Information Notice No. 93-88: Status of Motor-Operated Valve Performance Prediction Program by The Electric Power Research Institute
NUCLEAR REGULATORY COMMISSION
OFFICE OF NUCLEAR REACTOR REGULATION
WASHINGTON, D.C. 20555
November 30, 1993
NRC INFORMATION NOTICE 93-88: STATUS OF MOTOR-OPERATED VALVE PERFORMANCE
PREDICTION PROGRAM BY THE ELECTRIC POWER
All holders of operating licenses or construction permits for nuclear power
The U.S. Nuclear Regulatory Commission (NRC) is issuing this information
notice (IN) to alert addressees to preliminary results of motor-operated valve
(MOV) tests conducted by the Electric Power Research Institute (EPRI). 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 do not
constitute NRC requirements; therefore, no specific action or written response
On June 28, 1989, the NRC issued Generic Letter (GL) 89-10, "Safety-Related
Motor-Operated Valve Testing and Surveillance," to request that nuclear power
plant licensees and construction permit holders verify the design-basis
capability of their safety-related MOVs. In GL 89-10, the NRC staff requested
that licensees and permit holders test each MOV within the scope of the
generic letter under design-basis differential pressure and flow conditions,
where practicable. The recommended schedule in GL 89-10 would have licensees
and permit holders verify MOV design-basis capability by June 28, 1994, or
three refueling outages after December 28, 1989 (whichever is later).
In response to concerns regarding MOV performance, EPRI and its utility
advisors established a research program to develop a methodology to predict
the performance of MOVs under design-basis conditions. NUMARC coordinates the
interaction between EPRI, its utility Technical Advisory Group (TAG), and NRC
staff related to the EPRI program. The EPRI program includes detailed
analyses and testing of MOVs at test facilities and nuclear power plants. The
EPRI MOV Performance Prediction Methodology is intended to allow licensees to
demonstrate the design-basis capability of MOVs based on analytical
predictions combined with diagnostic tests conducted under static conditions.
In August and October 1993, EPRI presented the status and preliminary results
from its Flow Loop Testing Program to the NRC staff. The flow loop results in
November 30, 1993
Page 2 of 6
the EPRI presentation have not received full quality assurance verification,
but the preliminary information may be helpful to licensees as they implement
their MOV programs.
In a letter on September 16, 1993, NUMARC provided responses from EPRI to NRC
staff questions on the EPRI MOV Performance Prediction Program. Among the
information provided in the enclosure to the letter, EPRI stated that its
program is expected to cover about 90 percent of gate valves (about half with
its computer code and half with empirically-based data), essentially all globe
valves, and about 95 percent of butterfly valves. The globe and gate valves
are covered primarily by the computer code. EPRI also stated its method for
determination of operator output torque capability under degraded voltage
conditions is to apply standard methods as documented in EPRI NP-6660-D
(March 1990), "Application Guide for Motor-Operated Valves in Nuclear Power
Description of Circumstances
In conducting its MOV Performance Prediction Program, EPRI tested 28 gate,
4 globe, and 2 butterfly valves under a total of 62 test conditions. These
tests were performed at Wyle Laboratories and Siemens test facilities. EPRI
plans to obtain test data for an additional 35 valves being tested in nuclear
power plants. In addition, EPRI completed testing at Kalsi Engineering of
10 butterfly valve designs to assess flow and upstream piping configuration
effects. The results summarized below are based on the Wyle/Siemens MOV
1. Gate Valves
EPRI stated that all gate valves tested were initially preconditioned by
conducting a large number (50-1000) of short (no flow) strokes in cold water
under differential pressure loading. Initial "sliding friction coefficients,"
prior to preconditioning, generally ranged from 0.2 to 0.4. EPRI indicated
that, after preconditioning, "apparent friction coefficients" ranged from 0.3
to 0.6 for all but four valves tested under cold water pumped-flow conditions.
