PUBLISHED DOCUMENT PD ISO/TR

12767:2007

Measurement of fluid
flow by means of
pressure differential
devices — Guidelines
on the effect of
departure from the
specifications and
operating conditions
given in ISO 5167

ICS 17.120.10

12&23<,1*:,7+287%6,3(50,66,21(;&(37$63(50,77('%<&23<5,*+7/$:
PD ISO/TR 12767:2007

National foreword

This Published Document is the UK implementation of ISO/TR 12767:2007. It

supersedes BS ISO/TR 12767:1998 which is withdrawn.
The UK participation in its preparation was entrusted by Technical Committee
CPI/30, Measurement of fluid flow in closed conduits, to Subcommittee
CPI/30/2, Differential pressure methods.
A list of organizations represented on this committee can be obtained on
request to its secretary.
This publication does not purport to include all the necessary provisions of a
contract. Users are responsible for its correct application.

This Published Document was Amendments issued since publication

published under the authority
of the Standards Policy and
Strategy Committee Amd. No. Date Comments
on 28 September 2007

© BSI 2007

ISBN 978 0 580 55215 1

 PD ISO/TR 12767:2007

TECHNICAL ISO/TR
REPORT 12767

Second edition
2007-09-01

Measurement of fluid flow by means of

pressure differential devices —
Guidelines on the effect of departure from
the specifications and operating
conditions given in ISO 5167
Mesurage du débit des fluides au moyen d'appareils déprimogènes —
Lignes directrices relatives aux effets des divergences par rapport aux
spécifications et aux conditions de fonctionnement données dans
l'ISO 5167

Reference number
ISO/TR 12767:2007(E)
ii
 PD ISO/TR 12767:2007

Contents Page

Foreword............................................................................................................................................................ iv
Introduction ........................................................................................................................................................ v
1 Scope ..................................................................................................................................................... 1
2 Normative references ........................................................................................................................... 1
3 Terms and definitions........................................................................................................................... 2
4 Symbols and abbreviated terms ......................................................................................................... 2
5 Effect of errors on flowrate calculations ............................................................................................ 3
5.1 General................................................................................................................................................... 3
5.2 Quantifiable effects............................................................................................................................... 3
6 Effects of deviations in construction.................................................................................................. 4
6.1 Orifice-plate edge sharpness .............................................................................................................. 4
6.2 Thickness of orifice edge..................................................................................................................... 5
6.3 Condition of upstream and downstream faces of orifice plate........................................................ 6
6.4 Position of pressure tappings for an orifice ...................................................................................... 6
6.5 Condition of pressure tappings .......................................................................................................... 7
7 Effects of pipeline near the meter ....................................................................................................... 7
7.1 Pipe diameter ........................................................................................................................................ 7
7.2 Steps and taper sections ..................................................................................................................... 8
7.3 Diameter of carrier ring ........................................................................................................................ 8
7.4 Undersize joint rings .......................................................................................................................... 11
7.5 Protruding welds................................................................................................................................. 11
7.6 Eccentricity.......................................................................................................................................... 11
8 Effects of pipe layout.......................................................................................................................... 14
8.1 General................................................................................................................................................. 14
8.2 Discharge coefficient compensation ................................................................................................ 14
8.3 Pressure tappings............................................................................................................................... 16
8.4 Devices for improving flow conditions............................................................................................. 17
9 Operational deviations ....................................................................................................................... 17
9.1 General................................................................................................................................................. 17
9.2 Deformation of an orifice plate.......................................................................................................... 17
9.3 Deposition on the upstream face of an orifice plate ....................................................................... 19
9.4 Deposition in the meter tube ............................................................................................................. 23
9.5 Orifice-plate edge sharpness ............................................................................................................ 23
9.6 Deposition and increase of surface roughness in Venturi tubes .................................................. 24
10 Pipe roughness ................................................................................................................................... 26
10.1 General................................................................................................................................................. 26
10.2 Upstream pipe ..................................................................................................................................... 27
10.3 Downstream pipe ................................................................................................................................ 30
10.4 Reduction of roughness effects........................................................................................................ 30
10.5 Maintenance ........................................................................................................................................ 30
Bibliography ..................................................................................................................................................... 32

iii
PD ISO/TR 12767:2007

Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.

International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.

The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.

In exceptional circumstances, when a technical committee has collected data of a different kind from that
which is normally published as an International Standard (“state of the art”, for example), it may decide by a
simple majority vote of its participating members to publish a Technical Report. A Technical Report is entirely
informative in nature and does not have to be reviewed until the data it provides are considered to be no
longer valid or useful.

Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.

ISO/TR 12767 was prepared by Technical Committee ISO/TC 30, Measurement of fluid flow in closed
conduits, Subcommittee SC 2, Pressure differential devices.

This second edition cancels and replaces the first edition (ISO/TR 12767:1998), which has been technically
revised.

iv
 PD ISO/TR 12767:2007

Introduction
ISO 5167 (all parts) specifies methods for flowrate measurement using pressure differential devices.
Adherence to ISO 5167 (all parts) results in flowrate measurements whose uncertainty lies within specified
limits. If, however, a flow-metering installation departs, for whatever reason, from the conditions specified in
ISO 5167 (all parts), the specified limits of uncertainty may not be achieved. Many metering installations exist
where these conditions either have not been or cannot be met. In these circumstances, it is usually not
possible to evaluate the precise effect of any such deviations. However, a considerable amount of data exists
which can be used to give a general indication of the effect of non-conformity to ISO 5167 (all parts), and it is
presented in this Technical Report as a guideline to users of flow-metering equipment.

v
blank
 PD ISO/TR 12767:2007

Measurement of fluid flow by means of pressure differential

devices — Guidelines on the effect of departure from the
specifications and operating conditions given in ISO 5167

1 Scope
This Technical Report provides guidance on estimating the flowrate when using pressure differential devices
constructed or operated outside the scope of ISO 5167.

Additional tolerances or corrections cannot necessarily compensate for the effects of deviating from ISO 5167
(all parts). The information is given, in the first place, to indicate the degree of care necessary in the
manufacture, installation and maintenance of pressure differential devices by describing some of the effects of
non-conformity to the requirements; and in the second place, to permit those users who cannot comply fully
with the requirements to assess, however roughly, the magnitude and direction of the resulting error in
flowrate.

Each variation dealt with is treated as though it were the only one present. Where more than one is known to
exist, there may be unpredictable interactions and care has to be taken when combining the assessment of
these errors. If there is a significant number of errors, means of eliminating some of them have to be
considered. The variations included in this Technical Report are by no means complete and relate largely to
examples with orifice plates. An example with Venturi tubes has been placed at the end of its section. There
are, no doubt, many similar examples of installations not conforming to ISO 5167 (all parts) for which no
comparable data have been published. Such additional information from users, manufacturers and any others
may be taken into account in future revisions of this Technical Report.

