Professional Documents
Culture Documents
atura as y
,-----..,
o o
o
n
"r
1115191510161
1 171
1"
~v ....
1
U
I
Adh.
Catalog # XQ0601
The AGA does not, through its publications intend to urge action that is not in compliance with
applicable laws, and its publications may not be construed as doing so.
This report is the cumulative result of years of experience of many individual s and organizations
acquainted with the measurement of natural gas. However, changes to this report may become
necessary from time to time. If changes in this report are believed appropriate by any
manufacturer, individual or organization, such suggested changes should be cornmunicated to
AGA by completing the last page of this report titled, "Form for Suggestion to Change AGA
Report No. 7, Measurement of Natural Gas by Turbine Meters" and sending it to: Operations
& Engineering Services Group, American Gas Association, 400 North Capitol Street, NW,
th
4 Floor, Washington, DC 20001, U.S.A.
FOREWORD
This report is published in the fonn of a perfonnance-based specification for turbine meter for
natural gas flow measurement. It is the result of collaborative effort of natural gas users, turbine
meter manufacturers, flow measurement research organizations and independent consultants
fonning Task Group R-7 of AGA's Transmission Measurement Cornmittee (TMC). In addition,
comments to this report were made by the Committee on Gas Flow Measurement (COGFM) of
the American Petroleum Institute (API).
Research conducted in support of this report and cited herein has demonstrated that turbine
meters can accurately measure natural gas and, therefore, should be able to meet or exceed the
requirements specified in this report when calibrated and installed according to the
recommendations contained herein. Users should followappropriate installation, use and
maintenance ofturbine meter as applicable in each case.
This version of AGA Report No. 7 is intended to supersede aH prior versions of this document.
However, this document does not reference existing turbine meter instaIlations. The deCsion to
apply this document to existing instaHations shall be at the discretion of the parties involved.
Appendix B of this report contains the equations needed to convert volume measured at actual
(line) conditions to equivalent volume at base conditions, or to mass. These equations may be
used to perform such calculations with any type of positive displacement or inferential meter that
registers in units of volume.
IV
ACKNOVVLEDGEMENTS
Report No. 7, Measurement 01 Natural Gas by Turbine Meters, was developed by a Task Group
of the American Gas Association's Transmission Measurement Committee. Individuals who
made substantial contributions to the creation of this document are:
AGA acknowledges the contributions of the aboye individuals and thanks them for their time and
effort in getting this docurnent revised.
Lori Traweek
VI
TABLE OF CONTENTS
DISCLAIMER AND COPyRIGHT........................................................................................................................ 111
FOREWORD .......................................................................................................................................................... IV
ACKNOWLEDGEMENTS ....................................................................................................................................... V
TABLE OF CONTENTS .................................................................................................................................... VII
MEASUREMENT OF NATURAL GAS BY TURBINE METERS ................................................................ 1
1.
1.1
1.2
INTRODUCTION ............................................................................................................................................... 1
SCOPE .................................................................................................................................................................
PRINCIPLE OF MEASUREMENT .............................................................................................................................
1
1
2.
TERMINOLOGY ............................................................................................................................................... 2
3.
3.1
3.2
3.3
3.4
3.5
3.6
4.
4.1
COOES AND STANOAROS ......................................................................................................................................
4.2 METER BODY .......................................................................................................................................................
4.2.1 Meter Body End Connections ......................................................................................................................
4.2.2 Corrosion Resistance ....................................................................................................................................
4.2.3 Meter Lengths and Bores .............................................................................................................................
4.2.4 Pressure Tap .................................................................................................................................................
4.2.5 Sealing ..........................................................................................................................................................
4.2.6 Miscellaneous ...............................................................................................................................................
4.3 METER MARKINGS ...............................................................................................................................................
4.4 DOCUMENTATION ................................................................................................................................................
5.
5.1
5.2
5.3
5.4
6.
7
7
7
7
7
7
7
8
8
6.1
INTEGRITY TEST ................................................................................................................................................
6.2 LEAKAGE TEST ..................................................................................................................................................
6.3 CALIBRATION ....................................................................................................................................................
6.3.1 Calibration Conditions ...............................................................................................................................
6.3.1.1 Reynolds Number. ...............................................................................................................................
6.3.1.2 Density ................................................................................................................................................
6.3.1.3 CalibrationGases ................................................................................................................................
6.3.2 Calibration Guidelines ...............................................................................................................................
6.3.3 Calibration Configuration ..........................................................................................................................
6.3.4 Calibration Facilities .......... '" .....................................................................................................................
6.3.5 Calibration Results .....................................................................................................................................
Vll
12
12
12
12
12
13
13
14
14
14
14
6.3.5.1
6.3.5.2
6.3.5.3
6.3.5.4
6.3.5.5
6.4
6.5
7.
V1ll
APPENDIXA._....................................................................................................................................................... A-1
A.1
A.l.l
A.l.2
A.2
A.2.l
A.2.2
A.2.3
A.3
B.l.l
BASICGASLAWS ......................................................................................................................................... B-l
B.l.2
FLOW RATE AT FLOWING CONDITIONS ......................................................................................................... B-2
B.l.3
FLOWRATE AT BASE CONDITIONS ............................................................................................................... B-2
B .1.4
PRESSURE MUL TIPLIER ................................................................................................................................ B-2
B.l.S
TEMPERATURE MULTIPLIER ......................................................................................................................... B-3
B.l.6
COMPRESSIBILITY MUL TIPLIER .................................................................................................................... B-3
B.l.7
EQUATIONS FOR METER RANGEABILITY ...................................................................................................... B-3
B.l.7.l Maximum Flow rate .............................................................................................................................. B-3
B.2
C.2
C.3
CA
D.2
D.3
D.4
D.5
E.2
E.3
E.4
E.s.
IX
F.2
1.
Introduction
1.1
Scope
These specifications apply to axial-flow turbine flow meters for measurement of natural gas,
typically 2-inch and larger bore diameter, in which the entire gas stream flows through the
meter rotor. Typical applications include measuring single-phase gas flow found in
production, process, transmission, storage, and distribution and end-use gas measurement
systems. Typical use is the measurement of fuel grade natural gas and associated hydrocarbon
gases either as pure hydrocarbons or as a mixture of pure hydrocarbons and diluents.
Although not within the scope of this document, turbine meters are used to measure a broad
range of fluids other than natural gas.
This report does not address the characteristics of electronic pulse signal generating devices
within or attached to the meter, although it does address the use oftheir outputs.
AIso not addressed are the characteristics of mechanical or electronic instruments that convert
meter outputs from line conditions to base conditions. However, Appendix B do es contain the
equations establishing the mathematical basis for the conversion process. Although these
equations appear in this report, they may be used to convert volume registered by any type of
meter.
1.2
Principie of Measurement
Turbine meters are inferentiaI meters that measure flow by counting the revolutions of a
rotor, with blades, which turns in proportion to the gas flow velocity. From the geometry and
dimensions of the rotor blades and flow channeI, for a particular turbine meter size and
model, the gas volume at line conditions can be inferred trom counting the number of rotor
revolutions. The revolutions are transferred into digital readout or electronic signals by sorne
combination of mechanical gearing, generated e1ectronic or optical pulses, or frequency. The
accumulated line volume can be converted to base volume at standard or contract conditions
by accessory devices. Turbine meters can operate over a wide range of gas and ambient
conditions. Their upper flow capacities are established and limited by maximum local internal
gas velocities, noise generation, erosion, rotor speed, shaft bearing wear and pressure losses.
The maximum flow capacity at line conditions is fixed for a particular turbine meter
regardless of the operating pressure and temperature. The maximum base flow capacity
increases in accordance with Boyle's and Charles' laws. Minimum flow capacities are limited
by fluid and non-fluid drags (i.e., windage and mechanical friction los ses, respectively) that
cause a particular turbine meter design to exceed the desired or prescribed performance
limits.
2.
Terminology
Designer
Error
K-factor
MAOP
Manufacturer
Measurement cartridge
Meter factor
Pressure drop
The flow rate through the meter under a specific set of test or
operating conditions.
The maximum gas flow rate through the meter that can be
measured within the specified performance requirements.
The minimum gas flow rate through the meter that can be
measured within the specified performance requirement.
The transition flow rate. The flow rate through the meter at
which performance requirements may change.
Rangeability
Reference meter
Repeatability
Rotor factor
User
3.
Operating Conditions
Gas Quality
3.1
The meter should, as a minimum requirement, operate with any of the nonnal range natural
gas composition mixtures specified in Table 1 of AGA Report No. 8, Compressibility Factors
oiNatural Gas and Other Related Hydrocarbon Gases (Reference 1).
The manufacturer should be consulted if any of the following are expected:
Operation near the hydrocarbon or water vapor dew point of the natural gas mixture.
Total sulfur levels exceeding 20 grains per 100 cubic feet, including mercaptans, H2S
and elemental sulfur compounds, or exceeding those specified in the National
Association of Corrosion Engineers (NACE) guidelines for the materials ofwhich the
meter is manufactured.
Exposure to other contaminants that may affect the meter's error by reducing the
cross-sectional f10w area or building up on other sensitive features. Deposits may
also contaminate bearing lubrication and lead to reduced service life.
Operating Pressures
3.2
The operating pressure of the meter shall be within the range specified by the meter
manufacturer. The manufacturer shall specify the maximum allowable operating pressure for
the meter design and construction. Turbine meters, in general, do not have a minimum
operating pressure limit, although error may be increased if used under conditions for which
the meter has not been calibrated. Section 6 provides information on calibration requirements.
3.3
The meter shall be used within the manufacturer's flowing gas and ambient air temperature
specifications. Depending upon material of construction, turbine meters can operate over a
f10wing gas and ambient temperature range of -40 o p to + 165P (-40C to 74C). It is
important that the flowing gas temperature remain aboye the hydrocarbon dew point of the
gas to avoid possible meter damage and measurement error. The manufacturer shall provide
gas temperature and ambient air temperature specifications for the meter, as they may differ
from the aboye.
3.4
Gas density can have three principal effects on the performance ofthe gas turbine meter:
Pressure Drop - The pressure loss across a turbine meter increases as the gas density
increases.
