precision pressure measurement handbook

Precision Pressure Measurement

User Guide to Pressure Measurement

P-CP-2010-0200May 2013

2

Table of ContentsChapter 1—Fundamentals of Pressure Measurement

1.1 Gauge Versus Absolute Pressure ........................................................................................... 41.2 Nominal Versus Actual Pressure .............................................................................................. 41.3 Units of Pressure Measurement ............................................................................................... 41.4 Units of Mass Measurement ...................................................................................................... 51.5 Gravity ........................................................................................................................................ 51.6 Primary and Secondary Pressure Standards .......................................................................... 5

Chapter 2—Deadweight Pressure Testers2.1 Piston Gauge Type Deadweight Testers .................................................................................. 6

2.1.1 Simple Piston & Cylinder .................................................................................................... 62.1.2 Re-entrant Piston & Cylinder ............................................................................................. 62.1.3 Controlled Clearance Piston & Cyl. ................................................................................... 7

2.2 Floating Ball Type Deadweight Testers .................................................................................... 72.3 Other Types of Deadweight Testers ........................................................................................ 82.4 Intrinsic Pressure Correction Terms ......................................................................................... 8

2.4.1 Air Buoyancy ..................................................................................................................... 92.4.2 Thermal Expansion ............................................................................................................ 92.4.3 Surface Tension ................................................................................................................ 92.4.4 Tare Force ......................................................................................................................... 92.4.5 Area ................................................................................................................................... 92.4.6 Pressure Coefficient ......................................................................................................... 9

2.5 Site Location Correction Factors ............................................................................................ 102.5.1 Gravity .............................................................................................................................. 102.5.2 Temperature .................................................................................................................... 102.5.3 Test Level ........................................................................................................................ 11

2.6 Traceability and Uncertainty .................................................................................................... 112.6.1 Traceability ...................................................................................................................... 112.6.2 Laboratory Uncertainty ................................................................................................... 12

Chapter 3—Calibration of Deadweight Testers3.1 Methods of Calibration ............................................................................................................. 13

3.1.1 Fundamental Characterization ........................................................................................ 133.1.2 Calibrated Characterization ............................................................................................ 13

3.2 Test Equipment for a Pressure Standards Lab ...................................................................... 133.2.1 Introduction ...................................................................................................................... 133.2.2 Hydraulic Piston & Cylinder Type Testers ..................................................................... 143.2.3 Pneumatic Floating Ball Type Testers ............................................................................ 15

3.3 Frequency of Calibration ......................................................................................................... 163.3.1 Piston & Cylinder Hydraulic Deadweight Testers ......................................................... 163.3.2 Floating Ball Pneumatic Deadweight Testers ................................................................. 16

Chapter 4—AMETEK Deadweight Testers4.1 Deadweight Testers and Deadweight Gauges ..................................................................... 184.2 Features of AMETEK Hydraulic Deadweight Testers ............................................................ 18

4.2.1 Model T & R Hydraulic Testers ....................................................................................... 204.2.2 Model 10 Hydraulic Testers ............................................................................................ 214.2.3 Model HL Hydraulic Testers ............................................................................................ 224.2.4 Dual Column Hydraulic Testers ...................................................................................... 22

4.3 Features of AMETEK Deadweight Gauges ............................................................................ 224.3.1 Model HLG Deadweight Gauge ...................................................................................... 23

4.4 Features of AMETEK Pneumatic Deadweight Testers .......................................................... 24

3

4.4.1 Model PK II Pneumatic Tester .......................................................................................... 264.4.2 Model RK Pneumatic Deadweight Tester ....................................................................... 274.4.3 Model HK Pneumatic Deadweight Tester ....................................................................... 274.4.4 Model PK Medical Pneu. Deadweight Tester ................................................................. 28

4.5 Care of AMETEK Deadweight Testers ................................................................................... 284.5.1 Hydraulic Deadweight Testers ....................................................................................... 284.5.2 Floating Ball Pneumatic Testers ...................................................................................... 30

4.6 Procedures for the Use of AMETEK Testers ......................................................................... 314.6.1 Procedures for Cleaning Hydraulic Piston & Cylinder ................................................... 314.6.2 Procedures for Use of Hydraulic Deadweight Testers ................................................ 324.6.3 Procedures for Use of Hydraulic Deadweight Gauges ................................................ 334.6.4 Procedures for Use of Pneumatic Deadweight Testers ............................................... 34

Chapter 5—Manometers5.0 Principle of Operation .............................................................................................................. 365.1 Types of Manometers .............................................................................................................. 36

5.1.1 U Tube Manometers ......................................................................................................... 365.1.2 Inclined Tube Manometers .............................................................................................. 365.1.3 Well Type Manometers .................................................................................................... 37

5.2 Intrinsic Correction Factors ..................................................................................................... 375.2.1 Fluid Density Correction .................................................................................................. 375.2.2 Gravity Correction ........................................................................................................... 385.2.3 Pressure Medium Head Correction ................................................................................. 395.2.4 Scale Correction .............................................................................................................. 405.2.5 Compressability, Absorbed Gases and Capillary Considerations ................................ 405.2.6 Parallax (Readability) ...................................................................................................... 41

Chapter 6—Secondary Comparison Pressure Standards6.1 Types of Secondary Comparison Pressure Stds. ................................................................. 42

6.1.1 Mechanical Pressure Gauges ........................................................................................ 426.1.2 Electronic Pressure Gauges ........................................................................................... 42

6.2 Overall Application Accuracy ................................................................................................. 436.2.1 Application Task .............................................................................................................. 436.2.2 Accuracy Measurement ................................................................................................. 436.2.3 Hysteresis, Linearity, Repeatability ................................................................................ 446.2.4 Temperature Effect ......................................................................................................... 456.2.5 Resolution ........................................................................................................................ 466.2.6 Stability ............................................................................................................................. 466.2.7 Laboratory Uncertainty ................................................................................................... 466.2.8 Overall Accuracy ............................................................................................................ 47

Chapter 7—Selection of a Pressure Measurement Standard7.1 Application Considerations ...................................................................................................... 49

7.1.1 Test Fluid .......................................................................................................................... 497.1.2 Pressure Range ............................................................................................................... 497.1.3 Task to be Performed ...................................................................................................... 49

7.2 Cost of Measurement ............................................................................................................... 497.2.1 Custody Transfer ............................................................................................................ 507.2.2 Process Control ............................................................................................................... 517.2.3 Safety ............................................................................................................................... 517.2.4 Maintenance .................................................................................................................... 51

7.3 Summary & Conclusions .......................................................................................................... 52

Table of Contents

4

��

��

Pressure, by definition, is a derived para-meter.One cannot create an artifact of one pound persquare inch or any other measure of pressure.Pressure is derived by the combination of a massmeasurement imposed upon an area. It is com-monly expressed in terms of pounds force or perunit area (Pounds per square inch). Pressure canalso be expressed in terms of the height of a liquidcolumn (Inches of water or millimeters of mer-cury) that produces the same pressure at itsbase.

1.1 Gauge Versus Absolute Pressure

Pressure measurements are always expressedas the difference between the measured pres-sure and some base pressure. Gauge pressureis the pressure measured from, or in addition to,atmospheric pressure. Gauge pressure is nor-mally expressed in terms such as PSIG or poundsper square inch gauge. Absolute pressure ismeasured from a base of zero pressure and isexpressed as PSIA or pounds per square inchabsolute. Negative pressures such as vacuumare expressed as the difference between atmo-spheric pressure and the measured pressure.Vacuum or negative pressures are normally ex-pressed as inches or millimeters of mercury orwater vacuum.

1.2 Nominal Versus Actual Pressure

The accuracy of the generated pressure is mea-sured as the difference between the actualpressure produced by the pressure standardand a standard pressure to be used for thecalibration of a secondary standard pressuremeasuring instrument. AMETEK deadweighttesters are referred to “Nominal” or even unitpressures, such as 1000, 2000, 3000 PSIG etc.Pressure standards manufactured by other manu-facturers, Ruska, D.H., refer the output pressureto the actual output pressure as stated on thecertification or computed by an equation includedwithin the certification.

AMETEK accuracy is therefore the differencebetween the output pressure and the nominalpressure stated in percentage of the reading.AMETEK further states the ability of the instru-ment to repeat identical pressures with identicalweights and piston as the percentage of “Repeat-ability”. Ruska, D&H accuracy as stated is thestatistical ability of the instrument to repeat theidentical pressure with the same weights andpiston. This is comparable to AMETEK’s statedpercentage for instrument repeatability.

1.3 Units of Pressure Measurement

Pressure is measured in several different unitsdepending upon the application and the countryin which the measurement is taken. Within theUnited states, the most common unit of measureis Pounds (Force) per Square Inch, for lowpressure measurements a measure of Inches ofWater Gauge and for vacuum Millimeters of Mer-cury Vacuum. The official unit of pressure mea-surement within United States is the Pascal whichis defined as Newton per square meter. A detailedlisting of the various measures of pressure andthe equivalent pressure in pounds per squareinch is as follows:

Low PressureInches of Water Gauge (20oC) =0.036063 PSI (ISA RP 2.1)

Inches of Water Gauge (60oF) =0.036092 PSI (AGA Report 3)

Inches of Water Gauge (4oC) = 0.036126 PSIMillimeters of Water Gauge (20oC)= 0.0014198 PSIMillimeters of Water Gauge (4oC)= 0.00142228 PSI

VacuumInch of Mercury (0oC)= 0.49114999 PSITorr = 1 Millimeter of Mercury (0oC) = 0.019718 PSIMillimeters of Mercury (20oC) = 0.019266878 PSI

PressurePascal = 1 Newton/Meter2 = 0.0001450377 PSIBar = 100 Kilopascal = 14.50377 PSIKilogram (Force) per Square Meter = 14.22334 PSI

Chapter 1

5

1.4 Units of Mass Measurement

The term mass as used in the mathematical ex-pressions for pressure is understood to be thetrue mass or the mass value that would bemeasured in a vacuum. Although this is the valuerequired for the pressure equation, many differ-ent methods have been used by both deadweighttester manufacturers and calibration facilities.These methods fall into two categories: true massand apparent mass versus some material of adifferent stated density. Typical materials andconditions for apparent mass are brass with adensity of 8400 kg/m3 and stainless steel at 8000kg/m3, both measured at 20oC.

1.5 Gravity

The term force that is used within the deadweighttester mathematical expression for pressure isdefined as the mathematical product of the truemass and the local gravity. The total variation dueto gravity over the surface of the earth can varyas much as 0.5%. Acceleration due to gravity canbe calculated as follows:

g = 980.6160 (1 - 0.0026373 cos 2φ +0.0000059 cos2φ)

Where φ is the latitude at sea level and

g1 = gφ - .0003086 h + .0001118 (h -h

1)

Where h1 is the elevation, in meters, of the generalterrain for a radius of 70 km and h, in meters, isthe elevation of the location above sea level.

An exact measurement of a specific location canbe achieved by the conduct of a gravinometricsurvey. An estimate of local gravity for continen-tal United States locations is available at nominalcost from the United States Department of Com-merce, National Ocean Survey (301-713-3242).

1.6 Primary and Secondary PressureStandards

Primary pressure standards must be directlytraceable to the physical standards of length andmass, and any errors must either be eliminated orevaluated. Within the deadweight tester the areaof the piston and cylinder or the ball and nozzlecan be measured and directly traceable to thephysical standard of length. The weights can bemeasured directly traceable to the physical stan-dard of mass. The only other pressure measure-ment device that fulfills this definition of primaryis the U tube manometer wherein the columndifference is traceable to length and the fluiddensity is traceable to both mass and length.

All other devices for measuring pressure, re-gardless of accuracy, uncertainty, etc. are con-sidered as secondary. This includes electronic,quartz tube, vibrating cylinder, etc. type pressuremeasuring instruments.

6

��

Chapter 2

Deadweight pressure testers derive pressure bythe combination of a mass, usually weights,which are floated on a piston and cylinder com-bination with a defined area. The basic equationfor the deadweight tester is:

P = F/A

Where:

P = the pressure being derivedF = the force applied by the weightsA = the effective area of the piston cylinder

2.1 Piston Gage Type Deadweight TestersWithin a piston gage or deadweight tester, aplatform containing calibrated weights is bal-anced upon a piston which is floated within acylinder. The fluid can be either a liquid, a hydrau-lic deadweight tester, or a gas, a pneumaticdeadweight tester.

Piston gages commonly incorporate three de-signs of cylinders; the simple cylinder, the re-entrant cylinder and the controlled clearancecylinder.

2.1.1 Simple Piston and CylinderAs illustrated on Figure 2-1, within the simplepiston and cylinder design the test fluid is con-nected into a chamber below the interior of thecylinder, the weights are suspended upon thepiston and the piston is floated upon the pressur-ized test fluid.

Since the pressure is applied only to the interiorof the cylinder, increasing pressures result in theexpansion of the cylinder bore and reduction ofthe piston diameter, thus increasing the effectivearea of the combination. This combination ofeffects results in an increased clearance be-tween the piston and cylinder. This increasedclearance results in an increased leakage of thetest fluid, thus limiting the available test timebefore fluid replenishment. For these reasons,the simple piston cylinder design is used primarilywithin pneumatic piston and cylinder gages andhydraulic piston and cylinder gages below 10000PSI (690 bar).

2.1.2 Re-entrant Piston & CylinderAs illustrated on Figure 2-2, within the re-entrantpiston and cylinder design the test fluid is con-nected to a chamber on the outside of the cylinderas well as being connected to the interior of thecylinder. In operation, since the area of the out-side of the cylinder is larger than the insideincreasing test pressures will reduce the bore ofthe cylinder as well as reduce the diameter of thepiston. The net effect of these reductions will beto reduce the effective area and clearance be-tween the piston and cylinder at high pressures.The available test time is not restricted by theexcessive loss of test fluid.The re-entrant typepiston and cylinder design is used in all pressureranges, including high pressures exceeding 50000PSI (3450 bar).

2.1.3 Controlled Clearance Piston & CylinderThe controlled clearance piston and cylinder issimilar to the re-entrant piston and cylinder ex-cept that the chamber on the outside of the pistonand cylinder is connected to a separate sourceof calibration pressure. As illustrated on Figure 2-3, the calibration pressure can be adjusted ateach point of pressure measurement to maintainthe effective diameter of the cylinder.

In operation, this type of piston gage althoughvery accurate is very slow to operate and veryexpensive to purchase. It is used primarily by

Figure 2-1Simple Type Piston & Cylinder

PressurizedTest Fluid

7

standards laboratories such as the National Insti-tute of Standards and Technology (NIST) in theUnited States.

2.2 Ball Type Deadweight TesterBall type deadweight gages measure pressure interms of force over unit area. These instrumentsoperate identical to piston gages except that a ballis used instead of a piston.