2 Normative references
The following referenced documents are indispensable for the application of this document. For dated
references, only the edition cited applies. For undated references, the latest edition of the referenced
document (including any amendments) applies.

ISO 5167-1:2003, Measurement of fluid flow by means of pressure differential devices inserted in circular
cross-section conduits running full — Part 1: General principles and requirements

ISO 5167-2:2003, Measurement of fluid flow by means of pressure differential devices inserted in circular
cross-section conduits running full — Part 2: Orifice plates

ISO 5167-3:2003, Measurement of fluid flow by means of pressure differential devices inserted in circular
cross-section conduits running full — Part 3: Nozzles and Venturi nozzles

ISO 5167-4:2003, Measurement of fluid flow by means of pressure differential devices inserted in circular
cross-section conduits running full — Part 4: Venturi tubes

1
PD ISO/TR 12767:2007

3 Terms and definitions

For the purposes of this document, the terms and definitions given in ISO 5167-1 and the following apply.

3.1
square edge
angular relationship between the orifice bore of the flow-measurement device and the upstream face, when
the angle between them is 90° ± 0,3°

3.2
sharpness
radius of the edge between the orifice bore of the flow-measurement device and the upstream face

NOTE The upstream edge of the orifice bore is considered to be sharp when its radius is not greater than 0,000 4d,
where d is the diameter of the orifice bore.

4 Symbols and abbreviated terms

For the purposes of this Technical Report, the symbols given in Table 1 apply.

Table 1 — Symbols and units

Dimensions
M: mass
Symbol Quantity represented SI units
L: length
T: time
c Percentage change in discharge coefficient [ ≡ 100( ∆C / C )] dimensionless
C Discharge coefficient dimensionless
Cc Contraction coefficient dimensionless
d Diameter of orifice or throat of primary device at operating conditions L m
D Upstream internal pipe diameter at operating conditions L m
D1 Carrier ring diameter L m
D2 Orifice-plate support diameter L m
e Relative uncertainty dimensionless
E Orifice-plate thickness L m
Ee Orifice thickness L m
k Uniform equivalent roughness L m
Distance of upstream pressure tapping from upstream face of plate divided by
L1 dimensionless
pipe bore, D
Distance of downstream pressure tapping from downstream face of plate
L′2 dimensionless
divided by pipe bore, D
qm Mass rate of flow MT−1 kg/s
r Orifice-plate edge radius L m
Red Reynolds number based on throat bore of device dimensionless
ReD Reynolds number based on upstream pipe diameter dimensionless
u Local axial velocity LT−1 m/s
uCL Centreline axial velocity LT−1 m/s
U Mean axial velocity LT−1 m/s
Y Modulus of elasticity of orifice-plate material ML−1T−2 Pa
β Diameter ratio, (= d/D) dimensionless
∆p Differential pressure ML−1T−2 Pa
∆p y Differential pressure required to reach orifice-plate yield stress ML−1T−2 Pa

2
 PD ISO/TR 12767:2007

Table 1 (continued)

Dimensions
M: mass
Symbol Quantity represented SI units
L: length
T: time
ε Expansibility (expansion) factor dimensionless
λ Friction factor dimensionless
ρ Fluid density ML−3 kg/m3
ρ1 Fluid density at the upstream pressure tapping ML−3 kg/m3
σy Yield stress of orifice-plate material ML−1T−2 Pa

5 Effect of errors on flowrate calculations

5.1 General

In this Technical Report, the effects of deviations from the conditions specified in ISO 5167 (all parts) are
described in terms of changes in the discharge coefficient, ∆C, of the meter. The discharge coefficient, C, of a
pressure differential device is given by Equation (1):

4q m (1 − β 4)
C= (1)
επ d 2 (2∆p ρ 1)

The sharp edge of an orifice plate ensures separation of the flow and consequently contraction of the fluid
stream to the vena contracta. Defining the contraction coefficient, Cc, as the ratio of the flow area to the
geometric area the orifice produces Cc ≈ 0,6, which mainly accounts for the discharge coefficient, C ≈ 0,6.

The effect of change in the discharge coefficient is illustrated by the following example.

Consider an orifice plate with an unduly rounded edge. The result of this will be to reduce the separation and
increase Cc, leading in turn to reduced velocities at the vena contracta. The observed differential pressure will
therefore decrease. From Equation (1), it can be seen that the discharge coefficient would therefore increase.
Alternatively, as Cc increases, so does C. If no correction is made for this change in C, the meter reading will
be less than the actual value.

It can therefore be concluded that:

a) an effect which causes an increase in discharge coefficient will result in a flowrate reading lower than the
actual value if the coefficient is not corrected;

and conversely,

b) an effect which causes a decrease in discharge coefficient will result in a flowrate reading higher than the
actual value if the coefficient is not corrected.

5.2 Quantifiable effects

When the user is aware of such effects and they can be quantified, the appropriate discharge coefficient can
be used and the correct flowrate calculated. However, the precise quantification of these effects is difficult and
so any flowrate calculated in such a manner should be considered to have an increased uncertainty.

Except where otherwise stated, an additional uncertainty factor, equivalent to 100 % of the discharge
coefficient correction, should be added arithmetically to that of the discharge coefficient when estimating the
overall uncertainty in the flowrate measurement.

3
PD ISO/TR 12767:2007

6 Effects of deviations in construction

6.1 Orifice-plate edge sharpness

Orifice plates that do not have the specified sharpness of the inlet edge (edge radius r u 0,000 4d in
accordance with 5.1.7.2 of ISO 5167-2:2003), will have progressively increasing discharge coefficients as the
edge radius increases. Tests have shown that the effect on the discharge coefficient, C, is to increase it by
0,5 % for r/d of 0,001, and by about 5 % for r/d of 0,01. This is an approximately linear relationship (see
Figure 1 and Reference [1]). These values apply particularly to Red values above 300 000 and for β values
below 0,7, but they can be used as a general guide for other values.

Measurement techniques for edge radius are available, but in general it is better to improve the edge
sharpness to the required value rather than to attempt to measure it and make appropriate corrections.

The effect of nicks in orifice plates has also been measured in Reference [1].

Key
1 National Engineering Laboratory (NEL, UK) tests —– D = 300 mm
2 ISO limit —– r = 0,000 4d
3 others
4 NEL
5 D = 50 mm (Reference [56])
6 D = 100 mm (Reference [56])
7 D = 150 mm (Reference [34])
8 D = 75 mm (Reference [57])
9 D = 100 mm (Reference [58])

c change in discharge coefficient

r/d radius ratio

Figure 1 — Effect of edge radius on discharge coefficient

4
 PD ISO/TR 12767:2007

6.2 Thickness of orifice edge

For orifice plates, the increase in discharge coefficient due to excessive thickness of the orifice edge (see
5.1.5 of ISO 5167-2:2003) can be appreciable. With a straight-bore orifice plate in a 150 mm pipe, the
changes in discharge coefficient shown in Figure 2 were obtained (see Reference [2]).

