IS 9163-1 (1979): Dilution Methods for Measurement of ...

Disclosure to Promote the Right To Information

Whereas the Parliament of India has set out to provide a practical regime of right to information for citizens to secure access to information under the control of public authorities, in order to promote transparency and accountability in the working of every public authority, and whereas the attached publication of the Bureau of Indian Standards is of particular interest to the public, particularly disadvantaged communities and those engaged in the pursuit of education and knowledge, the attached public safety standard is made available to promote the timely dissemination of this information in an accurate manner to the public.

इंटरनेट मानक

“!ान $ एक न' भारत का +नम-ण”Satyanarayan Gangaram Pitroda

“Invent a New India Using Knowledge”

“प0रा1 को छोड न' 5 तरफ”Jawaharlal Nehru

“Step Out From the Old to the New”

“जान1 का अ+धकार, जी1 का अ+धकार”Mazdoor Kisan Shakti Sangathan

“The Right to Information, The Right to Live”

“!ान एक ऐसा खजाना > जो कभी च0राया नहB जा सकता है”Bhartṛhari—Nītiśatakam

“Knowledge is such a treasure which cannot be stolen”

“Invent a New India Using Knowledge”

है”ह”ह

IS 9163-1 (1979): Dilution Methods for Measurement ofSteady Flow, Part I: Constant Rate Injection Method [WRD 1:Hydrometry]

i8 : 9163 ( Part I ) - 1979

Indian Standard DILUTION METHODS FOR

MEASUREMENT OF STEADY FLOW

PART I CONSTANT RATE INJECTION METHOD

Fluid Flow Measurement Sectional Committee, BDC 17

SHRI K. K. FRAMJI Consulting Engineering Services ( India ) Pvt Ltd

57 Nehru Place, ‘ Manjusha ’ ( 5th Floor ) New Delhi 110019

Members Representing

SARI S. BANERJI Indian National Committee for the International Hydrological Programme ( CSIR ), New Delhi

DR BHARAT SINQR University of Roorkee CHIEF ENGINEER ( BRIDGES ) Roads Wing, Ministry of Transport & Shipping CHIEF ENGINEER ( DESIGN & Land Reclamation, Irrigation & Power Research

RESEARCH ) Institute, Government of Punjab, Amritsar DIRECTOR ( LRIPRI ) ( Alternate )

CHIEF ENGINEER ( F I 8~ T ) Central Water Commission, New Delhi DIRECTOR ( CSMRS ) ( Alternate )

CHIEF ENGINEER ( INVESTIGATION ) Irrigation Department, Government of Tamil Nadu, Madras

; DIRECTOR ( INSTITUTE OI

HYDRAULICS & HYD~OLOQY DrREo;;~~re )

Andhra Pradesh Engineering Research Labora- tories, Hyderabad

DIRECTOR DR 2. S. TARAPOI~E (Alternate

DIRECTOR

Central Water & Power Research Station, Pune 1

Irrigation Research Institute, Government of Uttar Pradesh, Roorkee

DIRECTOR River Research Institute, Government of West

DEPUTY DIRECTOR ( HYDRAU- Bengal

LICS GROUP A ) ( Alternate ) SHRI 0. P. GARG Ganga Basin Water Resources Organization

( Ministry of Agriculture & Irrigation )

( Continued on page 2 ) I

@ Cofiyright 1980

INDIAN STANDARDS INSTITUTION

This publication is protected under the Indian Cofiyright Act ( XIV of 1957) and reproduction in whale or in part by any means except with written permission of the publisher shall be deemed to be an infringement of copyright under the said Act.

c

\

( Continuedj?om page 1 )

Members Representing

SHRI N. K. GHOSH National Instrument ( Private ) Ltd, Calcutta HYDRAULIC ENGINEER Bombay Municipal Corporation

SHRI T. M. KANTAWALA ( Alternate ) JOINT DIRECTOR R E s E A= c a: Research, Designs & Standards Organization,

( BRIDGES & FLOODS ) Lucknow DEPUTY DIRECTOE ( BRIDGES

& FLOODS ) ( Alternate ) JOINT SECRETARY ( GB ) Ministry of Agriculture & Irrigation

MEMBER, INDO-BANGLA DESII JOINT RIVERS COWTISSION ( Alternate )

DR R. C. MALHOTRA Indian Institute of Technology, New Delhi SHRI R. H. MENDOXSA All India Instrument Manufacturers & Dealers

Association, Bombay SHRI J. MENDONSA ( Alternate )

METE~R~L~OIST India Meteorological Department, New Delhi PROF N. S. GOVINDA RAO In personal capacity ( 89 Diagonal Road, Visvesvara-

#ram, Bangalore 560004 ) PROF N. S. LAKSHMANA RAO Indian Institute of Science, Bangalore SECRETARY Central Board of Irrigation & Power Saail G. S. SHIVANA Public Works Department, Government of

Karnataka SHRI D. AJITRA SInsHA, Director General, IS1 ( Ex-o&cio Member )

Director ( Civ Engg )

Secretary

SHBI K. RAUEAVENDRAN Deputy Director ( Civ Engg ), IS1

Dilution Methods Subcommittee, BDC 17 : 7

Convener

SHRI C. V. GOLE Central Water Commission

Members

DIRECTOR Central Water & Power Research Station, Pune DR K. S. RAJAGOPALAN ( Alternate )

E)IBECT~R Power Research Institute, Bangalore Snnr N. V. KRIS~NAS~AMY ( Alternate )

SHRI K. K. FRAD~JI In personal capacity [ Consulting Engineering Services ( India ) Put Ltd, 57 Nehru Place, ’ Manjusha ‘, ( 5th Floor ) , .iVew Delhi 110019 ]

DR V. K. IYA Atomic Energy Commission, Bombay SHRI D. RAGHURAMAN National Environmental Engineering Research

SHRI S. K. SRIVASTAVA ( Alternate ) Institute, Nagpur

SECRETARY Central Board of Irrigation & Power DEPUTY SIXRETARY ( ID ) ( Alternate )

SHRI H. D. SHAEMA Irrigation Research Institute, Government of Uttar Pradesh, Roorkee

SHRI H. R. SIIARMA ( Alternate )

2

Indian

DILUTION E

MEASUREMENT

PART I CONSTANT F

0. F(

0.1 This Indian Standard ( Part ards Institution on 30 March 1979 Flow Measurement Sectional Corn Engineering Division Council.

0.2 Dilution methods are suitable hilly regions where turbulence ant be difficult, if not risky, to use con que is now tried in this countr after sufficient trials in many coun in ascertaining the minimum mix the chemical and the information only as a rough guide, particularl;

0.3 In the formulation of this star: international co-ordination among in different countries in addition 1 in this country. This hasibeen mei ( Part I ) Liquid flow measureme for measurement of steady flow : 1 issued by the International Orgar

0.4 This standard is one of the s for measurement of steady flow ir

0.5 In reporting the results of a tc this standard, if the final value, o off, it shall be done in accordance

*Rules for rounding off numerical v

IS : 9163 ( Part I ) -1939

Indian Standard DILUTION METHODS FOR

MEASUREMENT OF STEADY FLOW

PART I CONSTANT RATE INJECTION METHOD

0. FOREWORD

0.1 This Indian Standard ( Part I ) was adopted by the Indian Stand- ards Institution on 30 March 1979, after the draft finalized by the Fluid Flow Measurement Sectional Committee had been approved by the Civil Engineering Division Council.

0.2 Dilution methods are suitable for measurement of flow in streams in hilly regions where turbulence and velocity of flow are high and it would be difficult, if not risky, to use conventional methods. Though the technique is now tried in this country, the same has been well established, after sufficient trials in many countries. However, care should be exercised in ascertaining the minimum mixing length required for good mixing of the chemical and the information given in this standard should be used only as a rough guide, particularly for flows in excess of looms/s.

0.3 In the formulation af this standard due weightage has been given to international co-ordination among the standards and practices prevailing in different countries in addition to relating it to the practices in the field in this country. This hasbeen met by basing the standard on ‘ISO/R 555- ( Part I ) Liquid fl ow measurement in open channels - Dilution methods for measurement of steady flows: Part I Constanst-rate injection method’ issued by the International Organization for Standardization.

0.4 This standard is one of the series of standards on dilution methods for measurement of steady flow in open channels.

0.5 In reporting the results of a test or analysis made in accordance with this standard, if the final value, observed or calculated, is to be rounded off, it-shall be done in accordance with IS : 2-1960”.

*Rules for rounding off numerical values (revised).