3.5
The manufacturer shall provide the operating flow rate range at various pressures. The user
needs to consider the relationship between flow rate, error, pressure 10ss and service life. The
performance requirements for operation are stated in Section 5.1 of this document. The
pressure 10ss across a turbine meter increases with the square of a flow rate increase. Bearing
lubrication or visual inspection frequencies may need to be adjusted in accordance with the
operating flow rateo Flow limiting devices may be required to provide over-range protection
for the meter. Designers and users are cautioned to evaluate noise, piping safety and meter
integrity concems at maximum operating velocity. Refer to Section 7 of this document for
more inforrnation on installation considerations.
3.6
Research was conducted on the effects of installation configuration on turbine meter error in
2002 and the results published in Reference 2, Section 7 provides inforrnation on installation
requirements.
4.
The meter body and a11 other parts eomprising the pressure eontaining struetures shall be
designed and construeted of materials suitable for the service conditions for whieh the meter
is rated and in aceordance with any applicable eodes, regulations and speeifieations of the
designer. The meter body sha11 operate without leakage or pennanent defonnation over the
expeeted range of operating pressures, flowing gas temperatures and environmental
conditions.
4.2
Meter Body
4.2.1
The body end connections shaU be designed in accordanee with appropriate flange or
threaded connection standards.
4.2.2
Corrosion Resistance
AH wetled parts of the meter shaU be manufaetured of materials suitable for use in their
intended application. A11 external parts of the meter should be made of corrosion-resistant
material s or sealed with a corrosion-resistant coating suitable for use in environmental
conditions typically found in the natural gas industry aml/or as specified by the designer.
4.2.3
Manufacturers shaH publish their standard overa11 face-to-face length of the meter body
for each meter size and pressure rating. Turbine rneters are genera11y tolerant of minor
diameter differences, such as pipe schedule size changes. However, the designer sha11
make sure that the recommendations of Section 7 are followed.
4.2.4
Pressure Tap
The rnanufacturer shall provide at least one pressure tap on the meter body. The static
pressure from the meter tap provided and identified by the manufacturer shall be used for
pressure correction of the meter registration volume.
4.2.5
Sealing
The meter may be provided with sealing arrangements to prevent access to its internal
working parts, adjustments and reprogramming. The sealing arrangements shall be such
that they do not prevent access to routine maintenance features of the meter, such as
lubrication points. Where measurernent cartridges are interchangeable, the means of
sealing the cartridge shal1 be designed to prevent access to adjustment and
reprogramming when the cartridge is removed from the meter body. Any means provided
to seal the cartridge to the meter body sha11 be independent of any other sealing means
provided. Independent seaJing sha11 al10w the body-to-cartridge seal to be removed
without permitting access to the cartridge's intemal working parts or adjustments.
4.2.6
Miscellaneous
The construction shaIl be mechanicaIly and electricaIly sound, and the materials, finish,
etc., should be such as to provide assurance of long life and sustained accuracy. The
meter may provide one or more outputs (mechanical or e1ectrical), proportional to the
volume of gas that has passed through it, expressed at line conditions of pressure and
temperature.
The meter shaIl be designed in such a way that the body will not roIl when resting on a
smooth surface with a slope ofup to 10 percent. The meter design shall al so permit easy
and safe handling of the meter during transportation and installation. Threaded holes for
hoisting eyes or clearance for lifting straps shall be provided.
Meter Markings
4.3
A name platee s) containing the fo11owing information shall be affixed to the meter
Manufacturer
Serial number
4.4
Documentation
The manufacturer shall provide a11 necessary data, certificates and documentation for correct
configuration, set-up and use of the particular meter upon request by the user or designer. The
user or designer may also request that copies of hydrostatic-test or lcak-test certificates,
material certifications and casting or weld radiographs be supplied with delivery of the meter.
The manufacturer shalI provide or make available the following documents with the meter or
when requested; all documents shall be dated:
a) A description ofthe meter, giving technical characteristics and principIe of operation.
b) A perspective drawing or photograph ofthe meter
e) A list ofparts with a description oftheir constituent material s
d) A dimensional drawing
e) A drawing showing locations of seals
f) A drawing of the data plate or badge, showing arrangement of inscriptions
g) Instructions for installation, operation, and periodic maintenance
h) A general description of operation
i)
A description of available mechanical outputs and electronic output signals, and any
adjustment mechanisms
j)
5.
The manufacturer shall specify flow rate limits for Qmin, Qt and Qmax for each meter design
and size. Meter performance at atmospheric pressure sha11 be within the fol1owing tolerances
(see a1so Figure 1) after calibration.
Repeatability:
Maximum error:
Note 1.
The tolerances apply after adjustment ofthe change gears (if any) andlor
setting ofK-factors and application ofthe fmal meter factor.
The tolerances apply after any corrections perfonned within the meter
Note 2.
itself but prior to the application of any linearization algorithms by equipment
auxiliary to the meter.
Note 3.
These tolerances are applicable at atmospheric pressure. As operating gas
pressure increases, the perfonnance of the turbine meter can be expected to improve
dramatica11y, with sma11er values for repeatability and maximum peak-to-peak error,
provided the meter is calibrated for the intended operating conditions.
o
"ID
1.75
1.50
1.25
1.00
0.75
0.50
0.25
t====
Repeatabilily +/-0.2%
Maximum peak-to-peak error
1.0% (a ~ Qt)
0.00
eQ) -0.25
~ -0.50
-0.75
-1.00
-1.25
-1.50
-1.75
f
Qmn
al ~ 0.2 Qmax
10
5.2
The turbine meter shall meet the aboye performance requirements over the full operating
range of temperature and gas composition.
5.3
Pressure Influences
Research on the effects of pressure on turbine meter performance was conducted in 2002 and
2003, and the results published in Reference 3. To minimize error, turbine meters should be
calibrated for the applicable operating conditions. Guidance on calibration requirements is
provided in Section 6.
5.4
Meters with interchangeable measurement cartridges are designed so that the measurement
cartridge can be removed from the meter body without removing the body from the
installation. This design facilitates in situ inspection and replacement or upgrading of a
cartridge.
The construction of a meter with an interchangeable measurement cartridge shaIl be such that
the performance characteristics specified in Section 5.1 are maintained after installation of the
cartridge in other meter bodies of the same manufacturer, size and model, or after repeated
removal and instaIlation of the measurement cartridge in the same meter body. However,
slight differences in geometry from the body in which the cartridge was calibrated, body
wear, cartridge-body misalignment or other influences may affect the performance of the
cartridge and resuIt in measurement error.
An independent study (Reference 4) was conducted to assess measurement error due to
cartridge change-out practices. The study indicates that operating a cartridge in a body other
than the one in which it was calibrated can introduce random measurement errors from a
negligible amount to as much as O.35%. Turbine meter users should bear in mind that
calibration of measurement cartridges on a stand-alone basis, while convenient and less
expensive than calibrating a cartridge and body as a combination, can add to measurement
error.
11
6.
Integrity Test
The manufacturer shall test the integrity of a11 pressure-containing components for every
turbine meter. The test shall be conducted in compliance with the appropriate industry
standard, (ANSIIASME B16.1, B16.5, B16.34 or other, as applicable).
6.2
Leakage Test
Every turbine meter shall be leak-tested by the manufacturer after final assembly and prior to
shipment to the customer or flow-calibration facility. The test shall be conducted in
compliance with the appropriate industry standard. In the absence of specific standard(s), it is
customary for manufacturers to conduct the test as follows: The test medium sha11 be a gas,
such as nitrogen or airo The leak-test pressure shall be at least 1.10 times the MAOP and held
for a minimum offive minutes. To pass this test, the meter must not have detectable leaks.
6.3
Calibration
Calibra/ion Conditions
Research (Reference 3) has shown that the performance of turbine meter s vares with
changes in flow rate and operating pressure. These variations are related to changes in
Reynolds number and, in sorne cases density, and are particularly significant at low and
intermediate operating pressures and flow rates. Attention to these issues at the time of
calibration is crucial for optimal measurement. The following sections pro vide further
guidance in this regard.
6.3.1.1
Reynolds Number
12
(6.1)
p (D) (V) / JI
Reynolds number may also be calculated from either of the following formulae:
Re
4(Q) / 1f(D) (0
Re
p(rho)
D
V
where
Q
v (nu)
J.l (mu)
(6.2)
(6.3)
Reynolds number
Density
~eter diameter
BuIk (average) velocity offlowing fluid
Volumetric flow rate
Kinematic viscosity
Absolute viscosity
The aboye quantities must all be determined at the same conditions of temperature
and pressure.
The relationship between bulk velocity and flow rate is:
(6.4)
The relationship between absolute and kinematic viscosity and density is:
(6.5)
A meter calibration carried out in a test facility over a particular range of Reynolds
numbers characterizes the meter' s performance when used to measure gas over the
same range of Reynolds numbers when the meter is in service. Therefore, the Kfactors established during such a calibration, in most instances, can be used to
compute flow measured by the meter in service.
6.3.1.2
Densty
Research (Reference 3) has shown that the performance of sorne meters may al so be
sensitive to variations in gas density. Variations in calibration tend to be larger at
lower gas densities. Users with low-pressure, low-flow applications should consult
the meter manufacturer for meter performance characteristics and obtain calibration
data at the operating density to ensure that no significant measurement errors exist.
Additional information on density matching is provided in Appendix E.
6.3.1.3
Calibratan Gases
The research described in Reference 3 was conducted using natural gas and air as test
media. In addition, Reference 6 describes research that has been conducted to
establish the suitability of other gases for calibration of turbine meters. The data
show that turbine meters used in natural gas can be effectively calibrated in different
13
gases, and that satisfactory measurement will result provided calibration is conducted
over the range of Reynolds numbers ami/or density expected at operating conditions.
Further information on calibration in altemative gases is provided in Appendix E.
6.3.2
Calibration Guidelines
As discussed aboye, the expected operating Reynolds number range ami/or density for a
meter needs to be taken into account when designing a calibration programo This requires
establishing the expected range of flow rates and the properties of the gas to be measured
at the intended meter location. The gas properties may be determined directly by
measurement or by calculation from empirical equations.
Test points should be selected throughout the range offlow rates over which the meter is
to be tested. It may be decided to concentrate the majority of the test points in the range
ofthe meter's heaviest expected usage.
Further information and sample calculations appear in Appendix E.
6.3.3
Calibration Configuration
6.3.4
Calibration Facilities
Test facilities used for meter calibration shall be able to demonstrate traceability to
relevant national primary standards and provide test results that are comparable to those
from other such facilities.
