Asme 19.2 1987

COPYRIGHT American Society of Mechanical Engineers
Licensed by Information Handling Services

PART 2
Pressure INSTRUMENTS
AND
Measurement APPARATUS
ASME/ANSI PTC 19.2-1987

(REVISION OF PTC 19.2-19641
THE AMERICAN SOCIETY OF MECHANICAL

ENGINEERS
United Engineering
Center 345 East 47th Street NewYork, N.Y. 10017

Date of Issuance: August 15, 1988
This document willbe revised when the Society approves the issuance
of the nextedition,
scheduled for 1992. There will be no Addenda issued t o ASME/ANSI PTC 19.2-1 987.
of technical
Please Note: ASME issues written replies to inquiries concerning interpretation
aspects of this document.The interpretations are not partof the document.PTC 19.2-1 987
is being issued with an automatic subscription serviceto the interpretations that will
be is-
sued to it up to the publication of the 1992 Edition.
ASME is the registered trademark of The American Society of Mechanical

Engineers.
This code or standard was developed under procedures accredited as meeting the criteria for
American National Standards. The Consensus Committee that approved the code or standard was
balanced t o assure that individuals from competent and concerned interests have had an opportunity
to participate. The proposed code or standard was made available for public review and comment
which provides an opportunity for additional public input from industry, academia, regulatory agen-
cies, and the public-at-large.
ASME does not "approve," "rate," or "endorse" any item, construction, proprietary device, or
activity.
ASME does not take any position with respect to the validity of any patent rights asserted in
connection with any items mentionedin this document, and doesnot undertaket o insure anyoneuti-
lizing a standard against liability for infringement of any applicable Letters Patent, nor assume any
such liability. Users of a code or standard are expressly advised that determination of the validityof
any such patent rights, andthe risk of infringement of such rights, is entirely own theirresponsibility.
Participation by federal agency representative(sj or person(s1 affiliated with industry is not to
be interpreted as government or industry endorsementof this code or standard.
ASME accepts responsibility for only those interpretations issued in accordance with governing
ASMEproceduresandpolicies which precludetheissuanceofinterpretationsbyindividual
volunteers.
No part of this document may bereproduced in any form,

in an electronic retrieval system or otherwise,
without the prior written permission of thepublisher.
Copyright O 1988 by
THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS
All Rights Reserved
Printed in U.S.A.

FOREWORD
(This Foreword is not part of ASMUANSI PTC 19.2-1 987.)
This Instruments and Apparatus Supplement (PTC 19 Series) to the ASME Performance Test
Codes provides information on instrumentation and associated procedures for tests involv-
ing measurement of pressure. It is intended to promote results consistentwith the best engi-
neering knowledge and practice in industry.
The object and scopeof any testshould be agreed upon in writing by all parties to the test
prior tothe test.
ASMUANSI PTC 2 on Definitions and Values and ASME/ANSI PTC19.1 on Measurement
Uncertainty may be especially useful references when using this Supplement.
This Supplement replaces an older version published in 1964. This edition was approved
by the Boardon Performance Test Codes on September 23,1986 and adopted by the Ameri-
can National Standard Institute (ANSI) as an American National Standard on August 25,
1987.
Acknowledgement
The Committee wishes to acknowledge the contribution of past member P. Heydemann of
the National Bureau of Standards.
iii

All ASME codes are copyrighted, with all rights reserved to the Society. Reproduc-
tion of this or any other ASME code is a violation of Federal Law, Legalities aside, the
user should appreciate that the publishing of the high quality codes that havetypified
ASME documents requires a substantial commitment by the Society. Thousandsof vol-
unteers work diligently todevelop these codes. Theyparticipateon their own or with a
sponsors assistance and produce documents that meet the requirements of an ASME
concensus standard. The codes are very valuable pieces of literature to industry and
commerce, and the effort to improve these Iivingdocuments and develop additional
needed codes must be continued. The monies spent for research and further code de-
velopment, administrative staff support andpublication are essential andconstitute a
substantial drain on ASME. The purchaseprice of these documents helps offsetthese
costs. User reproduction undermines this system and represents an added financial
drain on ASME. When extra copies are needed,you are requestedto callor write the
ASME Order Department, 22 Law Drive, Box 2300, Fairfield, New jersey 07007-2300,
and ASME will expedite delivery of such copiesto you by return mail. Please instruct
your people to buy required test codes rather than copy them. Yourcooperation in this
matter is greatly appreciated.

PERSONNEL OF ASMEPERFORMANCE TEST CODE COMMITTEE NO. 19.2
ON PRESSUREMEASUREMENT
(The following i s the roster of the Committee at the time of approval of this Supplement.)
OFFICERS
J. A. Silvaggio, Chairman
J. Wyler, Vice Chairman
C. Osolsobe, Secretary
COMMITTEE PERSONNEL
C. W. Doran, Burr-Brown Corp., Foxboro Co.
T. A. S. Duff, Ametek - US Gauge, Inc.
C. W. Savery, Portland State University, Drexel University
J.A. Silvaggio, Transamerica Delaval, Inc.
J, A. Symonds, Private Consultant, Taylor InstrumentCo.
J. S. Wyler, Smiths Industries, General ElectricCo.

BOARD ON PERFORMANCE TEST CODES
J. S. Davis, Jr.,Vice President

W. O. Hays, Secretary
A. F. Armor W. L. Carvin S. P. Nuspl

R. L. Bannister G. J. Gerber R. P. Perkins
J. A. Booth K. G. Grothues R. W. Perry
B. Bornstein R. Jorgensen A. L. Plumley
W. A. Crandall D. R. Keyser J.A. Reynolds
H. G. Crim, Ir. J. E. Kirkland, Ir. C. B. Scharp
N. R. Deming W. G. McLean J. W. Siegmund
J. H. Fernandes J. W. Murdock R. E. Sommerlad
J. C. Westcott
vi

ASME P T C x 3 9 . 2 87 m O759670 0052093 T m
CONTENTS
Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
Committee Roster .........................................
Standards V
1 GENERAL CONSIDERATIONS ................................. 1

1.1 Definitions ........................................... 1
1.2 Units ................................................ 2
1.3 Dynamic Measurements ................................. 2
1.4
Use of Control
andOperating
Instrumentation ................ 3
1.5
Two-Phase Fluid Systems ................................ 3
1.6 Bibliography. ......................................... 4
2 PRESSURE STANDARDS ...................................... 5

2. 1 Inter-Laboratory
and
Transfer
Standards ...................... 5
2.2 Working Standards ..................................... 16
3 PRESSURE MEASUREMENT DEVICES ............................ 21

3.1
Ranges and
Accuracies .................................. 21
3.2 Piston
Gages. ......................................... 25
3.3 Manometers .......................................... 28
3.4
Pressure
Transmitters .................................... 32
3.5 LowAbsolute-Pressure
(Vacuum) Sensors .................... 55
3.6
Elastic
Gages .......................................... 65
3.7
Special Applications of AP Cells ........................... 73
4 PRESSURE MEASUREMENT INSTALLATIONS...................... 95

4.1 Pressure
Taps ......................................... 95
4.2 Pressure
Probes ........................................ 96
4.3 Connecting Piping., .................................... 102
4.4
DiaphragmSeals. ...................................... 107
4.5 Installation Effects ...................................... 114
4.6Uncertainties Measurement ......................
in Pressure 116
5 REFERENCES ............................................... 117
Figures
Basic
1.1
Pressure
Terms ......................................... 2
2.1
Pressure
Measurement
Calibration
Hierarchy ....................... 6
2.2Schematic Distortionof Pistonand Cylinder in a Simple
PistonGage .............................................. 7
2.3 Simple Cylinder Piston Gage. .................................. 8
2.4
Re-entrant Cylinder Piston
Gage ................................ 9

2.5 Gage Governed by the Controlled Clearance Principle . . . . . . . . . . . . . . . 9
2.6 General Principle of All Manometers. ............................ 12
2.7 Errors in Column Height Determination (Manometer Tilted by
Angle 0 About a Horizontal Axis Through the Manometer
Tubes) ................................................... 12
2.8 Errors in Column Height Determination (Manometer Tilted by
.............................
Angle 4 in the Plane of the Tubes) 14
2.9 Differential PressureTransducerConnected to the Manometer . . . . . . . . . 14
2.1 o McLeod VacuumManometer ................................... 17
2.1 1 Pivoting McLeod VacuumGage ................................. 18
3.1 . Simple Cylinder PistonGage................................... 27
3.2 Re-entrant Cylinder PistonGage ................................ 28
3.3 ........................
Controlled-Clearance Cylinder PistonGage 29
3.4 Pneumatic Deadweight Ball Gage ............................... 29
3.5 Absolute PressurePistonGage .................................. 30
3.6 Vacuum Piston
Gage ......................................... 30
3.7 PistonGageMeasurement With a Diaphragm Separator .............. 31
3.8 U-TubeManometer for Absolute Pressure ......................... 31
3.9 U-TubeManometer for Differential Pressure ....................... 31
3.10 CisternManometer .......................................... 32
3.1 1 Inclined Manometer.......................................... 32
3.12 Micromanometer (Null Reading) ................................ 33
3.1 3 Fortin Barometer ............................................ 34
3.14 Bell-Type Element ........................................... 36
3.1 5 Slack Diaphragm ............................................ 36
3.1 6A Schematic of a Pneumatic Force Balance Differential
PressureTransmitter ........................................ 38
3.1 6B Schematic of a PneumaticForceBalanceTape Drive Servo ............ 39
3.1 7 Schematic of an Electronic Force Balance Differential
Pressure Transmitter ........................................ 40
3.1 8 Typical Nozzle Baffle System ................................... 40
3.1 9 Nozzle Backpressure Versus Gap for an Elementary
Nozzle Baffle ............................................. 41
3.20 Schematic Representation of a Double-Sided Variable
Capacitance Sensor Head. ................................... 43
3.21 Schematic Representation of a Single-Sided Variable
Capacitance Sensor Head. ................................... 44
3.22 Inductive Displacement Detector. ............................... 45
3.23 PreferredSchematicRepresentation of theLVDT .................... 45
3.24 LVDT Used as a Pressure Transducer. ............................ 46
3.25 Potentiometric Detector ....................................... 48
3.26 StrainGage ................................................ 48
3.27 WheatstoneBridge Configuration of the StrainGage ................. 49
3.28 Full-Bridge Diaphragm Gage ................................... 50
3.29 Pressure Transducer With Vibrating Element ....................... 50
3.30 Piezoelectric PressureTransducer ............................... 52
3.31 Conventional Piezoelectric System............................... 52
3.32 ChargeSystem .............................................. 53
3.33 BasicICPTransducer ......................................... 54
3.34 Operating Ranges for PressureSwitches ........................... 54
3.35 Hickman Vacuum Gage ....................................... 56
3.36 Diaphragm Pressure Comparator ................................ 57
3.37 McLeodGage .............................................. 60
viii

.............. ............... aS W
3.38 ThermocoupleGage ......................................... 60

3.39 PiraniVacuumGage ......................................... 61
3.40 Bayard-Alpert Ionization Gage.................................. 63
3.41 Phillips-Penning Gage., ...................................... 63
3.42 Ionization Chamber of Alphatron Gage ........................... 65
3.43 Langmuir-Dushman Molecular Gage ............................. 66
3.44 BourdonCage .............................................. 68
3.45 BourdonTubes ............................................. 68
3.46 BellowsGage ............................................... 69
3.47 Slack Diaphragm Gage ....................................... 69
3.48 Flow Installation ............................................ 75
3.49 Schematic of Differential-Pressure Transmitter Primaryfor
Flow Measurement ......................................... 76
3.50 Schematic Diagram of Open-Tank Transmitter Primary ............... 78
3.51 Open-Tank Installation With Transmitter Mounted Directly
to Tank Nozzle ............................................ 79
3.52 Open-Tank Installations With Remote Seal Type of
Transmitter. .............................................. 80
3.53 Schematic Diagram of Closed-Tank Transmitter Primary.............. 81
3.54 ................................
Closed-Tank Installation, Dry Leg 82
3.55 Closed-Tank Installation, Dry Leg Transmitter Above
DatumLine ............................................... 83
3.56 Closed-Tank Installation, Dry Leg Transmitter Below
DatumLine ............................................... 84
3.57 Closed-Tank Installation, Wet Leg............................... 85
3.58 Closed-Tank Installation,Wet Leg Transmitter Above
DatumLine ............................................... 86
3.59 Closed-Tank Installation,Wet Leg Transmitter Below
DatumLine ............................................... 87
3.60 Closed-Tank Installation, Dry Leg Transmitter Above
Upper Process Tap ......................................... 88
3.61 A Repeater Type Level Measuring Device ....................... 90
3.62 Hydrostatic Head Provides One Method of Density
Measurement ............................................. 92
3.63 Differential Hydrostatic Head Increases Sensitivity of Density
Measurement ............................................. 92
3.64 Common Method of Measuring Density of a Process Liquid ........... 93
4.1 Pressure Tap Flow Field ....................................... 95
4.2 TapCeomet ry .............................................. 96
4.3 Errors for Different Size Taps in Fully Developed Pipe Flow . . . . . . . . . . . 97
4.4 .................
Relative Tap Errors as Percent of Dynamic Pressure. 98
4.5 ImpactTube ................................................ 99
4.6 Variation of Total Pressure Indication With Angle of Attack
and Geometry for Pitot Tubes (After NACA TN 2331,
Apri11951) ............................................... 100
4.7 KielProbe ................................................. 101
4.8 ................
Total Pressure Location on a Cylinder in a Flow Field 101
4.9 StaticTube ................................................. 103
4.1 O .............................................
Pitot-static Tube 103
4.1 1 .........................
Cylindrical Probe, Principle of Operation 104
4.1 2 Wedge-TypeProbe .......................................... 105
4.1 3 Spherical and Cone TypeProbes ................................ 105
4.14 Basket Probe ...................... ........................
; 106

ASME P T C * l 7 * 2 87 W 0757670 0052074 5 W
4.1 5 BasketProbe With a Pressure Transducer .......................... 106

4.1 6 Magnitude of Probe-BlockageEffects ............................. 108
4.1 7 Magnitude of Probe-BlockageEffects ............................. 1O9 .
4.1 8 Typical PressureGage Piping Arrangement ........................ 110
4.1 9 Differential PressureCross Connection ........................... 110
4.20 Tube Configuration Used in Frequency Calculation . . . . . . . . . . . . . . . . . . 110
4.2 1 Remote Seal With Measuring Elementfor a Gage Pressure
Transmitter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
4.22 RemoteSeal With a PressureGage .............................. 111
4.23 Remote Seals for Use With a Differential Pressure Transmitter .......... 112
4.24 Transducer Setup. ........................................... 114
4.25 Temperature Distributions in Connecting Tubing .................... 115
4.26 Flow MeasurementTransducer Application ........................ 115
Tables
1.1 Pressure Conversion Factors [3. 41 ............................... 3
2.1 Pressure Coefficient b [6] ...................................... 8
2.2 Density of ManometerFluids ................................... 10
2.3A Density. Thermal Expansion. and Vapor Pressure of Water
VersusTemperature in "C (IPTS-68) ............................ 13
2.38 Density. Thermal Expansion. and Vapor Pressure of Mercury
VersusTemperature in "C (IPTS-68) ............................ 13
2.4 Capillary Depression in Mercury in Glass Manometers
(SurfaceTension:
0.45
N/m) ................................. 15
3.1 Summary of Pressure-Measuring Device Characteristics. . . . . . . . . . . . . . . 22
3.2 SensingElements. ........................................... 26
3.3 VacuumMeasurement Units ................................... 58 .
3.4 Recording of Gage-TestDataSample ............................. 74
3.5 Seal Fluid SelectionChart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
3.6 Type of Calibration RequiredforVarious Applications ................ 91
3.7 Variations in Density for Different Liquids ......................... 94
Appendices
A and Gravity Data ...........................
Tables of Corrections 119
B Commonly Used Unit Conversion Factors ......................... 133
Tables
Al Corrections for Temperature to be Applied to Observed Height
of Mercury Columns (Assuming BrassScale is True at 32F) . . . . . . . . . . 121
A2 Corrections for Temperature to be Applied to Observed Height
of Mercury Columns (Assuming BrassScale is True at62OF) ........... 123
A3 Difference Between g,. the Gravitational Acceleration at Mean
Sea Level for Given Latitude O. and Standard Gravitational
.
Acceleration go = 32.1 740 ft/sec2 Table Gives (g, - go)as a
Function of Latitude ........................................ 125
A4Free-Air Gravity Correction Cf = 0.000003086 H ft/sec2.
Where Barometer Elevation H is in Feet Above Mean Sea level .
Correction Cf is to be Subtracted from (g, - go). . . . . . . . . . . . . . . . . . . 126
A5 Correction of Height of Column to Standard Gravity 6
(gl - go)/go .............................................. 127
A6 Gravity Data ............................................... 128
B1 Commonly Used Unit Conversion Factors ......................... 133

1
ASMUANSI PTC 19.2-1987
AN AMERICAN NATIONAL STANDARD
ASME PERFORMANCE TEST CODES

Supplement on
INSTRUMENTS A N D APPARATUS
PART 2
PRESSURE MEASUREMENT
SECTION 1 - GENERAL CONSIDERATIONS
The thermodynamic state of a simple fluid is speci- ambient pressure - force perunit area exerted by the
fied by two independent properties. in experiments in- atmosphere at a location (usually local barometric
volving fluids, pressure is customarily selected as one pressure) (see Fig. 1.1)
of the properties to be measured, Pressure measure- differential pressure - difference between any two
ments are alsoimportant in systems involving flowing pressures (see Fig. 1.1)
fluids as an indirect means of measuring velocity and gage pressure- force per unit area exerted by a fluid
flow rate. on acontainingwall with respectto local ambient pres-
Relevant staticand dynamic pressures span a range sure. Gage pressure canbe either positive or negative.
of 10 to l o 7 times atmospheric pressure. As such, Common practice is to refer to negative gage pressure
and because of associated dynamic pressure- as vacuum (see Fig. 1. I ) .
measurementproblems,pressure-measurement sys-
static pressure- pressureat a point where a fluid ele-
tems vary greatly in complexity and include a large
ment is in equilibrium
number of different devices.
This Section covers the definition ofpressure, funda- total pressure - pressure on a plane normal to local
mental thermodynamic and fluid-mechanic concepts flow direction. It is the maximum value of pressure as a
of pressure, pressure units and conversion amongdif- function of direction at a point. It is equal to the sum-
ferent units, pressureconsiderations in and pressure re- mation of static pressure and velocity pressure.
lations for flowing fluids, and the use of existing velocity pressure (fora flowing fluid) - expressed as
installed instrumentation in equipment tests. i eV2 where e is the fluid density and V is the fluid
velocity; also called dynamic pressure. Velocity pres-
sure (or head) is the net pressure increase that canbe
derived from complete conversion of the velocity (or
1.1 DEFINITIONS dynamic energy) to pressure.
pressure - force per unit area exerted by a fluid on a energy relationships - the steady-state microscopic
containing wall with respect to a reference one-dimensional conservation of energy equation
absohte pressure - force per unit area exerted by a (along a streamline in an irrotational steady flow) is
fluid on a containing wall withrespectto zero absolute
pressure. Absolute pressure can be positive only (see
Fig. 1.1).
1

ASME/ANSI PTC 19.2-1 987 PRESSURE M E A S U R E M E N T
I
Positive gage
Differential
pressure
pressure
Pressure B
Ambient pressure (zero gage pressure)

Absc,u 'te
pressure
I I
Negative
gage
(c
//////// Zero absolute pressure
FIG. 1.1 BASIC

PRESSURE
TERMS
Equation (1-1) applies to the flow of a frictionless fluid 1.2 UNITS

with no mechanical work performed on or by the sur-
The International System of Units (SI) will be used
roundings, constant velocities, anda constant gravita-
in this publication with U.S. customary units in paren-
tional acceleration g. In Eq. (1-1) thefluid has velocity
theses.
V, pressure P, density e, and elevation Z. The subscript A list of symbols used in this work, with the corre-
number denotes a position and corresponding fluid
sponding dimensions and conversion factors used to
state. convert from the absoluteengineeringto the SI system,
For the special case of an incompressible fluid, Eq.
are given in Ref. [2].
(1-1) reduces to the Bernoulli Equation
Pressure is expressed in units of pascal, Pa, which is
equivalent to newtons/meter2. Conversion factorsfor
1 1
-
2(h2- v,2)$. (P2 - P,) 4- g(Z2 - Z,) = o (1-2) commonly used pressure units are given in Table 1. I
[3, 41.
The International Standard Atmosphere i s 760 mm
In manometry, the fluid is static and the kinetic en- (29.921 in.)of mercury at0"C (32F). In SI units this is
ergy term vanishes. Then the sum of the second term,
101.325 kPa (1 4.69595lbf/ina2)at the standard gravita-
pressure head, andthe third term, elevation head, is a
tional acceleration of 9.806650 m/sec2(32.17406
constant. ft/sec2).
In a flowing system, the rise of the first term in Eq.
(1-21, velocity head, involves a corresponding de-
crease in either elevation head and/or static-pressure
1.3 DYNAMICMEASUREMENTS
head. Thus, after a flow-area contraction, thevelocity
head is necessarily increasedto accommodate the con- 1.3.1 Fluctuating Pressure. In many situations in test
stant massflow, and consequently the static pressure is work, flows are unsteady; thatis, velocity and pressure
reduced. After a flow expansion at subsonic flow ve- vary with time, either cyclically or as random fluctua-
locities, the static pressure is increased. In each case, tions. It i s usually necessary to determine the true aver-
the total pressure is the same before andafter the area agepressure in order to evaluate the time average
change, except for frictional losses, which would in- energy of the stream. The best way to do this is to re-
crease temperature andinternal energy at the expense duce the causes of pressure fluctuation to negligible
of mechanical energy. proportions.
2

ASME P T C * L . 7 * 2 8 7 W 0757b70 0 0 5 2 0 7 7 O
PRESSURE MEASUREMENT . ASME/ANSI

PTC 19.2-1987
TABLE 1.1 1.3.2 Fluctuating Flow. When pressure measurements

PRESSURE CONVERSION FACTORS [3,4] are taken for the purpose of evaluating flow rate as in
flowmeter work, it should be realized that average
Conversion pressure doesnot correspond to average flow rate (this
Multiplication
From To Factor
is due to the square-lawrelation between velocity and
velocity pressure). To obtain a true average flow rate, it
atmosphere (normal = 760 torr) pascal (Pa) 1.013 25 E+05* may be necessary to obtain a graphic record of velocity
atmosphere pascal (Pa) 9.806 650 E f 0 4 pressure with a high-frequency-response instrument,
(technical = 1 kgf/cm2)
derive from this a curve of the square root of velocity
bar pascal (Pa) +
1.000 O00 E 05*
centimeter of mercury(0C) pascal (Pa) 1.333 22 E+03 pressure and use the average of this square-rootcurve
centimeter ofwafer (4C) pascal (Pa) 9.80638 E +Olt to calculate velocity. It is possible to carry out this proc-
decibar pascal (Pa) +
1.000 O00 E O? ess automatically by electronic methods when an elec-
dyne/centimeter2 pascal (Pa) 1.000 O00 E-O1 tronic pressure-transducer is the primary element.
foot of water (392F) pascal (Pa) 2.98898 E+03*
However, the error encountered may be shown to be
gram-force/centimeter2 pascal (Pa) 9.806 650 E+01
inch ofmercury (32'0 3.386 389 E+03 negligibly small under some conditions and thus
pascal (Pa)
inch of mercury (60'F) pascal (Pa) 3.37685 E+03 ignored.
inch ofwater (39.2"F) pascal (Pa) 2.490 82 E+02
inch ofwater (6O00 pascal (Pa) 2.488 4 +O2
inch ofwater (68F) pascal (Pa) 2.486 E+02+
4
kilogram-force/centimeter* pascal (Pa) +
9.806 650 E 04*
1.4 USEOFCONTROLANDOPERATING
kilogram-force/mete? pascal (Pa) 9.806 650 E+00*
kilogram-force/millimeter2 pascal (Pa) 9.806 650 Ei-O6 INSTRUMENTATION
kip/inch2 (ksi) pascal (Pa) 6.894 757 E+06
millibar pascal (pa) I ,000 000 E + O Z * Equipment to be tested may be provided with pres-
millimeter of mercury (0C) pascal (Pa) 1.333 224 E+02 sure instrumentation, pressure connections, and gages
poundal/foot* pascal (Pa) 1.488 164 E O0+ for either control or operating information. It may be
pound-force/foot2 pascal (Pa) 4.788 026 E+Ol necessary or desirable for the test engineer to utilize
pound-force/inch2 (psi) pascal (Pa) 6.894 757 E+03
this instrumentation. However, the precision and ac-
psi pascal (Pa) 6.894 757 E -I-03
torr (mm Hg O O C ) pascal (Pa) 1.333 224 E+02 curacy of the installed instrumentation should be con-
sidered in designing a test.Separate calibration of
NOTE: installed pressure-measurement systems maybe desir-
*Relationships that are exact in terms of the base units. able. When doubts exist about the accuracy andpreci-
sion of installed instrumentation, provision of alternate
test instrumentation is recommended.
Where this cannot be done, two methods of obtain-
ing an average are possible. First, the pressure instru-
ment may be damped sufficiently to give a value which
is only fluctuating slightly and therefore easily read-
1.5 TWO-PHASEFLUIDSYSTEMS
able. This gives a true average only if the instrument re-
sponse is linearly proportional to the pressure signal Many applications require pressure measurementin
and if the damping forces are linearly proportional to two-phase fluid systems [5]. In these applications, care
pressure. This linearity may be closely approximated must be exercisedto avoid problems due to pumping
by a porous-plug type of damper in the pipe. of multiphase fluid into instrument lines, static-liquid
Second, a graphic instrument or digital storage-and- head-pressurecontributions, and vapor-pressureinter-
processing unit capable of responding to frequencies ference. Although such problems are more difficult
greater than themaximum frequency of pressure fluc- with differential-pressure measurement, they are miti-
tuation may be selected. The graphicor digital record gated by keeping the instrument lines full of either
of pressure as a function of time can then be analyzed phase (e.g., the liquid phase) or by employingpressure
to yield a true average. In many cases, this can be ac- transducersdirectly mounted to the measurement port.
complished by adisplay instrument, suchas acathode- Techniques of liquidpurging, or, in the case of single-
ray oscilloscope. The average can often be estimated component systems, cooling the instrument lines to
visually with sufficient accuracy or the display may be cause condensation of entering vapor,areused to
photographed and analyzedas a graphic record. maintain a liquid phase in the lines. Purging is also
Alternatively, digital processing may be directly em- used to maintain air or other suitable gas in the lines for
ployed to yield a true average, some two-phase measurements.
3

ASME/ANSI PTC 19.2-1987 PRESSURE MEASUREMENT
1.6 BIBLIOGRAPHY
(1) Beckwith, T.G.,andBuck,N. L., "Mechanical
Measurements," Third Edition, Chapter 14, Reading:
Addison-Wesley, 1981.
(2) Bird, R. B., Stewart, W. E., and Lightfoot, E. N.,
"TransportPhenomena,'' New York: John Wiley &
Sons, 1960, pp. 4-5.
(3) Holman, J. P., andGajda, W. J., "Experimental
Methods for Engineers," Third Edition, New York:
McCraw-Hill, 1978, pp. 51-55.
(4) Schlichting,H., "Boundary-Layer Theory," Seventh
Edition, New York: McGraw Hill, 1978, pp. 49-52.

MEASUREMENT PRESSURE PTC 19.2-1987
SECTION 2 - PRESSURE STANDARDS
All pressure measurements areultimately referred to pA on the piston, where A i s the effective area of the
devices which serve as primary standards of pressure piston. By balancing this force F with the weight of a
measurements.These, in turn,can be calibrated in number of masses loaded on tothe piston,the pressure
terms of the basic units of mass, length, andtime. In or- p can be determined provided that the effective areaA
der ofdecreasing pressure, the mostimportant primary and itschange with pressure and temperature are
pressure standards are: the piston gage, the manome- known. If all parameters entering into the determina-
ter, and the McLeod gage. From these devices, pres- tion of a pressure with a piston gage are considered,
sure values are transferred to the point of use, often one arrives at the following equation [6]
through long calibration chains, using various typesof n
gages and transducers. Most of the uncertainty of pres-
sure measurements atthe point ofuse comes from er-
c q 1 -G) +++N
i = l e Mi
rors accumulated in the transfer of the measurement P = (2-11
along a calibration chain and not from the primary
A, (1 + bp) 11 - (ac + ap)U - T,)]
standard. For the purposeof estimating uncertainty, or
whenparticularly accurate measurementsare in- where
tended, attention must be paid to the propagation of p = pressure at the reference level of the piston
measurement and associated errorsalong the calibra-
tion chain. Mi = mass of weight i
To demonstrate traceability of measurements to the g = local acceleration due to gravity
National Bureau of Standards (NBS), it is necessary to cair = density of the ambient air
establish calibration hierarchies. Each level in the hier- erMi= density of weight i
archy, including that corresponding to NBS, consti- y = surface tension of the pressure-transmitting
tutes an error source which contributes to the error in fluid
the final measurement. Figure2.1 is a typical pressure C = circumferenceofthepistonwhere it
transducer hierarchy. emerges from the fluid
Calibration of measurement instruments at NBS is N = tare weight
possible; however, such calibrations can be time con- A, = effective area of the piston gage, determined
suming, inconvenient, and expensive.Most industrial at atmospheric pressure and reference tem-
working standards are referred to inter-laboratory or perature
transfer standards. b = pressure coefficient of the effective area
This Section contains brief descriptions of the operat- ac, ap = linear thermal expansivities of the cylinder
ing principles of devices which serve as inter- and of the piston
laboratory and transfer standards. 7' = gage temperature
Tf = reference temperature for which the effec-
tive area A, is accurately known
In order to determine the effective area of a piston
2.1 INTER-LABORATORY A N D TRANSFER
STANDARDS gage, one may either directly measure the area of the
cylinder and that of the piston near the line of mini-
2,1.1 Piston Gages. A piston gage, also called a dead- mum clearance, provided one has anadequate dimen-
weight tester or pressure balance, consists essentially sional metrology laboratory, or simply compare the
of a cylinder and awell-fitted piston. (See paras, 2.2.1 piston gage with a primary pressurestandard. Like-
and 3.2 for further discussion of piston gages.) The wise, the elasticdistortion coefficient b may either be
pressure to be measured is connected into the cylinder. calculatedfrom elastictheory or from comparison with
The pressurizedfluid in the cylinder exerts a force F = a primary standard. Figure2.2 shows schematicallythe
5

ASMUANSI PTC 19.2-1987 PRESSURE MEASUREMENT
National Bureau of
Standards (NBS)
I I
Inter-Laboratory
Standards (ILS)
Transfer
Standards
Working
Standards
Measurement Pressure Pressure

Etc.
Instrument transducer gage
FIG. 2.1 PRESSURE MEASUREMENT CALIBRATION HIERARCHY

PRESSURE MEASUREMENT ASMVANSl PTC 19.2-1 987
I
x = r
I
P
x = o
FIG. 2.2 SCHEMATIC DISTORTION OF PISTON AND CYLINDER IN A-SIMPLE

PISTON GAGE
distortion of pisfon and cylinder in a simple piston where

gage.From elasticity theory for infinitely extended hol- R, and r, = outer and innerradii of the cylinders
low and solid cylinders, the following equation for the pcand ~p = ~~i~~~~~~ ofthe cylinderand piston
ratios
elastic distortion coefficient b may be derived [7] materials
,and E,, = Young's moduli ofthecylinderand piston
materials
p = pressure inside the cylinder
pe = pressure on the ends of the cylinder
po = pressure on the outside of the cylinder
piston
cylinder
distortion
due to (2-2) 'For anassemblymade of steel with E = 2 x IO'' N/
pressure
distortion
internal m2, p = 0.28, and R&, = 3, we have for the pressure
Coefficient
+ Pe PC
"
b = -4 X + 38 X - 112 X
P Ec P
cylinder distortion due cylinder distortion

due to end loading
+ 14 x = - 64 x [rn2/N]
to external pressure P P
7

ASME P T C * > 7 * 2 8 7 m. 0757670 0052302 O m
ASMVANSI PTC 19.2-1987 PRESSURE MEASUREMENT
TABLE 2.1 To auxiliary piston

and load carrier
PRESSURE COEFFICIENT b [6]
(Courtesy of Pergamon Press, Ltd.)
I
Pressure
Coefficient, Pistodcylinder Range, Cylinder
I I
l/kPa
+ 1.1 x10.
Material
Steel/brass
kPa
17000
TY Pe
Simple
I
-1.3 x IO-
$5.2 x IO-
-6.4 x 10.
Steel/brass
Steekteel
Steel/steel
70000
34000
83000
Re-entrant
Simple
Re-entrant
I I \f
-8.1 x
-6.1 x lo-
-2.9 x lo-
Carbidekteel
Carbidekteel
Carbidekarbide
28000
17000
83000
Reentrant
Reentrant
Re-entrant
I I
-5.1 x 10 Tungstedcarbide ... Controlled

clearance I
Two facts are obvious: (1) the distortion of the pis-
I
ton is usually very small, and (2)the total distortion of
the cylinder can be reduced to zero by an appropriate
I
adjustment ofp,, the pressure applied to the outsideof
the cylinder. I
Piston gagedesigns commonly incorporate three
types of cylinders: simple cylinder, reentrant cylin- I
l I
der, and controlled-clearance cylinder.
(a) Simple Cylinder Piston Gage. A schematic dia-
gram of a simple cylinder piston gage is shown in Fig.
2.3. Force is applied to the head of the piston either di-
rectly orthrough an auxiliary piston and a flexible
point to prevent imparting of a bending moment to the
gage piston. The operating area of the assembly is re- FIG. 2.3 SIMPLE CYLINDER PISTONCAGE
moved by several diametersfrom the lower part of the
cylinder, which is distorted by making a pressure seal.
The cylinder is stressed only by the pressureinside the the cylinder but which holds sufficiently well for the
crevice and the pressurecoefficient of the assembly is piston. If the external pressure po were used to make
positive. Table 2.1 lists examples of experimentallyde- the cylinder conform to the piston, then we would
termined pressure coefficients for piston gages of this need to determine the area and calculateonly the dis-
and other types [6]. tortion of thepiston.This is called the controlled
(b) Re-entrant Cylinder Piston Cage. Simple cylin- clearance principle and the design of such a gage is
der piston gages are limited in their range of operating illustrated in Fig. 2.5. In practice one cannot force the
pressures by the excessive leakage of pressure fluid cylinder to completely conform to the piston. The ex-
past thepiston at high pressures. Inspection of Eq. (2-2) ternal jacket pressure pi has to be reduced slightly
shows that if pressure is applied to the outside and to below the pressure pz at which the clearance between
the end of thecylinder, the inner diameter of the cylin- piston and cylinder is zero, in order to allowthe piston
der mayactually decrease fasterwith pressure than the to move. The determining equation for the pressure
diameter of the piston,thus reducing the clearance at generated or measured by a controlled clearance pis-
high pressures to zero. ton gage then is given by 161
Figure 2.4 showsa cross sectionthrough a re-entrant
cylinder where mostof the outside ofthe cylinder is ex-
posed to pressure [6].
(c) Controlled-Clearance Cylinder Piston Gage.
Equation (2-2) is derived for infinitely extended pis-
tons, a condition that cannotbe realized in practice for
8
.-
.i
ASNE P T C * L S - 2 87 m 075767 0 0 5 2 3 0 3 2 H
PRESSURE MEASUREMENT ASMUANSI PTC 19.2-1987
Pressure
t
P
fluid
FIG. 2.5 GAGEGOVERNED BY THE

FIG. 2.4 RE-ENTRANT CYLINDER PISTONCACE CONTROLLED CLEARANCE PRINCIPLE
In the denominator 2.1.2 Manometers. Manometers measure an un-

A, = area of the piston known pressure by balancing it against theweight of a
b, = pressure coefficient of the area of piston column of liquids. Since the weight of the liquid
d = jacket pressure coefficient column can often be measured in terms of the basic
pz = jacket pressure that reduces the clearance to units of mass and length, certain typesof manometers
zero can be consideredto be primary standards of pressure
p I. = operating jacket pressure measurement [8]. See paras. 2.2.2 and 3.3 for further
The terms in the numerator andthe temperature term discussion of manometers.
in the denominator are the same as in Eq. (2-1). Figure 2.6 illustrates the principle used in all ma-
All of the termsof Eq. (2-3) canbe determined by di- nometers: two vertical tubes areconnected at the bot-
rect measurement, except for d and pz, which require tom. The tubesare filled with a liquid of known
more involved proceduresfor their determination. The density. Pressures pl and p 2 are applied at the liquid
calibration of controlled-clearance piston gages is de- menisci. The horizontal connection at the bottom of
scribed in detail in Ref. [6]. the tubes does not contribute to the pressure generated
The National Bureau of Standards maintains a group by the manometerregardless of the density of the fluid
of controlled-clearancepiston gages covering the pres- contained in it. We arbitrarily measurethe column
surerange from about 35kPa (5 psi) to 1.4 CPa heights Hl and H2 from this level. p , is the pressurein
(200,000 psi). These gagesdefine the pressurescalesin the connection. At the bottom of column 1:
the United States, and they are usedto provide calibra-
tion services for American industry and government
agencies.

