Information Notice No. 93-82: Recent Fuel and Core Performance Problems in Operating Reactors

                                 UNITED STATES
                       NUCLEAR REGULATORY COMMUNICATION
                     OFFICE OF NUCLEAR REACTOR REGULATION
                            WASHINGTON, D.C. 20555

                               October 12, 1993


NRC INFORMATION NOTICE 93-82:  RECENT FUEL AND CORE PERFORMANCE PROBLEMS       
                               IN OPERATING REACTORS


Addressees

All holders of operating licenses or construction permits for nuclear power
reactors and all NRC-approved fuel suppliers.

Purpose

The U.S. Nuclear Regulatory Commission (NRC) is issuing this information
notice to alert addressees of recent fuel and core performance problems in
operating reactors resulting from deficiencies and oversights in evaluating
the impact of fuel design changes and mixed core behavior in the fuel and
reload core design process.  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  

Salem Unit 2 and Beaver Valley Unit 1

During shutdown inspection activities after Cycle 7 at Salem Unit 2 and  
Cycle 9 at Beaver Valley Unit 1, the licensees at both plants discovered
numerous fuel rods that had developed fretting wear and perforation.  The fuel
vendor attributed the degradation to grid-to-rod fretting resulting from flow-
induced vibration of the fuel bundles.  All but one of the affected fuel
assemblies were Westinghouse twice-burned VANTAGE 5H fuel located next to the
core baffle or with a history of previous operation at a peripheral location. 
The fretting wear occurred at the zircaloy mid-grid spacers rather than at
lower grid locations where debris-induced fretting wear typically occurs.  In
some of the affected assemblies, secondary hydriding also was evident.  

Wolf Creek

During the Cycle 5 shutdown at Wolf Creek, the licensee discovered 44 failed
fuel rods in three Westinghouse Batch F (standard) fuel assemblies that had
completed two cycles of operation with accrued assembly average burnups of
24,000 MWd/MTU.  The most severely degraded fuel rod fragmented into three
segments during fuel handling operations while offloading the core.  

9310070
.

                                                            IN 93-82
                                                            October 12, 1993
                                                            Page 2 of 5


Inspection of the failed fuel rods revealed the failures to be grid-to-rod
fretting wear at the lower three grid locations on in-core interior fuel
assemblies.  During Cycle 6 operation, additional fuel failures occurred in
three Batch F assemblies in their third cycle of operation with assembly
average burnups of 38,500 MWd/MTU.  The occurrence was a virtual repeat of the
Cycle 5 experience, including the discovery of a broken fuel rod during core
offloading.  These fuel failures have been attributed to high cross flows,
caused, in part, from mixed fuel designs which induced fuel rod vibration with
fretting wear at the lower grids.  The mixture of standard and VANTAGE 5H fuel
(with debris filter bottom nozzles) resulted in axial mismatches between the
bottom nozzles and the grid spacers of the two fuel types.  The failed
assemblies were all standard fuel and had been subjected to high cross flow. 
The transition to VANTAGE 5H fuel had commenced with Cycle 5.  

Palisades

Before Cycle 11 startup at Palisades, three separate segments of a damaged
fuel rod were discovered in the tilt pit area of the reactor cavity and a 
0.15 m [6 in] segment containing the upper end cap was found in the fuel
assembly upper grid spacer.  The upper 1.68 m [5.5 ft] segment in the tilt pit
was split with the plenum spring exposed and about 1/3 of the full fuel rod
inventory of fuel pellets was missing.  The second segment (1.37 m [4.5 ft] in
length) and the third segment (0.46 m [1.5 ft] long with the lower end plug)
appeared to have fuel pellets intact.  The fuel rod was from a corner rod
location in a peripheral fuel assembly which had been located in the corner of
the core adjacent to the core baffle during Cycle 10 operation.  

The defective fuel assembly was one of a batch of 16 fuel assemblies,
fabricated by Siemens Nuclear Power Co. (SNP), located on the core periphery
since Cycle 9 that had 8-rod hafnium absorber clusters inserted for vessel
fluence reduction.  Inspection of this fuel assembly and others from the batch
of 16 fuel assemblies revealed that the spacer springs had relaxed and there
was up to 90-percent through-wall fretting wear at mid-grid locations,
predominantly on the corner rods of affected fuel assemblies.  Although the
typical exposure for these fuel assemblies was only 80 percent of the design
life, they had been subjected to the reactor environment and fast flux
exposure during five cycles of operation.  Tests by SNP confirmed that the
lengthy service life was a contributory cause to the spring relaxation. 
 
During Cycle 10, the licensee operated the plant with two batches of the SNP
fuel assemblies with high thermal performance flow mixing spacers.  Flow
mixing spacers in neighboring fuel assembly locations and the low power of the
peripheral fuel assemblies adjacent to the core baffle produced a hydraulic
mismatch that caused cross flow into the peripheral fuel assemblies.  In
addition, the flow forces had increased when primary coolant flow increased 
7 percent after the replacement of steam generator tube bundles before Cycle 9
startup.  Limited inspection by the licensee indicated that corner fuel rods 
.

                                                            IN 93-82
                                                            October 12, 1993
                                                            Page 3 of 5


closest to the baffle were the most severely worn.  The licensee also found
corresponding wear locations between the most severely damaged fuel assembly
and the core baffle, indicating that the fuel assembly had rubbed against the
baffle, resulting in severe spacer damage. 