A schematic diagram of the AMETEK ball gage isshown on Figure 2-4. In this instrument, clean airis supplied from a flow regulator to an equalizingannulus and from there to a spherical chamberunder the ball and to the output port. In operationthe ball, with weight hanger and weights sus-pended from it, floats on a film of air with virtuallyno friction. The nozzle bore is tapered in the areawhere the circumference of the ball is located.This taper permits a variable exhaust flow whichfunctions in a pneumatic feedback loop togetherwith the flow regulator to maintain the ball in afixed vertical position. Pressure builds up withinthe spherical chamber below the ball until the balland the weights are suspended. The pressurewithin the spherical chamber is ported throughstabilizing tubing to eliminate pressure oscilla-tions.

2.3 Other Types of Piston GagesA vacuum-backed piston gage is designed tomeasure absolute pressure. This instrument issimilar to the conventional piston gage except thatboth the piston and cylinder and the weightsoperate within a vacuum evacuated bell jar. Thepressure within the evacuated chamber must bemeasured accurately to arrive at the total abso-lute pressure being measured by the instrument.

The tilted piston gage is designed to operate atlower pressures. Since lower pressures arelimited by the weight of the piston itself, the weightcan be reduced by using a hollow piston, backingit with a vacuum, and tilting the assembly.

Figure 2-2Reentrant Type Piston & Cylinder

Figure 2-3Controlled Clearance Type Piston & Cylinder

PressurizedTest Fluid Pressurized

Test Fluid

CalibrationPressure

8

2.4 Intrinsic Correction Terms

As described by P.L.M. Heideman and B.E.Welch, United States National Institute of Stan-dards and Technology1, the pressure P gener-ated by a piston gage at its reference level isgiven by the following equation:

Where:M,ρM = Mass and density of weightsgl = Local gravityρair = Density of air at the temperature,

barometric pressure and humidityprevailing in the laboratory

γ = Surface tension of the pressure-transmitting fluid

C = Circumference of the piston whereit emerges from the fluid

TW = Tare weight or errorA0 = Effective area of the assembly at

zero pressureαc, αp = Thermal expansivities of cylinder

and pistonT = Temperature of the assemblyTref = Temperature at which A0 is referredb = Pressure coefficient of the effective

area

2.4.1 Air Buoyancy Correction

The term ( 1 - ρair/ρM ) is the air buoyancy correctionfor weight M. The density of brass weights is 8.4grams/cm3 and the density of stainless steelweights is 8.0 grams/cm3. The density of moist airhas been tabulated 2,3. It can be computed from theequation

ρair = 1.2929 x 103

ρbar

- 0.3783 Hpsat

273.13T

( )

2.4.1.1 Air BuoyancyThe air buoyancy correction must be carefullyevaluated. For a measurement at 23°C, 30%relative humidity, 1000 mb barometric pressureand stainless steel weights, the correction wouldbe 147.5 ppm or 0.00147%.

2.4.2 Thermal Expansion

The term [1 + (αc+ αp)(T - Tref)] corrects the area forthermal expansion. This correction does not con-tribute significantly to accuracy. For a tungstencarbide piston inside a steel cylinder the correc-tion amounts to 17 ppm or 0.000017% withT-Tref = 1°C.

Figure 2-4Schematic Diagram - Floating Ball-type Deadweight Tester

Spherical Piston(Ceramic Ball)

Nozzle

Nozzle Body

SuspendedWeights

DynamicStabilization

System

Weight Carrier

Cylinder Core

EqualizationAnnulus

RegulatorCapacity Tank

ExponentialFlow Regulator

Gas Filter(Optional)

Supply Connection"In"Test Pressure Connection

"Out"

1 - ρairΣ Mg γC + Tw( ) +

Ao [1 + (αc + αp) (T-Tref)] [1 + bp] ρMP =

9

2.4.3 Surface TensionThe term γC describes the force generated by thesurface tension of the fluid acting on the pistonwhere it emerges from the fluid. is the surfacetension of the fluid and C is the circumference ofthe piston. The value of Spinesso 38 is represen-tative for many oils used in piston gages. For apiston of 32 mm2 cross sectional area this correc-tion amounts to about 10-4 Newtons. This consti-tutes a minimal correction at high pressures, butfor low pressure oil gages it should always betaken into account. This correction does not applyto air-operated gages.

2.4.4 Tare ForceTW is the tare force which may result from an errorin the determination of the mass of one of theweights or in computation of the head correc-tions; it might also be characteristic of a gage.

2.4.5 AreaA0 is the cross sectional area of the piston. Thisarea can be determined in two different methods.The first method is to perform numerous mea-surements of the diameter along the length of thepiston and transform these measurements usinga suitable mathematical model. An alternativemethod is to cross-float a piston against anotherpiston or instrument of known pressure or effec-tive area.

2.4.6 Pressure CoefficientThe term ( 1 + bP) describes the change ofeffective area with pressure. This is the mostimportant correction term. The pressure co-efficient is determined by the mathematicalexpression1:

Where:

µ = Poissons ratioE = Modulus of elasticityrc = Inside cylinder radiusRc = Outside cylinder radiusrp = Radius of a solid pistonp = End pressure on pistonpc = Pressure surrounding the piston in the

clearance between the piston andcylinder

po = Pressure on outside of the cylinderpe = Pressure on the end faces of the

cylinder

Depending upon the design and the materials ofconstruction of a piston and cylinder, the pres-sure coefficient can be either positive (Areaincreases) or negative (Area decreases), andthat under certain conditions it could approachzero (Constant area).

2.5 Site Location Correction FactorsAMETEK Certification of Accuracy states theaccuracy of each deadweight tester based onspecific conditions for the gravity and for thetemperature. Corrections to the pressures statedin the Certification can be made to compensate forthe actual conditions at the test site as follows:

1 - 3µ2 Ep

Piston Distortionb =

2Ec (Rc - rc )22

(1 + µc) Rc + (1 + µc) rc2 2

+

po

pEc( ) 2Rc

Rc - rc

( )2

22-

pe

p( )µc

Ec

( )+

Cylinder Distortion due to internal pressure

Cylinder Distortion due to external pressure

Cylinder Distortion due to end loading

10

2.5.1 GravityIf the local gravity at the test site is different fromthe gravity for which the deadweight tester wascalibrated, the actual pressure as shown on theCalibration Report can be converted to actualpressure at the test site through the use of thefollowing formula:

PL = PA x GL/G

Where:

PL

= Actual Pressure at the localgravity of the test site

PA = Actual Pressure as shown in theCalibration Report

GL = Local Gravity in Gals. (cm/sec2)G = Gravity for which the deadweight

tester was calibrated in Gals. (cm/sec2)

An example of this correction would be a readingof 3000 PSIG, taken at a site with a local gravityof 979.2 Gals., the tester calibrated for Interna-tional standard gravity 980.665 Gals.

PL = 3000 x 979.2/980.665or

PL = 2995.518 PSIG

2.5.2 TemperatureIf the temperature at the test site is different fromthe temperature for which the deadweight testerwas calibrated, the actual pressure as shown onthe Calibration Report can be converted to actualpressure at the test site through the use of thefollowing formula:

PL = PA + PA x A ( TS - T )

Where:

PL = Actual Pressure at the temperatureof the test site

PA = Actual Pressure as shown on theCalibration Report

A = Thermal Coefficient per °C as shown onthe Calibration Report

TS = Temperature in °C at the test siteT = 23°C, the Reference calibration

temperature

An example of this correction would be a readingof 100 Inches of Water (60°F), taken at a site witha local temperature of 40°C (104°F). The AMETEKModel PK II Deadweight Tester has a thermalcoefficient of 0.0000167 per °C.

PL = 100 + 100 x 0.0000167 (40 - 23)

= 100 + 0.02839

= 100.02839 inches W.C.

2.5.3 Test LevelIf the instrument being calibrated by a deadweighttester is not at the same height as the tester,compensation for the difference in height must bemade. This compensation can be made using thefollowing formula:

PL = PA + PA x ( HI - HT) x Df

Where:

PL = Actual pressure applied to theinstrument under test

PA = Actual pressure shown in theCalibration Report

HI = The calibration height of the deadweighttester (A height line is on the T & Rweight carrier, the major diameter of thePK II ball).

HT = The calibration height of the instrumentbeing calibrated.

Df = Density of the fluid being used forcalibration(Oil,water, etc.)

An example of this correction would be a mea-surement of 100 PSIG, using an AMETEK Model Ttester (H2O), where the instrument being cali-brated is located on a platform 20 feet (6 meters)above the tester:

PL = 100 +( 0 - 20) x 62.4 #/ft.3/144

= 100 - 8.666

= 91.334 PSIG

11

2.6 Traceability and UncertaintyThe parameters of both traceability and uncer-tainty are concerned with the accuracy of mea-surement within the laboratory that conductedthe testing on the pressure standard. Traceabilityis a measure of the ability of the laboratory to tracethe calibration of a pressure standard through adirect chain of calibrations to a national standardslaboratory such as the United States NationalInstitute of Standards and Technology. Uncer-tainty is an estimate of the measurement accu-racy of the traceability chain of calibrations.

2.6.1 TraceabilityTraceability in United states is defined as theability to demonstrate conclusively that a particu-lar instrument or artifact either has been cali-brated by NIST at accepted intervals or has beencalibrated against another standard in a chain orechelon of calibrations ultimately leading to acalibration performed by NIST4.

All testers produced by AMETEK are directlytraceable to NIST. This traceability of deadweighttesters is shown on Figure 2-5 .As shown on thisdiagram, AMETEK sends master testers to NISTon a regular interval for testing. These mastertesters are then utilized on an annual basis tocalibrate working master testers within theAMETEK Standards Laboratory. The workingmasters are used to test and calibrate all new andrepaired testers prior to shipment.

2.6.2 Laboratory UncertaintyIn general, the result of a measurement, such aspressure, is only an approximation or estimate ofthe specific quantity subject to measurement, andthe result is complete only when accompanied bya quantitative statement of its uncertainty5. Tostate this in a different way, a measurement, suchas accuracy, is only as good as the technique andequipment used to take the measure, and theaccuracy of the standard against which theaccuracy was measured. The overall uncer-tainty must take into account all testing betweenthe national standard (NIST) and the instrument towhich the uncertainty is assigned.

The measurement uncertainty of the deadweighttesters has been estimated for all of the dead-weight testers that are currently being produced6.Within this analysis all of the terms which effectthe calibration process were evaluated to estab-lish an estimate of the uncertainty of pressuremeasurement within the AMETEK Standards Labo-ratory. This measurement, when combined withthe NIST uncertainty, is an estimate of the overalluncertainty of testers produced by AMETEK.

The factors contributing to uncertainty of theAMETEK Laboratory included the following:

1. Laboratory conditions. The effect of tem-perature, pressure and relative humidity onthe air density in the vicinity of the weights.

2. Accuracy of the mass values that were usedto certify the balances used to measureweights at AMETEK.

3. The effects of surface tension, fluid head andthermal expansion on the piston during thetesting.

4. The sensitivity and uncertainty of the cross-float measurement instrumentation.

5. The uncertainty of the master testers asreported to AMETEK by NIST.

6. The local gravity at the AMETEK laboratory.7. The pressure range of calibration on each

dead weight tester.

Figure 2-5Traceability Path

AMETEK Deadweight Testers

AMETEK NISTMaster

NISTUSA Pressure Standard

AMETEKLaboratory Master

Customer'sTester

12

Currently, equipment specifications and adver-tisements include only claims on the accuracy ofdeadweight testers and do not include the uncer-tainty on the accuracy measurement. The trace-ability path and uncertainty measurement withinthe AMETEK Pressure Standards Laboratory asreported are well above industry average.

Future efforts on a United States national levelincluding laboratory accreditation (Already inplace in Europe and Asia) should require that boththe accuracy and uncertainty levels of instru-ments such as deadweight testers be reported.

13

Chapter 3

Deadweight testers can be calibrated using ei-ther the “Fundamental” or “Calibrated” methods tomeasure the pressure and effective area. Afundamental calibration involves having the ef-fective area of the gauge determined using onlymeasurements of the SI base units (e.g. mass,length) plus a suitable model. A calibrated calibra-tion has the effective area determined via calibra-tion against a gauge for which the effective areaor generated pressure is already known.

3.1 Methods of Calibration

3.1.1 Fundamental Characterization ofDeadweight TestersThe dimensional characterization of both thepiston and cylinder are usually performed underconditions of atmospheric pressure and roomtemperature. Depending upon the shape of boththe piston and the cylinder and the support struc-ture for the mass loading mechanism, the detaileddimensional characterization is usually confinedto that part of the piston that is inside the cylinderduring normal operation. Pistons of unusual shapeare not normally characterized using this tech-nique.

Once a set of dimensional measurements hasbeen performed on both the piston and cylinder,there are several ways in which the area of boththe piston and cylinder can be specified. Depend-ing upon the severity of the departure of thepiston surface from true cylindricity, an averageradius of the piston and cylinder may be calcu-lated, or more sophisticated approaches includ-ing distortion of the piston by the applied pressureand thermal expansion of the piston can beapplied. The overall area of the piston and cylin-der assembly is then computed using an appro-priate mathematical model.

This technique is used primarily by national stan-dards laboratories, such as the National Instituteof Standards and Technology, that are respon-sible for establishing reference measurementsfor a larger group such as the United States.

3.1.2 Calibrated Characterization ofDeadweight TestersThe calibrated characterization of deadweighttesters involves the transfer of effective areas ofone piston and cylinder to another utilizing pres-sure based cross-float techniques. To use thistechnique, identical piston and cylinders are placedin identical mountings with the output pressuresconnected. Means, such as a differential pres-sure meter are included to identify the time whena pressure balance between the two pressuregenerating components has been achieved at thereference levels of both the test and referenceunits. During the test the weights are exchangedon the columns and the piston and cylinders areexchanged in the mountings to reduce the uncer-tainty of measurement. This technique is usedprimarily by industry and calibration laboratories.The reference or master pressure generatingunits (Piston-Cylinder & Weights) are usuallytested at a standards laboratory (AMETEK mas-ters are tested at NIST).

3.2 Test Equipment for a Pressure StandardsLaboratoryMany different combinations and assemblies ofequipment may be utilized with varying degreesof uncertainty of measurement (See paragraph2.2). We will use for purposes of illustration, theequipment that is used within the AMETEK Pres-sure Standards Laboratory.