ASMElANSl PTC 19.2-1987 PRESSURE MEASUREMENT
where TABLE 2.2

g = local acceleration due to gravity DENSITY OF MANOMETER FLUIDS
el = average density of the liquid in this column
H, = height of the column Density,
Temperature,
Fluid OC g/cm3
pl = known reference pressure (usually atmo-
spheric pressure) Mercury O 13.5950
20 13.5458
At the bottom of column4 2 Water 0.99997

15.5 0.9989
20 0.99821
p0 = gezHz + PZ (2-5)
Tetrabromoethane 20 2.96
where (CH Br, CH Br,)
e2 = average density of the liquid in this column Bromine 20 3.1226
H, = height of the column
p2 = unknown pressure
The unknown pressure p2 is derived from Eqs. (2-4)
and (2-5) to be and certain stainlesssteelsare also sensitive to
mercury.
~2 = g(e14 - e A ) + PI (2-6) Mercury metal and compounds of mercury may
cause dangerous environmental problems. The very
Thevariousterms in this equation require further low vaporpressure of mercury presents a serious
discussion. health hazardif spillage occurs.Extreme carei s neces-
sary, and strict adherence mustbe given to all applica-
2.1.2.1localAccelerationDuetoGravity g
ble regulations concerning mercury.
[m/sec2]. For themostaccuratemeasurements,the
National Oceanographic and Atmospheric Adminis- 2.1.2.3 Height H [m]. To measure column height,
tration - Gravity, Astronomy and Satellites Branch, the meniscus should first be located and this position
will provide gravity data extrapolated to the latitude, should thenbe transferred to a suitable scale. Sighting
longitude, and elevation of the manometer site. The rings, pointers, electro-optical devices, floats, and ca-
uncertainty of this valuewill also be stated. pacitance pickups are among the many devices used to
locate theliquid surfaces. Theposition of the deviceis
2.1.2.2Density [kg/m3]. Among themanometer
then transferred to a scale with the help of a vernier,
fluids, mercuryis often favored because of its high and
leadscrew,gageblocks,etc.Cathetometersarefre-
well-known density. (Caution: See Note, below). Var-
quently used both to locate the meniscus and to mea-
ious oils, water, and aqueous solutions are also used.
sureits position. On primary standardmanometers,
The density of liquids depends on temperature, and
readings are taken in the balanced condition pl = p2
corresponding measurement dependence on tempera-
and in the pressurized condition so that the constant
ture needs to be considered for most measurements.
offset betweenfluid meniscus andscale, caused by the
Some typical density data are given in Table 2.2. The
locating device, can be eliminated from the data.
actual density of a sample fluid may differ from the tab-
A manometer with very high accuracy using an
ulated value due to variations in composition and con-
infrared interferometer to locate the menisci and to
tamination. In all but thecrudestmeasurements,an
measure their vertical movement is now under con-
effort should be made to determine the average tem-
struction at the National Bureau of Standards. A ma-
perature of themanometric fluid inall vertical compo-
nometer usingultrasonic transit time measurementfor
nents. For this purpose, thermometers should be
the determination of column length is commercially
attached to the columns or mounted veryclose to
available. A manometer using ultrasonic interferome-
them. Standard precautions shouldbe taken to ensure
try is used as the primary standard for the range
that the manometer i s at a uniform and constant
0.1 Pa to 1 O kPa at the National Bureau of Standards.
temperature.
While inthese manometers the difficult task of locating
NOTE: Mercury will alloy with many other metals the meniscushas moved from an observer to an appa-
such as copper,lead,tin,bronze, Monel and their ratus, and while the length measurementi s done auto-
alloys. There is evidence thatlnconel alloys, Zircalloy, matically in terms of the wave length of light or
10

PRESSURE MEASUREMENT ASMVANSI PTC 19.2-1987
ultrasound, many of the other difficulties, such as tem- consideration. Similarconsiderations should be ap-
peraturecorrection, tilt error,head correction, etc., plied to manometers of differentdesign and forvarious
remain. types of misalignment and tilt during or between
Scales, lead screws, and gageblocks are calibrated at measurements.
a reference temperature T, (Cl. If they areusedat Another important correction to the measured height
a temperature other than T,, this reading must be of a liquid-filledcolumn i s caused by the surface-
corrected by tension effects between the liquid and the solid walls
of the tube and the gas above it. This capillarity effect
causes the liquid surface to assume a convex shape (as
H = Hf*[1 - a(T - T,)] with mercury in glass), or a concave shape (as with liq-
uids wettingglass or metal). At the same time, the cen-
where ter of the meniscus is depressed below (with mercury)
HT* = reading taken with the scale at temperature T or raised above (with liquids wettingthe tube surface)
(II = thermal expansivity of the scale the level that it would assume in the absence of
H = corrected scale reading surface-tension effects. The capillary depression i s very
Errors in the column height determination are intro- sensitive to contamination of the liquid and of the tube
duced by misalignment and tilt ofa manometer. Figure surface. Tables have been published giving the capil-
2.7 illustrates the case for a manometer tilted byan an- lary depression as a function of meniscus height, sur-
gle 8 about a horizontal axis through the manometer facetension, andtubediameter (seeTables 2.3A, 2.3B,
tubes. The true difference AH in meniscus position and 2.4). These tables contain estimates at best and
is obtainedfrom the apparent meniscus position should be used only todetermine the size of the bore of
measurements Hl and H2/ by correcting as follows the tube for which the capillary-depressioneffect is less
(subscript zero indicates zero reading) than the required uncertainty of measurement.
AH = [ ( H ~-J H,~,) - ( ~ ~ 1] e
- H , ~ )cos 2.1.2.4 Reference Pressure pl [Pa]. Depending
upon the use ofthe manometer, the reference pressure
If the manometerwas vertically aligned for the zero pl in Eqs. (2-4) and (2-6) could be a vacuum, atmo-
reading and then tilted during the measurement, the spheric pressure, or any other pressure to which the
correction takes the form unknown pressure p2 shall be referenced. For other
than differential-pressuremeasurements, the reference
pressure must beknown with an uncertainty less than
AI-^ = ( ~ - ~ 1 COS e - - H~,)
the required uncertainty of the unknown pressure. For
J
)
a vacuum reference pressure, this can be achieved by
Figure 2.8 shows a manometertilted by an angle C$ i n continuous pumping with a mercurydiffusion pump or
the plane of the tubes. The correction takes again the an oil diffusion pump with a cold trap. The reference
form pressure should be measured with a McLeod or ther-
mocouple gage.
AH = [(Hl - Hlo) - (Hz - H2071 COS 4 Note that the calibration of thermocouple gages de-
pends upon the gas being measured. The vapor pres-
But, if the manometerwas vertical for the zerosetting sure of mercury of about 0.1 5 Pa (1.1 x 1O3 torr) at
and then tilted duringthe measurement, as is often the 20C (68F) sets a lower limit for the reference pres-
case due to the shifting load, the correction takes the sure in a mercury-filled manometer,
form
2.1.2.5 Head Corrections. For the calculation of gas
AH = (H, - H2)cos 4 - (Hlo - + [sind (air) heads, one muston occasion consider the factthat
a gas compresses under its own weight and that the
In these correction equations, the cosine error i s pro- density decreases with elevation as
portional to the pressure measured and,with a reason-
ably smallandconstantmisalignment, i s often dp = - egdh (2-7)
negligible. The sine error i s independent of the pres-
sure, proportional to the spacingbetweenthe where dp is the increment of pressure perincrement of
columns, and often large enough to warrant serious height dh.

A S M U A N S I PTC 19.2-1987 PRESSURE M E A S U R E M E N T
Column 1 Column 2
FIG. 2.6 GENERALPRINCIPLE O F ALL MANOMETERS
FIG. 2.7 ERRORS IN COLUMNHEIGHTDETERMINATION(MANOMETER

TILTED BY ANGLE e ABOUT A HORIZONTALAXIS
THROUGH THE MANOMETER TUBES)
12

PRESSURE MEASUREMENT ASMEIANSI PTC 19.2-1987
I
TABLE 2.3A TABLE
2.3B
DENSITY, THERMAL EXPANSION, A N D DENSITY, THERMAL EXPANSION, A N D
VAPOR PRESSURE OF WATER VERSUS VAPOR PRESSURE OF MERCURY VERSUS
TEMPERATURE IN OC (IPTS-68) TEMPERATURE IN OC (IPTS-68)
(Courtesy of the National Bureau of Standards) (Courtesy of the National Bureau of Standards)
~
Thermal Thermal
Expansion Vapor Expansion Vapor
Temperature, Density, Coefficient*10'6, Pressure, Temperature,
Density, Coefficient-10'6, Pressure,
OC kg/m3 K" Pa OC kg/m3 I
C
'~ ~~
Pa
~~
1 999.90 -50 657.1 181.5 2.6 x

2 999.94 - 33 705.9 181.5 2.95 X
3 999.96 -16 758.0 181.5 3.26 x
4 999.97 O 81 3.5 181.4 3.59 x 10-2
5 999.96 16 872.5 181.4 3.96 x
6 999.94 31 935.2 13582.75 181.4 4.36 x

7 999.90 46 1001.9 13580.29 181.4 4.79 x lo-2
8 999.85
1072.8 60 13577.82 181.4 5.27 x lob2
9 999.78 74 1148.1 13575.36 I 81.4 5.79 x lo-2
10 999.70 88 1227.9 13572.90 181.3 6.35 X
11 999.61 101 1312.7

12 999.50 114 1402.5
13 999.38 127 1497.7
14 999.24 139 I 598.6
15 999.1 o 151 1705.3 dP - mg
P RT
16 998.94 163 1818.3
17 998.77 174 1937.8
18 998.60 185 2064.1
which can be integrated to yield
19 998.41 196 21 97.6
20 998.21 207 2338.5
21 997.99 217 2487.4

22 997.77 228 2644.4
23 997.54 238 2810.1
24 997.30 248 2984.7
25 997.04 257 3168.7 A pressure p at level h can becalculated from the pres-
surep, at the level h = O using this equation. The cal-
996.78 26 267 3362.6 culation i s considerablysimplifiedbyusing the
27 996.51 276 3566.7
following approximation:
28 996.23 285 3781.5
29 995.94 294 4007.5
30 995.65 303 4245.2
GENERAL NOTE: Valuesarefor"ordinary" or "MeanOcean"

water, .e., standard isotopic composition of ocean water.
This approximation causes an error of less than 1 ppm

for h up to10 m ( - 33 ft) with air or nitrogen at room
temperature.
I the
From ideal gas law For differential-pressure measurementsat the refer-
ence level of a manometer, usually the level of the
mP lower meniscus, a head correction should be applied
e =-
RT to pl. If the reference port is open to the atmosphere,
then
where m i s the molecular weight, R is the universalgas
constant, and T is thetemperature of the gas. Inserting
Eq. (2-8) into Eq. (2-7) yields
13

FIG. 2.8 ERRORS IN COLUMNHEIGHTDETERMINATION(MANOMETER

TILTED BY ANGLE 4 IN THEPLANEOFTHETUBE)
Transducer
r1 P2
B H2
Reference level
FIG. 2.9 DIFFERENTIAL PRESSURE TRANSDUCER CONNECTED

TO THE MANOMETER
14

MEASUREMENT PRESSURE 19.2-1 ASMUANSI PTC 987
TABLE 2.4
CAPILLARY DEPRESSION OF MERCURY IN GLASS MANOMETERS
(SURFACE TENSION: 0.45 Nlm)
(Courtesy of the National Bureau of Standards)
Bore of Tube, Meniscus Height,

mm mm
~ ~~
... 0.2 0.4 0.6 0.8 1.o 1.2

10 0.069 0.137 0.202 0.264 0.322 0.375
12 0.040 0.079 0.117 0.1 53 0.1 a7 0.21 a
14 0.023 0.046 0.069 0.090 0.1 11 0.1 30
16 0.014 0.028 0.041 0.054 0.067 0.078
18 0.008 0.01 7 0.025 0.033 0.040 0.047
20 0.005 0.010 0.015" 0.020 0.024 0.029
22 0.003 0.006 0.009 0.01 2 0.01 5 0.01 7
GENERAL NOTE Capillary depression in mm at 20C and standard gravity.
where po is the atmospheric pressure at the reference Similar corrections can be derived for manometers
level of the manometer. Thedifferential height AH be- filled withdissimilar liquids or for manometers operat-
tween the menisci is counted positive upwards from ing at high pressures.
the reference level.
For another example, consider the connection of a
differential pressure transducerto the manometer(see
Fig. 2.9). The reference portsof manometer andtrans- 2.1.3 McLeod Gages
ducer are open. The transducer is at a level h above the
reference level of the manometer. The referencepres- 2.1.3.1 Principles. McLeod gages are based on the
sure for the manometer i s ideal gas law
PI = PO - geairAH
where Bair is the density of air at ambient pressure po where p and v are the pressure and volume, respec-
and temperatureT. The pressure at the test port of the tively, of a given quantity of gas before (subscript 1)
manometer is and after (subscript 2) isothermal compression. If the
volume of the gas before and after compression, and
PZ = ge@ -t PO - geairAH the pressure after compression are known, the initial
pressurep, can be calculated. Figure 2.10 shows sche-
where eLis the density of the manometer fluid. The matically a simple McLeod compression manometer,
pressure at the pressure port of the transducer is The vacuumsystem whose pressureis to bemeasured
i s connected through the tube A to a large volume B
with an attached,closed capillary C. The volume v in B
~3 = ~2 - gegh
and the attached tube and capillary can be separated
from the vacuum system by lifting the mercury reser-
where egis the density of the pressure-transmitting
gas.
voir D and raisingthe mercury meniscus in the flexible
The pressure at the reference port of the transducer is
hose E to the level a-a. The volume v1 of gas contained
i s then at the pressure pl. Subsequently thegas is com-
~4 = PO - geai$ pressed into the capillary C by further raising the reser-
voir untilthe meniscus in the tube A reaches the level
Finally,the differential pressureacross the trans- b-b of the top of the enclosed capillary. With the mer-
ducer is cury in the capillary at c-c,the volume ofthe gas is then
AP = ~3 - ~4 = - @air)- - eaiJ V, = bh (2-1O)
15

ASME P T C * l 7 - 2 8 7 m 0759670 0052LLO T W
where b is the capillary volume per unit length and h is 2.1.3.3 Ranges and Scales. McLeod gages are usable
the head of mercuryin tube A. The pressurein thecom- over therange from 2 kPa (15 torr) down to 1.3 x 1O-6
pressed gas is Pa (1O-* torr). Most commercial gages operate only in
the low vacuum rangefrom about 700 Pa (5 torr) down
~2 = PI + geh (2-1 1) to 0.1 Pa (7 x 1 torr).
Generally the distance between the meniscus of the
where g is the local acceleration of gravity ande is the mercury and the top of the capillary is measured with
densityofmercury at theappropriatetemperature. an attached length scale oracathetometer. For capillar-
From Eqs. (2-9) through (2-11) it follows that, for the ies of less than about 3.5 mm (0.14 in.) diameter and
pressure in the vacuum system, 1O0 mm (4 in.) length, the volume bh becomes a cor-
rection of less than 1O h to the volume v1 and the pres-
sure can be calculated from the simplified equation
gebh2
P1 =-
VI- bh
Figure 2.11 illustrates a simple, rotating McLeod

gage commonly used for the measurementof low vac-
uum as, for example, the reference pressure of manom- where K is the gage constant.
eters. The pressureto be measured is applied to part F, Simple types of McLeod gages frequently have a
which is normal to the plane of the paper and about pressure scale attached to or engraved into the capil-
which the entiregage can be rotated. To begin opera- lary tube C.
tion, the gage is rotated clockwise to collect all mer- 2.1.3.4 Gases andVapors. Within the precision
cury in the reservoir D. Sufficient time should be achievable with McLeod gages, the idealgas law holds
allowed in this position to let the pressurein the bulb B for all gases. Condensable vapors can generallynot be
and capillary C come to equilibrium withthe pressure measured with the McLeod gage. Corrosive gases and
at F. Then the gage isslowly rotated back into its verti- vapors should be avoided, since theywill cause rapid
cal position. This prompts mercury to run down deterioration of the performance of the gage through
through tube E and up into A thus cutting off the con- contamination of the capillary.
nection to the bulb B. Under the pressure of the mer-
cury column in A and D, mercury will begin to fill the
bulb B compressing the gas into the capillary C. As
soon as bulb B is filled, excess mercury will overflow 2.2 WORKING STANDARDS
from the pointed tip of tube A and collect in the reser-
Working standards convert an applied pressure into
voir around F until the level in D is atthe same height as
a suitable reading, voltage, frequency, or other signal
the tip of tube A. The uncertainty of measurements
that is a unique function of the applied pressure. Work-
made with this type of gage may be quite large dueto
ing standards are used to calibrate accurate pressure-
the somewhat unpredictable behavior of the mercury
measurement devices at one or more locations, and
at the top of tube A.
they are therefore required to have high long-term sta-
bility and a precision compatible with the intended
2.1.3.2 Capillaries, In order to reduce errors dueto purpose.Theymust be portable. This definition in-
capillar effects, a capillary by-pass of the same diame- cludes a wide variety of instruments, which may be
ter as C is attached to the tubeA. The meniscusin this classified as piston gages,manometers,transducers,
by-pass i s brought up to the level b-b. The volume per and gages. The advantages, expected precision, and
unit length of these capillaries is a critical parameter major operating characteristicsarediscussed in the
and is measured by determining the lengthof a known following paragraphs.
amount of mercury atvarious positions in thecapillary.
The shape of the inside of the closedtop should be as
flat as possible. 2.2.1 Piston Gages. Piston gages or deadweight testers
Sticking of the mercury in capillaries is a common consist essentially of a piston fitted into a cylinder.
problem. Capillaries with less than 1 mm (0.04 in.) Pressure applied to the cylinder exerts a force F on the
diameter are impractical for this reason. Extremeclean- piston:
liness of the gage and theuse of high purity mercuryare
necessary conditions for precise measurements. F = AeffP

ASME PTC*I7-2 87 Is 0 7 5 7 b 7 0 0052IiI I W
PRESSURE MEASUREMENT ASMVANSI PTC 19.2-1 987
f To vacuum system
b-
C.
About
760 mm
Open to
atmosphere
FIG, 2.10 McLEODVACUUMMANOMETER
17

FIG. 2.1 1 PIVOTING McLEOD VACUUM CACE

ASME P T C * 3 9 . 2 87 m 0 7 5 7 6 7 0 0052333 5 m
PRESSUREMEASUREMENT ASME/ANSI PTC 19.2-1 987
where A,ff is the effective cross-sectional area of the facturers. See Refs. [6],[8], and [9] for information on
piston-cylinder combination and p is the pressure. The calibration and use of piston gages.
force F is counterbalanced with deadweight loaded on
the top of the piston
2.2.2 Manometers, Manometers balance the
unknown
f = Mg (approximately) pressure against the weight per unit area of a liquid
column
where M is the mass of the weights and g is the local ac-
celeration of gravity. The effective area A,ff is deter- P = gen
mined by calibration against a transfer standard. See
paras. 2.1.1 and 3.2 for a detailed description of piston
gages. where is the density of the fluid used and H is the
Aeff is generally a function of temperature and height of the upper meniscus counting positive up-
pressure: wards from the lower meniscus of the fluid in a U-tube
or cistern manometer. This measurement can be done
by visually locating the menisci and transferring the
locations to a scale, or by means of capacitive, optical,
inductive, or ultrasonic devices, often in conjunction
where with lead screws and turns counters.
Ao = effective area at zero pressure and at the Major uncertainties can be caused by the tempera-
reference temperature ture dependence of the density of the manometric
a, and a,, = thermal expansion coefficients of the fluid, bycapillarityeffects, and bytiltofthe instrument.
cylinder and the piston, respectively For details see paras. 2.1.2 and 3.3. Mercury manome-
T = temperature ters foul easily when used with air or oxygen. Thisim-
Tref = reference temperature pairs their precision. Cleaning and refilling can usually
b, and b, = pressure coefficients for the area be accomplished without seriously affecting the cali-
The force generatedis accurately described by bration of the instrument.
Manometers normally cover the range from lo to
about 5 x lo5 Pa (1.5 x to 72.5 psi).Theaccu-
n
racy of the best commercial manometers, manual or
automatic, approaches 50 ppm at atmospheric pres-
sure. The precision of these instruments may reach1O
ppm. Like piston gages, manometers have excellent
where long-term stability provided they are carefully main-
Mi = mass values of weights i of density tained. Theyare difficult totransport. Some automatic
g = local acceleration due to gravity manometers haveelectrical outputs suitable for inter-
n = maximum number of weights i facing with automatic data-acquisition systems and
cair = density of the ambient air pressure controllers.
N = tare weight A detailed description of manometerpractices is
yc = surface tension correction available from the Instrument Society of America.
Piston gages have excellent long-term stability, varying
as little as a few parts per million (pprn) over 20 years
under favorable circumstances. The precision of the 2.2.3 Transducers. Pressure transducersconvert an ap-
best piston gages approaches 1 ppm. A precision of plied pressure into an analog or digital signal.The
better than0.1 O/O is achievable with most commercial transducing element may be a set of strain gages on a
instruments. membrane, a vibrating membrane or cylinder, an elas-
Piston gages cover the range from 2 x 1O3 to 1.5 x tically deformed capacitor, bellows, or a Bourdon tube
lo9 Pa (0.3 to 2.2 x lo5 psi). The calibration uncer- whosedeformation is measuredwith suitable means or
tainty of piston gages varies between 25 ppm and 200 any of a variety of similar schemes. Transducers range
ppm. Calibrations traceable to NBS are available from from simple strain-gage types to sophisticated force-
many manufacturers and certified calibration facilities. balanced Bourdon tubes. Many of these are discussed
Training courses on the care and operation of piston in Section 3. The selection of transducers suitable for
gages aregiven regularly at NBS and by several manu- use as transfer standards requiresreliable data on their
19

long-term performance. The National Bureau of Stan- 2.2.4 Gages. Pressure gages convert an applied pres-
dards provides a service for the determination of the sure into an analog pointer deflection that can be read
long-term performanceof transducers. on a scale.The transducing elements are most fre-
Zero drift, hysteresis, temperature, and altitude de- quently Bourdon tubes, bellows, or aneroid capsules,
pendence are major sources of error. A precision of the deflection of which i s coupled through gears and
100 parts per million of full-scale reading, excluding linkages to a pointer. Hysteresis in the transducing ele-
zero drift, can be attained by a few selected transduc- ment, friction in the gears and linkages, and tempera-
ers. Major advantages of transducers used as transfer ture effects limit the precision of pressuregages to
standardsare their portability, smallsize, automatic about 0.05%. Suchgagesare available for pressures
operation, and fast response. Transducers are available from about 1O3 to 7 x 10' Pa (0.15 to 1 x 1O5 psi), For
for the pressure rangefrom O. 1 to 1.5 x 1'O Pa (1.S X some types of gages, long-term performance i s well
1 to 2.2 X 1o5 psi). established.
20

ASME P T C * 1 9 - 2 87 l l B . 0 7 5 9 b 7 0 0052115 7 m
SECTION 3 - PRESSURE MEASUREMENT DEVICES
This Section is devoted to the characteristics of de- (b) Testing Environment

vices commonly used in the industrial environment to (7) Lab
measurepressure. It does not deal deeply with the (2) Controlled/field
theoretical characteristics of such devices,but rather i s (3) Field
intended as a guide in their selection and use. (c) Cost (Based on 1979 Dollars)
Because of the multiplicity ofdevices which can be (7) Inexpensive (under $250)
employed to measure pressure, this Section dealswith (2) Low (under $1 000)
the component parts of many such devices. For exam- (3) Medium (under $3000)
ple, a particular manufacturer may employ a bellows (4) High (above $3000)
coupled with an LVDT (Linear Variable Differential (d) Ease of Use
Transformer) to measure gagepressure, while an (I) Simple
equally capable device made by another manufacturer (2) Moderate
may employ a diaphragm andelectronic force-balance (3) Complex
techniques to achieve the measurement. Thus we have (e) Output
chosen to discuss the properties associated with each (I) Voltage
of the componentsto allow the user to understand the (2) Current
concepts employed in a device he may be considering (31 Frequency
for his particular measurement problems. Some de- (4) Analog Indicator
vices, particularly dial and piston gagesas well as (5) Digital Indicator
manometers, are complete unto themselves. Here the (6) BCD'
treatment is singular for the class of device. Once a (7) Pneumatic
choice of measuring device is made, i t i s recom-
mended that manufacturers be contacted for details of Tables 3.1 and 3.2 are presented as guides only. A
the applications and limitations of a particular unit. particular manufacturer may offer devicesof a particu-
lar type which fall outside these limits. Each such offer-
ing must be evaluated on its own merits; exclusion
from these tables doesnot necessarily indicate inferior
3.1 RANGES AND ACCURACIES performance with respect to listed devices. The listing
Usually the first considerations involved in selection of eachclass of device also indexes the paragraphs
of a pressure-measuringdevice are the magnitude of describing them, providing a handy cross-reference
the pressure to be measured and the accuracy of de- between ranges and device characteristics.
vices available to operate at the desiredpressure. (Ac- Not included in thesediscussionsare application
curacy includes the combined conformity, hysteresis, considerations such as temperature limitations or com-
and repeatability errors. See ISA S51-1). Other consid- patibility with corrosive media. These are so depen-
erations include theenvironment of intended use, ease dent upon manufacturing technique and device design
of use, and relative cost. The tables which follow are as to be beyond the scope of this Supplement, Here
intended to guide the user in the selection of appropri- again, manufacturers will provide assistance in selec-
ate pressure-measuring devices. The categories of en- tion,
vironment, ease of use, type of output, and relative cost As a general illustration for the use of Table 3.1 in ob-
are coded in the following ways. taining an instrument accuracy range, the following
(a) Type (see Fig. 3.45) procedure i s offered.
(1) Absolute (A)
(2) Differential (D)
'Binary coded decimal, using fourdata lines per digit, with 8-4-2-1
(3) Cage
(G) weighting of each data line.
21

TABLE 3.1 hl
SUMMARY OF PRESSURE MEASURING DEVICE CHARACTERISTICS I,
22
v
Paragraph Design Applicable Pressure Range, psi Accuracy Environment output Ease of Use
10-l lo3 106

I : : : ! : ! 4

3.2.1 Simple cylinder 200 - 1000 ppm 2 3 or4
3.2.2 Reentrant cylinder 50 - 1000 ppm 2 4

3.2.3 Controlled clearance 20-150 ppm 3 4
3.2.4 Pneumatic deadweight 100-1000 ppm 2 3or4
E 3.2.5 Absolute pressure 35-100 ppm 3 3or4

I
3.2.6 Vacuum 35-100 ppm 2 3
lo4 1 o-* IO0 10'

I I I : : : I
3.3.1 U-tube I I DorC 6-lkppm 1, 2, or 3 ... 1,2,or3 l-4
3.3.2 Cistern I I DorG 30-100 ppm 1 or2 1 or2 24
3.3.3 Inclined I I DorG 0.1 -10% full scale 1 or2 1
3.3.4 Micromanometer I I D or G 0.01 - 1% full scale 1 2 3
3.3.5 Fortin barometer H A 70 wm 1 or2 2 3

TABLE 3.1
SUMMARY OF PRESSURE MEASURING DEVICE CHARACTERISTICS (CONTD)
Paragraph Design Applicable Pressure Range, psi Type Accuracy Environment output Ease of Use cost
3.4.2.1 Force balance A, D, C 0.05-0.25% full 1 1,2,4,5 2 4

scale
A, D, G 0.25 -2.00% full 2,3 1,2,3, 7 1 2
scale
3.4.2.3(a) Nozzle baffle

Depends on Sensing
3.4.2.3(b) Capacitance n
Elements Used . *
3.4.2.3(c)(l) LVDT A, D, G 0.1 -1.0% full scale 2,3 1,2,3,7 1 1,2,3 5
.

3.4.2.3(c) Inductance devices l-u
CP
3.4.2.3(d) Potentiometric A, D, G l.O-10.0% full 2,3 1 1 1 -4
scale
I
g 3.4.2.3(e) Oscillating devices (See Table 3.2) A, D, G 0.02-0.50% full 1,2,3 1,2,3, 5,6 1,283 2,3,4
scale
1 o-l0 1 o9 100 IO3

I::::::::::::1
3.5.3.1 Mercury micromanometers - A 0.1 - 1.0% full scale 1 4 2 2
3.5.3.2 Butyl-phthalate manometer - A 0.1 - 1 .O% full scale 1 4 2 2
3.5.3.3 Diaphragm comparator A 1.0% full scale 1 1 2 2,3
3.5.3.4 McLeod gage I 1 A 0.05% full scale 1 4 2,3 2,3
3.5.4.1 Thermocouple gage I I A 1 .O - 5.0% full scale ,2 1,2 1 1
3.5.4.2 Pirani gage I I A 1 .O - 5.0% full scale 1,2 1,2 1 1
3.5.4.3 Ionization gage 4 I A 1 .O - 5.0% full scale 1 1 2 1,2,3,4
3.5.4.4 Molecular gage I I A 1 .O - 5.0% full scale 1 4 2 3

TABLE 3.1

SUMMARY OF PRESSUREMEASURING DEVICE CHARACTERISTICS (CONTD)
Paragraph Design Applicable Pressure Range, psi Accuracy Environment output Ease of Use

3.4.2.3 Pressure switch Depends on elements used A, D, C 1.0-5-O% full scale 1 1
2, 3 I,2
: 3.4.2.3 Piezoelectric I i A, D, G = 1 .O% full scale 1,2,3 1 2 3
ELASTIC [I 0][13] A, D,G 0.25-5.0096 full 2,3 4, 5 1 1

scale
3.6.1 Bourdon tube gage I 1 0.05 - 5.00% full 1,2 4,5 1 2
scale
3.6.2 Bellows gage t I A, D, G 0.25-5.00% full 2,3 4, 5 1 1 a

scale a
3.6.3 Diaphragm gage I I A, D, G 0.25-55.00% full 2,3 4, 5 1 1

scale
. ASME P T C * 1 7 - 2 e?' a'b7-59b70 0052337 b m
PRESSURE MEASUREMENT ASMWANSI PTC 19.2-1987

(a) Given a specific pressure level to be measured. (d)No typeof low absolute pressure (vacuum)
(b) Observe the applicable pressurerange of sensors.
pressure-measuring devices as listed on the left-hand (e) One type of elastic gage.
side of the Table. The above example suggests a range of pressure-
(c) From this, the design typesof available pressure- measuring devices for a specific set of pressure-
measuring devices that fulfill this requirement are measurement requirements. Further requirements
listed. would further narrow the list of devices.
(cf) The accuracy range of each design is listed on
the right-handside of the Tablealong with other impor-
tant information.
(e) The attainable accuracy is therefore known. 3.2 PISTON CACES
(0 Theexact accuracyforeach specificdevice, how- The piston gage is one of thefew measuring devices
ever, requires consultation with the manufacturer. which measures pressure in terms of the fundamental
A specific example for the use of Table 3.1 i s also units of force and area. Because it can also generate a
offered to show a detailed case study. pressure through applying a weight across a known
EXAMPLE: area, its use is frequently associated with a device
known as a "deadweight tester." While such use is an
A pressure-measuringdevice is required to measure
important application, i t i s not the one which falls
a pressure level of 10,000 psi under a controlled/field
within the scope of this Section. The basicequation for
environment [ll]. The question to beanswered is
the piston gage i s
which devices can be used and what accuracy levels
can be expected from each instrument.
ObservingTable3.1 permits a listingofdeviceswith P = F/A
their associated accuracy range and the paragraph of
where P is the pressure, F is force, and A is area,
PTC 19.2 where the device is described. This observa-
Measurement of pressure to an accuracy of 1 part in
tion gives the following.
10,000 or better can be made with certain types of pis-
(a) Two types of piston gages could be considered.
ton gages. In order to achieve this accuracy, the envi-
-
Design
Paragraph
Accuracy, ppm ronment in which the gage is to be used and certain
parametersof the insfrument itself must be considered.
Simple cylinder 3.2.1 200 - 1O00
Failure to consider these can introduce a considerable
Re-entrant cylinder 3.2.2 50 - 1O00
error. Acceleration due to gravity, air buoyancy, tem-
(b) No types of manometers. perature, surface tension of the fluid, weight of the
(c) Only pressure transmitters containing Bourdon fluid, and elastic deformation of the cylinder must be
tube sensing elementswill measure this pressure level. evaluated andcorrections made to reduce theerror in
The expected accuracyrange depends on the type of measurement,
detector that is used in the pressure transmitter. It must Piston gage designscommonly incorporate three de-
be remembered that a pressure transmitter or trans- signs of cylinders:
ducer is a device which contains a sensing element,a (a) simple cylinder
detector, and a means for transmitting the sensed sig- (b) re-entrant cylinder
nal to a remote location. An accuracy range of 0.02% (c) controlled-clearance cylinder
to 10% of fullscale canbe expecteddepending on the A platform with calibrated weights is balanced on a
type of detector. The types of detectors for controlled/ piston which is floated on the fluid withina cylinder. A
field environments are listed as follows: connection to the cylinder transmits thefluid pressure
from the process connection in whichpressure i s to be
Design
Paragraph
Accuracy, Oh of full scale
measured.
balance
Force 3.4.1.1 1.25 -2.00 See paras. 2.1 . I and 2.2.1 for detailed discussion of
piston gages.
Nozzle baffle 1
Capacitance
LVDT
Inductance
} 3.4.1 -3 0.1 -1.0
3.2.1 Simple Cylinder Piston Cage. The simple cylin-
der gage i s the mostoften used and is available with a
Potentiometric 3.4.1.3 1.0-10.0 range up to 83,000 kPa (1 2,000 psig),
In use, the piston gage is connected to the system un-
Oscillating devices 3.4.1 -3 0.02 -0.50 der test as shown in Fig. 3.1,
25

ASMVANSI PTC 19.2-1 987 PRESSURE MEASUREMENT
TABLE 3.2
SENSING ELEMENTS
Pressure
Applicable
Design
Paragraph Type
Range, psi
3.4.1.1 (a) Diaphragms

1
P
--
IO-^ 1OO lo6 lo3
A, D, G
3.4.1.1 (b)
3.4.1.1
3.4.1.2(a)
3.4.1.2(b)
3.4.1.2(c)
Bellows
Bourdon tubes
Rolling diaphragms
Bell-type element
Slack and limp diaphragms H

