Turbine Meter Training

Presented by Kevin Ehman 2008.10.08

Common Types of Gas Meters

TypesofGasMeters

Positive Displacement Meters

Inferential Meters

Material quoted in part from Sensus publication

Common Positive Displacement Meters

Positive Displacement Meters

Diaphragm Meters

Rotary Meters

Material quoted in part from Sensus publication

Common Inferential Meters

Inferential Meters

Turbine Meters

Orifice Meters

Ultrasonic Meters

Material quoted in part from Sensus publication

Calculating Flow Rate Measured by an Inferential Meter

Q=VxA
Where: Q=FlowRateinCFH V=GasVelocity V G V l it A=FlowArea

InferredFlowRate=Aflowratederivedindirectlyfromevidence (e.g.velocitythroughaknownarea)

Material quoted in part from Sensus publication

Advantages and Disadvantages of Turbine Meter

TurbineMeters

Advantages
GoodRangeability Compact,EasytoInstall DirectVolumeReadout NoPressurePulsations WideVarietyofReadouts Wide Variety of Readouts Willnotshutoffgasflow

Disadvantages
LimitedLowFlow Susceptibletomechanical wear Affectedbypulsatingflow

Material quoted in part from Sensus publication

Lets Start with Explaining a Few Key Definitions

Error Kfactor Meterfactor MAOP Pressuredrop Qmax Qmin Rangeability

Thedifferentbetweenameasurementanditstruevalue. Anumberbywhichthemeter'soutputpulsesaremultipliedto determinetheflowvolumethroughthemeter. Anumberbywhichtheresultofameasurementismultipliedto compensateforsystematicerror. Maximumallowableoperatingpressure Maximum allowable operating pressure Thepermanentlossofpressureacrossthemeter Themaximumgasflowratethroughthemeterthatcanbe measuredwithinthespecifiedperformancerequirement. Theminimumgasflowratethroughthemeterthatcanbe measuredwithinthespecifiedperformancerequirement. Theratioofthemaximumtominimumflowratesoverwhichthe metermeetsspecifiedperformancerequirement.Rangeabilityis alsoknownastheturndownratio.

Material quoted in part from AGA publication

Conversion to Base Conditions

Conversion to Base Conditions Baseconditionsisasetofgiventemperatureandpressurewhichdescribesthe physicalstateofgasinflowmeasurement. physical state of gas in flow measurement. Baseconditionsaredefinedjurisdictionally: InCanada InUSA Pb =101.325kPa,Tb =15C Pb =14.73psi,Tb =60F

The Ideal Gas Law

TheIdealGasLaw
Conversionofthemeasuredlinevolumetobasevolumereliesontheequation ofstatefortheparticulargas. ( ) PV=nRT(1) Inthisequation Pistheabsolutepressure Visthevolume n isthenumberofmolesofthegas Ristheuniversalgasconstantandequals8.31451J/molK. Tisthethermodynamic(orabsolute)temperature Thisequationisvalidfornmolesofgasanddescribestherelationbetween thevolumeV,the(absolute)pressurePandthe(absolute)temperatureT.
9

Gas Turbine Meter - a Well Established Technology

Reinhard Woltman was generally credited to be the inventor of the turbine meter in 1790 for measuring water fl i t flow. Modern gas turbine meters are very accurate and repeatable over a wide flow range. These meters have a very extensive installed base in the natural gas industry worldwide.

Sectional view of a turbine meter

Material quoted in part from Sensus publication

Cut-out View of a Turbine Meter

Flowvolumeregister Changegears IndexAssembly

Mainrotor

Lubricationfitting

Encoder/sensor Nosecone Topplate

Meterbody

Conditioningfins

Material quoted in part from Sensus publication

11

Cut-out View of another Turbine Meter

IndexAssembly

Flowvolumeregister

Conditioningplate

Coupling

Mainrotor

Mainshaft Meterbody Bearingblock

Material quoted in part from iMeter publication

12

Turbine Meter Operating at Various Pressure Ranges

Turbinemetersoperatingatatmosphericpressureshowa verynonlinearperformancecurve

50psig 175psig 720psig 1440psig

Turbinemetersoperatinginahighpressurelinedisplaysa muchmorelinearandpredictablecharacteristic

Material quoted in part from Sensus publication

13

Principle of Turbine Meters

TheLawofConservationofEnergy KineticEnergy=DynamicEnergyofMassinMotion KE=1/2 MV2

