic examination of certain special interest sleeved
)
tubes including leakers.
S/G "A"
S/G "B"
S/G "C"
S/G "D"
!
46
0
0
10
'
(5)
RPC examine two rows of tubes around leakers.
t'
Since the close of this -inspection on September 3,1993, the
inspector has been following the progress of this work
through telephonic discussions with the licensee. Through
these discussions, the inspector ascertained that the
i
sleeved segments of tubes R39-C72 and R7-C78 were pulled and
forwarded to BWNS, for failure analysis.
Preliminary
-!
results indicated that tube R39-C72 exhibited a
~
circumferential crack which covered approximately 270
degrees of the tube's periphery. On September 14, 1993
1
during a telephone conference the licensee /BWNS reported
i
that preliminary results from the metallurgical
.
I
investigation indicated that this tube exhibited an
unusually fine, ASTM grain size (a 11), a relatively high
hardness and relativity high mechanical properties
suggesting the tube was susceptible to primary water stress
corrosion cracking.
Also the inspector ascertained that the review of indications
selected through computer data screening had been completed - this
review identified no obvious miscalls.
Following is a tabulation of tubes in the U-Bend region exhibiting
indications and therefore designated for RPC examination.
S/G "A"
S/G "B"
S/G "C"
S/G "D"
,
'
Indications / Tubes
221/147
510/205
133/96
785/363
3.
Mockup Training of Welders (55050)
Towards the end of this inspection welders were undergoing mockup
training in preparation for welding plugs in the Tubesheet where tubes
were removed from service.
In addition to verbal instruction, the
i
. . - , _ .
.
.
-
-
-
.
. - . - -
-
-
-
-..
.
. . . -
-
l
'
l
.
!
.
.'
5
l
candidates demonstrated their ability to fabricate an acceptable weld in
the overhead position inside a mockup of a steam generator bowl with a
tubesheet.
The men were dressed with protective. clothing and
respiratory protection equipment. The test plug, was welded (fillet
'
weld), using the gas tungsten arc welding (GTAW), process. Both the
!
plug and filler metal were made of inconel material. The inspector
observed three welders take the subject test successfully. Welder
performance qualification records and the associated weld data sheet for
the applicable weld procedure were reviewed and found to be in order.
,
Within the areas inspected no violations or deviations were identified.
>
4.
Exit Interview
The inspection scope and results were summarized on September 3, 1993,
i
with those persons indicated in paragraph 1.
The inspector described
j
the areas inspected and discussed in detail the inspection results. The
i
licensee did not identify as proprietary any of the material provided to
,
!
or reviewed by the inspector during this inspection. Dissenting
comments were not received from the licensee.
-
i
,
!
!
i
i
f
.
I
!
'
i
i
i
!
!
,
!
!
!
,
!
!
!
!
i
!
. . .
.
-
. _ .
- .
. .
. _ _ . _ . . , .
.
. . .
. -
. . _ _ . ..,___ __ ..,......._._ .. . ...,,_.