1、Manometer TablesApproved 30 September 1978ISARP2.11978RECOMMENDED PRACTICEISA The Instrumentation,Systems, andAutomation Society TMCopyright 1978 by the Instrument Society of America. All rights reserved. Printed in the UnitedStates of America. No part of this publication may be reproduced, stored i
2、n a retrieval system, ortransmitted in any form or by any means (electronic, mechanical, photocopying, recording, orotherwise), without the prior written permission of the publisher.ISA67 Alexander DriveP.O. Box 12277Research Triangle Park, North Carolina 27709ISARP2.11978, Manometer TablesISBN 0-87
3、664-325-XISA-RP2.1-1978 3FOREWORDThis Recommended Practice has been prepared as a part of the service of ISA toward a goal of uniformity in the field of instrumentation. To be of real value this report should not be static, but should be subjected to periodic review. Toward this end the Society welc
4、omes all comments and criticisms, and asks that they be addressed to the Standards and Practices Board Secretary, ISA, 67 Alexander Drive, P.O. Box 12277, Research Triangle Park, North Carolina 27709, e-mail: standardsisa.org.This report was prepared and revised by the 8D-RP2 Committee of the Produc
5、tion Processes Standards Division on Manometer Tables, chaired by W. G. Brombacher, Consultant, National Bureau of Standards, Washington, D.C.The assistance of those who aided in the preparation of this Recommended Practice by answering questionnaires, offering suggestions, and in other ways is grat
6、efully acknowledged.The following reviewed the original practice and served as a Board of Review:E. A. Adler - United Engineers The American Gas Association, and perhaps other societies, uses 60F; and in orifice flowmeter work, 20C is commonly used. The temperature used to define the pressure does n
7、ot affect other conditions desired as standard, as for example, volume of gas flow measured in terms of the gas at 60F and 29.92 inches of mercury. Some have suggested the maximum density of water (3.98C) as the unit. This temperature is far from normal laboratory temperatures and will require the a
8、pplication of temperature corrections in many cases where none would be necessary with 20C as the standard. The proposal is attractive, however.It is intended that the definition apply to distilled water, free from absorbed gases. It is easier to eliminate absorbed gases than to determine the amount
9、 present. In mercury, absorbed gases ISA-RP2.1-1978 9give no trouble. Lack of purity fouls the mercury meniscus long before there is a significant change in density.No action on standards for liquids other than water and mercury was taken, since these other liquids are not often used as standards of
10、 pressure.3 Errors of liquid column manometers3.1 The errors in the indications of manometers are: (a) scale errors, (b) temperature errors, (c) gravity errors, (d) capillary errors, (e) compressibility, and (f) effect of absorbed gases. Of these, corrections are ordinarily applied only for the scal
11、e, temperature, and gravity errors. The other errors are corrected for only precise work or under special conditions of use.3.2 Scale errors. After a manometer reading has been corrected for temperature and gravity error, and for such of the other errors as are warranted, there usually remains a res
12、idual error which may be called the scale error. This can be determined only by calibration against a standard instrument. If corrections are made only for the temperature and gravity errors, the scale error will include (a) errors in graduating the scale, and (b) the effect of capillarity, and perh
13、aps of compressibility and of absorbed gases. Under (a) will be included to a high degree of accuracy the error introduced when the scale is graduated true at one temperature and is assumed correct at another temperature. Further, when the effects under (b) are included under the scale error, only t
14、he variation in these effects need to be accounted for.In general, the procedure of calibrating manometers is such that correction for scale error should be applied first in applying corrections to a manometer reading.3.3 Temperature Errors. Since the densities of liquids vary with temperature, any
15、deviation in manometer temperature from that selected as standard for the pressure unit will introduce an error in the manometer indication. Further, since the scale expands or contracts with changes in its temperature, an additional error is introduced. An expansion of the scale reduces the reading
16、 of the manometer held at constant pressure and conversely an expansion of the liquid increases the reading; the two expansions tend to balance one another, but the effect of the expansion of the liquid is usually much larger. Temperature corrections are given by the following relations:(3)(4)where
