Page 89 - Water Loss Control

Validation of Sour ce Meter Accuracy 71

To be sure that meters are accurate, compare the results of meter tests to applicable
AWWA standards and guidance manuals. If a meter measures incorrectly and the error
exceeds the standard for its category, repair and recalibrate the meter to function within
standard limits. If the meter has not been tested within the past 12 months, test the
meter.
If source meters are inaccurate, inspect each one in the field. Normal wear is not the
only cause of inaccurate meter readings. Check to be sure that the meter is the right type
and size for the application and that it is installed correctly. See the AWWA M33 publi-
2
cation for guidance on typical source meter types and applications. Check the size
against manufacturers’ recommended ranges. Be sure that the meter is level; most
meters are not designed for sloped or vertical operation. Inspect the meter to see if hard-
water encrustation is interfering with the measurement. Also check to verify that the
proper registers were selected and installed correctly. Finally, be sure that the register is
read correctly or the signal from the meter is properly transmitted through the SCADA
System. Have an employee familiar with metering instrumentation perform calibration
of the instrument and make a special reading of the source meter, or have a second
employee accompany the meter reader to verify sample readings. Check to be sure that
the meter is read and recorded correctly, and the correct conversion factor is used.
Check venturi meters for blockages in the throats of the meters or in the sensing
lines. Test the primary device by comparing it with a measurement taken from a pitot
rod or other insertion-type meter installed in series with the meter. Testing the meter
with a pitot rod shows whether or not the installation is adequate for nonturbulent
flows. The meter’s primary device should be tested at different flow ranges. If pressure
deflection for appropriate flows is adjusted without checking the venturi itself, the
meter may still record flows erroneously.

Testing Meters
There are four ways meters may be tested. Meter testing methods are listed here in
order of decreasing effectiveness.

• Test the meters in place. Some pipes may need to be replaced to make this
possible. Use of an insertion pitot rod will provide a measurement to compare
against that recorded by the meter.
• Compare meter readings with readings of a calibrated meter installed in series
with the original meter (Sec. 12.5).
• Record meter readings for a given flow over a specified time period. Remove
the meter and replace it with a calibrated meter. Record readings from the
calibrated meter using the same flow rate for the same duration; compare the
readings.
• Test the meter at a meter-testing facility. This is usually not feasible or cost-
effective for very large meters.
Meters can be tested with portable equipment. Pump efficiency flow testing can be
used to check meters; it is sometimes provided free of charge by electric utilities. Some
utilities use an averaging rod meter or anubar to test meters, but results may be off by
as much as 10 percent. A standard single-point pitot rod gives more accurate results,
generally ± 2%. Meter testing may be done by an outside agency. Consultants, meter
manufacturers, and special testing laboratories offer testing services.