3

IS : 9163 ( Part I ) - 1979

1. SCOPE

1.1 This lndian Standard specifies methods for the measurement of flow in channels under steady flow conditions by the dilution method using constant-rate injection. The method is applicable to the measurement of the flow in channels where the degree of turbulence is sufficiently high to ensure efficient mixing of the injected solution throughout the whole flow. This is generally the case in hill streams and torrents,

1.1.1 The apparatus and the chemicals to be used, the specifications of the techniques of injection and sampling, and also the methods of analysis are detailed.

2. DEFINITIONS

2.1 For the purpose -of this Indian Standard, the definitions given in IS : 1191-1971” shall apply.

3. UNITS OF MEASUREMENT

3.1 The units of measurement used in this Indian Standard are metres and seconds.

4. PRINCIPLE OF THE CONSTANT-RATE INJECTION M-ETHOD

4.1 A-solution of a suitably selected salt is injected at a constant rate, at a cross section, at entry of the measuring reach of the channel in which the rate of flow is constant over the period of test.

4.2 In a downstream cross section of this reach, far enough from the first for the injected solution to be uniformly diluted throughout this cross section, samples are taken at regular intervals of time. Provided that the concentration of the added chemical has attained a steady value, the rate of flow Qin the channel may be calculated from the equation:

Q = qJv

where

4 = rate of injection of the chemical salt solution;rand N= ratio of the concentration of the salt in the injected

solution to that at the downstream cross section of the channel, however, if traces of the chemical salt are already present in the stream prior to the injection, Nis defined by the equation given at the end of 4.3.1.

The ratio JV, which is called the dilution ratio, may be determined by one of the following methods.

*Glossary of terms and symbols used in connection with the measurement of liquid flow with a free surface (first revision ).

4

4.3 Method of Direct Measurem

4.3.1 The concentration of the :

a) upstream of the point of il state ( concentration Co )

b) in the injected solution ( t

c) in samples obtained fror concentration of the cherr

Equating the mass of the ( CoQ+ Clq ) and that passing th

Q=f

and if C’s is small compared

Q- --

4.4 Method of Comparative Dil made by diluting samples from the nts of water from the natural stre: injection station. These known di mately the same as those expecte section.

4.4.1 The standard solutions 2 compared by a similar technique unknown samples.

5. REQUIRED CHARACTERI!

5.1 The chemical substance to bi comply with the -following require1

4

b) C) d)

e>

It shall not give any react concerned or with any n which this may contain i material which forms the

It shall be stable to light

It shall not be toxic to fir

It shall be capable of beil tration level of the dilute

It shall be used only whe solution in the natural w

IS : 9163 ( Part I ) - 1979

4.3 Method of Direct Measurement of Concentration

4.3.1 The concentration of the salt is determined:

a) upstream of the point of injection where the water is in its natural state ( concentration Co ) [ see however, 7.4 (a) 1;

b) in the injected solution ( concentration C1 ); and

c) in samples obtained from the downstream section ( where the concentration of the chemical has attained a steady value Cs ).

Equating the mass of the salt passing the injection section ( CoQ+ Clq ) and that passing the sampling section C2 ( Q, + q ) gives:

Q= cz-_do q cl- (32

and if Cs is small compared with Cl

-QzNs ~'1 -- Q cz - co

4.4 Method of Comparative Dilutions - A set of standard solutions is

made by diluting samples from the injected solution with different amounts of water from the natural stream, taken from a point upstream of the injection station. These known dilution ratios Nl, JV2...shall be approximately the same as those expected of the samples from the downstream section.

4.4.1 The standard solutions and the samples taken downstream are compared by a similar technique to establish the dilution ratio Nof the unknown samples.

5. REQUIRED CHARACTERISTICS OF INJECTED SOLUTIONS

5.1 The chemical substance to be used for the injected solution shall comply with the following requirements:

a) It shall not give any reaction with the natural water of the river concerned or with any matter ( organic matter in particular ) which this may contain in solution or suspension, or with the material which forms the river bed;

b) It shall be stable to light; c) It shall not be toxic to fish in the dimtions used;

d) It shall be capable of being accurately determined at the concentration level of the diluted samples;

e) It shall be used only when the concentration of the chemical in solution in the natural water is relatively low.

5

IS : 9163 ( Part I ) - 1979

The following substances, which may not be suitable for all waters, are given as examples, together with final minimum concentrations at which they can be used after dilution:

Sodium dichromate ( Na&r3072Hz0 ) : 0.2 mg/l

Sodium chloride ( NaCl ) : 2 mg/l

Other salts have been used, in particular: Sodium nitrite( NaNO, ), and

Manganese sulphate (MnS044Hz0 ),

A chemical substance, such as fluorescein ( C,0H1,C5 ) having sufficient colouring power to allow the passage of the cloud in the river to be checked visually is useful, particularly in preliminary tests to check the delay period before sampling (see 6.2.1 ).

6. CHOICE OF THE MEASURING-REACH

6.1 General Considerations for Selection of Site

6.1.1 There shall be no loss or gain of water in the measuring-reach ( for example, a tributary joining or a distributary leaving the main flow, or overflow from or to the banks of the stream j, and its length shall be such that, allowing for the natural mixing action of the stream, the solution injected at its beginning is uniformly diluted throughout the sampling cross section.

6.1.2 The distance between the injection and sampling cross sections shall be as short as practicable and the dead-water zones shall be as small as possible, to reduce the duration of injection and the quantity of salt injected.

6.1.3 For rivers, this condition is easier to satisfy in relatively narrow channels; mixing is also improved by disturbances such as bends, narrows, shelves, falls, etc. In particular, very wide channels shall be avoided, and reaches in which the stream divides into a number of branches shall not be used.

6.1.4 Great care is required to make a satisfactory choice of a reach suitable for the measurement of discharge by the dilution method using constant-rate injection. This choice is facilitated by carrying out the preliminary tests described in 6.2.

6.2 Preliminary Tests- It is essential to the success of the method that the injected material have in constant concentration over the whole of the cross section at the downstream sampling station during the whole of the sampling period. Before selectin g the measuring-reach, it is advisable to

make preliminary tests to ensure th achieved, to choose the injection these have been chosen, to determi

6.2.1 Determination of Length of made by using a strong dye, such z of this dye can be injected for cross section of the upstream portic diffusion of the solution will make are any dead zones or similar E rough guide on the minimum dist; suitable sampling cross section.

6.2.2 Duration of Injection and OJ of injection shall be such that a stl be achieved for an adequate du sampling-section.

6.2.2.1 The duration of the the degree of turbulence, will vary measuring-reach and the extent o according to the mean velocity of

*The length of the measuring-reac cross section where mixing may be e: homogeneity, cannot be predicted with c has been evolved, though not widely estz

1=0*13.

where

b = average width of the wa

d = average depth of the wa

X = coefficient defined below

K= c(i

where

C = Chezy coefficient for th

g = acceleration due to gra

It is emphasized that the length o used as a preliminary guide, because t instances due to very low turbulence established by practical tests. For exalr at a single point in a straight reach, the across the cross section are used. In adc tion urlder-estimates the length for sma estimates the length for rivers of the or<

6

.

‘IS : 9163 ( Part I ) - 1979

make preliminary tests to ensure that efficient mixing in the river will he achieved, to choose the injection and sampling cross sections and, once these have been chosen, to determine the duration of the injection period.

~6.2.1 Determination of Length of Measuring-reach* -A first test can be made by using a strong dye, such as fluorescein. A concentrated solution of this dye can be injected for a relatively short time at a point on a cross section of the upstream portion of the measuring-reach. Study of the diffusion of the solution will make it possible to determine whether there are any dead zones or similar side-tracking of the chemical and be a rough guide on the minimum distance between the injection point and a suitable sampling cross section.

6.2.2 Duration of Injection and of the Steady-regime Condition - Duration of injection shall be such that a steady regime of the concentration may be achieved for an adequate duration, generally 10 to 15 min, in the sampling-section.

6.2.2.1 The duration of the injection, which is generally related to the degree of turbulence, will vary directly according to the length of the measuring-reach and the extent of the dead-water ~zones, and inversely according to the mean velocity of the water.

*The length of the measuring-reach 1, between the injection cross section and a cross section where mixing may be expected to be within 1 percent of complete homogeneity, cannot be predicted with certainty; but, as a guide, an empirical formula has been evolved, though not widely established on experimental data, as follows:

1 = 0.13 K$ (in SI units )

where

b = average width of the water-surface in the measuring-reach,

d = average depth of the water in this same reach, and

X = coefficient defined below : - ,=C(O.7C+2&?k?

s

where

C = Chezy coefficient for the measuring-reach, and 15 < C < 50;

g = acceleration due to gravity.