Where:KE=Kineticenergyofthemovinggas molecules M Mass of gas molecules M =Massofgasmolecules V=Velocityofgasmolecules

Velocity = V

Mass of gas molecules = M

Inanturbinemeter,aportionofthelinearkineticenergyofthemovinggasmoleculesis convertedintorotationalenergyoftherotor

Principle of Turbine Meters

Analysis of an Ideal Rotor

r Q A

is the average of the rotor radius is the volume flow rate is the annular flow area is the blade angle are the gas velocities at point (1) and (2) is the fluid velocity relative to the rotor blades is the ideal angular velocity

V 1, V 2 1, 2

(1)

(2)

The angular velocity of the rotor is proportional to the volume flow rate

Q i

(3)

Material quoted in part from Sensus publication

Turbine Meter Index Assembly

IndexAssembly
Change gears

Magnet reed sensor

Signal terminals

Theindexassemblytypicallyhousesa readoutregisterofflowvolumeand oneormoresetsofencoderdiscand sensorforgeneratingflowoutput pulsesforelectronicmeasurement systems.

Encoder disc

Sensor

Magnetic coupler

16

Dual-Rotor Turbine Meter

Theprimaryrotorofadualrotorturbinemeter isbasicallythesameasthatofasinglerotor design.Asecondrotorisaddedforchecking and/orimprovingthemeasurementintegrityof theprimaryrotorundervariousflowconditions.

CutoutviewofanAutoAdjustmeter

AdjustedVolumeatInitial Calibration BasicAdjustmentPrinciple OperatingChangesin RetardingTorque Retarding Torque SelfCheckingFeature

Material quoted in part from Sensus publication

Construction of a Turbine Meters

Material quoted in part from Sensus publication

Dual-Rotor Turbine Meter

Mainrotor Sensingrotor

Themainrotoriscalibratedtoregister110% oftheactualflowpassingthroughthemeter. Thesensingrotoriscalibratedtoregister 10%oftheactualflow.Bydesignofthetwo rotorsandtheirplacementinthemeter body,theflowerrorexperiencedbythe sensingrotormatchesthatofthemainrotor. sensing rotor matches that of the main rotor TheAdjustedVolumethereforeprovidesa veryaccurateaccountofthetrueflow.

Vadjusted = Vmain - Vsensing

CutoutdetailsofanAutoAdjust j dualrotorhousing Thesensingrotorcorrectionfactoris providedbyfactorycalibration.

Material quoted in part from Sensus publication

Dual-Rotor Turbine Meter

TheAutoAdjustTurbineMeterEquations:

V sensing V sensing A = 100 = 100 V adjusted V main - V sensing

(1)

V sensing = 100 A V main V sensing

Where: =volumebymainrotor Vmain Vsensing =volumebysensingrotor Vadjusted =adjustedvolume =averagevalueofthefactorysensingrotor%adjustment A=%deviationinfieldoperationfromfactorycalibration

(2)

Material quoted in part from Sensus publication

Dual-Rotor Turbine Meter

TheAutoAdjustselfcheckingIndicator:

V sensing = 100 A V main V sensing

TheparameterA(deltaA)isaselfcheckingindicatoroftheperformance j j ofanautoadjustturbinemeter.Itshowstheamountofadjustmentthe meterismaking,therebywarningtheuserofmeterorflowconditioning problems.

Material quoted in part from Sensus publication

Performance Curve of an Ideal Gas Turbine Meter

1.0

0.5 ERROR (%) R

0.5

1.0 0 25 50 75 100 125 CAPACITY (%Qmax)

AnidealturbinemeterhasaflaterrorcurveextendingfromQmin toQmax

Material quoted in part from iMeter publication

Performance curve of a Real Gas Turbine Meter

1.0

Causesfornonidealturbine meterbehaviours:
Dirtygas yg

0.5 Error (% %)

Mechanicalfriction Pertubations Densityeffect

-0.5

-1.0 0

Reynoldseffect
25 50 75 100 125 Capacity (%Qmax)

Typicalperformancecurveofaturbinemeter

Material quoted in part from iMeter publication

Of course Nothing is Perfect

1.0

0.5 Error (% %)
Ideal turbine meter

Real turbine meter

-0.5

-1.0 0

25

50

75

100

125

Capacity (%Qmax)

Performancecurveofarealgasturbinemeter

Material quoted in part from iMeter publication

Of course Nothing is Perfect

The accuracy of a gas turbine meter is influenced by mechanical friction at low flow rate y and Reynolds number at high flow rate. Recent research has shown that relatively large measurement errors can occur if a turbine meter was not calibrated at or near its operating pressure.