- H
A, D, G
A, D, G
D, G
D, G
D, G
Pressure is connected to the inlet line, which must in factors for this phenomenon. Generally, however,this
general be filled with a hydraulic oil. This oil is neces- is not a factor where accuracies2 0 . 5 % are involved.
sary to provide for proper operation of the piston and The optional gage (Fig. 3.1) is useful for estimating
cylinder. It also servesto keep corrosive or contaminat- the total weight to be used, but may be omitted for rea-
ing fluids from reaching theinternal parts of the mea- sons of economy.
suring system. If very large displacements hydraulic
of
fluid are required for a particular piston gage(e.g.,
large pistondiameter), anoil reservoir maybe added to
the system.This would beanexpandedarea of the inlet 3.2.2 Re-entrant Cylinder Piston Gage. The re-
pipe arranged so that largevolumetric changes would entrant cylinder gage i s usually used forhigher-
result in small height changes within the inlet column. pressure measurements; however, it can be used for
This is important because head effects can contribute lower pressures as well. The lower limit is usually de-
to reading errors. termined by the weight of the platform. Commercially
In use, calibration weights are addedto ortaken from available re-entrantcylinder gages areavailable with a
theweight platform until the piston rests somewhere in range of approximately 552 kPa to 276 MPa (80 psi to
its midposition; usually a fiduciary mark or scaleis 40,000 psi).
available to properly position the piston.It is also com- Figure 3.2 is a schematicrepresentation of a re-
mon practice to rotate the weights and piston while entrant cylinder piston gage. Note that a cavity is pro-
taking a final reading to minimize piston-to-cylinder vided around the outside of the cylinder so that the
friction effects. fluid pressure isexerted on the outsideas well as the in-
It must also be remembered that thetotal weight of side of the cylinder. This design reduces the clearance
piston, weight table, and calibration weights is used to between piston and cylinder at higher pressures and
calculate thepressure being measured. Manufacturer's thereby reduces the otherwise excessiveleakage of
literature will give the netweight of the table and pis- pressure fluid to tolerable levels.
ton. This must be added to the calibration weights for The operation of this piston gage is similar to that of
pressuremeasurement.Frequently,the calibration the simple cylinder gage. Because of the higher pres-
weights are corrected or calibrated for a particular pis- sures being measured, a motor-driven positive dis-
ton gage and maytherefore not be interchangeablebe- placement pump (A) is sometimes provided. This
tween gages of the same model or type. Manufacturer's pump is used to increase the system pressure to near
literature should again be consulted for this informa- the measurement value.A second hand-operated ver-
tion. nier pump (B) is used for the final adjustment. A moni-
Because the force of gravity varies with location and toring gage (F) is provided to allow tracking of the
altitude, some manufacturers will provide correction system pressure as it increases.
26

Pressure
under
7 Calibration weights test
t
Auxiliary gage
(optional) I
Inlet
inertline,
withfilled oil W
FIG. 3.1 SIMPLE CYLINDERPISTONCACE
3.2.3 Controlled-Clearance Cylinder Piston Cage, gages are easily operated and canbe used to measure
The controlled-clearancecylinder gage is, again, usu- down toabout 1.O kPa (4in. ofwater). A major advan-
ally used for higher-pressure measurement. However, tage of these gages is insensitivity to contamination.
as mentioned in Section 2, the National Bureau of Figure 3.4 i s a schematic diagramof the pneumatic
Standards maintains a group of controlled-clearance tester. In this type of construction a precision ceramic
piston gages covering the range from 35 kPato 2.5 GPa ball i s floated within a tapered stainless steelnozzle. A
(5 psi to 370,000 psi). Paragraph 2.1.1 should be con- flow regulator introduces pressure under the ball, lift-
sulted if amore detailed explanation of operating prin- ing it toward the annulusbetween the ball and nozzle.
ciples is desired. Equilibrium is reached when the vented flow equals
Figure 3.3 i s a schematicrepresentationof a the fixed flow from the supply regulator, and the ball
controlled-clearancepiston gage. It is similar to the re- floats. The pressure,which i s also the output pressure,
entrant type with the exception of the source of pres- is proportional to the load. During operation, theball is
sure for the external cylinder cavity. The motorized centered by a dynamic film ofair, eliminating physical
positivedisplacement pump (A) can be used to pressur- contact between the ball and nozzle.
ize the system to near the measurement point. By When weights are added or removedfrom the
opening valve (N) this pressure is exerted in the cylin- weight carrier, the ballrises or lowers, affecting the air
der (D) as well as the external cavity. When near the flow. The regulator senses the change in flowand ad-
measurement value, valves (N and M) are closed and justs the pressure under the ball to bring the system
the pump (A) stopped. The pressurein the cylinder (D) into equilibrium, changingthe output pressure accord-
is then adjustedby use of the hand pump (B). Similarly, ingly. Thus, regulation of output pressure is automatic
the pressure in the external cylinder cavity is adjusted with changes in load on the spherical piston (ball).
by use of hand pump (G). The internal and externalcyl-
inder pressure are monitored by gages (F) and (H), re-
spectively. The externalpressure is usually reducedto
3.2.5 AbsolutePressure Piston Cage. By enclosing the
just below the pressure in cylinder (D). Care must be
calibration weight platform and evacuating the cham-
taken to not allow the differential pressure between the
ber, the piston gage can be used to measure absolute
cylinder (D) and externalcavity to increase to a point
pressure (see Fig. 3.5). When testing in the region be-
where the cylinder is damaged.
low atmospheric pressure, the pressure sourcei s shut
off and the vacuumpump is used. The vacuum created
3.2.4 Pneumatic Deadweight Ball Cage. Ball gages us- allows the fluid (usually air or gas) in the piston to ex-
ing air or gas as the fluid are available with range capa- pand and lift the weight platform. The platform is ro-
bility of up to 7000 kPa (1000 psi). The use of such a tated by a motor drive, within the enclosure, to reduce
gage is usually limited by the available source of dry piston drag.
compressed gas, which mustbeabout 50% higher The accuracyof measurement is the same as for the
than the pressure to be measured. Pneumatic ball tester used in atmosphere, with the exception that the
27

A S M U A N S I PTC 19.2-1 987 PRESSURE M E A S U R E M E N T
External cylinder
FIG. 3.2 RE-ENTRANT CYLINDER PISTONCAGE
air buoyancy becomes a variable and introduces less The diaphragm separator can be one of a number
error as absolute zero pressure is approached. of devices such as those discussed in the paragraphs
on transducersandtransmitters in this Section, for
3.2.6 Vacuum Piston Gage. The vacuum piston gage measurement of differential pressure, or the separator
(Fig. 3.6) issimilar in design to the simplecylindergage may be integral to the piston gage.
used for measuring above-atmospheric pressure,with When using these devices, theweight applied to the
the exception that the weightsare hung from the pis- piston gage must bring the separator element td its null
ton. The vacuum range is dependent on the vacuum or zero-differential point. Selection of the diaphragm
pump used. Piston gagesfor commercial use are avail- devicedictates that it should be capable of measuring a
able to measure down to 25 mm Hg absolute (1 in. Hg low differential pressure compared to the pressure
absolute). The highest vacuumwhich can be measured measurement to be made. The accuracy of these de-
is dependent on the barometric pressure at the time vices is usually stated as a percentage of the range.
and place of use. Accuracies of 3 parts in 10,000 can be Therefore, if the range is low incomparison to the pres-
achieved if the piston gage reading is corrected for the sure being measured, the erroris less, Caremust beex-
environment in which is it used. ercised when applying pressure to the diaphragm
Control of the unit is by use of a valve to shut off the separator to avoid overranging the instrument, causing
vacuum source and a second valveto bleed air into the damage and introducing error in the measurement.
cylinder. The weight carrier must be rotated during
evacuation of thecylinder and at the time of measure-
ment reading. 3.3 MANOMETERS
The use of manometers, particularly in a field envi-
3.2.7 Piston Gage Measurement With a Diaphragm ronment, haslessenedas transducershavebecome
Separator. The piston gagecan be used to measure more accurate and accepted. However, thesimplicity
pressure in processes where the process fluid is not of construction and closeapproximation of many spe-
compatible with the gage by use of a diaphragm sepa- cialized designs to primary standards helps maintain
rator (see Fig. 3.7). their place in certain applications. A few of these

- ASME P T C * 1 9 * 2 8 7 m 0759b70 R052323 8 W
..
.
-,
-+

7
Reservoir r External cylinder

cavity
FIG. 3.3 CONTROLLED-CLEARANCE CYLINDERPISTONGAGE
In from
pressure
SlJPPlY
out to
instrument
under test
FIG. 3.4 PNEUMATICDEADWEIGHT BALL GAGE

[Courtesy of Ametek, Inc. (Mansfield & Green Division)]
(Reprinted from "Pressure Gauge Handbook," Marcel Dekker, Inc.,
New York, 1985, by courtesy of Marcel Dekker, Inc.)
29

ASME P T C * K L 9 * 2 8 7 W
ASMElANSl PTC 19.2-1 987 PRESSURE MEASUREMENT
J B From pressure
source
tr
FIG. 3.5 ABSOLUTE PRESSURE PISTON FIG. 3.6 VACUUM PISTON GAGE
GAGE
designs are described here. The detailsof manometer A refinement of the cistern manometer, which en-
operation are discussed in paras. 2.1.2 and 2.2.2, and ables the construction to be mostly from nontranspar-
in the reference publications. Specialversionsare ent materials, is to add a floating scale with suitable
available forlow absolute pressure (vacuum measure- low-friction bearings to the narrow leg. A small win-
ments) (see para. 3.5). dow with a fixed hairline index allows an operator to
take all readings at a convenient elevation. The addi-
tion of a vibrator to the outside of the narrow leg will
3.3.1 U-Tube Manometer. The basic form of the ma- eliminate errors due to bearing friction, but not those
nometer is a U-shaped tube with the legs vertical. Fig- due to surface tension on the float.
ure 3.8 shows the manometer being used to measure
absolute pressure in a pipe, and Fig. 3.9 shows its use
with a flowmeter element measuringdifferential pres-
3.3.3 Inclined Manometer,If a manometer is inclined
sure. Theliquid inthe U-tube must be more dense than at an angle with the vertical, thevertical displacement
is still the same, but the movementof liquid along the
the fluid in the pipe and immiscible with it.
tube is greater in proportion to the secant of the angle.
The common form of inclined manometer is made
with a cistern, as shown in Fig. 3.1 l .
3.3.2 Cistern Manometer. The cistern manometer is
The scale is graduated to take account of the liquid
shown in Fig. 3.1 O. In this device, thearea of one legis
density, inclination, and cisternlevel shift so that read-
made substantially larger than the other, in the form of
ings will be in convenient pressure units such as equiv-
a cistern into which the narrow leg dips. The advantage
alent vertical centimeters or inches of water. A spirit
of the cistern is that the liquid level inside it will vary
level and leveling screws are usually provided, so that
only slightly while substantial changesof level occur in
the designed angle can be reproduced in installation.
the narrowleg. This facilitatesapplying a scale to only
This form of manometer is usually used for gas pres-
the narrow leg and correcting the graduations on this
sures, as for draft gages. The graduation intervals are
scaleforvariationsof level in the cistern. Readingaccu-
commonly 0.25 mm water (0.01 in. water) with spans
racy is increased,relativeto a U-tube with the same fill-
up to about 250 mm (10 in.).
ing liquid, by the fact that one not does have to add two
readings to obtain the pressure. The manometer can
read pressures below atmospheric by leaving the cis- 3.3.4 Micromanometer. A micromanometer is a preci-
tern opento the atmosphere andmaking the test pres- sion device for measuring very smalldifferential pres-
sure connection to the top of the narrow tube. sures, Depending on the reference pressure,it can also
30

ASME P T C * L S - 2 8 7 W 0 7 5 9 b 7 0 0 0 5 2 3 2 5 L W
null
Test
connection
Pressure
indicator -c
Flaw P4
I
P3
I separator
Diaphragm
kump
Piston
gage
FIG. 3.7 PISTONGAGE MEASUREMENT

W I T H A DIAPHRAGM SEPARATOR
FIG. 3.9 U-TUBE MANOMETERFOR

DIFFERENTIAL PRESSURE
the well, independently of the micrometer, so that the

meniscus in the inclined tube coincides with the index
when the micrometer reads zero. With a differential
pressure applied between the well and the upperend
of the inclined tube, the meniscus moves away from
where the index. It is brought back to coincidence by raising
the well with the micrometer screw. The microma-
P = absolute pressure in pipe
nometer reading then measures the differential pres-
Pa =atmospheric pressure
g =magnitude of local acceleration due to gravity sure in terms of head of the manometer liquid. Since
P =density this i s a null method, it i s capable of refinement to high
precision.
FIG. 3.8 U-TUBE MANOMETERFOR

ABSOLUTE PRESSURE
3.3.5 Fortin Barometer. A Fortin-type barometer(Fig.
3.1 3) is an absolute pressure mercury manometer spe-
cifically designed for the purpose of measuring atmo-
be used to measure absolute pressure or near- spheric pressure. It comprises a vertical glass tube of
atmospheric pressure. 6.35 mm (0.25 in.) bore or larger for more precise in-
One form of micromanometer (Fig. 3.12) is essen- struments, sealed at its upper end, and with its lower
tially an inclined tube with a vertically moveable well. end immersed in a cistern of mercury. The upper end
The inclined tube i s short andset on a nearly flat slope of the tube is evacuated and the surface of the cistern
for high sensitivity. This tube is not graduated but is mercury i s exposed to ambient atmospheric pressure
merely provided with a fixed index. The well is moved which forces mercury to rise in the tube to a height cor-
by means of a micrometer screw and is connected to responding to the atmospheric pressure, The level of
the inclined tube indicator with a flexible hose. In use, the mercury meniscus in the tube is measured by aver-
the instrument is first zeroed by adjusting the height of nier index moveable relative to a fixed graduated scale.
31

ASMWANSI PTC 19.2-1987 PRESSURE MEASUREMENT
P1
P2
FIG. 3.1 1 INCLINED MANOMETER
P = ~ ( P ~ H ~ - P I H ~ ) + P ,
where notation is the same as in Fig. 3.8
How these pneumatic and electronic signals are de-
rived from a pressure signal is the subject of this Sec-
tion. A transmitter or a transducer is a device which
FIG. 3.10 CISTERN MANOMETER contains a sensing element,a detector, anda sender (a
means for transmitting the sensed signalto a remote lo-
cation). Three concepts are needed at this point:
(a) sensing element - a device which receives a
The level of mercuryin the cistern is adjusted to a fixed pressure signal and convertsit to information useful in
reference point ofivory by means of a displacer screw another form;
operating against the flexible bottom of the cistern. The (b)detector- a device which converts a sensing ele-
tip of the ivory point corresponds to the zero of the ment force or motion into a useful signal;
reading scale. Readings mustbe corrected for nonstan- (c) sender - a device which amplifies and/or con-
dard temperature,gravity,and capillary depression verts the detector signal into a transmittable signal.
which may be computed, and also for instrument im- Sometimes the detector and sender are one and the
perfectionswhich can only be detectedby comparison
same.
calibration. Implicit inthe use of a transmitter is a receiver to de-
tect the remote signal. Beyond mentioning the need for
such devices, they will not be discussed in this work.
3.4 PRESSURE TRANSMITTERS Basically, all pressure transmitters convert a pressure
The previous paragraphs dealt with devices for con- into either a force or displacement, and then use the
verting a pressure into a mechanical analog of that detectorhender to remotely indicate that force of
pressure. For example, the manometer exhibits a dif- displacement.
ference in the height of a liquid column, For practical
reasons, it is difficult totransmit these mechanical sig-
nals over large distances. With the advent of modern
3.4.1 Sensing Elements
control systems, however, the need developedfor this
transmission capability. Pneumatic transmission was 3.4.1.1 Elastic Sensing Elements.Elastic sensing ele-
first to be employed and still enjoys a worldwide ac- ments include diaphragms, bellows, and Bourdon
ceptance. Subsequently, with evolving refinements in tubes. There are many design variations of each, in-
components and techniques, electronic instrumenta- cluding hybrid designs that combine features of both
tion became a practical means for transmitting infor- diaphragms andbellows. The purpose, in each case, i s
mation over even greater distances, Transmission of one of the following: (1) to convert pressure changeto
signals representing measured pressure is frequently force change, (2) to convert pressure change to dis-
accomplished by varying an air pressure through tub- placement, or (3) to act as an interface betweenfluids
ing or an electrical current through wires to the remote while transmitting pressure, undiminished from one to
location. the other.
32

ASME P T C * 3 7 * 28 7 m 0757670 0052327 5 W
PRESSUREMEASUREMENT ASMVANSI PTC 19.2-1987
High-pressure
connection 7 -owpressure
connection
/Manometer scale
graduated inches
and tenths
Reference
calibration point
for fluid meniscus
\
- Well
position
indicator
Clear plastic
cover 7
/We11 position
indicator scale
Pyrex glass tube graduated inches
\ and tenths
Manometer scale
graduated inches
tenths
and 7
/Micrometer wheel
Operating
hand-wheel
Level
7'
4
Leveling
screw
I
P L
FIG. 3.12 MICROMANOMETER(NULLREADING)
33

ASME P T C * 3 7 - 2 87 sl 0759b70 0 0 5 2 3 2 8 7 m
ASME/ANSI PTC 19.2-1987 PRESSUREMEASUREMENT
3.4.2.21, and drift at the extremesof the rated pressure

Closed end range.
When the travelof a single metallic diaphragm is in-
sufficient for the application, pairsof diaphragms can
be assembled into capsules by connecting their outer
Reading level perimeters (e.g.,an aneroid-barometer capsule), and
several capsules canbe fastened together attheir cen-
ter disksto form a capsular stack.A hole through the in-
terior center disks allows the applied pressure to
Glass tube
simultaneously inflate all the capsules, and the stack
travel is equal to the sum of the individual capsule
travels. Thus, multiple diaphragms can be stacked to
act as a bellows, andwelded assemblies ofthis type are
commonly marketed as welded bellows.
When used as a force element, the most important
Glass performance characteristicis effective area. This i s the
cylinder Datum point area which, when multiplied by the applied pressure,
produces a force at the diaphragm center. It is different
from the actualprojected surface area, because,unlike
a piston, the outer perimeter of a diaphragm is sup-
ported, and some of the force on the actual area is re-
Leather sac sisted by this support. The remainder is the useful force
and it is resisted by whatever is attached to the dia-
phragm center. For a flat, circular, uniformly corru-
gateddiaphragm that is free from radial biasing forces,
Datum the effective areamay be calculated by the formula
adjusting [12]:
screw
FIG. 3.13 FORTIN BAROMETER
where
A, = effective area
D = outer clamping diameter
(a) Diaphragm. A diaphragm i s a pressure- d = diameter of rigid circular center disk
responsive membrane, supported at its outer perime- By inspection, it is seen that a large center disk pro-
ter. When used to generate force or displacement, the duces a large effectivearea. The centerdisk, however,
detector is attached to a central point or disk substan- reduces the surface available for flexing, thereby in-
tially in the same place as the perimeter. creasing the stiffness and reducing the travel.
Diaphragmsensingelements cover thepressure Most diaphragms are used in such a way that they
range from less than 250 Pa (1 in. of water) in draft can neither be classified as purely force or displace-
gages and transmitters to 140,000 kPa (20,000 psig) in ment elements, but must partially perform as both. This
instrumentation for plastic extruding equipment. For situation is handled by determining the effective area
very low pressures, thesensingdiaphragm will be and the spring rate. For an applied pressure, the useful
quite large and probably have a large-diameter rigid force at a required travel is found by subtracting the
center disk for maximum effective area. For the mea- product of spring-rate times travelfrom the product of
surement of high pressure,the diaphragm diameter effective area times theapplied pressure. It is assumed
may be as small as 7.5 mm (0.3 in.). that thespring rate remains constant over the required
Used as a displacement element, theimportant per- travel. Indeed, variations in either the spring rate or ef-
formance characteristics are travel(including its linear- fective area overthe required travel result in nonlinear-
ity withapplied pressure), hysteresis, energy (see para. ity of motion and force.
34

PRESSUREMEASUREMENT ASME/ANSI PTC 19.2-1987
Diaphragm materials areselected both for their would be too stiff, and as displacement elements
physical properties and for compatibility with the ma- where the required travel i s too great for a single dia-
terials which they contact, They include a large variety phragm or diaphragm capsule.
of elastomers and metals.
3.4.1.2. Inelastic Sensing Elements.Unlike the elas-
(I) Elastomers.Elastomersare well suited for tic sensing elements, aninelastic element does not op-
many applications,particularly when reinforced with a
pose the pressure applied to it. It is, in fact, designed to
fabric insert. They canbe molded with corrugations to
avoid opposing it. It acts only as a piston to sum the
increase their linear travel andcan be made extremely
pressure applied over its effective area and transmit
flexible to respond to small changes in applied pres-
that force to a secondary element.
sure. For sensorapplications, they are frequently used
(al Rolling Diaphragms. Rolling diaphragms can
with external spring systems.Elastomer diaphragms
function both as diaphragm and bellows. They are
are widely used in pneumatic valve actuators, pneu-
made from elastomers, often with fabric reinforce-
matic relays andcontrol equipment, andin devices op-
ment, to provide a positive seal in piston and cylinder
erating at very low gagepressures (draft range). An
applications. Clearance between the piston and cylin-
important concern to the user of instrumentation with
der i s kept large enough for a single diaphragm corru-
elastomer diaphragms i s the extreme operating tem-
gation to roll between the cylinder walland piston wall
peratures. Doesthediaphragm remain sufficientlyflex-
as the piston moves through its stroke. Typical dia-
ible at low temperature to allowa satisfactory response
phragm thickness atthe corrugation is of the order of
time? Will it permanently deform at high temperature,
0.25 mm (0.01 in.) depending upon the material, the
causing reduced service life or a deterioration of
maximum pressure differential, and the piston-to-
performance?
cylinder clearance.
(2) Metallic Diaphragms. Metallic diaphragms
The advantages of this type of diaphragm include a
are less flexible and have less travel than similar de-
high ratio of effective area-to-cylinderarea, a constant
signs made from elastomers, but they also have many
effective area, much more travel than available in other
superior features. Properly selected for the application,
diaphragm designs, low spring rate, and the ability to
they can provide excellent corrosion resistance and
be used with fairly high pressures to approximately
undiminished performance at extreme temperatures,
3.4MPa (500 psi). Limitations are found in the physical
They can be welded into liquid-filledsystems that must
and chemical properties of the elastomer that i s
withstand continuous vacuum service. They can also
selected,
be designed to perform with very low hysteresis and to
(b) Bell-Type Element. The bell-type element is gen-
have virtually no long-term drift. Applications include
erally useful in the range between0.25 to 2.5 kPa (1 to
field-mounted transmitters and transducers, recorders,
1O in. H,O). It is a simple device as shown i n Fig. 3.1 4.
barometers, andmiscellaneous process control
When PlA1 = P2A2,the weight of the bell and force
equipment.
exerted by the counterbalancing spring cause the bell
Ib) Bellows. Bellows used for pressuremeasure- to rest on the bottom of the vessel. A pressure applied
ment usually have theshape of a hollow cylinder with
to the undersideof the bell will cause it to rise to a new
one end closed and sidewalls deeply corrugated to al-
equilibrium point which just balances the increased
low longitudinal force or motion in response to the ap-
force exerted by the new pressure according to the
plied pressure. This pressure maybe applied to either
equation:
the inside or the outside, but the preferred loading for
best performance is from the outside, operating the
bellows in compression. P2 =
PIAl + Mg - K(X - X,J -B
Bellows can be molded from elastomers, be formed A2
from metal tubing either hydraulically or by rolling, or
be fabricated from center perforated sheet-metal dia- where
phragms, alternately welded together at the centers P, and P2 = applied pressures
and otsides. Very small, thin-walled bellows can be A, and A, = effective areas of top and bottom surface
made from certain metals, such as nickel, by plating or of bell
vacuum-depositing onto a formed core which is later K = spring constant of the counterbalance
melted, dissolved, or chemically removed, X - X, = vertical displacement of the bell
In size, bellows range from effective areas less than B = buoyant force
those available in diaphragms to those equally as large. M = mass of bell
They are used as force elements where diaphragms g = local acceleration of gravity
35

ASMUANSI PTC 19.2-1987 MEASUREMENT PRESSURE
A , = Area of top side Counter-balance spring

Atmosphere
or reference
I /
Seal fluid
A2 = Areaof
underside
FIG. 3.14 BELL-TYPEELEMENT
Note that all terms except X are constant and can be

nulled, leaving f 2proprotional to displacement. This
device is also capable of measuring differential pres-
sure if A2 = A , .
As with the manometer, the seal fluid must be com-
patible and immiscible with the process fluid. Fre-
quently, mercury is chosen.
(c) Slack or Limp Diaphragm. The slack or limp dia-
phragm i s usually employed for pressures between 0.1
and 5 kPa (0.5 to 20 in. H,O) and is frequently associ-
ated with furnace-draft and air-duct pressure measure- Pressure
ments. It employs a flexible, non-metallic diaphragm,
frequently of leather or a thin, neoprene-like material.
The pressure acts against the diaphragm(see Fig.3.1 5) FIG. 3.15 SLACK DIAPHRAGM
causing it to deflecf a flat spring, The design of the
spring and associated mechanism is such that deflec-
tion is proportional to pressure.By enclosing both
sides of the diaphragm, differential pressures less than system of low spring resistance to actuate a very sensi-
70 Pa (0.3 in H20)can also be measured. tive detector of displacement. In a pneumatic device,
this detectorwill be a nozzle-baffle system; in an elec-
tronic design it may be a vane thati s part ofa magnetic
3.4.2 Detectors. One of the basic means for pressure
inductor circuit. In either case, small changes in posi-
sensing is to convert pressure into a corresponding
tion cause the output signal to change, This changeis
force or displacement.Once in that form, a number of
used to produce a feedback force equalto and oppos-
devices can be employed to detect that displacement
ing the force caused by the unknown input pressure.
or force. Some of these devices are complete; that is,
Thus, the systemis maintained in continuous force bal-
they are a complete detectorhender. Others are em-
ance. It i s important to understand that the motion of
ployed as part of a system used to detect and balance
the sensing element, when used in a force-balance sys-
force.
tem, is less than 0.25 mm (0.01 in.).
3.4.2.1 Force-Balance Concept.This refers to a con- The purpose of using a force-balance system with an
cept in whichthe known effective area of a sensing ele- elastic sensing element might be to gain oneor more of
ment may be applied to advantage in externally the following advantages:
powered instrumentation. In such a system, theun- (a) extend the working pressurerange of the ele-
known pressureactsagainstthesensing element to ment;
produce an input force. This force acts through a lever (6) increase the fatigue life of the element;
36

(c) improve performance characteristics such as cal- k = spring rate, which may be dependent upon
ibration, linearity, and hysteresis; displacement but can usually be considered
(d) reduce volume transfer; constant over theworking range
(e) allow closer stops for better overrange protec- P = pressure
tion; The dependenceof spring rate on displacement is min-
(f) avoid the problem of friction in linkages; imized in the design of the sensing element. However,
(g) provide, in transmitters, theability togreatly sup- the motion must then be converted to a useful indica-
press or elevate the zero of the output signal, using tion. Inevitably, thereare frictional or reaction forces to
biasing springs in the lever system. This is important overcome and these losses must be supplied by the
when small variations in pressure, occurring over a sensing element. In order that these forces produce
limited range, are all that is of interest. small errors, theenergy from the element must be suffi-
Sometimes there is an amplifying device in the force- ciently high, andthe instrument manufacturer mustse-
balance circuit between the displacement detector lect a detector that will minimize the losses.
signalandthe instrument output. In pneumatic
transmitters, it is the output relay which increases the 3.4.2.3 Detector Types. Several types of detectors
open loop gain and also provides output air flow ca- exist which can be used to implementa force-balance
pacity (volume of air per unit time). This relay may it- or displacement-detectingtransmitter. These include,
self be a force-balance device (seeFigs. 3.16A and but are notlimited to, nozzle-baffle arrangements,
3.1 6B).The electrical equivalent may be an AC ampli- LVDTs, capacitance detectors, photocells, potentiom-
fier with electronic circuitry to convert the amplified eters, strain gages, and inductance devices. The most
AC signal to a D C output, usually 4 - 20 or 1O - 50 commonly used devices are discussed below.
milliamperes (see Fig. 3.1 7). (a) Nozzle-Baffle (Flapper-Nozzle). The mechanical
Feedback force, either linearly proportional to the simplicity and inherent reliability of nozzle-baffle sys-
output signal, or proportional by some other desired tems account for their extensive use as displacement
function such as square root, canbe generated by vari- detectors in the instrument field. A typical system,
ous means. In a pneumatic transmitter,the output shown diagramatically in Fig. 3.1 8, has a supply orifice
pressuresignal might simply pressurize a feedback and a nozzle connected by a nozzle tube and assoc-
bellows. In a pneumatic pressure recorderor indicator, ated capacity chamberC.
the position of a pen or pointer linkage can be used to When thedistancex between the nozzle and a baffle
extend a feedback spring connected to a servo force plate becomes sufficiently small, flow from the nozzle
beam. In an electronic pressure transmitter, theoutput is restricted, and thenozzle backpressurePz increases.
current signal can be fed into a force coil in a perma- It i s this pressure that i s the usefuloutput of the system,
nent magnetic field, Figure 3.19 shows how nozzle backpressure varies
as a function of the nozzle-baffle gap. Nozzle sensitiv-
3.4.2'2 Displacement Elements. If a sensingelement ity, defined as the ratio of change of nozzle backpres-
is used so that the applied pressure is dissipated primar- sure with change of baffle gap, is the slope of this
ily inmoving the element against some spring force, it curve. Its units might be kPdmm or psi/in. The useful
is termed a displacement element,(Frequently,the part of the curve is where the slope is high and nearly
misnomer "motion balance" i s also applied to this linear. Nozzle sensitivity can be increased by either in-
class of instruments.) Thiscondition is satisfied by the creasing the diameter of the nozzle or by decreasing
spring rate of elastic elements or by the external springs the diameter of thesupply orifice.
used with inelastic elements. A practical limitation, however, is that the baffle on
Pressure is then related to displacement of the ele- higher sensitivity systems must operate closer to the
ment in the following manner. nozzle, increasing the need for nearly perfect align-
ment of the nozzle with the baffle,
The time constant of a nozzle-baffle system takes the
p=- -kx form of a linear Lag identical to that obtained from a
A simple RC (resistance-capacitance) system. Equivalent
resistance, however, is not equal to total pressure drop
d'ivided by total flow.
where As nozzle backpressure varies with the distance of
x = displacement the baffle from the nozzle,so the nozzle reaction force
A = effective area of element is also a variable. This effect is probably best treated, in

Flexible strip Span screw
Bias spring 7
Vernier scale
/-
1 Follow-up bellows
Force beam w -0

-I
Span Zero screw A
scale *
5
.

nl
Zero spring
m
, Capacity 11 4
Vent or tank 11
drain v Thryst qivot II
f I
va Iv. and seal
I - output
Bleed
Relay stem
output
Relay
Damping
adjustment
FIG 3.16A SCHEMATIC OF A PNEUMATIC FORCE BALANCE DIFFERENTIAL

PRESSURETRANSMITTER
(Courtesy of Taylor Instrument Company)
ASME P T C * 3 9 * 2 8 7 P ' 0 7 5 7 6 7 0 0 0 5 2 3 3 3 O m
PRESSURE MEASUREMENT ASME/ANSI PTC 19.2-1 987
Y
2
M
n
Y
39

Light source
Photovoltaic cells
\ /
\ /
I
Mirrnr . \
\ /
/ Amplifier
Reference
pressure
\-
a
'C Precision
resis
Force balancing
coils
1000.00
Quartz Bourdon
I I
tube Test pressure Digital voltmeter
readout
FIG. 3.17 SCHEMATIC OF AN ELECTRONICFORCEBALANCEDIFFERENTIAL

PRESSURE TRANSMITTER
(Courtesy of Ruska Instrument Corp.)
O "
P2 u R b
backpressure
P2
I I
Air flow QI
R,, = orifice resistance

Rn = nozzle resistance
b = baffle resistance
FIG. 3.18 TYPICALNOZZLE-BAFFLESYSTEM

ASME P T C m L 9 - 2 87 m 0757670 0052335 LI
PRESSURE MEASUREMENT ASMUANSI PTC 7 9.2-1 987
140
l-
120
1O0
m
4 80
2-
2ln Sensitivity 2500 kPa/mm
EP (9200 psi/in.)
Y
o
m
'u 60 Nozzle diam. 0.64 mm
1
z
40
20
O
O 0.04 0.08 0.12 0.16 0.20
Nozzle-Baffle Gap, mm
FIG. 3.19 NOZZLE BACKPRESSUREVERSUS GAP FOR

AN ELEMENTARY NOZZLE BAFFLE

ASME P T C * 1 9 * 2 87 W C1757b70 0 0 5 2 1 3 b b W
ASME/ANSI PTC 19.2-1987 PRESSURE M E A S U R E M E N T
the design of a pneumatic system,asan equivalent plification of very small pressures (deflections). With
spring constant: proper choice of diaphragm thickness, sensors of this
construction aremade with full scalespans up to
k , = S,A, 20,000 kPa (3000 psi) anddown to 13 Pa (0.1 mm Hg)
for measuring gas pressures with a dielectric constant
where S, is the nozzle sensitivity and A, the nozzle near unity. Pressurechangesas small as 1 x 1O"j of
area. The airconsumption of a nozzle-bafflesystem is sensor full range can be detected.
dependent only upon the orifice size, the supplypres- Sensor nonlinearity can be made as low as 0.01 of
sure, and nozzle backpressure. When the flow is sub- reading by specialprocessingand linearizing tech-
sonic (P, > 0.53P,), as i n thenormal range of niques. More typically, sensors are made to 0.05/0 of
instrument operation, and the orifice is sharp-edged, reading nonlinearity,Hysteresis is typically 0.01 -
air consumption may be computed from 0.02% of reading.
A variation of the double-sided sensor i s a single-
sided construction that is useful in absolute pressure
Q, = . .SPZ(P-.) measurements of extremely corrosivegases (reference
Pa
side evacuated and sealed)or unidirectional differen-
tial pressure measurementswhere the reference side is
where exposed to a noncorrosive and clean gas (Fig. 3.21).
QI = flow of free air (referenced to atmospheric Since the electrode assembly is removedfrom the mea-
pressure) through the supply orifice suring side, the dielectric constant of the material be-
C, = velocity of sound in air = 344.2 m/sec ing measured is no'longer a factor.Thus, a single-sided
(13,550 in./sec) at 21 " C (70F) construction can be used with conductive, radioactive,
A, = orifice area or heavy organic media and gases at high pressure. The
pa = atmospheric pressure sensing technique can properly be called "curvature
P, = supply pressure (absolute) sensing" versus "deflection sensing" for the double-
P, = nozzle backpressure (absolute) sided sensors.
Successful operation of a nozzle-baffle systemde- Because of the very small diaphragm motions that
pends primarily upon proper alignment and a good air correspond to full pressure span in either of these de-
supply. The baffle should be capable of capping the signs, temperature will often have a significant effect
nozzle by normal driving means(.e., without addi- on zero stability, span, and linearity. The double-sided
tional force) so that maximum backpressure is very sensor design, however, will have some advantage in
close to supplypressureand,at maximum working that symmetrical structural changes tend to cancel at
backpressure, the nozzle and baffle arenot incontact. zero. Some sensors are maintained at a constant tem-
The supply air must be constant pressure for repeat- perature by integral heatersproportionally regulated to
able backpressure output and should be free of con- a temperature slightly above the ambient in whichthey
tamination thatcould plugthe supplyorifice orcause operate.
baffle buildup. For low-temperature operation, it is im- Pressureand differential-pressure transmitters for the
portant that the supply airdew point be low enough, process industries also often use capacitance detectors
about 10C (18F) below ambient; to avoid ice forma- to convert the measurement to an electrical output sig-
tion. nal. Typically, these transmitters enclose the sensing
(6) Capacitance Detector. A commonly used diaphragm and fixed electrodes in a liquid dielectric
variable-capacitance typeof pressure transduceris the filling material thati s separated from the process fluids
differential design, adouble-sided sensor shown in Fig. by means of integral sealdiaphragms.These dia-
3.20. It consists of a tensioned metal diaphragm posi- phragms and their enclosing flanges be canmade from
tioned between two fixed electrodes. The electrodes very corrosion-resistant materialsto handle practically
are excitedby an AC signal of fixed frequency, suchas any process fluid.
10 kHz. When pressure deflects the diaphragm, the (c) lnductance Devices [13]. The inductance of a
change in capacitancebetweenthediaphragmand coil is determined by its number of turns (squared), its
each fixed electrode resultsin an AC output voltage at geometric form, and the effective permeability of the
the applied fixed frequency and an amplitude propor- surrounding material. Using a single coil, variations
tional to the deflection. The result is a high output caused i n any of theseparameters will result in a
pressure-to-electrical signal conversionwith excellent change of the coil's self-inductance, making possible
signal-to-noise characteristics, allowing very high am- many designs of displacement detectors. Some use a

