Approval of Standard Rating for Polymer-Rotor Pygmy Meter with Cat-Whisker Head (P-Pygmy or PPG Meter)


In Reply Refer To:                               December 2, 1988
WGS-Mail Stop 415




OFFICE OF SURFACE WATER TECHNICAL MEMORANDUM NO. 89.01

Subject:  EQUIPMENT--Approval of Standard Rating for Polymer-
                     Rotor Pygmy Meter with Cat-Whisker Head
                     (P-Pygmy or PPG Meter)

A solid-cup polymer rotor has been developed as an eventual
replacement for the metal bucket-wheel in the Price pygmy meter
(Surface Water Branch (SWB) Technical Memorandum No. 85.07, dated
May 13, 1985).  Preliminary calibrations were performed on a set
of prototype rotors in 1986.  In 1987, a production lot of polymer
rotors of a slightly modified design was procured for field use.
Thirty-one calibration tests using these rotors with cat-whisker
heads were performed in April-June 1988 at the U.S. Geological
Survey tow tank at the Stennis Space Center, Mississippi.  The
calibration data were analyzed between July and November 1988.

This memorandum defines a standard rating for the P-Pygmy or PPG
current meter and releases the meter for general field use.

The PPG current meter has a Price pygmy yoke with cat-whisker head
and a yellow solid-cup polymer pygmy rotor.  The year of produc-
tion is indicated by the digits 87 molded into the top side of the
rotor web.  The standard rating for this meter is a two-leg
broken-line equation giving the water velocity, V, in feet per
second (ft/s), as a function of rotor speed R, in revolutions per
second (rps), as follows:

               V = 0.022 + 1.021 R  for R < 0.83 rps
                 = 0.056 + 0.980 R  for R > 0.83 rps

The rating is defined by calibration data for velocities between
0.16 and 12.0 ft/s.

The calibration data show root-mean-square (RMS) deviations
(sometimes called standard errors) of approximately 1 percent
between individual measured velocities and the standard rating for
velocities greater than or equal to 0.5 ft/s.  RMS deviations are
less than 2 percent and 3.5 percent at veloci-ties of 0.25 and
0.16 ft/s, respectively.  The statistical stan-dard error of these
estimates, based on a sample size of 31, is approximately
15 percent of the estimated value.  That is, the percentage RMS
devi-ations at velocities 0.5, 0.25, and 0.16 ft/s are 0.9+.1,
1.8+.2, and 3.2+.4 percent.  The RMS devia-tions for the PPG meter
are slightly smaller than those generally accepted for the metal-
bucket pygmy meter, but the difference is not significant.

The RMS deviation represents the magnitude of error that would be
expected in a single velocity measurement using a randomly
selected meter.  It combines the effects of both variability in
repeated measurements with the same meter (repeatability errors)
and meter-to-meter variability of individual ratings (bias error).
Repeatability test data suggest that meter-to-meter bias is a rel-
atively small component of the total.

As its name implies, the RMS deviation is a mean or average value.
Approximately one-third of the residuals in the calibration data
set are larger than the RMS deviation in absolute value.  Doubling
the RMS deviation yields a value called the uncertainty (at the
95-percent or 2-sigma confidence level).  Approximately 5 percent
of the residuals exceed this value in absolute magnitude.  The 2-
sigma uncertainties at 0.5, 0.25, and 0.16 ft/s are approxi-mately
2, 4, and 6.5 percent, respectively.  That is, approximately 1
velocity observation out of 20 may be expected to register errors
this large or larger.

As noted in SWB Technical Memorandum No. 85.07 and Water Supply
Paper 2175 (pages 86, 132, and 145), the pygmy meter is recom-
mended for depths between 0.3 and 1.5 feet, with the caveat that
there are unknown erratic underregistration errors when the meter
is placed closer than 0.3 feet to the channel bed or water sur-
face.  The pygmy meter also can be used with the 0.2/0.8-depth
method at depths of 1.5 feet or more.  Because the pygmy-meter
errors are greater at low veloci-ties, use of the AA meter is rec-
ommended, when depths are adequate, for velocities less than 1
ft/s; the AA meter underregisters when placed closer than 0.5 feet
to the channel bed or water surface.

When used with the Current-Meter Digitizer (CMD), the PPG meter,
like all cat-whisker meters, is subject to possible multiple-count
errors at velocities less than about 0.5 ft/s.  CMD audible tones
too closely spaced at low velocities indicate the occurrence of
this problem.

Adoption of a standard rating for the PPG meter does not imply
that the standard rating should be accepted for an individual
meter that consistently deviates significantly from the standard
rating.  Any meters put into service with new PPG rotors should be
adjusted to yield a satisfactory spin test.  Spin time for a prop-
erly adjusted PPG meter will approach 2 minutes and a minimum
acceptable time is 1 minute.  Proper adjustment of the cat whisker
is critical, especially at low velocity.  If the PPG meter has the
old-style lead-ball cat whisker, the cat whisker should be dis-
carded and a new one (stock number 1103006) installed.  Mea-
surement results should be scrutinized carefully for consistency
with results obtained from other meters.

If a PPG rotor appears to be giving questionable results, the com-
plete meter and a memorandum explaining the circumstances should
be sent to the Hydrologic Instrumentation Facility (HIF), Field
Service and Supply Section, John C. Stennis Space Center,
Mississippi, for inspection and possible repair or individual rat-
ing.  Copies of the memorandum should be directed to the Office of
Surface Water Hydraulic Laboratory, John C. Stennis Space Center,
Mississippi, and to the Chief, Office of Surface Water, Reston,
Virginia.  Any informa-tion thus obtained may be useful in assess-
ing the accuracy of new rotors.




                                       Ernest F. Hubbard, Jr.
                                       Acting Chief, Office of Surface
                                         Water

WRD Distribution:  A, B, S, FO, PO