Gas turbine meter

Impact of Dirt on Turbine Meter

Dirtaccumulatedontherotorbladeshasatendencytospeed upaturbinemeter,thusresultinginoverestimatedflow volume.

Dirtonthe rotorblades

1% Error -1% Flow rate Q

Material quoted in part from iMeter publication

Impact of Dirt on Turbine Meter

Dirtaccumulatedinbearingsslowsdownaturbinemeter, thereforeresultsinunderestimatedflowvolume.

1% Good bearings Error -1%

Flow rate Q

Damaged bearings

Material quoted in part from iMeter publication

27

Impact of Damaged Bearings

2 0
ERROR (%)

2 4 6 8

Ataconstantinlet pressure,increasein mechanicalfriction mechanical friction duetobearingwear hasmoresignificant effectonLOWFLOW accuracy.

NEW INOPERATION 75 100

10
0
25 50

CAPACITY (% Q Qmax) )

Damagedbearingsslowdownaturbinemeterconsiderably

Material quoted in part from iMeter publication

Typical Turbine Meter Spin Time Decay Curve

Thespintimeofaturbinemeterisaverygoodindicatorofitscondition

Material quoted in part from AGA publication

Spin Time Effect on Proof

Percent Error %

Flow Rate SCFH x 103

EffectofspintimeontheproofofaT35MarkIIturbinemeter
Quote from Sensus Turbo-Meter Installation & Maintenance Manual MM-1070 R9

Lubricating a Turbine Meter

Material quoted in part from iMeter publication

31

Lubricating a Turbine Meter

TurboMeterOil

AlemiteFitting

Material quoted in part from iMeter publication

Single K-factor Representation

AsingleKfactorisoftenusedtoexpressthecalibrationofaturbinemeter.Itis A i l Kf i f d h lib i f bi I i simplebutdoesnotrepresenttheoperatingcharacteristicsofthemeter throughouttheentireflowrange.

Material quoted in part from AGA publication

Meter Factors

Material quoted in part from AGA publication

Flow Weighted K-factor and Meter Factor

Material quoted in part from AGA publication

Typical Turbine Meter K-factors by Calibration

Material quoted in part from AGA publication

Shifting Error Curve by Change Gear

Material quoted in part from AGA publication

Fine Tuning K-Factor with Change Gear

Change Gear = 73/47

Calibrationadjustmentofthemechanicaloutputofaturbine meteristypicallyaccomplishedbychoosinganappropriateset ofchangegears.

Linearization

Linearizationofflowmeter , p Iftheerrorofaflowmeterisknown,itcanbecorrectedfor.Someflowcomputers havetheabilitytocarryoutthiscorrection.Firstthecorrectiondataresultingfrom calibrationarefedintotheinstrument.Next,theappropriatecorrectionfactorat theparticularflowrateisdeterminedandapplied.Theresultwillbeperfectly linear.

39

Typical Turbine Meter Calibration Certificate

AGA-7

Material quoted in part from AGA publication

AGA-7

Material quoted in part from AGA publication

AGA-7

Material quoted in part from AGA publication

AGA -7 General Performance Tolerances

Repeatability: Max peak-to-peak Error: Maximum error: Transition flow rate:

0.2% 0 2% from Qmin to Qmax 1.0% above Qt 1.0% from Qt to Qmax 1.5% from Qmin to Qt Qt not greater than 0.2 Qmax

Material quoted in part from AGA publication

AGA 7 - Installation for In-line Meter

Material quoted in part from AGA publication

AGA 7 - Typical Meter Set Assembly

Material quoted in part from AGA publication

AGA 7 - Short-Coupled Installation

Material quoted in part from AGA publication

AGA 7 - Close-Coupled Installation

Material quoted in part from AGA publication

AGA 7 - Angle-Body Meter Installation

Material quoted in part from AGA publication

Low Level Perturbation

AstraightAGA7compliantmeterrunproducesanuniformflow profilewiththesameflowvelocityacrossthecrosssectionofpipe

Anelboworteeintroducesalowlevelperturbationtotheflow
Material quoted in part from AGA publication

50

Low Level Perturbation

Anadditionaloutofplaneelbowaddsswirltothealreadyunevenflowprofile

Material quoted in part from AGA publication

51

High Level Perturbation

Anupstreamregulatorandoutofplaneelbowcauseahighlevelofswirland jettingatthemeterrun