ASME PTC*K17-2 8 7 m 0 7 5 7 6 7 0 0052337 8 m
Tensioned
metal
diaphragm
r1
Reference Measured
pressure pressure
Fixed
L Filter Filter
NOTE:
Filters used to prevent entry of particles that couldcreate shorts.
FIG. 3.20 SCHEMATIC REPRESENTATION OF A DOUBLE-SIDED VARIABLE

CAPACITANCE SENSOR HEAD
43

ASME P T C * L S * 2 8 7 m 0757b70 0052138 T m
ASMUANSI PTC 19.2-1 987 PRESSURE MEASUREMENT
r
3
Fixed
elecpodes
I
I
I
I
Reference
l
pressure
I
I
\
\
L Tensioned
metal
diaphragm
I
I
NOTE:
Baffle used to keep high-speed particles from impingingon the diaphragm.
FIG. 3.21 SCHEMATIC REPRESENTATION OF ASINGLE-SIDEDVARIABLE

CAPACITANCE SENSOR HEAD

ASME P T C * L 7 - 2 8 7 m
PRESSUREMMSUREMENT ASME/ANSI PTC 19.2-1 987
wiper action, as in a variable potentiometer, to change v
the number of turns. Others change the position of a

highly permeable core in the center of the coil, or
change the lengthof an air gap in an otherwise closed
magnetic pathof iron extending through the center of
the coil (variable reluctance). For the purpose of dis-
displacement of -+c3
-
Air gaps vary with
center member Iron cores
placement detection in instrumentation, thosedesigns

that avoid friction contact of the moveable member are
the more useful.
I'" ' O
Using multiple coils, the input displacement can be

made to vary the mutual-inductance (magnetic coup- FIG. 3.22 INDUCTIVE DISPLACEMENT
ling) between the coils.Mutual-inductance systems of- DETECTOR
ten differ from self-inductance systems only by the
choice of connecting the coils in series or parallel.
Higher sensitivity is obtained in inductive detectors
when they are designedso that an input change simul-
taneously causes an increase in one inductance and a
decrease in another. Then, either the difference or the
ratio of two inductances is used as a measureof the dis-
placement. Other advantages of using a difference or
ratio signal rather than the total inductance include
greater immunity to external magnetic fields and to
variations in temperature, supply voltage, and supply
frequency.Figure3.22illustrates an inductive dis-
placement detector that uses variable reluctance to FIG. 3.23 PREFERRED SCHEMATIC
provide differential or ratio output. REPRESENTATION OF THE LVDT
(d) The Linear Variable Differential 'Transformer
(LVDT). The LVDT is an electromechanical transducer
that produces an electrical output proportional to the tremely rugged moving element, and mechanical
displacement of a separate moveable core. As shown isolation of moving element.
in Fig. 3.23, three coils are equally spaced on a cylin- The electric features of the LVDT include voltage
drical coil form. A rod-shapedmagnetic core posi- output accurately proportional to core displacement,
tioned axially inside this coil assembly provides a path infinite resolution, and linear response characteristic
for magnetic flux linking the coils. When the primary (unlike potentiometer).
or center coil is energized with alternating current, (7) PressureTransducer.The LVDT is used to
voltages areinduced in the two outer coils. In the trans- measure the displacement caused by the expansion of
former installation, the outer or secondary coils are a Bourdon tube or diaphragm capsule when subjected
connected in series opposition, so that the two voltages to pressure. The design of the pressure-sensing ele-
in the secondary circuit are opposite in phase, the net ment is such that it causes a linear motion in the order
output of the transformer being the difference of these of 2 mm (0.080 in.) of travel. This travel is converted
voltages. For one central position of the core, this out- into an electrical signal by the LVDT. Typically, the pri-
put voltage will be zero. This is called the balance mary of the LVDT i s driven by a 12-volt square wave
point or null position. and the secondary produces an output amplitude mod-
When the core is moved from this balance point, the ulated by the stroke of the core, Thisoutput i s rectified
voltage induced in the coil toward which the core is and filtered to provide a DC level proportional to the
moved increases, while the voltageinduced in the op- pressure applied (see Fig. 3.24).
posite coil decreases. This producesa differential volt- The DC outputcan beeither current or voltage with a
age output from the transformer which, with proper common form being 4 mA to 20mA DC, linearly pro-
design, varies linearly with change in core position. portional to the pressure applied.
A preferred schematic representationof the LVDT is (2) Limitations. A basic limitation is the relatively
shown in Fig. 3.23. high amount of motion required to produce an out-
The mechanical features of the LVDT include com- put. This presents no particular problem for high-core
plete absence of friction and mechanical hysteresis, ex- travel elements (0.080 in. F.S.) but produces serious
45

ASME P T C * 1 7 * 2 87 m 0757b70 0 0 5 2 3 4 0 8 m
ASMWANSI PTC 19.2-1 987 PRESSURE MEASUREMENT
r """"""_ Servoamplifier -1
I
I
Phase detector
I
I
pressure 4
I
I
I
I
pressure -b I
I
I
l
I
_I
readout equipment
FIG. 3.24 LVDT USED AS A PRESSURETRANSDUCER

(Courtesy of IMO Delaval Inc., CEC Instrument Division)
(Reprinted from "Fundamentals of Temperature, Pressure and Flow Measurements,"
by R. P. Benedict, copyright O 1985
John Wiley& Sons, reprinted by permission of John Wiley& Sons, Inc.)
46

8 shortcomings in zero stability (mechanical andelectri-

cal) at small-core travels (< 0.01 O in. F.S.).
(c) Linearity. Linearity is essentially unaffected
by temperature changes.
(a) Zero hstability. Zero instability is not onlya (dl Potentiometric Detectors.This device can be
thermal but also a highly position-sensitiveerror due to made from a voltage-driven mechanism arranged so
the factthat the action of gravity on the core produces that the motion to be detected causes a changein the
motions that are a high percentage of the F.S. travel. output voltage. Thevoltage divider (potentiometer)is a
For example, when used with a core stroke of 0.010 in. three-terminal resistor. Two of the terminals are con-
F.S., an LVDT can exhibit zero shifts of -0.0001 in. nected to fixed points at either end of the resistor. A
due to position error. This, when converted to a volt- constant excitation voltage (AC or DC) i s applied be-
age or current output for pressure applied, could result tween them. The third terminal is connected to a slid-
in as much as 1OO/ F.S. position error. ing contact at theend of a wiper arm as shown in Fig.
(b) Thermal Effects. Thermal effectson span and 3.25. As the wiper arm moves, the output voltage be-
zero are linear and predictable and the effects can be tween the slidingcontact and the lower fixed terminals
compensated in the electronic circuit to better than changes accordingly to the relationship e, = vein,
0.1 O h F.S./lOOC with careful calibration. Tempera- where = R,/(R, + R*).
ture compensation in the order of 1 o/o F.S./I O O T are A potentiometric pressure transducer i s formed
regularlyrealized with standardtemperature- when the potentiometer is used to measure displace-
calibration techniques. ment caused by the expansion of a Bourdon tube,
(c) Zero Errors. Zero errors due to position are bellows, or diaphragm capsule subjectedto pressure.
generally not published by LVDT transducer manufac- The design of these sensing elementsis such that the
turers. However, this type of error is present in alltrans- full-span pressure change causesa linear motion in the
ducers using LVDTs due to the necessary mass of the order of 2 mm (0.080 in,) travel. This travel must be
core. It is a significant factor in capsule-typesensors be- amplified by mechanical linkage, sincethe potentiom-
low 24.9 kPa(1O0 in. H20).lt should be noted that the eter requires along stroke for accuracy and resolution
e position error is not significant in zero-based units

where the offset error can easily be adjusted out of the
output after installation. However, in non-zero-based
(on the order of centimeters). Sometimes pneumatic
a
servomotor i s used to provide the amplification.
(7) Features and Limitations. Potentiometric de-
elements acalibrated pressure sourcemust be applied tectors andthe pressure transducersusing them can be
after installation to remove the zero offset. small-size, lightweight, and easily installed. They are a
(3) Linearity. By carefullymatching the core simpleelectrical design, usuallyhigh-impedance
length to the selected LVDT, linearity correction can (5000 ohms typical), require low power drain, and
be introduced to correct for nonlinearity of the travel in may be operatedwith either AC or DC excitation,
thesensing element to theorder of 0.05% F.S. or The limitations are mostly mechanical. Friction and
better. mechanical hysteresis limit accuracy, and the potenti-
(4) Hysteresis. The LVDT has no measureable ometer is vibration-sensitive and subjectto wear of the
hysteresis. The hysteresis present i s a function of the resistive element. Normal wear can be predicted and
quality of the sensing element selected. the life made quite long in total number of cycles. An
(5) Span. The stability of the output span i s a func- unpredictable rapid variation called "dither" causes
tion of the electroniccircuit selection. The LVDThas a most early failures. In a pressure transducer, dither is
highly repeatable span with infinite resolution. caused by a rapid fluctuation in pressure, centered
(6) Thermal Effects about a given pressure. This tendsto dig a hole intothe
(a) Zero. The zero output when derived from resistive slide element.
the null ofthe secondarywinding is affected by normal (2) Typical Performanceo f a Potentiometric Pres-
changes induced in any transformer by temperature sure Transducer
changes. The largest effect is, however, due to me- (a) Linearity (independent): 1OO/ F.S. from 5% to
chanical shifts of the sensor, and circuit shifts, which 95% of span
must be compensatedover the operating range. (b) Resolution: 0.2% to 0.4% F.S. for wire-
(b) Span. Thespan output is affected by the wound elements
change in the windings due to thermal shifts in the ma- (c) Repeatability: Limited by resolution
terial selected. Thiserror i s of the same order of magni- (d) Hysteresis: 0.5% F.S., limited by mechani-
tude as the mechanical andcircuit shifts, and mustbe cal linkage of sensing element to resistive element and
compensated over the operating range. by resolution
47

ASME PTCm17.2 8 7 m d7'59b70 0 0 5 2 3 4 2 3 m
.. .
Wiper
Excitation
Output voltage
voltage
e, proportional to
o I ovoltage
excitation
FIG. 3.25 POTENTIOMETRIC DETECTOR

Wire under initial tension -
Electrical 1OOSa resistance
(e) Ambient TemperatureEffect: 3.6%/1OO0C connections
(2 %/I 00" F)
(e) StrainGage.Thesegagesarebased upon the
principle that a conductor's resistance changes with FIG. 3.26 STRAIN CACE
length. Whena conductor is stretched within its elastic
limits, and when the elongation is a small fraction of
the total length, the resistance change is quite linear or 0.05/0change in resistance.Letus now apply a
with length change and quite predictable for a given small current through the wire, say 10 milliamperes.
material. A convenient way to measure this phenome- Then, the initial voltage will be 0.01 x 100 = 1 volt.
non is a term called "gage factor." The voltage output after the pressure change will be
(100-0.05) (0.01) = 0.9995 Volt.
Gage factor k = ( 3( 3 Observe several things from this example: (1) the
output change is extremely small; (2) the output is not
zero-based; thatis, the initial value is not zero; (3) if no
pressure were applied, but,large temperature changes
where occurred, the resistance of the wire,as a general case,
AR - per unit resistance changefor unit change in
"
would change, This would give rise to a false indica-
tion of pressure, Asanexample, a 100C change in
AL length
temperature would produce a false signal of approxi-
R = element resistance
mately 760 timesthe above calculated value.
L = element length
For these and many other reasons, a different design
The symbolk is a dimensionless measure of the sensi- is chosen. A list of the changes would include the
tivity to the changeof resistance of the gage to changes
following.
in length. Typical metallic strain gages exhibit factors The element is folded on itself several times so
between 2 and5. Semiconductor types may exhibit that, in our example,the effective element length
factors as high as 100 to 200. might be 100 mm or 200 mm (4 in. or 8 in., respec-
Figure 3.26 illustrates the use of a strain gage for tively). Practical gages are, of course, much smaller.
pressure measurement. TheWheatstone bridge configuration i s used.
Assume that the wire has been sized and tensioned
Reasons for this will be discussed shortly.
so that its initial resistance i s 1O0 ohms. Also assume Higher voltage levels areused to obtain higher
that it is initially 1O mm (0.40 in.) long. Assume thata output voltages.
pressure applied to the diaphragm will move the dia- A higher initial resistance is generally used (120Q
phragm 0.001 mm (0.00004 in.) and that the wire will or 350Q are typical values).
not become slackwith this deflection. Recall thatgage Where practical, semiconductor elements exhib-
factors for metals run between 2 and 5. Thenthe
iting higher gage factors canbe used, with due consid-
change in resistance will be (assuming k = 51,
eration to thermal sensitivity.
Materials having lower thermal coefficients of
resistance are used.
ConsidertheWheatstone bridge configuration in
Fig. 3.27. Let R , = R, = R3 = R, and let V1 = -V2.
1O0 Then itfollowsthat V, = V, = O because
=5 - (0,001) = 0.05
R1, R,, R3, and R, constitute equal voltage dividers,
10
48
- -- .q"__ 2..
P
.
ASME P T C * 1 7 - 2 8 7 m 0.757b70 0052343 3 W
sates the full-scale output of the bridge for temperature

variations. An additional effect which may behandled
v3 /"" in this manner is the thermal characteristics of the

pressure-sensing element. The resistor is then chosen
to compensate the net effect of temperature upon the
pressure element and the bridge elements. Additional
compensation elements are frequently used to stan-
dardize full-scale outputs, compensate for changes in
effective modulus of elasticity of elastic-sensing ele-
ments, and a number of other factors. These schemes
are usually of a proprietary nature for a given manufac-
turer.
( I ) Methods of Application. Thestrain gage i s
FIG. 3.27 WHEATSTONEBRIDGECONFICURA- commonly applied directly to elastic diaphragms to
TlON OF THE STRAIN GAGE sense the displacement of thediaphragm. For this
purpose, specially shaped gage elements can be em-
ployed. Figure 3.28 is an example of this gage type.
and becauseV, = V,. If we also makeR , through R, of This gage is capable of measuring the tangential ten-
the same of material, the change in resistance value of sile strains developed at the center of the diaphragmas
each element will be equal with changes in tempera- well as the compressive radial strains present at the
ture. Therefore, the zero output will not change with outer diameter of the diaphragm.
temperature. Alternatively, the strain gages may be applied to a
If the elements are attached to a pressure element in flexed beam or similar member to measure deflection
such a manner thatR, and R, increase in value andR, of a spring. The elements may be individual elements
and R3 decrease in value when pressure is applied, V, or specifically designed for the application.
will approach V, and V, will approach V,. The net To isolate the element electrically and to provide
effect is an output voltage change four times greater support before use, the strain element i s attached to a
than if onlyone element were used. Typical outputval- nonconductive carrier. The materialsused rangefrom
ues are expressedas millivolts output(V, - V3) per volt paper to ceramics, with plasticor composites being the
of applied voltage (V, -V2). The applied voltage is most common. Obviously, the choice of carrier and
called the excitation voltage andusually ranges from 3 adhesiveused to secure it to the pressure element
to 15 volts. Typical values of full-scale output for metal- depends upon several factors.
lic strain gauges are 1.O -3.0 mvholt. Semiconductor Recently, two new developments havemodified the
types mayproduce up to1O0 mvholt. Variations from need for carriers and adhesive bonding. The first is
these values areof course possible. the development of diffused silicon strain gages. These
Earlier designsamplified these small signals and fre- devices incorporate the diaphragm,elements, and
quently required the use of AC excitation because AC- electrical isolation in one unit. The technology resem-
coupledamplifiersweremore gain-stable with bles that used in making transistors or integrated cir-
available techniques. One trade-off in using this cuits.
schemewas the need for careful compensation of The other development is that of vacuum deposition
phaseshiftscaused by capacitance within thestrain or sputtering. While the techniques usedto manufac-
gage and interconnecting cable. With the availability ture the gages are different, the results aresimilar. By
of modern semiconductor devices, the use of DC exci- depositing an insulator such as glass on the diaphragm
tation now predominates. and by then depositing the metallic alloy onto the insu-
The straingage, as described, is still sensitive to vari- lator, a structure is formed which provides the needed
ations in temperature when its output i s different from bonding and isolation. The element pattern is then
zero. Frequently, temperature-sensitive resistors, photographically etched onto the metallic layer.
matched to the characteristicsof the elements, are in- (0 Oscillating Devices. The needfor very high reso-
stalled in series with the excitation voltage for full-scale lution and for digital output has led tothe development
temperature compensation, of transducers in which the applied pressure varies a
The net effect is to increase or decrease the voltage suitableresonancefrequency. A simple example is
on the bridge in an inverse mannerto an increaseorde- given in Fig. 3.29. A steel wire is stretched between a
crease in the resistanceof the elements. This compen- diaphragm and a fixed reference point. The wire is
49

. .
ASME P T C m L 9 . 2 87 61 0?5j7b70..
. .
005231.11.1 5
ASME/ANSI PTC 19.2-1987 FRESSUREMEASUREMENT
/ \
I I
l I
\ /
FIG. 3.28 FULL-BRIDGE DIAPHRAGM GAGE

(Courtesy of Micro-Measurements Division, Measurements Group, Inc.,
Raleigh, North Carolina, USA)
Diaphragm
I
Vibrating steel wire
B 4
-
P
Driver Pick-up
FIG. 3.29 PRESSURE TRANSDUCER WITH VIBRATING ELEMENT
50

excited in its fundamental resonance frequency with where fT and f, are the limits of a suitable band for te-
the aid of a magnetic driver and a pick-up coil. When lemetry, Great convenience for the user results from
pressure is applied to the membrane, the tension on coupling vibrating-element transducers with hand-
the wire is reduced and the resonance frequency i s de- held calculators. The calculator can be programmed
creased. The resonance frequency i s a function of pres- with the calibration equation relating pressure and fre-
sure. By making the membrane very stiff, hysteresis can quency (or period) to calculate pressure in any one of a
be reduced below levels normally encountered in variety of pressure units.
strain gage pressure transducers. The best presentlyavailable vibrating-element pres-
Among the vibrating elements usedin commercially suretransducershavean uncertainity on the order
available transducers arewires, diaphragms, cylinders, of 0.05/0. This valuewas estimated from the standard
rim-loaded piezoelectric elements, beam-loaded pi- deviation of the residuals of a least-squares fit of a
ezoelectric elements, and ultrasonic resonators. All of polynomial in pressure and temperatureto data taken
thesetransducers produce a sinewave or pulse se- in the standard NBS transducer characterization test.
quence whose frequency (or period) is related to pres- (g) Piezoelectric Transducer.Pressuretransducers
sure. The transducers areall sensitive to temperature; of this type can measure quasi-static and dynamic pres-
careful temperature compensation, measurement, or sures from a few millipascals up tovalues greater than
control are needed. Therelationship between pressure 7.5 x 1O5 kPa (1O5 psi). Frequencies in excess of 500
p and period Tis often given by polynomials of the type kHz may be measured with high-stiffness quartz ele-
ments. Therefore, very short pressure transients or very
steep pressure rises (rise time of a few microseconds)
can be recorded. Piezoelectrictransducers canbe used
over a wide temperature range from cryogenic to over
400C (752OF).
where A piezoelectric transducer utilizes a piezoelectric
T = period of oscillation at pressure p crystal thathas the ability togenerate a charge when a
T~ = period of oscillation at reference pressure force is applied to it. Figure 3.30 diagrammatically
or, if temperature is measured separately, shows the basic construction of a piezoelectric pres-
sure transducer.
In a constructed transducer, the pressure being mea-
sured exerts aforce, usually through a diaphragm upon
the piezoelectric crystal element (usually quartz be-
cause of its stablecharacteristics),and causes the
Polynomials sometimes require more than a dozen crystal to be mechanically loaded and strained. When
terms to characterize a transducer fully. Pressure- the crystalis strained by an external force, adisplaced
period or pressure-frequency relations for vibrating- electrical charge accumulateson opposing major sur-
element transducers are rarelylinear within the faces, forming avoltage signal according to the laws of
precision of the transducing element. Some transduc- electrostatics. The diaphragm can be hermetically
ersare, therefore, available with digital linearizers sealed to the transducer body to protect the crystal
which multiplythe output signal with suitable polyno- element.
mials, If these polynomials do not contain a sufficient A limitation ofthis transducer is that it responds only
number of terms, the precision of the transducing ele- to changes in pressure. A voltage generated by an
ment is degraded. Linearizers convert the transducer applied steadypressuredecays or leaksoff,unless
output to a frequency f related to pressure by the crystali s connected to an amplifier of infinite impe-
dance.
p = af A piezoelectric transducer output signal usually re-
quires amplification before connecting to a readout
or device. This can be achieved with either a charge or
voltage amplifier. On voltage-amplifier systems, the
p = a(f - fl) amplifier can be external or built into the transducer
housing using integrated circuitry. Some ceramic trans-
with ducershave exceptionallyhigh values of internal
capacitanceandcanbe used directly into high-
impedance readout devices such as oscilloscopes.
I-
K- --
-
ASME P T C * L 7 * 2 47' m
ASMVANSI PTC 19.2-1 987 PRESSUREMEASUREMENT
Diaphragm
Pressure /- Piezoelectric element
r""
Crystal
Charge
plates
,Transducer' body
Cable L"
,
Readout Transducer
instrument
FIG. 3.30 PIEZOELECTRIC PRESSURE

TRANSDUCER
FIG, 3.31 CONVENTIONAL PIEZOELECTRIC
SYSTEM
Others which have low internal capacitance require
voltage amplifiers.Each systemwill be discussed sepa- cable mandatory and precludesuse theof such systems
rately. in moist or dirty environments unless extensive mea-
(I) Voltage Systems. Figure 3.31 shows schemati- sures are taken to seal cables and connectors. Addi-
cally a conventional piezoelectric system including tionally, frequency response i s affected by cable
transducer, cable, and readout or amplifier. The open- capacitance.
circuit voltage sensitivity V, is (measured with cable (2) ChargeSystems.The problem ofcapacitive
disconnected) attenuation is solved by the use of the chargeamplifier
since the output voltage is dependent only upon the
v, = -
4 ratio of input charge to feedback capacitor. Refer to
C1 Fig. 3.32.
However, there are limitations on the use of charge
where amplifier systems, especially in field environments or
q = basic charge sensitivity (pcblpsi) when driving long cables from transducer to amplifier.
C, = transducer internal capacitance The electrical noise at theoutput of a charge amplifier
The overall system voltage sensitivitymeasured at the is directly related to the ratio of total input capacitance
readout instrumentis the reduced value: (C,+ +
C, CJ, to feedback capacitance Cf. Because
of this, cable lengthis limited as in the voltagesystem.
(3) The ICP (Impedance Converter Piezoelectric)
v, =
4
Concept. TheICP concept involves the combining of a
c, + c, + c3 miniature IC voltage amplifier (impedance converter)
intothe same package withthepiezoelectric
where element. Figure 3.33 shows a fundamental ICP trans-
C, = cable capacitance ducer schematically.
C3 = input capacitance of readoutinstrument or An input pressure acting upon the piezoelectric ele-
amplifier ment produces a quantity of charge AQ. This charge
The dependency of system voltage sensitivity upon collects in shuntcapacitance C forming voltage V
total shunt capacitanceacross the crystal, as defined by equal to AQIC.
the abovetwo equations, severely restrictsinput cable The basic ICP amplifier shown in Fig. 3.33 is a non-
length. This explains why the voltage sensitivity of inverting DC amplifier, with a frequency response be-
high-impedance type of piezoelectric transducers is *
yond l MHz at a signal level of 5 volts. However, the
measuredand specified with a givencablecapaci- frequency response of specific transducers is limited.
tance. If the cable length and/or type is changed, the This type of transducer is particularly applicable to a
system must be recalibrated. The very high impedance measurement requiring long cables between trans-
level of the crystal makes the useof low-noise coaxial ducer and readout.
52

ASME P T C * L S * Z 87 m 8957b70 0052347 O M
. .......
.
..-
PRESSURE MEASUREMENT ASMEYANSI PTC 19.2-1987
small differential between the actuation values on ris-

ing and falling pressures. This means that a pressure
4 CCT
-
I V switch, unlike an analog transmitter, cannot operate
close to 0% of its operating range. If a pressure switch
with a 3 kPa differential were used at a setpoint of 2
kPa, it would require a - 1 kPa pressure to actuate. If
the processpressuredoes not attain this value, the
switch would always be activated and therefore be use-
L Transducer Cable Amplifier less for control or alarm.
This fact, combined with nonlinear behavior near
zero input, suggests that pressure switches not be used
FIG. 3.32 CHARGE SYSTEM below about 10% of their rated working range. At
higher pressures, elements can be subjected to stresses
which lead to element fatigue and premature failure;
manufacturers generally discourage operation above
Each system has advantages and disadvantages, de- 70% -75% of rated pressure. A trade-off then exists
pending on the application. Manufacturers' literature between accuracyand operating life (seeFig. 3.34)
should be consulted for application specifics. Piezo- which lies between 30% and 70% of working range.
electric pressure transducers have uses in a multitude If a switch i s operated between these limits and the
of applications such as: pressuremeasurements in electrical load is not severe, then some generalizations
hydraulic and pneumatic systems, pressure measure- can be made about service life (number of switch oper-
ments in infernal combustion engines, and ballistic ations). If a servicelife ofone million pressure cycles or
pressure measurements in shock tubes. switch operations or less is expected, adiaphragm,bel-
(bJ fressure Switches. Combining any of the previ- lows, or Bourdon tube element i s satisfactory. If greater
ously discussed sensing elements with an electrome- service life is expected, a piston switch should be used.
chanical switch forms a pressureswitch. Usually used However, where pressure variations are small (20% or
where a limit or alarm indication is required, pressure less of the adjustable range of the switch), a Bourdon
switches may alsobe employed when simple "on-off" tube, bellows, or diaphragm switch can be expected to
control is required. Generally speaking, they are avail- provide a usefullife of up to two to fivemillion cycles
able to sense gage, absolute, or differential pressures before metal fatigue failure of the sensing element or
with magnitudebetween 1 kPa(5 in. H20) and switch mechanism occurs,
700,OO kPa (100,000 psi). These rule-of-thumb estimatescan be drastically
Broadly speaking, they may be classified with reduced if the speed of cycling is high. If a cycle is less
displacement-type devices, with the sensing element than once every few seconds, a piston switch should
constrained by the usual considerations for elements in be considered. When rapid cycling occurs with Bour-
this class of service. These sensing elements may sup- don tube, bellows, or diaphragm elements, fatigue will
ply the return spring force or may be used only as occur in much shorter times than would normally be
pressure-summing members, with the restoring forces expected.
supplied by external springs. It is important that the For elements used above 7000 kPa (1000 psi), the
forceconsumed in actuating the electromechanical materials from which they are constructed are gener-
switch be a verysmall fraction of the total spring rate of ally workingcloser to the ultimate strength of the mate-
the system. This point is quite important because pres- rial. This will cause a serious degradation of operating
sure switches employ a snapacting switch for positive life and suggests that these elements should be used at
operation and therefore subject the pressureelement perhaps only 30% - 35% of their rated working pres-
to a sudden reduction in opposing force. If the switch sure to achievegreater life. These estimates assume the
mechanism representsa very large fraction of the sensing element to be the limiting factor in pressure
spring rate, intolerably large hysteretic errorsor appar- switch life. However, if the switch is inadequately
ent deadband will exist. That is, actuation values on rated for the electrical circuit, failure can occur prema-
rising and falling pressures will be different by unac- turely in the electromechanical switch. Also, wear in
ceptably large amounts. However, even with the this mechanism may cause a drift of setpoint. Should
switch contributing small amounts of force to the sys- such a drift be unacceptable, estimates of useful life
tem spring rate, the need for snap-action requires some must be reviseddownward.
53

Adapter and in-line

amps., this sect. only
r-""" f------------ 1
I I
I
Piezoelement
I
I - I W
AV
G
I I
I 4- I Mosfet
c I
-
R
I Ezl IC
"
T\
I I
I I
I I - A h
"
i
I I
L"""- """" ---- A
FIG. 3.33 BASIC ICP TRANSDUCER
Proof
pressure r Yield
point
Working range
Max. incr. set point
Min. decr. Max. decr.
Pressure,
Life
NOTES:
(1) For accuracy and life, select Zone A.
(2)For life, select Zone C.
FIG. 3.34 OPERATING RANGESFORPRESSURESWITCHES

(Courtesy of IMO Delaval, Barksdale Controls Division)
54

. .
-. .
It is difficult to predict a derating factor; user experi- nate parallax,andadequate illumination. Suchan
ence and/or manufacturer's datashould be examined instrument could therefore be used to determine pres-
for these factors. sure difference of 130 Pa (1 mm Hg) to 1O/O precision,
and 1.3 kPa ( I O mm Hg) to 0.1% precision. Instru-
ments of this grade areavailable commercially as well-
type manometers or barometers with a span of 1 O0 kPa
3.5 LOW ABSOLUTE PRESSURE (VACUUM) (30 in. Hg) or on special order up to 340 kPa (100 in.
SENSORS Hg) span. This is, of course, not a convenient test in-
strument but i s useful for calibration purposes. If the
3.5.1 Units and Terminology, Historically, two mean- reference leg of the manometer is carefully filled, the
ings of the term vacuum have evolved. Both meanings reference pressure will be the vapor pressureof mer-
refer to absolute pressures below normal atmospheric cury of about 0.5 Pa (0.004 mm Hg) at atmospheric
pressure, butdiffer in their reference points.For exam- temperature, which is less than the reading limit of the
ple, when an auto mechanic describes"20-incha vac-
gage.
uum," he is discussing a negative gage pressure
equivalent to 20 in. Hg. A vacuumtechnologist speaks
3.5,3.2 Butyl-Phthalate Manometer. Water is not
of a "hard" vacuum of I Torr. Here, the technolo-
useful as a manometric liquid for low absolute pressure
gist means an extremely low absolute pressure. To add
because of its high vapor pressure. However, butyl
to the confusion, note that different units areused.
phthalate, as used in the Hickman vacuum gage shown
Eacharea of usagehas its own set of "customary"
in Fig. 3.35, is a liquid withvapor pressure much less
terms to quantify vacuum measurements. Table 3.3
than that of mercury and density of the same order as
lists the more common units, conversion factors, and
water. With a manometer arrangement of similar preci-
area of usage. Note particularly the termTorr (= 1 mm
sion, butyl phthalate can measure absolute pressures
Hg). This unit is in very common use, but such use is
about one order of magnitude lower than mercury.
being discouraged. Ultimately the pascal should dis-
It has the disadvantages that its temperature coeffi-
place all other units and its use is being encouraged.
cient of expansion is high and many gases and liquids
are highly soluble in it, so that the low-pressure refer-
3,5.2 Technology. The choice of measuring devices ence side of the manometer must be continuously
becomes progressively more restricted as the absolute pumped, This also means that the liquid is easily con-
pressure level decreases. Indicating gages may mea- taminated, with a consequent changein density.
sure down toabout 10 kPa (3 in. Hg) absolute, By care-
3.5.3.3 Diaphragm Comparator. A special modifi-
ful and innovative design, other direct-measuring
cation of the diaphragmpressure gageis commercially
devices may be ableto measure down to 0.1 Pa (0.75
available for the measurementof very low pressure dif-
micron). To measure pressureslower than this limit, in-
ferentials, with a sensitivity of about 0.1 Pa (1 micron)
ferential measurements are available which can be re-
(see Fig. 3.36). The referencepressure, usually a high
lated to pressure for a known gas or mixture of gases.
vacuum, is applied to one sideof a diaphragm and the
Devices used to measure vacuum referenced to at-
unknown higher pressure to the other side. The dia-
mosphere (sometimes called "suction vacuum") are
phragm forms one plate of an electrical capacitor. An
similar to those discussed earlier in this Section and
adjustable direct-current voltage i s applied to bringthe
will not be discussed here. Instead, only devices in-
diaphragm back to its original position by electrostatic
tended for low absolute pressures will becovered.
attraction, The balancepoint is indicated by a capaci-
tance bridge circuit. The value of the balancing DC
3.5.3 DirectMeasuring Devices. The gages which voltage i s read from a potentiometer, and is the mea-
measure pressure directly include mercury microma- sure of pressure difference. The spanof the instrument
nometer, butyl-phthalate manometer, diaphragm com- is 20 Pa (150 microns). The reference base may be a
parator, and McLeod gage. high vacuum or atmospheric pressure. Theinstrument
is a true pressure gage,but it is affected by the dielectric
3.5.3.1 Mercury Micromanometer.If a mercury ma-
constant of the gasand, of course, by temperature.
nometer i s refined for the best possible precision, it is
These effects become more important at higher pres-
possible to read levels accurately to about 1 Pa (0.01
sure levels.
mm Hg). This requires tubes atofleast16 mm(0.63 in.)
to minimize capillary effects, precision-scale engrav- 3.5.3.4 McLeod Gage. This gage comprises means
ing, vernier reading, sighting edges arrangedto elimi- for compressing a volume of the rarefied gas into a
55

ASMVANSI PTC 19.2-1987 PRESSUREMEASUREMENT
Air-cooled
acetone
condenser
Acetone boiler
To mechanical phthalate condenser
manometer
FIG. 3.35 HICKMANVACUUM CACE
56

MEASUREMENT PRESSURE PTC 19.2-1 987
7
Capacitance
brldge
Variable r"l
H
DC source
Pressure
c
3
Diaphragm J
L Reference pressure
port
FIG. 3.36 DIAPHRAGM PRESSURE COMPARATOR
57

TABLE 3.3
VACUUM MEASUREMENT UNITS
Where Conversion Factors [Note (211

Used
Units [Note (111 psi kPa in. H,O in. Hg mm Hg Ear
in. H20(4C) GP0.249 0.0361 1 0.0736 1.87 0.06249

(39.2F)
GP/LA (32F)
in. Hg (0C) 3.39 0.491 13.6 1 0.00339 25.4
psi GP/LA 1 6.89 27.7 2.04 51.7 0.0689
mm (of Hg) (OOC) LA 0.0193 0.133 0.535 0.0394 1 0.00133

