United States Nuclear Regulatory Commission - Protecting People and the Environment

Information Notice No. 94-18: Accuracy of Motor-Operated Valve Diagnostic Equipment (Responses to Supplement 5 to Generic Letter 89-10)

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

March 16, 1994


NRC INFORMATION NOTICE 94-18:  ACCURACY OF MOTOR-OPERATED VALVE DIAGNOSTIC
                               EQUIPMENT (RESPONSES TO SUPPLEMENT 5 TO
                               GENERIC LETTER 89-10)


Addressees

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

Purpose

This information notice is intended to alert addressees to information
submitted by various nuclear power plant licensees related to the accuracy of
motor-operated valve (MOV) diagnostic equipment.  The information was
submitted in response to Supplement 5, "Inaccuracy of Motor-Operated Valve
Diagnostic Equipment," to Generic Letter (GL) 89-10, "Safety-Related
Motor-Operated Valve Testing and Surveillance."  It is expected that
recipients will review the information for applicability to their facilities
and consider actions, as appropriate, related to operability of MOVs.
However, suggestions contained in this information notice do not constitute
NRC requirements; therefore, no specific action or written response is
required. Background

In GL 89-10 (June 28, 1989), the NRC staff asked holders of operating licenses
and construction permits for nuclear power plants to provide additional
assurance of the capability of safety-related MOVs and certain other MOVs in
safety-related systems to perform their intended functions by reviewing MOV
design bases, verifying MOV switch settings initially and periodically,
testing MOVs under design-basis conditions where practicable, improving
evaluations of MOV failures and necessary corrective action and trending MOV
problems.  The NRC staff issued several supplements to GL 89-10 to clarify or
modify its recommendations.

As an integral part of most GL 89-10 programs, licensees are relying on MOV
diagnostic equipment to provide information on the thrust required to open or
close the valve, as well as the thrust delivered by the motor actuator.  The
various types of MOV diagnostic equipment estimate stem thrust using different
methods, such as spring pack displacement or strain in the stem, the mounting
bolts, or the yoke.  Because some licensees make decisions regarding the
operability of safety-related MOVs on the bases of diagnostic equipment




9403100159.                                                            IN
94-18                                                             March 16,
1994                                                             Page 2 of 5


thrust readings, the accuracy of MOV diagnostic equipment can have a
significant effect on the safe operation of a nuclear power plant.

During the implementation of GL 89-10, the NRC staff became aware of new
information on the accuracy of MOV diagnostic equipment.  This new information
raised a generic concern regarding the reliability of the data produced by MOV
diagnostic equipment.  For example, on February 3, 1992, the MOV Users Group
(MUG) of nuclear power plant licensees released "Final Report - MUG Validation
Testing as Performed at Idaho National Engineering Laboratories" (Volume 1).
The report stated that the MOV diagnostic equipment that relied on spring pack
displacement to estimate stem thrust was not as accurate as its vendors
claimed.  In addition, the NRC staff learned that specific MOV diagnostic
equipment that relies on valve yoke strain to estimate stem thrust was less
accurate than had been previously reported.

On March 2, 1992, the NRC staff held a public meeting with representatives of
ITI-MOVATS to discuss the accuracy of the ITI-MOVATS thrust measuring device
(TMD) to estimate stem thrust on the basis of spring pack displacement.  At
this meeting, the representatives of ITI-MOVATS described the results of their
field validation program which showed that, in some instances, the TMD may be
less accurate than licensees had assumed.  The ITI-MOVATS representatives also
discussed the results of their activities to resolve concerns about the fact
that the TMD is calibrated in the valve opening direction, although it also is
used to predict the thrust delivered by the actuator in the valve closing
direction.  ITI-MOVATS prepared Engineering Report 5.2 (March 13, 1992) to
provide guidance to its licensee customers for evaluating the capability of an
MOV to perform its safety function under design-basis conditions in light of
the decreased accuracy of the TMD.  The Nuclear Management and Resources
Council (NUMARC) developed guidelines for licensees to use in evaluating MOVs
that had been set up using the TMD.