3.2.1 Hydraulic Piston-Cylinder DeadweightTestersThe equipment, as illustrated on Figure 3-1, con-sists of four identical column assemblies to holdthe master piston and cylinders and one identicalcolumn to hold the piston and cylinder under test.Each of the master columns contains a masterpiston of a single range (0.01,0.020,.05,0.1in2) tominimize disassembly after calibration. Masterweights with known mass are used on both thetest and master piston and cylinders. A analogdifferential pressure meter is used to indicatewhen a pressure balance is achieved. A cross-over valve permits interconnection of the output

��

14

pressure generating components. The pressurebalance is illustrated by the following equation:

Which simplifies to:

Mu x g1

Au

Mm x g1

Am

=

Where:

Au = Area of the unknown piston and cylinderAm = Area of the master piston and cylinderMt = Known mass of the test weightsMm = Known mass of the master weightsg1 = Local gravity

The procedure for calibration of a piston andcylinder with an unknown effective area con-sists of the following steps:

1. The unknown piston and cylinder is cleanedand installed in the test column.

2. Pressure is applied to both the unknownpiston and cylinder and the master piston andcylinder to bring both pistons to position in themiddle of the vertical float range.

Am (Mu x g1)

(Mm x g1)=Au

3. Both pistons are rotated approximately 30RPM and the cross-over valve is closed.

4. The pressure differential between the mas-ter and the test piston-cylinders is indicatedon the null meter.

5. Tare weights are added to the piston-cylinderthat is low in pressure until the null metershows a pressure balance. The required tareweights are recorded on the test log.

6. A small tare weight is placed upon the oppo-site piston and cylinder to unbalance thepressure to the opposite side. This test as-sures that the piston was free and sensitivewhen the pressure balance was measured.

7. The steps 2-6 are repeated for each requiredtest pressure.

8. The effective area of the unknown piston andcylinder is determined using the equation inparagraph 2.4 (AMETEK uses a proprietarycomputer program for this task).

Figure 3-1Test Equipment for a Hydraulic Piston and Cylinder Pressure Calibration Laboratory

15

3.2.2 Pneumatic Floating Ball DeadweightTesters

The equipment, as illustrated on Figure 3-2, con-sists of an AMETEK Pneumatic Tester CalibrationSystem, a master floating ball pneumatic tester,with a known effective area, of the identicaldesign (Model PK or RK) as the tester beingtested, master weights with known mass andtest weights with known mass. The calibrationsystem incorporates a cross-over valve, a digitalreadout differential pressure meter and inputflow meters for both the test and master testers.

The procedure for calibrating the floating balltesters is as follows:

1. The ball, nozzle and nozzle body of thefloating ball tester with unknown effectivearea are thoroughly cleaned(See Section 4.for the recommended cleaning procedure).

2. The tester is leveled and he bulls-eye level isadjusted accordingly ( See Section 4. for therecommended procedure for leveling).

3. Input pressure is applied to both testers withthe cross-over valve in the open position.

4. Weights are added to both the test unit and themaster unit for the initial test point.

5. The cross-over valve is closed. The pres-sure differential between the test and themaster tester is displayed on the differentialpressure meter display.

6. The steps 3 and 4 are repeated for eachrequired test pressure.

7. The effective area of the unknown ball andnozzle is determined using the equation inparagraph 2.4 (AMETEK uses a proprietarycomputer program for this task).

3.3 Frequency of Calibration

3.3.1 Frequency of Recertification andRecalibration of Piston & Cylinder TypeDeadweight TestersThe precision piston and cylinder assembly, theoperating heart of the hydraulic deadweighttester, is a device which given proper care anduse has a long service life. M&G has manycustomers who have pistons and cylinders in usefor over 10 years.

Recertification of a deadweight tester is a pre-cautionary measure required by each user toassure that each device has not worn sufficientlyto produce inaccurate pressures.As discussedin paragraphs 2.6.1 and 3.3, many different

Figure 3-2Test Equipment for a Pneumatic Floating Ball Deadweight Pressure Calibration Laboratory

16

pressure between the output on the master andfactors can contribute to wear of the piston andcylinder. No specific time period for recertifica-tion is appropriate for each application. If thedevice is exposed to heavy usage and factorsthat contribute to wear such as dirty instrumentsbeing tested, frequent calibration is appropriate.Infrequently used instruments used only in labo-ratory conditions need not be calibrated fre-quently.

The United States National Institute of Technologyhas suggested that7"For piston gages for use asplant or laboratory standards (Accuracy within0.1%) when the gage is connected to a leak tightsystem and the piston is set into rotation that thepiston should fall slowly as a result of the leakbetween it and the cylinder. At maximum pres-sure, the fall rate of a good instrument should beless than 0.1 inch per minute. We regard this asthe most important single index of quality”.

Confirming the NIST findings, AMETEK has foundthat the leakage rate of the piston and cylinder isthe best measure of wear. This is an easy test toperform on AMETEK testers:

1. Load the tester with 100# of weights.2. Pump piston to the top of the stroke.3. Spin the weights and measure the time until

the weights descend to the bottom of thestroke.

4. If the time, thus the leak rate, increasessignificantly (+5%) wear is indicated and thetester should be recalibrated.

AMETEK suggests that for users with unknownhistory of usage with deadweight testers con-sider conducting leakage tests monthly and havethe testers recertified annually. This annual re-certification should include the actual perfor-mance data. As the user accumulates a historyof usage and wear, the recertification intervalcan be adjusted accordingly.

3.3.2 Frequency of Recertification of FloatingBall Type Pneumatic Type deadweight TestersFrequency of recertification of a pneumatic dead-weight tester is a precautionary measure under-taken by each user to assure that the device isfunctioning accurately and has not worn in sucha way as to produce inaccurate pressure. asdiscussed in Section 4.5, many factors can con-tribute to malfunction and wear. No standardperiod of recertification is appropriate for everyapplication. If the device is used for portableservice in dirty or dusty environment, frequentrecalibration is appropriate. Infrequently usedinstruments used only under laboratory condi-tions need not be calibrated as frequently.

AMETEK has found that the input flow rate to thetester for a specific output pressure is the bestmeasure of the operation of the tester. A flowmeter capable of reading flow in Standard cubicFeet per Hour is placed in the input nitrogen lineprior to the tester. the tester is cycled throughseveral output pressures within its pressurerange and the input flow rate is recorded. Thisflow recording should be initiated when the testeris new or immediately after a recalibration iscompleted. If a flow rate for a given pressurechanges significantly (+5%), wear or malfunc-tion are indicated.

Formal recertification should be done at regularintervals by cross-floating the tester against aNIST Certified Master Tester in a pressure labo-ratory. AMETEK recommends that users with anunknown history of usage with ball type pneu-matic testers initially consider conducting flowtests monthly and have the testers recertifiedannually. The annual test recertification shouldinclude performance data (Area, mass, correc-tion factors,etc). As the user accumulates ahistory of usage and wear, the recertificationcycle can be adjusted accordingly.

17

Chapter 4

AMETEK produces an extensive line of bothhydraulic and pneumatic deadweight testers anddeadweight gauges for both laboratory and fieldportable applications.Within this chapter we willdescribe the unique features of the AMETEKinstruments, the features of each tester, operat-ing procedures and care of the testers.

4.1 Deadweight Testers and DeadweightGaugesAll deadweight testers and deadweight gaugesare primary pressure standards. These instru-ments are one of only two instruments that canbe used to generate pressure using the funda-mental S.I. units. Pressure within both the dead-weight tester and the deadweight gauge is gen-erated using the fundamental expression:

Deadweight testers incorporate a source forgenerating pressure, such as a hand pump.Instruments that are to be tested are attached todeadweight testers and the pump is used topressurize both the instrument under test and thedeadweight piston and cylinder.

��

Figure 4-1Deadweight Tester Schematic

Pressure =Force

Area

Deadweight gauges, on the other hand, do notincorporate a pressure source. A deadweightgauge is attached to a source of pressure, suchas a pressurized line, and is used to measure thepressure within the line very accurately.

4.2 Features of AMETEK Hydraulic Piston & Cyl-inder Deadweight TestersThe AMETEK hydraulic deadweight testers areshown schematically on Figure 4-1. These instru-ments consist of a re-entrant type piston andcylinder mounted in a pressure containing columnassembly, a weight carrier suspended on thepiston, weights, a pump to supply pressure anda vernier assembly.

AMETEK hydraulic deadweight testers and dead-weight gauges incorporate many features thatare designed to improve the performance andsafety of these instruments:

Dual Volume Hand PumpThe lever action hand pump that is supplied withboth the Type R & T Hydraulic Deadweight Testersincorporates a dual volume control valve. Thispermits the pump to deliver a high volume torapidly fill systems and build pressure. The lowvolume permits easy pumping at high pressuresand a more gradual approach to the point ofpressure calibration.

18

Overhung Weight CarriersThe weights are suspended from a weight tubewhich is suspended from the piston assembly.As shown on Figure 4-2, this design lowers thecenter of gravity of the suspended weights andreduces side thrust and friction on the measuringpiston and cylinder assembly thus improving themeasurement accuracy.

Re-entrant Type Piston & Cylinder Assem-blyThis design of piston and cylinder, as describedin Section 2.1.2, has a reduced clearance be-tween the piston and cylinder at higher pres-sures. The advantage of this design is to reducethe rate of fluid leakage at high pressures, thusincreasing the time available for the technician tocalibrate instruments prior to pumping to restorefluid loss.

Enclosed Piston & Cylinder AssemblyThe piston and cylinder assembly is enclosed ina unique column design, see Figure 4-2. Theweight carrier is floated on the piston assemblyto prevent side thrust. These features preventaccidental damage to the piston assembly.

Positive Over-pressure Protection of theMeasuring Piston AssemblyThe vertical movement of the measuring pistonassembly is restricted by a positive stop asillustrated on Figure 4-2. This positive stop willprevent the piston from damage caused by acci-dental removal of the weights when pressurized.These stops are rated at the maximum testerpressure.

Interchangeable Piston & Cylinder Assem-bliesThe column assemblies will accept several differ-ent size piston & cylinder assemblies as illus-trated on Figure 4-3. The Type T or R columnaccepts piston and cylinder assemblies withcross section areas of 0.1, 0.05, 0.02 and 0.01square inch. The Type HL column accepts pistonand cylinders with 0.1 and 0.02 square inchareas. With the use of a single weight set, thisfeature increases the test pressure range capa-bility of the testers.

Pressure VernierAn auxiliary screw type piston on the Type T &R deadweight testers permits fine adjustment ofpump pressure. The pump in the Type HL dead-weight testers is a screw type piston pump whichpermits fine adjustment.

AccuracyThe standard accuracy of the Type T, R & HLdeadweight testers is +0.1% of the indicatedpressures at nominal readings. An optional accu-racy of +0.025% of the indicated reading isavailable on the T & R testers.

Figure 4-2Piston & Cylinder Mounting Assembly

Figure 4-3Interchangeable

Piston & Cylinder Assemblies

19

Certification of Accuracy and TraceabilityA certification of accuracy and traceability toNIST is included with each deadweight tester. Anoptional certification of accuracy with area, mass,intrinsic correction factors and pressures isprovided with the optional +0.025% accuracy.

Weight MaterialsThe weights for all AMETEK deadweight testersare manufactured of non magnetic materials toprevent inaccuracy. The standard model T,R,10and HL weights are manufactured of a hardnon-magnetic zinc alloy. Optional weights for theModel T & R testers are manufactured from forgedbrass conforming to the material requirements ofNIST Class Q.

Interchangeable WeightsThe weights for standard accuracy (0.1%) dead-weight testers are interchangeable and lost ordamaged weights may be replaced.

Test FluidThe test fluid for the Type T is distilled water. Thetest fluid for the Types R ,10 and HL is AAAtester oil.

4.2.1 AMETEK Type T & R HydraulicDeadweight TestersThe AMETEK Type T & R are pictured on Figure 4-4. These are high accuracy pressure standardsdesigned for both laboratory or portable use.Specifications and features of these testers areas follows:

Pressure Range:Type T- 10 to 15000 PSI (0.7 to 1034.3 bar)Type R- 10 to 10000 PSI (0.7 ti 689.5 bar)

Pressure Increments:High range 50 PSI (3.5 bar)Low range 5 PSI (0.4 bar)

Standard Models:Type T- 11 English, 6 metricType R- 11 English, 6 metric

Accuracy:+0.1% of indicated reading+0.025% of indicated reading (optional)

Features:• Dual Volume Hand Pump• Overhung Weight carriers• Re-entrant type Piston and Cylinder

Assemblies• Enclosed Piston and Cylinder Assembly• Positive Over-pressure Protection of the

Piston Assembly• Interchangeable Piston & Cylinder

Assemblies• Interchangeable weights on standard

accuracy testers• Pressure Vernier• Test Fluid- Type T, Distilled Water or AAA oil;

Type R, AAA oil

Figure 4-4Type T & R Hydraulic Deadweight Tester

20

4.2.3 AMETEK Type HL Hydraulic DeadweightTestersThe AMETEK Type HL, Hydra Lite, hydraulic dead-weight tester is illustrated on Figure 4-6. Thistester incorporates a unique single case designfor portable applications. The entire tester, in-cluding weights, is contained in a 9.0 x 9.0 x 10.0-inch (228.6 x 228.6 x 254.0 mm) case. Specifica-tions and features are as follows:

Dual Pressure Range:10 to 600 and 50 to 3000 PSI(0.7 to 41.4 and 3.5 to 206.9 bar)

Pressure Increments:High range 0.5 PSILow range 0.1 PSI

Standard Models:9 English, 18 metric

Accuracy:+0.1% of indicated reading

Features• Screw type single piston pump with

ratchet handle• Interchangeable weights• Tripod mountable for field use• Overhung weight carriers• Re-entrant type piston and cylinders• Enclosed piston and cylinder assembly• Positive over-pressure protection of the

measuring pistons• Test fluid, AAA oil

4.2.2 AMETEK Type 10 Hydraulic DeadweightTesterThe AMETEK Type 10 hydraulic deadweight testeris illustrated on Figure 4-5. This tester is designedfor both laboratory and portable applications.Specifications and features are as follows:

Dual Pressure Range:5 to 2000 and 25 to 10000 PSI(0.4 to 138 and 1.7 to 689.5 bar)

Pressure Increments:High range 25 PSI (1.7 bar)Low range 5 PSI (0.4 bar)



Features• Interchangeable coaxial design piston and

cylinder assemblies for rapid changeover.• Interchangeable weights• Single volume pump• Simple Piston and Cylinder assembly• Pressure Vernier• Test Fluid, AAA oil

Figure 4-5Type 10 Hydraulic Deadweight Tester

Figure 4-6Type HL Hydraulic Deadweight Tester

21

4.2.4 AMETEK Dual Column HydraulicDeadweight TestersThe AMETEK dual column hydraulic deadweighttesters are illustrated on figure 4-7. These testers,which are manufactured in both Type T and TypeR design, are designed for both laboratory orshop installation. Separate columns are providedfor both the high and the low pressure piston andcylinder assemblies. Range changes are accom-plished rapidly by actuating a crossover valve.

The features and specifications are identical tothe Type T and Type R described in paragraph4.2.1.

