United States Nuclear Regulatory Commission - Protecting People and the Environment

Information Notice No. 97-46: Unisolable Crack in High-Pressure Injection Piping

                                 UNITED STATES
                         NUCLEAR REGULATORY COMMISSION
                     OFFICE OF NUCLEAR REACTOR REGULATION
                         WASHINGTON, D.C.  20555-0001

                                 July 9, 1997


NRC INFORMATION NOTICE 97-46:  UNISOLABLE CRACK IN HIGH-PRESSURE INJECTION     
                               PIPING


Addressees

All holders of operating licenses or construction permits for nuclear power
reactors.

Purpose

The U.S. Nuclear Regulatory Commission (NRC) is issuing this information
notice to alert addressees to the discovery of a leaking cracked weld in an
unisolable section of a combined makeup (MU) and high-pressure injection (HPI)
line at Oconee Unit 2.  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 April 22, 1997, at 12:50 p.m., Oconee Unit 2 was shut down because of
unidentified reactor coolant system (RCS) leakage exceeding the technical
specification limit of 1 gpm.  From the time of initial leak indications on
April 21, at approximately 10:45 p.m., until reactor pressure was sufficiently
reduced, the leakage rate rose from approximately 2 gpm to a maximum leakage
rate of approximately 12 gpm.  A subsequent containment entry identified an
unisolable leak in the MU/HPI line 2A1 from a through-wall crack in the weld
connecting the MU/HPI pipe and the safe-end of the 2A1 reactor coolant loop
(RCL) nozzle.

Discussion

The Oconee 2A1 MU/HPI nozzle assembly consists of the MU/HPI 2� inch diameter
pipe/safe-end/thermal sleeve (see Figure 1 - Original Design).  The sleeve is
attached by contact rolling to the inner surface of the safe-end.  A 1-inch
diameter "warming" line taps into the bottom of the MU/HPI pipe immediately
upstream of the pipe/safe-end weld where the through-wall crack was found. 
This line permits a small continuous MU flow (3 gpm) to reduce nozzle thermal
transients due to changes in normal MU flow.  All Oconee units have two
combined MU/HPI lines and two additional HPI lines connected to the RCS. 
However, the thermal sleeve configuration in Oconee Unit 1 is different from
that in Units 2 and 3. 

Preliminary analysis indicates that crack initiation and propagation in the
weld was caused by high-cycle fatigue due to a combination of thermal cycling
and flow induced vibration.  The 

9707020306.                                                            IN 97-46
                                                            July 9, 1997
                                                            Page 2 of 4

metallurgical examination of the weld determined that the crack consisted of a
360� inside surface flaw.  The flaw depth increased gradually from about 30
percent into the wall until it became through-wall over a 77� arc length (see
Figure 2).  The examination found a gap in the contact area between the
thermal sleeve and the safe end, indicative of loss of contact that caused the
thermal sleeve in this line to be loose (see Figure 1).  The thermal sleeve
was found to be cracked, with portions missing from the end that extends into
the RCS flow path.  Significant wear damage was observed at both the upstream
(the rolled end) and the downstream end.  Cracking was also found in the pipe
in the vicinity of the "warming" line nozzle.  Video examinations of the other
thermal sleeves of the HPI system showed no evidence of damage.  Ultrasonic
Testing (UT) and Radiographic Testing (RT) of the welds and the thermal
sleeves in the other HPI nozzles showed no indications of cracking or
loosening, or other signs of degradation.  Figure 1 shows a comparison of the
original and new thermal sleeve designs.  The thermal sleeve in the 2A1 MU/HPI
line was replaced during the current outage with the new design thermal
sleeve.

Although the root cause of the cracking is not well understood, the licensee
has identified a number of thermal/mechanical conditions that may have
contributed to the crack propagation of the 2A1 pipe to safe-end weld.  The
precise contribution to cracking of each of these conditions is not presently
known.  However, the licensee has hypothesized that, in addition to the
thermal cycling experienced at the nozzle during heat up/cool down and other
plant transients, a likely contributor to the fatigue may have been the
alternate heating and cooling of the weld by intermittent mixing of the hot
reactor coolant leaking through the gap in the contact area between the loose
thermal sleeve and the safe-end, and the cooler normal makeup water flowing
through the associated MU/HPI line.  Although the precise contribution of the
gap is unknown, it is believed that a gap may be a prerequisite for cracking
in the piping since the cracked pipes also had gaps between the thermal sleeve
and the safe end.
  
This phenomenon was identified as the probable cause for similar safe-end
cracking observed at Crystal River and other B&W plants (including Oconee) in
the early 1980's. This issue was previously addressed in Information Notice
82-09 and Generic Letter 85-20.  

