Information Notice No. 94-40, Supplement 1:Failure of a Rod Control Cluster Assembly to Fully Insert Following a Reactor Trip at Braidwood Unit 2

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

                               December 15, 1994


NRC INFORMATION NOTICE 94-40, SUPPLEMENT 1:  FAILURE OF A ROD CONTROL CLUSTER  
                                             ASSEMBLY TO FULLY INSERT FOLLOWING 
                                             A REACTOR TRIP AT BRAIDWOOD UNIT 2


Addressees

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

Purpose

The U.S. Nuclear Regulatory Commission (NRC) is issuing this supplemental
information notice to alert addressees to new events involving detached guide
funnels underneath the reactor head.  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.

Background

The original information notice, dated May 26, 1994, described an event at
Braidwood Station, Unit 2, on April 5.  After receiving a valid reactor trip
signal, the control rod located in core position K-2 inserted only to step 210
because a loose part prevented complete insertion.  The loose part was a
mechanical pin detached from a threaded connection joining a funnel and
thermal sleeve for a thermocouple column penetration of the reactor head.  The
funnel guides the thermocouple column in aligning with connections below the
reactor head, the pin prevents the threaded parts from rotating, and the cap
weld keeps the pin in place.  Subsequently, the licensee recovered a loosened
but not detached pin from its position adjacent to the original position of
the loose pin and a second loose pin from another funnel.  The results of a
metallurgical evaluation of these three failed pins are provided in the
discussion section of this supplement.  Also described below are events that
involve similar funnels used to align control rod drive shafts instead of
thermocouple columns.

Description of Circumstances

Byron Station, Unit 2

On October 8, 1993, the licensee, the Commonwealth Edison Company, discovered
during a refueling outage that the control rod drive shaft funnel at position
B-8 was uncoupled and resting on the upper internal structure.  The licensee
determined that the funnel had not been welded during original construction. 

9412090238.                                                            IN 94-40, Supp. 1
                                                            December 15, 1994
                                                            Page 2 of 4


The licensee welded the funnel to the thermal sleeve using a plug weld
according to design, inspected all remaining control rod drive shaft funnels,
and found them secure.

Sequoyah Nuclear Plant, Unit 2

On July 22, 1994, the licensee, the Tennessee Valley Authority, observed
during a refueling outage, while placing the reactor head onto the reactor
vessel, that the control rod drive shaft at core position H-4 failed to align
into the thermal sleeve guide tube and was severely damaged as the head
continued to be lowered.  The licensee stopped lowering the reactor head,
inspected, and found that the H-4 drive shaft was misaligned and bent into an
"S" shape.  The associated funnel was detached from its thermal sleeve and was
resting around the drive shaft on the upper internal structure.  The licensee
machined the replacement funnel to remove the internal threads and welded the
funnel to the thermal sleeve.

The licensee sent the H-4 guide funnel to a hot cell facility for a failure
analysis.  The licensee concluded that the funnel had become detached as a
result of wear of the retainer pin due to flow-induced vibration and
subsequent loss of preload on the funnel.   

The Sequoyah licensee used a remotely controlled device to inspect all the
component guide funnels underneath the reactor head and verified that the
threads were engaged, the pins were intact, the thermal sleeves had not been
drilled through, and the funnels had no lateral play.  The results indicated
that two control rod drive shaft funnels had slight lateral play of about     
3.2-1.6 mm [1/8-1/16 inch].  The licensee concluded that portions of the pin
for that funnel had likely been worn away because of flow-induced vibration,
that all pieces of the failed H-4 pin were accounted for, and that the funnel
could become unthreaded and detached if the funnel was not fastened properly
with the pin.

Diablo Canyon Nuclear Power Plant, Unit 1

On September 29, 1994, the licensee, the Pacific Gas and Electric Company,
found during an inservice inspection that the control rod drive shaft funnels
at core positions N5, N9, H2, H6, C7, and B6 could move laterally as much as
4.8 mm [3/16 inch].  The licensee repaired each loose funnel, using two stitch
welds spaced 180 degrees apart.  The licensee believes the funnels were
loosened by flow-induced fretting of the dowel pin-hole interface.  