6.3.5
Calibration Results
During calibration, the appropriate K-factor(s), meter factors, change gears ratios and
rotor factors will be established. The applicable factors will be established for each output
for meters with more than one output. Refer to Appendix D for detailed information and
examples of determining and applying these factors.
6.3.5.1
Change Gears
For turbine meters with mechanical output(s), intemal gearing is typically used to
adjust the registration to produce a (nearIy) fmite indicated volume (e.g., 100 cubic
feet, 10 cubic meters, etc.) for each revolution of the output shaft. Differing change
gear sets, comprised of two replaceable mating gears incorporated within the gear
train, perrnit adjustments to be made to the overall gear ratio. While change gear sets
with many ratios are available, it is not always possible to install gears with the
precise ratio needed. Thus, there may be sorne residual bias in the meter's calibration
14
even after the best available change gears have been installed. The change gears are
usualIy located in a non-pressurized region of the meter that is accessible during
calibration, but that can be sealed to prevent unauthorized access. When an
interchangeable measurement cartridge is moved to a new body, the change gears
shalI be moved also.
6.3.5.2
K-Factor(s)
For turbine meters with electronic output(s), the appropriate K-factor(s) is established
at the time of calibration. These value(s) are then entered into an electronic accessory
device. The K-factor(s) is expressed in units of pulses/unit volume. By dividing the
accumulated pulses by the K-factor or by dividing the instantaneous pulse frequency
by the K-factor, the accumulated volume or the instantaneous flow rate, respectively,
can be determined.
6.3.5.3
Meter factors are non-dimensional multiplier values. They are derived from
calibration data by dividing the true volume of the reference meter by the indicated
volume of the test meter, both volumes having first been corrected to the same base
conditions. Alternatively, meter factors can be calculated from the percent error
values provided at each calibration flow rate, by the formula:
M eter factor = 100 / (1 00 + percent error)
Thus, the meter factor example of 1.005 would be the same as -0.5 percent error. The
mechanical or electronic outputs of a turbine meter may be adjusted by the
application of individual meter factors for specific flow rates or by a single final
meter factor over the range of flow rates. This may be done omine manualIy or
online in an electronic accessory device. The calibration facility may provide meter
factors in addition to or in place of percent error values for each test flow rate of a
meter.
6.3.5.4
For dual-rotor turbine meters, with associated algorithms for enhanced performance
and diagnostics, the manufacturer will supply unique K -factors for each rotor' s
electronic pulse output. These are referred to as "rotor factors" to distinguish them
from K-factor, which is the term historically used to apply to the single-rotor
electronic output of a meter. Refer to Appendix A, Sections A.2 and A.3 and to
Appendix D, Section D.5 for more details.
6.3.5.5
Following an adjustment, at least one test point shall be repeated to verify that the
adjustment was calculated and applied correctly. If a linearization technique is
applied in secondary or companion electronics, then at least two test points shalI be
repeated.
15
6.4
Test Reports
The resuIts of each test required in Section 6.3 shall be documented in a report including, as a
mmlmum:
a) The name and address of the manufacturer
b) The name and address of the test facility
c) The model, size and serial number of the meter
d) The date(s) ofthe test
e) The name and title of the person who conducted the tests
f)
j)
k) The value of any adjustment made and the results of the verification test
A copy of the report shall be available from the testing organization for a period of five years.
6.5
Quality Assurance
The manufacturer shall establish and follow a comprehensive quality-assurance program for
the assembly and testing of the meter and its electronic system (e.g., ISO 9000, API
Specification Q 1, etc.). The user shall have access to the quality-assurance documents and
records.
16
7.
Installation Specifications
The impact on measurement has been assessed for the configurations described below. Various
organizations have published test data. Configurations other than those described below may
result in unacceptable measurement errors and are not recommended without further testing.
7.1
General Considerations
7.1.1
Flow Direction
Turbine meters, designed for flow in one direction only, shall be installed accordingly.
Reverse flow may not damage the meter internals but may result in registration error. Tbe
manufacturer may be consulted if reverse flow has occurred. Where reverse flow is
expected, additional valving is necessary to allow gas to flow through the meter in the
forward direction only, unless the turbine meter is recommended for bi-directional flow.
7.1.2
The meter and adjacent pipe sections should have the same nominal diameter, but
schedule changes are acceptable provided satisfactory meter performance has been
demonstrated. Meter inlet and outlet connections and companion pipe flanges shall be
aligned concentrically. Gaskets shall not protrude into the flowing gas stream. Gasket
protrusion or flange misalignment can affect meter perfonnance.
7.1.4
Internal Surfaces
The intemal surface of the meter should be kept cIear of any deposits that may affect the
meter's cross-sectional area. The meter's perfonnance depends on a known crosssectional area. Pipe interior surfaces should be of commercial roughness or better. Welds
on piping at the meter inlet and outlet should be ground flush with the internal surface of
the pipe so that they do not protrude into the gas stream.
7.1.5
The temperature well shall be located downstream of the meter to keep disturbances to a
minimum. Generally temperature wells are installed between one and five nominal pipe
diameters from the meter outlet but upstream from any valve or flow restrictor. It is
important that the temperature well be installed to ensure that heat transfer from the
adjacent piping and radiation effects of the sun do not influence the temperature reading
of the flowing gas.
17
7.1.6
The pressure tap provided by the manufacturer on the meter shall be used as the point of
pressure sensing for recording or integrating instruments and during calibration.
7.1. 7
Flow Conditioning
A flow conditioner may be used upstream of the turbine meter to reduce or eliminate the
effects of swirl amI/or asymmetric flow. Headers, pipefittings, valves and regulators
preceding the meter inlet may cause perturbed flow conditions. Flow conditioners shall
be installed as specified in the following sections. There shall be no protrusions into the
piping between the flow conditioner and the meter.
7.1.7.1
For specifications for these devices, refer to the latest revision of AGA Report No. 3,
Orifice Metering 01 Natural Gas and Other Related Hydrocarbon Fluids (Reference
10). This design has demonstrated its effectiveness in the reduction of swirl but does
not eliminate asymmetric flow.
7.1.7.2
Isolating flow conditioners offer an alternative to tube bundles. They are recommended for use if the contracting parties agree. Isolating flow conditioners general1y
consist of perforated plates in various parterns, sometimes accompanied by vane
assemblies. Several of these devices have been evaluated for performance and found
to be effective in reducing swirl and asymrnetric flow.
7.1.7.3
7.2
Research (Reference 2) shows that turbine meters may be operated according to the
recornmendations in this section with acceptable results, while more severe piping
arrangements may result in considerable error. The magnitude of the error, if any, will be a
function of the extent of the flow disturbances, the meter' s design, the quality of external and
integral flow conditioning, amI/or the meter's ability to adjust for such conditions. However,
other configurations may be used provided they are shown to be acceptable based on
published experimental data.
18
7.2.1
Optional
- Rel. Temperature Well
Turbine Meter
- Sample Probe
Pressure Tap
Recommended
- Blow Down Valve [3]
--1
0""'....
'-,
l,
c==-==~:=J
---Ir1I
Temperature Well
'
.,J\f.ii
7 _
I
.!-\ ':'f[J'
r-~--~~~~
NOTES:
Diameters
'.
Device
,,, ,,
,, ,,
r-------------~, :
L.
L.
Optional _______---l
- 19 Tube Bundle
- or Flow Conditioning Element
Optional
-Flow Limiting
Minimum
5 Nominal Pipe
Diameters
[1] Recommended spacing, unless otherwise supported by published test data lor the fiow conditioning elemen!.
[2] No pipe connections or protrusions allowed within this upstream section.
[3] For recommended size 01 blow down valve, see Table 1. Locate downstream 01 meter.
of the meter. There shaIl be no pipe connections or protrusions within the upstream or
downstream piping other than pressure taps, temperature wells or flow-conditioning
elements.
A typical recornmended installation meter run with accessories and optional devices is
shown in Figure 3. The maximum pipe-size difference upstream or downstream of the
recornmended installation should be one nominal pipe size. Valves, filters or strainers
may be instaIled upstream or downstream of the recommended instaIlation piping. Any
valve immediately upstream of the installation shall be fuIly open during meter operation.
Strainers and filters should be kept c1ean for optimum performance.
19
Turbine
Optional
- Filter
90 Elbow or Tee - or Strainer
Maximum
Pressure
Tap
Reduction
Lo
~e~perature
10 Nominal Pipe
: [:---1
OH
~L - -
r-
~,. ~"'~\
__
::
.,
.7
5 Nominal
Pipe[1][2]
-,
_c_
":. . :' j
Recommended
Blow Down
V~~~!~)stream
:10;J':lJ--"'-M~[):
l.
Optional
- 19 Tube Bundle
psigl31
-1
- - - - - --- - - - - - - - --- - - -
Minimum
+- 5 N~;:nal I
Lol..lj
~I
~"";':' ...."j
Optional
[ . _ Bypass Une _
",,_
-l f~(->''.
"""V....
,'v
~~ I
lo - - - - - - - - - - - - - - - -
NOTES:
... 1
[1) Recommended spacing, unless olherwise supported by published test data lor the flow conditioning element.
[2) No pipe connections or protrusions allowed within this upstream section.
[3) Size 01 pressure loading line and valve to be the same as recommended blow down valve sizing, (see Table 1).
7.2.2
The use of the following optional installation configurations may result m relatively
higher, but still acceptable, measurement uncertainty.
7.2.2.1
Short-Coupled Installation
20
- Lo
Diameters"" ~"'~\
l'
Diameters
-
DO.
t-_-..,.
~~~"j
1- 5
~"~"~
Optional
- Flow limiting
Device
Optional
- Ref. Temperature Well
- Sample Probe
the inlet riser shall be fully open during meter operation, and strainers and filters
should be kept c1ean for optimum performance.
Optional
_Filler
_or Strainer
90 o Elbow or Tee
Maximum Reduction
li
F J.
Inlet,L--'
.. 1..'
- -
-:
Optional
- Valve
~.: ~"
:,14..
=- _
(~~I'. ~ ~',
11
1 \\.....~ "1
L~~
Optional
19 Tube Bundle
- or Flow Conditioning Element
9~~
.. L.'