Recent re-examination of radiographs made in April 1996 of the Oconee 2A1
nozzle revealed that the licensee had failed to identify the gap which had
developed in the safe-end/thermal sleeve contact area.  The licensee also had
failed to follow the original recommendations for augmented ultrasonic testing
(UT) as listed in NRC Generic Letter 85-20, "High Pressure Injection/Make-Up
Nozzle Cracking in Babcock and Wilcox Plants," issued November 8, 1985.  The
licensee performed the recommended UT of the safe ends of the MU/HPI lines;
however, they did not inspect the adjacent piping as recommended.  In
addition, the licensee failed to UT the weld between the safe-end and pipe, a
discontinuity where cracking would be expected, and did, form. Also, NRC
Bulletin 88-08, Supplement 1, "Thermal Stresses in Piping Connected to Reactor
Coolant Systems," issued August 4, 1988, emphasized that, because of the
difficulty in identifying the types of cracks that were occurring due to
thermal stresses, the need exists for enhanced UT and for experienced
examination personnel to detect the cracks..                                                            IN 97-46
                                                            July 9, 1997
                                                            Page 3 of 4


The licensee also reviewed the 1996 radiographs of the safe-ends in Oconee
Unit 3.  The 3A1 MU/HPI line was found to have a gap in the safe-end/thermal
sleeve contact area.  As a result of the gap in the 3A1 safe-end, Oconee Unit
3 was shut down on May 2, 1997.  UT examinations identified apparent cracking
in the 3A1 safe-end.  This safe-end has been removed and is presently being
metallurgically examined, but a visual examination has also revealed cracks in
the thermal sleeve.  Minor gaps in the other safe-end/thermal sleeve contact
areas were determined not to have grown, the rolled area of the thermal sleeve
was acceptable, and UT examinations of the other Oconee Unit 3 HPI nozzle
assemblies revealed no cracking.

The Oconee Unit 1 nozzles have a double thermal sleeve design (Figure 3). 
Radiographic inspection in the period from 1983 to 1989 indicated that no gap
existed in three of the four thermal sleeves.  The thermal sleeve in the 1B2
(HPI) line had a gap; but, the gap had not grown during the inspection period. 
Advantages of the double thermal sleeve as stated by the licensee include: 
(1) greater stiffness; (2) greater thermal resistance; and (3) reduced flow
area, with corresponding increased flow velocity.

General Design Criterion 14 of Appendix A to Part 50 of Title 10 of the Code
of Federal Regulations requires that the reactor coolant pressure boundary be
designed so as to have an extremely low probability of abnormal leakage, of
rapidly propagating failure, and of gross rupture.  The related generic
communications listed below discuss several other similar events, and the
actions that licensees were requested to take to reduce the probability of
additional similar events occurring.  

Similar Recent Events

On December 14, 1996, a non-isolable leak on piping connecting the safety
injection system to the reactor coolant system was found in Dampierre Unit 1
in France.  The damaged pipe length was examined and a through wall crack
located on an uninterrupted portion of straight piping (not on a stressed area
such as a weld or a bend).  The licensee has not identified the root cause of
the cracking, but concluded that the most probable cause was temperature
variations produced by cold water coming from leaking valves located upstream
in the safety injection system.  The licensee also concluded that the presence
of a through-wall defect on a straight portion of a pipe is likely to raise
questions about previous assumptions made regarding the root cause of the
cracking. 

Related Generic Communications

NRC INFORMATION NOTICE 82-09, "CRACKING IN PIPING OF MAKEUP COOLANT LINES AT
B&W PLANTS," dated March 31, 1982.

NRC GENERIC LETTER 85-20, "RESOLUTION OF GENERIC ISSUE 69: HIGH PRESSURE
INJECTION/MAKEUP NOZZLE CRACKING IN BABCOCK AND WILCOX PLANTS," dated November
11, 1985.

.                                                            IN 97-46
                                                            July 9, 1997
                                                            Page 4 of 4


NRC BULLETIN NO. 88-08, "THERMAL STRESSES IN PIPING CONNECTED TO REACTOR
COOLANT SYSTEMS," dated June 22, 1988.

NRC BULLETIN NO. 88-08, Supplement 1, "THERMAL STRESSES IN PIPING CONNECTED TO
REACTOR COOLANT SYSTEMS," dated June 24, 1988.

NRC BULLETIN NO. 88-08, Supplement 2, "THERMAL STRESSES IN PIPING CONNECTED TO
REACTOR COOLANT SYSTEMS," dated August 4, 1988.

NRC BULLETIN NO. 88-08, Supplement 3, "THERMAL STRESSES IN PIPING CONNECTED TO
REACTOR COOLANT SYSTEMS," dated April 11, 1989.

NRC INFORMATION NOTICE 97-19, "SAFETY INJECTION SYSTEM WELD FLAW AT SEQUOYAH
NUCLEAR POWER PLANT, UNIT 2," dated April 18, 1997

This information notice requires no specific action or written response.  If
you have any questions about the information in this notice, please contact
one of the technical contacts listed below or the appropriate Office of
Nuclear Reactor Regulation (NRR) project manager.


                                          signed by S.H. Weiss for

                                    Marylee M. Slosson, Acting Director
                                    Division of Reactor Program Management
                                    Office of Nuclear Reactor Regulation

Technical contacts:  Barry Elliot, NRR       Eric Benner, NRR
                     301-415-2709            301-415-1171
                     E-mail:  bje@nrc.gov    E-mail:  ejb1@nrc.gov

                     Kamal Manoly, NRR       Mark Hartzman, NRR
                     301-415-2765            301-415-2755
                     E-mail:  kam@nrc.gov    E-mail:  mxh@nrc.gov

Attachments:  
1.  Figure 1 - Thermal Sleeve
2.  Figure 2 - Warming Line Flow
3.  Figure 3 - Unit 1 Thermal Sleeve
4.  List of Recently Issued NRC Information Notices
Page Last Reviewed/Updated Tuesday, December 03, 2013