(32F)
micron of Hg (P) LA 1.93 x 1.33 X 10 5.35 X 3.94 X 1.33

10~ 10 x
Torr (sec mm of Hg) LA 0.0193 0.133 0.535 0.0394 1 0.001 33
Pascal LA 1.45 X 10 10 4.01 X 10 2.95 X 10 7.50 X 10
Millibar LA 0.0145 0.1 O0 0.401 0.0295 0.756 I
Bar GP/LA 14.5 1O0 40.1 29.5 750 1
NOTES:
(1) GP = measurement referenced to atmospheric pressure; LA = low absolute.
(2) Rounded to three places.
much smaller volume. From the dimensions of the ap- Ah

paratus, and a reading of a substantial mercury-level P, =
difference, the pressureof the original sample in terms V1 - hA
of a heightof a mercurycolumn .is calculated. The ar-
rangement is shown in Fig. 3.37. The compression is which reduces to, approximately
essentially isothermal because of the time involved
and the large surface-to-volume ratio. Themeasure- Ah2
ment starts with mercury drainedout ofthe instrument P1 = -
V1
and the gage filled with the gas to be measured. The
mercury is raised by any of a numberof possible meth-
ods cutting off thevolume V1 in the measuringbulb. As since hA < < V1.
the mercury continues to rise, this gas is compressed Alternatively, the gage maybe arranged to compress
into the measuring capillary extension, until the level the gas only to a fixed volume V, identified by a refer-
in the exactly similar reference capillary reaches azero ence zero mark at the base of the measuring capillary.
point corresponding to zero volume in the measuring Then the mercury will stand higher in the reference
capillary. The mercury levelin the measuring capillary capillary by the heighth. The original pressure is then:
will be lower because of the trappedgas. Thelevel ref-
erence h is related to the original pressure Pl (both in
linear units) in the following way: P1 = P2 ();
P,V1 = P2V2for isothermal compression
P2 = P1 +h
V, = hA
Pl = h (L)
Pz = P1 +h v1 - v2
58

PRESSURE MEASUREMENT ASMUANSI PTC 19.2-1 987
-
Since [V2/(V, V2)]is a constantof the instrument,P, For a given gage, the readingof the microammeter i s
is a direct linear function of h. constant for a constant heater input and constantpres-
Combining these two methods provides a double sure. The reading depends upon the gas composition.
range in one instrument for high and low pressures. At pressures higher than 30 Pa (225 microns), the mi-
If condensible components are present in the origi- croammeter reading is very low and may correspond
nal sample, they will be partially condensed by the to about 10% of full-scale value. The reason for this ef-
compression and will not contribute to the final gas fect is that the thermal conductivity through the gas is
volume, The McLeod gage measures essentially only high and essentially independent of pressure above
the fixed gases in the original sample (see para. 2.1.31, 130 Pa (1 mm Hg). However, as thepressure i s re-
The range of the usualcommercial forms ofMcLeod duced below 130Pa, the gas conductivity begins to de-
gage covers from 0.001 Pa to 7 kPa (0.01 to 50,000 crease with pressure down to about l .3 Pa (10
microns), not,of course, in the sama gage.Atthe lower microns), Sincethe thermal conductance of the gas de-
end of this range, it is necessary to provide a cold trap creases for decreasing pressures, the temperature of
between the gage and the system to prevent contami- the heating element (and thus thethermocouple junc-
nation of the system by the mercury vapor from the tion as well) increases.
EmF. This increase in temperature of the thermocouple
junction with decreasingpressureresults in an in-
crease in the voltage output ofthe thermocouple, Thus,
3.5,4 Inferential Measuring Devices. When the pres- the deflection of the microammeter is greatest for the
sure to be measured falls below that covered by the lower pressures. In some thermocouple gages, the mi-
previous devices, it is necessary to use detectors that
croammeter reading is about 80% of the full scale at
respondto a pressure-relatedproperty. Two such prop-
1.3 Pa (10 microns).
erties are thermal conductivity and ionization. As pressure is reduced below 1.3 Pa, the tempera-
Thermal-conductivity devices rely upon the fact that, ture change at thethermocouple junction is compara-
for several decadesof pressure i n the region of interest, tivelysmall. Thus themicroammeterreading
the heat loss from a thin wireis nearly linear with pres- approachesanasymptote for decreasing pressures.
sure. Thermocouple and Pirani gages are two devices This asymptoteis due to two major factors: thermal ra-
using this phenomenon.
diation and thermal conduction through the support-
The principal advantages of the thermocouple and
ing leads of the heater andthermocouple elements. At
Pirani vacuumgages are their simplicity and low cost, pressures below 1.3 Pa, the thermal radiation and heat
Improvements in their performance are being con- conduction through the leads are essentially constant
stantly made. Their principal disadvantagesarethe and are considerably greater in magnitude than theef-
shift in calibration caused by contaminating vapors fect of thermal conduction through the gas. For these
from the vacuum system and slow response. The shift
reasons,pressuremeasurementsless than 1 Pa (7.7
in calibration i s more severe nearthe low-pressureend microns) are not attempted with the thermocouple
of the scale, This is caused primarily by the change in
gage.
emissivities of the heating element, thermocouple
junctions, and surrounding walls of the container. Re-
3.5.4.2 Pirani Cage. The Pirani gage i s similar in op-
sponse of the thermal-conductivity gages i s relatively
eration to the thermocouple gage. The same factors
slow, because of thermal inertia. These gages must be
that limit the performanceof the thermocouple gage at
calibrated for the gas mixture to be encountered.
pressures above 130 Pa (1 mm Hg) and at pressuresbe-
3.5.4.1 Thermocouple Gage, In the usual form of low l Pa (1 O-2 mm Hg) also limit the measurablepres-
construction of a thermocouple gage, a shortlength of surerange of the Pirani gage. In the Pirani gage,
resistancewire is heated to perhaps 200C. At the mid- however, only a heating element is usedand the
point of this heaterwire,a thermocouple is spot- change in resistance of this element is measured as a
welded. A sensitivemicroammeter (of theorder of 200 function of pressure.Theusual detecting-circuit ar-
microamperes) and low internal resistance (of the or- rangement for a Pirani gage is to use the heating ele-
der of 50 ohms) is used to measure the current pro- ment in one arm of an electrical bridge network (see
duced by thevoltage at the thermocouple. The Fig. 3.39). To compensate for ambient effects, includ-
assembly of the thermocouple and heater element is ing supply-voltagevariations,another Pirani element is
usually mounted in a metal or glass envelope, as enclosed in a sealed and evacuated chamber andused
shown in Fig. 3.38. Ashortconnection of tubing i s pro- as the balancing element in the bridge circuit. The
vided for connection to the vacuum system. power is supplied to twoopposite corners of the bridge
59

ASME P T C * L 7 - 2 8 7 m
t To vacuum
space
II
ry
acer
Reservoir
FIG. 3.37 McLEODCAGE

Heater
Thermocouple Support
junction
5
1
7
3
Prongs 1 and 5 - heater inputs

Prongs 3 and 7 -thermocouple
FIG. 3.38 THERMOCOUPLECACE
60

PRESSURE MEASUREMENT ASME/ANSI PTC 19.2-1987
YSealed-off Pirani element

To vacuum system
Pirani
element
I balance adjustment
J
1 1 1 1
Voltage source t o supply
current t o Pirani elements
FIG. 3.39 PlRANlVACUUM GAGE
61

.i . ,
and an indicator, typically a DC microammeter, is con- A pressure and atmosphere limitation exists for the
nected to the remainingcorners. Initial bridge zero bal- hot filament gage. Too high pressure or an atmosphere
ance is obtained at anabsolute pressure no greater than excessively rich in O, H 2 0 , or CO, or other gases
0.01 Pa ( l g 4mm Hg). As the pressure increases from which can react with the hot filament will destroy the
about 1 Pa to about 100 Pa, resistance of the sensing filament. Generally these conditions are avoided for
Pirani element decreases. This unbalances the bridge pressures below 0.1 to 0.01 Pa (10" tomm Hg).
and causes indication on the microammeter corres- Frequently, emission or collector current is automati-
ponding to pressure. cally monitored for evidence of excessive pressure and
provision is made toautomatically shut off the filament
3.5.4.3 Ionization Gages. Ionization gages measure when excess pressure exists.
the frequency ofcollection and dischargeof ions at an As indicated before, the upper pressure rangefor the
electrode. They include a means for producing ions hot-filamentgage is generally 0.1 Pa (1O" mm Hg),The
and a means for collecting them. Associated instru- lower limit is influenced by design of the gage, but gen-
mentation is then usedto measure the ion current. This erally corresponds to about 1O 5 to 1O-6 Pa (1 to 1o-'
current is, for constant conditions, proportional to gas mm Hg). One order-of-magnitude decrease below this
density, which is in turn related by the ideal gas law to level is possible through careful design and selection of
gas pressure. Several types of gages exist. Their func- materials.
tion is similar; only the details ofoperation differ. (b) Phillips-PenningGage. This isa type of commer-
(a) Bayard-AlpertGage. The hot-filament or Bayard- cially available ionization gage. Ionization of the gas in
Alpert fypegenerates ions by collision of energetic this gage is caused by the electrons and ions createdin
electrons with the gas molecules. Thermionic emis- a glow discharge. In order to achieve even greatereffi-
sion, as employed in an electron vacuum tube, i s used. ciency of ionization of the gas molecules, the electrons
Bias of the individual elements within the gage deter- created in the glow discharge are constrainedto move
mine proper operation.Refer to Fig. 3.40. The filament in helical paths by the proper application of electric
is heated by voltagesupplied through R,. R, is adjusted and magnetic fields (see Fig. 3.41). The amount of ion-
until current I, through the grid circuit is equal to a ization produced in a given gas by this method i s a
value dependent upon physical dimensions of the function of the numberof molecules perunit volume.
gage. Thevoltage between the grid and filament acts to The ions thus formed are collected at the cathode.
acceleratetheelectrons toward the grid. Collisions An electroniccurrent flow is thereby set up in theexter-
with gas molecules in this area produce positively na1 circuit. A microammeter is used to measure this
charged ions. They are,in turn, attracted to the collec- current flow. For pressures below 0.1 Pa (1 micron),
tor, M, measures the ion current I, and is calibrated in the microammeter reading is closely proportional to
pressure units.If the voltage betweenfilament and col- pressure. However, at pressures above 0.1 Pa (1 mic-
lector is not set high enough, electrons which escape ron) andup to about 60 Pa (5 mm Hg), the relation be-
the grid would impinge upon the collector, subtracting tween the microammeter reading and pressure departs
from the ion current in an unknown manner. widely from linearity. Thepresent commercial
Due to exposure to dirty atmospheres or other fac- Phillips-Penningtype ionization gages available do not
tors, it is possible during operation or storage for the read pressures of air much above 60 Pa (0.5 mm Hg).
gage to become contaminated, and therefore require The Phillips-Penning type of ionization gage is not
cleaning. To ensure proper operation, the other volt- too costly or complicated and thereis no danger of de-
ages are turned off and S, is closed. Anelectric current struction if the gage is accidentally exposed to atmo-
i s passed through thegrid heatingthe entire gage, caus- spheric pressure. Its principal disadvantages are, first,
ing accelerated outgassingofthe gage, thereby in effect its sensitivity to pressure changes above 13 Pa (100
cleaning it. microns) is low, and second, the glow-discharge phe-
During normal operation, thegage is heated by the nomena involved in its operation are dependent on the
filament and somematerial is also evaporatedfrom the condition of the anode and cathode surfaces. This lat-
filament, This combination of outgassing andpumping ter effectresults in errors in calibration when contami-
(gettering)at elevated temperatures can cause the indi- nants cover the cathode and anode surfaces, However,
cated pressureto be in error if the gage is not coupled for pressure readings below 13 Pa, thePhillips-
closely to the gasvolume whose pressureis of interest. Penning ionization gage performs quite satisfactorily.
The reading obtainedwill be valid for the pressure in-
side the gage, but will be inaccurate for the vacuum
system if this caution is not observed.
(c) Alphatron Gage. There is one commercial form
of ionization gage available thatwill measure gas pres-
sures from 0.01 Pa (O.? micron) up to atmospheric
e-
PRESSURE MEASUREMENT ASMVANSI PTC 19.2-1987.
Vacuum
FIG, 3.40 BAYARD-ALPERT lONlZATlONCACE
From system 7
Microammeter
"
2000 V D
FIG. 3.41 PHILLIPS-PENNING CAGE

63

. .. .
ASME/ANSI PTC 19.2-1 987 PRESSURE MEASUREMENT
pressure.This is the alphatron vacuum gage which Pa would be very smallbecause of the verylow rate of
uses a small quantityof radium as an alpha source (see momentum transfer when the mean free path is less
Fig. 3.42). The alpha particlesemitted from this source than the separation between the surfaces.
ionize the gas molecules. The positive ions thus pro- Since this vacuumgage depends for its operation on
duced are accelerated by an electric field to a nega- the transfer ofmomentum of the gas molecules, its de-
tively charged collector probe. The accumulated flection will be a function of the molecular weight of
positive charge on this probecauses anelectronic cur- the gas as well as the temperature, Thegage is custom-
rent flow which i s measured by an electrometer ampli- arilyfurnishedto bedirect-readingfordryair. Forgases
fier, The output of this amplifier-operates a microam- heavier than air, the deflection will be greater for a
meter or a strip chart recorder. given pressure. The reverse is true for gases lighter than
Six pressurescalesare available on the alphatron air. Correction factors are available for some of the
gage. The lowest full-scalerange is 13 Pa (10 microns). more common gases. The gage is also available with a
The otherfive scales increase in factors of 1O up tothe linear scale of arbitrary unitsso that the user maycali-
highest full-scale pressure reading of 130 kPa (1000 brate thegage more conveniently when it i s to be used
mm Hg). The output indications of this gage are quite for measurement of other gases. In addition to being
linear as a function of pressure over this entire range. dependent for its calibration on the gas mass and tem-
This gage iscalibrated to read pressurecorrectly for dry perature, it i s also dependent on the power-line fre-
air at normal room temperatures. As is the case with all quency. Thisi s because a small synchronous motor is
ionization types of pressure-readinggages, corrections used to drive the driver surface at a constant speed.By
must be made if the gas composition is different from referring to the equation in Fig. 3.43, it can readily be
that of dry air, or the temperature of the gases being seen that the molecular momentum transfer between
measured i s different from that for which the gage is the surfaces i s a function of the speed of the driver
calibrated. For gasesother than air,the scale factors are surface.
provided for making the necessary conversion. Since For some installations, it may not be convenient to
the alphatron is linear over mostof its pressure range use a gage of this type because the dial indicator has
for gases heavier than air, and linear over the entire to be located in close proximity to the system under
range for air andgases lighter than air, theapplication measurement.
of these correction factors is simple. Figure 3.43 i s a sketch of theearlyLangmuir-
Although great care has been taken to make the al- Dushman molecular vacuum gage. It i s limited to
phatron gage as free as possiblefrom the effectsof con- about 130 Pa (1000 microns) maximum pressure.
tamination in the vacuumsystem, it is important to take However,recentimprovementshaveextendedthe
reasonable precautions to keepthevacuumsystem pressure range up to 2.6 kPa (20 mm Hg) by adding
from depositing vapors on the radium source and the vanes on the moving surface to produce windage
probe insulators. If a gage becomes contaminated by effects.
vapors from the vacuum system, a simple cleaning
with a solvent and a few minutes drying time will re-
store the original calibration of the gage. 3.5.5 Application Considerations. Even if the preci-
(cf) Molecular Gage. Another very useful vacuum sion of a vacuum gage is high, the readings obtained
pressuregage is the molecular vacuum gage. One will be in error if certain precautions and corrections
model of this gage is calibrated to read pressuresfrom are not made. If a leak exists at the vacuum-gage con-
0.26 Pa (2 microns)up to26 kPa (20 mm Hg). Its opera- nection to the vacuum system, a pressure drop could
tion depends on the transferof molecular momentum easily result in the direction of molecular flow in the
transmittedfrom a moving surface to another surface in vacuum system under measurement. If the molecular
close proximity (see Fig. 3.43).At pressures below 130 conductance between the vacuum gage and the point
Pa (1O00 microns), the angular deflection of thedial in- at which the pressure measurement desired is high,
dicator is almost linearly proportional to pressure. This then it is quite likely that a correction in the reading
i s because the mean free path of a molecule at pres- may not be necessary. However, if the molecular con-
sures below 130 Pa is larger than the distance between ductance of the pipe, or tubing, connecting the vac-
the two surfaces. In order to extend the range of the uum gage to thevacuum system is verylow,then
gage above130 Pa up to 2.6 kPa(20 mm Hg) of air, the serious errorscould beobtained. For these reasons,it is
designers have included vanes on one of thesurfaces alwaysgood practice to place the vacuum gageas
to produce windage effects. If it were not for these close as possible to the point in the vacuum system
vanes, the response of the gage to pressures above130 where the pressure information is desired. In this
64

O a
Radiur
FIG. 3.42 IONIZATION CHAMBER OF ALPHATRON GAGE
regard, due consideration must be given to the pos- constants of the vacuumsystem itself are considerably
sibility of contaminants such as oil vapors from larger.
a back-streaming vacuum pumps. These contaminants

could result in large errors in the gage readings. In
many cases a simple, right-angle elbow-pipe connec-
In general, themeasurement problem should be
carefully considered before a vacuum gage is selected
for any particular application. Careful consideration
tion from the gage to the vacuum system helps consid- must be given to the particular kind of gas variable
erably in reducing gage contamination. which i s being measured. In many cases, pressure is
Another point that is often overlooked when using the most important quantity. In other cases, the gas
vacuum gages in systems occurs when there is a large density is a much more important factor thanthe pres-
difference in temperature between the vacuum gage sure. From economic considerations it may be found
and the point in the system where pressure information that thethermal-conductivity vacuum gages have more
is required. As mentioned earlier, this can be a subtle than adequate accuracy and speed of response to sat-
source of error in the hot-filament ionization gage. In isfy the measurement requirement. Where high accu-
the case of high-vacuum furnaces, temperatures may racy i s required, it may benecessary to use someof the
be elevated by several hundred degrees centigrade. more expensive vacuum gages. However, expensive
Elementary considerations of the gas laws clearly vacuum gages do notnecessarily meanmore accurate
indicate the correction factors involved. measurements if the gage is improperly applied and
In addition to accuracy andthe application of correct necessary correction factors are not made.
scale factorsin the use of the vacuum gages, it is impor-
tant to knowsomething about the speed of response of
the vacuum gage to sudden changes in pressure. In
3.6 ELASTIC CACES
the case of the thermal-conductivity gages, the time
constants involved are of the order of seconds. Most Positive and negative (vacuum) pressuregages,as
composition-dependent gages are considerably more ordinarily used, are instruments for measuring the
rapid in responding to a pressure change, and their re- difference between ambient pressure (atmospheric
sponse is usually limited by the recording device used pressure) and the pressurein a pipe orvessel. The pres-
to measure the output signal of the vacuum gage. In sure to bemeasured is transmitted to the interior of an
a many applications, however, it is usually found that

the speed of response of even the slow thermal-
conductivity gages is entirely adequate since thetime
elastic element (Bourdon tube, bellows, or diaphragm
capsule) and the resultant motion translated into
pointer motion. Pointer motion over a calibrated dial
65

. ..
ASMUANSI PTC 19.2-1 987 PRESSUREMEASUREMENT
Mirror fordeflecting light

beam used far measurement
of angular deflection of
Connection t o
vacuum system
Motor coilwhen energized

produces rotating magnetic
field to turnsurface a at
constant speed
B = K u P d /M
_r
where
B = rate of momentum transfer per unit area
between the rotatingsurface a and the
suspended surface b
K = constant for a given gas
u = angular velocity of the surface a
P = absolute pressure
M = molecularweight
T = Kelvintemperature
FIG. 3.43 LANGMUIR-DUSHMAN MOLECULAR GAGE
provides an indication of the applied pressure. Ambi- 3.6.1 Classification by Sensing Element Type
ent pressure exterior to the elastic chamber is usually 3.6.1.1 Bourdon Tube Gage. The Bourdon tube
atmospheric pressure. ?he pressure relationships are gage illustrated in Fig. 3.44 involves a curved elastic
shown in Fig. 1.1. tube, closed at one end, The tendency of the tube to
The output motion of the elastic element under pres- straighten out when pressure is applied to the interior
sure usuallyrequires a mechanism to amplify and causes motion of the closedend, This motion is ampl-
translate thismotion into easily detectablecircular ro- fied and transmittedto the pointer. In indicating gages,
tation of a pointer. Thenecessity for mechanical multi- the usual Bourdon tube is curved through an arcof 200
plication and translation devices, plus the sensitivity of deg. to 300 deg, In other types, the Bourdon tube may
the elastic element itself, requiresperiodic evaluation be in the form of a spiral or helix having a number of
of gage accuracy.Accuracy i s determined by compari- complete turns, as illustrated in Fig. 3.45. Combina-
son of gage-pressure readings with transfer standards tions of these forms may be used with or without a
such as manometers, deadweight-loaded piston gages, motion-multiplicatlon mechanism. Bourdontube
pressure transducers, and appropriate test gages. gages are made forbothpositive andnegative
66

- .
. .
. .
(vacuum) pressure, compound,anddifferential- tition separating the elastic chamberfrom the dial and
pressure measurement. some may have features which permit the enclosure to
Differential-pressure gages may be constructed u5- be filled with liquid. Filling the enclosure with liquid
ing twoelastic elementswith the mechanism arranged can reducevibration effects and reduce pointer oscilla-
to subtract the motion of one from the motion of the tion as well as protectthe internals from external corro-
other. An alternate construction may be made with a siveatmospheres.Gageenclosures should always
pressure-tight case so that pressure may be applied provide a means of venting to reduce thepossibility of
both inside and outsidethe elastic element. This
type i s internal case pressure buildup should a leak occur in
limited to low pressures by the strength considerations the elastic element.
forthe enclosure, the clarity of the fluid media
3.6.2.2 Recording Gage. Recording gages are usu-
emp[oyed, and its compatibility with the internal
ally used only for secondary and incidental pressure
mechanism.
measurements and not for the precise measurements
3.6.1.2 Bellows Gage. The bellows gage, illustrated normally required by ASME Performance Test Codes.
in Fig.3.46, utilizes an elastic element formed with However, if they meetthe accuracy requirements, they
two or more welded diaphragms or a formed bellows. may be used to measure important pressures. Record-
This construction is used for low pressure gages, up to ing gages come in two general forms, the round
abaut 350 kPa (50 psi). chartandthe miniature strip chartinstruments,
Both types provide a scale length of 75 - 1 O0 mm
3.6.1.3 Diaphragm Gage, A slack diaphragm gage,
(3 - 5 in.) andthechoice normally depends on the type
illustrated in Fig. 3.47, utilizes a flexible diaphragm,
of record which i s preferred and theamount of panel
The motion of the diaphragmi s transmitted andampli-
space available.
fied by a suitable linkage and gears to operate a
pointer. The diaphragm gage i s suitable for very low
pressures.
Metallic diaphragm gages utilize an elastic element
made up of a series of capsules. Each capsule com- 3.6.3 Commercial Gages. Commercial gages .having
prises two corrugated diaphragms joined together by Bourdon tube, bellows,anddiaphragmelasticele-
various means, such as brazing, soldering, or welding. ments vary in construction and accuracy (error) limits,
This type of gage is available in ranges of O - 25 mm Hg dependingonintended use.Gages are normally
(0- 1 in, Hg) up to 1400 kPa (200 psi)and is well graded by the accuracy limits of their factory calibra-
adapted to the measurement of pressures lower than tion. Accuracy limits (permissible error) are defined in
the practical limit of the Bourdon tube. ANSVASME B40,l and are expressed in terms of the
pressure span over a nominal 270 deg. arc of pointer
motion.
Grade 4A: * O . 10% of span for the total span
3.6.2 Classification by Display Format Grade 3A: -t0.25% of span for the total span
3.6.21 indicating Gage. Indicating gages (analog Grade 2A: +0.50% of span for the total span
type) as represented in Figs. 3.44,3.46, and 3.47 givea Grade 1A: + 1.O% of span for the total span
direct visual presentation of pressure, For constant or Grade A: k 1.O% of span for the middle 50% of
slowly varying pressure applications, thistype of gage span and 2.0% for the balance
presentation is usually preferred. Indicating gages are Grade B: *2.0% of span for the middle 50% of
provided in numerous sizes and configurations. The span and 3.0% for the balance
general construction consists of an elastic element, an Grade C: +3.0/o of span for the middle 50% of
operating mechanism, a dial and indicating pointer, a span and 4.0% for the balance
transparent window, and an applicable enclosure or Grades3Aand 4A are usually considered where
case. The details of construction are too numerous to accurate pressure measurements are required for eval-
discuss in this document; recommendedstandard con- uating critical processesand for testpanelsused to
struction and other details may be found in American evaluate plant equipment. Thesegradesare recom-
National Standards Institute documents ANWASME mended for precise measurementbut are generally not
B40.1. employed for universal and continuous industrial ap-
Gages are available for panel mounting, surface plications because they may incorporate pressure-
mounting, and mounting by means of the pressure sensing elements having appreciably shorter life (see
connection (stem). Enclosuresor cases may have a par- para. 3.6.9.1).
67

Oval
FIG. 3.44 BOURDONCACE
Motion
Motion
C-Shaped
t Pressure
Helical Bourdon Spiral Bourdon

Bourdon Tu.be Tube Tube
FIG. 3.45 BOURDON

TUBES
68

I Pinion
and sector
gage movement
Connecting link
Spring
:> +ICase
connections
FIG. 3.46 BELLOWS CACE
i
Scale
Spring
lexible
Pressure
connections
L Slack diaphragm
FIG. 3.47 SLACK DIAPHRAGM CAGE
69

Sinceaccuracyand readability arecloselyassoci- gage is continuously operateddoes not exceed 75% of

ated, gage size is restricted by the grade classification. the full-scale pressure. The range.selected should have
Minimum gagesizes recommendedforthevarious a full-scalepressure of approximately twice the in-
grades are: tended operating pressure.It is desirable that all read-
Grade 4A: 8/2 in. nominal size ings be between one-quarter and three-quarters of the
Grade 3A: 4% in. nominal sire pressure span. Test gages (Grade 3A and 4A) may in- be
Grade 2A: 2/2 in, nominal size termittently used for readings outside theselimits pro-
Grade 1A: 1/2 in. nominal size vided the gage manufacturer recommends such usage.
Grade A: 1/2 in, nominal size Special consideration must be given when gages may
Grade B: 1/2 in. nominal size be installed in applications where pressure pulsation,
To reduce parallax reading errors, Grade 3A and 4A vibration, corrosive atmosphere, or corrosive and/or
gages arenormally provided with mirror sectors on the oxidizing pressure media exist. Safety as applicable to
dial and a special pointer. indicating gages is covered in ANSVASME B40.1, and
gage users should become familiarwith this Standard.
3.6.4 Pressure Ranges. Pressure gages are made in a 3.6.4.2 Reading. Gages are normally furnished hav-
wide variety of ranges. Preferred ranges for pressure ing dial numbersand division spacing in the form:
gages are: 1 X lon, 2 x lon, and 5 x I O n times the unit of pres-
Preferred SI Ranges (kPa) sure measurement (nis a whole number, positive, neg-
0/1 0/10 01100 0/1000 0/10 O00 0/100 000
ative, or equal to zero), It is desirable that graduation
0/1.6 0/16 0/160 Oh600 0/16 O00 0/160 O00 spacing be equivalent to the numerical error (accuracy)
Positive 0/2.5 0/25 0/250 0/2500 0/25 O00 0/250 O00 rating of the gage andshould not exceed two times the
0/4 0/40 0/400 0/4000 0/40 O00 0/400 O00 error (accuracy) permitted in the middle half of the
0/6 0/60 0/600 0/6000 0/60 O00 0/600 O00
scale. Thus, pressure input tothe elastic element may
be readily estimated when the pointer is between
graduations.
Graduation and pointer-tip widths are normally fur-
nished as narrow as practical and commensuratewith
theaccuracygrade.Thechange in position of the
Receiver
20/1 O0 pointer which can be detected hasminimum a value of
Cage
about 0.25 mm (0.01 in.) but for most pressure gages
Preferred U.S. Customary Ranges this value is nearer to 0.75 mm (0.03 in,). Discretion
0/10 in. H 2 0 0/3 psi 0/200 psi 0/10 O00 psi
must be used in deciding to what degree the pressure
0/15 in, H,O 0/5 psi 0/300 psi 0/15 O00 psi will be recorded. The eye should be directly in line
0/30 in. H 2 0 0/10 psi 0/600 psi 0/30 O00 psi with pointer and perpendicular to the gage faceto min-
0/60 in. H20 0/15 psi 0/1000 psi 0/60 O00 psi
(in Oh00 in. H 2 0 0/30 psi Oh500 psi 0/100 O00 psi imize parallax error. Thegage should belightly tapped
and 0/200 in. H 2 0 0/60 psi 0/3000 psi with the fingers at each reading to minimize friction
0/300 in. H 2 0 0/100 psi 0/6000 psi error.
Negative Special caution must be exercised when readings are
Pressure taken near zero. Stop pins insertedin the dial or inter-
(vacuum,
- 30/0
in. Hg) nal minimum stops mounted on the sensing element or
mechanism may prevent free pointer motion at zero
30-0-1 5 30-0-1O0 pressure. This feature element, called take up, is not
30-0-30
30-0-1 50
(in. Hg 30-0-60 recommended on Grade 3A and 4A gages but may be
and psig) incorporated on other grades. The take-up feature, if in-
Receiver 3/1 corporated, is usually equal to the numerical value of
Gage (psig) the accuracy; for example, a Grade B gage having an
Ranges other than these ranges maybe available. Dia- accuracy of f 3.0% in the first quarter of the span, take-
phragm gages are normally used for ranges less than up may be 3.0% of the span and the stop pin orinternal
1O 0 kPa (15 psi). stop would be setto stop the pointeror mechanism at a
pressure equivalent to 3.0% of the span above true
3.6.4.1 Range Selection. The rangeof a gage should zero. Additional details of this feature are outlined in
be selectedso that themaximum pressure atwhich the ANSVASME B40.1.
70

3.6.4.3 Installation, Gages should be installed in the 3.6.4.4 Connections. Pressure gageconnections are
same position and the same orientation as when cali- generally 1/8-27, 1/4-18 or 1/2-14 American Standard
brated. Thenormal position is with the dial in a vertical external taper pipe threads (NPT) depending on the
plane and mid-scale at the 12 o'clock position. If gages size andmethod of mounting. NPT threads are used for
are to be mounted in other than normal positions, it pressures through 160,000 kPa (20,000 psi), Above
may be necessary to recalibrate the gage in the in- this pressure, 9/16-18 female super-pressure compres-
tended mounting position. When gagesare installed in siontubefittings or equal are employed.Other
a system where a liquid head exists in the connecting appropriately-sized connections, employing sealing
line, it may be necessary to compensate for this static means other than tapered pipe threads,are coming
head. The compensation may be negative or positive into use for specific applications where thread-sealing
depending on the location of the gage above or below gage replacement or gage orientation may be critical.
the pressure tapin the system. Gagesshould be located
where the temperature is as near as practical to 23C
( 7 3 O F ) andwherethere is no continuous vibration
which may cause premature mechanismfailure. 3,6.5 Gage Attachments. When added to a pressure
Temperature variationsin excess of 5OC (10F) must
gage, special accessories improve its ability to with-
be considered, particularly for the more accurate stand adverseconditions and broaden itsusefulness by
gradesused as testgages. Temperature errors for performing functions not normally required of a pres-
normal materials encountered will range between sure gage alone.
1.5% and 2.5/a of the pressure span for 55OC ( 1 O O O F )
change in temperature. For certainspecial applica- 3.6.5.1Chemical Diaphragm Seals. Diaphragm
tions, temperature-compensated gages have been sealsare used to prevent the measured fluid from
manufactured which embody materials anddesigns to reaching thegage elastic element.Chemical seals may
effectively reduce temperatureerror over specific tem- be required to:
perature ranges. (a) seal off a corrosive chemical that would seriously
Vibration effects may be reduced or eliminated by attack or perhaps destroy the gage pressure element;
mounting the gage remotely using flexible piping. (6) prevent the entry of contaminants, solid parti-
Gages having internal damping, such as liquid filling cles, or liquids that might solidify in the pressure-
the case, may also have reduced vibration effects. sensingelement,therebyrenderingthe gage
Permanently installed gages mayhave long pipes inaccurate or inoperative.
leading to a central gage panel. It is not unusual to find The gage manufacturer should be consulted regard-
that more attention has been given to appearance than ing the selection of chemical seals. Proper selection of
to accuracy.Long pipe lines are not good practice materials and themethod of attachment may be criti-
where good accuracy is essential, Test gagesin particu- cal. Also, the accuracy of a gage will be affected by
lar shouldbe installed with pipingas short as practical. adding a chemical seal,
It may be necessary to temporarily insert a test gage at
3.6.5.2 Pulsation Dampers. Pulsation dampers are
the line pressure tap and calibrate the remote panel
frequently utilized to reduce the magnitude of line
gage accordingly,where long piping cannot be
pressure pulsations. Rapidly pulsating pressurecan
avoided.
quickly destroy gage accuracy by producing abnormal
Transmitters, both pneumatic and electric, are fre-
wear on moving parts. The elastic element is a metal
quently employed in industrial plants to economically
spring member and rapid pulsations can cause exces-
and safely transmit the process pressureto some more
sive reversing stresses and eventual metal-fatiguefail-
convenient location. While the accuracy of transmit-
ure. In addition, pulsating pressures often make it
ters is often more than adequate for industrial opera-
virtually impossible to obtain pressure readings. Nu-
tions, precise measurements for tests should preferably
merous types of pulsation dampers are available. The
be made with a suitable, calibrated gage atthe point of
gage manufacturer can provide valuable assistance
measurement. Wherethis is not possible, the transmit-
with the proper selection.
ter and receiver gage should be calibrated as a unit, and
this unit treated the same way as the calibrated test 3.6.5.3 Gage Cocks. Gage cocks may be installed in
gage. The reasons for preferring the use of a separate the gage line to shut off the gage from the systemeither
gage are ease of calibration and the greater reliability of to provide isolation to reduce unnecessary wear or to
a single element. permit gage replacement without system shutdown.
71

0 7 5 9 b 7 0 005Z-Lbb 4 m
. .
3.6.5.4 Siphons. Siphons (vapor traps) are installed the gage be given a complete calibration-verification
in pressure lines ahead of the gage to provide a seal test before further use. Accidents include but are not
. against steam or other condensable vapors of high tem- limited todropping, jarring, application of pressure in
perature. The siphon is a device designed to permit excess of its calibrated range, or application of a fluid
cooling of condensable vapors and retention of the considered corrosiveto the pressure element.
condensate when installed in series between the gage Testgages shouldalways bear a calibration-
and the pressurized fluid. Condensate may be added at verification chart indicating date of last verification and
installation or condensation induced by rapid-cooling due date of the next calibration verification. It i s also
the siphon. Pressure should always be applied slowly desirable to indicate the applicable correction values.
so that the interior of the gage i s kept cool by the
3.6.6.2 Corrections. It may be necessary to apply
condensate.
correction values to gage readings. These values may
3.6.5.5 Bleeders. Bleeder devicesmay be utilized to be obtained during calibration verification. A sug-
permit flushing, liquid-filling, or draining of the pres- gested form for recording this information i s provided
sure element. The bleeder can alsobe used to ensure in Table 3.4.
proper filling of the pressure element, especially when The corrections are the numerical values to be added
precise measurements at low pressures are required. algebraically to the gage reading to give the correct
Bleeder devices must be incorporated at gage assem- pressure. The readings should be correctedfor temper-
bly; they cannot be added at a later time. ature and head effects.
3.6.6.3 Adjustments. Gages having high accuracy
3.6.5.6 Heaters. Heaters may be required to protect
ratings are frequentlyprovided with adjustment
gage elements and linesfrom solidification of fluids in
means, rotatable dials, adjustable pointers or move-
the line or gage. Caution must be exercised to prevent
ments, and span adjustments. The first three arereadily
overheating.
available to the user and may be used to adjust for er-
3.6.5.7 Maximum or Minimum Pointers. For record rors that areapproximately equal over theentire pres-
purposes, it may be desirable to establish the highest or sure span. In general, they should be used to set the
the lowest applied pressure, so an additional pointer scale (dial) or pointer at one point on the scale for a spe-
(lazy-hand ortelltale pointer) may be mounted through cific pressure input.
the gage window in such a mannerthatthe gage- Except in high accuracy test gages, these adjustments
pressure-indicating pointer will move the second should not be used to simply reset the pointer to zero.
pointer to position of maximum or minimum applied Span adjustments arenot readily accessibleto the gage
pressure (usually only one value, not both) andremain user. This adjustment is usually found on the internal
atthat position until manually reset.Thesedevices components of the movementor link, Span adjustment
are frequently available as add-on or replacement may be used to correct a progressive error, uniformly
assemblies. increasing or decreasing over the pressure span. The
gage manufacturer's instructions should be consulted
and adjustments shouldonly be madeby experienced
3.6.6 Gage Calibration, Gage-operating parameters personnel. Calibration verification must be repeated
and gage testing (calibration verification) are covered after adjustments have been completed.
in detail in ANSVASME 640.1 Standard for Gages. It is
essential thatall gage usersand manufacturers operate
3.6.7 Precautions. Thepressure-sensing element in
using the same base standards. Therefore, copies of
most pressure gagesis an elastic element subjectedto
this American National Standard should be obtained
high internal stresses. Applications exist where the
and utilized.
possibility of catastrophic failure is present.ANSI/
3.6.6.1 Test Gages. Test gages or inspector's gages ASME 640.1 covers most of the considerationsessen-
are commonly used as working standards for gage cali- tial in a pressure gage.
bration verification. These gagesmust be carefully han- Protective shields should be providedbetween
dled to ensure reliable accuracy. Handling mustbe gages and the person taking readings, particularly in
done with the knowledge that a gage suitable for accu- systems where compressedgas and high pressures-are
ratepressuremeasurement is a delicate instrument. involved.
During its removal and replacement, avoid dropping It shouldbenotedthat testgagesbear a proper
or jarring the gage or subjecting it to excessive temper- current-calibration chart,Proper compensation for
ature. An accident involving a test gage requires that fluid head should be indicated, preferably with a
72