It is emphasized that the length obtained from the above equation may only be used as a preliminary guide, because the equation may not apply at all in certain instances due to very low turbulence and the actual length required should be established by practical tests. For example, since the equation was based on injection at a single point in a straight reach, the length will be shortened if multiple injections across the cross section are used. In addition, some tests seem to show that the equation under-estimates the length for small streams of the order of 5 m in width and over estimates the length for rivers of the order of 50 m in width.

7

IS : 9163 1 Part I ) - 1979

6.2.2.2 When the measuring-reach has been selected, preferably where the river bed is stable: the duration of injection may be determined from preliminary tests at different values of the rate of flow. These tests consist of determining the variation, as a function of time, of the concentrations of an injected chemical substance at a number of sampling cross sections.

6.2.2.3 An instantaneous injection of Auorescein may also be used for this purpose. For a given flow, observation at the time of appearance and disappearance of coloration in each cross section allows curves 1 and 2 of Fig. 1 to be plotted, If it is desired to obtain, at a selected sampling

MEASURING REACH (SCHEMATIC)

2

3 w a

E

iTi ln I

DISTANCE FROM INJECTION CROSS-SECTION

FIG. 1 CURVE OF TIMES OF APPEARANCE AND DISAPPEARANCE OF COLOIJRED CLOIJD AS A FIJNCTION OF DISTANCE FROM POINT OF INJECTION

8

cross section, a steady-regime condi to add the time At to the time of di (see Fig. 2 ), and to trace throut curve (1’) parallel to the curve (1) c ordinate at the origin of this curve gj to be made. The same figure di beginning of the injection and the b

/-MINIMUM MIXING

SAMPL

DISTANCE FROM INJEt

W4lNlMUM DURATION OF INJECTION

FIG. 2 DETERMINATION

6.2.3 Non-absorption of Chemical - that there is no absorption of the in either by matter in suspension or this purpose, samples shall be take atleast one other cross section fart; no systematic difference in the m section to another.

7. PROCEDURE

7.1 Preparation of the Concem the concentration of the injected SC

7.1.1 Homogeneity of the solu performed with a mechanical stirre

b” *

IS : 9163 j Part I ) - 1979

cross section, a steady-regime condition of duration At, it ‘is only necessary to add the time At to the time of disappearance of coloration at this point ( see Fig. 2 ), and to trace through the point obtained in this way a curve (1’) parallel to the curve (1) of appearance of the coloration. The ordinate at the origin of this curve gives the minimum duration of injection to be made. The same figure directly gives the time-lag between the beginning of the injection and the beginning of the steady regime ( AtI ).

MINIMUM MIXING LENGTH

SAMPLING CROSS-SECTION

ZONE OF STEADY REGIME OF CONCENTRATION

At DURATION OF STEADY REGIME AT SAMPLING

CROSS-SECTION

At, INTERVAL OF TIME BETWEEN

INJECTION AND SAMPLING

DISTANCE FROM INJECTION POINT

b4lNlMUM DURATION OF INJECTION

FIG. 2 DETERMINATION OF DURATION OF INJECTION

6.2.3 Non-absorfition of Chemical - It is particularIy necessary to check that there is no absorption of the injected chemical in the measuring-reach either by matter in suspension or by the material of the river-bed. For this purpose, samples shall be taken from the sampling cross section and atleast one other cross section farther downstream to check that there is no systematic difference in the mean concentration from one sampling section to another.

7. PROCEDURE

7.1 Preparation of the Concentrated Solution - It is essential that the concentration of the injected solution be as homogeneaus as possible.

7.1.1 Homogeneity of the solution is achieved by vigorous mixing performed with a mechanical stirrer or with a closed-circuit pump.

within the tank. The crest of the \ tank shall be sharp, horizontal and on the outlet pipe within limits of -J of minor fluctuations in the rate o shall be smooth. The weir shall ext tank to avoid ‘contractions’ and sl tely above the outlet pipe. Arrange

? tion of any solution which overflows of injection flow rates may be obtai nozzles of appropriate dimensions.

IS : 9163 ( Part I) - 1979

7.1.2 It is recommended that the injection solution be prepared in a separate tank from the supply tank ( see Fig. 3 ). Water taken from the o.pen channel shall be used for preparing the concentrated solution. If, however, the mixing is effected in the supply tank, this tank shall be of sufficient capacity to avoid the need for the addition of water or salt during an injection, The solution shall be drawn from a level above the bottom of the tank and provision shall be made to prevent particles of undissolved salts passing out with the injected solution.

7.2 Injection of the Concentrated Solution - The concentrated solution shall be introduced into the stream of which the discharge is to be measured at the chosen injection cross section. The concentrated solution shall be injected at a constant rate of flow which may be controlled by one of the following devices.

a) A constant-level Tunk - This tank ( for example as shown in Fig. 3 ) shall be of such dimensions and volume as will permit suitable arrangements to be made internally to disperse effectively the flow emanat- ing from the inlet pipe and to avoid any short circuiting of flow between the inlet and outlet pipes with the consequent creation of ‘dead’ zones

SUPPLY TANK7 f LIQUID-LEVEL INDICATOR

CONSTANT LEVEL TANK

It=-

------ - ---- --- -

HEAb ON ORIFICE

1

- &OLATING VALVE

i”\

MIXIN,G TANK--/

ORIFICE (FULLY OPEN DURING INJECTION)

INJECTION PIPE

FIG. 3 ARRANGEMENT FOR INJECTING CONCENTRATED SOLUTION BY MEANS OF A CONSTANT LEVEL TANK WITH A REGULATING WEIR

b) A Volumetric Pump Driven by pump ( for example as shown in Fi type and great care shall be taken t constant. The speed can be detern synchronous electric motor, from a

SUPPI

INJECTION PIPE

FIG.~ ARRANGEMENT FOR IN BYROTARY\

10

IS : 9163 ( Part I ) - 1979

within the tank. The crest of the weir controlling the water-level in the tank shall be sharp, horizontal and of sufhcient length to control the head on the outlet pipe within limits of f 0.25 percent of this head regardless of minor fluctuations in the rate of inflow. The back of the weir plate shall be smooth. The weir shall extend completely across one side of the tank to avoid ‘contractions’ and should preferably be situated immediately above the outlet pipe. Arrangements shall be included for the collection of any solution which overflows from the tank during a test. A range of injection flow rates may be obtained by using a set of orifice plates or nozzles of appropriate dimensions.

b) A Volumetric Pump Driven by a Constant-speed Motor - If a volumetric pump ( for example as shown in Fig. 4 ) is used, it shall be of the rotary type and great care shall be taken to ensure that the speed of this pump is constant. The speed can be determined either directly or, if driven by a synchronous electric motor, from a knowledge of the supply frequency.

SUPPLY TANK

f MOTOR

PUMP LINJECTION PIPE

FIG.~. ARRANGEMENTFOR INJECTING CONCENTRATED SOLUTION BYROTARYVOLUMETRIC PUMP

IS : 9163 ( Part I ) - 1979

c) A Mariotte Vessel - ( for example as shown in Fig. 5 ).

d) A Floating Siphon - ( for example as shown in Fig. 6 ).

f- FILLING PLUG

HERMATIC SEAL

GAUGE GLASS

ADJUSTABLE AIR INLET TUBE

DATUM (ATMOSPHERIC PRESSURE)

HEA-D ON ORIFICE PLATE

1

ORIFICE PLATE

ISOLATING VALVE

DETAIL OF AIR INLET TUBE

FIG. 5 MERIOTTE VESSEL FOR INJECTING CONCENTRATED SOLUTION

12

CONSTAN RlFlCE PLATE

FIG. 6 FLOATING SIPHON FOR 1

7.2.1 General Recommendations - injection rate can be calculated fr details of one of the other devices, incorporated in the injection appal and repeatability of the methods.

7.2.1.1 Where the natural tu ensure uniform and homogeneous predetermined mrasuring-reach, solution may be assisted by the injection cross section. Care shall 1 of concentrated solution reaches th on the banks or on rocks above the

c

/ L COLLECTING FUNNEL FOR

CONSTANT FLOW

FIG. 6 FLOATING SIPHON FOR INJECTING CONCENTRATED SOLUTION

7.2.1 General Recommendations - It is generally desirable, even when the injection rate can be calculated from the pump dimensions or from the details of one of the other devices, that a second measurement method be incorporated in the injection apparatus to serve as a check on the accuracy and repeatability of the methods.