Material quoted in part from AGA publication

52

HIGH Level Perturbation

Expandingfromasmallerdiameterpipeintoalargeroneintroducesjetting Expanding from a smaller diameter pipe into a larger one introduces jetting whichcannotberemovedbyatubebundleflowstraightener

Additionofanoutofplaneelbowupstreamcompoundstheproblembyadding aswirlcomponenttotheflow
Material quoted in part from AGA publication

53

AGA 7 - Flow Conditioning for Turbine Meter

19tubebundlestraighteningvanes

Flowconditioningplate
Material quoted in part from AGA publication

AGA 7 - Meter- Integrated Flow Conditioner

Material quoted in part from AGA publication

Turbine Meter with Integral Flow Conditioner

Integralconditioningplate typicallyallowsaturbinemeter tobeinstalledinanonideal b i ll d i id l meterrun(e.g.shortmeter run,elbows.)andmaintainits accuracy

Exampleofaturbinemeterwith integralconditioningplate

56

Pressure Loss Across a Turbine Meter

Thepressurelossofaturbine meterisdirectlyproportionalto theflowpressureandspecific gravityandtothesquareofthe flowrate:

Pm Pm G Q 2
Where
Constant Pm and G

Pm = pressure drop across meter Pm = absolute flow pressure G = specific gravity of gas Q = flow rate

Pressure Loss Across a Turbine Meter

45 Rotor Meter Characteristics

Thepressurelossacrossaturbinemeteris directlyproportionaltothelinepressure andspecificgravityandtothesquareofthe flowrate:

Pm Pabs G Q 2
Inwhich Pm isthepressurelossacrossthemeter Pabs istheabsolutelinepressure Gisthespecificgravityofthegas Qistheflowrate

Material quoted in part from iMeter publication

58

AGA 7 - Recommended Blow Down Valve Size

Properlysizedblowdownvalvepreventoverspinning ofturbinemeterduringlinepurgeoperation

Material quoted in part from AGA publication

Effect of Rapid Rate of Pressure Change

Pipeline pressure vs Time

Turbinemetermanufacturers oftenspecifyamaximumrateof pressurechangeallowedfortheir products. Exposuretorapidpressure Exposure to rapid pressure changecancausedamagetothe electronicsensorsinaturbine meter.

P~ 240 psig

T~ 30 sec

Rate of pressure change = P t Where P = maximum pressure change t = time period during which P occurs

Typicalmaximumrateof pressurechangeratingfor pressure change rating for turbinemeter: 100psig/minute

Intermittent Flow Characteristic of Turbine Meter

TurbineMetersdisplaydifferentresponsecharacteristics T bi M t di l diff t h t i ti whilespeedingupandslowingdown.

Underestimated volume on rapidly increasing flow Overestimated volume on rapidly decreasing flow

Flow Rate (ACFH)

Actual flow Flow registered by turbine meter

Time (in minutes)

IntermittentFlowResponseofTurbineMeter
Material quoted in part from iMeter publication

Intermittent Flow Characteristic of Turbine Meter

Duetotheunsymmetricaltransientresponseofturbinemeters,theyare susceptibletooverestimatingtheflowvolumeofpulsatingdevicessuchas susceptible to overestimating the flow volume of pulsating devices such as compressorsandregulators.
Turbine meter can track the rising edge of a pulsating flow

Turbine meter cannot track the falling edge of a pulsating flow because of the inertia of its rotor

Flow Rate (ACFH)

Overestimated volume

Actual flow Flow registered by turbine meter

Time (in minutes)

IntermittentFlowResponseofTurbineMeter
Material quoted in part from iMeter publication

Reynolds Number

D Reynolds Number =
= fluid density = flow velocity D = pipe diameter = fluid viscosity
RecentresearchconductedatCEESIandSwRIonbehalfofAGA Recent research conducted at CEESI and SwRI on behalf of AGA hasdemonstratedthatcommerciallyavailablegasturbine metershavemarkedlydifferentresponsestogivenvolumesof naturalgasatdifferentReynoldsnumber.