17、Hois the height of the liquid column at the standard temperatureHtis the indicated height of the liquid column at temperature t, corrected for scaleerrorsC is the temperature corrections is the coefficient of linear expansion of the scalem is the coefficient of cubical expansion of the liquidt is th
18、e manometer temperaturetsis the temperature at which the scale indicates the true heightHoHtC+=Cst ts()mt to()1 mt to()+-Ht=10 ISA-RP2.1-1978tois the standard temperature at which the height of the liquid column is in termsof a pressure unit.When the correction is negative in sign, the correction is
19、 subtracted from Ht as indicated by equation (3).The coefficient of expansion is usually available with reference to 0C or 32F, but no significant error is ordinarily introduced by using this value if todiffers from 0C. If warranted, the proper value of s can be computed.Both the cubical and linear
20、thermal expansions of most manometer liquids and scales are hyperbolic in character and therefore in general cannot be accurately represented by the single terms s and m used in equation (4). For the small temperature ranges over which manometers are generally used, the errors introduced are not usu
21、ally significant. Consideration of this point is essential in very precise measurements.Exceptions to the above are mercury, for which m requires no modification over the temperature range of interest, and water, for which m varies with temperature considerably and significantly for most measurement
22、s at all temperatures ordinarily of interest. Therefore, for water, equation (4) does not apply, and corrections must be applied in two parts, first for the temperature error for the scale =s(tts) and then for the error for the liquid by a multiplier obtained from Table 12.If tobe substituted in equ
23、ation (4) for the temperature tsat which the scale is calibrated true and the resulting error be incorporated into the scale correction, considerable simplification in constructing tables is obtained. The error thus introduced is usually not significant. On this basis, equation (4) becomes(5)For cis
24、tern manometers where the level of the liquid in the cistern shifts with change in the pressure while the scale remains fixed, equations (4) or (5) do not apply. Usually the temperature correction can be put in the form(6)where C1is the temperature correction, Cois defined by equation (5), and h is
25、an addition to the indicated column height Htwhich in general must be determined by calibration tests at various temperatures of a manometer of any given design.3.4 Gravity Errors. The effect of variation of gravity from the standard value on manometer indications is as follows:(7)(8)where P is the
26、height of the liquid column subjected to standard gravity. This is thepressure if, as is common practice, no further corrections are applied.Hois the indicated height of the liquid column at the standard temperatureC is the correction to be applied to Hog is the ambient value of gravityCst to()mt to
27、()1 mt to()+- HtCoHt=C1Hth+()Co=Pggs- Ho=or P HoHoggs()gs-+ HoC+=ISA-RP2.1-1978 11gsis the standard value of gravity, 980.665 cm/sec2The ambient value of gravity at any location can be obtained as a function of latitude from the Smithsonian Meteorological (or Physical) Tables and more accurate value
28、s for many locations from the U. S. Coast there is little overall advantage in bores exceeding 15 to 20 mm (0.6 to 0.8 in.). The above recommended minimum and maximum bores apply equally well to water manometers and probably also to those with other liquids. Equal size of bore is obviously indicated
29、 for U-tube manometers.Bore of tube, mm Height of meniscus in mm0.6 0.8 1.0 1.5 2.060.730.941.121.43-8.370.480.580.780.8910 0.20 0.26 0.32 0.44 0.5212 0.12 0.15 0.19 0.26 0.3115 0.05 0.07 0.09 0.12 0.1520 0.015 0.02 0.024 0.034 0.042ggo0.000094h=12 ISA-RP2.1-1978b) Cistern manometers should also hav
30、e the recommended tube bores. The capillary effect in the cistern is minimized by making the area of the cistern large. In mercury manometers of this type especially, it is necessary to tap or vibrate the cistern vigorously in order to reduce scatter in the readings at a constant pressure. It also h
31、elps to vibrate the tube, but not to the same degree.c) The gradual accumulation of corrosion products and dirt at mercury surfaces changes the capillary correction. This change is larger in its effect on the reading the smaller the bore of the tube or of the cistern. Corrosion seems easier to preve
32、nt with other liquids.d) After a pressure change, drainage of liquids which wet the surface of tubes is a source of error, particularly in manometers in which the measured pressure depends upon the indication of one surface level only, as in the case of most cistern manometers. In effect a time lag
33、in reading occurs while drainage takes place. The effect reduces with increase in the bore of the manometer tubes.e) The addition of a few drops of a wetting agent, such as Aerosol OT100, or Dreen, to the water helps greatly in obtaining a symmetrical meniscus in water manometers. This decreases the
