Information Notice No. 95-08: Inaccurate Data Obtained with Clamp-On Ultrasonic Flow Measurement Instruments

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

                               January 30, 1995


NRC INFORMATION NOTICE 95-08:  INACCURATE DATA OBTAINED WITH CLAMP-ON
                               ULTRASONIC FLOW MEASUREMENT INSTRUMENTS

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 difficulties in accurately measuring flow rates
with clamp-on ultrasonic flow measurement instruments.  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 October 14, 1993, while performing a high head safety injection (HHSI) flow
balance test, the licensee for North Anna Unit 2 determined that the as-found
safety injection cold leg branch line flow rates did not meet the technical
specification requirements.  The as-found cold leg branch line flow rates had
also failed to meet the technical specification requirements during the
previous test in April 1992.  

The North Anna Technical Specifications give limits related to both minimum
and maximum safety injection line flow rates.  For North Anna, these two
technical specification limits resulted in an very narrow allowed operating
band, which required the use of highly accurate flow measurement instruments
for flow tests.

The licensee performed the flow balance using System 990 Uniflow Universal
Transit-Time Flowmeters manufactured by Controlotron of Hauppauge, New York. 
The Uniflow flowmeter is a clamp-on non-intrusive portable flowmeter that uses
multi-pulse, transit-time ultrasonic technology to measure precisely the flow
rate.  The flowmeter induces alternate upstream and downstream ultrasonic
signals and measures the difference in transit times.  This information is
then used to determine the axial fluid velocity and the flow rate through the
pipe.  Inaccuracies associated with the use of these instruments at North Anna
combined with the narrow band of allowable flows caused problems which
contributed to the failure to meet technical specification requirements.




9501240226.                                                            IN 95-08
                                                            January 30, 1995
                                                            Page 2 of 4


Discussion

The licensee examined the 1993 North Anna Unit 2 flow balance test data and
observed that some of the data indicated that the instrument errors
significantly exceeded accuracies assumed for the tests.  Numerous factors may
affect the accuracy of this flowmeter including pipe dimensions, fluid flow
effects, transducer mounting, and instrument setup and operation.  After the
test, the licensee investigated and concluded that the instrument inaccuracy
most likely was caused by flow effects associated with the piping
configuration.  The piping arrangement, which contained several out-of-plane
bends that forced the flow to change direction numerous times before entering
the straight section of pipe containing the ultrasonic transducers, may have
caused a stable swirl-type flow profile corkscrewing down the pipe.  The
licensee postulated that the swirl flow added a radial flow error to the test
instruments.  The flowmeter was calibrated assuming a primarily axial flow
profile.  In subsequent testing at a national laboratory, Controlotron
ultrasonic flowmeter data varied up to 5% at different orientations around the
pipe at a position approximately 15 pipe diameters downstream of a single
bend.  The test configuration did not include any out-of-plane bends,
indicating that it is also possible for a non-uniform flow profile to exist in
the absence of out-of-plane bends, causing flow measurements with larger than
expected errors. 

The licensee noted that the instrument output during the test did not give any
indication of measurement problems.  The instruments displayed little noise
and only the expected small-magnitude random flow fluctuations.  The lack of
erratic instrument readings could easily be interpreted as meaning the
readings were valid (i.e., within the stated accuracy).  However, the licensee
tested the instrument and determined that it could register a stable reading
while a significant error was occurring.  Therefore, the licensee concluded
that repeatability of the data does not prove that the flow has been measured
accurately.  

After the test failure, the licensee found several inconsistencies in the test
data which indicated problems in data accuracy.  As an example, when the
A HHSI pump replaced the C HHSI pump as the operating pump, without any other
changes in system resistance, the B cold leg flow rate increased by 
8.4 percent while the A cold leg flow rate increased by 1.5 percent and the 
C cold leg flow rate did not change.  The licensee concluded that these values
corresponded to an implausible change in the highest cold leg flow path from
loop A to loop B when the sole change was which injection pump was operating. 
In addition, the head curve for the A HHSI pump was lower than that for the 
.                                                            IN 95-08
                                                            January 30, 1995
                                                            Page 3 of 4


C HHSI pump; thus, the A HHSI pump would be expected to supply flow rates less
than the C HHSI pump for the same system resistance.  However, the ultrasonic
instruments indicated that the A HHSI pump delivered 3.3 percent more flow
than the C HHSI pump.  This was contrary to the previously measured pump head
curve data.  At the same time, installed plant process instruments indicated
that the A HHSI pump delivered 3.4 percent less flow than the C HHSI pump, a
measurement consistent with the pump curves.

The licensee could not determine with certainty the factors that resulted in
the unanticipated measurement errors in this event.  Conditions specific to an
application can cause deviations from the vendor-supplied instrument
accuracies.  However, during the investigation the licensee developed methods
that could be used to determine when the ultrasonic flowmeter may not yield
the assumed accuracy.  These methods include comparing ultrasonic flowmeter
data from several locations or different orientations at the same location or
by comparing the data from these instruments with data from permanently
installed instruments.  Test results can also be compared with expected system
behavior.

The NRC staff has approved a request from the licensee of the North Anna plant
to modify the technical specification that specifies the acceptance criteria
for the HHSI pump flow rates.  The change removes the specific fixed values
for the flow from the technical specifications and permits the licensee to
establish allowable values by analysis, providing greater operational
flexibility.  The licensee is also planning to (1) install permanent flow
instrumentation that will supply highly accurate data for future flow balance
testing and (2) supply training concerning the accuracy of clamp-on ultrasonic
flowmeters to the personnel who use them.

Licensees frequently use flow measurements to demonstrate that safety-related
component performance is adequate to meet the demands indicated by the
accident or transient analyses.  Therefore, obtaining accurate flow
measurements is important to safety.  When balancing the flow of HHSI
injection paths, flow rates need to be accurately measured to demonstrate that
the unit complies with the assumptions in safety analyses and is operated in
an analyzed condition.  At North Anna the licensee initially failed to detect
inaccurate flow data because of a failure to verify the consistency,
repeatability and reasonableness of the data from the ultrasonic flow
measurement instruments.  Although technical specification requirements were
not met at North Anna, the licensee later analyzed the issue and concluded
that the HHSI system would have performed its design safety function.
.                                                            IN 95-08
                                                            January 30, 1995
                                                            Page 4 of 4


This information notice requires no specific action or written response.  If
you have any question 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 BKGRIMES


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

Technical contacts:  Donald R. Taylor, RII       Larry W. Garner, RII
                     (703) 894-5421              (404) 331-4663

                     Hukam C. Garg, NRR
                     (301) 504-2929

Attachment:  List of Recently Issued Information Notices