4.3 Features of AMETEK Hydraulic Piston &Cylinder Deadweight GaugesA deadweight gauge is designed to be connectedto a source of pressure, such as a feed water linein a steam power plant or a gas well head, andmeasure the pressure with a high degree ofaccuracy. In operation, after the tester is con-nected to the pressure source, weights areadded onto the piston until the piston floats freelyin mid stroke. The features of the deadweightgauges are identical to the comparable dead-weight testers (Type T and Type HL) with thefollowing additions:

Figure 4-7Dual Column Hydraulic Deadweight Testers

• Elastomer Membrane Fluid SeparatorAn elastomer membrane is included to permitisolation of the piston and cylinder from theprocess fluid to permit operation with bothcontaminating or abrasive fluids and also topermit operation with pneumatic pressure.

• Conversion Capability to a Deadweight testerThe Type HLG may be converted to a dead-weight tester with the addition of a conver-sion kit.

4.3.1 AMETEK Model HLG Deadweight GaugeThe AMETEK Model HLG deadweight gauge isshown on Figure 4-8. This instrument is designedas a portable instruments to take accurate pres-sure measurements under field conditions. Therange and features are identical to the Type HLdeadweight tester. This instrument can be up-graded to a Model HL deadweight tester by theaddition of an AMETEK conversion kit (P/N T-535).

Figure 4-8Model HLG Hydraulic Deadweight Gauge

22

4.4 Features of AMETEK Floating Ball TypePneumatic Dead Weight TestersThe AMETEK pneumatic deadweight testers areself-regulating primary type pressure standards.An accurate calibrating pressure is produced bybringing into equilibrium the pneumatic pressureon the under-side of the ball with known effectivearea by weights of known mass on the top.

The ball type pneumatic deadweight tester isshown on the schematic diagram on Figure 4-9.In this type of construction a precision ceramicball is floated within a tapered stainless steelnozzle. A flow regulator produces pressureunder the ball, lifting it in the stainless steel nozzleuntil equilibrium is reached. At this point the ball isfloating and the vented flow equals the fixed flowfrom the supply regulator. The pressure, whichis also the output pressure, is proportional to theweight load. During operation the ball is centeredby a dynamic film of air, eliminating physicalcontact between the ball and nozzle.

When weights are added or removed from theweight carrier, the ball rises and lowers, affect-ing the air flow. The regulator senses thechange and adjusts the pressure under the ballto bring the system under equilibrium, changingthe output pressure accordingly. This regula-tion of output pressure is automatic withchanges in weight mass on the spherical piston(ball). When minimum supply pressures aremaintained, the unit equilibrium is unaffected byvariations in supply pressure.

AMETEK floating ball type pneumatic dead weighttesters incorporate many features that aredesigned to improve the performance and safetyof these instruments:

Floating BallIn operation, the ball and weights float freely,supported only by a film of air which is virtuallyfrictionless. This feature eliminates the neces-sity to rotate the weights during testing, thusfreeing the technician to concentrate on thetask of instrument calibration.

Figure 4-9Schematic AMETEK Floating Ball Pneumatic Deadweight Testers

23

Self RegulatingThe built-in flow regulator automatically adjuststhe input air flow to maintain the ball and weightsin a float position. The regulator also compen-sates for variations in pressure from the airsupply. These features eliminate the necessityof the technician continually adjusting the supplyduring the test, thus permitting the technician tocalibrate instruments more efficiently.

Rugged Ceramic Measuring BallThe floating ball/piston is manufactured fromaluminum oxide ceramic. This material has neardiamond hardness. The ball, unlike steel andcarbide pistons, can be dropped on hard sur-faces without damage.

Quick Setup and OperationSet up is completed by connecting two tubes andadding the appropriate weights. Operation is fastwith no valves to adjust or regulation between setpoints. Pressure regulators are not required if theair supply is within the tester operational require-ments.

Non Contaminating Test FluidThe test fluid is nitrogen or instrument quality aircomplying to the ISA Standard S7.3. This fluid isnon contaminating to virtually all processes, thuseliminating the need to clean instruments aftercalibration before use.

Tripod MountingThe Models PK II and RK are designed for bothlaboratory and portable use. Both instrumentsincorporate built-in tripod mounts.

Closed Cover OperationThe Model PK II is designed to operate with thecover closed, thus eliminating the effects of windduring outside portable operation.

Ball ValvesAll of the AMETEK floating ball testersincorporate multi-position ball valves for both theinlet and outlet valve connections. These valvesafford quick, easy and trouble free operation.

Pre-fill ValveA pre-fill valve connection is provided on theModel HK high pressure pneumatic testers. Thisvalve permits the inlet pressure to bypass the

flow regulator to permit rapid filling of the systemvolume to be tested.

AccuracyThe standard accuracy of the Models PK II and RKis +0.05% of the indicated pressures at nominalreadings. Optional improved accuracies of+0.025% and +0.015% of the indicated pres-sures at nominal readings are available.

The standard accuracy of the Model HK tester is+0.025 % of the indicated pressure or +0.025 PSIwhichever is greater.

GravityStandard weights for all of the ball type dead-weight tester models are manufactured from nonmagnetic 300 series stainless steel. Theseweights can be calibrated to operate at Interna-tional Standard Gravity (980.665 Gals.) or at thecustomers local gravity.

An economic model of the PK II tester is availablewith cast weights. These weights are calibratedto U.S. Mean Gravity (980.000 Gals.)

Certification of Accuracy and TraceabilityA certification of accuracy and traceability toNIST is included with each floating ball typedeadweight tester. An optional certification ofaccuracy with area, mass and intrinsic correc-tion factors is available.

Weight MaterialsThe weights for all floating ball pneumatic dead-weight testers are manufactured from non mag-netic materials to prevent inaccuracy. The stan-dard Model PK II, RK and HK weights are manufac-tured from 300 series stainless steel. The eco-nomic Model PK II uses cast weights manufac-tured of a hard nonmagnetic zinc alloy.

Interchangeable WeightsThe cast weights for the economy Model PK IItester are interchangeable and lost or damagedweights may be replaced. AMETEK maintains apermanent file on all deadweight testers. All dam-aged or lost stainless steel weights can bereplaced to the original mass when manufac-tured. In most instances, it is recommended thatthese testers be recalibrated prior to weightreplacement.

24

4.4.1 AMETEK Model PK II Portable DeadweightTesterThe AMETEK Model PK II floating ball deadweighttesters are pictured on 4-10. These instrumentsare high accuracy instruments designed for bothlaboratory and portable operation. This instru-ment is self contained in a rugged ABS plasticcase designed for “Closed case operation”, elimi-nating wind problems. The weights are individu-ally stacked vertically in pockets in the case forease of operation. Specifications and features ofthese testers are as follows:

Pressure Range:4" H2O to 30 PSI


Accuracy:+0.05% of indicated reading+0.025% or +0.015% of indicatedreading (optional)

Features• Floating Ball• Self Regulating• Rugged Ceramic Measuring Ball• Quick Setup and Operation• Non Contaminating Test Fluid• Tripod Mount is Standard• Closed Cover Operation• Ball Valves for Inlet and Outlet• Stainless Steel weights Calibrated to

Local or International Standard Gravity• Interchangeable Weights on Economical

Model Tester

Figure 4-10Model PK II Pneumatic Deadweight Tester

25

4.4.2 AMETEK Model RK Portable Floating BallDeadweight TesterThe AMETEK Model RK floating ball deadweighttesters are pictured on Figure 4-11. These instru-ments are high accuracy instruments designedfor both laboratory and portable operation. Thisinstrument is self contained in a cast metal basewith an ABS plastic cover. It is self contained withweights for pressures to 50 PSI (3.5 bar), withadditional cases to carry the remaining weights.The weights are stacked vertically in the casesfor ease of operation. Specifications and fea-tures of these testers are as follows:

Pressure Range:4" H

2O to 300 PSI

Accuracy:+0.05, +0.025% or +0.015% of indicatedreading

Repeatability:+0.005% of output reading

Temperature Coefficient:0.00167% output pressure per °C

Models10 English, 15 metric

Features• Floating Ball• Self regulating• Rugged Ceramic measuring ball• Quick Setup and Operation• Non Contaminating Test Fluid• Tripod Mount is Standard• Ball Valves for Inlet and Outlet• Stainless Steel Weights

Figure 4-12Model HK Pneumatic Deadweight Tester

4.4.3 AMETEK Model HK High Pressure FloatingBall Deadweight TesterThe AMETEK Model HK floating ball pneumaticdeadweight testers are pictured on Figure 4-12.These instruments are high accuracy instru-ments designed for laboratory operation. All in-struments are supplied with stainless steelweights with a protective storage case. Thetester itself is supplied with a laminated woodgrained storage case. Specifications and fea-tures of these instruments are as follows:

Pressure Range:Standard 10 to 1000 PSI (0.7 to 69 bar)Optional 100 to 1500 PSI (6.9 to 103.4 bar)


Repeatability:+0.005% of Output Reading

Temperature Coefficient:0.00167% output pressure per °C

Models5 English, 3 metric

Features:• Floating Ball• Self Regulating• Rugged Ceramic Measuring Ball

Figure 4-11Model RK Pneumatic Deadweight Tester

26

• Quick Setup and Operation• Non Contaminating Test Fluid• Ball Valves for Inlet and Outlet• Pre-fill Valve• Stainless Steel Weights Calibrated for

Local International Standard Gravity

4.4.4 AMETEK Model PK II Medical Floating BallDeadweight TesterThe AMETEK Model PK II Medical floating balldeadweight tester, shown on Figure 4-12, is aself regulating, portable, primary type pressurestandard designed to provide an accurate pres-sure calibration standard for physiologic pres-sure ranges. Standard hospital wall oxygen out-lets or therapeutic oxygen pressure regulatorsprovide a convenient source of supply pressure.The instrument will also operate on compressedair or dry nitrogen. Specifications and features ofthese testers are as follows:

Pressure Range:10 to 325 mm Hg

Figure 4-12Model PK II Medical

Pneumatic Deadweight Tester

Accuracy:+0.05%, +0.025% or +0.015% of IndicatedReading

Repeatability:+0.005% of indicated reading

Temperature Coefficient:0.00167% of output pressure per °C

Features• Floating Ball• Self regulating• Rugged Ceramic Measuring Ball• Quick Setup and Operation• Non Contaminating test fluid• Closed Cover Operation• Oxygen Swivel Valve and Hose on

Inlet, 1/8" Male Luer Lock on Outlet• Stainless Steel Weights Calibrated to

Local or International Standard Gravity

27

4.5 Care of AMETEK Deadweight Testers

4.5.1 Care of AMETEK Piston & Cylinder TypeDeadweight TestersThe United States National Institute of Standardsand Technology has suggested that7 “For pistongages for use as plant or laboratory standards(Accuracy within 0.1%), the following points arepertinent to the instrument performance:

1. The weights should be non magnetic, solid,preferably of a hard, nonporous metal, such asbrass or stainless steel. They should mesh accu-rately, and should have a smooth finish, prefer-ably polished.

2. The piston should turn freely in the cylinder.When the weights are loaded on top of the pistonand set into rotation, they should continue torotate for several minutes. Some instrumentshave means for rotating or oscillating the piston.These devices should produce negligible longitu-dinal thrust.

3. When the gauge is connected to a leaktightsystem, and the piston is set into rotation, thepiston should fall slowly as a result of the leakbetween it and the cylinder. At maximum pres-sure, the fall rate of a good instrument should beless than 0.1 inch per minute. We regard this asthe most important single index of quality.

4. The structure should be such that the level ofoil surrounding the piston is known at all positionsof the piston, in order to correct for air buoyancy.The weight of oil displaced by the submergedportions of the piston, and the weight of airdisplaced by the rest of the piston and load, mustbe subtracted from the load. The pressure thusmeasured is that which obtains at the level of theoil surface. If the reference level of the gaugebeing calibrated is higher or lower that this level,corrections should be made for the head of oil,about 0.03 PSI (2 mbar) for each inch of differ-ence in height.

5. Occasionally the direction of rotation will affectthe pressure developed by the piston gauge. Thishas been attributed to helical tool marks on thepiston or cylinder or guide bearing. Observationsshould be taken, at least occasionally, reversingthe direction of rotation, but changing nothingelse.

6. Erratic behavior has sometimes been observedif the weights are stacked off center. There maybe fluctuation of pressure in synchronism withthe rotation, or a change of pressure whichdepends on the speed of rotation or a change ofpressure which depends on the speed of rota-tion, but not the direction. These problems aremost severe in instruments with the weightsstacked in a tall pile on top of the piston. Asuspended, non rotating load may oscillate ab-normally when the piston is rotated or oscillatedat a particular speed. Such speeds should beavoided.”

AMETEK has found the following items to beimportant factors affecting performance and lifeof piston gauges8:

1. FLUID CLEANLINESS-Deadweight testers areclosely fit devices with clearances of 1 or 2millionths of an inch and finishes exceeding 1micro inch. The principle source of fluid contami-nation is the interior of instruments being cali-brated. When pressure is relieved, contaminantswithin the instrument being calibrated are ex-hausted into the deadweight tester. If instrumentscontain dirty fluid or particles, they should beflushed prior to testing. Individual users mustconsider the cleanliness of the instruments whenestablishing the frequency of changing the testfluid.

2. FLUID LUBRICITY-The test fluid creates alubricating film between the closely fit piston andcylinder. The ability of the fluid to lubricate at highpressure is an important criteria in determining thewear rate of pistons and cylinders. A high pres-sure spindle type oil such as AMETEK AAA TesterOil affords maximum protection. Other test fluidssuch as water which is used within the AMETEKType T testers afford satisfactory operation butreduce service life slightly. Silicone oils must beavoided as they tend to crystallize at high pres-sures causing the piston and cylinder to seize.

28

3. STORAGE TECHNIQUES-The storage of theclose fitting piston and cylinder assembly is im-portant to continued service life. AMETEK hasfound that the piston and cylinder should beseparated and individually coated with light oilduring storage. After storage, both the piston andcylinder must be cleaned prior to use.

4. INSTALLATION-It is important to install thepiston and cylinder so that the piston does notbind within the cylinder.the cover that securesthe piston and cylinder within the AMETEK TypeT & R testers contains eight cap screws. Thesescrews should be tightened in a criss-crosspattern to minimize distortion. Frequent changesof the piston and cylinder should be avoidedbecause of the attendant risk of damage. If theusers task involves frequent calibration of instru-ments, the AMETEK Dual Column Assembly shouldbe considered. in this device, the high and lowpressure pistons are separately mounted andrange changes are accomplished by adjustingthe high pressure valves.

5. CLEANING- The most important care that canbe given to a deadweight piston and cylinderassembly is cleaning. Cleaning is particularityimportant during the initial use of pistons andcylinders. Although efforts are made to clean allparts after manufacture, some abrasive particlesremain within the pores of the material. afterapplication of pressure, these particles are dis-charged. It is recommended that new pistons andcylinders be cleaned prior to each usage for thefirst month of operation. Monthly cleaning there-after is satisfactory. AMETEK recommendedcleaning procedures are included in paragraph4.6.1.