7.2.1.1 Where the natural turbulence of the stream is not such as to ensure uniform and homogeneous mixing of the two liquids within the predetermined measuring-reach, the distribution of the concentrated solution may be assisted by the use of perforated pipes spanning the injection cross section. Care shall be taken to ensure that the whole flow of concentrated solution reaches the stream, that is that none is deposited on the banks or on rocks above the water level. ‘\

IS : 9163 ( Part I ) - 1979

7.3 Measurement of Rate of Injection - The rate of injection of the concentrated solution may be measured by direct observation if the regulating device used has been previously calibrated. Where the rate of injection is controlled so that it is constant but is not exactly determined, it may be ascertained indirectly by measuring the total volume of solution discharged over a measured period of time. The degree of accuracy with which the rate of injection can be measured depends on the particular measuring device used. Suitable devices are described in 7.3.1 and 7.3.2, and the estimates of accuracy associated with the particular device shall be taken into consideration when estimating the overall accuracy of the flow measurement in the channel.

7.3.1 Direct Measurement of Rate of Injection - The rate of injection shall be determined from readings, taken during the injection period, of one of the following characteristics, depending on the device used:

a) Level of constant-level tank in relation to terminal orifice,

b) Speed of pump,

c) Level of orifice in relation to the atmospheric pressure datum in the Mariotte vessel,

d) Height between solution surface and orifice of floating siphon arrangement, and

e) Reading of flow-meter.

For procedures (a), (c) and (d) described above, the rate of injection of the concentrated solution can then be determined from the calibration curves for these devices. The devices shall be calibrated both before and after the tests, and if the two calibrations differ by more than 1 percent, the tests shall be repeated. If a volumetric pump is used as in procedure (b), it is advisable that a calibration check be carried out before and after the tests to check that there is no leakage or, alternatively, to determine the leakage if this exists, that is to calibrate the pump under the same conditions as those experienced during the test. There shall not be a difference of more than 1 percent between the two calibrations. A flow- meter may be installed in the injection apparatus for the specific purpose of measuring the rate of injection. It shall be calibrated before and after the test -and these calibrations shall be carried out with the meter installed in the apparatus. Calibration is not necessary:

a) if the meter has been manufactured and installed in conformity with the specifications laid down in ISO/R%l-1965” in particular if sufficient lengths ofstraight pipe have been installed on either side of the meter, and

“A4easuremcnt of flrlid flow by means of orifice plates and nozzles.

b) If it can be shown that been incorporated to ensu data are unaffected by an: tion apparatus.

7.3.1.1 The flow-meter may between the flanges of two lengths apparatus to the injection point. chemical solution being used.

7.3.1.2 The average of the t\ carried out before and after the tes rate of injection.

7.3.2 indirect Measurement of An Indirect measurement of the rate ~1 of two factors : the duration of tl solution delivered.

a) The time which elapses bl and its termination shall f O* 1 percent of the peril

b) The volume may be meas previously calibrated feed cement meter.

7.3.2.1 Calibration-feed tank - soIution is to be measured in a ca to be calibrated shall be erected so vertical. If calibration is made L ensured that no distortion takes pl: Horizontal cross sections shall be - at intervals sufficient to enable the mined with an accuracy of 5 0.2: nuts shall be measured for each t deductions made at the requisite 11 structural members shall be me, also made. These measurement! before the tank is used. Altern: may be determined by the use of of which shall be added to the m successive steps. The smaller vess f 0.1 percent by weighing the c

14

IS : 9163 ( Part I ) - 1979

TV) If it can be shown that an adequate straightening device has been incorporated to ensure that the available meter-calibration data are unaffected by any flow disturbances caused by the injection apparatus.

7.3.1.1 The flow-meter may conveniently be an orifice plate placed between the flanges of two lengths of straight pipe connecting the injection apparatus to the injection point. Its parts shall not be affected by the chemical solution being used.

7.3.1.2 The average of the two calibrations of the device or meter, carried out before and after the test, shall be used in the calculation of the rate of injection.

7.3.2 Indirect Measurement of Rate of Injection ( Volumetric Method) - Indirect measurement of the rate of injection involves the determination of two factors : the duration of the discharge and the total volume of solution delivered.

a) The time which elapses between the commencement of injection and its termination shall be measured to an accuracy of within f. 0.1 percent of the period of injection.

b) The volume may be measured by observing the drop in level in a previously calibrated feed-tank or by the use of a positive-displacement meter.

7.3.2.1 Calibration-feed tank -Where the volume of concentrated solution is to be measured in a calibrated feed-tank, the tank which has to be calibrated shall be erected so that the vertical axis of the tank is truly vertical. If calibration is made by mensuration of the tank, it shall be ensured that no distortion takes place when the tank is subsequently filled. Horizontal cross sections shall be measured to an accuracy of 0.1 percent at intervals sufficient to enable the height/volume relationship to be determined with an accuracy of & 0.25 percent. Bolt heads or bolt shanks and nuts shall be measured for each type and counted and the appropriate deductions made at the requisite levels. Internal flanges, tie-rods or other structural members shall be measured and the appropriate deductions also made. These measurements shall be carried out immediately before the tank is used. Alternatively, the height/volume relationship may be determined by the use of a smaller calibrated vessel, the contents of which shall be padded to the main tank, and the height measured in successive steps. The smaller vessel shall be calibrated to an accuracy of f O-1 percent by weighing the contents. Where the rate of injection is

15

IS : 9163 ( Part I ) - 1979

controlled by a constant-head tank, the spillage therefrom shall be collected and measured in a vessel calibrated to an accuracy depending on the ratio of the amount of spillage to the amount injected. If this ratio is a, the spillage shall be measured to an accuracy of

& & percent

Alternatively, the spillage may be collected and returned to the main tank before the final reading is taken.

7.3.2.2 Positive displacement meter- Wllere the volume of concentrated solution is measured by means of a volumetric meter, the meter shall be of a positive-displacement type and shall have been calibrated recently to give an accuracy of j, 1 percent or better. The meter shall be connected in the pipelines between the constant-head tank and the injection point and shall be of such size and SO placed, in relation to the constant-head tank, that sufficient ~operating head is available to sustain flow through the meter to the injection point at the greatest injection rate required. Precautions shall be taken to remove all suspended impurities from the concentrated solution before it is delivered to the constant-head tank. The volume of solution delivered shall be ascertained from the difference between the readings on the meter index before and after the injection.

7.4 Sampling - Samples shall be taken as follows:

a)

b)

C)

7.4.1

Upstream from the injection cross section, generally two or three samples before and after the injection. However, if any variation of the background concentration is anticipated along the measuring-reach, the samples of the natural water shall be taken at the downstream sampling station before and after the passage of the salt solution.

.4t the outlet of the injection apparatus, three to five samples are recommended, either just before and just after the injection period or, alternatively, during the injection period, It should be noted that sampling during the period of injection will affect the total amount of the salt injected and the constancy of the injection rate. However, this source of error is generally negligible.

At two or three points in the sampling cross section, at regular intervals during the steady regime, five to ten samples at each point.

These samples shall be taken by means of immersing bottles or by pumping. With a view to avoiding the influence of more or less rapid

chance variations of the concent recommended that the time taken as possible.

8. PRINCIPAL CHEMICAL Sl ANALYSIS IN PRESENT-DA

8.1 Method of Calorimetric An

8.1.1 Choice of Chemical Substan Measured - The salt at present f: injection dilution method with co11 Its solubility in water is relatively tely 600 g/l of water may be used of the requirements of section : measurement of very low concentr concentrations between 0.2 and 2 depend upon the concentration u the calorimetric apparatus. Nat1 chromium ions in solution. The pr in the measurement. The nature sion in the natural water can seril

8.1.2 Method of Analj,sis - The compare, by means of a colorime ( the natural dilution of the samp‘ dilutions ), by their absorption of each solution.

The recommended reagent 1

Diphenylcarbazide ( crystallized ) [ (

f;nh;thhank;y:ride( c

I

Ethyl alcohol, 95 percent ( v/v) ((

To 50 ml of the sample concentrated sulphuric acid ( Hz should be added to obtain an 2 reagent mentioned. The action c acid medium for the calorimetric after introduction of the reagent i dure shall be carried out in the fc

c

IS t 9163 ( Part I ) - 1979

chance variations of the concentration in the sampling section, it is recommended that the time taken between successive samples be as short as possible.

8. PRINCIPAL CHEMICAL SUBSTANCES AND METHODS OF ANALYSIS IN PRESENT-DAY USE

8.1 Method of Calorimetric Analysis

8.1.1 Choice of Chemical Substance and Minimum Concentration that can be Measured - The salt at present favoured for applying the constant-rate injection dilution method with calorimetric analysis is sodium dichromate. Its solubility in wa.ter is relatively high ( concentrations up to approximately 600 g/l of water may be used in practice ) and the salt satisfies most of the requirements of section 5. Calorimetric analysis permits the measurement of very low concentrations of sodium dichromate. With final concentrations between 0.2 and 2 mg/l, the accuracy of the analysis will depend upon the concentration used and the sensitivity and accuracy of the calorimetric apparatus. Natural waters do not generally contain chromium ions in solution. The presence of such ions would lead to errors in the measurement. The nature and quantity of any matter in suspcn- sion in the natural water can seriously affect the accuracy of analysis.