Key
1 section of an orifice plate
2 symbol
3 limit of standard

c change in discharge coefficient

Ee/D orifice thickness to upstream internal pipe diameter ratio

Figure 2 — Change in discharge coefficient as a function of orifice thickness

5
PD ISO/TR 12767:2007

6.3 Condition of upstream and downstream faces of orifice plate

The upstream face should be flat and smooth. Excessive roughness leads to an increase in the discharge
coefficient. Tests have indicated that a surface roughness of 0,000 3d will cause an increase in discharge
coefficient of the order of 0,1 %. Since the requirement for edge sharpness is r u 0,000 4d, an increase in
plate roughness will make it difficult to define the edge sharpness or to confirm that the sharp edge
requirement has been met.

Local damage to the upstream face or edge of an orifice plate does not adversely affect the discharge
coefficient, provided that the damage is kept as far away from the pressure tapping as possible (see
Reference [1]). The discharge coefficient is much less sensitive to the surface condition of the downstream
face of the plate (Reference [1]).

Large-scale lack of flatness, e.g. “dishing”, leads to flow-measurement errors. A “dishing” of 1 % in the
direction of flow causes the reading to be below the actual value, i.e. an increase in C of about 0,2 % for
β = 0,2 and of about 0,1 % for β = 0,7. Distortion against the direction of flow also causes errors which could
be either positive or negative depending on the amount of distortion.

6.4 Position of pressure tappings for an orifice

6.4.1 General

Values of the orifice-plate discharge coefficient for the three standard tapping positions (corner, flange, D
and D/2) can be calculated using Equation (4) of ISO 5167-2:2003 (see Reference [55]). Where the tapping
positions fall outside the tolerances permitted in ISO 5167-2 for the three positions, the discharge coefficient
may be estimated as described in 6.4.2. It should be emphasized that an additional uncertainty factor needs to
be associated with the use of non-standard tapping positions.

6.4.2 Calculation of discharge coefficient

Calculate the actual values of L1 and L′2 . The discharge coefficient can be estimated only if L1 u 1 and
L′2 u 0,47.

Using the actual values of L1 and L′2 , estimate the discharge coefficient using Equation (4) of
ISO 5167-2:2003.

6.4.3 Estimation of additional uncertainty

If tappings lie between the flange and the corner tappings, the additional uncertainty, e, expressed as a
percentage, can be estimated from:

CF
e = 25 −1 (2)
C CT

where

CF is the discharge coefficient for flange tappings;

CCT is the discharge coefficient for corner tappings.

If tappings lie between the D and D/ 2 tappings and the flange tappings, the additional uncertainty, e,
expressed as a percentage, can be estimated from:

C D and D / 2
e = 25 −1 (3)
CF

6
 PD ISO/TR 12767:2007

where

CD and D/2 is the discharge coefficient for D and D/2 tappings.

6.4.4 Example

Consider an orifice meter with β = 0,6, ReD = 106, D = 250 mm and tappings at 0,15D upstream and
downstream of the plate.

To estimate the discharge coefficient, use Equation (4) of ISO 5167-2:2003 with L1 = L′2 = 0,15.

The tappings in this example lie between the flange tapping and D and D/ 2 tapping positions. From
Tables A.8 and A.2, respectively, of ISO 5167-2:2003: CF = 0,605 1; CD and D/2 = 0,607 0. Therefore

0,605 1
e = 25 − 1 = 0,078
0,607 0

The uncertainty in the discharge coefficient is 0,5 % (see 5.3.3.1 of ISO 5167-2:2003).

Therefore, overall uncertainty is 0,5 + 0,078 ≈ 0,6 % (i.e. the uncertainties have simply been added together).

6.5 Condition of pressure tappings

Experience has shown that large errors can be created by pressure tappings which have burrs or deposits on,
or close to, the edge where the tapping penetrates the pipe wall. This is particularly the case where the
tappings are in the main flow stream, such as throat tappings in nozzles or Venturi tubes, where small burrs
can give rise to significant percentage errors. Upstream corner tappings and downstream tappings in relatively
dead zones are much less liable to cause this problem.

The installation shall be inspected before use and at regular intervals to ensure that these anomalies are not
present.

7 Effects of pipeline near the meter

7.1 Pipe diameter

The internal diameter of the pipe upstream and downstream of the primary device should always be measured
to ensure that it is in accordance with 6.4 of ISO 5167-2:2003, 6.4 of ISO 5167-3:2003 or 6.4.1 of
ISO 5167-4:2003. Errors in the upstream internal diameter measurement cause errors in the calculated
flowrate, which are given by:

δq m −2β 4 δD
= (4)
qm (1 − β 4) D

These errors become significant for large β, e.g. with β = 0,75, a positive 1 % error in D will cause a negative
1 % error in qm.

The downstream pipe is far less critical, as for an orifice plate, an ISA 1932 nozzle or a long radius nozzle its
diameter need only be within 3 % of that of the upstream pipe (see 6.4.6 of ISO 5167-2:2003 or 6.4.6 of
ISO 5167-3:2003) and for a Venturi nozzle or a Venturi tube its diameter need only be W 90 % of the diameter
at the end of the divergent section (see 6.4.6 of ISO 5167-3:2003 or 6.4.1.3 of ISO 5167-4:2003).

7
PD ISO/TR 12767:2007

7.2 Steps and taper sections

Sudden enlargements of the pipe in the vicinity of the primary device should always be avoided as large errors
in flow measurement result from their use. Similarly, tapering sections of pipe can lead to significant errors, as
can be seen from Table 2 which gives the order of errors to be expected when an orifice plate with corner
tappings is immediately preceded or followed by a taper piece.

The information in Table 2 indicates that a taper piece divergent in the direction of flow, and placed
immediately upstream, is not recommended, since discharge-coefficient increases of up to 50 % result. On the
other hand, a convergent taper piece, whether installed before or after the orifice plate, and provided it is not
of a steeper angle than those shown, results in coefficient changes of generally less than 2 %.

Table 2 — Effect of taper pieces

Order of the discharge coefficient

Position of orifice plate β
change to be expected
%

a) Immediately downstream from a divergent taper piece

0,4 + 10

0,7 + 50

b) Immediately downstream from a convergent taper piece

0,4 − 0,5

0,7 −2

c) Immediately upstream from a convergent taper piece

0,4 0 to − 1

0,7 +1

7.3 Diameter of carrier ring

The requirements for the sizing and concentric mounting of carrier rings for orifice plates and nozzles are
specified in 6.4 and 6.5 of ISO 5167-2:2003, 6.4 and 6.5 of ISO 5167-3:2003 and Figure 4 of ISO 5167-2:2003.
If the requirement of 6.5.4 of ISO 5167-2:2003 and 6.5.4 of ISO 5167-3:2003 (i.e. that the centred carrier ring
should not protrude into the pipe) is not met, relatively large flow-measurement errors will be introduced.
Figure 3 shows such an installation and Figure 4, using the same notation, shows the approximate errors
introduced for the given conditions, where a is the width of the portion of the carrier ring upstream of the
upstream face of the orifice plate or nozzle. It is emphasized that in arriving at these errors, the internal carrier
ring diameter, D1, and not the diameter of the main line, has been used in determining the calculated flowrate
and is to be used for D in determining the correction factor when making use of the values shown.

