Information Notice No. 93-02: Malfunction of a Pressurizer Code Safety Valve

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

                                January 4, 1993


NRC INFORMATION NOTICE 93-02:  MALFUNCTION OF A PRESSURIZER CODE SAFETY VALVE


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 failure of a pressurizer code safety valve
to maintain set pressure and reseat properly during a plant transient.  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 3, 1992, after an electrical transient, the reactor at Fort Calhoun
Station tripped on high primary pressure.  Both power-operated relief valves
opened and valve RC-142, one of two pressurizer code safety valves, lifted
prematurely at a pressure below 16.75 MPa [2430 psia], as opposed to the
proper setpoint pressure of 17.24 Mpa [2500 psia] +/- 1 percent.  The relief
valves shut automatically when the reactor coolant system pressure decreased
to 16.20 Mpa [2350 psia].  Because a safety valve was still open, the pressure
continued to decrease and RC-142 subsequently reseated at approximately 
12.03 MPa [1745 psia].  The pressure then increased and RC-142 lifted again at
approximately 13.27 MPa [1925 psia].  RC-142 partially reseated, as pressure
again dropped, at approximately 6.89 MPa [1000 psia].  RC-142 continued to
leak, as indicated by the tail pipe temperature, until the plant was brought
to cold shutdown.  The licensee removed RC-142 and sent it to Wyle Laboratory
(Wyle) for inspection and testing.

Discussion

The code safety valves installed at Fort Calhoun are "3-inch inlet by 6-inch
outlet", Size 3K6, Style HB-86-BP, Type E valves (Figure 1) manufactured by
the Crosby Valve and Gage Company (Crosby).  In early 1980, the Electric Power
and Research Institute (EPRI) tested Crosby safety valves that have loop seals
and are subjected to back pressure.  The EPRI test results indicated that the
initial discharge of the loop seal or a transition from discharging steam to
discharging water could cause the valve to chatter.


9212280132.

                                                            IN 93-02
                                                            January 4, 1993
                                                            Page 2 of 4


The licensee believes that RC-142 chattered during its initial lift from the
discharge of the loop seal.  Apparently, the chatter loosened the locknut on
the adjusting bolt and allowed the adjusting bolt to partially back out. 
Later, primary water discharged through RC-142 for approximately 5 minutes
(pressurizer level reached 100 percent) during its second lift and subsequent
partial reseat at approximately 6.9 MPa [1000 psia].  The discharging water
induced further chattering, apparently causing the adjusting bolt to back out
even further, reducing the valve lift setpoint to approximately 
10.18 MPa [1477 psia].   

To ensure that the adjusting bolt would not back out again, Crosby designed a
special mechanical locking device and installed it on the two valves at 
Fort Calhoun.  Crosby also specified a torque value of "400 foot-pounds" for
the adjusting bolt locknut.  This value had not been previously specified in
procedures used by Wyle for inspecting and testing pressurizer code safety
valves.

An NRC augmented inspection team monitored licensee activities at Fort Calhoun
and Wyle Laboratory.  At Wyle, the locknut for RC-142 was found to have backed
off from the top of the valve bonnet by approximately 3 to 6 mm [1/8 to 
1/4 inch] and could be turned by hand.  The adjusting bolt was determined to
be 19.5 flats of bolt revolution from the zero compression position of the
spring.  Crosby representatives calculated that this position corresponded to
a setpoint value of approximately 10.18 MPa [1477 psia].  In March 1992, the
valve had been set to 17.24 MPa [2500 psia] +/- 1 percent at Wyle Laboratory.

When the valve internals were removed, the bellows assembly was found to have
failed on each end at the first weld after the transition weld.  Also, the
disc insert was found jammed into the disc holder.  The disc insert was
recessed approximately 0.05 mm [0.002 inch] below the top surface of the disc
ring.  The disc ring was seated on the nozzle ring, which indicated that the
valve had not reseated properly.  NRC Inspection Report 50-285/92-18 contains
additional information on this event.

On August 22, 1992, an additional problem was revealed when RC-142 again
lifted prematurely.  This premature lift occurred as the reactor coolant
system pressure increased to approximately 16.53 MPa [2397 psia].  This
pressure was approximately 4 percent below the normal setpoint of 
17.24 MPa [2500 psia] +/- 1 percent.  However, the valve reseated properly
with no leakage detected before or after the valve lift.

Because of the premature lift, both valves were returned to Wyle Laboratory
for additional inspection and testing.  Wyle inspected RC-142 and found the
valve to be in good condition with only minor nicks on the nozzle seat.  When
Wyle attempted to test RC-142, certain conditions and difficulties were noted.

�     RC-142 has a stainless steel nozzle and a carbon steel body.

�     The initial attempt to test the valve under "cold" conditions to
      simulate the normal operating conditions of the plant could not be 
.

                                                            IN 93-02
                                                            January 4, 1993
                                                            Page 3 of 4


      performed as planned because of increases in the valve internal and
      external temperatures before and after the lift of the valve.  

However, the testing revealed that the lift setpoint would increase with
increasing nozzle temperature and then decrease as the valve body temperature
began to increase.  Licensee personnel concluded that the valves would require
testing at "hot" upper bound temperatures with saturated inlet steam near
setpoint pressure and with the valves insulated with the actual plant
insulation.  The insulation used by Wyle differed somewhat in composition and
fit from the plant insulation and apparently affected the distribution of heat
within the valve.  The data, obtained from the tests performed with saturated
steam, confirmed that the setpoint value initially increased due to the
thermal expansion of the valve nozzle, then decreased as the temperature of
the valve body increased.  The setpoint value stabilized once the valve
temperature stabilized.  This condition was also observed during the "in-situ"
Trevitest testing performed at the plant.  As a result of this testing, the
licensee performed a safety analysis and determined that the reactor coolant
system could withstand an overpressure transient with a code safety valve
pressure setpoint deviation of +6 percent (1.03 MPa [150 psi]) and that a
pressure setpoint deviation of -4 percent (0.69 MPa [100 psi]) would not cause
unnecessary challenges to the safety valves.  The licensee also reduced the
power-operated relief valve and reactor high pressure trip setpoint by 
0.35MPa [50 psi].  The licensee is considering the removal of the loop seal as
a long-term action.  

After completing the testing, Wyle reset both code safety valves to the
technical specification setpoint of "2500 psia +/- 1 percent."   The valves
were then returned to Fort Calhoun and were reinstalled.

The licensee believes that the premature lift of RC-142 on August 22, 1992,
resulted from using Wyle insulation during testing rather than the actual
plant insulation.  The tests using Wyle insulation resulted in a lower
stabilized valve temperature than resulted from using the actual plant
insulation.  The difference in the test methods resulted in an approximate
3 percent difference in setpoint.  NRC Inspection Report 50-285/92-21 contains
additional information on this event.

.

                                                            IN 93-02
                                                            January 4, 1993
                                                            Page 4 of 4


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. 


                                        ORIGINAL SIGNED BY


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

Technical contacts:  T. F. Westerman, Region IV
                     (817) 860-8145

                     P. A. Goldberg, Region IV
                     (817) 860-8168

Attachments:  
1.  Figure 1, Fort Calhoun Pressurizer Safety Valve 
2.  List of Recently Issued NRC Information Notices
.