8.1.2 Method of Anal_),sis - The principle of calorimetric analysis is to compare, by means of a calorimeter, the dilutions of sodium dichromate ( the natural dilution of the samples from the channel and the standard dilutions ), by their absorption of light after a reagent has been added to each solution.

The recommended reagent has the following.composition:

p;;;;ny;;e2;zide [ ( CsHs NH.NH )2 CO ] : 0.25 g

;;;$$$~~)dride( CsH/;;>o ) : 4.0 g

Ethyl alcohol, 95 percent ( v/v ) (CJ&OH) : 100 ml

To 50 ml of the sample which is being analyzed, 10 drops of concentrated sulphuric acid ( HzSOd ) having a relative density of 1.84 should be added to obtain an acid solution, followed by 2 ml of the reagent mentioned. The action of this reagent is sufficiently rapid in an acid medium for the calorimetric measurement to be made about 2 min after introduction of the reagent into the solution. The analytical procedure shall be carried out in the following manner.

IS: 9163 ( Part I ) - 1979

8.1.2.1 Calibration of the calorimeter - From one of the samples taken at the outlet of the injection device, a number of standard solutions ( with a minimum of four ) shall bemade, having a known dilution approximating to the dilution occurring in the channel. For this purpose, a sample of the concentrated injection solution is diluted by means of calibrated flasks and pipettes with water from the channel, taken from a point upstream of the injection section. These samples can then be measured in the calorimeter and the readings, which are a function of the dilution ratio, plotted on a curve. It is recommended that a calorimeter be used which gives a linear relationship between colour intensity and meter reading.

The calibration in the flasks and pipettes is one of the major sources of error in this method. With commercial grade equipment, this error may reach -f 1 percent for a dilution ratio of 50 000. To reduce the systematic influence of this error, a precise calibration of the apparatus used for carrying out these standard dilutions shall be made. The concentration of the injected solution shall be checked for uniformity by analyzing identical dilutions of the samples taken [ see 7.4 (b) 1.

8.1.2.2 Measurement of dilution of the sambles from the downstream section - The samples taken downstream are then analyzed in the colori- meter. The calibration which has already been made makes it possible for the dilution ratio to be obtained, and the average value of the dilution ratio for the samples as well as their standard deviation can be computed. It is recommended that a further calibration of the apparatus, with the help of standard solutions, be effected at this stage. Variation between the two calibrations, which may occur because of drift in the calorimeter or instability of the reagent, shall not be greater than 1 percent.

8.1.2.3 Recommended precautions

a) Because of the influence of the passage of time on the diluted solution of sodium dichromatc, it is necessary, where an accuracy of the order of fr 1 percent is required, to make the analysis within a few hours of the actual test. If, however, some days must elapse between the sampling and the making-up of standard dilutions and their analysis, it is recommended that the following precautions be taken against the risk of changes in the dichromate solutions:

1) The samples should be kept in the dark, and

2) Atleast three standard dilutions, approximately equal to the dilution in the actual sample, should be made on the spot and kept under exactly the same conditions as all the other samples.

b) Any error caused by than estimated by comparing I trated solution prepared a is particularly recommen matter which would reduc

c) If, as happens frequently, contains solid matter in settle before the colorimet! of turbidity between the I and the standard dilution this, a correction curve cl calorimeter after the intro as a function of the readin reagent. This last readi technique may lead to errs or in certain cases where recommended that, in sue than usual be used in t taken downstream be dilu after stirring.

d) An alternative procedure samples before the additib. standard dilutions made I water taken from the cro( point. For the standard s before addition of the rea absorption of sodium dich this procedure is adopted.

8.2 Conductivity Method

8.2.1 Choice of Chemical Substant Measured - The substance to be stated in 5, and, in particular, it order that when present in smal resulting increase in conductivil accuracy.

8.2.1.1 The chemical salt n ( NaCl ) which has a solubility of to 50 mg/l, the relation of its condl In aqueous solutions, at a tempers vity is 1*86pS/cm for 1 mg/l. This which the salt is dissolved alread) of dissolved salts.

18

b)

C>

d

1s : 9163 ( Part i ) - 1919

Any error caused by changes in the samples with time can be estimated by comparing standard dilutions of the same concentrated solution prepared at different times. This latter precaution is particularly recommended when the water contains organic matter which would reduce the sodium dichromate.

If, as happens frequently, the water is not perfectly clear and contains solid matter in suspension, the samples shall be left to settle before the calorimetric analysis. In addition, any difference of turbidity between the samples taken at the sampling section and the standard dilutions shall be taken into account. To do this, a correction curve can be obtained of the reading of the calorimeter after the introduction of the reagent into the solution as a function of the reading obtained before the addition of the reagent. This last reading characterizes the turbidity. This technique may lead to errors if the degree of turbidity is very high, or in certain cases where the particles are extremely small. It is recommended that, in such cases, a much stronger concentration than usual be used in the injection solution and that samples taken downstream be diluted with equal amounts of clear water after stirring.

An alternative procedure for silted waters is to filter the dilute samples before the addition of the reagent and also to filter the standard dilutions made up from the concentrated solution and water taken from the cross section upstream from the injection point. For the standard solutions, filtering is again carried out before addition of the reagent. It is necessary to check that any absorption of sodium dichromate by the silt is negligible before this procedure is adopted.

8.2 Conductivity Method

8.2.1 Choice of Chemical Substance and Minimum Concentration that can be Measured - The substance to be injected shall satisfy the requirements stated in 5, and, in particular, its ionizing power has to be very high in order that when present in small concentrations in natural water, the resulting increase in conductivity can be measured with sufficient accuracy.

8.2.1.1 The chemical salt most frequently used is sodium chloride ( NaCl ) which h as a solubility of 360 g/l at 15°C. For concentrations up to 50 mg/l, the relation of its conductivity to concentration is almost linear, In aqueous solutions, at a temperature of 18%, the variation of conductivity is 1*86$/cm for 1 mg/l. This coefficient is smaller when the water in which the salt is dissolved already contains an appreciable concentration of dissolved salts.

IS : 9163 ( Part 1 ) - 1979

8.2.1.2 The conductivity of the natural water also affects the minimum limiting value of concentration of salt at which the method can be used with confidence.

8.2.1.3 The minimum concentrations of sodium chloride which may be measured with an error of the order of & 1 percent by the conductivimetric method are given below as a function of the conductivity of natural water.

8.2.1.4 A very sensitive conductivity-meter shall be used, capable of detecting variations of conductivity of about l/l0 000 of that of the natural water. The instrument shall be capable of balancing any capacity component while the conductivity measurement is being made.

Conductivity of natural water 1 000 200 100 20

( $/cm ) Mimimum concentration of 10 2 1 1

NaCl, measurable with f 1 percent accuracy (mg/l )

Corresponding relative varia- 2 2 2 10 tion of conductivity of water as a percentage

8.2.2 A4cthod.r of Anulysis

8.2.2.1 Comparison of the variations of conductivity - By continuous recording of the conductivity of the water during the test, the mean value of the variation of conductivity during the steady-regime condition produced by the injection of salt can be determined. The procedure used is either interpolation in time from the conductivity measurements made before and after the passage of the saline solution, or measurement during the test of the conductivity of the water at a section upstream from the injection section. This latter procedure is the only one possible if the variations of conductivity with the passage of time are relatively large, but allowance must then be made for the time during which the water passes from the upstream measurement sect,ion to the sampling cross section. A calibration of the conductivity-meter is then made in the laboratory. This operation consists of determining the variation in the conductivity produced by adding to a sample of natural water taken upstream of the injection cross section a measured quantity of the concentrated solution so as to obtain a known dilution ratio approximating to that obtained in the test. The conductivity of this sample is measured immediately before and after the addition of the concentrated solution, all other conditions of mcasurcment being identical. Because of the great influrricc of temperature on the specific conductivity of electrolytes, it is necessary to carry out the calibration at the same temperature as that measured during the test. For a solution of common salt, the

relative variation of conductivity is SO that, for accurate measuremc difference between that in the test more than -& O*l”C. In addition, the test and during the calibratior Such accuracies can be obtained The same instrument shall be u measurements. The temperature during the period of calibration al duction of the concentrated solutic are usually used, the variation of chloride is nearly proportional to I cal of the dilution ratio; it is reca of the conductivity curve be plotte dilution ratio,

8.2.2.2 Comparison of Conduct; to compare the conductivity of the passage of the salt cloud with that ned by diluting, in known propo water taken upstream from the inj

This method is not generally differences of natural resistivity be

It is found that the natural cl the period of time of a test an’ completely all variations of resisti sampling and the time of analysis.