Where the carrier is oversize, experimental results indicate that for β = 0,74 a carrier 11 % oversize and
extending 0,05D upstream from the plate increased the discharge coefficient by approximately 0,5 %.
However, for a similar geometry but with β = 0,63, no effect was found.

8
 PD ISO/TR 12767:2007

a) Orifice plate

b) Nozzle

Key
1 flow

Figure 3 — Carrier having internal diameter, D1, less than pipe diameter, D

9
PD ISO/TR 12767:2007

a) Orifice plate

b) Nozzle

Key
1 a = 0,2D to 0,3D

c change in discharge coefficient

β diameter ratio

Figure 4 — Effect of incorrect carrier diameter

10
 PD ISO/TR 12767:2007

7.4 Undersize joint rings

When the inside diameter of a joint ring or gasket is smaller than the pipe diameter, especially on the
upstream side of an orifice plate or nozzle, very large flow-measurement errors may occur. The magnitude
and sign of the effect in relation to the measurement of flowrate is dependent on the combination of a number
of variables, e.g. the thickness of the joint ring upstream of the orifice plate, the extent of its protrusion into the
flow, its position relative to the orifice plate and pressure tappings, and the degree of roughness of the
upstream pipe.

7.5 Protruding welds

The effect of an undressed circumferential weld protruding into the pipe bore adjacent to the primary device
will be similar to that of an undersize joint ring. Such an effect may arise from the fitting of a weld-neck flange,
and the magnitude of the effect will depend on the height uniformity, or otherwise, of the protruding weld, and
its position in relation to the single or multiple pressure tapping arrangement employed to measure the
differential pressure across the primary device. To quantify the resulting error in a specific situation is difficult
without a direct calibration.

It should be noted that seamed pipe may be used, provided that the internal weld bead is parallel to the pipe
axis throughout the entire length of the pipe required, to satisfy the installation requirements for the primary
device being used. Any weld bead shall not have a height greater than the permitted step in diameter. Unless
an annular slot is used, the seam shall not be situated within any sector of ± 30° centred on any individual
pressure tapping to be used in conjunction with the primary device. If an annular slot is used, the location of
the seam is not significant. If spirally wound pipe is used, then it shall be machined to a smooth bore.
(See 7.1.4 of ISO 5167-1:2003.)

7.6 Eccentricity

The requirements for concentric mounting of the device are given in 6.5.3 and 6.5.4 of ISO 5167-2:2003, 6.5.3
and 6.5.4 of ISO 5167-3:2003 and 6.4.3 of ISO 5167-4:2003. The geometric measure of eccentricity is the
distance between the pipe and orifice-plate centrelines and is often expressed as a percentage of the pipe
diameter, D. Deviations from the permitted eccentricity values for the mounting of an orifice plate relative to
the upstream and downstream pipe sections will result in errors in the measurement of flowrate.
Figure 5 shows the eccentric mounting of an orifice plate in a sideways direction relative to the upstream
pipeline. The displacement is to the right and the eccentricity is a combination of the dimensional tolerances
arising from the bolt-hole pitch-circle diameter, the bolt diameter, the bolt-hole diameter and the outer
diameter of the orifice plate.

Experimental evidence on the effects of eccentricity is limited, but it has been shown that for orifice plates, the
effect on discharge coefficient is a function of β , pipe size and roughness, pressure-tapping type, location and
magnitude, as well as the position of the orifice centre relative to the pressure tapping.

Experimental work indicates that the errors due to eccentricity increase in general with β . For β = 0,2 and
eccentricity up to 5 % of D, discharge coefficient increases are unlikely to exceed 0,1 %. For larger β , the
changes are best shown graphically as in Figure 6.

Below 3 % eccentricity, the error varies with type of tappings and direction of eccentricity. The meter is least
sensitive to eccentricity perpendicular to the tappings. Above 3 % eccentricity, errors for all tappings and
directions increase rapidly.

NOTE No data are available for corner tappings, but the errors are probably similar to those for flange tappings since
the above data were obtained from a test line with D = 150 mm.

A further effect of eccentric positioning of an orifice plate is an increased unsteadiness of the differential
pressure signal obtained. Observations have shown, for example, a marked increase in differential pressure
reading fluctuations with increasing eccentricity for all values of β between 0,4 and 0,7.

11
PD ISO/TR 12767:2007

Because of the number of variants contributing to the effect of eccentricity on the measurement of flow, the
effect is difficult to quantify. Every effort should be made to restrict eccentricity to less than 3 % of D,
particularly in the direction of the tappings.

The effect may be minimized by employing four equally-spaced upstream and downstream tappings on the
flowmeter, as illustrated in Figure 1 of ISO 5167-1:2003. The pressure lines from these are then coupled in the
widely used triple-T tapping arrangement in order to obtain an average differential pressure reading.

As a general guide, it may be assumed that the effects of eccentric mounting for multi-tapped nozzles will be
less than those for orifice plates of equivalent β . Venturi tubes are less likely to be installed off-centre.

NOTE Combined installation faults: it is recommended that errors arising from the combined effects of eccentricity,
carrier ring steps, etc., are not taken into account additively. The total possible error will be governed by the strongest of
the effects present.

Key
1 bolt-hole pitch circle
2 flange centreline
3 orifice bore
4 orifice-plate outside diameter
5 flange bore
6 pipe inside diameter
7 pipe centreline
8 orifice centreline
9 eccentricity

Figure 5 — Possible orifice-plate eccentricity resulting from specified tolerances on bolt hole, bolt
hole pitch circle, pipe outside diameter and flange bore

12
 PD ISO/TR 12767:2007

a) β = 0,75 b) β = 0,66

c) β = 0,57
Key
1 D and D/2 tappings
2 flange tappings
3 ± 0,3 %
4 (away from tapping 1) ← → (towards tapping 1)
5 ± 0,5 %
6 ± 0,7 %

c change in discharge coefficient

x eccentricity

Figure 6 — Discharge coefficient error vs. eccentricity for an orifice plate with D and D/2
and flange tappings

13
PD ISO/TR 12767:2007

8 Effects of pipe layout

8.1 General

Minimum values of the straight lengths required between the primary device and various upstream fittings are
given in 6.2 of ISO 5167-2:2003, 6.2 of ISO 5167-3:2003, and 6.2 of ISO 5167-4:2003. Minimum straight
lengths are given both for zero additional uncertainty and for 0,5 % additional uncertainty in the discharge
coefficient.