4.5.2 Care of AMETEK Floating Ball TypePneumatic Deadweight TestersAMETEK has found the following items to beimportant factors affecting performance and lifeof the floating ball type pneumatic deadweighttesters8.

1. AIR CLEANLINESS- The AMETEK floating balldeadweight tester contains several small diam-eter bores. Free passage of air within thesebores is important for the correct operation of thisdevice. the air used to operate the tester must beboth clean and dry to preclude problems. M&Grecommends the use of instrument quality air asdefined by Instrument Society of America Speci-fication ISA-S7.3 “Quality Standard for Instru-ment Quality Air”.

2. CLEANING- Cleaning of the ball and nozzle isimportant prior to the use of the pneumatic deadweight tester. This cleaning is easily done byremoving the tester nozzle and cleaning the ball,nozzle and nozzle body with a non residuesolvent such as grain alcohol and a lint free clothor paper. Tester weights should be cleanedperiodically with alcohol and a lint free cloth ortissue paper.

3. LEVELING- For proper operation, the air streamexhausting around the ball should be equallydistributed around the periphery of the ball andthe weight carrier is not in any contact with thenozzle body. Each tester is equipped with a bulls-eye level which is pre set during manufacture.Proper level may be verified by either of thefollowing methods:

A. The tester should be leveled and operatedwith only the weight carrier (4"H2O or 10"H20) inplace. The carrier should be carefully set intorotation. This rotation should continue for at leasttwo (2) minutes. If the rotation stops prematurely,the level should be corrected.

B. The hand may be placed above the nozzle andfeeling the direction of the exhaust gas. Thisstream should be approximately vertical whenthe testeris in operation. The weight carrier maybe rotated to aid in the determination of the airstream direction.

29

It should be noted that once the tester is calibratedand the level is adjusted, the level should not bereset unless the accuracy of calibration is re-tested.

4. OPERATING TECHNIQUE- The pneumatic dead-weight tester is designed as a self regulatingdevice which produces an accurate pressureindependent of operator technique. The operatormust, however, exercise care during operation.The weight carrier is suspended from the ball onthree prongs. Care should be taken not to drop thecarrier or drop weights onto the carrier in such away as to bend one or more prongs.

5. LEAKAGE- The accuracy of the pneumatictester is seriously affected by leaks in the outputor input connections and/or within the instrumentbeing calibrated. Leaks within the output connec-tion or instrument are easily determined by clos-ing the tester output valve and determiningwhether the instrument under test shows pres-sure loss. Internal tester leaks are identified byplacing an O-ring under the ball and closing thepressure inlet. Again, pressure loss shown onthe instrument being calibrated indicates a leak.

6. TEST GAS- The pneumatic deadweight testeris calibrated for pressure accuracy with nitrogengas. The use of gases with other density willaffect the output pressure. The testers can becalibrated for some gas densities other thannitrogen on special order. Gases used with thetester must also be compatible with the materialsused for construction. AMETEK can assist users,on request, to evaluate alternate gases.

4.6 Procedures for the Use of AMETEKDeadweight Testers

4.6.1 Procedures for Cleaning HydraulicPistons and CylindersEach piston and cylinder should be cleaned priorto use. If a piston and cylinder has been storedoutside the tester, it should be recleaned prior touse. It is very important that new pistons andcylinders be cleaned repeatedly prior to use toremove all evidence of the abrasive particlesused during manufacture. Periodic recleaning ofpiston and cylinders is necessary. A lack ofsensitivity to small pressure changes is an indi-cation that cleaning is necessary.

The suggested cleaning procedure for pistonsand cylinders is as follows:

1. Carefully wipe off any visible dirt or foreignmatter from the protruding part of the piston andslowly withdraw the piston from the cylinder. Donot use force, but be sure that all dirt is removedso that the piston will slip out easily.

2. Boil both the piston and cylinder for at least 15minutes in distilled water. The cylinder bore shouldbe wiped with a small wood handled wiper suchas a cotton swab to remove all evidence of dirt.Wipe the piston dry and clean with a lint freetissue paper.

3. Rinse both the piston and cylinder in grainalcohol.

4. Wipe the cylinder bore and the piston again toremove any dirt or grit.

5. Pick up the piston by the piston cap and dip itinto clean fluid to be used in the tester, thencarefully insert the piston into the cylinder. If anyfeeling of roughness or what might be grit in theannulus area is suspected, disassemble andrepeat the cleaning procedure.

6. At the same time, the deadweight column outputpost and tubing should be drained and flushedwith a residue free solvent such as grain alcohol,then cleaned, dried and refilled using clean testfluid.

30

7. The piston and cylinder assembly then can beinstalled carefully in the mounting column.

CAUTIONDo not touch the piston and cylinder with the

fingers or other soiled or contaminatingsurfaces after cleaning. Extremely minute

particles may cause trouble in a closely fittedassembly.

4.6.2 Procedures for the Use of HydraulicDeadweight TestersThe following procedure is typical of methodsused to test pressure instruments using hydraulicdeadweight pressure testers. The user is en-couraged to modify these procedures to testcritical instrument characteristics.

The step by step procedure is as follows13:

1. Remove the pressure instrument from serviceand blowdown (Vent to atmosphere to removetrapped pressure from the pressure side of theinstrument).

2. Set up the deadweight tester and level theinstrument.

3. Connect the pressure input of the pressureinstrument to the pressure test port on the dead-weight tester.

4. Put weights on the deadweight tester carrierequal to 90 to 100% of the range of the instrumentbeing tested.

5. Build up pressure with the tester hand pumpuntil the weights float freely. Rotate the weights10 to 30 RPM. Check to be sure that no leaks arepresent.

6. If no leaks are present, proceed with thecalibration. If leaks are apparent, they must belocated and repaired before proceeding with thecalibration.

7. Remove all pressure from the tester handpump. Vent the high and low side of the pressureinstrument to atmosphere. Adjust zero on theinstrument.8. Leave the low side of the instrument open toatmosphere, open the high side of the instrumentto the deadweight tester.

9. Begin calibration by placing weights on thedeadweight tester weight carrier equal to 25% ofthe range of the instrument being tested.

10. Build up pressure with the tester hand pumpuntil the weights float freely. Rotate the weights10 to 30 RPM. Allow the instrument to settle forone (1) minute. Tap case of the instrument.Record the reading.

11. Place additional weights on the deadweighttester to read points at 50%, 75% and 100% of theinstrument range.

12. Close valve between the deadweight testerand the instrument being tested.

13. Remove weights from the deadweight testerto read 75% of the range of the instrument beingtested.

14. Open the valve between the deadweighttester and the instrument being tested. Allow thedeadweight tester weights to settle. Weightsmust be floating freely and rotating 10 to 30 RPM.Allow the tester to stabilize for one (1) minute. Tapcase of the instrument. Record reading.

15. Repeat steps 12 through 14 to read points at50% and 25% of the instrument range.

16. Remove all pressure from the instrumentbeing tested. Recheck the zero on the instrument.

31

4.6.3 Procedures for the Use of HydraulicDeadweight GaugesDeadweight gauges are used to measure pres-sures contained within an external source withhigh precision. Deadweight gauges do not con-tain a pump or any self contained source ofpressure.

It is important to emphasize that all tubing, fittingsand chambers within the deadweight gauge berated at pressures higher than any pressuresource that may be attached to the gauge. dead-weight testers should not be used as deadweightgauges unless so certified by the manufacturer.

The following procedure is typical of the methodsused for hydrostatic testing on pipelines and wellstatic pressure testing.

The step by step procedure is as follows13:

1. Set up the deadweight gauge and level theinstrument.

2. Connect the pressure port on the device to betested to the inlet port of the deadweight gauge.A shut off valve should be placed at the pressureport and at the inlet of the deadweight gauge. Aseparator may be connected between the pres-sure tap and the inlet of the deadweight gauge.Some deadweight gauges contain isolation de-vices to prevent contamination from coming intocontact with the piston and cylinder (See 4.3.1 ,Elastomer Membrane).

3. Shut the valve on the deadweight gauge inlet,open the valve on the pressure port. Check forleaks; if leaks are present they must be repairedbefore proceeding, if no leaks are present pro-ceed with the test.

4. Place weights on the deadweight gauge equalto 100% of the pressure within the line or well tobe tested.

5. Open the inlet valve to the deadweight gaugeslowly. Rotate the weights 10 to 30 RPM.

6. Observe the direction of the float of the pistonand cylinder. If the piston is falling, weight mustbe removed. If the piston is rising, weight must beadded. Weights should be added or deleted untilequilibrium within the mid one-third of the pistonand cylinder float range is achieved.

7. Sufficient time must be allowed for pressureconditions to stabilize. The higher the pressurethe longer the period required for the conditionsto stabilize.

8. Close the valve on the pressure port on thedevice being tested. Open the deadweight gaugevent valve to relieve pressure. Close the inputvalve on the deadweight gauge.

9. Remove the weights on the deadweight gauge.Use the manufacturer’s data to determine themeasured pressure. For example:

Piston & Cylinder Area- 0.01 inches2

Weight of Piston/Cylinder Carrier Assembly - 0.5 lbsWeights on Carrier: 2 - 10 lb

1 - 5 lb3 - 1 lb4 - 0.1 lb

Pressure =

=

Weight

Area

28.9 lbs

0.01 inches2

= 2890 lbs/inch2 (PSI)

32

10. The pressure measured by a piston gaugeloaded with a particular value of weights isproportional to the local value of gravity. Once thelocal value of gravity is known (See paragraph1.4), the nominal values can be converted toactual values by the equation:

Where:

M = Massg

l= Local value of gravity in cm/sec2

gs = 980.665 (International Standard Gravity)

Under standard conditions, the pounds weightis equal to the pounds mass.

Continuing the example:

Local Gravity = 979.628 Gals (cm/sec2)Calibrated Gravity of Deadweight GaugeWeights = 980.665 Gals.

Force =M x g1

gs

4.6.4 Procedures for the Use of Floating BallType Pneumatic Deadweight TestersA suggested step by step procedure to be usedwhen calibrating a flow measuring instrument(Meter) with a PK tester is as follows14:

1. Remove meter from service and blowdown.

2. Level PK and connect gas supply (30-35 PSIG)to inlet of the PK. PK should be close to meter toeliminate long tubing line on outlet. (Be sure skirtof the weight table is not dragging on the nozzlebody.)

3. Connect the PK outlet to the high side of themeter. vent the low side of the meter to atmo-sphere. Turn on the supply gas to the PK and openthe outlet valve. provide filtered gas, free ofliquids.

4. Put weight equal to above 90% to 100% of themeter range of the PK and allow the pen to travelto a full stop on the chart (At least one minute).

5. Close the outlet of the PK and observe thereading on the chart for about two or moreminutes.

6. If no leaks are present, proceed with thecalibration. If leaks are present they must belocated and repaired before proceeding with thecalibration.

7. With the PK outlet valve closed, vent the highand low side of the meter to atmosphere andadjust the zero on the chart.

8. Leave the low side of the meter open toatmosphere, connect the high side to the PK.

9. Begin meter calibration by placing the table andthe necessary weights an the PK to read about6" on a 50" meter (12" on a 100" Meter). Allowmeter and PK to settle in for at least one full minuteminimum time before pulling point on the chart.

(Note: Use weights which allow reading to fall onthe narrow line of the calibrating chart)

10. Place additional weights in increments of 10"on table to read points at 16",26" and 46" on a 50"

P = 2890 x979.628

980.665

P = 2886.944 lbs/inch2 (PSI)

33

meter (32", 52", 72" and 92" on a 100" meter,etc).Care should be taken when placing weights onthe table to see that all points are reached with themeter travel in the up direction, in other words donot overshoot the points. Again, wait a minimumof one full minute before pulling any points.

13. Check down points at 42", 32", 22" and 12" ona 50" meter (84",64",44", and 24" on a 100" etc.).Care should be taken to remove weights in theproper manner so that all down points are reachedfrom the downward travel of the meter. This isimportant as it will tell if there are binds in themeter.

14. Recheck zero of the meter.

34

Chapter 5

Principle of OperationManometers derive pressure by the combinationof a height differential of a liquid column and thedensity of the fluid within the liquid column. The Utype manometer, which is considered as a pri-mary pressure standard, derives pressure utiliz-ing the following equation:

P = P2 - P1 = hw ρ g

Where:

P = Differential pressureP1 = Pressure applied to the low pressure

connectionP

2= Pressure applied to the high pressure

connectionhw = is the height differential of the liquid

columns between the two legs of themanometer

ρ = mass density of the fluid within thecolumns

g = acceleration of gravity

5.1 Types of Manometers

5.1.1 U Tube Manometers“The principle of operation of the U type manom-eter is shown on Figure 5-1. It is simply a glasstube bent to form the letter U and partially filledwith some liquid. With both legs of the instrumentopen to atmosphere or subjected to the samepressure, Figure 5-1, the liquid maintains exactlythe same level or zero reference. As illustrated onFigure 5-2, if a pressure is applied to the left sideof the instrument, the fluid recedes in the left legand raises in the right leg. The fluid moves until theunit weight of the fluid as indicated by “H” exactlybalances the pressure. This is known as hydro-static balance. The height of fluid from one sur-face to the other is the actual height of fluidopposing the pressure.

The pressure is always the height of fluid fromone surface to the other regardless of the shapeor size of the tubes, as illustrated in Figure 5-3.The left hand manometer has a uniform tube, thecenter one has an enlarged leg and the right onehas an irregular leg. Manometers at the top areopen to atmosphere on both legs so the indicatingfluid level in both legs is the same. Imposing anidentical pressure on the left leg of each manom-eter, as shown on Figure 5-4, causes the fluidlevel in each manometer to change. Because ofthe variations in volume of the manometer legs,the distances moved by the fluid columns aredifferent. However, “H” the total distance be-tween the fluid levels, remains identical in thethree manometers”15.

5.1.2 Inclined Tube Manometers“Many applications require accurate measure-ments of low pressure such as drafts and verylow differentials. To better handle these applica-tions, the manometer is arranged with the indicat-ing tube inclined, as in Figure 5-5, providing forbetter resolution. This arrangement can allow 12”of scale length to represent 1" of vertical height.With scale subdivisions, a pressure of 0.00036PSI ( 1/100 inch of water) can be read”15.

��

Figure 5-1U-Tube Manometer Zero Pressure

35

5.1.3 Well Type ManometersThe well type manometer is illustrated on Figure5-6. In this design, the pressure is applied to a fluidwell attached to a single indicating tube. As thefluid moves down in the well, the fluid displacedinto the smaller indicating leg of the manometer.This permits direct reading on a single scale.