8.3 Method of Volumetric Che

8.3.1 Choice of Chemical Substan Measured - As an example of the to determine the dilution ratio, dichromate is given in detail. SC detected by titration against silver to determine the end-point of t to serious error if there is any natural water as the technique sample, and the colour change of Electrical means of detecting tht conditions.

8.3.1.1 Volumetric analysis out with better than h 1 percent ; as low as 3 mg/l.

20

IS 19163 ( Part I ) - 1979

relative variation of conductivity is of the order of 2 to 3 -percent per *C SO that, for accurate measurement ( & 1 percent ), the temperature difference between that in the test and that in the calibration shall not be more than f 0.1%. In addition, the fluctuations of temperature during the test and during the calibration shall be checked to within + O*Ol”C. Such accuracies can be obtained with instruments, such as thermistors. The same instrument shall be used for both the field and laboratory measurements. The temperature shall not vary by more than O*dl@C during the period of calibration and in particular over the period of introduction of the concentrated solution. In the range of concentrations which are usually used, the variation of conductivity of the solutions of sodium chloride is nearly proportional to the concentration, that is to the reciprocal of the dilution ratio; it is recommended therefore that the variation ofthe conductivity curve be plotted as a function of the reciprocal of the dilution ratio.

8.2.2.2 Comfiarison of Conductivities - The quickest method would be to compare the conductivity of the water of the river at the moment of passage of the salt cloud with that of a series of standard solutions obtained by diluting, in known proportions, the concentrated solution with water taken upstream from the injection section.

This method is not generally of high accuracy, because of the great differences of natural resistivity between different samples.

It is found that the natural conductivity can vary appreciably during the period of time of a test and, furthermore, it is difficult to avoid completely all variations of resistivity in the samples between the time of sampling and the time of analysis.

8.3 Method of Volumetric Chemical Analysis

8.3.1 Choice of Chemical Substance and Minimum Concentration that can be Measured - As an example of the volumetric method of chemical analysis to determine the dilution ratio, that used for the detection of sodium dichromate is given in detail. Sodium chloride may also be used and detected by titration against silver nitrate ( Ag NOs ) using an indicator to determine the end-point of the titration, but the results are liable to serious error if there is any appreciable colouring matter in the natural water as the technique normally requires evaporation of the sample, and the colour change of the indicator is then difficult to detect. Electrical means of detecting the end-point may be used under these conditions.

8.3.1.1 Volumetric analysis of sodium dichromate can be carried out with better than & 1 percent accuracy on final sample concentrations -as low as 3 mg/l.

21

IS : 9163 ( Part I ) - 1979

8.3.1.2 As stated in 8.1 the use of sodium dichromate is advanta- geous as natural waters do not generally contain chromium ions. It should be noted, however, that this chemical is easily affected by reducing agents in the water and initial tests shall be carried out to ensure that there will be no serious effect on the samples between the time of sampling and the time of analysis. The samples can be re-oxidized by boiling with persul- phate but the time of analysis is considerably lengthened if this procedure is adopted.

8.3.2 Method of Analysis -The principle of the method is that the total amount of the salt present in the sample should react with the ferrous ammonium sulphate solution used for the titration, the end-point of the reaction being determined by colour change in the indicator, .N-phenylanthranilic acid, which has been added to the sample. The amount of ferrous ammonium sulphate used in the titration is directly related to the amount of dichromate present in the sample.

8.3.2.1 The recommended reagents are as follows:

a) Ferrous ammonium @hate - A stock N/IO solution of this salt should be prepared by weighing out 39.22 g of pure ferrous ammonium sulphate [ FeS04 ( NH4 ),S04.6H,O ] and dissolving it in 500 ml of distilled water containing 10 ml of concentrated sulphuric acid ( HzS04 ) having a relative density of 1.84. The resulting solution is then made up to 11. For titrating against the samples, an N/200 solution should be made up by taking 100 ml of the stock solution and diluting to 2 1 with distilled water.

b) Xphenylanthranilic acid indicator - 1.07 g of fl-phenylanthranilic acid ( CGH,.NHCsHI COOH ) should be dissolved in 20 ml of a 5 percent sodium carbonate ( Na,COa ) solution and made up to 1 1. 0.5 ml of this solution should be used for each titration.

8.3.2.2 To 500 ml of the sample which is being analyzed, 10 ml of concentrated sulphuric acid shall first be added and the sample stirred continuously throughout the titration.

8.3.2.3 The general precautions recommended in 7.1.2 should be followed and the standard solutions should be prepared as described therein. The results of the t&rations of the standard solutions can be plotted and the best straight line drawn through the points with the condition that it passes through a point close to the origin of the axes of the graph. The position of this point depends upon the amount of indicator used in the titration but, ifthe recommended procedure is followed, the point should be about + O-5 ml on the ferrous ammonium sulphate axis.

8.3.2.4 It is important that 2 analysis of all the samples. This the acid and indicator added and nium sulphate is added during the

8.3.2.5 The samp&s taken at in the same manner by titration : the dilution ratio can be determi mining the average of the titration deviation of the samples may be c

9. ERRORS IN FLOW MEASU

9.1 No measurement of a physic; which may be associated with c terrors in the standardizing equip lack of repeatability of the measul ted by repeated measurements an equipment is used for the mea reduce the error caused by randon age of n repeated measurements i! the individual measurements.

9.2 When considering the possib charge in an open channel, it is I! but an analysis of the individual : obtain the discharge can be made likely 6‘ tolerance “. The tolerant measurement may be defined stal calculated value, which, on can expected to include the true value dence limits can be taken to equa’ squares of the deviations, provid various measurements are small ai

If the different independent are x1, x2, x3,... and the standard 8x1, 8x2, &s,... then the tolerance c

where ag azL ax1 ax2”’

are partial deriv,

the manner in which Q is a functio.

22

.

IS : 9163 ( Part g ) - i979

S-3.2.4 It is important that a standardized procedure be used in the analysis of’all the samples. This applies particularly to the amounts of the acid and indicator added and to the rate at which the ferrous ammonium sulphate is added during the titration.

8.3.2.5 The samphs taken at the sampling cross section are analyzed in the same manner by titration against ferrous ammonium sulphate and the dilution ratio can be determined from the standard graph by determining the average of the titration values of all the samples. The standard deviation of the samples may be computed from the individual results.

9. ERRORS IN FLOW MEASUREMENTS

9.1 No measurement of a ~physical quantity can be free of uncertainties which may be associated with either systematic deviations caused by errors in the standardizing equipment or a random scatter caused by a lack of repeatability of the measuring equipment. The former is unaffected by repeated measurements and can only be reduced if more accurate equipment is used for the measurements. Repetition does, however, reduce the error caused by random scatter. The likely error of the average of n repeated measurements is 6 times smaller than that of any of the individual measurements.

9.2 When considering the possible error of any measurement of the discharge in an open channel, it is not possible to predict this error exactly, but an analysis of the individual measurements which were required to obtain the discharge can be made and a statistical estimate made of the likely “ tolerance “. The tolerance with 95 percent confidence limits on a measurement may be defined statistically as the bandwidth around the calculated value, which, on an average of 19 times out of 20, can be expected to include the true value. The tolerance with 95 percent confidence limits can be taken to equal twice the square root of the sum of the squares of the deviations, provided that the individual deviations on the various measurements are small and independent.

If the different independent quantities which have been measured are x1, sc2, x3,... and the standard deviations on these measurements are sx1, 8x2, sxs,... then the tolerance on the measurement of flow is :

2x= Q 2 J($-!$%)“+ (g.gy+...

where a& a’ 3x1 ax2”’

are partial derivatives, the values of which depend on

the manner in which Q is a function of ~1, ~2 . . . .

23

tS t 9163 ( Part I) - i979

If an independent quantity y has been obtained by n repeated measurements, as for example when the average dilution ratio of a large number of samples has been determined, and if the results of these measurements areyl,yZ p, then the standard deviation of the average ye of the n measurements may be defined by:

I i=n

2/c (Y1- YO I2 8Yyo = i=l

n(n- 1)

If the value of an independent quantity x1 or xs or._. is based on a~single measurement, then the standard deviation cannot be calculated statistically. For the purposes of this Indian Standard, however, the standard deviation of such a ilow measurement may be taken as half the estimated maximum possible error.