When the minimum requirements for even 0,5 % additional uncertainty cannot be satisfied, the user should
make a correction to compensate for the change in the discharge coefficient and should also increase the
value of the percentage uncertainty.

Corrections and additional uncertainties for square-edged orifice plates with corner, flange and D and D/ 2
tappings are given in Tables 3 and 4 for a variety of upstream pipe bends and fittings. Shifts in columns 4
and 5 are particularly variable, depending on the exact details of the double bend.

Additional data on shifts in orifice-plate discharge coefficients for a large number of upstream fittings are given
in References [3-6].

8.2 Discharge coefficient compensation

8.2.1 Corrections

The discharge coefficient can be corrected using the data in Table 3 as illustrated in the following examples:

a) percentage change in coefficient is + 1,1 %, therefore the coefficient should be multiplied by 1,011;

b) percentage change in coefficient is − 2,3 %, therefore the coefficient should be multiplied by 0,977.

14
 PD ISO/TR 12767:2007

Table 3 — Percentage change in discharge coefficient, c, when the straight pipe lengths before the
orifice are less than those specified in ISO 5167-2

Upstream
straight β Type of fitting (for details of nomenclature, see key)
length
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

4D 0,5 −1,4 −1,4 −0,5 +2,9 +2,9 −0,4 +8,2 — +0,2 +0,2 −1,0 −0,8 +0,3 +0,5 +0,2 — —
0,6 −2,3 −2,2 −1,1 +1,7 +1,3 −1,2 +8,5 — −0,2 −0,3 −2,4 −1,7 +0,3 0 −0,2 — —
0,7 −3,8 −3,2 −1,8 +0,1 +0,4 −2,1 +8,2 — −0,9 −0,7 −4,4 −2,3 +0,3 −0,6 0 — —
0,8 −5,6 —b −2,6 −2,4 —b −3,1 +3,4 — −2,2 —b −7,5 —b +0,3 −1,3 —b — —
8D 0,5 −0,7 −0,7 −0,3 +2,4 +2,4 0 +6,3 +6,4 −0,2 −0,2 −0,6 −0,4 —a −0,2 −0,2 −0,8 −0,7
0,6 −1,4 −1,2 −0,7 +1,4 +1,2 −0,7 +5,6 +6,1 −0,6 −0,4 −1,3 −1,2 — a −0,7 −0,8 −1,3 −1,2
0,7 −2,2 −1,9 −1,2 +0,3 +0,4 −1,3 +4,4 +6,1 −1,1 −0,8 −2,1 −1,9 +0,1 −1,2 −1,2 −1,7 −1,7
0,8 −3,2 — b −1,8 −1,7 — b −2,0 +2,3 — b −1,9 — b −3,1 — b +0,1 −1,8 — b −2,0 −2,1
12D 0,5 —a —a — a +2,0 +2,0 0 +5,5 +5,5 -0,2 −0,1 −0,4 −0,3 — a −0,3 −0,2 — —c
0,6 −0,8 −0,8 −0,4 +1,2 +1,0 −0,4 +3,9 +4,3 −0,4 −0,3 −0,9 −0,9 — a −0,7 −0,6 −0,8 −0,8 c
0,7 −1,4 −1,4 −0,8 +0,3 +0,3 −0,8 +2,6 +3,2 −0,8 −0,7 −1,3 −1,3 — a −1,1 −1,0 −1,2 −1,1
0,8 −2,0 — b −1,3 −1,3 — b −1,3 +1,5 — b −1,3 — b −1,7 — b — −1,5 — b −1,5 −1,4
16D 0,5 —a —a — a +1,7 +1,7 0 +5,1 +5,0 −0,1 0 −0,2 −0,2 — a −0,2 −0,2 — —
0,6 —a —a −0,3 +1,1 +0,9 −0,3 +3,5 +3,6 −0,3 −0,2 −0,6 −0,6 —a −0,4 −0,4 — —
0,7 −0,8 −0,8 −0,5 +0,3 +0,3 −0,5 +2,1 +2,4 −0,5 −0,5 −0,9 −1,0 — a −0,7 −0,6 −0,9 —
0,8 −1,3 — b −0,7 −1,1 — b −0,8 +0,8 — b −0,8 — b −1,0 — b — −1,0 — b −1,2 —
a Refer to Table 3 of ISO 5167-2:2003.
b For D and D/2 tappings, discharge coefficient changes measured for β > 0,75 should not be used for interpolation to give discharge
coefficient changes for β u 0,75, as the downstream tapping is in the pressure recovery region if L′2 > 2(1 − β ).
c For a concentric expander 0,5D to D over a length of D to 2D refer to Table 3 of ISO 5167-2:2003.

Key

No. Type of upstream fitting Type of No. Type of upstream fitting Type of
tappings tappings

1 Single short radius 90° bend Corner, flange 10 Butterfly valve, fully open D and D/2

2 Single short radius 90° bend D and D/2 11 Butterfly valve, 52° open Corner, flange

3 Two 90° bends in the same plane, U- or S- All 12 Butterfly valve, 52° open D and D/2
configuration, 0D to 10D spacer

4 Two 90° bends at right angles, no spacer Corner, flange 13 Gate valve, fully open All

5 Two 90° bends at right angles, no spacer D and D/2 14 Gate valve, ⅔ open Corner, flange

6 Two 90° bends at right angles, 5D to 11D All 15 Gate valve, ⅔ open D and D/2
spacer

7 Two 90° mitre bends at right angles, no spacer Corner, flange 16 Gate valve, ¼ open and globe valve All

8 Two 90° mitre bends at right angles, no spacer D and D/2 17 Symmetrical enlargement, tapered or abrupt All

9 Butterfly valve, fully open Corner, flange

15
PD ISO/TR 12767:2007

Table 4 — Formulae for additional uncertainty in the orifice discharge coefficient, to be used with the
percentage changes given in Table 3, for all tapping arrangements

Type of upstream fitting Additional uncertainty formulae

Piezometer ring Single tapping a
e.g. Triple-T

Single short radius 90° bend. Bend radii 1D to 1,5D 0,5(1 + 0,6⏐c⏐) 0,5 + 0,6⏐c⏐
Two 90° bends, U- or S-configuration, in same plane 0,5(1 + ⏐c⏐) 0,5 + ⏐c⏐
Two 90° bends at right angles, no spacer (where X is the 0,5(1 + ⏐c⏐) + (10D/X) 0,5 + ⏐c⏐ + (10D/X)
distance from the orifice plate to the nearest bend)
Two 90° bends at right angles, 5D to 11D spacer 0,5 + ⏐c⏐ 0,5(1 + 3⏐c⏐)
Two 90° mitre bends at right angles, no spacer 0,5 + ⏐c⏐ 0,5(1 + 3⏐c⏐)
Butterfly valve, fully open 0,5 + ⏐c⏐ 0,5(1 + 3⏐c⏐)
Butterfly valve, 52° open 0,5 + ⏐c⏐ 0,5(1 + 3⏐c⏐)
Gate valve, fully open 0,5(1 + ⏐c⏐) 0,5 + ⏐c⏐
Gate valve, ⅔ open 0,5(1 + ⏐c⏐) 0,5 + ⏐c⏐
Gate valve, ¼ open and globe valve 0,5 + ⏐c⏐ 0,5 + ⏐c⏐
Symmetrical restriction or enlargement, tapered or abrupt 0,5 + ⏐c⏐ 0,5 + ⏐c⏐
a The tapping axis should be at right angles to the plane of the nearest upstream bend.