34、 difficulty in making readings and also eliminates in large measure the effect of films of foreign material on the glass.3.6 Compressibility effects. The absolute pressure acting on a liquid compresses it and thereby changes its density. It is of importance for manometers only in the cases where the
35、 difference between two high pressures is measured.The compressibility coefficient C, or change in volume per-unit pressure, is defined as follows:(10)where Vois the volume at pressure P or more practically is the volume at the lowestpressure dV/dP is the rate of change of volume with pressure norma
36、lly negative since the volumeusually decreases with the application of pressureA few values of the coefficient are given in the table for water, mercury, and ethyl alcohol. The pressures are absolute values. The values for water were obtained from “Properties of Ordinary Water Substance“ by E. N. Do
37、rsey; those on mercury from the “Smithsonian Physical Tables;“ and those on ethyl alcohol from the “Handbook of Chemistry and Physics.“C1Vo-dVdP-=ISA-RP2.1-1978 13Compressibility of liquidsPetroleum oils with a specific gravity 60/60F between 0.80 and 0.90 have a compressibility coefficient of appro
38、ximately 70 x 106at 20C (68F) for a gage pressure change from 0 to 50 atmospheres. Reference: R. S. Jessup, BS Jnl. of Research, Vol. 5, 1930, p 985; RP244. Most organic liquids have compressibilities of the same order as that of oil.As a consequence of compressibility it requires 4 x 106more pressu
39、re to obtain a change in indication of one mm of mercury at an absolute pressure of 760 mm than it does at one mm of mercury. This difference is 12 times as great for water and 38 times as great for ethyl alcohol. Thus the unit of pressure defined in terms of a liquid column is also a function of pr
40、essure, but the effect in manometers as ordinarily used to measure low gage pressures is less than 0.001 percent.3.7 Effect of Absorbed Gases. Air absorbed in water decreases the density, contrary to what might be expected. At 100C the decrease of water density presumably fully saturated with air is
41、 1.6, and at 20C, 0.4 parts per million. The weight of absorbed air is 27 at 10C and 22 parts per million at 20C. The effect is quite variable and requires consideration for each gas in contact with a particular liquid.Liquid Temp. deg C *Coefficient Pressure range, atmospheresWater 10 48.0 1-2520 4
42、7.2 1-2510 49.2 25-5020 47.6 25-5010 46.0 100-20020 44.2 100-20050 42.8 100-20020 37.4 500-10020 32.6 1000-150020 29.2 1500-200020 25.9 2000-250020 23.6 2500-3000Mercury 20 3.95 30022 3.97 50022 3.91 1000Ethyl Alcohol 20 112 1-5028 86 150-20028 81 150-40014 ISA-RP2.1-1978The “Handbook of Chemistry a
43、nd Physics“ and the “International Critical Tables“ contain data on the solubility of many gases in a number of liquids. These references, however, contain relatively little information on the effect of solubility on the density of those liquids that are used in manometers.Mercury is an exception in
44、 that small amounts of gases go into solution. No data were found, but no sensible outgassing is evident when mercury is subject to a sudden decrease in pressure, as is the case with most other manometric liquids.Since there is considerable time lag in both outgassing and in the dissolving of gases
45、in liquids, considerable uncertainty usually exists as to the precise amount of gas in solution. For this reason, densities used in computing pressures should be for gas-free liquids if available, and when necessary, corrections made for the change in density due to the gas in solution. The correcti
46、on is rarely made except in precise work or under unusual conditions of measurement4 Basis, description and use of tables4.1 Table 1. Abbreviations. Abbreviations used are those proposed by the American Standards Association (Z10.1-1941).4.2 Table 2. Fundamental constants and common factors. As indi
47、cated in the table the values are largely taken from “Units of Weight and Measure,“ NBS M233, although the number of digits in the values is far in excess of that needed in manometry. The conversion factors are primarily from metric to English units and vice versa.4.3 4.4 Tables 3, 4, and 5. No comm
48、ent is required.4.5 4.6 Table 6. Density of mercury. The values of the density of mercury given here are based on the mean value of the thermal coefficient of volume expansion. Accurate (not mean) values of the coefficient of cubical thermal expansion of mercury are given by Beattie, Blaisdell, Kaye
49、, Gerry and Johnson, Proc. Am. Acad. Arts and Sciences, 74, 370, 1941. The two values of the density do not differ more than one part in 100000. At 0C, the density is generally considered accurate to one part in 100,000 for virgin natural mercury. There is no great difficulty in purifying mercury to the degree necessary to realize the densities given; even minute impurities of the base metals make the mercury unusable in a ma