63

Turbine Meter Performance vs Reynolds Number

EffectOfFluidAndNonfluidRetardingTorquesOnGasTurbineMeterPerformanceForReynoldsNumberBelow 100,000(Source:InvensysMeteringSystems)

Flow Profiles at Various Reynolds Number

Laminar if Re < 2000 Transient if 2000 < Re < 4000 Turbulent if Re > 4000
Reynolds Number examples: 12 Standard Capacity Meter at 350 psia at 10% of capacity Re = 700,000 at 95% of capacity Re = 6,800,000 o capac ty e

Velocity Profiles in Laminar and Turbulent Pipe Flow

Flow Measurement Engineering Handbook R.W. Miller, McGraw-Hill

Equation of State

TheStateofagas Tocalculatequantityintermsofbaseorstandardvolumeoneneedsto knowthequantityofmatter,e.g.thenumberofmoles,thatoccupies theactualvolumemeasuredunderoperatingconditions. ThisisdonebyusingasuitableEquationofStateforthetypeofgas measuredandbyusingmeasuredpressureandtemperature.

66

Equation of State Composition of Natural Gas Compositionandcompressibility ThecompositionofthegasinfluencestheconstantsintheEquationof State.ThisismostlytranslatedintheCompressibilityfactororZ.

Material quoted in part from AGA publication

67

Elevated Pressure Operation of Turbine Meter

45 Rotor Meter Characteristics

ElevatedPressureOperation

1.MaximumCapacityinSCFHincreasesdirectly asdoestheBoylesLawpressuremultiplier factor. 2.Minimum(LowFlow)Capabilitiesincreases directlywiththesquarerootoftheBoylesLaw pressuremultiplierfactor.

Material quoted in part from iMeter publication

Calculating Rangeability

Rangeability calculation example c

69

Material quoted in part from Sensus publication

Calculating Rangeability

PressureMultiplier =(LinePressure+AverageAtmospheric)/BasePressure*CompressibilityRatio =(500psig+14.48psi)/14.73*1.0863 =37.942 MaximumFlowRate =MeterRating*PressureMultiplier =18,000acfh*37.942 =682,956scfh=683,000scfhfromtable MinimumFlowRate =MeterRating*SquareRootofPressureMultiplier =1200acfh*(37.942)0.5 =7391scfh=7400scfhfromtable Range =Maximum/MinimumFlowRater =683,000/7400=92:1

Material quoted in part from Sensus publication

70

Typical Turbine Meter Installation

HazardousArea

NonhazardousArea

Intrinsically f I t i i ll safe NAMUR sensor or dry contact

Pulse Amplifier

Flow Computer / RTU

Power

Turbine Meter

PulseamplifierconvertingNAMURsignaltoastandard24Vdigitalsignal

71

NAMUR Signal

InductiveSensor

CapacitiveSensor

Supply Voltage = 8 2 VDC 8.2 Source impedance ~ 1 k

Typicalsensorcurrentversussensingdistance

72

Turbine Meter Output Signal Format

Low flow

N NAMUR Detection Lim mits

LowFlow

HighFlow

High flow

NAMUR Signal

Digital Signal

Material quoted in part from iMeter publication

Turbine Meter Pulse Signal Conditioning

Normal turbine meter signal

Turbine meter

NAMUR pulse amplifiers

Incorrect turbine meter signal

Incorrectsupplyvoltageorsourceimpedanceresultsinmissedpulses

Cost of Measurement Error

Turbine Meter Operating at 50 psig

Meter Size Inches 4 Energy Delivered in a 6 year Calibration Cycle * MMBtu 1,271,208 1 271 208 Cost of Energy Delivered * US$ 8,898,458 8 898 458 Cost of 0.5% Measurement Error US$

Turbine Meter Operating at 500 psig

Meter Size Inches 4 Energy Delivered in a 6 year Calibration Cycle * MMBtu 10,990,320 10 990 320 Cost of Energy Delivered * US$ 76,932,238 76 932 238 Cost of 0.5% Measurement Error US$

44,492 44 492 86,732 149,246 223,588 348,054 , 571,641

384,661 384 661 747,921 1,281,828 1,923,306 2,991,684 , , 4,914,990

2,478,052

17,346,361

21,369,172

149,584,204

4,264,180

29,849,258

36,623,671

256,365,699

8 HC

6,388,224

44,717,567

8 HC

54,951,598

384,661,188

12

9,944,389

69,610,722

12

85,476,688

598,336,817

12 HC

16,332,613

114,328,289

12 HC

140,428,286

982,998,005

Note 1: Turbine meters operating at 30% of Qmax average 2. Energy content of natural gas based on 1.0205 MBtu/cu.ft. 3. Cost of energy calculated based on $7.00 USD per MMBtu (including delivery)

Questions ?

Oct. 8 2008

References:

Sensus repair manuals. Sensus Turbine Meter hand book. iMeter Presentation on Turbine Meter Instromet System Handbook AGA Report #7 AGA Report #8

Oct. 8 2008