From the preceding considerations, it follows that the overall tolerance, with 95 percent confidence limits, for a flow measurement by the constant-rate injection method, is given by :

the partial derivatives being unity here. The standard deviations on the quantities q and J%‘, which themselves may have been determined by -any of the diKerent methods contained in this lndian Standard, shall be calculated by assessing all the possible sources of both systematic and random errors in each of the measurements used in the computation of these quantities and combining the squares of these with the appropriate partial derivatives, to obtain the total squared standard deviations, that is

($)‘and (g)”

Numerical examples of the estimation of the overall tolerance on a flow measurement are given in Appendices A and B. The examples are only illustrations of the procedure and individual figures should be estimated by the user for each particular test.

24

APPE: ( Cla;

EXAMPLE OF CALCULA ESTIMATION OF TOLE

VOLUMETRIC CE

A-l. MEASUREMENT OF INJE

A-l.1 The concentrated solution through a sprinkler system into a river, at an average rate of 5.960 X timing the discharge recorded b meter and the uncertainties were ae

a) Standard deviation on mt b) Standard deviation on me

beginning and end of tin c) Standard deviation on

measurement

Overall standard deviation o:

+= 2/( 0.15 )” + (

A-2. MEASUREMENT OF SAIL!

A-2.1 Ten 500-ml samples were ta in the~cross section at the first sam: second station farther downstream described in 8.3 and two typical se

Ferrous Ammon r----------- Station 1

20.70

20.77 20.65 20.67

20.75

20.63 20,80 20.77 20.70 20.72

.

IS : 9163 ( Part I ) - 1979

APPENDIX A ( Clause 9.2 )

EXAMPLE OF CALCULATION OF FLOW RATE AND ESTIMATION OF TOLERANCES IN THE CASE OF

VOLUMETRIC CHEMICAL ANALYSIS

A-l. MEASUREMENT OF INJECTION RATE

A-l.1 The concentrated solution of sodium dichromate was injected through a sprinkler system into a narrow turbulent cross section of the river, at an average rate of 5.960 x 10-4ms/s. This rate was determined by timing the discharge recorded by a calibrated positive-displacement meter -and the uncertainties were as follows:

a) Standard deviation on meter calibration : 0.15 percent b) Standard deviation on meter readings at : 0.07 percent

beginning and end of timed period c) Standard deviation on stop - watch : 0.02 percent

measurement Overall standard deviation on injection rate:

-$= 2/( 0.15 )2 + ( 0.07 )s + ( O-02 )a-= O-17 percent

A-2. MEASUREMENT OF SAMPLE CONCENTRATIONS

A-2.1 Ten 500-ml samples were taken at 1 min intervals from five points in the cross section at the first sampling station and from one point at the second station farther downstream. These samples were analyzed as described in 8.3 and two typical sets were as follows:

Ferrous Ammonium Sulphate ( ml ) ~~~~~__~~_~~A~_~~_~_~__~~ Station 1 Station 2

20.70 20.57 20.77 20.67 20’65 20.75 20.67 20.52 20.75 20.57 20.63 20.72 20.80 20.70 20.77 20.67 20.70 20.40 20.72 20.70

25

IS : 9163 (Part I ) - 1979

The average values from these and the other sampling points were 20.716 - 20,627 - 20,492 - 20.601 - 20.624 - 20.775 ml and the standard deviations of these samples ranged from 0.28 to 0.63 percent showing that variations between the different points were not significant. The overall mean value of these averages is 20,639 ml with a standard deviation of 0.10 ml, The standard deviation of the mean is therefore:

0.10 ml -~ %G-

=I 0.04 ml which means 0.2 percent

Reading errors were & 0.01 ml and do not add to the possible error.

A-3. MEASUREMENT OF STANDARD DILUTIONS

A-3.1 Samples of the concentrated solution were diluted with water taken upstream of the injection cross section during the test. Calibrated pipettes and burettes were used and it was estimated that the possible error in making up these dilutions was rfr 0.5 percent. The standard deviation was half this value, therefore, and equal to 0.25 percent.

Nine standard dilutions were titrated in the same manner as the samples, and gave the following results:

Dilution Ratio 105/Dilution Ferrous Ammonium Ratio Sulphate

ml

f 2.96 19.62 33 784 4 2.96 19.50

(2.96 19.69

C 3.08 20’31 32 468 { 3.08 20.43

(3.08 20.21

(-3.20 21.16 31 250 < 3.20 21.05

i3.20 20.98

A-3.2 The straight line passing through the centroid of these values at 3.08 x 10~sand 20,329 ml, and a point zero and + 0.5 ml to allow for chromate absorption by the indicator, was determined. The equivalent dilution ratio for the mean value of the samples ( 20.64 ml ) was found to be 31 969. The standard deviation of the individual standards measurements from the straight line is 0.09 ml, equivalent to 0.45 percent. The dilution ratios calculated from the straight line will be subject to an uncertainty therefore, the standard deviation being :

$$L= O* 15 percent

26

In this particular instance, v

value of chromate absorption by the by 0.1 percent and its standard de1

A-3.3 Similar tests carried out a another sample of the concentrat injection period gave similar fluct for the samples ( 20.64 ml ) was in that previously estimated by 0.3 pel

The average of these two esti and an additional standard devia cover the uncertainty of variations i: The overall standard deviation OX

therefore:

J?= q’( 02)s. + (025 )” i- ( JV

= 0.39 percent

A-4. FLOWRATE AND TOLER,

A-4.1 The volume flow rate of the

where

Q = qN = 5.960 x 1

= 19.02 ms/s.

The overall tolerance, with ! twice the overall standard deviatio

26Q - = i 22/( 0.17 )“+ (0.: Q

and hence

Q = 19.02 ms/s to the n

IS : 9163 ( Part I ) - 1979

In this particular instance, variation of even 0.5 ml on the zero value of chromate absorption by the indicator only affects the value 3 1969 by 0.1 percent and its standard deviation may be taken as 0.05 percent.

A-3.3 Similar tests carried out on standard dilutions prepared from another sample of the concentrated solution taken at the end of-the injection period gave similar fluctuations. The calculated dilution ratio for the samples ( 20.64 ml ) was in this case 31 865 which differed from that previously estimated by 0.3 percent.

The average of these two estimates of the I dilution ratio was 31 917 and an additional standard deviation of 0.15 percent was estimated to cover the uncertainty of variations in samples of the concentrated solution. The overall standard deviation on the estimate of the dilution ratio is therefore:

&IQ - = 1’( 0.2)s + (0.25 )” + (0.15 )2 + (0.05 )2 + (0.15)s JV

= 0.39 percent

A-4. FLOWRATE AND TOLERANCE

A-4.1 The volume flow rate of the river is equal to Q

where

Q = pjY= 5.960 x 10-4 x 31917

= 19.02 ms/s.

The overall tolerance, with 95 percent confidence limits, is equal to twice the overall standard deviation:

2 ‘2 = i 22/( 0.17 )“+( 0*39)2= f 0.84 percent

and hence

Q = 19.02 ms/s to the nearest 0.84 percent

I8 : 9163 ( Part 1) - 1979

APPENDIX B

( Clause 9.2 )

EXAMPLE OF CALCULATION OF FLOW RATE AND ESTIMA- TION OF TOLERANCES IN THE CASE OF

CONDUCTIVIMETRIC ANALYSIS

B-l. MEASUREMENT OF THE CHANGE PRODUCED BY THE INJECTION

B-l.1 An injection of a concentrated solution of sodium chloride produced the following changes in the electrical resistance of the measuring cell which was placed in the flowing water:

a) Average resistance of~the water, measured at a number of points in the sampling cross section during the steady regime of the salt cloud

& - 5 86OQ

corresponding temperature, 8s = 3.50%

b) Average resistance of the water in its natural state

R,, = 6 495Q

corresponding temperature, 80 = 3.50%

The values R,, and 0s were determined from measurements made before and after the passage of the salt solution.

The average change in conductance was therefore:

A (l/R) = -&-+ 16.68/&S 2

After the measurements were made, the standard deviation of the fluctuations in the resistance during the passage of the salt solution, as variations from the mean values, was estimated to be 1OQ. The change in resistance produced by the injection was 635Q, so that the ,standard deviation was equal to 10/635= 1.6 percent. The number of readings which were made was of the order of 10 so that the standard deviation on the average change in conductance A ( l/R ) was:

1.6 1 luoX 7z= O-5 percent

A small difference between the temperatures e2 and fl,, could intro- duce a relatively large error. Assuming that the equality of the two tempertures was verified to within &O.O5”C, and that the relative change

in the conductivity of the water w corresponding error in the conduct

1 2.5 5 860 x 100~

28

from which the possible error is

0.21 i_6T68 =

The corresponding standard 1 this, that is 0.6 percent.

B-2. MEASUREMENT OF DILI

B-2.1 The standard dilution value value was in the neighbourhood of measurement. The possible error i and pipettes, was -f 0.5 percent gi cent.