ASME P T C * 2 7 - 2 8.7 m 0757b70 0052367 b W
sketch showing any head of liquid inthe gage. Check 3.7.1 Flow[l4]. The method of interest relates the pres-
that the gage pointer is fastened securely. This may be sure drop across a metering restriction, suchas an ori-
accomplished bywatchingwhile the pointer is in afree fice plate, a flow nozzle, or a venturi tube to the rateof
position and the gage i s tapped. If loose, remove the fluid flow. A differential-pressure transmitter i s com-
gage for recalibration. monly used to measure this pressure drop and send n-
Check to determine if a bleederhas been provided in formation to a centralcontrol room (see Fig. 3.48 as an
the gages or pipingand, if incorporated, that thesystem example).
has been bled and closed. The differential-pressurespan is often very small rel-
If an adjustable throttling device has been installed ative to the static pressureinside the pipe. For example,
to control pressure pulsations, it shall be adjusted to it may be on the order of 25 kPa (1O0 in. H,O) with a
leave some pulsation. static pressureof 49 MPa (5800 psi), a ratio of 1 part in
All readings shall be taken while the gage is being 1600. This limits the choiceof transmittersto those that
lightly tapped. Thepoint of viewshould be located so have been specifically designed for operation up to
as to avoid parallax. Actual readings should be re- high static-pressure levels with minimum effect of
corded on log sheets(see Table 3.4). Corrections static-pressure variation on the measurement of the
should be made when the test has been completed. pressure differential.
Check that the pressure-element materials arecom- Associated with high static-pressureoperation is the
patible with the pressure media. overrange characteristic of the flow transmitter. The
Make sure that thegages arenot subjected to temper- ability to have full staticpressure applied indepen-
atures in excess of manufacturers recommended val- dently to either process port (high-side or low-side
ues, and that materials used in the gage arecompatible overrange) during startup, or in the event4of a system
for temperatures to be encountered. malfunction, is essential. A suitable transmitter will
maintain its calibrated accuracy.
Figure 3.49 shows schematically the primary of a
3.6.8 Form for Reporting Gage-Calibration Verifica-
transmitter that has been designed for flow measure-
tion. The record of the observations of a test whereby
ment.Thediaphragms,process-cavityflanges,and
gage-calibration is verified should be made in some
suitable form such as that in Table 3.4. The readingsin flange-fittings (vent or drain valve and process adap-
tors) are selected from a choice of corrosion-resistant
the column marked Standard Pressure are the pres-
materials offered by the manufacturer. These are usu-
sures corresponding to the deadweight tester input
ally the only parts that come into direct contact with
pressure (weights including the weight tray). The read-
the process fluid.
ings in the column marked Up are obtained in the
The schematic shows connecting rods from the dia-
order given and the readings in the column marked
Down are recorded from the bottom upward. The phragms deflecting a force beam that extends into the
transmittersecondaryhousing. Here, one of many
column marked Average i s then computed as a
types of motion detectors or force-balancing mecha-
mean between theascending and descendingcolumns
nisms provide either a pneumatic or an electronic out-
to the nearest value which can be estimated. Thegage
p u t signal.Sometimes, the force beam w i l l be
correction is the amount to be added algebraically to
the average gagereadingto give the pressure shown by completely eliminated in electronic forms of the trans-
the standard. If the tests were done before the first use mitter which detect diaphragm motion inside the pri-
of the gage, hysteresis should then be computed. The mary capsule and sendan electrical signal to an
amplifier in the secondary housing. Some of these de-
gage corrections are copied on a tag attached to the
signs are covered in other paragraphs.
gage, with date, and the gagepassed for use.For a
The centercavity of the primary capsule is filled with
gage correction test after use, the variance should be
silicone oil or other suitable liquid to support the high
computed.
staticpressure of the process. Thecenter cavity is
sometimes hermetically sealed by means of an all-
welded construction to enable long-term full-vacuum
3.7 SPECIAL APPLICATIONS OF AP CELLS
operation. A further function of the filling liquid i s to
Some very important special applications of pressure provide damping (in combination with a suitable inter-
and of differential-pressure transmitters,particularly in nal resistance restriction and volume transfer). Flow
the process industries, include use in making flow, liq- signalsare often noisy,and without this damping
uid level, and density measurements. These applica- would cause noisy output signals to be transmitted to
tions are briefly described. the controlroom. Excessive damping, however,
73

ASMEYANSI PTC 19.2-1987 PRESSURE MEASUREMENT
TABLE 3.4
RECORDING OF GAGE-TEST DATA SAMPLE
DESIGNATION OF CACE NO. OWNER

Make and Type:
Size and Range:
CONDITIONS DURING TESTS
Before
Use During
Use
After
Use
Date and hour of test

Temperature of gage
Pressure standard used
TEST DATA
(All in lb per sq in.) I Variance
Standard
Pressure
T Cage Reading Before Use.
Flap Gage Before Reading.
Down Average Correction
T Cage Reading
Down
After
Average
Use.
Correction of
Before and
Difference
Between
Before and After
Corrections
-
REMARKS:
(1) Range of pressure during test: ..
(2) Pressure characteristics during use:

(a) Steady - Less than 1% per second and 5% per minute, the percentage referring to the full range of the gage.
(b) Fluctuating - Changes faster than for "steady" - not regular in occurrence.
(c) Pulsating - Changes faster than for "steady" and characterized by cyclic regularity.
(3) Equipment and location where gage was used:
(4) Operator making tests:

(5) Other remarks:
74

MEASUREMENT PRESSURE . ASMVANSI PTC 19.2-1987
valve
t- Low pressure
valve
FIG, 3.48 FLOW INSTALLATTON

(Courtesy of Taylor lnstrument Company)
75

r------ 7
Secondary
I l
housing 4 I
I Force beam
I I
I l
I Thrust pivot
.I and seal
._
adjustment
FIG. 3.49 SCHEMATIC OF DIFFERENTIAL-PRESSURETRANSMITTER PRIMARY

FOR FLOW MEASUREMENT
76

ASME PTC*L7..2 87' m.0 7 5 9 6 7 00 0 5 2 3 7 3 8 W
should be avoided because of the square-lawrelation- The same differential-pressure transmitters as were
ship between flow velocity and differential pressure described in para. 3.7.1 may also be used to measure
that will lead to erroneous flow measurements. Most liquid level. An impulse line from below the minimum
instrument manufacturers, therefore, place an upper liquid level in the tank is connected to the high-side
limit onthe adjustment theyprovide. transmitter-connectionand, following the manufactur-
Overrange protection of the diaphragmsis provided er's instructions, air i s purged from this line and from
by mechanical support from a contoured back-up pro- the process cavity of the transmitter, allowing liquid
file in the body block. There areother designs in which from the tank to enter. The Jow-sidetransmitter con-
the diaphragmsare supported hydraulically by closing nection is vented to atmosphere in an open-tank appli-
a valve to prevent complete volume transfer. cation or connected to the vapor space abovethe
Other requirements relateto, and vary substantially maximum liquid level in closed tanks, as indicated for
with, the operating environment. They include com- the direct-mounted installations that follow. Using a
pensation for outdoor ambient temperature andproc- differential-pressure transmitter withimpulse-line
ess temperature variations, immunity to vibration and process connections is generally the less expensive ap-
mechanical shock, the needto maintain calibrated ac- proach if process fluid can be tolerated in these lines
curacy with power-supply variations andenvironmen- and in the process cavities of the transmitter.
tal interferences (radio frequency and magnetic), and 3.7.2.1 Open-Tank Installations. An example of a
sometimes the need to operate safely in explosive at- direct-mounted liquid-level transmitter capsule for
mospheres (dust, hydrocarbons, etc.). open-tank serviceis shown schematically in Fig. 3.50,
The adjustmentsof concern to the user are for zero, and itsapplication, mounted to a tank nozzle,i s shown
span, and elevation/supression. in Fig. 3.51, The flush diaphragm is suitable for applj-
3.7.1.1 Adjustment for Zero. Zero is adjusted with cations where the process liquid is free from sus-
the inputat the lower rangevalue ofthe pressurediffer- pendedsolids.Anextendeddiaphragmform
ential (usually zero) by first closing the valve to the eliminates the pocket at the transmitterconnection and
low-pressureimpulse line and then opening the equal- should be used for slurries and viscousliquids. Figure
izing valve to ensure that the process cavities are atthe 3.52 shows a variation of the transmitter with the ex-
same pressure (see Fig. 3.48). This procedure is per- tended diaphragm and a remote seal element.
formed after installing the transmitter,to correct instal- 3.7.2.2Closed-TankInstallations. Closed-tank
lation effects, and after the process cavities have been liquid-level applications differ from open-tank applica-
properly filled or drained (depending upon the type of tions in that the pressure over the liquid may be differ-
installation). ent from atmospheric. Figure3.53 i s a schematic
3.7.1.2 Adjustment for Span. Adjustment for span is diagram of a level transmitter for closed-tank service,
normally made by the factory or instrument shop. It re- and Figs. 3.54 through 3.60 show examplesof closed-
quires an application of a known pressure differential, tank installations, using both integral and remote-seal
usually equal to the upper range value, in addition to transmitter forms.These aredifferential-pressure trans-
zero. An exception is when the transmitteri s being re- mitters with one side connectedthrough a compensat-
turned to a previously calibrated span setting. ing leg to measure pressure above the liquid.
The compensating leg canbe either wet or dry de-
3.7.1.3AdjustmentforElevation/Supression. The pending on the characteristics of the process vapor.
elevationhppression adjustment operates like a Any change in liquid level in the compensating leg,
coarse zeroing adjustment andi s used only toachieve however, w i l l cause measurement error. Also,
ranges that are not zero based. ambient-temperature changes can result in excessive
errors dueto changing specific gravitiesin the wet leg.
3.7.2 Liquid level. Many of the liquid-level measuring A dry leg is used when the process vapor i s not readily
devices used by industry for accuracies to about 0.5% condensable or when the compensating leg is at a
of span depend upon the fundamental equation higher temperature than the tank interior (seeFigs.
3.54, 3.55, and 3.56).A trap installed at the bottom of
Pressure = (Density) (Height] the leg minimizes the possibility of condensate collect-
ing inthe compensating-diaphragm cavity.
Pressure and fluid height bear a direct relationship if When the process vapor i s condensable, a wet leg is
density remains constant, and for most applications recommended(see Figs. 3.57,3.58, and 3.59). Theleg
this is a valid assumption. Density compensation will can be filled with process liquid or a suitable seal liq-
be discussed later (see para. 3.7.2.4). uid(seeTable3.5),usingafillingteeinstalledatthetop
77

ASME/ANSIPTC 19.2-1987 PRESSUREMEASUREMENT
Force beam +
Vented to atmosphere
L Liquid fill
Process diaphragm
FIG. 3.50 SCHEMATIC DIAGRAM OF OPEN-TANK TRANSMITTER PRIMARY

78

ASME P T C * L S * 2 87 M - 0 7 5 7 b 7 0 . 0 0 5 2 3 7 3 3 W
REMENT PRESSURE PTC 19.2-1987
Datum
line
FIG. 3.51 OPEN-TANK1NSTALLATlON WITH TRANSMITTERMOUNTED

DIRECTLY TO TANK NOZZLE
79

ASME P T C * : L 7 * 2 8 7 I0759b70 0 0 5 2 3 7 4 3 m
line
FIG. 3.52 OPEN-TANK INSTALLATIONS WITH REMOTE SEAL

TYPE OF TRANSMITTER
80

ASME PTCmi7.2 87 R I 0757670 0 0 5 2 3 7 5 5 m
MEASUREMENT PRESSURE 19.2-1 987
Force beam
Vent or
drain Thrust pivot
valve and seal
Primary
diaphragm
Liquid fill
Process diaphragm
FIG. 3.53 SCHEMATIC DIAGRAM OF CLOSED-TANKTRANSMITTER PRIMARY

81

ASMWANSI PTC 19.2-1 987 PRESSUREMEASUREMENT
I Datum line
~.
FIG. 3.54 CLOSED-TANKINSTALLATION,DRY LEG

82

ASME P T C m 1 7 - 2 8 7 U 0 7 5 q h 7 00 0 5 2 3 7 7 7 W
r Gate
FIG. 3.55 CLOSEDUTANKINSTALLATION,DRY LEC: TRANSMITTER

ABOVE DATUM LINE
83

ASME/ANSI PTC 19-2-1967 PRESSURE MEASUREMENT
Gate
l-""l I
I I
1 I
I I
I I
Drain
va've
I x Condensate trap
(optional)
FIG. 3.56 CLOSED-TANK INSTALLATION, DRY LEGTRANSMITTER

BELOW DATUM LINE
84

ASME P T C * 3 9 * 2 87 II 0 7 5 7 b 7 0 0052377 2 m
Filling
tee \
Datum
line
FIG. 3.57 CLOSED-TANKINSTALLATION,WETLEG

85

S Datum
c line
FIG. 3.58 CLOSED-TANK INSTALLATION, WET LEGTRANSMITTER

ABOVE DATUM LINE
86
COPYRIGHT American Society of Mechanical Engineers Ir-

. . .
A S I E P T C * L 7 - 2 8 7 IBBl 0 7 5 4 L 7 0 0052LBL O m
Filling Gate
tee
7
T
1
S
Datum
line
FIG, 3.59 CLOSED-TANKINSTALLATION,WETLEGTRANSMITTER

BELOW DATUM LINE
a7

r 1
valve
Minimum level
FIG. 3.60 CLOSED-TANK INSTALLATION, DRY LEG TRANSMITTER

ABOVE UPPER PROCESS TAP
88

of the leg. The wet leg can beavoided if the transmitter . TABLE 3.5
can be installed near thetop of the tank(see Fig. 3.601, .. SEAL FLUID SELECTION CHART
so that the condensate drains backinto the tank. (Courtesy of Taylor Instrument Company)
In allof the installations shown, the minimum mea-
sured level must be at or above the datum line. Also, Freezing Point/
High-Temperature
Specific
Recommended
when the installation is to be used on vacuum service,
Gravity Limit Liquid Service
the transmittershould usually be mounted at or below
the datumline (see Figs. 3.54,3,56,3.57, and 3.59) to Water o/ + 93 1 .O0 General usage
keep thefilling liquidpressurized above its vapor pres- 50% Water - -28/+ 96 1.14 Anti-freeze
for
sure, However, this requirement varies somewhat
hydrocarbon 50/0 Glycerin
amongmanufacturers, depending upon their filling service
pressure, so emphasis is placed on following the manu-
facturer's recommendations. 50h Water - - 36/+
149 1.O7 Anti-freeze
for
50% Ethylene than other
To determine the span and range valuesfor a specific Glycol
application, use the following equations. service
Below
Fluorinated 1.90
alkalies,Acids,
Span = AG, Hydrocarbon - 1 B/+ 260 [Note (r)] strong salts
high(generally at 38C
viscosity)
Lower Range Value = SG, + SfGf - Gs - EfGf Silicone

(high +21/+316
a . . e . .
temperature)
Silicone (low -51/+149 ... ...

Upper Range Value = (A i- S) C, + SfGf - GS - EfGf temperature)
NOTE:
where ( 1 ) Changes greatly with temperature, affectingmeasurement
accuracy.
A, S,
Sf' E,
and
E f = length as shown in Figs. 3.51, 3.52, and 3.54
through 3.60 The useful working range is determined entirely by
G, = specific gravity of the liquid in the tank the air supply and the required output pressure, which
G, = specific gravity of the liquid in the wet leg can be biased relative to the measured pressureusing a
Gf = specific gravity of the liquid inthe tube system zeroing adjustment. A sensing diaphragm contacts the
A negative upperor lower range value would indicate process liquid ona nearly flat surface. There is no fill-
that positive pressure must beapplied to the compen- ing liquid, which simplifies maintenance, Linearity
sating sideof the measuring element when calibrating varies somewhat with the choice of supply pressure
a transmitter for this range. Refer to Table 3.6 for the and with the transmitter design, but itis usually best at
midrange. The manufacturer should be consulted for
type of calibration required.
Note that specificgravity terms have been usedas a details of operation, application, associated equip-
convenience in the preceding equations rather than ment, and accuracy.
density, causing the span values to be in terms of height
3.7.2.4 Density Compensation, It may be important
of an equivalent column of water. This is because of
to accurately know the level under conditions of vary-
the extensive use of water and its role as a reference
ing specific gravity. The differential-pressure type of
standard for many pressure measurements.
level meter measures the product of height and aspe-
3.7.2.3Repeaters. Another device often used for cific gravity. If a second instrument i s added which
level measurement i s a direct-mounted pneumatic measures specific gravity and the reading on the level
force-balancetransmitterthatreproducesa process meter i s divided by the gravity reading, the actual level
pressure on a one-to-one basis (see Fig. 3.61). Some- of the liquid is the result.A small computer (pneumatic
times called a "repeater," it has no particular range or or electronic) can be used to perform the division and
calibration inherent in its construction. provide a compensated level signal.
89
COPYRIGHT American Societyn of Mechanical Engineers

ASMVANSI PTC 19.2-1 987 PRESSUREMEASUREMENT
.-m
Y-

I
p-
m
C
4-
+-
P
W
m
+-
C
\ SU
90
COPYRIGHT American Society of Mechanical Engineers r

ASME PTC*K17.2 8 7 111 0 7 5 7 b 7 00 0 5 2 1 8 5 8 m
senfed by the reference water column of Fig. 3.63)

TABLE 3.6
could be eliminated. The low-side pressure connec-
TYPE OF CALlBRATlON REQUIRED
tion would simply be vented to the atmosphere.
FOR VARIOUS APPLICATIONS
(Courtesy of Taylor Instrument Company) 3.7.3.2Sample-Column Method. A common
method of measuring thedensity of a process liquid is
Transmitter Application by means of a samplecolumn shown schematically in
Initial Level Transmitter Type of Fig. 3.64. The sample enters the column at the bottom
Service Calibration
Elevation
Condition and overflows into a return line toestablish a constant
Atdatum line Zero based sampling height. A bubbler tube connected to the
Open tank differential-pressure instrument makes possible the
or Minimum level
Above datum
Elevated zero measurementof pressure at thebottom of this column
closed tank
at datum line
line without bringing the process liquid into contact with
with dry leg
Below datum Suppressed zero the instrument. The resulting backpressure is related to
line the liquid level measured vertically from the base of
Abovedatum Elevated zero the dip tube to the liquid surface. This arrangement
Minimum level line [Note (111 makes it easy to measure or adjust the head.
above datum The required height of the sample column is deter-
below
orAt
line Suppressed zero mined by dividing the differential-pressure span of the
datum line
instrument (height of water) by the density span. Here,
Closedtank Minimum levelAnytransmitter Elevatedzero the temptation is to use a short instrument pressure
with wet or above elevation span in order to make the samplecolumn short; how-
1% line
datum ever, the air-pressure change in the tube is about 3mm
~~
(0.01 in.) water during the formation of every bubble,

NOTE:
(1) Can be suppressed zero depending on relative head pressures. thereby placing a practical minimum length on the
column for signal stability.
The diameter of the sample column is also impor-
3.7.3 Liquid Density (Specific Gravity). Density is tant. For good speed of response, it is desirable to
mass per unit volume and i s usually expressed as kilo- change the sample at least once per minute. However,
grams per cubic meter (kg/m3). Specific gravity is just one must rememberthat flow i s always accompanied
one of many numerical scales which may beapplied to
by pressure drop; also,fluctuations in flow will change
density-measuring instruments. It is the ratio of the
the liquid head above theoverflow weir which is the
density of a liquid tothe density of water, the tempera-
perimeter of the top of the sample-column pipe. A
ture of both liquids being stated. Thus, sp. gr. 0.904
good approach to the problem is to picka pipe size and
20C/40C means that the densityof the liquid sample
then calculate its performance using the sample flow
at 20C divided by the densityof water 4C is 0.904.
rate, its variation, and
the volume and top perimeter of
Since specific gravity is a ratio, it is dimensionless.
the proposedpipe. The variablesto be determined are
3.7.3.1 Basic Concepts. The application of a the rate of change of the sample, the pressure change
differential-pressure instrument to density measure- due to flow variation, and the head change of liquid
ment develops from the factthat pressure atthe bottom above the crest of the weir (Francis formula2) due to
of a vertical column of liquid is the product of liquid flow variation.
density multiplied by the height of the column. It is
readily seen, however, that asimple hookup such as is Francis Formula
shown in Fig. 3.62 would be unsatisfactoryformostap-
plications because of insufficient sensitivity. A more
useful differential arrangement (see Fig. 3.63) - load-
ing one side of the instrument with a constantknown where
h= I*( 213
h = head of liquid above crest of weir, m

pressure - would allowthe pressure span to beshort- L = perimeter of top of sample column, m
ened. By making its span one-tenth that of the instru- Q = flow rate over weir, m3/s
ment in Fig. 3.62, its readability is increasedby a factor or
of 10. If the instrument was provided with a suppres-
sion adjustment (usuallyavailable in differential-
pressuretransmitters), the constant pressure to the h = (S)
for h and L in ft, and Q in ft3/s
2/3
low-pressure sideof the differential instrument (repre-

91

ASME P T C * 1 9 - 2 8 7 0 7 5 9 b 7 0 0052LBb T W
FIG. 3.62 HYDROSTATICHEADPROVIDESONEMETHODOF

DENSITY MEASUREMENT
"
"
-" "
"
".
Reference
"+
column
"
FIG. 3.63
=e-
DIFFERENTIALHYDROSTATICHEADINCREASESSENSITIVITY
OF DENSITY MEASUREMENT
92

-3 140 kPa air

D/P fransmitter
with suppression
O
J
~ O
column
Sample in
FIG. 3.64 COMMONMETHOD OF MEASURINGDENSITY

OF A PROCESS LIQUID
93

TABLE 3.7
VARIATIONS IN DENSITY
FOR DIFFERENT FLUIDS
Change in Density
Liquid Per OC Per OF
Water at 15C (59F) 0.00016 0,00009

Water at 31"C (88F) 0.00032 0.00018
Water at 80C (1 76F) 0.00062 0.00034
Sugar, 50% solution at 20C (68F) 0.00036 0.00020
Caustic Soda (NaOH),50% solution 0.00073 0.00041
at 20C (68F)
if a bubble tube is used,pressure on the bubbles

must beheld constant. This requires a good quality air-
regulator upstream from the adjustable restriction,
Temperature fluctuation is by far the greatest source
of error in density measurements. Thisis apparent from
a few examples shown in Table 3.7. For petroleum
products and organicsolvents, the temperature effects
are much greater than for aqueous liquids, and range
from 0.0007 to 0.0014/C. Therefore it may be neces-
sary toapply a heat-exchangerand temperature-
controller to the incoming sample.
94

ASME P T C * 1 9 . 2 87 m 0757670 0052187 5 W
SECTION 4 - PRESSURE MEASUREMENT INSTALLATIONS

While the importance of the proper choice of a
pressure-measuringdevice and itscalibration is not to
be understated, it is the application of the selectedde-
vice to the actualpressuremeasurementthatoften
poses the greatest difficulty. As any experienced test
engineer canattest, it is quite a different matter to make
accuratepressuremeasurements in a field environ-
ment (which is characterized by a combination of
noise, vibration,moisture,andtemperature fluctua-
tions) thanto make acalibration under controlled labo-
ratory conditions. The ideal constant-pressure source
used for calibration purposes is replaced by a pressure
sensor of some sort, usually placed in a moving fluid
stream which may contain pressure oscillations result- FIG. 4.1 PRESSURE TAP F L O W F I E L D
ing from the flow field. The pressure sensor may range
in complexity from a tap in a pipe wall to a probe
*
which must becalibrated for compressibility and flow sure tap i s a function of th: Reynolds number Rd
angleeffects.Betweenthepressuresensorandthe based on the shear velocity v and the tap diameterd.
pressure indicator there may be many feetof connect- The shear velocity equals the square root of the ratio of
ing tubing introducingadditional problems, the local wall shear stress to the fluid density at the
It is the intention of this Section to guide the test engi- wall e.
neer in the set-up of his instrumentation so as to mini-
mize the additionaluncertaintyintroducedin a
pressure measurement madein the field environment.
"* =J"
e
The dataof Shaw [15], Rainbird [16], and Franklin and
Wallace [I71 for taps with geometry as shown in Fig.
4.1 PRESSURETAPS
4.2, are correlated by the following expressions [18]:
The basic pressure sensor is the pressure tap or pi-
ezometer. A pressure tap usually takes the form of a
= 0,0001 57
hole drilled in the side of a flow passage and i s as- 70
sumed to sense the true static pressure.When the fluid
is moving pastthe tap, which is usually the case, the tap *
For Rd = v*dlu 385
will not indicate the true static pressure. The stream-
lines are deflected into the hole as shown in Fig. 4.1,
setting up a system of eddies. The stream-line curvature -
AP = 0.269 (Rdt)0'353
70
results in a pressure atthe tap "mouth" different from
the true fluid pressure. These factors in combination * *
for Rd > 385 (extrapolation beyond Rd E 1 O00 may
result in a higher pressure at the tap"mouth" than the
be unreliable)
true fluid pressure, a positive pressure error.
Figure 4.3 shows the errors for different-size taps in
fully developed flow in a smooth pipe of diameter D.
4.1.1 Velocity-Induced Errors. The magnitude of the Theerrorsare nondimensionalized by the dynamic
pressure error of a carefully made square-edged pres- pressure q = 112 (eV2)and are a function of the pipe
95

l
4.2 PRESSUREPROBES
4.2.1 Total Pressure Probes.Total pressure probes are
used to determine the total pressure at a specific loca-
tion. Total pressure is used to determine head-loss data
and to establish velocities,state points, andflow rates.
By definition, total pressure canbe sensed only by stag-
nating the flow isehtropically.
4.2.1.1 impact Tube. An impact tube or Pitot tube is
an open-end tube placed in the flow field pointing di-
rectly upstream (see Fig. 4.5). The pressure in the tube
i s total pressure P,. The maximum velocity can be de-
FIG. 4.2TAP GEOMETRY termined by changing the orientation of the Pitot tube
until a maximum total pressure is observed. If the static
pressure PS is known and the fluid is incompressible,
the velocity pressure P, can be calculatedas the differ-
Reynolds number. Larger tap diameters and higher ve- ential between thetotal and static pressure. This can be
locities givelargererrors.Similarcalculationshave used to calculate the velocity V of the fluid at the im-
been carried out for throat taps in an ASME nozzle [19, pact tubes location. For incompressible flow of den-
201. sity @:
The above information represents a correlation of
availableexperimental data for a limited Reynolds
number range. Other correlations havebeen found to
v=J-
be more representative at higher Reynolds numbers,
such as those encountered in throat tap nozzles (see
Ref. [21]). The impact tube can be traversed acrossaduct todeter-
mine the velocity profile. The shape of the tip deter-
mines the sensitivity of the probe to flow angularity
4.1.2 Other Sources of Errors. The effectof compress- (flow not parallel to the head). Figure4.6 gives the vari-
ibility on tap errors i s not well understood or demon- ation of total pressure indication with angle of attack
strated, even though correlations for this effect have and geometry forPitot tubes, where AP, is the change
been suggested [16,22]. The only conclusion that can in total pressure and P, is the velocity pressure [27].
be reached is that at Mach numbers nearunity, the tap
error is greatly magnified and measurementsin this re- 4.2.1.2 Kiel Probe. A Kiel probe resembles an im-
gion should be avoided if possible. pact tube surrounded by a cylindrical shroud to direct
When a pressure tap i s located in an accelerating the flow parallel to the headof the impact tube (see Fig.
flow field, the external pressure gradient is the signifi- 4.7). Kiel probes are used because they are relatively
cant parameter for correlating tap error. It hasbeen insensitive to pitch-and-yaw angles up to angles of 40
found [16, 23, 24, 251 that the effect of the pressure deg. or more measured from the axis of the head. They
gradient is to move the effective location of the tap up- are suitable for measuringtotal pressure in cases where
stream from 0.30 to 0.37 tap diameters. The lower the exact flow direction i s unknown or varies with
number correspondsto incompressible flow whilethe operating conditions,
higher number correspondsto nearly sonicflow. Other types of probes are usedto measure total pres-
In most casesit i s possible to reduce the pressure-tap sure, All of these probes operate on the principle of
error by using taps of smaller diameter. Thelimitation stagnating theflow isentropically (as occurs at the up-
is, however, that it is more difficult to make a smaller stream sideof a cylinder oriented perpendicular to the
tap that is free from burrs. The presence of burrs of a flow field) (see Fig. 4.8).
height greater thanabout 0.008 times the tap diameter
will greatly magnify the taperror [I 51. Similarly, round-
ing of the tap, or locating the tap at positions other than 4.2.2StaticPressureProbes. Staticpressureprobes
normal to the surface will also affect the tap error as sense the static pressure of a fluid field whether the
shown in Fig. 4.4 [26], whererelative errors are shown fluid is in motion or at rest. Static pressureis required to
as percentages of the dynamic pressure. determine fluid velocity and is useful in obtaining flow
96

ASME PT-C*(SL7-2 8 7 0757b70 0052373 3 m
1.2
1.o
0.8
0.6
0.4
0.2
O
104 2 5 105 2 5 1o6 2 5
Pipe Reynolds Number, RD
FIG. 4.3 ERRORS FOR DIFFERENT SIZE TAPSIN FULLY DEVELOPED PIPE FLOW
97

Reference form
+ 0.2%
%
30 deg.
+ 0.3%
+
t 1.1%
?Ir
R = 1/40
-b 0.9%
%
45 deg.
t 0.4%
&
- 0.1% - 0.5% W. 0.0%

&
30 deg.
- 0.3%
W - 0.1%
45 deg.
FIG. 4.4 RELATIVETAPERRORSASPERCENT OF DYNAMIC PRESSURE
98

ASME P T C * L 7 - 2 87 m 0759670 0052373 7
Yaw angle
FIG. 4.5 IMPACT TUBE
99

Licensed by Information Handling Services c
ASME P T C * 3 7 - 3 8 7 m 0757670 0052374 7
o.1
o.2
0.3
*I
% 4

0.4
0.5
0.6
0.7
40 30 20 10 O 10 20 30 40
Angle of Attack, deg.
2o 22-112 deg.
I 15 deg.
W in.
114 in.
0 1318 in. i- (Ogival tubes)
FIG. 4.6 VARIATION OF TOTAL PRESSURE INDICATION WITH ANGLE OF ATTACK

AND GEOMETRY FOR PITOT TUBES (AFTER NACA T N 2331, APRIL 1951)
1O0

ASME PTC*K17*2 8 7 Ba 075.9b70 0052195 O W
Impact tube
\
f Flow direction
I
rc---------
I m .
FIG. 4.7 KIELPROBE
Total pressure
-
/- location
n /
Flow
direction
Section A - A
A A
~ Total pressure
location
FIG. 4.8 TOTAL PRESSURE LOCATION ON A CYLINDER IN A FLOW FIELD
101

direction (as used on sphere, cylinder, wedge,and probe, andit is often not feasible. If only the static pres-
cone-type probes), sure is of interest, a basket probe can be used, This
probe measures the static pressure independent of the
4.2.2.1 Static Tube. A static tube, similar to an im- flow direction for low-to-moderate fluid velocities. An
pact tube, is used to determine the staticpressure in a example of this probe i s shown in Fig, 4.14 [30]. The
fluid stream (see Fig. 4.9).Theaccuracy of static- basket probe is particularly useful for static pressure
pressuremeasurement with a static tube depends
measurements in thecondenser of a steam turbine
mainly on the location of the sensing taps. The nose of
where the flow direction is not well defined. Another
the probetends to accelerate theflow whichlowers the type of basket probe is shown in Fig. 4.15 [31]. This
tap pressure, while the stem tends to stagnate the flow version of the probe has a pressure transducer located
raising the static pressure.Both effects should be com- in the probe head itself to avoid the buildup of water
pensated for whenconstructingthis type of probe. Cal- legs.
ibration of the static tube is necessary before use. The
static tube can be combined with an impact tube to 4.2.24 Probe Blockage Effects. Standard pressure
give a Pitot-static tube which samples both the static probes such as cylinders and spheres are commonly
and total pressure (see Fig. 4.10). used to measure total and static pressures as part of per-
formance tests of flow elements, lt is usually assumed
4.2.2.2Aerodynamic Probes. Accuracyofstatic- that the probes presence in the flow field does not
pressuremeasurement using static-pressuretaps in change the characteristics of the flow, If the probe is
aerodynamic probes depends onthe tap location, tap small comparedto the flow area, its effecton the flow is
size, Mach number, and direction and changeof direc- usually small and can be neglected. However, larger
tion of the flow field [28, 291. Many probe configura- probes will noticeably influence the flow characteris-
tions fall underthegeneral title of aerodynamic tics such thatthe measurements will no longer indicate
probes. Among these are the spherical, cylindrical, the correctflow parameters. This phenomenonis gen-
wedge, and cone-type probes. To exemplify their prin- erally referred to as the probe blockage effect.
ciple of operation, consider the cylindrical probe, The blockage effectcan be looked upon as a pertur-
sometimes referredto as a Fechheimer probe,in a flow bation of the velocity in the vicinity of the probe.These
field as shown in Fig. 4.11. Two taps are located on the perturbations are important both when the probe is cal-
cylinder in a plane perpendicular to the cylinders cen- ibrated andwhen it is used to make pressure measure-
ter line but separated by a certain angle. Calibration of ments. If the significant blockage i s due to the probe
the cylindrical probe dependson the tap location with stem, Figs. 4.1 6 and 4.17 can be used to estimate the
respect to the flow direction. As the probe is rotated in magnitude of the probe blockage effect[32]. These fig-
a flow field, an orientation can be obtained where the ures are based on a cylinder of diameter d immersed
pressures sensed at both taps are equal. This can be midway into the circular free-jet or pipe of diameterD.
done, for example, by connecting the taps to opposite Note that in a free-jet, the effect of blockage is to in-
legs of a manometer. The pressure sensed at these taps crease the static pressurewhile ina pipe, the opposite
can then be determined and calibrated with respect to is true.
static pressure, In this way, a probe that senses static
pressure and two-dimensional flow direction can be
obtained.
Using the same principle as described above, static 4.3 CONNECTING PIPING
pressurecanbemeasured by wedge-type probes
which similarly give a two-dimensional flow direction In evaluating a pressure measurement,it is necessary
(see Fig. 4.12). to knowthe densityof the fluid inthe connecting pip-
Three-dimensional flow directions can bedeter- ing. If this fluid is supposed to be gas, then there must
mined using five-hole spherical and cone-type probes be no possibility of collecting or trapping liquid in the
(see Fig. 4.1 3). These probes use the same principle but connecting piping. On the other hand, if the piping is
in two perpendicular planes. Therefore, a static. supposed to be filled with liquid, then trapping of gas
pressure-balance between two holes in each of thetwo pockets should be avoided.These precautions are par-
perpendicular planes i s required. A yaw-and-pitch ticularly necessary at lower pressures.
angle is then determined. For pressures which are below atmospheric pres-
sure, the connecting piping should contain a means for
4.2.2.3 BasketProbes, A static-pressuremeasure- bleeding air or suitable gas nearthe pressure-
ment can be quite cumbersome using an aerodynamic measurement device.A very small rate of gas flow may
102