8.2.2 Additional uncertainty

The formulae for calculating the additional percentage uncertainty in discharge coefficient are given in Table 4
for each type of fitting. This is in addition to the basic uncertainty in the discharge coefficient of: 0,5 % for
0,2 u β u 0,6, (1,667β − 0,5) % for 0,6 < β u 0,75. In deriving the formulae, the quantity of data, its
consistency and corroboration from different sources have been taken into account. Their use is illustrated in
the following examples.

a) If the equation to be applied is:

(
e = 0,5 1 + c ) (5)

where ⏐c⏐ is the modulus of percentage change (i.e. the magnitude irrespective of sign) and if the change in the
coefficient is + 1,4 %, then e = 1,2 %.

b) If the equation to be applied is:

e = 0,5 + c (6)

and if c = - 2,8 %, then e = 3,3 %.

8.3 Pressure tappings

It is emphasized that the change in the coefficient when D and D/ 2 tappings are used is often different from
those obtained with corner or flange tappings.

When the upstream straight pipe length is less than that required for zero additional uncertainty, it is
recommended that multiple tappings with triple-T connections, as shown in Figure 1 of ISO 5167-1:2003, be
used. If single tappings are used, their axes should be at right angles to the plane of the nearest upstream
bend.

16
 PD ISO/TR 12767:2007

8.4 Devices for improving flow conditions

Flow conditioners should be used where asymmetric or swirling flow has to be measured. Descriptions of
various flow conditioners are provided in Annex B of ISO 5167-2:2003. Even where the installation
requirements of ISO 5167-2:2003 [6.3 or Annex B (informative)] cannot be met, the use of a flow conditioner
may reduce errors especially in swirling flow.

9 Operational deviations

9.1 General

Metering systems that conform to ISO 5167 when new or recently maintained may be subject to a significant
degradation in accuracy over the passage of time.

This degradation may result from several causes:

a) deformation of the orifice plate;

b) deposition on the upstream face of an orifice plate;

c) deposition in the meter tube;

d) rounding of the orifice-plate edge;

e) deposition in the pressure tappings;

f) deposition and increase of surface roughness in a Venturi tube.

An indication of the effect of sources of error a) to d) and f) is given in 9.2 to 9.6.

It cannot be emphasized too strongly that the continued achievement of high accuracy requires the
expenditure of considerable effort. In particular, regular inspection and maintenance are essential. Inspection
periods depend on the nature of the fluid being metered and on the manner of operation of the system in
which the meter is installed, and can only be determined from experience.

9.2 Deformation of an orifice plate

9.2.1 General

An orifice plate may be said to be deformed when it deviates beyond the 0,5 % value specified in 5.1.3.1 of
ISO 5167-2:2003. The deformation may be in the upstream or downstream direction, and possible causes are
defects in manufacture, poor installation or incorrect use. Manufacturing and installation faults should be
rectified before use.

Deformation arising from the manner of use may be either temporary (elastic) or permanent (buckling). This is
discussed in References [7-9]. Information regarding the necessary thickness of orifice plates when metering
systems are being designed is given in 8.1.1.3 of ISO/TR 9464:1998[10].

17
PD ISO/TR 12767:2007

9.2.2 Elastic deformation

Elastic deformation arises when the differential pressure due to flow deforms the plate by a small amount in
the downstream direction, such that the induced stresses remain within the elastic limit of the plate material.
For a plate simply supported at its rim, a first approximation to the percentage increase in discharge
coefficient is given by:

2
100 ∆p ⎛ D2 ⎞ ⎛ a1D2 ⎞
c= ⎜ ⎟ ⎜ − a2 ⎟ (7)
Y ⎝ E ⎠ ⎝ E ⎠

where

a1 = β (0,135 − 0,155β )

a2 = 1,17 − 1,06 β 1,3

For American Iron and Steel Institute grades 304 or 316 stainless steel (ISO/TS 15510[11]), Y can be taken as
193 × 109 Pa.

In virtually all cases, the result of the deformation is to cause an increase in the discharge coefficient.

Errors due to elastic bending may be additional to those arising from initial lack of flatness. Only when the
combination of both effects results in a slope greater than 1 % under flowing conditions does the plate depart
from the requirements of ISO 5167-2.

Since the plate will return to its undeformed state when the flow is zero, elastic bending cannot be detected
during routine inspection of a metering system.

9.2.3 Plastic deformation

Where an orifice plate has been subjected to excessive differential pressures it may deform permanently.
When the deformation is known, the error may be estimated from Figure 7. Such deformation may occur
during over-rapid pressurization or venting of a line containing a compressible fluid, or through an abnormal
flow condition. It should be emphasized that a permanently deformed plate should be discarded.

The differential pressure required to reach orifice-plate yield stress, ∆py, may be estimated from:

2
⎛ E ⎞ ⎛ 1 ⎞
∆p y = σ y ⎜ ⎟ ⎜ ⎟ (8)
⎝ D 2 ⎠ ⎝ 0,681 − 0,651 β ⎠

18
 PD ISO/TR 12767:2007

Key
1 experimental
2 experimental
3 theoretical
4 theoretical
5 flow
6 D = 200 mm

c change in discharge coefficient

δ/D ratio of deflection to upstream internal pipe diameter

Figure 7 — Effect of orifice-plate deformation on flow-measurement accuracy

9.3 Deposition on the upstream face of an orifice plate

The effect of deposits on the upstream face of an orifice plate is similar to that of upstream face roughness
and always causes the discharge coefficient to increase.

Table 5 shows the effect of a uniform layer of sand one grain thick (grain size 0,4 mm) and the effect of
grease spots (each nominally 6,3 mm diameter and 2,5 mm high) on an orifice plate in a 100 mm diameter
meter tube measuring air at atmospheric pressure. Table 5 shows the importance of the annular region
around the entrance to the orifice bore. As this region is usually scrubbed by the flow, the actual errors are
probably smaller than those indicated.