Hatch Unit 2, Peach Bottom Unit 3, and Perry

Recent fuel failures found at Hatch Unit 2 and Peach Bottom Unit 3 in General
Electric (GE) zirconium-barrier fuel have been attributed to debris fretting
or to undetected manufacturing defects.  At Perry, leaking fuel rods in two
bundles during Cycle 1, one bundle during Cycle 2, and two bundles during
Cycle 3 are believed to have been the result of bad end cap welds.  However,
during the recently completed Cycle 4, one severely damaged fuel rod (with a
50.8 cm [20 inch] long large axial crack as a result of secondary hydriding)
and possible leaking fuel rods in several other bundles were noted.  These
fuel bundles were fabricated to avoid repetition of the end cap problem.  
GE believes that the root cause of the failures is undetected manufacturing
defects, possibly exacerbated by the Perry operating practice of using control
rod movement rather than flow control for minor power adjustments.  

DISCUSSION  

Recent operating experience of pressurized-water reactors has identified
debris-induced fretting as a leading cause of fuel failure.  However, current
experience also indicates that a new type of vibrational fretting is emerging. 
During ultrasonic examination of VANTAGE 5H failed fuel assemblies at Salem
and Beaver Valley, Westinghouse confirmed grid-to-rod fretting at the zircaloy
low-pressure drop mid-grids.  Most of the failures occurred near the core
baffle or in fuel that had resided near the core baffle in a preceding cycle. 
Through several tests, Westinghouse determined that the occurrence of flow-
induced vibrational fretting was likely.  This vibrational fretting involves
the natural frequency and flow condition for fuel assemblies adjacent to the
core baffle.  To avoid this type of fretting wear, Westinghouse proposed
corrective actions for existing fuel assemblies adjacent to the core baffle to
dampen the vibrations.  The long-term corrective actions will include design
modifications and more extensive vibration testing to ensure that the natural
frequency of the fuel assembly does not correspond to flow conditions that may
exist in the reactor.  

For Wolf Creek, the fretting wear involved a mixed core of standard fuel and
VANTAGE 5H fuel with inclusion of intermediate flow mixer grids and debris
filter bottom nozzles.  The failures were located at the lower grids,
especially near debris filter bottom nozzles.  Westinghouse performed a
sensitivity study to analyze the flow condition near the debris filter bottom
nozzle region for mixed cores.  The analysis indicated that excessive
vibration of the fuel rods of the standard fuel assembly may have been caused
by cross flow from VANTAGE 5H fuel to the standard fuel and that the mixed
core of different bottom nozzle designs may have contributed to such a cross
flow phenomenon.  

.

                                                            IN 93-82
                                                            October 12, 1993
                                                            Page 4 of 5


An issue involving the compatibility of mixed cores in a boiling-water reactor
was identified during the NRC evaluation of the Washington Nuclear Power 
Unit 2 (WNP-2) instability event of August 15, 1992.  At that time, it was
noted that an abnormally large mismatch in flow and pressure drop between
neighboring fuel assemblies in a mixed core was further exaggerated by the
core reload operating scheme and may have contributed to thermal-hydraulic
instability of the core.  Similarly, it now appears that core designs
comprised of mixed-fuel types sometimes result in physical mismatches of
neighboring fuel assemblies that may be contributing to flow-induced fretting
wear failure of fuel rods in pressurized-water reactors.  The mixing of fuel
of different types is further complicated by design variations introduced by
fuel reconstitution and by the purchase options offered by at least one fuel
vendor (Westinghouse) to include design features such as intermediate flow
mixer grids and debris filter bottom nozzles with the fuel type of customer
choice.  This has resulted in one known instance (at Turkey Point Unit 3) of
an axial offset in power distribution as a result of a small axial mismatch
between burnable poison rods and the active core.  The mismatch occurred when
debris filter bottom nozzles were included with the fuel type chosen by the
licensee, but a corresponding revision of burnable poison rod specifications
was overlooked.
  
The above problems raise questions about the effectiveness of design quality
assurance, as implemented by some licensees and vendors to ensure that the
reactor core continues to conform to its design bases when new and modified
fuel designs are introduced during core reload.  The degree of core design
responsibility exercised by the fuel supplier varies significantly depending
to a large extent on the wishes of each individual licensee.  Defining reload
core design safety evaluation and quality assurance responsibilities,
especially the interaction with fuel suppliers, can help assure that licensee
and vendor design responsibilities are properly implemented.
.

                                                            IN 93-82
                                                            October 12, 1993
                                                            Page 5 of 5


No specific action or written response is required by this information notice. 
If you have any questions regarding this matter, please contact one of the
technical contacts listed below or the appropriate Office of Nuclear Reactor
Regulation (NRR) project manager.



                                    Brian K. Grimes, Director
                                    Division of Operating Reactor Support
                                    Office of Nuclear Reactor Regulation

Technical contacts: S. L. Wu, NRR
                    (301) 504-3284

                    L. E. Phillips, NRR
                    (301) 504-3232

Attachment:  List of Recently Issued NRC Information Notices
.
 

Page Last Reviewed/Updated Tuesday, March 09, 2021