Information Notice No. 94-05: Potential Failure of Steam Generator Tubes with Kinetically Welded Sleeves
NUCLEAR REGULATORY COMMISSION
OFFICE OF NUCLEAR REACTOR REGULATION
WASHINGTON, D.C. 20555
January 19, 1994
NRC INFORMATION NOTICE 94-05: POTENTIAL FAILURE OF STEAM GENERATOR TUBES
WITH KINETICALLY WELDED SLEEVES
All holders of operating licenses or construction permits for pressurized
water reactors (PWRs).
The U.S. Nuclear Regulatory Commission (NRC) is issuing this information
notice (IN) to alert addressees to the potential failure of steam generator
tubes sleeved with kinetically (explosively) welded sleeves supplied by B&W
Nuclear Service Company (BWNS). 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.
For certain types of defects, a steam generator tube may be repaired using an
approved sleeving method as an alternative to plugging the tube and removing
it from service. In this process, the sleeve is positioned inside the steam
generator tube so that it bridges the defect. The sleeve is then joined to
the parent tube on both sides of the defect to serve as a new primary coolant
interface and allow the tube to be returned to service. The sleeve can be
joined to the tube wall by a mechanical seal or a weld. In the case of the
kinetic welding process used by BWNS, an explosive charge expands a narrow
band of the sleeve, fusing the outer sleeve wall to the inner tube wall. The
process leaves residual stresses in the parent tube in the vicinity of the
seal or weld which necessitates a post-weld heat treatment to relieve the
stresses. The heat treatment is necessary in the parent tube because the tube
is constructed of nickel Alloy 600 which is susceptible to stress corrosion
cracking. Heat treatment is not necessary for the repair sleeve because it is
made of nickel Alloy 690 which is more resistant to stress corrosion cracking.
Description of Circumstances of the McGuire 1 Incident
On August 22, 1993, operators at Unit 1 of the William B. McGuire Nuclear
Station (McGuire) shut down the reactor because of a primary-to-secondary leak
of about 760 liters [200 gallons] a day in steam generator A. This amount of
leakage was within the technical specification limits but exceeded the
administrative limit. Duke Power Company, the licensee, determined that a
tube containing a BWNS kinetically welded sleeve was the source of the leak.
January 19, 1993
Page 2 of 4
The licensee removed the tube and the sleeve and found a circumferential crack
in the parent tube just above the upper weld that joined the tube and the
sleeve. The sleeve had been installed in 1991 and had been reinspected in
April 1993 with no indications of cracking. The licensee used a rotating eddy
current probe to examine the tube in situ and found an indication of a
120-degree to 180-degree circumferential defect. Destructive examination of
the tube found a through-wall circumferential crack extending 270 degrees
around the tube. The remaining 90 degrees of the tube was cracked
approximately 50 percent through the wall. The crack had initiated from the
inner surface (primary coolant side) and was characteristic of primary water
side stress corrosion cracking (PWSCC). This type of stress corrosion
cracking is a well-known failure mechanism in steam generator tubes. The
mechanism is discussed in greater detail in the NRC information notices
referenced at the end of this notice.
Another tube that had been sleeved at the same time as the leaking tube was
also removed because an eddy current indication was found in the same area as
in the failed tube. The indication was not a defect but was found to be the
result of a variation in the surface geometry. Metallographic examination of
the sleeve and parent tube also showed no signs of cracking. A review of
process records showed that both tubes had received the same stress relief
temperature and time. Hardness measurements confirmed that both tubes had
been stress relieved after the kinetic welding.
Physical and chemical tests performed on the two tubes that were removed
showed significant differences in the yield strength, carbon content,
microstructure, and PWSCC susceptibility of the tube material. Tests
sponsored jointly by Studsvik Power, the Swedish State Power Board, and
AB Sandvick Steel show that a strong correlation exists between the carbon
content and yield strength of the material and its susceptibility to PWSCC.