Seven samples of the concent injection equipment were diluted : cell as in the test on the river. Th obtained.

SAMPLE

TABLE 1 ANAL

RESISTAECE R’. OB

NATURAL WATER

sz

RESIS R'2

L TAX: s DILU

9

6 550.3 584 6 557.2 584 6 559.4 58: 6 556.6 5 8’

6 558.4 5 84 6 556-l 5 84 6 561.4 5 8l

The average change in con standard dilution was therefore:

A ( l/R’

and the standard deviation on this

IS : 9163 ( Part I ) - 1979

in the conductivity of the water with temperature is 2.5 percent/C, the corresponding error in the conductance of the salt water would be:

1 2.5 -x 100 x 0.05 = 0.21 ps 5 860

from which the possible error is

0.21 i@ti8-

- I.2 percent

The corresponding standard deviation can be taken as equal to half this, that is 0.6 percent.

B-2. MEASUREMENT OF DILUTION STANDARDS

B-2.1 The standard dilution value .NI equal to 10 100 was chosen, as this value was in the neighbourhood of the dilution ratio realized during the measurement. The possible error in this dilution operation, using flasks and pipettes, was & O-5 percent giving a standard deviation of 0.25 percent.

Seven samples of the concentrated solution taken at the outlet of the injection equipment were diluted and analyzed using the same resistance cell as in the test on the river. The following table shows the readings obtained.

SAMPLE

TABLE 1 ANALYSIS OF SOLUTION

RESISTAXCE R’. 06

NATURAL WATER

B

6 550.3

6 557.2 6 559.4 6 556.6

6 558.4

6 556-l

6 561.4

RESISTANCE R’s OF

STANDARD DILUTION

sz

5 840.5

5 846.9 5 850.6

5 845.1

5 848.1

5 846.2

5 849.0

A (l/R’) = 1 1 -- .-_

R’* R’,,

KS 18.55

18.53

18.47

15.57

18.52

18.52

18.56

TEMPERATUBE 8’

OC

3.50 3.50 3.50 3.50 3-50 3.50

3.50

The average change in conductance, A ( l/R’ ) produced by the standard dilution was therefore:

A ( l/R’ ) = 18.53~s

and the standard deviation on this average was 0.1 percent.

29

5S : 9163 ( Part I ) - 1979

The influence of temperature on the readings was as follows :

4

b)

The temperature 8’ ofthe samples during the calibration should be equal to the temperature of the water in the river during the test ( t&, 8, ) and these temperatures were measured with a possible error of + O*l”C. The deviation in the conductance, n ( l/R’ ), varies 2.5 percent/% so that the corresponding percentage error would be & ~0.25 percent equivalent to a standard deviation of 0.12 percent.

On the other hand, it is necessary to take account of the variation in temperature during a single calibration, that is the difference between the temperature of the natural water ( at the moment of measuring RrO ) and that of the water when the~salt has been added ( reading RI2 ). The possible error with the equipment used was 0.01%; supposing that this error was systematic for all seven samples analyzed and assuming a relative change in conductivity of 2.5 percent/% the corresponding error in the conductance l/Rrz of the diluted solution would be:

This gives a relative possible error on the change A ( l/R’ ) equal to

f 0.043 18.5 = -& 0.2 percent. The standard deviation corresponding to this

is 0.1 percent.

B-3. CALCULATION OF THE DILUTION RATIO N

B-3.1 The changes in conductance in the range considered are proportional to the concentration and inversely proportional to the dilution ratio so that:

.? _ A ( l/R’ ) j”“~- A( l/R)

where N = 11 2 10 with a standard deviation given by:

x= 1/ (0.5 )” + ( 0.6 )” + ( 0*25)2 + ( 0.1 )” + ( 0.12 )” -i- ( O.l)2

= 0.85 percent.

30

B-4. CALCULATION OF THE I

B-4.1 The injection flow rate wa red with a standard deviation of C

The flow rate ofthe river is :

Q = q.N= 3.70 x 10-J x

The overall tolerance, with ’ twice the standard deviation:

2 @ r= rf: 22/( 0.85 )” +

so that finally

Q_= 4.15 ms/s to the

fS:9163(PartI)-1819

B-4. CALCULATION OF THE FLOW RATE

B-4.1 The injection flow rate was O-370 l/s. This flow rate was measured with a standard deviation of0.2 percent.

The flow rate of the river is given by the relation: Q = qN= 3.70 x IO-4 x 11210 = 4_15ms/s.

The overall tolerance, with 95 percent confidence limits, is equal to twice the standard deviation:

2 sQ c= f 22/( 0.85 )Z + ( 0.2 j2 = f. 1.8 percent

so that finally

Q= 4.15 ms/s to the nearest I.8 percent.

31

ON

FLUID FLOW MEASUREMENT

1191-1971 Glossary of terms and symbols used in connection with the measurement of liquid flow with a free surface (jirst reuision )

1192-1959 Velocity-area methods for measurement of flow of water in open channels

1193-1959 Methods for measurement of flow of water in open channels using notches, weirs and flumes

1194-1960 Forms for recording measurement of flow of water in open channels

2912-1964 Recommendation for liquid flow measurement in open channels by slope area method ( approximate method )

2913-1964 Recommendation for determination of flow in tidal channels

2914-1964 Recommendations for estimation of discharges by establishing stage-discharge relation in open channels

2915-1964 Instructions for collection of data for the determination of error in measurement of flow by velocity area methods

2951 ( Part I )-1965 Recommendation for estimation of Aow of liquids in closed conduits : Part I Head loss in straight pipes due to frictional resistance

2951 (Part II )-1965 Recommendation for estimation of flow of liquids in closed conduits : Part II Head loss in valves and fittings

2952 (Part I )-1964 Recommendation for methods of measurement of fluid flow by means of orifice plates and nozzles : Part I Incompressible fluids

2952 (Part II )-1975 Recommendation for methods of measurement of fluid flow by means of orifice plates and nozzles : Part II Compressible fluids

3910-1966 Current meters ( cup type ) for water flow measurement

3911-1966 Surface floats

3912-1966 Sounding rods

3913-1966 Suspended sediment load samplers

3917-1966 Scoop type bed material samplers

3918-1966 Code of practice for use of current meter (cup type) for water flow measurement

4073-1967 Fish weights

4080-1967 Vertical staff gauges

4477 ( Part I )-1967 Methods of measurement of fluid flow by means of venturi meters : Part I Liquids

4477 (Part II )-I975 Methods of measurement of fluid flow by means of venturi meters : Part II Compressible fluids

4858-1968 Velocity rods

4890-1968 Methods for measurement of suspended sediment in open channels

6059-1971 Recommendation for liquid flow measurement in open channels by weirs and flumes - weirs of finite crest width for free discharge

6062-1971 Method of measurement of flow 0; wave flume-fall

6063-1971 Method of measurement of flow of wave flume

6064-1971 Sounding and suspension equipmen

6330-1971 Recommendation for liquid flow n and flumes-end depth method fc channels with a free overfaIl ( apprc

6339-1971 Methods of analysis of concentratio. gravity of sediment in streams and

19:

6062-1971 Method of measurement of flow of’ water in open channels using standing wave flume-fall

6063-1971 Method of measurement of flow of water in open channels using standing wave flume

6064-1971 Sounding and suspension equipment

6330-1971 Recommendation for liquid flow measurement in open channels by weirs and flumes-end depth method for estimation of flow in rectangular channels wifh a free overfall ( approximate method )

6339-1971 Methods of analysis of concentration, particle size distribution and specific gravity of sediment in streams and canals

INTERNATiONAl. SYSTEM

Base Units Quantity

Length

Mass

Time

Electric current

Thermodynamic temperature

Luminous intensity

Amount of substance

Supplementary Units

Quantity

Plane angle

Solid angle

Derived Units

Quantity

Force

Energy

Power

Flux

Flux density

Frequency

Electric conductance

Electromotive force

Pressure, stress

Unit

metre

kilogram

second

ampere

kelvin

candeia

mole

Unit

radian

steradian

Unit

newton

joule

watt

weber

tesla

hertz

siemens

volt

Pascal

OF UNITS (SI UNITS)

Symbol

m

kg S

A

K

cd

mol

Symbol

rad

sr

Symbol

N

J

W

Wb

T

HZ

S

V

Pa

Definition

1 N=i kg.m/s*

1 J =l N.m

1 W=lJ/s

1 Wb= 1 V.s

1 T = 1 Wb/me

1 Hz=1 c/s (s-l)

1 S = 1 A/V

1 V=lW/A

1 Pa= 1 N/m2

c

IS 9163-1 (1979): Dilution Methods for Measurement of ...

Documents

Transcript of IS 9163-1 (1979): Dilution Methods for Measurement of ...