::::::::C~:f::]
-"
I
r,~
---, i '
,, ii ,'
_..J i '
Minimum
_.:~
C!-==J
,' 1'----I ,
I
,1'
~
, I '- -
Pressure Tap
Recommended
Pressure-Ioading Une
and valve lor operation
over 200 psig [3)
OPtiona~--1
c.A::!
-:
"j
, ~.: ~ ~~
':' lc< \ 11 Recommended
1 \ ....
\\~_'"
-BlowDown
~ ".-,:...J
Valve [3)
,-"-,
Optional
- Flow Limiting
Device
-V. . .
"1
-~
NOTES; [1) Recomrnended spacing, unless otherwise supported by published test data lor the flow conditioning elemen!.
[2) No pipe cooneclions or prolrusions allowed wilhin this upstream section.
[3) Size 01 pressure loading line and valve lo be the same as recommended blow down valve sizing, (Table 1).
[4) Turbine meler musl have integral flow conditioner.
7.2.2.2
Close-Coupled Installation
21
Pressure Tap
Turbine Meter (1)
90 o Elbow or Tee
Maximum Reduction One
Nominal Pipe Size
'-1-
1
--- 1 ,
, I 1
I-rl
1 I 1----1 I 1
1 I1
1 I 1__
... LI
, I 1
- -1 I 1
...,.~ l.
f'-=1.
-11 ... ,......-...-rl
'1'\..!.."/ ,'~
.~---"
II~ 1" r~ I)I~
~:.-..::_ ;~\.':::{;J~}'
.~_-""''1
Optional
,'~
J\\\
-Valves ..........,..," ~ 1)1
,,,( .. " '
J\H
r.. ~~
'~~( _
>.,'//
a_...
Optional
L\S
..' ..-:,.>,. . (
...,:::..- ......1
Recommended ~I ~ _ _
_~
Pressure-Ioading
1
line and valve for~ 90 o Elbow
operatlon over
200 psig [2)
!!.-_-
Recommended
Blow Down Valve
(Downstream) [2)
..
NOTES: [1) Turbine meter must have integral flow condilioning elemenl.
[2) Size of pressure-Ioading line and valve to be the same as recommended blow down valve sizing, (Table 1).
7.2.2.3
Research (Reference 2) has confirrned that turbine meters with integral flow
conditioning in the nose-cone flow passages operate satisfactorily in short and closecoupled installations. Those integral flow conditioners tested were similar in design
to that shown in Figure 6 and to those evaluated in Reference 8. For this design, the
aspect ratios are BID < 0.15 and SIL < 0.35. These parameters are illustrated in
Figure 6.
1--
"'1
22
7.2.3
\ -r-
90 o Elbow or Tee
Maximum Reduction
One Nominal Pipe
Size
,-
Pressure Tap
Inlel Piping 10 Nominal Pipe Diamelers Long
(5 Nominal Pipe Diamelers wilh 19 lube bundle)
(1)[2)
-1
Angle-Body
Turbine
Meler
!!~~~~~~~~~~"~~~~J r--~~~~
o
,--- -;-;
--J~; . . ",- "
....
Oplional
- Valve
90
:::::
,,
"-,..:-..."', ,.......
,~
::)1:,-"
"0
:~
Temperalure
Well
-)--;'':.~Q
1",__ 'p
"
\'....
.-.1----1
Recommended ~~~,
- Blow Down Valve
I },
,
_ _
_,
Downslream [3) / '
II
,A. '\" ~
'" tC<
~
~
~
Oplional
- 19 Tube Bundle
- or Flow Conditioning Elemenl
,',..- '\\\
'" rC< ')' Optional
"" ' .... , , - Valve
\ .........:-::,
Optional
- Flow Lim~ing
Device
Recommended
Pressure-Ioading
line and valve
for operalion
over 200 psig (3)
I '....
-~
NOTES: [1) Recommended spacing, unless olherwise supported by pubfished lesl dala lor Ihe f10w conditioning elemenl.
[2] No pipe connections or prolrusions allowed within Ihis upslream section.
[3] Size of pressure loading fine and valve lo be Ihe same as recommended blow down valve sizing, (see Table 1).
The meter inlet piping may be connected using a 90 elbow or tee, There are no
restrictions on the downstream piping except that the flange attached to the meter outlet
shall be full-size. Any valve immediately upstream ofthe installation shall be open fully
during meter operation.
The installation may be oriented vertically,
Caution: Users are cautioned that the error of the angle-body configuration has not been
confinned by published research. Contact the manufacturer for supporting experimental
data for specific installation requirements,
23
7.3
Environmental Considerations
7.3.1
Temperature
The meter shall be installed and used within the ambient and flowing gas temperature
limits specified by the manufacturero
7.3.2
Vibration
Associated Devices
7.4.1
Filtration of the flowing gas may not be necessary in all cases but is recornmended for
most meter applications. The accumulation of deposits due to a mixture of dirt, milI scale,
condensates and/or lubricating oils will deteriorate meter performance. Bearing wear and
measurement cartridge damage aml/or failure can be caused by foreign material in the
flowing stream. Normal pipeline gas quality may deteriorate during peak demands, plant
upsets and new tie-ins, or from normal internal pipeline corrosion resulting in dust, dirt
andJor scale. Under such conditions, it is recornmended that a strainer with a basket of
3/32 inch maximum hole size and 40 mesh wire liners be installed upstream of the meter
to catch the major part of this foreign material. In sorne instances, it may be preferable to
install lO-micron filters for the removal of fine dust, thus increasing bearing life and
minimizing deposits on the meters internal parts. A differential pressure gauge should be
installed across the filter or strainer to indicate an increase in pressure drop resulting from
a build-up of foreign matter in the filter or strainer. Normal pressure drop should be
observed and recorded at various flow rates when the strainer or filter is clean.
24
Inspection of the devices should be performed whenever higher than normal pressure
drops are indicated on the differential pressure gauge.
A greater degree of meter protection can be accomplished through the use of a dry-type
or separator-type filter installed upstream ofthe meter inlet piping.
When cornmissioning a pipeline, it is recornmended that the meter be bypassed or a
temporary strainer element installed to protect the meter from dirt and debris entrained
within the initial flow.
7.4.2
Throttling Devices
The installation of a throttling device, such as a regulator or partially closed valve, is not
recornmended in close proximity, especially upstream, to the meter. Where such
installations are necessary, the throttling device should be placed an additional eight
nominal pipe diameters upstream or an additional two nominal pipe diameters
downstream of the in-line recornmended installation in Figure 2. In the configurations
illustrated in Figures 3, 4, 5 and 7, the throttling device should be placedeight additional
nominal pipe diameters upstream of the inlet vertical riser or an additional two nominal
pipe diameters downstream of the oudet vertical riser. Placement of such a device in
c10ser proximity to the meter may result in increased uncertainty amI/or reduced bearing
life.
7.5
Precautionary Measures
7.5.1
Installation Residue
Va/ve Grease
Grease can flow from sorne pipeline val ves into the gas stream during lubrication. Val ve
grease can adhere to turbine meter blades, thereby affecting meter performance. Such
valve types should not be located irnmediately upstream of a turbine meter.
7.5.3
Over-Range Effects
Surges of high-velocity gas through a turbine meter can severely damage the rotor.
Extreme gas velocities can occur when pressurizing, blowing down or purging the meter
runo The operation of flow- or pressure-control devices in the downstream piping system
can also create extreme gas velocities.
7.5.3.1
Run Pressurization
It is good practice to provide isolation block valves for meter runs so that the meter(s)
can be maintained and calibrated without service interruptions. F or single meter run
stations, a flow bypass line should also be considered (see Figure 3). The isolaton
block valves must be operated in the proper sequence and slowly to avoid reverse
25
spinning andlor over-speeding the meter during startup. If operating pressures are
over 200 psig, a small pressure-Ioading line and valve around a large or fast acting
inlet block valve will allow the meter mn to be pressurized slowly to avoid overspeeding damage to the stationary rotor. Recornmended sizes for pressure-Ioading
lines and valves are the same as those for blow down valves in Section 7.5.3.2.
7.5.3.2
While most turbine meters can be operated beyond rated capacity for short periods of
time with no damaging effects, oversized blow down valves can cause rotational
speeds greatly in excess of the rated capacity. Therefore, blow down valves should be
sized as shown in Table l.
TABLE 1-BLOW DOWNVALvE SIZING
Meter Run
Valve Size
mm
Inches
mm
Inches
50
80
100
150
200
300
2
3
4
6
8
12
6
13
13
25
25
25
0.25
0.50
0.50
1.0
1.0
1.0
Consult the manufacturer for information on valve sizes for meters not covered in the
Table 1.
Sorne meters and secondary devices may be damaged when they are operated in a
reverse direction. In such cases, the blow down valve shall be located downstream of
the meter.
7.5.3.3
In those installations where excessive flow can occur as a result of the operation of
the downstream piping system or as a result of the operation of flow- or pressurecontrol equipment, a restrictive device may be installed in the piping downstream of
the meter ron to prevent meter over-ranging. An over-range protection device may be
sized to limit the flow through the meter to approximately 120% of the maximum
rated meter capacity. Meters shall not be operated beyond their rated capacity under
normal circumstances. Refer to the sonic nozzle and critical orifice sizes in Table 2.
A perrnanent pressure loss will occur even at sub-critical flow rates when one of
these devices is installed. Therefore, adequate pressure needs to be available at the
location. A critical orifice may result in up to 50% perrnanent pressure loss at critical
conditions. Any flow-limiting device may generate significant noise.
26
nD"
TURBINE
METER
RATING
cubic
cubic
meters/h
feeUh
100
3500
115
4000
130
4500
250
8800
255
9000
280
10000
450
16000
510
18000
680
24000
760
27000
850
30000
1000
35000
1020
36000
1420
50000
1620
57000
1700
60000
2550
90000
2830
100000
4000
140000
4250
150000
6230
220000
6520
230000
7650
270000
SONIC
120% of
RATING
cubic
~ubic feetlh
meters/h
120
4200
138
4800
156
5400
10600
300
306
10800
336
12000
540
19200
612
21600
816
28800
912
32400
1019
36000
1200
42000
1224
43200
1704
60000
1944
68400
2040
72000
3060
108000
3396
120000
4800
168000
5100
180000
7476
264000
7824
276000
9180
324000
27
VENTURI
mm
inch
13.0
13.7
14.7
20.6
20.8
21.8
27.7
29.5
33.8
35.8
37.8
40.9
41.4
49.0
52.3
53.6
65.5
69.1
81.8
84.6
102.6
104.9
113.5
0.51
0.54
0.58
0:81
0.82
0.86
1.09
1.16
1.33
1.41
1.49
1.61
1.63
1.93
2.06
2.11
2.58
2.72
3.22
3.33
4.04
4.13
4.47
nD"
"T"
CRITICAL
ORIFICE
CRITICAL
MAX.