ASME P T C * 3 7 - 2 8 7 'm: 07.57670 0052377 4
Static holes
>-o- / 4- Flow direction
FIG. 4.9 STATICTUBE
Static holes
Impact Flow
hole direction
Stat
CO
I Impact
connection
FIG. 4.10 PITOT-STATIC TUBE
103

ASMUANSI PTC 19.2-1 987 PRESSURE MEASUREMENT
1 Flow direction
Static tap
Cylindrical
probe
FIG. 4.11 CYLINDRICAL PROBE,PRINCIPLE OF OPERATION
104

, ASflE P T C * L q * 2 B7 M O759b.70 0052197 B W
Section A -A
Static
pressure
taps
- Flow
direction
- Flow
direction
FIG. 4.12 WEDGE-TYPEPROBE
Four static
pressure taps
c+r x
Spherical Type Probe Cone Type Probe
FIG. 4.13 SPHERICAL AND CONE TYPEPROBES
105

Flow direction
3/4 in. extra heavy pipe

(stainless steel)
Eighty 1/16 in. holes

1/4 in. pitch
Screen (stainless steel)

No. 7 mesh, 0.065 in.
diameter wire
v
FIG. 4.14 BASKET
PROBE
/-Ball bearings
Protective gauze
Lattice of holes drilled
Stainless steel &Transducer

sphere
FIG. 4.15 BASKETPROBE WITH A PRESSURE TRANSDUCER
106

be metered for purging except when readings are be- where

ing made. Such purging may be used for any higher- c = velocity of sound
pressure connections for which asuitable uniform D = tube inside diameter
source of air or other suitable gas under pressure is L = tube length
available. The aboveequation assumes no damping as i s the case
Liquid-filled piping should slope upward continu- in most short-tube configurations. This equation pre-
ously from the instrument to the tap. If this is not possi- dictsfrequencies 2% to 5% above thosefound by more
ble, the high points must be vented, to eliminate gas rigorous analyses, Higher-order resonant frequencies
pockets, before reading the instrument. Gas-filled pip- can be calculated with more involved calculations.
ing should slope downward continuously from the in-
strument to the tap to automatically drain any liquid. If
this is not possible, the low points mustbedrained, be-
fore reading the insfrument. See Fig. 4.18. 4.4 DIAPHRAGM SEALS
A valved cross-connection between the two pipes of
Direct pressuremeasurement, using connecting
a differential pressure meter i s a frequent sourceof er-
lines to a gage or a transmitter,is sometimes hampered
ror. A leak through. the cross-connection reducesthe
by the nature of the process fluid. The process fluid
differential. Caution should be used with such an ar-
may be highly viscous, corrosive, contain slurries that
rangement. An alternate arrangement involves the
could plugthe interior of an instrumentortend tosolid-
cross-connection to include two valves, with a drain
ify, as in high-pressure polymer extruders.
valve between, as a telltale for leakage detection, as
In the food and dairy industry, ease of cleaning is a
shown in Fig. 4.19.
concern. Crevices or pockets in the instrumentation
must be avoided, for they trap materials that support
4'3.1 Transient Pressure Measurement.Measurement the growth of bacteria. Standardconstruction materials
of transient and average pressure in a flowing stream are often not acceptable.
may be required in many test and monitoring applica- Diaphragm seals isolate the process fluid from the
tions. The best way to makesuchameasurement measuringelementwhile communicating the pressure
would be to place the sensing transducer right at the signal through connecting liquid-filled capillary tub-
sensing location. However, many test configurations ing. The seal is normally a very flexible membrane,
will not allowsuch a test setup. In these cases, tubing such as an elastomeror thin metallicdiaphragm capa-
will be required to connect the sensinglocation to the ble oftransmitting the process-fluid pressure with mini-
sensingtransducer. In such cases the frequency re- mum pressure drop to the liquid inside the capillary.
sponse of the tubing must be considered so that thein- The liquid, in turn, pressurizes the measuring element.
fluence on the test results is known. There areno lines that require either draining orpurg-
In setting up connecting tubing one should avoid ing orrequire special precautionsto keep vertical runs
configurations that will cause wavereflectionsand completely fui1 or empty of process fluid.
hence system resonances. Sharp bends, sudden expan- Some applications of diaphragm seals areillustrated
sions, and sudden contractions of the connecting tube in Figs. 4.21, 4.22, and 4.23.
should be avoided. Seal elements are available for flush mounting to
The task of calculating the system resonances of a pipe flanges or chemical tees, welding to pipe, or ex-
tubing system can be quite complicated and involved. tending into the process. Only the seal membrane and
Refer to Refs. [33] through [39] for guidelines in per- trim come into contact with the process.
forming such a calculation. Several precautions and choices areinvolved in the
An example of a resonant-frequencycalculation of a use of seals. They involve size, pressure, temperature,
relatively simple connecting tubing configuration will filling liquid, capillary length and design, welded or
be considered (see Fig. 4.20). bolted construction, quality of fill, and elevation rela-
The equation stated below gives an adequateap- tive to the measuring element. Carefulengineering is
proximation of the resonant frequency f using the required, but when done, some bothersome measure-
quarter-wavelength equation with an end correction ment problems will be solved, and the system reliabil-
[331. ity will be extremely high.
4.4.1 Size. A seal membrane mustbe capable of trans-

ferring the required volume of liquid to actuate the
107

ASME P T C * L 7 * 2 8 7 U 0757670 0 0 5 2 2 0 2 LI m
Mach number = 0.9 I
25
20
S
(v'
>
P
x
15
L
E
h
E!
i
a
E!
10
Diameter Ratio, d / D
FIG. 4.16 MAGNITUDE OF PROBE-BLOCKAGEEFFECTS
108

ASME P T C * 1 9 * 2 87 M 07596700052203 b W
MEASUREMENT PRESSURE PTC 19.2-1 987
1.0
0.8
0.6
z
i
d
$
z
.c
S
0.4
0.2
0.1%
O 1 I I I
O 0.05 0.10 0.15 0.20
Diameter Ratio, d l 0
FIG. 4.17 MAGNITUDE O F PROBE-BLOCKAGEEFFECTS
I o9

Pipe
Gage
Liquid-Filled in Slope up
Gage Piping
Gas-Filled
Gage Piping
Pipe
FIG. 4.18 TYPICAL PRESSURE GAGE PIPING ARRANGEMENT
FIG. 4.19 DIFFERENTIAL PRESSURE CROSS CONNECTION
Connecting tube
Sensing
Sensing
transducer
FIG. 4.20 TUBE CONFIGURATION USED IN FREQUENCY CALCULATION
110

PRESSURE MEASUREMENT ASMVANSI PTC 19,2-1987
Liquid fill
Membrane
Measuring
element
FIG. 4.21 REMOTE SEAL WITH MEASURING ELEMENT FOR A

CACE PRESSURE TRANSMITTER
Membrane -/
FIG. 4.22 REMOTE SEAL WITH APRESSURE CACE

111

Wafer-Type Element
Welding Element
n
t"-l
U
Chemical Tee Element
Extended Diaphragm Element
FIG. 4.23 REMOTE SEALS FOR USE WITH ADIFFERENTIAL

PRESSURE TRANSMITTER %

measuring element through its full range. The seal the volume-transfer requirement, viscosityof the filling
membrane mustalso be able to accommodate changes liquid, and capillary bore size determine the response
in volume that result from thermal expansion or con- time. Thecapillary bore diameter should be the small-
traction of the filling liquid. est practical size that will still give a satisfactory re-
It should be kept in mind that any pressure change sponse time. This will provide the least sensitivity to
across the seal that is caused by expansion or contrac- ambient temperature changes.
tion of the fill fluid due to ambient or process- With some filling fluids such as mercury, tempera-
temperature changes represents an error i n the ture compensation of the capillary can be obtained
measurement. A perfect seal membrane would be like using low or zero-expansion filler wireinside thecapil-
a soap bubble in that itcould accommodatevolumetric lary bore.
changes in the system-filling liquid without signifi-
cantly changing the pressure of the liquid. While a
small-diameter seal diaphragm may be a convenience, 4.4.4 Welded or Bolted Construction. Several manu-
a larger one may perform better. Also, the lower the facturers offer seals designed to bolt onto most existing
pressure span, the larger the seal required for satisfac- measuringelements which havesuitablysmall
tory temperature performance. volume-transfer requirements. Theyusually provide a
means of adding the liquid fill. Their advantages in-
Diameter alone, however, canbe misleading. Corru-
gation design, material, and thickness are equally as clude the ability to modify existing measuring equip-
important. The performanceof a seal i s best described ment for more demanding applications and the ability
by a volume-displacement-versus-pressure plot. The to make repairs in the field.
slope of this plot i s capacitance (volume change/ Hermetically sealed, all-welded systems, however,
pressure change). Its value should be as high as possi- are less susceptible to leakage and require no mainte-
ble over the range of volume displacement required by nance. They are particularly suited to vacuum applica-
the measuring element. If there are discontinuities in tions in which bolted constructions would be subject
the curve, this is an indication of oil-canning, and a to long-term migration of air across nonmetallic parts.
seal with such a malfunction could cause large errors in
the measurement. 4.4.5 Quality of Fill. System filling is extremely impor-
tant. It must be free from air pockets that contract and
4.4.2 Filling liquid. Selection of a filling liquidshould expand with changes of temperature or applied pres-
take into account itscoefficient of thermal expansion, sure. A vacuum fill is recommended for optimum per-
compressibility, viscosity (including constancy of vis- formance.Whenthe system i s t ob e usedat
cosity over the working temperature range, since that temperatures below the filling temperature or for vac-
affects response time), its freezing and boiling points, uum service, additional tensioning fill must be
tendencies to decompose at the maximum operating added to keep the seal from hitting tt-e overrange stop
temperature, compatibility with the materials of the or the filling-liquid pressure from dropping below its
measuringelement and seal system, and its vapor pres- vapor pressure. No leakage can be tolerated, because
sure if it is to be used in a vacuum application. Density this would result in driftof the measurement andulti-
can alsobe important if the seal is to be located at a dif- mate failure of the system. Gasketedor threaded joints
ferent elevation from the measuringelement (see para. should be avoided.
4.4.6)or if there is danger that the liquid mass could
cause mechanical shock forcesduring handling oran
4.4.6 Elevation. Diaphragm-sealsystems aresubjectto
excessive static head that would damage the seal by
static-head effects when the measuring element and
overstressing. It should also be a safe material in con-
seal are mounted at different elevations. For a particu-
tact with the process in case of seal rupture. This i s par-
lartemperature, the static head can be corrected by the
ticularly important with sanitary seals in the food and
user during installation. This involves re-zeroing the
dairy industry and with certain chemical applications
measuring element.
where an incompatible filling fluidcan even cause ex-
Installing a seal at an elevation lower than the mea-
plosions (e.g., silicone and chlorine).
suring element is quite risky unless the seal element
has been specially designed to support the increased
4.4.3 Capillary. Capillary length is often dictated by head of filling liquidthat will be acting outward. This
the application, but when there is a choice, length risk i s even greater when vacuum measurements are
should be kept to a minimum. With length established, involved.
113

!
When seals are installed either above or below the Transducer

measuring element, it is also necessary to be aware of
theeffecfof changing density of the filling liquid
l
caused by ambient temperaturechange. This effect is U
"7-
different from the volumetric temperature effect and is
normally not compensated for by the instrument man-
ufacturer. Thermostatically-controlled heat tracing or
vertical capillary runs maybe warranted on outdoor in-
stallations where high accuracy is required. The den-
sity variation effect can be calculated as follows, I
(Cf)L (A T ) (100)
E =
S
where
E = effect of filling-fluid density changewith tem- FIG. 4.24 TRANSDUCERSETUP
perature in percent of instrument span
a = coefficientofexpansionoffillingfluidatmean
temperature temperature of the surroundings, and thenmaking the
G, = specific gravity offilling fluidat mean temper- required vertical run.
ature Estimating the required length of horizontal run can
L = elevation of seal element relative to instru- be done quite easily andreliably with an equation de-
ment body (or relative to the other seal ele- rived from one-dimensional heat transfer [40].
ment on instruments with two remote seals)
AT = ambient temperature change
S = instrument span
'O
4.5 INSTALLATION EFFECTS where

Most pressure indicatorsare located remotefrom the T = temperature of the.fluid in the tubing a dis-
pressure sensor,often at a different elevation. Thisdif- tance x from the sensor
ference in height introducesa bias in the pressure mea- To = temperature of the fluid at the pressure
surement for which there must be a correction factor sensorltap
introduced. A typical setup is shown in Fig. 4.24. K, = thermal conductivity of surroundings
The pressure at the transducerwill be less than that at K2 = thermal conductivity of tubing material
the centerline of the pipe by the amountTh, where y is b= tubing outer radius
the specificweight of the fluid inthe vertical leg, and h x = length of tubing
is the vertical heightfrom the centerline of the pipe to = ratio of inner to outer tubing radii
the center of pressure of the transducer. Normally this N, = Nusselt number
bias is small forgases and can be neglected unless very Here all temperatures have been referenced to a sur-
accurate measurements are required. rounding temperature of zero. A correlation for the free
To apply thiscorrection requires a knowledge of the convection from a horizontal tube has been given as
specificweightofthefluid inthevertical leg. Ifthefluid [41]:
is at a high temperature relative to the surroundings,
then therewill be a temperature gradientin the vertical
leg dueto heat transfer. Since the specific weight of any
fluid is a function of the fluid's temperature,there where G, is the Grashof number and P, is the Prandtl
would then be a variation in the specificweight in the number of the surrounding fluid, Using the Nusselt
vertical leg.Estimatingtheeffectivespecific weight number for free convection will ensurean ample
would be difficult if notimpossible. Thisproblem can length of tubing for all conditions. A comparison of the
be avoided byrunning the connecting tubing horizon- calculations for200"C (392F) waterin 1.27 cm in.) ('h
tally for a short distanceuntil the fluid has reached the tubes of stainless steel and steel is shown in Fig. 4.25.
114

ASME P T C * L 7 * 2 87 'M 0759b70 0 0 5 2 2 0 7 7 W
ASMVANSI PRESSURE MEASUREMENT PTC 19.2-1987
1.o
I I I I I
- ---
A
= 1 - D theory
0.8 = 2 - D theory
h 0.6
.-0'
i-
oc"
o
L 1/2 in. stainless steel tubing
3
i-
F?
o
0
0.4
I-
1/2 in. steel pipe
0.2
O
O 4 8 12 16 20
Axial Distance, x , in.
GENERAL NOTE: 1 in. = 25.4 mm
FIG, 4.25 TEMPERATURE DISTRIBUTIONS IN CONNECTING TUBING
/ Transducer
T,
sin e
FIG. 4.26 FLOW MEASUREMENTTRANSDUCER APPLICATION

115

- ASME P T C * 3 9 * 2 8 7 m
ASME/ANSI PTC 19.2-1987 PRESSUREMEASUREMENT
Also shown for comparison with Eq. (4.1) are the two- measurement is composed of two types of errors: ran-
dimensional calculations. The two calculations agree dom errors and bias errors. Random errors are variable
so well because the ratio of the thermalconductivities from measurementto measurement and are character-
of the water and thetubing material is small, 0.04 and ized by words like precision or repeatability.
0.02 for stainless steel and steel,respectively. The cal- Bias errors are fixed errors which do not vary from
culations indicate that approximately 30 cm (1 ft) measurement to measurement. It is the bias type of
and 60 cm (2 ft) of horizontal tubing are required for error which is reduced during calibration of an instru-
the fluid to attain essentially the temperature of the sur- ment. A detailed procedure for treating the measure-
roundings, The parameters of this exampleare typical ment errors will not be given here, but the reader i s
ofthoseencountered in flowmeasurements of hot con- referred to ASMVANSI PTC19.1 on Measurement
densate from a large steam turbine. Uncertainty which treats this subject thoroughly. The
The subject offlow measurement has an interesting present intent is to make the reader aware of the fact
application of these ideas because while a difference that there are various sources of measurement uncer-
between two pressures is measured, the two pressure tainty.
sensors are sometimes located at different elevations.
Figure 4.26 shows such anapplication.
If it is assumed that the temperatureof the fluid inthe 4.6.1 Error Propagation.Any measurementis the final
pipe is uniform at T, and that the horizontal legs of the step in a series of steps, a measurement chain. At the
connectingtubing are adequateto ensure that theverti- beginning of the chain i s the National Bureau of Stan-
cal legis at the temperature of the surroundings TS (dif- dards to which all important engineering measure-
ferentfrom T,), then therewill be a pressuredifference ments must be traceable. Each link in the chain adds to
between ports1 and 2 of the transducerwhen there is the final uncertainty of themeasurement. Startingwith
no flow across the fluid meter. The amount of this pres- the NBS reference, whichmight be a deadweight
sure difference is given by tester, for example, there is a small uncertainty contri-
bution to calibration of the transfer standard. The trans-
P1 - P2 = [rr(T,)- r(Ts)lh fer standard has an uncertainty as does the working
standard that is actually used to calibrate the pressure
measurement system used. Finally, the calibrated sys-
merely the differencein the specific weights y of the
tem itselfwill have someuncertaintyassociated with it.
fluid evaluatedat the two temperatures, times theverti-
It is usually thecase that the contribution of each link
caldistancebetweenthe tubing connections. This
increases with the smallestcontribution from the NBS
pressure difference is a bias which must be subtracted
reference and the largest from the calibrated system.
from all AP measurements across the fluid meter.
The measurement chain,as described, endswith the
instrument being calibrated in ideal laboratory condi-
tions (see Fig.2.1). To obtain uncertainty values for ac-
4.6 UNCERTAINTIES IN PRESSURE MEASURE-
tual pressuremeasurements, other variables such as
MENT installation effects, environmental effects, time varia-
All measurements, including pressure, have errors tion, and spatial variationsof the pressure being mea-
associated with them. Error isthe difference between a sured must be considered. Installation effectsare
measured value and the true value.difference
This can- described in Section 4, while the reader is referred to
not be known, but an estimate of its probable magni- manufacturers literature for environmental effects on
tude can be made. The estimate is usually referredto as specific instruments, and to ASMEIANSI PTC 19.1 for a
the measurement uncertainty. The uncertainty of a discussion of temporal and spatial-variationeffects.

SECTION 5 - REFERENCES
[l]Reynolds, W. C., Thermodynamics,SecondEdition, New York: McGraw-Hill, 1968,pp. 67-69,

168-171,226-227.
[2] ASME Orientation and Guide for Use of SI (Metric) Units, ASME Guide SI-1, Eighth Edition, 1978.
[3] ASTM Metric Practice Cuide, ASTM Guide E380-72, June 1972.
[4] Murdock, J. W., editor, PTC 2-1 971, Definitions and Values, New York: ASME, 1971
[5] Hewitt, G. F., The Role of Experiments in Two-Phase Systems with Particular Referenceto Measure-
ment Techniques, Progress in Heat and Mass Transfer, Vol. 6, Oxford: Pergamon Press, 1972, pp.
213-240.
[6] Heydemann, P. L. M., and Welch, B. E., Experimental Thermodynamics, B. LeNeindre and B. Vodar,
editors, Butterworth and Co.,Ltd., 1975.
[A Johnson, D. P., et al., Industr Eng Chem. 49, 2046 (1957).
[8] Lewis, S., and Peggs, G. N,, ThePressureBalance: A Practical Guide To Its Use, National Physical
Laboratory, United Kingdom, 1979.
[9] Dadson, R. S., Lewis, S. L., and Peggs, G. N., The Pressure Balance: Its Theory and Practice, Her Maj-
estys Stationary Office, London, England, 1982.
[lo] Pressure Gauge Handbook, New York: Marcel Dekker Inc., 1985.
[l11 ANSVASME B40.1-1985, Gauges- Pressure Indicating Dial Type- Elastic Element, New York: ASME,
1985.
[12] Newell, F. B., Diaphragm Characteristics, Design and Terminology, New York: ASME, 1958
[13] Lion, K. S., Instrumentation in Scientific Research, First Edition, New York: McGraw-Hill, 1959, p.44.
[14] ASME Fluid Meters, Sixth Edition, New York: ASME, 1971.
Il 51 Shaw, R., The Influence of HoleDimensions on Static Pressure Measurements, Trans. ASME, lou&al
o f Fluid Mechanics, Vol. 7, 1960, pp. 550-564.
[16] Rainbird, W. J., Errors in Measurement of Mean Static Pressure of a Moving Fluid Due to Pressure
Holes,Quart. Bull. Div. Mech. Engrg., Nat. Aero, Est., Nat. Res. Council, Canada. Rep. DMWNAE 1967
(31, PP. 55-89.
[17] Franklin, R. E., and Wallace, J. M., Absolute Measurements of Static-Hole ErrorUsing Flush Transduc-
ers, Trans. ASME, lournal of Fluid Mechanics, Vol. 42, Part 1, 1970, pp. 33 -48.
[18] Benedict, R. P., and Wyler, J. S., Analytical and ExperimentalStudies of ASME Flow Nozzles, Trans.
ASME, lournal of Fluid Engineering, September 1978.
[19] Benedict, R. P., Wyler, J. S., Analytical and ExperimentalStudies of ASME Flow Nozzles, ASME Paper
77 WNFM.
[20] Wyler, J. S., and Benedict, R. P., Comparisons Between Throat and Pipe Wall Tap Nozzles, Trans,
ASME, lournal ofEngineering for Gas Turbines and Power, 1975.
117

ASME P T C * & 7 * 2 8 7 M Q757b70 0 0 5 2 2 1 2 7 W
ASMVANSI PTC 19.2-1987 MEASUREMENT PRESSURE
[21] Benedict, R. P., The Plenum Inlet Discharge Coefficient of an ASME Nozzle in Flow, Its Measure-
ment and Control in Science and Industry, vol. 2, ISA, 1981, p. 363.
[22] Peto, J. W., and Pugh, P. G., The Effects of the Presence of Static Holes on the Measurement of Static
Pressures on Models at Supersonic Speeds, NP1 AERO Report 1292, March 1969.
[23] Jaivin, G. I., Effect of Hble Size on Pressure Measurements Made with a Flat-Plate Dynamic-Head
Probe, JPL-TR-32-617, Jet Propulsion Laboratory, California Institute of Technology, June 1964.
[24] Thrasher, L. W., and Binder, R. C., Influence of Compressibility on Cylindrical Pitot-Tube Measure-
ments, Trans, ASME, July 1950, pp. 647 - 650.
[25] Morrison, D. F., Sheppard, L. M., and Williams, M. J., Hole Size Effect on Hemisphere Pressure Distri-
butions, j , Roy, Aero. Soc., Vol. 71, Royal Aeronautical Society, April 1967, pp. 31 7- 319.
[26] Rayle, R. E., An Investigation of the Influence of Orifice Geometry on Static Pressure Measurements,
Masters Thesis, Massachusetts Institute of Technology, 1949.
[27] Benedict, R. P., Fundamentals ofTemperature,Pressure, and Flow Measurements;Third Edition, New
York: John Wiley & Sons, 1984.
[28] Wuest,W., Measurement of Flow Speed and Flow Direction by Aerodynamic Probes and Vanes,
AGARD No. 32, 1967.
[29] Chue, S. H. Pressure Probes for Fluid Measurement, Prog. Aerospace Sci., Vol. 16, no. 2, 1975.
[30] ASME PTC 12.2-1 983, Steam Condensing Apparatus, New York: ASME, 1983.
J.
[31] Todd, K. W., and Fallow, B., Erosion Control in the Wet Steam Turbine, froc. lnst. Mech.Engrs., Vol.
180, part 30, 1965-66, pp. 50- 63.
[32] Wyler, J. S., Probe Blockage Effectsin Free Jets and Closed Tunnels, Trans. ASME,
journal of Engineer-
ing for Gas Turbines and Power, October 1975, pp. 509 - 515.
[33] Nyland, T. W., Englund, D. R., and Anderson, R. C., /Frequency Response of Short Pressure Probes,
Instruments andControl Systems, August 1973.
C341 Nyland, T. W., Englund, D. R., and Anderson, R. C., On the Dynamicsof Short Pressure Probes: Some
Design Factors Affecting Frequency Response, NASA TN D-6151, February 1971.
C351 Bergh, H., and Tijdeman,H., Theoretical and ExperimentalResults for the Dynamic Response of Pres-
sure Measuring Systems, Rep. NLR-TR-F 238, National Aero and AstronauticalResearch Inst. Amster-
dam, January 1965.
[36] Nyland, T. W., Englund, D. R., and Gebben, V. D., System for Testing Pressure Probes Using a Simple
Sinusoidal Pressure Generator, NASA TM X-1 931, 1970.
[37] Grandall, I. B., Theory of Vibrating Systems and Sound, D. Van Nostrand, 1926.
[38] Doebelin, E. O., Measurement Systems: Application and Design, New York: McGraw-Hill, 1966.
[39] ASMEPTC 19.2-1 964, Pressure Measurement, New York: ASME, 1964.
[40] Kreith, F,, Principles of Heat Transfer, Second Edition, International Textbook Co., Scranton, PA, 1967,
p. 49.
[41] McAdams, W. H., Heat Transmission, Third Edition, New York: McGraw-Hill, 1964.
118

ASME P T C * 3 9 - 2 87. m 0 7 5 9 6 7 00 0 5 2 2 3 3 7 m
APPENDIX A
TABLES OF CORRECTIONS A N D GRAVITY DATA
Tables begin on the following pages.
119

ASME PTC*KL9.2 87 W 0757670 0 0 5 2 2 1 4 O m
TABLE A I
CORRECTIONS FOR TEMPERATURE TO BE APPLIED TO OBSERVED HEIGHT OF
MERCURY COLUMNS (ASSUMING BRASS SCALE IS TRUE AT 32OF)
Height of Mercury Height of Mercury

Attached Column, in. Attached Column, in.
Thermometer, rhermometer,
OF 1 20 30 OF 1 20 30
- 20 +0.004745 +0.095 +0.142 15 +0.001 546 3.0.031 f0.046

-19 ,004654 ,093 ,140 16 .O01455 .O29 .044
-18 .O04562 .o91 .137 17 .O01364 .O27 .041
-17 ,004470 .O89 .134 18 ,001273 .O25 ,038
-16 .O04379 .O88 .131 19 .O01 182 ,024 .035
-15 +0.004287 + 0.086 +0.129 20 +0.001091 e 0.022 +0.033

-14 .O041 95 .O84 .126 21 .001000 .o20 .030
-13 .O04104 ,082 .123 22 .o00909 .O18 .027
-12 ,00401 2 .O80 .120 23 .O00818 .O1 6 .025
-11 .O03920 .O78 .118 24 .O00727 .O1 5 .022
-10 +0.003829 +0.077 +0.115 25 +0,000636 + 0.01 3 "0.019

- 9 -003737 .O75 ,112 26 ,000545 .o1 1 .O1 6
- 8 .O03646 .O73 .lo9 27 .O00454 .o09 .O14
- 7 .O03554 .O71 $107 28 .O00363 .O07 .o1 1
- 6 ,003463 .O69 ,104 29 .O00272 .O05 .008
- 5 +0.003371 +0.067 +0.101 30 +0.000182 f0.004 +0.005

- 4 .O03280 .O66 .O98 31 .o00091 .o02 .003
- 3 .O03188 ,064 .O96 32 .oooooo .o00 .000
- 2 .O03097 ,062 .O93 33 -0.000091 - 0.002 -0.003
- 1 .O03006 .O60 .o90 34 .O001 82 ,004 ,005
O +0.002914 +0.058 +0.087 35 -0.000272 -0.005 -0.008

" 1 .O02823 .O56 .O85 36 ,000363 .O07 .o1 1
2 ,002732 .O55 .O82 37 .O00454 .o09 .O14
3 .O02640 .O53 .O79 38 .O00544 .o1 1 .O1 6
4 .O02549 .O51 .O76 39 ,000635 .O1 3 .o1 9
5 +0.002458 + 0.049 +0.074 40 -0.000726 -0.01 5 -0.022

6 .O02366 .O47 .O71 41 .O00816 .O16 ,024
7 .O02275 .O46 ,068 42 .O00907 ,018 .027
8 .O02184 .O44 .O66 43 .O00997 .o20 .030
9 .O02093 .O42 .O63 44 ,001088 .o22 .033
10 +0.002002 + 0.040 +0.060 45 -0.001 179 - 0.024 -0.035

11 .o0191o .O38 .O57 46 .O01269 .O25 .038
12 .U01 81 9 .O36 .O55 47 .O01360 .O27 .041
13 .O01 728 .O35 .O52 48 ,001450 ,029 .044
14 .O01637 .O33 .O49 49 ,001541 .O31 .046
121

' ASME PTCm37.2 6 7 W 0 7 5 7 b 7 0 0 0 5 2 2 3 5 2 W
ASMEIANSI PTC 19.2-1987 PRESSURE M E A S U R E M E N T
TABLE A I
CORRECTIONS FOR TEMPERATURE TO BE APPLIED T O OBSERVED HEIGHT OF
MERCURY COLUMNS (ASSUMING BRASS SCALE IS TRUE AT 32OF) (CONT'D)

Thermometer, Thermometer,
OF 1 20 30 OF 1 - 20 30
50 -0.001631 -0.049
-0.033 a5 -0.004786 -0.096 -0.144
51 .O52
.O01 721 .O34 a6 .o04875 ,098 .146
52 .O01812
.O54 .O36 87 .O04965 ,099 ,149
53 ,001902 ,038 ,057 8a .O05055 .IO1 ,152
54 .o01993 ,040 .O60 a9 .O05145 ,103 ,154
55 -0.002oa3 -0.042 -0.062 90 -0.005234 -0.105 -0.157

56 .o021.O65
74 .O43 91 .O05324 .lo6 .160
57 .O02264 ,045 ,068 92 ,005414 .lo8 .162
58 .O71
.O02354 .O47 93 .O05504 -110 .165
59 .O73
.O02444 .O49 94 ,005593 ,112 .I 68
60 -0.002535 -0.051 -0,076 95 -0.005683 -0.114 -0.170

61 .O79
.O02625 .O52 96 .O05772 ,115 ,173
62 .O02715 .O54 .o81 97 .o05862 ,117 -176
63 .o02805 ,056 .o84 98 .O05952 .119 .179
64 ,002a96 ,058 ,087 99 .O06041 ,121 .181
65 -0.002986 -0.060 -0.090 1O0 -0.006131 -0.123 -0.184

66 .O03076 .O62 ,092 1o1 ,006220 -124 .1a7
67 ,0031.O95
66 .O63 102 ,00631O .126 .I a9
68 ,003256 .O65 ,098 103 ,006399 .12a .192
69 .O03346 .O67 .loo 104 .O06489 .130 .195
70 -0.003436 -0.069 -0.103 105 -0.00657a -0.132 -0.197

71 .O03526 ,071 ,106 106 ,006668 ,133 ,200
72 .O03616 .O72 ,108 107 ,006757 ,135 ,203
73 .O03707 .O74 .111 1oa ,006847 .137 .205
74 .114
.O03797 ,076 1o9 ,006936 ,139 ,208
75 -0.003887 -0.078
-0.1 17 110 -0,007025 -0.141 -0.21 1
76 .O03977 .oao ,119 111 .O071 15 ,142 ,213
77 .O04067 .O81 .122 112 .O07204 .144 -216
78 .O04156 .O83 .125 113 ,007293 .146 ,219
79 ,004246 .o85 ,127 114 .O07383 .I 48 .221
ao -0.004336 -0.087 -0.130 115 -0,007472 -0.149 -0.224

81 ,004426 .O89 .133 116 .O07561 ,151 .227
a2 ,004516 .o90 .135 117 ,007650 .153 ,230
83 ,004606 .O92 ,138 118 ,007740 ,155 .232
84 ,004696 .O94 ,141 119 .O07829 .157 .235
120 .O07918 ,158 .238
122

ASME P T C * L S * 2 8 7 W 0 7 5 7 6 7 0 00522Lb 4 m
UREMENT PRESSURE PTC 19.2-1987
TABLE A2
MERCURY COLUMNS (ASSUMING BRASS SCALE I S TRUE AT 62OF)

Thermometer, Thermometer, ~~ . ~
OF 1 20 30 OF 1 20 30
- 20 +0.004439 3.0.089 +0.133 15 + 0.001240 +0.025 +0.037

-19 ,004347 .O87 .130 16 .O01 149 .O23 .034
-18 .O04255 .O85 .128 17 .O01058 .O31 .032
-17 .O041 64 .O83 .125 18 .O00967 .o1 9 ,029
- 16 .O04072 .O81 .I22 19 .O00876 .O1 8 .026
-1.5 . +0.003981 + 0.080 "0.1 19 20 +0.000785 + 0.016 +0.024

-14 ,003889 ,078 .117 21 ,000694 .O14 .o2 1
-13 ,003797 .O76 ,114 22 ,000603 .o12 .O18
-12 -003706 .O74 ,111 23 .O0051 2 .o1o .O1 5
-11 ,003614 .O72 ,108 24 ,000421 .O08 .O1 3
-10 +0.003523 +0.070 +O.T06 25 +0.000330 + 0.007 +0.010

- 9 .O03431 .O69 .lo3 26 ,000239 ,005 .007
- 8 ,003339 .O67 .1O0 27 .O00148 .O03 ,004
- 7 ,003248 .O65 ,097 28 +0.000057 f0.001 +0.002
- 6 .O03157 ,063 .O95 29 -0.000034 -0,001 -0.001
- 5 +0.003066 "0.061 +0.092 30 - 0.0001 24 -0.002 -0.004

- 4 .O02974 ,059 .O89 31 .o00215 .O04 .006
- 3 .O02882 ,058 ,086 32 .000306. ,006 .009
- 2 .O02791 .O56 .O84 33 .O00397 .O08 .o12
- 1 .O02699 .O54 .O81 34 .O00488 ,010 .O1 5
O +0.002608 +0.052 "0.078 35 -0.000578 -0.012 -0.01 7

1-1 .O02517 .O50 .O76 36 .O00669 .o1 3 .020
2 .O02425 .O49 .O73 37 ,000760 .O1 5 .023
3 .O02334 .O47 .O70 38 .O00850 .O1 7 .026
4 ,002243 .O45 .O67 39 .O00941 .o1 9 .028
+0.002151 +0.043 +0.065 40 -0.001032 -0.021 -0.031

.O02060 ,041 ,062 41 .o01 122 .o22 .034
.O01 969 .O39 .O59 42 .o01213 .O24 .036
.O01878 .O38 .O56 43 ,001 303 .O26 .039
.O01 787 ,036 .O54 44 ,001 394 .O28 .042
10 +0.001695 +0.034 +0.051 45 -0.001485 - 0.030 -0.045

11 .O01604 .O32 ,048 46 .0Q1575 ,031 .047
12 .O01 51 3 .O30 ,045 47 .O01665 .O33 .050
13 ' .O01422 .O28 .O43 48 .O01 756 ,035 . .O53
14 ,001331 .O27 .O40 49 ,001846 .O37 .055
123

ASME P T C * L S - 2 87
~~ m 0 7 5 7 b 7 0 OOS2217 b
TABLE A2
MERCURY COLUMNS (ASSUMING BRASS SCALE IS TRUE AT 62'F) (CONT'D)

