1、BSI Standards PublicationBS 7965:2013Guide to the selection,installation, operation andcalibration of diagonal pathtransit time ultrasonicflowmeters for industrialgas applicationsPublishing and copyright informationThe BSI copyright notice displayed in this document indicates when the documentwas la
2、st issued. The British Standards Institution 2013Published by BSI Standards Limited 2013ISBN 978 0 580 80061 0ICS 17.120.10The following BSI references relate to the work on this document:Committee reference CPI/30/5Draft for comment 13/30268440 DCPublication historyFirst published April 2000Second
3、edition, October 2009Third (present) edition, October 2013Amendments issued since publicationDate Text affectedBS 7965:2013 BRITISH STANDARDContentsForeword ii1 Scope 12 Normative references 13 Symbols and abbreviations 24 Selection 35 Installation recommendations 146 Operational considerations 227
4、Ultrasonic flowmeter verification and calibration 358 Condition-based maintenance (CBM) 53AnnexesAnnex A (informative) Application of flow calibration results 56Annex B (informative) Methods for correcting a USMs flow measurement errorat flow calibration 59Annex C (informative) Thermowell installati
5、ons in bi-directional flowinstallations incorporating flow-conditioners 66Bibliography 67List of figuresFigure 1 General ray geometry for transit time velocity measurement 4Figure2Transducer designs 7Figure3Variations in path configuration 8Figure 4 Some typical cross-sectional acoustic path configu
6、rations 11Figure 5 Performance specification summary for Class 1 meters 23Figure 6 Performance specification summary for Class 2 meters 24Figure 7 Performance specification summary for Class 3 meters 25Figure 8 Performance specification summary for Class 4 meters 26Figure B.1 Un-corrected and FWME-c
7、orrected flow calibration 62Figure C.1 Bi-directional flow installation employing two USMs 66Figure C.2 Bi-directional flow installation employing one USM 66List of tablesTable 1 Symbols and abbreviations 2Table 2 Class performance criteria Meter requirements 12Table 3 System uncertainties and appli
8、cations System requirements 13Table 4 Common thermal expansion coefficients in the 0 to 100 C range 27Table 5 Calibration options and applications 36Table 6 Internal diameter tolerance 36Table7Tolerance for beam path length 37Table 8 Minimum nominal test flow rates 43Table 9 Spread or repeatability
9、(max. error minus min. error) versus number oftest points for a specified uncertainty in the mean 45Table 10 Effects of number of readings 45Table A.1 Sequential flow calibration results 58Table B.1 Example flow calibration data table for 200 mm diameter USM 60Table B.2 Flow weighted mean error calc
10、ulation 61Table B.3 FWME-corrected flow calibration data summary 61Table B.4 Example results from a typical flow calibration 63Table B.5 Example calibration results 64Table B.6 Calculation of percentage flowrate 65Summary of pagesThis document comprises a front cover, an inside front cover, pages i
11、to ii,pages 1 to 68, an inside back cover and a back cover.BRITISH STANDARD BS 7965:2013 The British Standards Institution 2013 iForewordPublishing informationThis British Standard is published by BSI Standards Limited, under licence fromThe British Standards Institution, and came into effect on 31
12、October 2013. Itwas prepared by Subcommittee CPI/30/5, Velocity based methods, under theauthority of Technical Committee CPI/30, Measurement of fluid flow in closedconduits. A list of organizations represented on this committee can be obtainedon request to its secretary.SupersessionThis British Stan
13、dard supersedes BS 7965:2009, which is withdrawn.Information about this documentThis is a full revision of the standard, and introduces the following principalchanges: further division of meter sizes into large, medium and small to incorporatemeter sizes with a nominal bore last viewed 14 October201
14、3.BRITISH STANDARD BS 7965:2013 The British Standards Institution 2013 13 Symbols and abbreviationsFor the purposes of this British Standard, the following symbols andabbreviations apply.Table 1 Symbols and abbreviationsSymbol Meaning UnitsA cross-sectional area of meter body m2Cfspeed of sound in a
15、 fluid m/sCI confidence interval d diameter of meter spool MEiindicated flow rate error in percent (%)F single calibration factor frfrequency Hzfivariable weighting factor for ith velocity path khtheoretical correction factor kruniform equivalent roughness mmK meter factor typically the ratio of ref
16、erence/meter flows l acoustic path length between two ultrasonic transducers MmA current in milli-amps mAP absolute static pressure of gas barqiflow rate m3/sqmaxmaximum volumetric flowrate of meter m3/sqminminimum volumetric flowrate of meter, as specified by themanufacturer, as being the flowrate
17、below which theexpanded error limit for the meter class is exceeded.m3/sqttransitional volumetric flowrate where the specifiedperformance of the meter changesm3/sqvvolumetric flow rate m3/sRaarithmetic mean deviation of the (roughness) profile mT temperature of fluid Ctabtransit time for ultrasound