ORIFICE THICKNESS
mm
inch
15.2 0.60
16.3 0.64
17.3 0.68
24.1 0.95
24.4 0.96
25.7 1.01
32.3 1.27
34.3 1.35
39.6 1.56
42.2 1.66
44.2 1.74
47.8 1.88
48.5 1.91
57.2 2.25
61.2 2.41
62.7 2.47
76.7 3.02
81.0 3.19
95.8 3.77
99.1 3.90
120.1 4.73
122.7 4.83
132.8 5.23
mm
inch
1.88
2.03
2.13
3.00
3.02
3.20
4.04
4.29
4.95
5.26
5.54
5.99
6.07
7.16
7.65
7.82
9.60
10.11
11.96
12.40
15.01
15.34
16.61
0.074
0.080
0.084
0.118
0.119
0.126
0.159
0.169
0.195
0.207
0.218
0.236
0.239
0.282
0.301
0.308
0.378
0.398
0.471
0.488
0.591
0.604
0.654
D (inch) ]
0.00893
=..J1ifi:6 x
= 1.291
air rate
[D (ineh)]2
0.00893
. h) ( V ' )
(0.00893)2 x Gas rate
D ( me
entun =
1.291
D(ineh) (Venturi) = (0.00786) .JGas rate
D (ineh) (Orifice) = 1.17 x D (in eh) (Venturi)
Note: To be sure that the orfice perfonns as a thin-plate, sharp-edged orifice in
critical flow, the ratio of orifice pI ate thickness to the hole diameter shall be less than
or equal to 0.125. Refer to Reference 9 for additional infonnation. If the thicknessto-diameter ratio is larger than 0.125, then discharge coefficients can have large and
uncertain values within the range of 0.8 to 0.95 (vs. 0.73). This is caused by
boundary layer/shock interactions within the orfice.
---1 r--
T or less
T or less
Flow
Diameter - D
Symmetric plate,
both orifice edges sharp
28
7.6
Accessory Installation
7.6.1 Density Measurement Devices
When using den sitometers , it is desirable to sample the gas as cIosely as possible to the
meter. Care should be exercised not to disturb the meter inlet flow or to create an
unmetered bypass. Densitometers should be installed downstream of turbine meters.
Refer to manufacturers' installation instructions for further information.
7.6.2
29
8.
General
In addition to sound design and installation practices, turbine meter performance is dependent
on good maintenance procedures, regular inspections and periodic field checks. The
frequency of mantenance is dependent on the flowing gas conditions, station operation
and/or contract requirements. Meters that operate under dirty flowing gas conditions will
require more frequent inspections. In addition, the flowing gas condition will influence the
oiling frequency for lubricated turbines. Changes in the performance of a meter may be
detected by self-checking features, by visual inspection of the internal mechanism, by spin
time tests, or by calibration. Section 6.3 recornmends matching in-service conditions during
calibration to determine the best indication of meter error. However, periodic calibration
using atmospheric air can be useful for monitoring ongoing meter performance.
Sorne dual-rotor turbine meters have output readings that can be used for periodic or
continuous checking. These readings can be used to determine the need for maintenance or
inspection. A turbine meter can al so be field-checked by either another meter in series or a
check rotor in tandem with the metering rotor in a two-rotor turbine meter. In the case oftwo
meters in series, the check meter (a turbine meter or other suitable meter) must be installed
relative to the field meter so that there is no effect on either meter's performance from the
presence of the other. The effects of flowing pressure and temperature on both meters should
be considered along with the error of the check meter at the operating conditions. In the case
of a dual-rotor meter, both rotors can be calibrated, and field checking achieved by
comparing the ratio of the two rotor outputs.
8.2
Visual Inspection
A visual inspection of flow conditioners, upstream and downstream piping, and the meter
internals should be performed periodically to ensure there is no accumulation of debris,
particularly in the flow passage area, drains, breather holes and lubricaton systems.
Removing the measurement cartridge from the meter body facilitates an inspection of the
internal mechanisms. If the cartridge is not removable, remove the meter from the piping or
use a borescope. The closures on the ends of the ron may be removed or opened for internal
inspecton of the piping.
An inspection of the measurement cartridge consists of examining the rotor for damaged or
missing blades, accumulation of solids, erosion or other damage that would affect rotor
balance and blade configuration. When a meter is disassembled for any purpose, the
mechanism should be thoroughly cleaned to remove dirt or foreign material. Meters in
operation can ofien yield information by the noise they generate or by vibrations felt through
the body. If the meter has severe vibration, it usually indicates damage. This condition will
lead to complete rotor failure. Rotor rubbing and deteriorated bearings can ofien be heard at
relatively low flow rates where such noises are not masked by normal flow noise. Concurrent
with the internal meter inspection, checks should be performed to ensure that gaskets are
properly aligned and that flow conditioners are free of obstructions.
30
8.3
Spin time tests are not intended to take the place of inspection, maintenance or periodic
assessment of the meter's errors via a calibration check. However, a spin time test can be a
practical indicator of the relative level of mechanical friction in the meter. Increased
mechanical friction can result in degradation of meter performance and registration errors,
especially at low flow rates and low operating pressures. As mechanical friction inereases, the
potential for bearing or other component failure increases.
Spin time is not indicative of overall meter performance. Conditions, such as damage or wear
to the rotor and internal components, or debris and foreign material inside the meter, can
affect meter performance with minimal change to the spin time. A thorough inspection should
also be carried out when conducting a spin time test.
The manufacturer provides spin times for individual meters and may provide spin times for
the meter at various stages of disassembly. Such guidelines may also inelude minimum spin
times for various models and sizes of meters. The manufacturer's published guidelines and
procedures for conducting spin time tests should be followed. An example of a spin time test
procedure appears in Appendix F.
31
It is recornmended that an initial spin time test be conducted to establish a baseline for the
meter with the meter or measurement cartridge completely assembled except for register or
recording gauges where gear-driven. When accessory devices (register, integrating gauge,
pulse generators, etc.) are installed, care should be taken to ensure that no excess friction is
introduced. A spin time test should be performed to ascertain that the accessory device has
not affected the meter.
After oiIing, the meter should be operated according to the manufacturer's guidelines and
procedure to reduce any drag from excess oH before performing a spin time test. When a
meter that has been idIe for a long period of time faiIs to meet the manufacturer's specified
minimum spin time it should be oiled and then operated for a period of time before repeating
the spin time test.
Spin time tests may be conducted on complete meters or on measurement cartridges alone. If
either the meter or the cartridge is removed from the run for testing, the test shouId be
conducted in a draft-free environment with the mechanism supported in its nonnaI operating
position. Conducting a spin time test with the meter in line requires depressurizing the meter
runo Ensure that the meter run shutoffvaIves do not Ieak because Ieakage ofthe vaIves andlor
drafts in the meter run will affect the test. Low-pressure gas from a hose or tubing can be
used to rotate the turbine rotor at a sufficient rate to begin the in-line spin time test. Bypasses
around the shutoff vaIves can be installed for this purpose. Care should be taken to ensure
that vented gas does not accumulate.
Regardless of location, the test is conducted by setting the rotor in motion, manually or by a
jet of air or gas, in the same direction as under flowing conditions. The rotor is timed until it
stops rotating. When the rotor is set in motion by a jet, significant time may be added if the
rotor is tumed at exceptionally high speeds. Ambient temperature, lubrication, the presence of
accessories, the manner of initiating the blade rotation and other factors affect spin times and
they must be considered to obtain repeatable and comparable results from test to test. It is
recommended that records of spin time test be maintained for the purpose of detecting
changes in bearing integrity over time. A typical decay curve for meter spin time is provided
in Figure 9.
1000
900
800
~e
':lE
lO
100
600
e
o
500
72o
400
:>
300
200
"'-
100
O
20
i'--.
---
40
--60
BO
100
120
140
Time - Seconds
32
A spin time test should be repeated three times, with deviations les s than 10% from the
average of spin time. If spin times are les s than those recornmended by the manufacturer, the
tests may be repeated at various levels of disassembly until the source of the abnormal
friction is determined. Cleaning, oiling or replacing the bearings or other components may
bring the spin time back to an acceptable value. Bearings, shafts, magnetic coupling
assemblies or gearing may be replaced on sorne meters without affecting the meter's
performance. Consult the manufacturer for specific recornmendations. A follow-up spin time
test should be performed after repairs or component replacement. If an acceptable result is not
obtained, the meter should be removed from service.
8.5
Dual-rotor turbine meters may offer the ability to check the operation of the meter in situ by
comparing the rotor outputs. Consult the manufacturers' literature for further information.
8.6
Retesting Considerations
Meters, or their measurement cartridges, should be retested on a periodic basis. The period
between tests should be cornmensurate with meter usage and line conditions. In sorne cases,
regulatory agencies establish the test intervals. The decision to perform periodic transferproving or flow calibration is left to the users.
Often, when a meter or cartridge is retumed for repair/recalibration, the user requests an asfound calibration in order to have a record of the meter errors when it was removed from
service. Such information is useful in the event of a measurement dispute and may be helpful
in establishing recalibration intervals.
33
APPENDIXA
Schematics ofaxial-flow single-rotor gas turbine meters are shown in Figures A.I and A.2.
Gas entering the meter increases in velocity through the annular passage formed by the nose
cone and the interior wall of the body. The movement of gas over the angled rotor blades
imparts a force to the rotor, causing it to rotate. The ideal rotational speed is directly
proportional to the flow rateo The actual rotational speed is a function of the passageway size
and shape, and the rotor designo It is also dependent upon the load that is imposed due to
internal mechanical friction, fluid drag, extemalloading and the gas density.