The "apparent friction coefficients" for the remaining four valves ranged from
0.66 to 1.93. EPRI results demonstrated "apparent friction coefficients"
ranging from 0.34 to 0.41 for hot water pumped-flow conditions, 0.35 to 0.8
for hot water blowdown conditions, and 0.25 to 0.64 for steam blowdown
conditions. EPRI's "apparent friction coefficients" reflect all valve
internal phenomena and are not necessarily indicative of a "sliding friction
coefficient." The major difference between the "apparent friction
coefficient" used by EPRI and the "valve factor" used historically by valve
vendors in sizing motor operators is the consideration of the valve disc angle
in determining the EPRI "apparent friction coefficient."
Most valve vendors have used a "valve factor" of 0.3 for flexible wedge gate
valves and 0.2 for parallel disc gate valves in sizing motor operators.
Therefore, the EPRI test results indicate that the thrust required to operate
November 30, 1993
Page 3 of 6
gate valves could be significantly greater than the thrust predicted by the
valve vendors. The EPRI blowdown test results are generally consistent with
those obtained in the limited testing program conducted by the Idaho National
Engineering Laboratory (INEL) for the NRC Office of Nuclear Regulatory
Research in 1989.
EPRI reported that the valve sliding friction coefficient tends to decrease
with increasing differential pressure which lends support for linear
extrapolation of reduced differential pressure results when there is a low
potential for valve damage (for example, under nominal flow velocity
EPRI reported that several gate valves were damaged during hot water and steam
blowdown testing. These included a 6-inch Anchor-Darling valve (disk and seat
damage); a 6-inch Crane valve (guide damage); a 10-inch Velan valve (guide
damage); a 6-inch Walworth valve (guide damage); and a 10-inch Edward valve
(disk and seat damage). Two of the damaged valves exhibited "apparent
friction coefficients" exceeding 0.6.
Two gate valves were damaged under cold water pumped-flow conditions. These
included a Velan 6-inch valve (plastic bending of body guides at high flow
velocity greater than 30 feet per second) and an 18-inch Anchor-Darling valve
(valve disk forced through seating area resulting in leakage above disk).
EPRI test results revealed that it is generally not possible to determine
accurately the point of flow isolation prior to disk wedging based on the
thrust diagnostic trace alone.
EPRI stated that it had not observed differences in thrust requirements for
valve operation between valves installed in horizontal pipes with the stem
either vertical or horizontal. This finding differs from some operating
experiences in nuclear power plants.
EPRI stated that, for incompressible flow conditions, globe valve thrusts are
consistent with industry calculational-method predictions only if the
appropriate area is chosen for differential pressure application. The
appropriate area (disk mean seat area versus disk guide area) appears to be
unique to valve design. It was determined that use of disk mean seat area
rather than disk guide area can result in significant underestimation of
required thrust for some globe valve designs. Specifically, one globe valve
tested under cold water pumped-flow conditions required approximately twice as
much thrust to close using disk mean seat area and a valve factor of 1.0.
A two-inch Rockwell/Edward globe valve, tested under hot water blowdown
conditions, exhibited thrust requirements exceeding predictions based on disk
guide area by approximately 35 percent. This valve sustained damage to the
portion of the body bore that guides the disk.
November 30, 1993
Page 4 of 6
Current industry practice for determining the required thrust for globe valves
varies by manufacturer. Many manufacturers assume disk mean seat area
multiplied by a valve factor in the 1.0 to 1.1 range. Others use disk guide
area in making thrust predictions. Therefore, the EPRI results indicate that
actual thrust requirements may exceed those predicted using current industry
practice for some globe valve designs.
EPRI stated that the Wyle flow loop testing revealed torque requirements to
operate Pratt butterfly valves which were bounded by the most current torque
predictions of the manufacturer. However, butterfly valves at some nuclear
power plants (for example, Catawba and Palo Verde) have demonstrated torque
requirements that exceed vendor predictions. EPRI is currently evaluating
data from testing conducted at Kalsi Engineering to assess butterfly valve
torque requirements for a wide range of butterfly valve designs.