ASEA-Brown Boveri (ABB) Impell manufactures MOV diagnostic equipment (known as
OATIS) that relies on spring pack displacement to estimate stem thrust.
Following the release of the MUG report, Impell representatives stated that
they would be working with their licensee customers to develop new accuracy
values.

Liberty Technologies has manufactured MOV diagnostic equipment, referred to as
valve operation test and evaluation system (VOTES), that estimates the thrust
required to open or close a valve based on the strain of the valve yoke.
The VOTES system does this by measuring the diametral strain (the change in
diameter) of the valve stem for a given yoke strain.  The stem thrust is then
calculated using the diametral strain and the nominal engineering properties
of the stem.  This stem thrust is compared to the yoke strain to provide a
correlation between the yoke strain and the stem thrust.  Once this
correlation has been established, the stem thrust can be more easily
determined by measuring the yoke strain.  On October 2, 1992, Liberty
Technologies notified the NRC, in accordance with 10 CFR Part 21, that it had
found two new factors that can affect the thrust values obtained with its
VOTES equipment.  Those factors are (1) the possible use of improper stem   .
IN 94-18                                                             March
16, 1994                                                             Page 3
of 5


material constants and (2) the failure to account for a torque effect when the
equipment is calibrated by measuring the strain of the threaded portion of a
valve stem.

In its October 2 submittal, Liberty Technologies provided guidance to
licensees for correcting the thrust data by performing hand calculations, and
stated that the new Version 2.3 of its VOTES software would be of help in
performing the corrections.

On June 28, 1993, the NRC staff issued Supplement 5 to GL 89-10 requesting
licensees and construction permit holders (1) to re-examine their MOV programs
and to identify measures taken to account for uncertainties in properly
setting valve operating thrust to ensure operability and (2) to evaluate the
schedule necessary to first consider the new information on MOV diagnostic
equipment inaccuracy and then to take appropriate action in response to that
information.  Within 90 days of receipt of Supplement 5 to GL 89-10, licensees
were required (1) to notify the NRC staff of the diagnostic equipment used to
confirm the proper size, or to establish settings, for safety-related MOVs,
and (2) to report whether they had taken actions or planned to take actions
(including schedule) to address the new information on the accuracy of MOV
diagnostic equipment.

Description of Circumstances

The NRC staff has reviewed the responses to Supplement 5 to GL 89-10 submitted
by licensees and construction permit holders.  The staff has found that, for
the most part, licensees and permit holders have been actively addressing the
uncertainties regarding the accuracy of MOV diagnostic equipment.  The newly
recognized reduced accuracy of MOV diagnostic equipment can raise questions
regarding (1) the adequacy of torque switch settings to provide sufficient
thrust while not exceeding thrust or torque structural limits and (2) the
capability of actuator motors with the present settings.  In their responses,
licensees and permit holders indicated that many MOVs had the potential for
underthrusting or overthrusting as a result of the less-than-expected accuracy
of MOV diagnostic equipment.  Consequently, some licensees reported that MOVs
have been retested, adjusted, or modified to resolve the concerns regarding
the accuracy of MOV diagnostic equipment.  The staff will discuss specific
aspects of the response to Supplement 5 to GL 89-10 with individual licensees
during future inspections.

The following is a summary of the issues relating to the accuracy of MOV
diagnostic equipment as reported by licensees in their responses to
Supplement 5 to GL 89-10.

1.ITI-MOVATS Engineering Report 5.2 discussed the calibration of the
ITI-MOVATS TMD in the open direction with reliance on the calibration to
measure thrust in the close direction.  This ITI-MOVATS report focuses on
accuracy corrections for the TMD under static test conditions.  The NRC staff
knows of no ITI-MOVATS guidance related to accuracy corrections for the use
of the TMD under dynamic test conditions.