Body
Rotor
Mechanical or
Electrical
Readout
End Connectio
Inlet
Outlet
Electronic
Pickup
Mechanism
Housing and Tail
Cone
A-l
Body
Electronic
Pickup
Rotor
Inlet ~ Annular
Passage
+-I~Outlet
Tail
Figure A.2. Single Rotor Turbine Meter (Low Torque Design)
A.1.2
The basic designs ofaxial-flow gas turbine meters differ significantly from liquid turbine
meters due to density, viscosity and compressibility differences of the fluids. The need to
extract sufficient kinetic energy trom the flow to provide the torque to overcome internal and
external frictional los ses results in the proportions of the nose cone and annular passages
typical of those shown in Figure A.l. However, gas turbine meter designs similar in
proportions to liquid turbine meters, as shown in Figure A.2, have been successfully used for
particular sizes and applications (i.e., sizes smaller than 4 inches operating at higher flow
rates or pressures). Typically, these designs provide low torque at similar flow rates and
pressure and cannot drive mechanical readout devices or instruments.
A.2
Dual-Rotor Designs
Schematics of various dual-rotor turbine meters are shown in Figures AJ, A.4, A.5 and A.6.
The primary rotor or main metering rotor of each of these designs is basically the same as that
of a single-rotor turbine meter as shown in Figure A.l. The blades ofthe primary rotor will
typically have pitch angles in the range of 30 to 60 degrees. This rotor may have an output
drive for a mechanical register or for an accessory device.
A-2
Readout
Body
Inlet
Main
Rotor
Master Rotor
or Proving Rotor
Annular
Passage
Outlet
Electronic
Pickups
Flow Guides
Readout
Body
Main
Rotor
Inlet
Sensing
/ ' Rotor
Outlet
Electronic
Pickups
A-3
Main
Rotor
Body
Secondary
Rotor
Inlet
Outlet
Electronic
Pickups
Body
Secondary
or Slave Rotor
Main
Rotor
Outlet
Inlet
Electronic
Pickups
l....-__
Piggy-Back
Bearing
Arrangement
A-4
.2.2
The secondary rotors are downstream of the main rotors in Figures A.3, A.4, A.S and A6.
They may be separated from the primary rotors and isolated from them by flow conditioners
between the two rotors (Figures A.3 and A.6). Sorne designs provide for fluid coupling of the
secondary rotor to the primary rotor by positioning the rotors in close proxirnity to each other
(Figures A.4 and A.S). In either case, rotation of the secondary rotor may be in the same or
opposite direction as that of the primary rotor. Typically, the secondary rotor operates at a
lower speed than the primary rotor in order to extend its service life and to differentiate the
measurements of the two rotors for checking purposes.
.2.3
The secondary rotor is provided for checking amI/ or improving the measurement integrity of
the primary rotor under various flow and metering conditions. In sorne dual-rotor turbine
meters, the secondary rotor can provide measurement adjustments to improve the output error
of the primary rotor and provide diagnostics under certain operating conditions.
A.3
Electronic pulse outputs corresponding to the speed of the rotors in dual-rotor turbine meters are
provided by sensors that detect the passage of individual turbine blades, spaces in chopper disks
or the teeth of gears that are driven by the rotors. These signals are fed to a manufacturer's
electronic accessory device or to a user device with appropriate algorithms that ca1culates and
compares volumes from both rotors, ancl/or perfonns diagnostics.
A-5
APPENDIX B
The turbine meter is a velocity-measuring device. It depends upon the flow of gas to cause the
meter rotor to tum at a speed proportional to the flow rateo Rotor revolutions are counted
mechanically or electrically and can be converted to a continuously totalized volumetric
registration. Since the registered volume is at flowing pressure and temperature conditions, it
must be corrected to the specified base conditions for accounting purposes. The register of the
turbine meter indicates volume at flowing conditions so this value needs to be corrected to the
base conditions.
B.1.1
The subscripts "b" denoting base conditions, ''J' denoting flowing conditions and "r"
denoting rated conditions are used in this appendix.
The basic gas law relationship is expressed as follows:
(P) (V) = (ZrJ (N) (R) (T J
F or flowing conditions
Eq. (B.1)
Eq. (B.2)
or
where
P = Absolute pressure
V = Volume
Z = Compressibility
N = Number of moles of gas
T = Absolute temperature
R = Universal gas constant
Since R is a constant for the gas regardless of pressure and temperature, and for the same
number of moles of gas (N), the two equations can be combined to yield:
Vb
= V,
T,) Z,)
B-l
Eq. (B.3)
B.1.2
Q = V
Eg. (B.4)
t
where
B.1.3
= Time
Eg. (B.5)
B.1.4
Pressure Multiplier
P
Eg. (B.6)
p = Pg + Pa
where
Pb
In instances where the atmospheric pressure value is not defined by Federal Energy
Regulatory Commission Tariff or contract terms, atmospheric pressure can be determined
using the following eguations which are based on the National Oceanic and Atmospheric
Administration publication, U. S. Standard Atmosphere, 1976 (Reference 11).
English units
p"
where
Elevation y2554
Eg. (B.7)
SI units
Pa = 10 1.325 x (1 - 0.00002256 x Elevation y2554
where
B-2
Eg. (B.8)
B.1.5
Temperature Multiplier
_ Tb
Temperatur Multiplie
where
Tf
Tf
Absolute temperature:
B.1.6
R = F + 459.67, or
K = oC + 273.15
Compressibility Multiplier
Compressibilit Multiplie
where
Eq. (B. 9)
Zb
Zf
=Z
Eq. (B.10)
Zf
Compressibility at base conditions
Compressibility at flowing conditions
The compressibility multiplier can be evaluated from the supercompressibility factor Fpv , as
follows:
Z b =(F )2
Z
pv
r
Eq. (B.ll)
Compressibility values may be determined from the latest edition of AGA Report No. 8
Reference 1), or as specified in contracts or tariffs, or as mutually agreed upon by both
parties.
B .1.7
B.l.7.1
Turbine meters are generally designed for a maximum flow rate in order to not exceed a
certain rotor speed. This maximum flow rate remains the same (unless stated otherwise
by the manufacturer) for all pressures within the operating range.
Eq. (B.12)
The maximum flow rate at base conditions Qbmax can be expressed as:
B-3
The minimum flow rate (or minimum capacity rating) for a turbine meter is the lowest flow rate
at which the meter will operate within the specified error limit. Generally the minimum flow rate
depends on the magnitude ofnon-fluid drag and the density ofthe measured gas.
The minimum flow rate at base conditions is:
Eq. (B.14)
where:
The range of operating flows for accurate measurement increases approximately as the square
~P
Generally, the rated temperature and pressure are close to the base temperature and pressure. In
this case:
Eq. (B.15)
Eq (B.l6)
Frequently the temperature and compressibility ratios are close to unity and can be neglected for
purposes of approxmaton. The operating range of the gas turbne meter is the flow range over
whch the meter wll operate wthn its specfied performance. In general, the turbne meter range
wll vary drectly wth the square root ofthe gas density. As the density ncreases, the lnearity of
the meter wll be extended to a lower flow rate whle the upper lmit remans fixed by the desgn
consideration stated aboye. Thus:
QjmCLT
'l' = R angeabllty
-=
Qjmin
QbmaT
Qbmm
QrmCLT
Qrmm
B-4
() (;) [i J [~r J
r
Eq. (B.17)
B.2
Mass flow measurement can be employed to arrive at base volume (VbJ or base volume flow rate
(Qb) through the use of a turbine meter and densitometer or calculation from compositional
analysis. The mass or mas s rate of flow is:
M= (V) (PI)
where
VI
PI
and
where
Q ro
Q
Q
P
(Q) (p)
ro =
Eg.(B.18)
Eg. (B.19)
Since the mass or mas s rate of flow at flowing conditions eguaIs the mass at base conditions it
can be stated that:
or,
( V, )( Pb ) =(Vr ) (P r )
Eq. (B.20)
Eq. (B.21)
(Qb) = (Qr)!!..L
Pb
Eq. (B.22)
The aboye eguations show that the base volume (VbJ or base volume flow rate (Qb) can be
calculated by knowing the density of the fluid at both flowing and base conditions without the
need to measure the flowing pressure (p) or the flowing temperature (T) and calculating the
compressibility multiplier.
B-5
APPENDIXC
Computing Flow
C.I
When computing total uncorrected volume from the turbine meter register, two register readings
are taken over a period of time as defined by the contract; e.g., one reading at the first of the
month and the second reading at the end of the month. The first reading is subtracted from the
second to obtain the uncorrected volume measured during the month. Ifthe smallest unit of
volume that can be read from the register is greater than 1 cubic foot; e.g., 10,100, l,000, etc.,
then the difference ofthe two readings is multiplied by the smallest volume unit shown on the
register.
C.2
Electronic Computation
Electronic outputs from meters may be applied to computers in conjunction with temperature and
pressure transducers to obtain volumes for billing aml/or telemetering at base conditions.
C.3
These instruments apply a pressure, or combined pressure and compressibility factor, to the
metered gas volume correcting it to base pressure. An additionaI mechanism may aIso apply a
temperature factor, thus providing registration at base conditions.
C-I
APPENDIX D
D.l
Change Gears
Master Meter
Test Meter
Test Meter
Test Meter
Point
"As-Found"
"As-Left"
% ofQmax
0/0 of Qmax
% Error
10.025
10.00
-0.25
-0.49
20.000
20.00
0.00
-0.24
49.875
50.00
0.25
0.01
74.750
75.00
0.33
0.09
99.650
100.00
0.35
0.11
D- 1
1.00
C 0.80
o
:; 0.60
...
'iii 0.40
'0,
~ 0.20
.!el) 0.00
:2-0.20
.....,
...5- 0.40
U-0.60
c#! -0.80
-1.00
------ ------
----------
-""'\..
10
20
30
?f~
fll"'"
..
40
50
60
70
80
90
100
Flowrate (% Qmax)
D.2
K-Factor(s)
Turbine meter(s) K-factor(s), in pulses/volume units, is established at the time of calibration. Kfactors are used to convert pulses, accumulated from the electronic output of a meter, into
indicated volume units. (See equation D.1.)
where:
= _P_u_l_s_es-=.f_
(K - factor)
Eq. (D.1)
Vf
This calculation is usually accomplished in an electronic accessory device by dividing the pulses,
accumulated over a time period, by the K-factor. Note that, there may be unique K-factors for
each electronic pulse output ofthe meter. Also, there may be different K-factors associated with
D-2
specific flow rates as detennined by calibration. The manufacturer or calibration facility will
provide the K-factor or K-factors for the meter electronic output(s). These values must be
properly entered into an accessory electronic device in order to produce the correct registration
volume(s) from the meter electronic pulse output(s).