4.Data Interpretation and Assessment
In July 1993, EPRI sent a Quarterly Status Report to all utilities
participating in the EPRI MOV Performance Prediction Program. This report
summarized preliminary flow loop test results. After the completion of
Wyle/Siemens quality assurance checks, EPRI plans to update this information
in its next Quarterly Status report scheduled for late 1993. Detailed test
reports documenting these results are scheduled for delivery to participating
utilities between October and December 1993. EPRI stated that, in
interpreting the EPRI flow loop test results, utilities need to understand the
assumptions and equations that were used by EPRI in presenting the data. For
example, the EPRI calculated "apparent friction coefficient" for gate valves
is based on the equation provided in EPRI Report NP-6660-D, "Application Guide
for Motor-Operated Valves in Nuclear Power Plants." This equation is solved
for "apparent friction coefficient" using (1) the maximum measured stem thrust
which occurs prior to the initiation of wedging (for valve closing) or the
maximum thrust which occurs after cracking (for valve opening); (2) full
(valve closed) tested differential pressure; (3) mean seat area; (4) valve
disk angle; (5) full (valve closed) upstream tested pressure for stem
rejection thrust; and (6) measured values of packing load.
EPRI stated that valve design and test conditions, maintenance history, and
operating experience may be important in assessing the applicability of EPRI
test results to plant MOVs.
EPRI uses the greatest thrust requirement to overcome differential pressure
and flow to determine its "apparent friction coefficient." EPRI assumes the
highest differential pressure observed during the test regardless of the stem
position where the greatest differential pressure/flow required thrust occurs.
This results in a lower calculated friction coefficient than would be
determined if the actual differential pressure at the point of greatest thrust
was used in determining the friction coefficient.
November 30, 1993
Page 5 of 6
EPRI plans to submit sections of a topical report for NRC review as they are
completed between November 1993 and April 1994. EPRI intends to submit
supporting reports in advance of the final topical report to allow the staff
to raise questions with EPRI early in the review process.
EPRI stated that its final methodology is scheduled for delivery to utilities
in April 1994 as a tool that may be used to confirm many aspects of MOV
calculations and setup. Further, the EPRI flow loop test results provide
licensees with information which might be helpful in supplementing other "best
available" data in establishing MOV switch settings.
Since EPRI initiated its MOV Performance Prediction Program, the NRC staff has
conducted public meetings with NUMARC and EPRI to discuss the goals of the
EPRI program, the development of the program activities to accomplish those
goals, the tests conducted in support of the program and the results of those
tests, and the completion schedule for the program. The staff has provided
questions and comments to NUMARC and EPRI on the EPRI MOV program as a result
of these meetings. For example, in a public meeting on October 6-7, 1993, the
staff emphasized the need for EPRI to ensure that licensees clearly understand
the application of the EPRI test data and methodology. Also at this meeting,
contents of this notice were discussed and the comments from EPRI have been
considered. The staff expressed concern about the valves damaged during the
EPRI testing and the apparent lack of action by some valve manufacturers in
response to the valve damage. The staff also discussed the need for EPRI to
ensure that adequate peer review of the EPRI methodology is conducted.
Although some issues remain to be resolved, the EPRI testing program should
provide a significant amount of MOV test data that can assist nuclear power
plant licensees in demonstrating the design-basis capability of MOVs that
cannot be tested under dynamic conditions as installed. The preliminary test
information provided in this notice is provided for licensee consideration in
implementing programs in response to GL 89-10 . The staff plans to conduct
additional public meetings with NUMARC and EPRI to discuss the status of the
EPRI MOV program. The staff will consider the need for additional generic
communications to nuclear power plant licensees and construction permit
holders as additional information is obtained from the EPRI MOV program.
Related Generic Communications
NRC Information Notice 90-40, "Results of NRC-Sponsored Testing of
November 30, 1993
Page 6 of 6
This information notice requires no specific action or written response. If
you have any questions about the information in this notice, please contact
the person listed below or the appropriate Office of Nuclear Reactor
Regulation (NRR) project manager.
/s/'d by BKGrimes
Brian K. Grimes, Director
Division of Operating Reactor Support
Office of Nuclear Reactor Regulation
Technical contact: Thomas G. Scarbrough, NRR
List of Recently Issued NRC Information Notices
Page Last Reviewed/Updated Friday, May 22, 2015