Information Notice No. 94-69: Potential Inadequacies in the Prediction of Torque Requirements for and Torque Output of Motor-Operated Butterfly Valves

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

September 28, 1994


NRC INFORMATION NOTICE 94-69:  POTENTIAL INADEQUACIES IN THE PREDICTION OF
 TORQUE REQUIREMENTS FOR AND TORQUE OUTPUT OF      MOTOR-OPERATED BUTTERFLY
VALVES


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 potential inadequacies in the prediction of
torque requirements for and torque output of motor-operated butterfly valves.
It is expected that recipients will review the information for applicability
to their facilities and consider actions, as appropriate, to avoid similar
events or problems.  However, suggestions contained in this information notice
are not NRC requirements; therefore, no specific action or written response is
required.

Description of Circumstances

Several problems regarding inadequacies in the prediction of torque
requirements to operate butterfly valves and available torque from their motor
actuators have been reported through Generic Letter (GL) 89-10 reviews.
A significant example pertains to an event that occurred at the Catawba
nuclear power plant.

On February 25, 1993, the licensee for the Catawba facility determined that
both loops of the service water systems had been inoperable since August 1992,
because three of the four service water pump discharge valves could not open
against the pump discharge pressure.  In August 1992, the licensee had reduced
the torque switch settings on two of the four service water system discharge
valves (one in each train) to address a concern that, under minimum voltage
conditions, the motor operators for these valves might stall at a torque value
lower than the maximum torque setpoint.  The new torque switch settings were
measured on a test bench and judged to be adequate based on manufacturer
sizing equations.  Following this modification of the two valves, differential
pressure testing was not conducted to ensure the valves would function under
design-basis conditions.






9409210211.                                        IN 94-69
                                        September 28, 1994
                                        Page 2 of 4


The licensee determined that these valves failed to open at the established
torque output because the seat material had hardened with age and bearings had
degraded with raw water service.  Such other variables as gearbox efficiency,
application-specific flow characteristics, and packing friction also affect
the torque required to operate a butterfly valve.  This event indicated that
the engineering analysis performed to derive the torque requirements of the
two referenced valves did not adequately address age degradation of the
valves.

Some significant problems at other nuclear power plants relating to torque
requirements of butterfly valves and available torque from their motor
actuators are discussed below:

-In March 1994, the licensee for the Perry facility found that certain
butterfly valves required higher seating torque than expected.  Dynamic
testing showed that the bearing friction coefficient for certain
butterfly valves with stainless steel bushings was higher than the value
originally assumed (0.25) in the calculations.  Specifically, the
bearing coefficient of friction was 0.634 for a fuel pool cooling and
cleanup valve with stainless steel bushings, and 0.334 for an emergency
service water valve with graphite-lubricated stainless steel bushings.
The licensee investigated the effect of this information on 17 motor-operated
valves, the entire plant population of motor-operated butterfly valves with
stainless steel bushings.  No operability problems were found.  However, the
licensee is evaluating possible modifications to improve margins for the four
butterfly valves in the fuel pool cooling and cleanup system.

-In November 1993, the licensee for the Palo Verde facility determined
that the condensate storage tank isolation valve would not have opened
or closed completely under design-basis conditions with the as-found
torque switch settings.  The safety function of the valve is to isolate
the condensate storage tank should non-seismic piping rupture
downstream.  The licensee also found that a containment isolation valve
would not have closed under design-basis conditions with the as-found
torque switch settings because the measured closing torque at torque
switch trip was less than the minimum torque required in accordance with
the design-basis calculation.

-In March 1993, the licensee for the Comanche Peak facility reported that
a valve in the containment hydrogen purge system had demonstrated a
lower torque output at torque switch trip during loaded testing on a
torque test stand than that delivered at torque switch trip during
static insitu testing.  The licensee demonstrated that no operability
concern existed for this valve.  However, this apparent reduction in
torque output at torque switch trip under loaded conditions for a
motor-operated butterfly valve could have implications for other
motor-operated valves.


.                                        IN 94-69
                                        September 28, 1994
                                        Page 3 of 4


-In May 1992, the licensee for the Zion facility determined that 12 of
the 15 butterfly valves in the service water system were inoperable
because their torque switch settings did not support the original design
basis maximum differential pressure of 862 kPa [125 psid].  The
remaining three valves would have supported a maximum differential
pressure of 862 kPa [125 psid]; however, it would not have supported the
current design requirement of 971 kPa [141 psid].  This event was caused
by deficiencies in the pre-service design installation testing.

-In March 1992, the licensee for the Grand Gulf facility determined that
the actuators for the suppression pool makeup system dump valves were
undersized (they could not meet the torque requirements).  It was found
that the actuators were undersized because the flow rate specified in
the design specification for the valves, which was used by the valve
manufacturer to size the actuators, was the minimum required flow rate
to ensure a "post-loss of coolant accident" drywell vent coverage of at
least two feet above the top row vents with maximum emergency core
cooling systems pump-down of the suppression pool.  The size of the
valves should have been determined using maximum expected design-basis
flow rate based on the actual configuration of the system.

-In March 1991, while performing diagnostic tests on a butterfly valve of
the primary containment purge and exhaust system, the licensee for the
Limerick facility discovered that the "as-found" torque switch setting
for the valve motor actuator was too low.  This would have resulted in
the motor actuator tripping on high torque before the valve could
adequately close and seat if this valve was required to close during a
design-basis differential pressure condition.  In addition, the testing
indicated that the actual motor actuator torque output for this valve
was lower than the design torque output requirement.  As a result, the
valve would not have been capable of performing its primary containment
isolation design function.  When the licensee reviewed the original
startup test records for all motor-operated butterfly valves in the same
system, it found eight other valves that did not meet the design torque
specifications for adequate valve seating in the event of a design basis
differential pressure condition, or for which no test data were
available to support the adequacy of butterfly valve testing.

Discussion

These examples of recent problems illustrate that many variables affecting the
application of butterfly valves, including flow, pressure, seat material
degradation, gearbox efficiency, turbulence in the system, as well as packing
and bearing frictions, can have an impact on the required torque to operate a
butterfly valve and on the torque output of their motor actuators.  A primary
basis for the NRC issuance of Generic Letter (GL) 89-10, "Safety-Related
Motor-Operated Valve Testing and Surveillance," with its request to test
safety-related motor-operated valves under design-basis conditions where


.                                        IN 94-69
                                        September 28, 1994
                                        Page 4 of 4


practicable, is the uncertainties surrounding the prediction of torque/thrust
requirements by valve vendors.

Related Generic Communications

NRC Information Notice 93-88, "Status of Motor-Operated Valve
Performance Prediction Program by the Electric Power Research
Institute," November 30, 1993

NRC Information Notice 90-21, "Potential Failure of Motor-Operated
Butterfly Valves To Operate Because Valve Seat Friction Was
Underestimated," March 22, 1990

NRC Information Notice 84-04, "Failure of Elastomer Seated Butterfly
Valves Used Only During Cold Shutdowns," January 18, 1984

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

Technical contacts:  Thomas G. Scarbrough, NRR
               (301) 504-2794

               William T. Orders, RII
               (803) 383-4571

               Perry C. Hopkins, RII
               (803) 831-2963

               Christine Lipa, RIII
               (319) 851-5111

Attachment:
List of Recently Issued NRC Information Notices