Protecting People and the EnvironmentUNITED STATES NUCLEAR REGULATORY COMMISSION
UNITED STATES
NUCLEAR REGULATORY COMMISSION
OFFICE OF NUCLEAR REACTOR REGULATION
WASHINGTON, D.C. 20555
December 23, 1994
NRC INFORMATION NOTICE 94-88: INSERVICE INSPECTION DEFICIENCIES RESULT IN
SEVERELY DEGRADED STEAM GENERATOR TUBES
Addressees
All holders of operating licenses or construction permits for pressurized-
water reactors.
Purpose
The U.S. Nuclear Regulatory Commission (NRC) is issuing this information
notice (IN) to alert addressees to recent findings from steam generator (SG)
tube inspections and investigations at Maine Yankee Atomic Power Station
(Maine Yankee). A number of tubes were degraded to the point where they
potentially no longer retained adequate structural margins to sustain the full
range of normal operating, transient, and postulated accident conditions
without rupture. This occurrence is the result of inservice inspection
deficiencies during past inspections. On December 7, 1992, NRC informed
addressees of a similar problem at Arkansas Nuclear One, Unit 2 (ANO 2), by
issuing NRC IN 92-80, "Operation With Steam Generator Tubes Seriously
Degraded." It is expected that recipients will review the information for
applicability to their facilities and consider actions, as appropriate, to
avoid similar problems. However, suggestions contained in this information
notice are not NRC requirements; therefore, no specific action or written
response is required.
Description of Circumstances
On July 15, 1994, Maine Yankee Atomic Power Company, the licensee for Maine
Yankee, shut down the plant when the measured primary-to-secondary leak rate
approached 189 liters [50 gallons] per day. The technical specification limit
is 818 liters [216 gallons] per day. Beginning at a low rate before the last
refueling outage, this leakage had slowly increased during the operating
cycle, which began in October 1993.
After shutting down the plant, the licensee tested for leaks and found four
leaking tubes. The licensee conducted an eddy current test (ECT) inspection
of the leaking tubes using a motorized rotating pancake coil (MRPC) probe and
found circumferential cracks, initiating from the tube inner diameter surface,
at the hot leg expansion transition location, which is near the top of the
tubesheet. One of these tubes contained a circumferential indication with an
average depth of 94% and extending 360 degrees around the tube circumference.
Inner diameter circumferential cracking at the tube expansion transition
location had been observed previously in other tubes at Maine Yankee during
inspections dating back to 1990.
9412160204. IN 94-88
December 23, 1994
Page 2 of 4
Reanalysis of previous inspection data for the leaking tubes, performed with
the benefit of "hindsight," indicated that circumferential indications were
present in these tubes since at least 1990 when MRPC inspections were first
performed at Maine Yankee. These indications had not been previously reported
due to the difficulty of discriminating the flaw indication from the
interference signal associated with probe "liftoff" effects caused by the
transition geometry and denting. This problem was intensified by the fact
that cracks initiating from the inner diameter surface produce a signal with a
phase angle rotation that is small in comparison to the geometry- and denting-
induced interference signal such that the composite signal was interpreted as
a normal, non-flawed indication. However, terrain plots of these signals,
generated as part of the data reanalysis, revealed the presence of the
circumferential crack indications. These terrain plots had not been generated
as part of the original field analysis for these tubes.
The licensee performed a 100% MRPC inspection of all expansion transition
locations on the hot leg side of each steam generator. The licensee employed
a 3-coil MRPC consisting of a 2.9-mm [0.115-inch] diameter pancake coil which
is sensitive to both axial and circumferential cracks, an "axial" coil which
is primarily sensitive to axial cracks, and a "circumferential" coil which is
primarily sensitive to circumferential cracks. The licensee reported that
improved signal-to-noise performance was achieved through the use of the
larger 2.9-mm [0.115-inch] diameter pancake coil in lieu of the previously
used 2.0-mm [0.080-inch] diameter coil together with the use of larger
diameter copper cables ("low loss cables") for the MRPC probe.
The licensee also employed improved data analysis procedures to reflect
information learned from the reanalysis of the previous inspection data for
the leaking tubes and from recent experience at ANO-2. All primary and
secondary analysts were trained and tested on the 1992 and 1993 data for the
Maine Yankee tubes that leaked in 1994. The primary and secondary analysts
were required to generate a pancake coil terrain plot for each tube at the
expansion transition. Any circumferential indication from the terrain plot
was to be recorded as a possible crack. Level III resolution analysts
reviewed all indications of possible cracks using voltage ratios from the
axial and circumferential coils to assist in interpretation of the pancake
coil signals. The voltage ratio criterion is based on the premise that
pancake coil indications caused by geometry or dents are likely to produce a
response on both the axial and circumferential coils whereas circumferential
crack indications are likely to produce a significant response only on the
circumferential coil. The licensee considered a voltage ratio of 2.0
(circumferential coil response divided by the axial coil response) to indicate
a circumferential crack. The licensee stated that it would likely have found
the cracks in previous inspections and plugged the tubes if it had used these
data analysis methods.