The well type manometer utilizes the principle ofvolume balance wherein the fluid displaced fromthe well is equal to the added fluid in the smallerindicating column. The well area and the internaldiameter of the indicating type must be carefully

5.2 Intrinsic Correction Factors

5.2.1 Fluid Density CorrectionManometers indicate the correct pressure at onlyone temperature. This is because the indicatingfluid density changes with temperature. If wateris the indicating fluid, an inch scale indicates oneinch of water at 4°C only. On the same scalemercury indicates one inch of mercury at 0°Conly. If a reading using water or mercury is takenat 20°C then the reading is not an accuratereading. The error introduced is about 0.4% ofreading for mercury and about 0.2% of readingfor water. Since manometers are used at tem-peratures above and below the standard tem-perature, corrections are needed. A simple wayfor correcting for density changes is to ratio thedensities.

(Standard) ρo g ho = (Ambient) ρt g ht

Where:

ho = Corrected height of the indicating fluid tostandard temperature

ht = Height of the indicating fluid at thetemperature when read

ρo = Density of the indicating fluid at standardtemperature

ρt = Density of the indicating fluid when read

Using this method is very accurate, when den-sity/temperature relations are known. Data isreadily available for water and mercury.

Density (g/cm3) as a function of temperature (°C)for mercury:

= 13.556786 [1 - 0.0001818 (T - 15.5556)]

controlled to insure the accuracy of the instru-ment.

The well type manometer does not fulfill therequirements of a primary standard as describedin paragraph 1.5 and can be considered as oneform of a secondary standard.

Figure 5-2U-Tube Manometer Pressure Applied

ρt

ρo h

o = x ht

36

Density ( g/cm3) as a function of temperature forwater:

= 0.9998395639 + 6.798299989 x 10-5 (T)- 9.10602556X10-6 (T2) + 1.005272999 x10-7 (T3) - 1.126713526 x 10-9(T4) +6.591795606 x 10-12 (T5)

For other fluids, manometer scales and fluiddensities may be formulated to read inches ofwater or mercury at a set temperature. Themanometer still only reads correct at one tem-perature, and for precise work the temperaturecorrections cannot be overlooked.15

5.2.2 Gravity CorrectionThe need for gravity corrections arises be-cause gravity at the location of the instrumentgoverns the weight of the liquid column. Likethe temperature correction, gravity correctionis a ratio.

(Standard) ρogoho = (Ambient) ρtgtht

Where:

go = International Standard Gravity(980.665Gals.)

gt = Gravity at the instrument’s location (InGals.)

A 10° change in latitude at sea level will introduceapproximately 0.1% error in reading. At the Equa-tor (0° Latitude) the error is approximately 0.25%.An increase in elevation of 5000 feet (1524 m) willintroduce an error of approximately 0.05%.

For precise work you must have the value of thegravity measured at the instrument location. Grav-ity values have been determined by the U.S. Coastand Geodetic Survey at many points in the UnitedStates. Using these values, the U.S. GeodeticSurvey may interpolate and obtain a gravity valuesufficient for most work. To obtain a gravityreport, the instruments latitude, longitude andelevation are needed. Similar agencies are avail-able in countries outside the United States. Con-tact local authorities for the agency and proce-dures to determine local gravity.

Figure 5-3U-Tube Manometers Non-uniform Tubes, No Pressure

ht =ρogo

ρtgtx h

o

37

Where a high degree of accuracy is not neces-sary and values of local gravity have not beendetermined, calculations for differences in localgravity can be obtained. Gravity at a knownlatitude is:

Gx= 980.616 [1 - 0.0026373 cos(2x) + 0.0000059cos2(2x)]

Where:

Gx = gravity value at latitude x, sea level(cm/sec2)

x = latitude (degrees)

The relationship for inland values of gravityat elevations above sea level is:

Gt = G

x - 0.000094H + 0.00003408(H - H

1)(cm/sec2)

Where:

H = Elevation (feet) above mean sea levelH1 = Average elevation (feet) of the general

terrain within a radius of 100 miles ofthe point15

5.2.3 Pressure Medium Head CorrectionCommonly, a differential pressure is measured bythe height of the fluid column. Actually the differ-ential pressure , measured by the indicating fluidheight, is the difference between the density ofthe fluid column and the density of equal height ofthe pressure medium.

The relationship is:

ρogoho + ρpmgtht = ρtgtht

ho= x h

t

Where:

ρpm = density of the pressure medium

The significance of the pressure medium correc-tion effect on the manometer reading varies withthe indicating fluid and pressure medium. Whetherthis correction is necessary depends upon theuser’s accuracy requirements. The most com-mon pressure medium is air. Not correcting for airover water yields an error of 0.12% (using the

gt (ρ

t-ρ

pm)

goρo

Figure 5-4U-Tube Manometers Non-uniform Tubes, Pressure Applied

38

density of air as 0.0012 g/cm3). In precise work,air density can be determined exactly knowingthe temperature, pressure and relative humidityof the air. The correction for air over mercury isextremely small (0.008% error) and therefor mayusually be ignored. Another application, oftenused in flow applications, is water over mercury.The pressure medium correction in this situationis mandatory. An error of 7.4% is introduced if thecorrection is not applied. In many instances ma-nometer scales can be designed with this correc-tion built-in.

5.2.4 Scale Corrections“Another factor governing manometer’s accu-racy is the scale. As with indicating fluids, tem-perature changes affect the scale. At highertemperatures the scale will expand and gradua-tions will be further apart. The opposite effect willoccur at lower temperatures. All Meriam scalesare fabricated at a temperature of 22°C (71.6°F).A 10°C shift in temperature from that temperaturewill induce an error in the reading of about 0.023%in an aluminum scale. All Meriam scales are madeof aluminum.

ho = ht [1 + α( T - To)]

Where:

α = Coefficient of linear expansion for thescale material (0.0000232/°C foraluminum)

T = the temperature when the manometerwas read

To = temperature when the scale wasmanufactured

5.2.5 Compressibility, Absorbed Gases andCapillary ConsiderationsCompressibility of indicating fluids is negligibleexcept in a few applications. For compress-ibility to have an effect, the manometer must beused in measuring high differential pressures.At high differential pressures the fluid shrink-age (Increase in density) may begin to beresolvable on the manometer. At 250 PSI thedensity of water changes approximately 0.1%.Depending upon accuracy requirementscompressibility may or may not be critical. Therelationship between pressure and density ofwater is as follows:

ρ = 0.00000364 p + 0.9999898956

Where:

ρ = density of water(g/cm3) at 4°C andpressure p

p = pressure in PSIA

Figure 5-5Inclined Tube Manometer

39

Since the need to correct is very rare, otherindicating fluid’s compressibilities have not beendetermined. Mercury’s compressibility is negli-gible.

Absorbed gases are those gases found dis-solved in a liquid. The presence of dissolvedgases decreases the density of the liquid. Air isa commonly dissolved gas that is absorbed bymost manometer fluids. The density error ofwater fully saturated with air is 0.00004% at20°C. The effect is variable and requires con-sideration for each gas in contact with a particu-lar fluid. Mercury is one exception in whichabsorbed gases are not found. This makesmercury an excellent manometer fluid in vacuumand absolute pressure applications.

Capillary effects occur due to the surfacetension or wetting characteristics between theliquid and the glass tube. As a result of surfacetension, most fluids form a convex meniscus.Mercury is the only fluid that does not wet theglass, and consequently forms a concave me-niscus. For consistent results, you must al-ways observe the fluid meniscus in the sameway, whether convex or concave. To helpreduce the effects of surface tension, manom-eters should be designed with large bore tubes.This flattens the meniscus, making it easier toread. A large bore tube also helps fluid drainage.The larger the bore the smaller the time lag whiledrainage occurs. Another controlling factor isthe accumulation of corrosion and dirt on theliquid surface. The presence of foreign materialchanges the shape of the meniscus. With mer-cury, it helps to tap or vibrate the tube to reduceerror in the readings. a final note to capillaryeffects is the addition of a wetting agent to themanometer fluid. Adding the wetting agent helpsin obtaining a symmetrical meniscus.15

Figure 5-6Well-type Manometer

5.2.6 Parallax (Readability)In order to achieve consistent results, the level ofthe meniscus on a manometer must be read withthe eyes level to the meniscus. Placing the eyeslevel with the meniscus eliminates reading distor-tions caused by angle of reading,parallax, etc. Ifa mirror back is available, it will aid in placing theoperators eyes in the proper position beforetaking a reading.

To duplicate the factory calibration procedure,read the lowest indicated liquid level as measuredby the hairline at which the original zero was set.

40

Chapter 6

A secondary comparison pressure standard isany device that measures pressure which doesnot meet the criteria of a primary pressure stan-dard. Secondary pressure standards thereforecomprise both mechanical and electronic instru-ments with varying rates of measurement accu-racy.

In general, secondary pressure standards havethe following advantages in comparison to pri-mary pressure standards:

1. Faster and easier to use2. Usually no measurement corrections3. Non-incremental measurements4. Generally less expensive

In general, secondary pressure standards havethe following disadvantages in com-parison toprimary pressure standards:

1. Must be periodically recalibrated bya primary pressure standard traceable tonational standards

2. Pressure measurements cannot bereduced to measurements of mass, lengthor temperature

6.1 Types of Secondary Pressure Standards

6.1.1 Mechanical Pressure GaugesThe most commonly used secondary pressurestandard is the mechanical pressure gauge. Thisinstrument utilizes a hollow tube (Bourdon) that isformed into the shape of a C, a spiral or a helix.When pressure is applied within the tube, the tubetends to straighten out. This movement is trans-ferred to a display pointer either through a gearrack and pinion mechanism or a direct drive.

The measurement accuracy of mechanical pres-sure gauges varies generally from 3% full scaleto 0.10% full scale. The higher accuracy gaugesare known as “Test gauges”. The hollow tubesare generally manufactured from nonferrous ma-terials such a beryllium copper, Monel, stainless

steel and Nispan “C”. Other mechanical pressuregauges use diaphragms and capsules to expandwith the applied pressure.

The bourdon tube can also be constructed froma stable material such as quartz to achieve higherlevels of measurement accuracy. Accuracieswill vary with the manufacturer, but accuraciesas high as 0.015% Full Scale are available.

6.1.2 Electronic Pressure GaugesThe electronic pressure standards comprise alarge, divergent group of instruments with widelyvarying features and accuracy. Each of theseinstruments incorporates a pressure transducer.the function of which is to convert an inputpressure signal to an electronic signal such aresistance, voltage, current etc. The most com-mon transducers incorporate an electronic straingauge and a wheatstone bridge generally locatedon a silicone chip. Other transducers incorporatepiezo electric, vibrating wire or cylinders andother technologies.

The advantages of these instruments is that theydisplay the pressure on easily read digital dis-plays, are easy to use and are generally small,light in weight and portable. On the negative side,the secondary standards are generally not asaccurate as primary standards and the accuracyis seriously affected by temperature variations.In most cases, the overall accuracy of the instru-ment is divided into several elements such as:

1. Accuracy, % Full Scale or % IndicatedReading

2. Precision, Repeatability, Hysteresis3. Temperature Effect4. Resolution

It is important when selecting a secondary pres-sure standard, therefore, to consider first theapplication task to be performed and then toevaluate the overall accuracy capabilities of eachinstrument.

��

41

6.2 Overall Application AccuracyMany different elements must be combined toestablish the overall application accuracy of aninstrument when performing a specific test. Theseelements include the accuracy measurement,repeatability, hysteresis, linearity, temperatureaffect, resolution and laboratory uncertainty. Eachof these elements is described within the follow-ing paragraphs.

6.2.1 Application TaskWhen evaluating the overall accuracy of aninstrument, the first item that must be consideredis the application or task for which the pressurestandard will be used. Elements that should beconsidered are the minimum and maximum pres-sures to be measured and the minimum andmaximum ambient temperatures within which theinstrument will be used.

For purposes of illustration we have selected anapplication where the customer wishes to mea-sure pressure over a range of 4" H2O (20°C) to200" H2O with ambient temperature variationsfrom 20° to 110°F (-7° to 43°C).

6.2.2 Accuracy MeasurementThe measurement accuracy of instruments isgenerally expressed as “Per Cent of Full Scale”

or “Percent of Indicated Reading”. As illustratedon Figure 6-1 if the accuracy of a 200 inch ofwater column capacity instrument is stated as+0.05% full scale, the guaranteed accuracy is+0.1 inch of water column throughout the oper-ating range. If the instrument is used at 10 inchesof water, the accuracy of measurement is 1%. Ifon the other hand the accuracy of the same 200inch of water column instrument is stated as+0.05% indicated reading, the accuracy at 200inches of water column would still be +0.1 inch ofwater but the accuracy at 10 inch of watercolumn would be +0.005 inch of water column.

As an example of this calculation, consider anelectronic pressure gauge (A) with a specifiedrange from 0 to 280" H2O, a stated accuracy of0.1% full scale plus 1 Least Significant Digit. Forthe application of 4" H

2O to 200" H

2O, the accuracy

portion of the overall application accuracy at the4" and 25" H2O test points would be:

Specified error:

280" H2O (Full Scale) x 0.1% = 0.28" H

2O

Least Significant Digit = 0.1" H2OTotal = 0.28 + 0.1 = 0.38" H2O

Figure 6-1Accuracy % Full Scale versus % Indicated Reading

42

Application error @ 4" H2O:

0.38" H2O / 4 x 100 = 9.5 %


0.38" H2O /25 x 100 = 1.52%

As a further example of an electronic pressuregauge, consider an instrument (B) with an oper-ating range of 0 to 5 PSIG, a stated accuracy of+0.03% full scale plus 0.07% of the indicatedreading plus 1 Least Significant Digit. The accu-racy portion at the 4" and 25" H2O test pointswould be:

Specified error:

5 PSI x 27.72" H2O/PSI x 0.03% =0.04158" H

2O

0.07% Indicated Reading x 4" H2O = 0.0028" H

2O

or0.07% indicated Reading x 25" H2O = 0.0175" H2O

Least Significant Digit = 0.001" H2O

Total @ 4" H2O = 0.04538" H2OTotal @ 25" H2O = 0.06008" H2O


0.04538" H2O / 4 x 100 = 1.1345%


0.06008" H2O /25 x 100= 0.24032%

Consider as a further example a deadweighttester with a specified range of 250" H2O and astated accuracy of 0.015% of the indicated read-ing.

Since the accuracy is stated as percent of indi-cated reading, the application error at both the 4"and 25" H2O test points would be 0.015%.