Table 6 shows the effect of a layer of Audco1) grease on an orifice plate of thickness 6 mm and of diameter
ratio 0,6 in a 300 mm pipe. The pressure tappings were in the horizontal plane on the left of the drawings. The
pipe Reynolds number was approximately 107. For the tests, the orifice plate was removed from the carrier
and moved into a laboratory area where contamination was added with the plate in a horizontal position. The

1) Example of a product available commercially. This information is given for the convenience of users of this Technical
Report and does not constitute an endorsement by ISO of this product.

19
PD ISO/TR 12767:2007

contamination level given in Table 6 is that applied in the laboratory. The plate was then moved to a vertical
position to allow any liquid to drain off before being reinserted into the carrier in the test line. During the
subsequent 2 h test over a range of flowrates, the maximum increase in discharge coefficient (around the
beginning of the test) and the saturation increase in discharge coefficient once the increase had become
constant were recorded. Further information on the effect of contamination is given in References [12], [13].

Table 5 — Effect of deposits on β = 0,2 and β = 0,7 orifice plates

Change in discharge coefficient, c

Deposit β = 0,2 β = 0,7
%
Sand 1 sand quadrant + 1,0 + 0,8

2 sand quadrants + 2,8 + 1,9

3 sand quadrants + 3,9 + 2,4

4 sand quadrants + 6,2 + 3,0

4 sand quadrants with 6 mm ring + 0,3 + 0,3

removed from around orifice bore

Grease 4 grease deposits + 1,0 + 0,1

8 grease deposits + 2,8 + 1,3

16 grease deposits + 2,1 + 1,2

32 grease deposits + 2,6 + 0,6

20
 PD ISO/TR 12767:2007

Table 6 —Increase in discharge coefficient, c, of an orifice plate, D = 300 mm, β = 0,6, due to Audco2)
grease coating

Upstream face profile Cross-section Contaminant Maximum Saturation Coating

profile pattern increase in increase in thickness
discharge discharge
coefficient, c coefficient, c
% % mm

2,85 2,15 0,6

Full upstream
3,49 2,13 1,2
face
5,15 3,70 2,0

Full upstream 1,2

face tapered
1,30 1,00 to
off towards
centre 0,0

Full upstream
face with
1,00 0,61 1,2
10 mm clean
ring at centre

Full upstream
face with
0,60 0,50 1,2
20 mm clean
ring at centre

2) Example of a product available commercially. This information is given for the convenience of users of this Technical
Report and does not constitute an endorsement by ISO of this product.

21
PD ISO/TR 12767:2007

Table 6 (continued)

Upstream face profile Cross-section Contaminant Maximum Saturation Coating

profile pattern increase in increase in thickness
discharge discharge
coefficient, c coefficient, c
% % mm

Half circle at
top of
1,70 — 1,2
upstream
face

Half circle at
bottom of
1,80 1,30 1,2
upstream
face

Half circle
vertical near 2,20 0,85 1,2
tappings

Half circle
vertical away 3,70 2,84 1,2
from tappings

22
 PD ISO/TR 12767:2007

9.4 Deposition in the meter tube

In an exercise to simulate the effect of deposition in the meter tube, welding rods were stacked axially against
the upstream face of an orifice plate as shown in Figure 8. The rods caused an increase in the discharge
coefficient.

Key
1 orifice plate
2 welding rods laid axially against orifice plate (rod diameter = 0,016D, rod length = 0,5D)

c change in discharge coefficient

n number of welding rods

Figure 8 — Effect of welding rods in meter tube

Figure 9 shows the results of tests carried out to investigate the effect of a smooth horizontal build-up of
material in a meter run. When the material is below the dam height, the discharge coefficient increases. When
the build-up exceeds the dam height, the orifice bore cross-sectional area is reduced, leading to a decrease in
discharge coefficient.

9.5 Orifice-plate edge sharpness

9.5.1 Deterioration

The sharp edge of an orifice plate may deteriorate with time. Possible causes of this deterioration are:

a) erosion;

b) cavitation;

c) mechanical damage;

d) careless handling.

Orifice-plate discharge coefficients are sensitive to edge sharpness, and, where any of the above effects may
occur, regular quantitative inspection of the edge should be made.

23
PD ISO/TR 12767:2007

The effect of loss of sharp edge is described in 6.1.

Key
1 fraction of dam height

c change in discharge coefficient

2l/(D − d) obstruction level — fraction of dam height

Figure 9 — Effect of debris in meter tube (on both sides of plate)

9.5.2 Plate reversal

Particular care should be taken to ensure that bevelled orifice plates are inserted into the meter line with the
bevel on the downstream face.

In a 100 mm diameter meter, a plate bevelled at 45° and facing upstream can give the following percentage
increases in discharge coefficient:

a) 0,25 mm bevel width: c = 2,0;

b) 0,5 mm bevel width: c = 4,0;

c) 1,25 mm bevel width: c = 13,0.

These values should be taken simply as indicative of changes which can occur by incorrect installation and
should not be taken as precise.

9.6 Deposition and increase of surface roughness in Venturi tubes

9.6.1 General

Two effects may occur in a Venturi tube which has been in use for a period of time. These are deposition of
material in the contraction and the bore, and an increase in the surface roughness. Both effects result in a
decrease in the discharge coefficient and both effects may occur together. They are, however, considered
separately in 9.6.2 and 9.6.3.

24
 PD ISO/TR 12767:2007

9.6.2 Deposition

If material is deposited smoothly and uniformly in the contraction and bore of a Venturi tube, the change in
discharge coefficient, expressed as a percentage, c, may be estimated theoretically from the reduction in area
as:

c = − 400(l /d ) (9)

where l is the thickness, in metres, of the annular deposit in the bore of the Venturi tube.

9.6.3 Surface roughness

The chemical nature of the fluid and the material of the Venturi tube may be such that the surface roughness
of the Venturi tube increases with time (Reference [14]). This increase in roughness leads to a reduction in the
discharge coefficient. An indication of the error involved is given in Figure 10.

The rate of increase of surface roughness is dependent on the chemical reactions occurring in the metering
system, and is outside the scope of this Technical Report.

Key
1 effective roughness for new meters (k = 0,056 mm)

C Venturi tube discharge coefficient

k uniform equivalent roughness of Venturi tube

Figure 10 — Variation of Venturi meter coefficient with surface roughness

25
PD ISO/TR 12767:2007

10 Pipe roughness

10.1 General

The discharge coefficients given in 5.3.2.1 of ISO 5167-2:2003, 5.1.6.2, 5.2.6.2 and 5.3.4.2 of
ISO 5167-3:2003, and 5.5 of ISO 5167-4:2003 assume conformity to specified installation conditions. In
particular, the velocity profile immediately upstream of a primary device should be similar to that in the
experiments on which the equation is based.