Results of accelerated corrosion tests indicate that the time to cracking for
reverse U-bend tubes is shorter for tubes constructed of materials with
elevated yield strengths and carbon content. Based on a yield strength of
51.7 MPa (72.5 ksi) and a 0.05-percent carbon content, the material heat of
the leaking tube would be ranked as one of the most susceptible to PWSCC in
the plant. Metallographic examination of the tube material confirmed that it
had a susceptible microstructure.
The BWNS kinetic sleeve stress-relieving process was originally qualified for
a range of material corrosion susceptibilities. The stress relief
temperatures were selected for what was believed to be the worst-case
material. As evidenced by the destructive examination of the tube that
leaked, material properties of steam generator tubes can be significantly
different than the properties listed in a Certified Material Test Report
(CMTR). Based on the CMTR, the yield strength of the leaking tube should have
been 44.1 MPa (64 ksi), whereas the actual yield strength was 51.7 MPa
(72.5 ksi). As a result of this and other industry data, BWNS will be
evaluating the appropriateness of using alternate stress relief cycles for
tubing of higher yield strengths.
The destructive examination of the sleeved tube at McGuire indicates that the
root cause of the parent tube leak was the high susceptibility of the parent.
January 19, 1994
Page 3 of 4
tube material to stress corrosion cracking. Since 1990, approximately 4500
sleeves manufactured by BWNS have been installed worldwide using the kinetic
weld process. According to BWNS, the defective tube sleeve at McGuire is the
first confirmed case of cracking in a sleeved tube that had received the
post-weld stress relief treatment required in the process qualification. A
kinetically sleeved tube cracked and leaked at the Trojan Nuclear Plant in
1992 but that tube had not received the required post-weld stress relief
In September 1993, B&W Nuclear Technologies contacted affected domestic
licensees to inform them that the destructive examination of the sleeved tube
at McGuire indicated that the root cause of the leak in the parent tube was a
high susceptibility of the parent tube material to stress corrosion cracking.
Five domestic nuclear units have installed tube sleeves using the BWNS process
and a number of others are licensed to install them.
After the incident at McGuire 1, B&W Nuclear Technologies made recommendations
to licensees with BWNS kinetically welded sleeves concerning (1) the
identification of highly susceptible parent tube material, (2) procedures for
dealing with primary-to-secondary leakage, and (3) operator readiness to
respond to a tube leak such as the one that occurred at McGuire. Duke Power
Company implemented these recommendations at McGuire 1 and for preventive
purposes plugged sleeved tubes that were not axially restrained (peripheral
tubes not completely surrounded by other tubes).
This failure mechanism has the potential for introducing difficult-to-detect
circumferential stress corrosion cracks in steam generator tubes which could
lead to rapidly increasing primary-to-secondary leakage. The NRC staff has
contacted all affected licensees concerning the implications of these findings
and is continuing to monitor this issue.
Related Generic Communications
1. NRC IN 92-80, "Operation With Steam Generator Tubes Seriously Degraded,"
December 7, 1992.
2. NRC IN 91-43, "Recent Incidents Involving Rapid Increases in
Primary-to-Secondary Leak Rate," July 5, 1991.
3. NRC IN 90-49, "Stress Corrosion Cracking in PWR Steam Generator Tubes,"
August 6, 1990.
4. NRC IN 88-99, "Detection and Monitoring of Sudden and/or Rapidly
Increasing Primary-to-Secondary Leakage," December 20, 1988.
January 19, 1994
Page 4 of 4
No specific action or written response is required by this information notice.
If you have any questions about this matter, please contact the person listed
below or the appropriate Office of Nuclear Reactor Regulation (NRR) project
/s/'d by BKGrimes
Brian K. Grimes, Director
Division of Operating Reactor Support
Office of Nuclear Reactor Regulation
Technical contact: H. Conrad, NRR
List of Recently Issued NRC Information
Page Last Reviewed/Updated Thursday, March 25, 2021