6.2.3 Hysteresis, Linearity, RepeatabilityAs illustrated on Figure 6-2, hysteresis is definedas the maximum difference for the same mea-sured quantity between the upscale and thedown scale readings during a full range traversein each direction. Linearity is defined as themaximum deviation of the instrument character-istics from a corresponding point on a specifiedstraight line. If the straight line does not passthrough zero, the difference between the loca-tion of the line and zero is defined as “ZeroOffset”. Repeatability is defined as the closenessof agreement among a number of consecutivemeasurements of the output of an instrument forthe same value of the input under the sameoperating conditions, approaching the measure-ment from the same direction, and for full rangetraverses. This term is identical to the statistical“Precision” of a measurement. In general, thecombined error from hysteresis, linearity andrepeatability is stated as a single error percent-age in instrument specifications.

Continuing the example, the hysteresis, linearityand repeatability are included within the accu-racy specification of most of the electronic digitalgauges, therefore the incremental contribution ofboth the example (A) and (B) gauges to the overallaccuracy of these instruments is zero.

For the deadweight tester, the repeatability orprecision is usually specified as a separate por-tion of the accuracy statement. For the example,the repeatability of the deadweight tester is es-timated at 0.005% of the indicated reading.

6.2.4 Temperature EffectThe “Temperature Coefficient” is a measure of thechange of a parameter as a result of a change inambient temperature. It is generally expressed asa percentage change in the parameter per degreeof change ( Percent/ °C or Percent/°F).

The electronic digital gauge (A) example is nottemperature compensated. The temperature co-efficient is specified as 0.01% full scale per °F.The calibration temperature is 68°F. The maximum

43

application temperature difference is:

68°F - 20°F = 48°F

Temperature error = 48°F x 0.01% = 0.48% fullscale

0.48 % full scale x 280" H2O (Full scale) = 1.344" H2O

Application temperature error @ 4" H2O =

1.344" H2O/ 4" H

2O(Low Application Pressure) x 100

= 33.6 %

Application temperature error @ 25" H2O =

1.344" H2O/25" H

2O x 100 = 5.376 %

Electronic pressure gauge (B) is temperaturecompensated over a range of 20° to 130°F (-7to 54°C). The temperature coefficient portion ofthe overall efficiency is zero.

The deadweight tester is not temperature com-pensated but has a very low temperature coef-ficient. The temperature coefficient is 0.0009278%per °F. The calibration temperature is 73.4°F.

The maximum application tempemperature differ-ential is:

73.4°F - 20°F = 53.4°F

Temperature error =

53.4°F x 0.0009278% = 0.049%

Application temperature error =

(4" and 25" H2O) = 0.049%

6.2.5 ResolutionResolution is defined as the least incrementalvalue of input or output that can be detected,caused, or otherwise discriminated by the mea-suring device. This parameter is best illustratedby the following example of a 100 PSI capacity

Figure 6-2Precision- The Random Error Observed in a set of Repeated Measurements

44

instrument. At 100 PSI, if the instrument has 3digits, it can display only 99.6 to 100.4 PSI with apotential resolution error of .4%. If the display is4 digits, the error is reduced to 99.96 to 100.04 PSIor 0.4%. A five digit display reduces the error to99.996 to 100.004 PSI or 0.004%. However, withthe same instrument, if the measurement is mea-suring 10 PSI the 3 digit measurement is 9.6 to 10.4PSI or 4%, the 4 digit instrument is 9.96 to 10.04PSI or 0.4% and the 5 digit measurement is 9.996to 10.004 PSI or 0.04%.

Resolution error is not additive to other errorparameters but the overall error can never be lessthan the instrument resolution at the lowest pointof measurement.

The electronic digital pressure gauge (A) has a 31/2 digit display with one decimal maximum (0.1"H2O). The maximum resolution error is therefore:

4" H2O:

0.1" H2O / 4" H2O x 100 = 2.5%

25" H2O:

0.1 H2O / 25" H2O x 100 = 0.4%

Electronic pressure gauge (B) has a 5 digit dis-play with a 3 decimal maximum (0.001" H2O).Maximum resolution error is therefore:

4" H2O:0.001" H2O / 4" H2O x 100 = 0.025%

25" H2O:0.001" H2O /25" H2O x 100 = 0.004%

The deadweight tester does not have a display,therefore the resolution error component of theoverall application accuracy is not applicable.

6.2.6 StabilityStability is defined as the ability of an instrumentto maintain the measurement accuracy over aspecified time period. All instruments must becalibrated after some time interval to assure thatthe accuracy performance has not degraded.The shorter the stability time period, the morefrequently the instrument must be recalibrated.Recalibration of some instruments can be done bythe customer while other types must be returnedto the manufacturer.

The stability or frequency of calibration require-ment is seldom included within the advertisedliterature for secondary pressure standards.Users frequently must consult the manufacturerfor recommended intervals.

6.2.7 Laboratory UncertaintyLaboratory uncertainty is defined as the ability orcertainty of a laboratory to measure the specificparameters necessary to establish the variouselements of the instrument accuracy. Currentlythis measure is only applied to national standardslaboratories such as United States National Insti-tute of Standards and Technology and a fewmanufacturers of equipment for primary stan-dards laboratories. A world wide program isunderway to accredit laboratories with regard totheir capability. This measure should becomemore important in the future.

45

6.2.8 Overall AccuracyIn order for the user to determine the overallaccuracy of measurement for each application itis necessary to consider all of the elementscontributing to the accuracy of measurementtogether with the minimum and maximum pres-sure and ambient temperatures for the plannedapplication.

A Worksheet for Estimated Overall Accuracy isshown on Figure 6-3. This worksheet can beused to evaluate alternate instruments for indi-vidual applications.

For the three instruments that have been used forthe example, the overall accuracy at the 4 and 25inch H2O test conditions are as follows:

Accuracy Deadweight Tester Digital (A) Digital (B) Element 4"H2O 25"H2O 4"H2O 25"H2O 4"H2O 25"H2O

Accuracy 0.015% 0.015% 9.5% 1.52% 1.1345% 0.24032%

Hysteresis 0.005% 0.005% - - - -

Temperature 0.049% 0.049% 33.6% 5.376% - -

Total 0.069% 0.069% 43.1% 6.986% 1.1345% 0.24032%

Resolution - - 2.5% 0.4% 0.025% 0.004%

Application Accuracy 0.069% 0.069% 43.1% 6.986% 1.1345% 0.24032%

Table 6-1Overall Accuracy and Characteristics of Test Devices

46

Figure 6-3Worksheet Estimated Overall Accuracy for

Pressure Calibration Instruments

Instrument Model Range

I. Application

Measurement Pressure: High Low

Ambient Pressure: High Low

II. Instrument Accuracy

A. Accuracy %: � Indicated Reading �� Full Scale � +1 LSD

Pressure Errors: Indicated Reading

Full Scale

Least Significant Digit

TotalError

LowMeasuredPressure

x 100 %

B. Precision, Hysteresis, Repeatability: � Included in Accuracy

C. Temperature Compensated � YES � NO Range

Temperature Effect:

to

% °F

� Full Scale � Indicated Reading Cal Temp.

Temperature Differential: High Temp. - Cal Temp. or,

Cal Temp. - Low Temp.

1. If % Full Scale: Temp. Diff. x Temp. Effect =

2. If % Indicated Reading:Temp. Diff. x Temp. Effect =

% FS

% FS x FS

Low Measured Pressure

%

%x 100 =

III. Resolution

Instrument Display Digits Decimal, Maximum

Low Measured Increment

Low Measured Pressurex 100 = %

IV. Overall Accuracy for Application

1. SUM of IIA, IIB, IIC1 or IIC2

2. Resolution III

3. Overall Accuracy (Greater IV1 or IV2)

%

%

%

� NOT Included in Accuracy%=

47

��

��

Chapter 7

The selection of a pressure calibration standardis a critical decision in the success of manyindustries. If the standard is not sufficiently ac-curate, significant losses both of revenue and/orprocess control can occur. On the other hand, ifthe standard is excessively accurate for the taskbeing performed, cost of the standard may be toohigh and the labor time to calibrate may be high.It is important, therefore, for the potential user ofa pressure standard to investigate both the needsof the task requiring the pressure standard andthe overall costs associated with each pressurestandard that is being considered.

7.1 Application ConsiderationsMany different types of pressure calibration stan-dards are available. In order for the user to selectthe correct tester, several aspects of the task tobe performed should be considered.

7.1.1 Test FluidAn important element is the test fluid. Since thetest fluid will enter the pressure sensing elementof the instrument being tested, the test fluid mustbe compatible with the process fluid to which theinstrument will be attached. Otherwise, all instru-ments must be cleaned after testing, an expen-sive operation. The most common test fluid isinstrument grade mineral oil. Where useable, oilprovides an outstanding combination of corro-sion resistance with lubrication of the close fittingparts of the pressure standard. Distilled wateralso provides an excellent test fluid which is inertto most process fluids. Clean, dry air or nitrogengas however, eliminates the problem altogether.

7.1.2 Pressure RangeA survey should be made of the pressure rangeof all instruments to be tested. The pressurestandard should be capable of producing pres-sures in excess of the highest instrument to betested. Care must be taken to consider the accu-racy of the pressure standard at the lowestpressure to be tested. Several pressure stan-dards of different ranges may be required tomaintain acceptable levels of accuracy.

7.1.3 Task To Be PerformedConsideration should be given to the task to beperformed. If most of the instruments to be testedare fixed in place, such as recorders, transmit-ters, etc. the portability of the instrument is impor-tant. If air transportation is required, such as oil/gas platforms, the size and weight is important.If many instruments are to be tested, dual columndead weight testers which change test rangequickly should be considered. If many technicianswill use the tester, the tester should be ruggedand relatively independent of operator technique.High performance tasks, such as testing of in-struments at manufacture, require custom de-signed testers.

7.2 Cost of MeasurementThe most important consideration in the selectionof a pressure standard should be the potentialloss of revenue resulting from the use of theinstrument. If the pressure standard is used tomeasure the quantity of product either beingpurchased or sold then the accuracy of pressuremeasurement can be directly related to the profitor loss to the user. In other instances, the pres-sure measurement is related to the efficiency ofa piece of equipment or process and can again bedirectly related to profit or loss. Another use ofpressure standards may be the test or calibrationof safety related equipment. In this instance theconsequences of failure should be consideredwhen selecting the pressure standard. On theother hand, some pressure measurements aremade for routine maintenance activities with mini-mal consequences of failure. Each of theseconsiderations will be discussed in more detail inthe following paragraphs.

A Worksheet for determining the economic analy-sis is shown on Figure 7-1. This worksheet canbe used to evaluate various instruments for anydesired application.

7.2.1 Custody TransferThe custody transfer involves the use of a pres-sure standard to determine the quantity of a

48

commodity to be purchased or sold by a cus-tomer. A typical example is a natural gas pipelinewherein all natural gas entering the pipeline isbeing purchased and all gas leaving the pipelineis being sold. The quantity of gas in each trans-action is determined by a measurement of flowacross a meter, usually an orifice plate.

For purposes of this analysis the flow rate at asingle metering station is assumed as 1,000,000cubic feet per day at a 25" H2O test point, the priceof natural gas is estimated at $2.00 per 1000 cubicfeet.

The equation for flow through an orifice meter is:

Flow = C x SQRT(∆P x Ps)

Where:

C = Flow constantP = Differential pressure over the orifice

platePs = Static pressure at the meter

At the 4" H2O calibration point, the flow rate is:

Flow(4"H2O) =

1,000,000 Ft3/Day x SQRT (4" H2O/25" H2O)= 400,000 Ft3/Day

As computed in Chapter 6, the overall accuracyof the deadweight tester at 4" H2O was 0.069%and at 25" H2O was 0.069%. The overall accuracyof digital (A) at 4" H2O was 43.1% and at 25" H2Owas 6.986%. For digital (B) the accuracy at 4" H2Owas 1.1345% and at 25" H2O was 0.24032%.

The estimated measurement accuracy lossesper test station for these three examples aretherefore:

A. Deadweight Tester

Error (4" H2O) = 4" H2O x 0.069% = 0.00276" H2O

Error (25"H2O) = 25"H2O x 0.069% = 0.01725" H2O

Loss(4" H2O)= 400,000 - 400,000 SQRT(4 - 0.00276)/

SQRT(4) = 138 Ft3/Day x 365 days x $2.00/MCF =$100.74 per Year

Loss (25"H2O) =

1,000,000 - 1,000,000 SQRT (25 - 0.01725)/SQRT (25)= 345 Ft3/Day x 365 Days x $2.00/MCF =$251.85 per Year

B. Digital Electronic Tester (A)

Error (4" H2O) = 4" H2O x 43.1% = 1.724" H2OError (25"H2O) = 25" H2O x 6.986% = 1.7465" H2O

Loss (4" H2O) =

400,000-400,000 SQRT (4 - 1.724)/SQRT (4) = 98,272 Ft3/Day x 365 Days x $2.00/MCF =$71,738.56 per Year

Loss(25" H2O)=

1,000,000 x 1,000,000SQRT(25 -1.7465)/SQRT(25)= 35,562 Ft3/Day x 365 Days x $2.00/MCF =$25,960.99 per Year

C. Digital Electronic Tester (B)

Error (4"H2O) = 4"H

2O x 1.1345% = 0.04538" H

2O

Error(25"H2O) = 25"H

2O x 0.24032% = 0.06008" H

2O

Loss(4" H2O) = 400000-400000 SQRT(4-0.04538)/SQRT(4)

= 2275 Ft3/Day x 365 Days x $2.00/MCF =$1660.75 per Year

Loss (25" H2O) = 1000000-1000000 SQRT(25-0.06008)/

SQRT(25) = 1202 Ft3/Day x 365 Days x $2.00/MCF =$877.46 per Year

7.2.2 Process ControlProcess control involves the use of a pressurestandard to measure pressures that are critical tothe efficiency and/or control of processes. Anexample of this type of application is the measure-ment of feed water pressure in steam poweredelectric generating plants. The operating effi-ciency of the plant is affected by the accuracy ofmeasurement of this pressure. Other applica-tions involve the uniform control of pressureswithin a process to assure optimum operatingefficiency.

The worksheet for determining the economicanalysis of various pressure standards is usedin the identical manner as that illustrated in Section7.2.1 except that the algorithm for process effi-ciency is substituted for the orifice plate flowalgorithm.

49

7.2.3 SafetyPressure standards are frequently used to testand/or calibrate safety related control systems.These systems may be designed to preventoverpressure within tanks or lines, overfilling oftanks, etc. with the potential of severe environ-mental damage, employee injury, etc.. If the pres-sure standard is not sufficiently accurate, thesafety system may be certified as accurate andmalfunction in the event of an emergency.

Using the three pressure standard examplesfrom Section 7.2.1, consider a safety applicationwherein the actuation pressure is 4 inches ofwater column with a set point tolerance of onepercent. The maximum pressure tolerance ac-ceptable would be 1% x 4 inch Water Column or0.04 Inches of Water. If a 4:1 margin of safety isrequired, only instruments with .04/4 or 0.01inches of water accuracy would be acceptable.The deadweight tester (0.069%) would then bethe only acceptable instrument for the test.