The licensee found indications of circumferential cracks in a total of 303
tubes, including the four leaking tubes. All of these tubes were plugged and
staked, irrespective of the measured depth of the indications. Several of
these indications were quite large, including 23 indications with average
. IN 94-88
December 23, 1994
Page 3 of 4
depths (over the 360-degree circumference) exceeding 79 percent of the tube
wall thickness and 10 indications with average depths exceeding 89 percent of
the tube wall thickness. The licensee reports that 79 percent is the
allowable average crack depth, consistent with the most limiting burst
pressure criterion from Regulatory Guide (RG) 1.121, "Bases for Plugging
Degraded PWR Steam Generator Tubes," August 1976. Analyses performed for a
similar plant (ANO-2) indicate that a tube with an average crack depth
exceeding 89 percent may not sustain a postulated main steam line break (MSLB)
pressure of 17,340 kPa [2,500 psi]. Most of the indications found were
determined to date back to at least 1990.
Discussion
The recent eddy current test results at Maine Yankee indicate that certain
tubes may have degraded to the point that they did not meet the structural
margin criteria of Regulatory Guide 1.121 and that, in addition, some of these
tubes may not have been capable of sustaining the differential pressures
associated with a postulated MSLB. That this was, in fact, the case is not
conclusive, given the large degree of uncertainty associated with eddy current
depth measurements for cracks and the inability of eddy current testing to
resolve small ligaments of sound material between crack segments.
The licensee performed in-situ pressure tests of ten tubes at Maine Yankee
containing some of the largest indications to assess their actual burst
integrity. The results of the in-situ pressure tests indicate that the eddy
current test measurements for at least some of the cracks were conservative.
The licensee concluded on the basis of these tests that the majority of tubes
with MRPC sized indications exceeding 79 percent average through wall had
structural margins consistent with the most limiting Regulatory Guide 1.121
criterion and that all tubes were capable of sustaining an MSLB without
rupture. The staff is evaluating the results of the in-situ pressure tests,
but has not yet reached a conclusion regarding the validity of the tests to
simulate an actual pressure transient in the steam generators.
Inadequate eddy current test procedures from 1990 or earlier appear to be the
primary reason the tubes at Maine Yankee became severely degraded before their
recent discovery. In INs 90-49, "Stress Corrosion Cracking in PWR Steam
Generator Tubes," of August 6, 1990, and 92-80, "Operation with Steam
Generator Tubes Seriously Degraded," of December 7, 1992, NRC stressed the
importance of using appropriate probes such as pancake type coils when
inspecting locations that are potentially subject to circumferential cracks.
The Maine Yankee findings, however, indicate that use of pancake type coils is
not necessarily sufficient to ensure the timely detection of circumferential
cracks.
This difficulty in obtaining accurate eddy current test results also
demonstrates the importance of (1) optimizing the test methods to minimize
electrical noise and signal interference and to maximize flaw sensitivity; (2)
anticipating potential sources of interfering signals, such as from probe
liftoff caused by tube transition geometry and from dents and understanding
their potential effect on flaw detection; (3) developing test and analysis . IN 94-88
December 23, 1994
Page 4 of 4
procedures that will allow the flaw signal to be discriminated from any
unavoidable signal noise or interference; and (4) being alert to plant unique
circumstances (e.g., dents, copper deposits) which may necessitate special
test procedures found not to be necessary at other similarly designed steam
generators or not included as part of a generic technique qualification.
Recent information from ANO-2, where circumferential cracks initiated from the
outer diameter surface at the expansion transition, indicates that procedures
which rely on what nondestructive examination analysts refer to as "good or
expected phase angle correlation" between base test frequencies can lead to
missed indications. Data from pulled tube specimens are useful for developing
and validating effective inspection methods for each plant. Appropriate
training and performance demonstration testing of the data analysts on the
test and analysis procedures are essential elements of an effective inspection
program.
Circumferential cracks have been reported at the tube expansion transition
locations of several Combustion Engineering (CE) designed plants (including
Maine Yankee and ANO-2) and at several Westinghouse plants. Tubes in the CE
steam generators and at most of the affected Westinghouse plants were
explosively expanded against the tubesheet (called an "explansion" process at
CE units and a "WEXTEX" process at Westinghouse units). Tubes in the
remaining affected Westinghouse plants were roll expanded against the
tubesheet.
The staff is continuing to evaluate the generic implications of the Maine
Yankee occurrence.
This information notice requires no specific action or written response. If
you have any questions about the information in this notice, please contact
the technical contact listed below or the appropriate Office of Nuclear
Reactor Regulation (NRR) project manager.
/S/'D BY BDLIAW/FOR
Brian K. Grimes, Director
Division of Project Support
Office of Nuclear Reactor Regulation
Technical contact: Emmett L. Murphy, NRR
(301) 504-2710
Attachment:
List of Recently Issued NRC Information Notices