Table D.2 shows an example ca1culation of a single K-factor, related to change gear and intemal
gear ratio of a turbine meter. Also shown is a K-factor that is determined as an average of
individual K-factors from five flow rates of a calibration of a turbine meter that has only an
electronic output.
..
K-factor
(Average of
5 calibration values1
103.5303 pulses/cu ft
%
Test flow rate
Test flow rate
Test flow rate
Test flow rate
Test flow rate
No.
No.
No.
No.
No.
1
2
3
4
5
Qmax
10
25
50
75
100
10000
10000
10000
10000
10000
pulses
collected
from
Test Mtr
/
/
/
/
/
96.9650 = 103.1300
96.7227 = 103.3883
96.4810 = 103.6474
96.4004 = 103.7340
96.3840 = 103.7517 pulses/cu ft
cubic ft
Volume*
of
Ref. Mtr
*Volumes corrected to P & T
conditions ofthe test meter.
There may be a single K-factor for use with all flow rates over the operating range of the meter.
A single K-factor may be based on change gear ca1culations, as in Table D.2, or upon arithmetic
average of individual K-factors determined from calibration at various flow rates, or the single Kfactor may be weighted for an operating flow rate range. Table D.3 shows a single K-factor
ca1culated from change gears and an example of an "as-found" meter test.
D-3
Test Meter
Test Meter
Point
RAs-Found"
Kfactor
% Qmax
% Error
(pulses/cubic foot)
10.00
-0.25
20.00
0.00
103.3883
103.3883
50.00
0.25
103.3883
75.00
0.33
103.3883
100.00
0.35
103.3883
Note: The single K-factor shown here in Table D.3 is based on high-frequency pulse
output from a sensor on a rotor shaft (see example 1 in Table D.2).
A single K-factor for all flow rates, or individual K-factors for each calibration flow rates, may be
entered into an electronic accessory device or flow computer for use in converting turbine meter
output pulses to volume. Table D.4 shows an example of individual K-factors for each flow rate
of the test meter. These individual K-factors have be en adjusted by the "as-found" percent errors
to bring each flow rate test point to zero error.
Individual Kfactors
"As-Found"
(single value)
% Error
Jpulses/cubic foot}
0.9975
(pulses/cubic foot)
103.1298
1.0000
103.3883
Test
Test Meter
Point
1
-0.25
0.00
103.3883
103.3883
0.25
103.3883
1.0025
103.6468
0.33
103.3883
1.0033
103.7295
0.35
103.3883
1.0035
103.7502
Note that the multiplier values shown in Table D.4 are derived from the as-found meter error and are used
to calculate the individual K-factors. These muItiplier values are equivalent to the (l/meter factor) values
shown in Section D.3. Also, note that the techniques of curve fitting or linearization, as shown in Section
D.3 for meter factor, may be applied to the individual K-factors, as weIl, for use within an accessory device
or flow computer.
D.3
Meter Factor
A calibration facility may provide meter factors in addition to or instead of percent error values
for a meter. These may apply either to the mechanical or to the electronic pulse output(s) of a
meter. Meter factors are non-dimensional numeric multipliers. The meter factor is the ratio of the
reference volume to the indicated test meter volume, for a particular flow rateo Volumes for both
test meter and the reference must be at the same conditions. The meter output is multiplied by the
meter factor to adjust the meter output, in an effort to eliminate known errors. Table D.5 shows
meter factors for the example test meter.
D-4
Test Meter
Master Meter
Test Meter
Meter faclors
%Qm3.
% Qm
10.025
10.000
1.0025
20.000
20.000
1.0000
49.875
50.000
0.9975
74.750
75.000
0.9967
99.650
100.000
0.9965
Measurement can be' improved by programming a flow computer to curve fit or linearize the
meter factor calibration curve. Multi-point linearization or polynomial curve-fitting techniques
may be used to apply the meter factors across the operating flow rate range. Tables D.6 through
D.8 and Figures D.2 through D.6 show examples ofvarious curve-fitting techniques: AIso shown
are examples of the relative percent errors that may result from the imperfections inherent to each
technique.
TABLE D.6. EXAMPLE - POLYNOMIAL CURVE FIT AND ERRORS
Qi
Flow
Rate
%
Test Meter
Adjusted K-factors
"As-found"
Meter factors
Meter factors
% Error
Resulting % Error
10
-0.25
1.0025
1.0021
103.1716
-0.04
20
0.00
1.0000
1.0006
103.3263
0.06
50
0.25
0.9975
0.9975
103.6474
-0.00
75
0.33
0.9967
0.9964
103.7618
-0.03
100
0.35
0.9965
0.9966
103.7410
0.01
Qma.
1.0050
1.0040
Meter Factor = 108E -06(O/OQm axf - 0.000 179(O/OQm ax) + 10037399
1.0030
...o
-...
u
1.0010
LL
1.0000
IV
1.0020
"
G)
0.9990
0.9980
~
~
0.9970
0.9960
0.9950
20
f--
G)
:::
--
.......
40
60
F10wrate (o/lOmax)
D-5
80
100
1.0050
1.0040
1.0030
'- 1.0020
o
-O 1.0010
IU
IJ..
1.0000
'Q)
a 0.9990
~
'" "'-
"
.~
0.9980
0.9970
0.9960
0.9950
r--
20
--..-
40
60
Flowrate (%Qmax)
10
80
Flow Rate
Meter factors
Linear interpolation
Curve Fit
% Qmax
10
1.0025
1.0025
103.1305
Resulting % Error
0.00
1.0013
103.2541
0.04
1.0000
103.3883
0.00
0.9988
103.5125
0.02
0.9975
103.6474
0.00
0.9971
103.6907
0.01
0.9967
103.7306
0.00
0.9966
103.7410
0.00
0.9965
103.7514
Qi
15
20
1.0000
35
0.9975
50
62.5
0.9967
75
87.5
100
0.9965
Adjusted K-factors
Linear Interpolation
0.00
'2
o 0.60
~<n
0.40
.~ 0.20
.A.-..._.8 .._....
0.00
2:l
W
Ct~
~-020
..~ h~.
~"As-Found"
~-0.40
w
~
tvBter Error
-0.60
--fr-
-0.80
-100
O
10
20
30
40
50
60
70
80
D-6
90
100
Figure DA shows comparison plots of polynomial to linear interpolated curve fitting for the
example test meter.
For "piecewise" curve fit linearization, the meter factors are applied in steps that are constant
over regions offlow rates. An example is shown in Table D.8 and in Figure D.5.
QI
Adjusted Kfactor
Test Meter
(with Meter factor)
Flow Rate
Meter factors
%Qmu
1.0025
103.1305
Resulting % Error
-0.25
1.0025
103.1305
0.00
14.9999
1.0025
15
1.0000
103.1305
103.3883
-0.09
1.0000
103.3883
0.00
1.0000
103.3883
0.14
0.9975
103.6474
-0.11
0.9975
103.6474
0.00
0.9975
103.6474
0.06
-0.03
10
1.0025
20
1.0000
34.9999
35
50
0.9975
62.4999
62.5
75
0.9967
87.4999
87.5
100
0.9965
105
0.16
0.9967
103.7306
0.9967
103.7306
0.00
0.9967
103.7306
0.01
0.9965
103.7514
-0.01
0.9965
103.7514
0.00
0.9965
103.7514
0.00
--+-Meter Factors
1.0030
"\
1.0020
Lo.
.....oo 1.0010
ns
u.
1.0000
Lo.
.....Q) 0.9990
Q)
" """ --
0.9980
0.9970
0.9960
1-
-.
0.9950
20
40
60
Flowrate (%Qmax)
80
100
The resulting error for the application of piecewise curve fitting to the example data shows step
changes for flow rates where the meter factor changes can be seen in Figure D.6.
Turbine Meter Calibration
Piecewise Curve Fit - Example
1.00
0.80
eo
OBO
.~ DAD
t'
.0,
~
020
.
/-/~
ID 0.00
ID
~-0.20
e.....-040
/
o/
----
_._-Q
...
[]!-----
'if. -0.60
-8-
-0.80
-1.00
10
3D
20
40
50
Flowrate (% Orrax)
60
70
80
90
100
DA
The final meter factor may be the arithrnetic average or mean of the meter factors, (see Section
D.3), over the range of flow rates at which the meter is to be used. The final meter factor may
also be weighted more heavily toward the individual meter factors at the higher flow rates at
which the meter is to be used. The final meter factor is a single multiplier value that can be
applied either to the mechanical output or to the single or individual electronic outputs. The final
meter factor may be applied to compensate for systematic calibration offsets. It is applied to any
volume output of the meter by multiplying by the final meter factor. It is applied to any Kfactor(s) by dividing by the final meter factor. Table D.9 and Figures D.7 and D.8 illustrate an
example of percent error performance for a test meter before and after the final meter factor is
applied. Both an arithmetic average and a flow weighted final meter factor are shown.
TABLE D.9. EXAMPLE - FINAL METER FACTOR
Qi
Test Meter
Flow
Rate
"As-found"
Meter
% Qma.
% Error
factors
Final meter
factor
(Arithmetic
average)
K-factor
Test Meter
(103.38831
Final meter
factor)
pulses I cubic
foot
Average final
meter factor
applied
% Error
Final meter
factor
Flow weighted
(see Note 1)
K-factor
Test Meter
Flow
(103.38831
weighted
Final meter final meter
factor applied
factor)
pulses I
% Error
cubic foot
10
-0.25
1.0025
0.9986
0.9975
0.00
1.0000
0.9986
103.5332
103.5332
-0.39
20
-0.14
0.9975
103.6474
103.6474
-0.25
50
0.25
0.9975
0.9986
103.5332
0.11
0.9975
103.6474
0.00
0.9986
103.5332
0.19
0.9975
103.6474
0.08
103.6474
0.10
75
100
0.33
0.35
0.9967
0.9965
0.9986
103.5332
0.21
0.9975
-0.50
Note 1: In this example, the meter factor has been weighted by normalizing percent error at 50 percent Qma.
to zero. Different flow weighting methods may be used for other applications.
D-8
1.0020
o
o 1.0010
ro
u.. 1.0000
Lo.
Lo.
Q)
Q)
-+-Meler Factors
-a-A-..erage Final Meter Factor
---s!r-Weighted Final Meter Factor
0.9990
"'-~
...............