18、to travel from transducer a totransducer bstbatransit time for ultrasound to travel from transducer b totransducer asUSMtolpredetermined tolerance %ttba tabstmean of taband tbast average transducer/electronic delay time. sviaveraged axial component of velocity along the pathbetween two transducersm/
19、sv mean pipe velocity m/sV Voltage Vwiconstant weighting factor for ith velocity path Wfiflow rate weighting factor X distance between transducers in the axial plane mZ compressibility of fluid angle between the direction of ultrasound and the axis ofthe pipea.c. alternating current Ad.c. direct cur
20、rent ABRITISH STANDARDBS 7965:20132 The British Standards Institution 2013Table 1 Symbols and abbreviationsAGC automatic gain control CBM Condition-based maintenance CFD Computational fluid dynamics EU-ETS European Union Emissions Trading Scheme FWME flow weighted mean error LDA LaserDoppler Anemome
21、try MSOS measured speed of sound PTZ pressure, temperature and compressibility SOS speed of sound m/sSNR signal-to-noise ratio SPU signal processing unit TSOS theoretical speed of sound USM ultrasonic flowmeter USMP USM package, including meter, meter tubes, flowconditioner and thermowellWME weighte
22、d mean error 4 SelectionCOMMENTARY ON CLAUSE 4There are a number of USMs currently in service and under development. Thisclause aims to present a generic overview of the meters available and offer guidanceas to their selection for specific applications.4.1 Principles of measurement4.1.1 Gas velocity
23、USMs are based on the measurement of the propagation time of acoustic wavesin a flowing medium. Generally, the assertion that the apparent velocity along apath is given by the speed of sound in the fluid at rest, Cf, plus the averagedaxial component of fluid velocity along the path between the trans
24、ducers, vi,isapplied. To eliminate the speed of sound from the subsequent derivation, transittimes are determined both in the direction of flow and against it. Consideringthe general ray geometry shown in Figure 1, the upstream and downstreamtransit times are given by the following equations:tab5L(C
25、f1vicosH20850 (1)tba5L(Cf2vicosH20850 (2)BRITISH STANDARD BS 7965:2013 The British Standards Institution 2013 3Figure 1 General ray geometry for transit time velocity measurementKeyX Distance between transducers in theaxial plane Angle between direction of ultrasound path andaxis of pipeD Diameter o
26、f meter spool a Transducer (upstream)L Acoustic path length between twoultrasonic transducersb Transducer (downstream)There are four basic methods by which transit time velocity measurement can beperformed: direct time differential, phase differential, phase control, andfrequency differential as def
27、ined in BS ISO/TR 12765:1998, Clause 3 and Clause 5.In modern ultrasonic flowmeters the direct time differential method is mostcommon. Short pulses are propagated upstream and downstream and the timeinterval for each excitation/detection is measured against an accuratehigh-frequency clock. In this c
28、ase, the expressions for the upstream anddownstream transit times are then solved for vias follows:1tab21tba5H20849Cf1vicosH208502H20849Cf2vicosH20850L (3)vi5L2cosS1tab21tbaD5L2costtabtba5L22Xttabtba(4)Where:cos5XLTo obtain the volumetric flow rate, qv, individual velocity measurements arecombined b
29、y a mathematical function to yield an estimate of the mean pipevelocity and multiplied by the cross-sectional area of the measurement section,A, as follows:qv5Av (5)Where:v5fH20849v1,vnH20850 (6)Where:n is the total number of paths.BRITISH STANDARDBS 7965:20134 The British Standards Institution 2013
30、Even for a given number of paths, the exact form ofv5fH20849v1,vnH20850differs due to variations in path configuration and different proprietaryapproaches to solving Equation 6 (see 4.3).4.1.2 Density/molecular weightThe dynamically measured average speed of sound, after correcting for pressureand t
31、emperature, or possibly some other disruptive variables, suffices todetermine the molecular weight and density of the flowing medium.An alternative indirect mass method is to enter a fixed static density into anultrasonic flow meter that assumes a well defined and constant gas composition.The dynami
32、c function is often employed in flare gas mass flow measurement.The accuracy of density determination is generally stated by manufacturers asbetter than 2.0%; a value that can be improved upon if the composition of theflare gas is well defined and relatively constant. An example of a disruptivevaria
33、ble for a flare gas measurement is that of nitrogen as a non-hydrocarbon.The determination of density from speed of sound measurement, in anultrasonic flow meter, has been demonstrated as capable of matching traditionalvibrating tube densitometers when employed on single component, ultra-purefluids