.                                                             IN 94-18
                                                            March 16, 1994
                                                            Page 4 of 5


2.The temporary installation of the ITI-MOVATS torque thrust cell (TTC) for
diagnostic testing can affect the actuator output thrust.  ITI-MOVATS
Special Test Report 6.0 addresses the actuator repositioning effect.  The
licensee of the Oyster Creek Nuclear Power Plant stated in its response to
Supplement 5 to GL 89-10 that this special test report indicates that
ITI-MOVATS recommends that actuator repeatability values and thrust/torque
measurement error values be revised when using a temporarily installed TTC.

3.The torque effects and the material characteristics that are relevant to
the use of the Liberty Technologies' VOTES diagnostic equipment is
discussed in this company's Part 21 notification of October 2, 1992.

4.In addition to the issues involving torque effects and material
characteristics, Liberty stated in its Part 21 notification of October 2,
1992, that (a) incorrect thrust readings can arise with the use of long
cable lengths between the signal conditioning box and the breakout box
when calibrating a VOTES sensor with a U-clamp or D-clamp; (b) under
certain conditions, one or more of the four operational amplifiers in the
breakout box oscillates and causes a thrust indication which can be lower
or higher than actual by a factor of about two or four; and (c) CB 23-100
cables were mistakenly shipped without the offset resistor (which results
in a lower-than-actual torque indication from the VTC load cell).

5.The licensee of the Cooper Nuclear Power Plant reported in its response to
Supplement 5 to GL 89-10 that Liberty Technologies is evaluating the issue
of calibration of its equipment in one direction and reliance on thrust
measurements in the other direction.

6.The Cooper licensee reported that Liberty Customer Service Bulletin
    CSB-030 (May 6, 1993), "Proximity Probe Type Calibrators With a Possible
3% Shift in Sensitivity," alerts VOTES users to possible changes in the
sensitivity of the proximity probe-type calibrator that can overpredict
thrust readings.

7.The licensee of the Susquehanna Nuclear Power Station reported in its
response to Supplement 5 to GL 89-10 that factors that can affect VOTES
accuracy are (a) the change in stem transition areas resulting from
Liberty Technologies' refinement of its finite element model, (b) changes
in effective stem diameters, (c) the need to differentiate between General
Purpose ACME and Stub ACME threads when determining torque correction
factors and effective stem diameters, and (d) new accuracy values based on
torque correction values and percent extrapolation beyond calibration
ranges.

8.The licensee of the Maine Yankee Nuclear Power Plant noted in its response
to Supplement 5 to GL 89-10 that it had submitted a notice in accordance
with 10 CFR Part 21 on July 21, 1993, regarding its determination that the
accuracy cited by Liberty Technologies for its VOTES equipment is only
    .                                                            IN 94-18
                                                            March 16, 1994
                                                            Page 5 of 5


appropriate for torque switch trip when the Best-Fit-Straight-Line (BFSL)
calibration method is used.  As a result of this question regarding the
accuracy of the VOTES calibration method, the Susquehanna licensee
reported that its diagnostic tests will need to be repeated using the BFSL
calibration method.  Liberty Technologies is in the process of issuing a
customer service bulletin that will give guidance on the inaccuracies of
the results obtained by using the BFSL method.  The service bulletin will
also include a summary of a statistical analysis that was done to verify
the accuracy of the BFSL method.

Related Generic Communications

The NRC has issued other generic communications on the accuracy of MOV
diagnostic equipment.  For example, the NRC issued Information Notice (IN)
92-23, "Results of Validation Testing of Motor-Operated Valve Diagnostic
Equipment," to alert licensees to the potential decreased accuracy of MOV
diagnostic equipment manufactured by ITI-MOVATS and ABB Impell; and IN 93-01,
"Accuracy of Motor-Operated Valve Diagnostic Equipment Manufactured by Liberty
Technologies."

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.

/s/'d by BKGrimes


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

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

                     Allen G. Hansen, NRR
                     (301) 504-1390

Attachment:
List of Recently Issued NRC Information Notices.
Page Last Reviewed/Updated Friday, November 15, 2013