7.2.4 MaintenanceMany pressure measurements are made only toverify that the operating pressure recording de-vices are operating. Accuracy of measurementis generally not a major concern for this type ofapplication. Of concern may be the size, weight,portability and cost of the pressure standard.

As an example, consider that a customer has anapplication that utilizes process gauges withaccuracy of +3.0%, and a minimum range of 25Inches of water. The actual error of this equip-ment would be 3.0% x 25 inch Water Column =0.75 inch Water Column. Of the three examplesevaluated in Section 7.2.1 both the deadweighttester (0.069%) and the digital gauge (B)(0.24032%) would be acceptable solutions. Thefinal consideration would then be the size, weight,portability, ease of use and cost of each instru-ment.

7.3 Summary and ConclusionsMany different factors must be considered whenselecting a pressure standard. The customermust consider the requirements of the task to beperformed, the overall accuracy of the pressurestandard for the application, any specific applica-tion considerations including the test fluid, pres-sure range and the task to be performed, the costof inaccurate pressure measurement either onthe purchase or sale of product, the adequacy ofthe pressure standard to test or adjust safetyrelated systems or specific requirements forroutine maintenance inspections.

50

II. Process Control

Test Instrument: (% Overall Accuracy x Test Pressure = Actual Error)

x = Actual Error

Dollar Loss: Using Your Process Efficiency Algorithm, compute:

1. Process Error resulting from Actual Pressure Error

2. Cost of Process Error

Figure 7-1Worksheet Economic Analysis Pressure Measurement Instrument

Instrument Model Range

Overall Accuracy:

Test Point:

Cost:

Application: � Custody Transfer

� Process Control

� Safety

� Maintenance

I. Custody Transfer

Product Test: � Rate of Flow � Pressure � Other

A. Rate of Flow (% Overall Accuracy x Test Pressure = Actual Error)

x = Actual Error

Dollar Loss: Flow Equation (Orifice Meter): F = C x SQRT (∆P x Ps)

Flow = C x SQRT TestPoint

= ActualError

ActualError

= IndicatedError

Flow = C x SQRT

ActualError

- IndicatedError

x 100ActualError

=

% Error

Flow/Day = x Cost x % Error = Loss/Day

Error =

ProcessError =

x Cost = Loss/Test

Test/Day = x Loss/Test = Loss/Day

51

IV. Maintenance

Test Instrument:

III. Safety

% OverallAccuracy

x Test Pressure = Actual Error

Safety Parameters:

Maximum Pressure

Minimum Pressure

Process

Setpoint

Potential Loss:

� Employee Injury Equipment

Safety Analysis:

� Instrument is adequate

� Instrument is not adequate

Test Instrument:

% OverallAccuracy

x Test Pressure = Actual Error

Special Requirements:

� Portable

� Size

� Weight

� Other _____________________________________

Maintenance:

Task: _______________________________ Setpoint ___________ Tolerance ____________

Maintenance Evaluation:

� Instrument is adequate

� Instrument is not adequate

52

References

� ��!

9 Dadson, R.S.,Lewis, S.L., Peggs, G.N.,"The Pressure Balance, Theory andPractice", Her Majesties Stationary Office,London, 1982.

10 Delajoud, P.,Girard, M.,"Increasing theAccuracy of pressure MeasurementsThrough Improved Piston Gauge AreaDetermination", Instrument Society ofAmerica.

11 Legras, J.C., Schatz, B., Delajoud, P., "LaReference National de Pression Du BNMdans la Domaine de 10 a 400 kPa",Reference National de Pression BNM.

12 Bridge, M.L., "Internal Calibration Chain",Proceedings for Measurement ScienceConference, January 1993.

13 Reed, C.J., "Effective Use of DeadweightTesters”, Proceedings for The InternationalSchool of Hydrocarbon Measurement, May1993

14 Drake, Charles P., Natural Gas Pipeline ofAmerica, Proceedings of AGA Gulf CoastMeasurement Short Course, Houston,Texas, 1974.

15 "Manometer Principles", Meriam Instru-ment, a Scott Fetzer Company

16 Harland, P.H., "Pressure Gauge Handbook",by AMETEK, Marcel Dekker Incorporated

1 P.L.M. Heideman, B.E. Welch, Part 3, PistonGages, Pure and Applied Chemistry,Experimental Thermodynamics, Volume IIExperimental Thermodynamics of Non-reacting Fluids

2 Handbook US National Bureau of Stan-dards No. 77, Vol.III, "Precision Measure-ment and Calibration".

3 Bowman, H.A. and R.M. Schoonover. J.Res. Nat. Bur. Std. 71C,179 (1967).

4 Belanger,Brian C., "Traceability, AnEvolving Concept", ASTM StandardizationNews, Vol.8, Number 1.

5 Taylor, Barry N & Kuyatt, Chris E., NISTTechnical Note 1297, "Guidelines forEvaluating and Expressing the Uncertaintyof Uncertainty of NIST MeasurementResults".

6 Fawcett, L.W. Jr., AMETEK Engineering.Report 312,"Measurement Uncertainty inthe Calibration Process for DeadweightTesters".

7 Factors Affecting Performance of PistonGauges”, National Bureau of Standards,6.2 February 1957.

8 Reed, C.J., AMETEK Engineering Report247, "Calibration of Pressure Standards".

53

GENERAL TERMS

AMETEK CERTIFICATION - The document statingthe device, either a tester, piston & cylinder, orweights, the guaranteed accuracy of thedevice, the device serial number, certification(not the actual test date), and the recommendedrecertification date (normally 1 year), the NationalBureau of Standards trace-ability report num-bers, and the test conditions. The purchase ordernumber is included so that proof of certificationcan be directly connected to a specific device anda specific Purchase Order.

CERTIFICATION WITH DATA - This is a secondpage to the certification. The document reports allof the testers measured particulars, (e.g. effec-tive area, "B" pressure coefficient, etc.) Themasses of each weight in the weight set is alsoreported as well as nominal and actual pres-sures. The customer can use this infor-mation toconduct more accurate tests than simply justworking to the accuracy of the tester. With thisinformation the customer can calculate his ownactual pressures per the equations on the backof his certification.

CROSSFLOAT - A testing method or procedurewhich compares a known pressure for the pur-pose of determining the effective area of theunknown pressure source.

DEADWEIGHT GAUGE - A device used to mea-sure pressure, using a known area and mass(deadweights).

DEADWEIGHT TESTER - A device used to gener-ate or produce a known pressure, using a knownarea and masses (deadweights).

GAUGE - This term is normally used with refer-ence to pressure requirements.

GAGE - this term is normally used with referenceto dimensional requirements, such as a "Go-NoGo" Plug Gage.

INSTRUMENT AIR - Air defined by the ISA - S7.3standard.

NIST REPORT NO. - This number appears on thecertificaton for each tester. Numbers precededby "P" e.g. P-7800 represent press-ure test re-port numbers assigned by and on file with theNIST temperature and pressure measurementsand standards division center for absolutephysical quantities. Numbers such as 212.31/193712 are NBS numbers used in calibrating theweights used to calibrate AMETEK scales. All ofthese numbers are part of the traceability re-ported to the customer.

TEMPERATURE COEFFICIENT - As the piston andcylinder or ball and nozzle heat up or cools, theirlinear dimension changes, thus affecting theeffective area. The temperature coefficientsare well documented for the materials AMETEKuses and are reported in case the customerwishes to use his own computations.

TRANSDUCER - A device for converting me-chanical stimulation into an electrical signal. It isused to measure quantities like pressure, tem-perature and force.

��"��

54

ACCURACY

ACCURACY - The closeness or agreement be-tween a measured value and a standard or truevalue; uncertainty as used herein, is the maxi-mum inaccuracy or error that may reasonably beexpected.

PRECISION ERROR - The random error observedin a set of repeated measurements. This error isthe result of a large number of small effects, eachof which is negligible alone.

TRUE VALUE - The reference value defined bythe National Institute of Standards & Technologywhich is assumed to be the true value of anymeasured quantity.

UNCERTAINTY (U) - The maximum error reason-ably expected for the defined measurement pro-cess.

CALIBRATION

CALIBRATION - The process of comparing andcorrecting the response of an instrument to agreewith a standard instrument over the measure-ment range.

CALIBRATION HIERARCHY - The chain of calibra-tions which link or trace a measuring Instrumentto the National Bureau of Standards.

LABORATORY STANDARD - An instrument whichis calibrated periodically at the NIST. The labora-tory standard may also be called an interlabstandard.

NIST - National Institute of Standards & Technol-ogy. The reference or source of the true value forall measurements in the United States of America.

TRACEABILITY - The ability to trace the calibrationof a measuring device through chain ofcalibrations to the NIST.

TRANSFER STANDARD - A laboratory instrumentwhich is used to calibrate working standards andwhich is periodically calibrated against the labo-ratory standard.

WORKING STANDARD - An instrument which iscalibrated in a laboratory against an interlabor transfer standard and is used as a standard incalibrating measuring instruments.

55

MASS TERMS

AIR BUOYANCY - When considering the load onthe piston and cylinder in a deadweight tester, theload has to be reduced by an amount equal to themass of the air dis-placed by the weights.

APPARENT MASS - Mass of a weight as refer-enced to the density of a comparison mass. Whenweights are weighed on scales, the internalcalibration standard of the scale is normally brassor stainless steel. AMETEK uses brass thus allmasses are reported as apparent mass againsta brass standard.

CALIBRATION GRAVITY - The gravity of thelocation for which the tester was calibrated.Note, this is not necessarily the same as thegravity of the location in which the tester wascalibrated.

GRAVITY - Since dead weight testers rely onknown masses to generate known pressures itis necessary to correct the mass of the weight fordifferent gravities. The weights for all dead weighttesters are refer-enced to or calibrated at aspecific gravity. When correcting for a change ingravity, it is only necessary to take a ratio of thelocal gravity to the calibrated gravity and multiplyit by the pressure. See the handbook for a furtherexplanation.

LOCAL GRAVITY - The gravity of the area oflocation in which the tester is being used.

TRUE MASS - Mass of a weight as referenced toits own density.

HYDRAULIC TESTER TERMS

ACTUAL AREA (PISTON) - The area as deter-mined by measuring the piston with a dimensionalgage.

CONTROLLED-CLEARANCE PISTON & CYLIN-DER - This design permits an external pressure tobe exerted around the cylinder thus controllingthe clearance between the piston and cylinder.While this device is sophisticated and cumber-some to use, it is well documented and is theprimary standard used by the NIST.

EFFECTIVE AREA (PISTON & CYLINDER OR BALL& NOZZLE) - The area as determined by a calibra-tion process, such as crossfloating.

NOMINAL AREA (PISTON) - An ideal specific areawhich is not measured, but is used as a refer-ence. In the case of AMETEK pistons, nominalareas are as follows: 0.01, 0.05, 0.02, 0.1 squareinch.

"B" PRESSURE COEFFICIENT - This term repre-sents the piston's non-linearity. As the pressureincreases in a hydraulic tester the piston andcylinder experience large stresses which changethe effective area. Depending on the piston andcylinder shape the value of "B" can be (+) or (-).

SURFACE TENSION - As fluid leaks up betweenthe piston and cylinder it puts an upward force onthe piston, this is a result of the fluid surfacetension acting on the circumference of the piston.

SIMPLE PISTON & CYLINDER - A piston andcylinder design which permits pressure to beapplied only to the piston. This is represented byhydraulic cylinders used in machinery.

RE-ENTRANT CYLINDER - This design allows thepressure being exerted on the piston to also acton the cylinder. As pressure increases the cylin-der is being forced to reduce the fluid leakagebetween the piston and cylinder. The AMETEKType T & R series testers and gauges are of thisdesign.

56

PRESSURE TERMS

ACTUAL PRESSURE - The pressure measured bya gauge or generated by a tester.

ABSOLUTE PRESSURE (PSIA) - Pressure mea-sured with reference to zero gauge pressure, ora perfect vacuum. Atmospheric pressure is ab-solute pressure or about 14.7 Measurement PSIA.0 PSIA would be a perfect vacuum.

BACKGROUND, LINE, OR STATIC PRESSURE -The magnitude or pressure in a differential pres-sure gauge, or transmitter. This pressure couldbe at 3000 PSIG, but the differential pressure mayonly be 1 PSID. This means that one side of thedevice could be 3000 PSIG and the other side at3001 PSIG.

DIFFERENTIAL PRESSURE (PSID) - The differ-ence between two pressures as measured by adifferential pressure gauge.

GAUGE PRESSURE (PSIG or PSI) - Pressuremeasured with reference to atmospheric pres-sure. 14.7 PSIA is equal to 0 PSIG.

NOMINAL PRESSURE - An ideal specific pressureused as a reference. Generally this would be thepressure desired by a gauge user. The differ-ence between the nominal pressure and theactual pressure represent the accuracy of thatpressure.

VACUUM - A negative gaugepressure, with themaximum (or perfect vacuum) having noatmosphere pressure, i.e. about -14.7 PSIG or 0PSIA.

PNEUMATIC TESTER TERMS

ACTUAL WEIGHT FACTOR - The weight Factorcomputed by using a pneumatic tester's effectivearea, or by an actual calibration at a specific testpoint.

HI/LO WEIGHT FACTOR - The average weightfactor of the highest and lowest actual weightfactors determined by calibration at AMETEK. TheHI/LO weight factor is also called the averageweight factor or the reported weight factor. Thisvalue is used to calculate all actual pres-sures. Reported on the certification with data.

NOMINAL WEIGHT FACTOR - This term corres-ponds to the hydraulic nominal area terminology.The nominal weight factor for the pneumatictesters are as follows:

PK : 200.51 grams/PSIRK: 100.25 grams/PSIHK: 20.05 grams/PSI

NOTE: The RK is 1/2 the PK and HK is 1/10 the PK sothat all the areas are multiples of each other.

WEIGHT FACTOR - A term used by AMETEK todescribe the relationship between the ball areaand the mass required to generate 1 PSIG. Weightfactor is defined as grams/PSI.

www.ametekcalibration.com

United KingdomTel +44 (0)1243 833 [email protected]

FranceTel +33 (0)1 30 68 89 [email protected]

GermanyTel +49 (0)2159 9136 [email protected]

DenmarkTel +45 4816 [email protected]

USAFloridaTel +1 (800) 527 [email protected]

CaliforniaTel +1 (800) 444 [email protected]

IndiaTel +91 22 2836 4750 [email protected]

Singapore Tel +65 6484 [email protected]

ChinaShanghaiTel +86 21 5868 5111

BeijingTel +86 10 8526 2111

GuangzhouTel +86 20 8363 [email protected]

Information in this document is subject to change without notice. ©2013 by AMETEK, Inc., www.ametek.com. All rights reserved.

precision pressure measurement handbook

Documents