Braidwood Station, Unit 2

On October 14, 1994, the licensee, the Commonwealth Edison Company, discovered
during a refueling outage that the control rod drive shaft funnel at position
B-8 was uncoupled and resting on the upper internal structure.  The licensee
determined by visual inspection that the design plug weld on the uncoupled
funnel was intact but that weld residue was on the thermal sleeve.  The
licensee observed no other loose funnels..                                                            IN 94-40, Supp. 1
                                                            December 15, 1994
                                                            Page 3 of 4


Donald C. Cook Nuclear Power Plant, Unit 2

On October 14, 1994, while inspecting the reactor vessel head, the licensee,
the Indiana/Michigan Power Company, found that 42 control rod drive shaft
funnels could move approximately 1.6 mm [1/16 inch] and that the thermocouple
column funnel at core position R-11 did not have a hole and dowel pin as
designed.  The licensee repaired the control rod funnels using two stitch
welds, spaced 180 degrees apart, and repaired the thermocouple column funnel
using a pin-and-cap weld, according to design.

Discussion

On September 9, 1994, the Braidwood licensee received the results of the
metallurgical examination of the three thermocouple column funnel anti-
rotation locking pins that had failed at Braidwood, Unit 2, in April 1994 (the
subject of the original information notice).  The components were constructed
of the following materials:

      Locking pins                  ASME SA-479, Type 304 stainless steel
      Funnels                       ASME SA-240, Type 304 stainless steel
      Thermocouple columns          Inconel 600, nickel based alloy

The pins were 2.2 cm [0.88 inch] long by 0.95 cm [3/8 inch] in diameter with a
0.038-cm [0.015-inch] chamfer at each end.  

Metallographic examination verified that all three pins had been plug-welded
to the housing guide during original construction.  The welds were applied in
multiple passes and were not porous.

Energy dispersive x-ray analysis indicated that the field-welded pins loosened
because of combined vibration-induced wear and fatigue cracking.  All three
pins were worn on the ends, which were smooth and polished in a manner typical
of wear caused by vibration-induced movement.  Further, the wear was
predominantly along one side of the pins, indicating unidirectional loading. 
The licensee postulated that as pin wear increased, the vibration-induced
movement increased, creating high stresses at the plug welds, which eventually
caused the welds to crack.  Cracking then continued to propagate through the
welds between the pins and the guide to form a circumferential crack.  The
licensee attributed the cause of the weld failure to fatigue cracking because
no secondary cracking or tearing was observed.  The additional vibration-
induced movement eventually wore away enough of the weld metal to allow the
pins to back out of the funnels.

On January 28, 1994, in response to the 1993 Byron event, the nuclear steam
supply system vendor, Westinghouse Corporation, issued a Nuclear Safety
Advisory Letter, concluding that a separated control rod drive shaft funnel
does not significantly increase the probability or consequences of a licensing
basis accident or create any accidents or malfunctions that are not already
addressed by the existing licensing basis analyses.  However, the potential
for foreign objects in the top of the reactor exists for both the plug-weld
and pin-and-cap weld designs for funnel-thermal sleeve attachment pins.  The
former design is used at the Braidwood and Byron plants and the latter design
                                                            IN 94-40, Supp. 1
                                                            December 15, 1994
                                                            Page 4 of 4


is used at the Cook, Diablo Canyon, and Sequoyah plants.  Therefore, jamming
of control rod(s) by loose pins, as happened at Braidwood, may still occur. 
This condition could be detected by rod drop testing, performed as required by
technical specifications, at each startup.  Westinghouse recommended that each
of its nuclear steam supply system customers consider an inspection and
recovery action program for these attachments as part of their next outage
activity.

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.

                                    /s/'d by BDLiaw/for


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

Technical contacts:  S. M. Shaeffer, RII        Edward D. Kendrick, NRR
                     (615) 842-8001             (301) 504-2891

                     Rolf A. Westberg, RIII     Vern Hodge, NRR
                     (708) 829-9732             (301) 504-1861

Attachment:  
List of Recently Issued NRC Information Notices