34、such as Ethylene.Determination of molecular weight of flare gas may be employed in identifyingspecific flare gas sources in common flare gas headers such as in an oil refineryor petro-chemical complex. Specific molecular weights or changes from theaverage molecular weight can signal and help identif
35、y leakages into the flaregas stream from passing relief valve systems.4.2 Factors affecting performanceThe performance of a USM is dependent on a number of intrinsic and extrinsicfactors.Intrinsic factors (i.e. those related to the meter and its calibration, as applicableto the class of meter, prior
36、 to delivery) include the: geometry of the meter body and ultrasonic transducer locations and theuncertainty with which these are known; accuracy and quality of the transducers and electronic components used inthe transit time measurement circuitry (e.g. the electronic clock stability); techniques u
37、sed for transit time detection and computation of meanvelocity (the latter of which determines the sensitivity of the meter tovariations in the flow velocity distribution); calibration applicable to the class of meter (for static calibrations thisincludes proper compensation for signal delays in ele
38、ctronic componentsand transducers); thermal expansion coefficient of the meter body; pressure coefficient of the meter body.Extrinsic factors (i.e. those related to the flow and environmental conditions ofthe application) include the: flow velocity profile; temperature distribution; flow pulsation;
39、gas composition;BRITISH STANDARD BS 7965:2013 The British Standards Institution 2013 5 noise (acoustic and electromagnetic); solid and liquid contamination; impurities in the gas.4.3 Description of generic types4.3.1 GeneralThis subclause is a generic description of USMs for gases. It recognizes the
40、 scopefor variation within commercial designs and the potential for newdevelopments. For the purpose of description, ultrasonic meters are consideredto consist of several components, namely: transducers, see 4.3.2; meter spool (and acoustic path configuration), see 4.3.3; electronics, data processin
41、g and presentation unit(s), see 6.5.4.3.2 TransducersTransducers can be supplied in various forms. Typically, they comprise apiezoelectric element with electrode connections and a supporting mechanicalstructure with which the process connection is made. The process connection canbe welded, flanged o
42、r threaded, or can be more mechanically complex, forexample, to allow removal of transducers from a pressurized line. When insertedin the line, the active element transmits ultrasonic waves at an angle to themeter spool axis in the direction of a second transducer or reflection point inthe meter spo
43、ol interior. The active element is usually isolated from the fluid bya mechanical barrier.For specific applications, special transducers might be needed to cope with theconditions detailed below. These special transducers can differ from the norm interms of frequency, construction materials and mech
44、anical arrangement.Transducer specification and mounting should be given careful consideration forextreme or difficult application conditions, such as: high gas flow velocities; high, low and/or rapid cyclic temperature; high, low and/or rapid cyclic pressure; corrosion/erosion; wet or “dirty” gas;
45、close proximity to high ultrasonic noise sources; the gas composition.Two general examples of transducer design are depicted in Figure 2.4.3.3 Meter spool and acoustic path configurations4.3.3.1 GeneralThe meter spool is a conduit comprising one or more pairs of ultrasonictransducers. An acoustic pa
46、th is the transmission link between two transducersthrough the flowing fluid. USMs are available in a variety of pathconfigurations. A particular configuration is generally chosen based on arequirement with respect to variations in velocity distribution and/or built inredundancy requirements.BRITISH
47、 STANDARDBS 7965:20136 The British Standards Institution 2013As well as variations in the location of the measurement paths in thecross-section, the path configuration can be varied in orientation to the pipeaxis. By using reflection of the ultrasonic wave from the interior of the meterspool or from
48、 a fabricated reflector, the path can traverse the cross-sectionseveral times.4.3.3.2 Elementary acoustic pathsElementary acoustic paths are illustrated in Figure 3. The number of traversesand three-dimensional geometry of each traverse describe an arbitrary path.Certain path configurations might of
49、fer reduced susceptibility to disturbed flowprofiles. Claims for any such reduced susceptibility will need to have beendemonstrated by the manufacturer with appropriate supporting test data.In some specific applications the path length may be reduced, e.g. flare gasmeters.Figure 2 Transducer designsa) Intrusive b) Non-intrusiveBRITISH STANDARD BS 7965:2013 The British Standards Institution 2013 7Figure 3 Variations in path config