0.9980
---...
0.9970
0.9960
-----
0.9950
100
80
40
60
Flowrate (%Qmax)
20
.~
......
060
ro
b 0.40
'5J
------
If)
0.20
l...
2 000
......
/~
~~
~-0.20
l...
t-0.40
tr
'#. -060
f--
t--
-0.80
-1.00
O
10
20
30
40
50
FJow rate (% Qmax)
60
70
80
90
Figure D.8. Example - Shifts oC Averaged and Weighted Final Meter Factors
D-9
100
D.S
For dual-rotor meters, there may be individual and unique rotor factors, in units ofpulses/volume
unit, associated with each rotor's electronic pulse output. Examples of rotor factors for the
primary and secondary rotors of a dual-rotor turbine meter areRotor factor (for primary rotor)
= 95.2000 pulses/cubic foot
Rotor factor (for secondary rotor) = 143.4000 pulses/cubic foot
Rotor factors are to be used with the proprietary algorithms as directed by the dual rotor turbine
meter manufacturero Note that the rotor factor(s) should not be confused with the K-factor(s).
Even though rotor factors have the same units as K-factors and may even be similar in value, their
applications are very different and they should not be used interchangeably. For example, the Kfactor for the primary rotor ofthe meter aboye may be 103.3883 pulses/cubic foot.
Users should consult with the manufacturer of dual-rotor turbine meters for specific instructions
on the correct use ofrotor factors and/or K-factors.
D-IO
APPENDIXE
Calibration Guidelines
E.l
Reference 3 shows that meters should be calibrated under conditions that are similar to those
expected in service. For many meters and service conditions, calibrating over the expected range
of Reynolds numbers and flow rates will provide the best results. Reynolds numbers may be
calculated from equation 6.3. (Repeated here for convenience.)
Re
where
Re =
Q =
D =
p =
p =
4 (Q) (p) /
7r (D)
(jJ)
Eq. (E.I)
Reynolds number
V olumetric flow rate
Meter diameter
Density
Absolute viscosity
The subscripts "j', denoting field conditions, and "t", denoting test conditions, are used in this
appendix.
To match both Reynolds numbers and flow rates requires the relationship: Re, = Re, Q, = Qr,
and:
Eq. (E.2)
p/jJ = p,/p,
From this relationship, the density of the test gas can be calculated. From the test density, the
necessary pressure of the test gas can be determined.
The resulting conditions will be such that Reynolds numbers found in the field will be matched in
the test facility. The K-factors measured at each test flow rate can then be prograrnrned directly
into a flow computer for linearization purposes or used to calculate the meter's final meter factor.
K -factors determined in this manner can be used directly for measurement of natural gas
regardless of the calibration gas used. In the event that natural gas is used as the test medium, the
densities and visco sities between field conditions and test conditions need be adjusted for
temperature and composition differences only. In the event an altemative test gas is used, the
properties of the test gas will determine its density and viscosity characteristics. In either case, the
pressure of the test gas is adjusted to create equivalent conditions of Reynolds numbers and flow
rateo An example calculation using air as the test gas is provided in E.5.1.
E-I
E.2
The Reynolds matching approach described in Section E.l differs from a commonly used method
of establishing test conditions for turbine meters with alternative gases. Traditionally, the practice
has be en to match both in-service pressure and flow rates during tests. When the test medium is a
gas ofhigh relative density, this practice results in density and, in sorne cases, Reynolds
conditions in the meter that are significantly higher than those in service. Testing with heavy
gases at matched pressure and flow rates results in high momentum at the rotor. This means that
the rotor thrust bearing could be overloaded, with detrimental effects on accuracy and meter life.
The manufacturer should be consulted before conducting calibrations in this manner.
E.3
Reference 3 shows that sorne turbine meters can exhibit density-related performance effects. In
the research described in Reference 3, this effect was observed only at low pressures and low
flow rates in a few meters (Refer to Section 6.3.1.2 for a discussion). It is possible to use a similar
approach to that described in Section E.l to assess this aspect of meter performance. Density and
Reynolds number would be matched instead of flow rate and Reynolds number. Using the same
notation as aboye, equating the field and test Reynolds number and density, and dropping the
constant tenns leads to the relationship:
Eq. (E.3)
As can be seen from the aboye, the test flow rate range will differ from that in the field by the
ratio of the viscosities involved. The example calculation shown below also illustrates this option.
Note that when using density and Reynolds number matching in altemative calibration gases, it
may not be possible to calibrate the meter over its designed flow rate range. This limitation is
further discussed in the example in E.5.2.
E.4
E.5
Example Calculations
A four-inch meter is to be installed in 150-psig natural gas pipeline, at flow rates from 1,600 to
16,000 actual cubic feet per hour (acfh). lt is intended to calibrate the meter at a facility that uses
air as the test medium. The properties ofthe natural gas are:
Pressure
Density (PI)
Viscosity (jil)
Range of flow rates
150 psig
0.5191lbn/fe
0.01086 cp
1,600 - 16,000 acfh
E-2
The resulting Reynolds number range is 120,762 to 1,207,602. (Note that it is not necessary to
calculate Reynolds number to use this method. Solving equation E.2 or E.3 is sufficicnt. The
range ofReynolds numbers is shown here for reference purposes only.)
The test gas is air. The viscosity (;..tr) of air at 60 0 P is 0.01774 cp.
E.5.l
Pr
(p) (flt/fl)
Eq. (E.4)
The resulting density ofthe test air (PI) would therefore be 0.8479 lbm/ft3. Air has this density
(at 60 0 P) at a pressure of about 148 psig. Thus, the meter would be calibrated in 148-psig air
at the in-service flow rates and the resulting K-factors applied directIy to correct the meter's
output.
E.5.2
The field gas density to be matched is 0.5191 lbm /ft3. In air, this density results in a test
pressure of about 85 psig. Equation E.3 can be rearranged in the forro of:
Eq. (E.5)
The resulting flow rate range would therefore be 2,614 to 26,136 acfh.
CAUTION: The resulting upper flow rate is beyond the maximum for this meter. Tests
cannot be conducted aboye the maximum rated flow rate without severe risk of damaging the
meter. With density and Reynolds matching in typical (i.e., heavy) altemative calibration
gases, the field flow rate range can be matched only if it does not approach the maximum for
a particular meter size. This limitation on the range of calibrated flow rates will reduce the
range of calibrated Reynolds number.
E-3
APPENDIX F
Testing In-Line
1.
2.
3.
4.
5.
6.
7.
Meter Size
(Inches)
2 and3
4 and 6
8 to 12
F.2
(% Rated Capacity)
45
25
10
F-I
b.
6.
7.
8.
9.
10.
11.
Calculate the as-found freestanding spin time as the average ofthe results ofthe
three trials.
1. Compare the as-found free standing spin time with the manufacturer's
specified spin time.
11. Replace the meter's bearings or module ifthe as-found freestanding spin
time is not within 90% ofthe manufacturer's specified spin time.
lll. Replacement of a dual-rotor module requires a change in the standard
electronic readout prograrnmable read only memory (PROM).
Perform an as-Ieft spin time test ifthe as-found spin time test required corrective action.
Compare the as-Ieft freestanding spin time with the manufacturer's specified spin time.
Replace the meter module ifthe as-Ieft time is less than 90% ofthe manufacturer's time,
even after corrective action has been taken.
Reinslll the meter module in the meter runo
Purge and pressurize the meter runo
Control the rate of pressurization to prevent damage to the primary device and to avoid
creating an excessive differential pressure across the turbine meter.
F-2
REFERENCE LIST
Reference 1: A.G.A. Transmission Measurement Cornmittee Report No. 8, Compressibility
Factors ofNatural Gas and Other Related Hydrocarbon Gases, American Gas Association,
Washington, D.C., November 1992.
Reference 2: George, D. L., GRI Topical Report GRI-OI-0226, Metering Research Facility
Program: Turbine Meter Research in Support olthe Planned AGA Report No. 7 Revision, Gas
Technology Institute , Des Plaines, Illinois, January 2003.
Reference 3: George, D. L., GRI Topical Report GRI-03-0172, Metering Research Facility
Program: Effects 01Line Pressure and Gas Density on Turbine Meter Measurement Accuracy at
Conditionsfrom Atmospheric Air to 700 psig in Natural Gas, Gas Technology Institute ,Des
Plaines, Illinois, August 2004.
Reference 4: Siebenaler, S. P., and George, D. L., GRI Topical Report GRl-04-0207, Metering
Research Facility Program: Effects ofTurbine Meter Cartridge Changeout on Measurement
Uncertainty, Gas Technology Institute, Des Plaines, Illinois, USA, March 2005.
Reference 5: International Organization of Legal Metrology R-6 General Provisions for Gas
Volume Meters -1989, Intemational Organization ofLegal Metrology, Paris, France, 1989.
Reference 6: George, Fraser, Nored and Tang, Carbon Dioxide as a Test Fluid for Calibration of
Turbine Meters, American Gas Association Spring Conference 2004, American Gas Association,
Washington, D.C., May 2004.
Reference 7: ISO 9951: Measurement ofGas Flow in Closed Conduits - Turbine Meters,
International Organization for Standardization, Geneva, Switzerland, November 1994.
Reference 8: Peace, D.W., 1nstallation EjJects on Turbine Meters, AGA Operations Conference,
May 1995, American Gas Association, Washington, D.e.
Reference 9: Arnberg, B.T., Review ofCritical Flowmeterslor Gas Flow Measurements, ASME
Journal ofBasic Engineering, December 1962.
Reference 10: A.G.A. Transmission Measurement Committee Report No. 3 Orifice Metering of
Natural Gas and Other Re/ated Hydrocarbon Fluids, American Gas Association, Washington,
D.C., Apri12000.
Reference 11: National Oceanic and Atmospheric Administration, U. S. Standard Atmospher, U.
S. Department of Commerce, National Technical Information Service, October 1976.
REF-l
Send to:
Name: ________________________________________________________________
Company: _________________________________________________________
Address: _________________________________________________________
Phone: ____________ Fax: ____________~E-maiI: _ _ _ _ _ _ __
PIe ase Indicate Organization Represented (if any): _______________
1.SectionlParagraph:__________________________________________________
2. Proposal Recommends: (check one):
new text
revised text
o deleted text
s-