Thermometer, Thermometer,
OF 1 20 30 OF 1 20 30
50 -0.001 937 -0.039 -0.058 85 -0.005091 -0.102 -0.153

51 ,002027 .O41 ,061 86 .O05181 .lo4 .155
52 .O021 18 .O42 .O64 87 ,005271 ,105 -158
53 .O02208 .O44 .O66 88 .O05360 ,107 ,161
54 ,002298 ,046 ,069 89 .O05450 ,109 64 -1
55 -0.002389 - 0.048 -0.072 90 -0.005540 -0.111 -0.166

56 .O02479 .O50 .O74 91 ,005630 ,113 .169
57 .O02570 ,051 .O77 92 .O05719 .114 -172
58 .O02660 .O53 .O80 93 .O05809 ,116 -174
59 .O02750 ,055 .O83 94 .O05899
.177 ,118
60 -0.002840 -0.057 -0.085 95 -0.005988 -0.120 -0.180

61 .O02931 ,059 .O88 96 .O06078 .122 ,182
62 .O03021 ,060 ,091 97 .O061 68 ,123 .185
63 .O031 11 .O62 .O93 98 .O06257 ,125 ,188
64 .O03201 .O64 .O96 99 ,006347 .127 .190
65 "0.003291 -0.066 - 0,099 1O 0 -0.006436 -0,129 -0.193

66 .O03382 ,068 ,101 101 .O06526 .131 .196
67 .O03472 .O69 .lo4 102 .O06615 ,132 ,198
68 .O03562 .O71 .lo7 103 ,006705 .134 ,201
69 ,003652 .O73 *110 104 .O06794 .136 .204
70 -0.003742 -0.075 -0.112 105 - 0.006884 -0.207

-0.138
71 .O03832 .O77 .115 106 .O06973 .139 .209
72 .O03922 ,078 .118 107 .O07063 ,141 ,212
73 .O040 12 .O80 .120 108 ,007152 .143 .215
74 ,004102 .O82 .123 1o9 ,007241 .145 .2 1 7
75 -0.004192 - 0.084 -0.126 110 -0.007331 -0.147 -0.220

76 .O04282 .O86 .128 111 .O07420
,223 ,148
77 .O04372 .O87 .131 112 .O07509
,225 .150
78 .O04462 .O89 .134 113 .O07599 ,152 ,228
79 .O04552 ,091 .137 114 .O07688 .154 .231
80 -0.004642 - 0.093 -0.139 115 -0.007777 -0.233

-0.156
81 ,004732 .O95 ,142 116 .O07866 ,157 ,236
82 .O04822 ,096 ,145 117 .O07956 -159 .239
83 .O04912 ,098 ,137 118 .O08045 ,161 .241
84 .O05001 .loo ,150 119 ,008134 .163 .244
120 ,008223 .164 ,247
124

ASNE P T C * L 7 * 2 8 7 m 0757670 0 0 5 2 2 1 8 8 W
REMENT PRESSURE 19.2-1987
TABLE A3
DIFFERENCE BETWEEN g,, THE GRAVITATIONAL ACCELERATION AT MEAN
SEALEVEL FOR GIVEN LATITUDE @), AND STANDARD GRAVITATIONAL
ACCELERATION, (90 = 32.1740 ft/sec2)
Table gives (g,-
go)as a function of latitude.
Latitude &!r - Latitude h-gp latitude @I -gp

Wsec deg. Wsec deg. ftlsec deg.
O -0.0863 30 -0.0440 60 +0.0409

,0862 1 31 .O414 61 .0434
,0861 2 32 .O388 62 .0459
0.858 3 ,0361 33 ,0483 63
,0854 4 34 .O334 64 .0507
5 -0.0306
-0.0850 35 +0.0530 65
6 .O844 36 ,0278 66 .0552
,0838 7 37 ,0250 67 .0574
8 .o221 ,0830 38 68 .0595
1 ,082 9 39 .O1 92 69 5 ,061
10 -2 0.081 40 63 -0.01 70 +0.0635

11 ,0801 41 ,0654 .O1 34 71
12 .O790 42 .O1 05 72 .0672
13 ,0777 43 ,0689 .O075 73
14 ,0764 44 ,0046 74 .0705
-0.0749 15 45 -0.0016 75 +0.0720

16 .O734 46 f0.0013 76 .0735
17 .O718 47 .O043 77 ,0748
18 ,0701 48 ,0073 78 .0760
19 .O683 49 .o102 79 .0772
20 -0.0665 50 +0.0131 80 +0.0783

21 .O646 51 .O1 60 81 .0793
22 .O625 52 .O1 89 82 ,0801
23 .O605 53 .O2 18 83 .0809
24 ,0583 54 .O246 84 .O81 6
-0.0561 25 +0.0274 55 85 +0.0821

26 .O538 56 ,0302 86 ,0826
27 ,0514 57 ,0329 87 .0830
28 .O490 58 ,0356 88 .0832
29 .O465 59 .O383 89 .0834
90 .0834
GENERAL NOTE: gl is computed on basis of Meteorological Gravity System:
g, = 32.1 724 (1 - 0.0026373 cos 2 4 + 0.0000059 cos2 2 $Il in Wsec where 4 = latitude.
125

TABLE A4
FREE-AIR GRAVITY CORRECTION, Cf = 0.000003086 H FT/SEC2, WHERE
BAROMETER ELEVATION ( H ) IS IN FEET ABOVE MEAN SEALEVEL.
CORRECTION Cf IS T O BE SUBTRACTED FROM (gl - 8 0 ) .
Barometer Barometer
Elevation, Elevation,
H, c,, H, Cf,
ft ftlsec ft ftlsecZ
- 1,000 -0.0031 10,000 0.0309
O 0.0000 1 1,000 0.0339

+ 1,000 +0.0031 12,000 ,0370
2,000 .O062 13,000 .0401
3,000 ,0093 14,000 ,0432
4,000 .O123 15,000 ,0463
5,000 0.0154 16,000 0.0494

6,000 .O185 17,000 ,0525
7,000 .O2 16 18,000 .0555
8,000 .O247 19,000 .0586
9,000 .O2 78 20,000 ,0617
126

ASME P T C * : 1 9 * 2 8 7 0757b70 0052220 b
MEASUREMENT PRESSURE PTC 19.2-1987
TABLE A5
CORRECTION OF HEIGHT OF COLUMN TO STANDARD GRAVITY,
B (91 gdgo -
Observed heightof manometer or barometer
@l-
Wsec 1 20 30
-0.1500 -0.004662 -0.0932 -0.1399

-0.1400 -0.004351 -0.0870 -0.1305
-0.1 300 -0.004041 -0.0808 -0,1212
-0.1200 -0.003730 -0.0746 -0.1 119
-0.1100 -0.003419 -0.0684 -0.1026
-0.1000 - 0.0031 08 -0.0622 -0.0932

-0.0900 -0.002797 -0.0559 -0.0839
- 0.0800 -0.002486 -0.0497 -0.0746
-0,0700 -0.002176 -0.0435 -0.0653
-0.0600 -0.001665 -0.0373 -0.0559
-0.0500 -0.001554 -0.031 1 - 0.0466

-0.0400 -0.001243 -0.0249 -0.0373
-0.0300 -0.000932 -0.0186 -0.0280
-0.0200 -0.000622 -0.0124 -0.0186
-0.0100 -0.00031 1 -0.0062 -0.0093
0.0000 0.000000 0.0000 0.0000
+0.0100 +0.000311 +0.0062 +0.0093

+0.0200 +0.000622 +0.0124 +0.0186
f0.0300 +0.000932 4-0.01 86 +0.0280
+0.0400 +0.001243 +0.0249 +0.0373
f0.0500 3-0.001554 f0.0311 +0.04-66
+0.0600 +0.001865 +0.0373 +0.0559

+0.0700 +0.002176 +0.0435 +0.0653
+0.0800 +0.002486 f0.0497 + 0.0746
+0.0900 +0.002797 +0.0559 3-0.0839
+0.1000 +0.003108 +0.0622 +0.0932
GENERALNOTES:
(a) Tabular values give the correction for gravity, B(gl - go)lgoas a function of (gr - go)and B, where
gl = local gravitational acceleration
go = standard gravitational acceleration
B = observed height of column of liquid inbarometer or manometer
(b) Since the ratio (gl - go)/gois dimensionless, the correction givenin table always is in terms of the
units used for B. This permits table to be applied forany units, for example mm as well as inches,
127

ASMEIANSI PTC 19.2-1987 PRESSURE MEASUREMENT
TABLE A6
GRAVITY DATA
latitude (N) longitude 0 Elevation, Meteorological

Gravity, g
location deg. min deg. min ft ft/sec2 @ - g0)ko
Alabama
Birmingham 33 30.8 86 48.8 587 32.1 367 -0,001 161
Spring Hill 30 43.1 88 09.1 2 04 32.1 302 -0,001362
Wetumpka 32 31.2 86 12.4 202 32.1355 -0.001197
Arizona
Phoenix 33 26.6 112 06.8 1067 32.1349 -0.001218
Tucson 32 14.8 110 50.1 2546 32.1270 - 0.001462
Winslow 35 00.4 110 37.4 4867 32.1278 - 0.001 438
Arkansas
Fort Smith 35 23.3 94 25.5 442 32.1423 -o.ooogaa
Little Rock 34 44.9 92 16.4 292 32.1427 -0.000973
California
Burbank 34 11.1 118 I 8.9 61 2 32.1385 -0.001 106
Fresno 36 46.6 119 50.2 300 32.1465 -0.o00856
Highland 34 07.5 117 12.5 1288 32.1347 -0.001223
lsleton 38 10.2 121 35.6 20 32.1 509 -0.00071 9
Long Beach 33 46.3 118 11.6 27 32.1 393 - 0,001080
Maricopa 35 03.8 119 24.0 842 32.1 388 -0.001096
Palo Alto 37 26.6 122 09.7 48 32.1 504 -0.000735
Pasadena 34 08.1 118 07.6 750 32.1379 - 0.001 123
Sacramento 38 34.8 121 29.8 19 32.1525 - 0.000670
San Diego 32 42.8 117 09.9 22 32.1364 -0.001 170
San Francisco 37 47.5 122 25.7 375 32.1508 -0.000724
Sisson 41 18.3 122 19.6 3438 32.1510 -0.00071 7
Tehama 40 01.6 122 07.2 214 32.1 559 -0.000564
Ventura 34 16.8 119 17.6 78 32.1 386 -0.001 1O 0
Colorado
Colorado Springs 38 50.8 104 49.5 6043 32.1 352 - 0.001 208
Denver 39 40.6 104 57.1 5379 32.1391 -0.001087
Falfa 37 14.1 107 47.5 6960 32.1271 -0.001458
Grand Junction 39 04.2 1o8 33.9 4587 32.1 399 -0.001063
Connecticut
Hartford 41 44.8 72 41 .a 123 32.1 630 -0.000345
New Britain - Bristol 41 41 .O 72 58.1 640 32.1621 -0.000370
New Haven 41 19.1 72 55.4 69 32.1 622 -0.000368
Orange 41 18.2 72 59.5 299 32.1623 - 0.000366
Wolcott 41 33.4 72 58.0 32.1620 - 0.000374
Delaware
Dover 39 09.7 75 32.0 38 32.1 552 -0.o00587
Wilmington 39 43.0 75 18.2 88 32.1572 -0.000525
District of Columbia
Washington 38 49.4 77 01.5 23 32.1554 - 0.000579
Florida
Italia 30 37.0 81 43.1 21 32.1 320 -0.001 306
Leesburg 28 48.6 81 53 98 32.1268 - 0,001 468
Miami 25 42.8 ao 15.1 23 32.1202 -0,001673
Monticello 30 32.7 a3 51.1 158 32.1309 - 0.001 340
Pensacola 30 24.5 a7 12.9 6 32,1309 -0.001341
Riverview 27 52.4 a2 20.4 20 32.1253 -0.001515
St. Petersburg 27 48.9 82 40.2 49 32.1 253 -0.001516
West Palm Beach 26 42.8 ao 02.8 7 32.1233 -0,001 576
128

ASME P T C * l t 7 - 2 8 7 M 0 7 5 7 6 7 0 0 0 5 2 2 2 2 T
TABLE A6 GRAVITY DATA (CONTD)
latitude (N) Meteorological

Elevation,
Gravity,
longitude g
location min deg.
deg. min ft ft/sec2 @ - digo
Georgia
Atlanta 33 45.3 84 23.5 1063 32.1363 -0.001 174
Augusta 33 54.8 82 18.0 532 32.1396 -0.001072
Columbus 32 27.0 84 57.6 241 32.1363 -0.001 175
Macon 32 50.3 83 37.6 326 32.1 372 -0.001 145
Savannah 32 54.9 80 40.0 79 32.1377 -0.001129
Idaho
Boise 43 37.2 116 12.3 2697 32.1 589 -0.000472
Sandpoint 48 16.4 116 33.3 2090 32.1742 +0.000005
Illinois
Chicago 41 47.4 87 35.9 597 32.1610 -0.000405
Dixon 41 49.9 89 27.6 765 32.1 609 -0.000410
Keithsburg 41 06.4 90 56 546 32.1 588 -0.000473
Springfield 39 47.7 89 39.5 599 32.1 548 -0.000598
Streator 41 09.1 88 49.5 623 32.1 592 -0.000461
Urbana 40 06.7 88 13.6 728 32.1 557 -0.000571
Indiana
Angola 41 37.7 85 00.6 1043 32.1 599 -0.000439
Indianapolis 39 45.9 86 08.8 713 32.1 549 -0.000597
Princeton 38 19.2 87 34.0 472 32.1513 -0.000707
Salem 38 35.2 86 08.7 807 32.1510 -0.00071 8
Terre Haute 39 28.7 87 23.8 495 32.1 543 -0.000615
Wabash 40 47.9 85 49.6 683 32.1573 -0.000522
Iowa
Davenport 41 06.4 90 56 546 32.1588 -0.000473
Dallas Center 41 41.O 93 57.8 1067 32.1583 - 0.000490
Iowa City 41 39.6 91 32.2 697 32.1600 -0.000436
Osage 43 16.8 92 48.5 1167 32.1630 - 0.000343
sioux city 42 23.0 97 19 1690 32.1 596 -0.000448
Kansas
Boyle 39 20.0 95 19.8 1129 32.1513 -0.000706
Kansas City 39 05.8 94 35.4 913 32.1516 -0.000699
Newton 38 04.5 97 18.0 1496 32.1474 -0.000828
St. Marys 39 12.4 96 05.4 965 32.1515 -0.000702
Kentucky
Danville 37 38.9 84 46.4 984 32.1471 -0.000836
Louisville 38 35.2 86 08.7 807 32.1510 -0.00071 8
Louisiana
Baton Rouge 30 26.8 91 09.6 39 32.1305 -0.001352
New Orleans 29 56.9 90 04.3 8 32.1297 -0.001379
Sarepta 32 53.5 93 27.1 260 32.1373 -0.001 143
Maine
Fort Kent 47 15.0 68 35.6 524 32.1 770 +0.000092
Rockland 44 06.3 69 06.9 31 32.1695 -0.000142
Maryland
Baltimore 39 17.8 76 37.3 1O0 32.1 555 -0.000575
Massachusetts
Boston 42 21.6 71 03.8 72 32.1 649 -0.000285
Bridgewater 41 58.4 70 58.5 75 32.1633 -0.000333
Cambridge 42 22-8 71 07.8 46 32.1650 -0.000282
Fall River 41 41.9 71 09.4 154 32.1 619 -0.000376
Great Barrington 42 11.2 73 18.8 971 32.1610 -0.000405
Lawrence 42 45.8 71 27.5 105 32.1 657 -0.000260
Lowell 42 45.8 71 27.5 105 32.1657 -0.000260
-
Springfield Holyoke 42 05.0 72 34.4 171 32.1 635 -0.000327
Worcester 42 16.5 71 48.5 558 32.1 626 -0.000357
129

ASME P T C * L 7 - 2 8 7 E 0753b70 0 0 5 2 2 2 3 L H
ASME/ANSl PTC 19.2-1987 PRESSURE MEASUREMENT
TABLE A6 GRAVITY DATA (CONT'D)
latitude (N) longitude (W)

~ l ~ ~ Meteorological
~ ~ Gravity,
i ~ ~g ,
deg. min deg. location 63 - dkl
Michigan
Bay City 43 39.9 83 54.2 584 32.1664 -0,000239
Grand Rapids 42 58.0 85 39.5 774 32.1641 - 0.000309
Mason 42 34.6 84 25.8 900 32.1623 -0.000365
Seney 46 20.8 85 57.6 733 32.1 744 +0.000010
Three Rivers 41 56.5 85 38.1 790 32.1614 -0.000395
Traverse City 44 45.8 85 37.2 591 32.1699 -0.000127
Minnesota
Duluth-Superior (Wis.) 46 47.0 92 06.4 708 32.1 768 +0.000085
Minneapolis-St. Paul 44 58.7 93 13.9 840 32.1 71 5 -0.000080
Mississippi
Forest 32 21 .8 89 27.7 473 32.1 348 -0.001219
Missouri
Columbia 38 56.2 92 19.8 74 7 32.1512 -0.000709
Forsyth 36 41 .O 93 06.2 677 32.1457 -0.000881
Kansas City 39 05.8 94 35.4 913 32.1516 -0.000699
St. Joseph 39 20.0 95 19.8 1129 32.1513 -0,000706
st. Louis 38 38.0 90 12.2 505 32.1519 -0.000687
Montana
Boulder 46 14.2 112 07.3 4898 32.1 602 -0.000431
Hinsdale 48 23.8 107 05.3 2169 32.1761 +0.000065
Miles City 46 24.2 1os 50 2356 32.1 696 -0.000139
Nebraska
Dorchester 40 39.3 97 10.2 1496 32.1 543 -0.000614
Fremont 41 27.1 96 34.1 1217 32.1 579 -0.000503
Hershey 41 04.8 1O 0 57.5 3091 32.1518 -0.000691
Randolph 42 23.0 97 19 1690 32.1 596 -0.000448
Nevada
Boulder Dam 36 01.7 114 46.9 1209 32.1 409 -0.001031
Ely 39 14.9 114 53.4 6437 32.1 355 -0,001 197
Reno 39 32.4 119 48.8 4584 32.1 416 - 0,001 o1o
Winnemucca 40 58.4 117 43.8 4301 32.1468 - 0.000847
New Hampshire
Lancaster 44 29.5 71 34.3 859 32.1 678 -0.000193
Nashua 42 45.8 71 27.5 105 32.1 657 -0.000260
New Jersey
Atlantic City 39 21.9 74 25.0 10 32.1 556 -0.000574
Bordentown 40 10.1 74 41.4 57 32.1 586 -0.000479
Carnden 39 57.1 75 11.7 52 32.1583 -0,000488
Glen Ridge 40 48.1 74 12.2 191 32.1591 -0,000466
Hoboken 40 44 74 02 35 32.1 607 -0.000414
Plainville 40 28.0 74 40.6 114 32.1577 -0,000508
Swedesboro 39 43.0 75 18.2 88 32.1572 -0.000525
New Mexico
Abiquiu 36 12.0 106 15.0 5915 32.1275 -0.001446
Albuquerque 35 05.1 106 37.4 5156 32.1 262 -0.001486
Nara Vista 35 38.4 103 03.2 4213 32.1328 -0.001281
Roswell 33 23.6 104 30.7 3565 32.1272 -0.001455
130

ASME P T C * L 9 - 2 87 W 0759670 005222LI 3 m
MEASUREMENT PRESSURE ASMVANSI PTC 19.2-1987
TABLE A6 GRAVITYDATA(CONTD)
latitude (N) longitude (W Elevation,Meteorologicar

Gravity, g
location deg. min deg. min ft fuse2 (i-
govgo
?
New York
Albany 42 39.1 73 46.1 200 32.1632 -0.o0033a
Buffalo 42 57.1 78 49.3 689 32.1637 -0.000321
lthaca 42 27.1 76 29.0 alo 32.161 7 -0.o003a2
Little Falls 43 02.7 74 51.2 448 32.1642 -0.000307
New York City 40 48.5 73 57.7 725 32.1607 -0.000416
Schenectady 42 47.8 74 02.3 702 32.1631 -0.000342
Troy 42 40.6 73 30.7 945 32.1614 -0.000393
Whitehall 43 33.0 73 2x8 138 32.1660 -0.000251
Wilson 43 18.4 78 49.6 285 32.1 660 -0,000249
North Carolina
Asheville 35 35.9 82 33.3 21 99 32.1 389 - 0.001 093
Charlotte 35 13.8 80 50.8 748 32.1429 -0,000967
Durham 36 00.2 78 56 41 3 32.1465 -0.000857
Raleigh 35 47.5 78 40.3 410 32.1446 -0,00091 6
Wilmington 34 14.2 77 56.6 28 32.1409 -0.001030
Winston-Salem 36 06.1 80 14 932 32.1427 -0.000976
North Dakota
Bismarck 46 48.5 1O0 47.1 1688 32.1 724 -0,000051
Fargo 46 53.4 96 48.0 900 32.1 753 +o.o0003a
Ray 48 20.4 103 09.8 2269 32.1 762 +0.000066
Ohio
Cincinnati 39 08.3 84 25.3 ao4 32.1520 -0.000684
Cleveland 41 30.4 a1 36.6 689 32.1598 -0.000443
Columbus 39 57.8 a2 59.4 758 32.1 548 -0.000598
Dayton (Wrighf Field) 39 46.6 84 05.9 813 32.1 549 -0.000596
Dover 40 31.9 a1 28.4 906 32.1564 -0.000548
Kent 41 09.3 81 21.3 1168 32.1 577 -0.000507
Oberlin 41 17.5 82 13.2 814 32.1586 -0.000479
Steubenville 40 04.0 80 43.3 673 32.1 547 -0.000602
Tiffin 41 06.9 83 10.0 763 32.1 583 - 0.000490
Oklahoma
Ardmore 34 13.2 97 07.3 828 32.1 385 -0.001 103
Guymon 36 40.7 1o1 28.7 3113 32.1378 -0.001 126
Oneta 36 o1 .o 95 42 .O 709 32.1438 - 0.000940
Seminole 35 14.7 96 34.5 951 32.1413 -0.001017
Oregon
Eugene 44 02.7 123 05.6 423 32.1 680 -0,0001 89
Glendale 42 44.2 123 25.8 1424 32.1618 -0.000380
Heppner 45 21.4 119 33.2 1962 32.1 662 -0.000243
Portland 45 31.4 122 40-7 26 32.1731 -0.000030
Pennsylvania
Allentown - Bethlehem 40 36.5 75 22.6 328 32.1 569 -0.000532
Chester 39 50.7 75 23.9 95 32.1578 -0.000504
Erie 42 07.8 ao 04.8 650 32.1610 -0.000405
Harrisburg 40 16.0 76 53.1 340 32.1 565 -0.000547
Philadelphia 39 57.1 75 11.7 52 32.1 583 -0.000488
Pittsburgh 40 27.4 80 00.6 771 32.1 558 -0.000568
Pocono Lake 41 06.2 75 29.4 1698 32.1560 -0.000561
Reading 40 19.2 75 53.9 753 32.1551 -0.000590
University Park 40 47.9 77 51.8 1174 32.1 560 -0.000562
Rhode Island
Cranston 41 47.0 71 27.1 66 32.1 624 -0.000361
Sooth Carolina
Charleston 32 47.2 79 56.0 20 32.7370 -0,001 151
McCormick 33 54.8 82 18.0 532 32.1 396 -0,001 072
Walterboro 32 54.9 80 40.0 79 32.1377 -0.001 129
131

TABLE A6GRAVITYDATA(CONTD)
latitude (N) Longitude (W)

Elevation,
Meteorological
Gravity, g
location deg. min ft/sec2 deg. ft min @ - g0)lgo
South Dakota
Canton 43 17.7 96 35.4 1229 32.1635 -0.000328
Pierre 44 21.9 1O0 20.8 1190 32.1659 -0.000253
Rapid City 44 05.7 103 16.9 3396 32.1 605 -0.000420
Tennessee
Cleveland 35 09.4 84 52.9 863 32.1404 -0.001046
Columbia 35 36.7 87 02.5 679 32.1440 -0.000934
Knoxville 35 5 7.7 83 55 919 32.1425 -0.000982
Memphis 35 08.7 90 03.3 263 32.1434 -0,000953
Texas
Amarillo 35 38.4 103 03.2 421 3 32.1328 -0.001281
Austin 30 16.5 97 44.3 539 32.1286 -0.001413
Beaumont - Port Arthur 30 05.2 94 06:O 18 32.1297 -0.001 378
Corpus Christi 27 49.2 97 23.3 4 32.1238 -0.001563
Crosbyton 33 22.9 101 33.9 3048 32.1296 -0.001381
Del Rio 29 21.8 1O0 53.2 956 32.1253 -0.001514
El Paso 31 46.3 106 29.0 3 760 32.1232 -0.001 582
Galveston 29 18.2 94 47.5 8 32.1 281 -0.001430
Grand Saline 32 40.1 95 45.2 469 32.1 360 -0.001 184
Houston 29 41.4 95 25.1 59 32.1285 -0.001415
Kerrville 30 02.0 99 08.2 1609 32.1262 -0.001488
Laredo 27 30.5 99 31.2 412 32.1218 -0.001625
Marlin 31 18.3 96 53.1 395 32.1329 -0.001278
Port Isabel 26 04.7 97 12.5 10 32.1216 -0.001 630
San Angelo 31 28.7 1O0 24.7 1867 32.1279 -0.001436
Sweetwater 32 28.5 1O0 23.7 2136 32.1289 - 0.001 403
Vernon 34 09.1 99 16.3 1196 32.1362 -0.001 177
Utah
Salt Lake City 40 46.1 111 53.8 4337 32.1454 -0.000889
Vermont
Montpelier 43 33.0 73 23.8 138 32.1 660 -0.000251
North Hero 44 49.1 73 17.5 115 32.1712 -0.000089
Virginia
Clifton Forge 37 49.1 79 49.6 1066 32.1468 -0.000847
Newport News - Hampton 37 02.5 76 25.8 20 32.1484 -0.000798
Oakwood 36 55.7 76 14.6 14 32.1483 -0.000801
Richmond 37 32.2 77 26.1 98 32.1506 -0.000729
Washington
Olympia 47 03.4 122 52.7 62 32.1 790 +0.000153
Seattle 47 36.5 122 19.8 243 32.1 757 +0.000051
Spokane 48 16.4 116 33.3 2090 32.1 742 +0.000005
Takoma 47 15.2 122 26.3 85 32.1 778 +0.000115
West Virginia
Charleston 38 20.9 81 37.7 604 32.1498 -0.000754
Wheeling - Weirton 40 04.0 80 43.3 673 32.1 547 -0.000602
Wisconsin
Franksville 42 44.2 87 54.3 749 32.1629 -0.000347
Madison 43 04.6 89 24.0 886 32.1639 -0.0003 16
Oconto 44 53.2 87 52.0 594 32.1694 -0.000146
Wyoming
Bufora 41 07.4 105 18.3 7861 32.1 398 -0.001066
Lander 42 50.0 108 43 5365 32.1491 -0.000776
Sheridan 44 48.0 106 58.7 3773 32.1 598 - 0.000443

C ASME/ANSI
PRESSURE MEASUREMENT 987
APPENDIX B
COMMONLY USED CONVERSION FACTORS
Conversion
Multiplication
Quantity From To Factor
Acceleration - Linear ft/sec2 3.048* e-01

standard gravity 9.806 65* E+OO
Area in2 6.4516 E -04
ft2 9.290 304* e-02
Density Ibm/ft3 1.601 846 E+01
slugdft3 5.153 788 E+02
Energy, work, heat Btu (IT) 1.055 056 E+03
ft-lbf 1.355 818 E+OO
Flow rate, mass Ibm/sec 4.535 924 e-01
Ibm/min 7.559 873 e-03
Ibmhr 1.259 979 e-04
slugs/sec 1.459 390 E+01
Flow rate, volume @/m in 4.71 9 474 e-04
ft3/sec 2.831 685 e-02
gallon (US liquid)/min 6.309 020 e-05
Force Ibf (avoirdupois) 4.448 222 E+OO
Frequency sec" 1* E+OO
Gas constant ft-lbf/lbm-"R 5.380 320 E+OO
Length in 2.54* e-02
ft 3.048* e-01
mile (US) 1.609 344* E+03
Mass Ibm (avoirdupois) 4,535 924 e-01
slug 1.459 390 E+Ol
Plane Angle degrees 1.745 329 E- 02
Power Btu(lT)/hr 2.930 71 1 e-01
ft-lbf/sec 1.355 818 E+OO
hp(550 ft-lbf/sec) 7.456 999 E+02
Pressure standard atmosphere 1.O13 25* E+05
bar 1* E+05
Ibf/ft2 4.788 026 E+01
Ibf/in2 6.894 757 E+03
Rotational frequency min" 1.666 667 e-02
Specific heat Btu/lbm-'.R 4.186 8* E+03
Specific volume ft3/lbm 6.242 797 e-02
Specific weight (force) Ibf/ft3 1.570 875 E+02
Surface tension lbf/ft 1.459 390 E+01
Temperature interval OF 5.555 556 e-01
Temperature, measured "F tc = (tF - 32)/1,8
Temperature, thermodynamic "C +
TK = tc 273.15
"F +
T, = (tF 459.67)/1.8
"R T K = T~/1.8
Time hr 3.6* E+03
min 6* E+01
Torque Ibf-in 1.129 848 e-01
Ibf-ft 1.355 818 E+OO
133

ASMEIANSI PTC 19.2-1987 PRESSURE MEASUREMENT
COMMONLY USED CONVERSION FACTORS (CONTD)
Conversion
Multiplication
Quantity
To From Factor
Velocify far mls 8.466 667 e-05

Wmin m/s 5.08*, e-03
Wsec mls 3.048* e-01
knot (international) mls 5.144 444 e-01
mile (US)/hr mls 4.470 4* e-01
Viscosity,
centipoise
dynamic Pals 1* e-03
poise Paas 1* e-01
Ibm/ft-sec Paas 1.488 164 +00
Ibf-sedft* Paes 4.788 o26 E+01
slug/ft-sec Pa-s 4.788 026 E+01
Viscosity, kinematic centistoke m21s 1* e-06
stoke m1s 1* e-04
ft2/sec m2/s 9.290 304* E-O2
gallon Volume (US liquid) m3 3.785 412 e-03
ft3 m3 2.831 685 E-02
in3 m3 1.638 706 E-05
Iiter m3 1* E -03
GENERAL NOTE: The factors are written as a numbergreater than one andless than ten with six decimal
places. Thenumber is followed bythe letterE (for exponent), a plus or minus
symbol, and two digits which
indicate to power of 10 by which the number must be multiplied to obtainthe correct value.
Example:
3.785 412 E-03 is 3.785 412 x 10 or 0.003 785 412
NOTE:
*Exact relationships in terms of the base unit.
134

A7 'c1 7596 90 0052228 0
SUPPLEMENTS ON INSTRUMENTS AND APPARATUS

NOW AVAILABLE
PTC 19.1 . Measurement Uncertainty. ............................ 1985

PTC 19.3 - Temperature Measurement. ........................... 1974
(R19861
PTC 19.5 - Application. Bart II of Fluid Meters: Interim
...............1 9 7 2
Supplement on Instruments and Apparatus
PTC 19.6 - . . . . . . . . . . . . . . . . .1 9 5 5
Electrical Measurements in Power Circuits
PTC 19.7 - .......................... 1 9 8 0
Measurement of Shaft Power
PTC 19.8 - ...................1 9 7 0
Measurement of Indicated Horsepower
(R1985)
PTC 19.10 - lue and Exhaust Gas Analyses......................... 1981
PTC 19.9 1 - Water and Steam in the Power Cycle (Purity and Quality.
Lead Detection and Measurement)..................... 1970
PTC 19.1 2 - Measurement of Time ............................... 1958
PTC 1 9.13- Measurement of Rotary Speed ......................... 1961
PTC 19.14 - Linear Measurements ................................ 1958 3
PTC 19.1 6- Density Determinations of Solids and Liquids. . . . . . . . . . . . . .1.9 6 5
PTC 19.1 7- Determination of the Viscosity of Liquids..................1 9 6 5
PTC 19.22 - Digital Systems Techniques ........................... 1986
PTC 19.23 - Guidance Manual for Model Testing ..................... 1980
(R1 985)
D02987


Asme 19.2 1987

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Asme 19.2 1987

Uploaded by

Copyright:

Available Formats

COPYRIGHT American Society of Mechanical Engineers

Licensed by Information Handling Services

ASME/ANSI PTC 19.2-1987

THE AMERICAN SOCIETY OF MECHANICAL

COPYRIGHT American Society of Mechanical Engineers

ASME is the registered trademark of The American Society of Mechanical

No part of this document may bereproduced in any form,

COPYRIGHT American Society of Mechanical Engineers

(This Foreword is not part of ASMUANSI PTC 19.2-1 987.)

COPYRIGHT American Society of Mechanical Engineers

COPYRIGHT American Society of Mechanical Engineers

COPYRIGHT American Society of Mechanical Engineers

J. S. Davis, Jr.,Vice President

A. F. Armor W. L. Carvin S. P. Nuspl

COPYRIGHT American Society of Mechanical Engineers

1 GENERAL CONSIDERATIONS ................................. 1

2 PRESSURE STANDARDS ...................................... 5

3 PRESSURE MEASUREMENT DEVICES ............................ 21

4 PRESSURE MEASUREMENT INSTALLATIONS...................... 95

5 REFERENCES ............................................... 117

COPYRIGHT American Society of Mechanical Engineers

COPYRIGHT American Society of Mechanical Engineers

3.38 ThermocoupleGage ......................................... 60

COPYRIGHT American Society of Mechanical Engineers

4.1 5 BasketProbe With a Pressure Transducer .......................... 106

COPYRIGHT American Society of Mechanical Engineers

ASMUANSI PTC 19.2-1987

AN AMERICAN NATIONAL STANDARD

ASME PERFORMANCE TEST CODES

SECTION 1 - GENERAL CONSIDERATIONS

COPYRIGHT American Society of Mechanical Engineers

Ambient pressure (zero gage pressure)

FIG. 1.1 BASIC

Equation (1-1) applies to the flow of a frictionless fluid 1.2 UNITS

COPYRIGHT American Society of Mechanical Engineers

PRESSURE MEASUREMENT . ASME/ANSI

TABLE 1.1 1.3.2 Fluctuating Flow. When pressure measurements

COPYRIGHT American Society of Mechanical Engineers

COPYRIGHT American Society of Mechanical Engineers

SECTION 2 - PRESSURE STANDARDS

COPYRIGHT American Society of Mechanical Engineers

Measurement Pressure Pressure

FIG. 2.1 PRESSURE MEASUREMENT CALIBRATION HIERARCHY

COPYRIGHT American Society of Mechanical Engineers

FIG. 2.2 SCHEMATIC DISTORTION OF PISTON AND CYLINDER IN A-SIMPLE

distortion of pisfon and cylinder in a simple piston where

cylinder distortion due cylinder distortion

COPYRIGHT American Society of Mechanical Engineers

ASMVANSI PTC 19.2-1987 PRESSURE MEASUREMENT

TABLE 2.1 To auxiliary piston

-5.1 x 10 Tungstedcarbide ... Controlled

PRESSURE MEASUREMENT ASMUANSI PTC 19.2-1987

FIG. 2.5 GAGEGOVERNED BY THE

In the denominator 2.1.2 Manometers. Manometers measure an un-

COPYRIGHT American Society of Mechanical Engineers

where TABLE 2.2

At the bottom of column4 2 Water 0.99997

COPYRIGHT American Society of Mechanical Engineers

COPYRIGHT American Society of Mechanical Engineers

FIG. 2.6 GENERALPRINCIPLE O F ALL MANOMETERS

FIG. 2.7 ERRORS IN COLUMNHEIGHTDETERMINATION(MANOMETER

COPYRIGHT American Society of Mechanical Engineers

1 999.90 -50 657.1 181.5 2.6 x