The uniform equivalent pipe roughness, k, Reynolds number, ReD, and friction factor, λ, are interrelated and
determine the velocity profile (see Reference [15]). Experimental results suggest that the velocity profile,
defined as the ratio of the local axial velocity at y from the pipe wall, u, to the velocity at the centreline (y/R = 1),
uCL, can be described approximately by:

1
u⎛ y⎞ n
=⎜ ⎟ (10)
u CL ⎝ R ⎠

where

y is the distance from the pipe wall;

R is the pipe radius, D/2;

n is a number whose reciprocal gives the power (dependent on ReD and k/D) to which y/R must be
raised to give the velocity profile.

The ratio of the mean axial velocity, U, to the velocity at the centreline (y/R = 1), uCL, is then given by:

U 2n 2
= (11)
u CL ( n + 1)(2n + 1)

In smooth pipe, n increases with the Reynolds number (see Table 7). In fully rough pipe, n decreases with
increasing relative roughness (see Table 8).

A more uniform profile (U/uCL → 1) reduces the discharge coefficient and a more peaked profile (U/uCL
decreasing) increases C.

The extent to which the discharge coefficient varies is also influenced by β , being less for smaller β .

Table 7 — Values of n, U/uCL and λ for smooth pipe

ReD n U/uCL λ
4 × 103 6,0 0,791 0,04
2,3 × 104 6,6 0,807 0,025
1,1 × 105 7,0 0,817 0,0175
1,1 × 106 8,8 0,850 0,0115
2× 106 10 0,866 0,0105

26
 PD ISO/TR 12767:2007

Table 8 — Values of n, U/uCL and λ for rough pipe

R/k k/D n U/uCL λ

507 0,986 × 10−3 6 0,791 0,020
126 3,97 × 10−3 5 0,758 0,028
31 16,1 × 10−3 4 0,711 0,045

10.2 Upstream pipe

For an orifice plate the change in discharge coefficient, ∆C, due to pipe roughness is approximately
proportional both to the change in friction factor, ∆λ, and to β 3,5. The friction factor, λ, can be measured
directly, using:

2D∆p
λ= (12)
ρU 2 Z

where

D is the pipe diameter, in metres;

∆p is the pressure difference, in pascals, between two tappings;

ρ is the fluid density, in kilograms per cubic metre;

U is the mean axial velocity, in metres per second;

Z is the distance, in metres, between two tappings.

It is simpler to measure the arithmetic mean deviation of the roughness profile, Ra, to deduce the uniform
equivalent roughness, k ≈ πRa, and to calculate λ using the Colebrook-White equation {see 7.4.1.5 of
ISO 5167-1:2003 and Equation (20.35a) of Reference [15]}:

1 ⎛ 2k 18,7 ⎞
= 1,74 − 2lg ⎜ + ⎟ (13)
λ ⎜ Re D λ ⎟
⎝ D ⎠

If an estimate of the shift in discharge coefficient from Equation (4) of ISO 5167-2:2003 is desired, it is also
necessary to estimate the friction factor for the discharge coefficient equation. This has to be done on the
basis of the measured roughness or friction factor of the pipes in which the standard data (to which the
discharge coefficient equation was fitted) were collected; these are given in Table 9. Both k/D and λ depend
on ReD; k/D reduces with ReD because the higher Reynolds numbers generally occur in larger pipes, which are
generally relatively smoother.

Table 9 — Values of k/D and λ associated with Equation (4) of ISO 5167-2:2003

Pipe Reynolds No. 104 3 × 104 105 3 × 105 106 3 × 106 107 3 × 107 108
ReD

Ratio of uniform equivalent 1,75 1,45 1,15 0,9 0,7 0,55 0,45 0,35 0,25
roughness to pipe diameter
k/D × 104
Friction factor 0,031 0,024 0,018 5 0,015 5 0,013 0,011 5 0,010 5 0,010 0,009 5
λ

27
PD ISO/TR 12767:2007

Figure 11 gives measured and computed (using computational fluid dynamics) values of ∆C as a function of
β 3,5∆λ (see Reference [16] for complete references). The computed values and the European experimental
data were obtained using corner tappings. The North American experimental data (References [17-19]) were
obtained using flange tappings. For corner tappings, the following approximate equation to calculate the
change in discharge coefficient, ∆C, has been plotted:

∆C = 3,5β 3,5∆λ (14)

Key
1 computed [16]
2 experiment [20]
3 experiment [21]
4 experiment [22]
5 experiment [23]
6 experiment [24]
7 experiment [17]
8 experiment [18]
9 experiment [19]
10 ∆C = 3,5β 3,5∆λ

β diameter ratio, d/D

λ friction factor
∆C change in discharge coefficient

Figure 11 — The effect of rough pipe on discharge coefficient

28
 PD ISO/TR 12767:2007

From computational work, the effect of roughness on discharge coefficients using D and D/2 tappings has
been found to be about 25 % less than its effect on those using corner tappings. ∆C using flange tappings lies
between ∆C using corner tappings and ∆C using D and D/2 tappings.

In a swirling flow at a fixed Reynolds number, increasing the roughness of the upstream pipe reduces the swirl
at the flowmeter.

In extreme cases, roughness can change the diameter of the pipe and consequently β . The following
information (see Reference [20]) is for such an extreme case.

Figure 12 relates to orifice plates with corner tappings and gives the discharge coefficient change for pipes
with a roughness corresponding to surfaces encrusted with closely spaced spherical nodules. These averaged
6,3 mm in diameter reducing the effective diameter of the pipe by at least 6,3 mm. The changes shown would
be applied to the flow using the larger clean pipe diameter. [The dotted curve for β = 0,71 applies to a sanded
surface (0,5 mm to 1,0 mm diameter particles)].

Key
c percentage change in discharge coefficient
D upstream internal pipe diameter, in metres
e additional uncertainty

Figure 12 — The combined effect of abnormal roughness and diminution in pipe bore

Figure 13 shows the coefficient changes based on similar pipe conditions to those above, but calculated on
the smaller effective pipe diameter De (= D − 6,3 mm) and β e where β e = d/De. The changes due to sand
particles of about 1 mm diameter are about one-third of those given in Figure 12.

If a measurement of the flow needs to be made under such adverse conditions, the corrected discharge
coefficients given above should be used with an additional uncertainty of half the percentage change in
discharge coefficient.

29
PD ISO/TR 12767:2007

Key
c percentage change in discharge coefficient
De effective upstream internal pipe diameter, in metres
e additional uncertainty

Figure 13 — The effect of abnormal roughness on orifice plates

10.3 Downstream pipe

Even severe encrustation adjacent to the downstream side of an orifice plate has no significant effect on the
discharge coefficient.

10.4 Reduction of roughness effects

Experiments have shown that if a relatively short upstream length of pipe adjacent to the orifice plate is
cleaned to remove the encrustations, the error is significantly reduced. Table 10 gives recommendations
regarding the extent of such cleaning for various pipe sizes, values of β and types of roughness. For pipes of
internal diameter greater than 300 mm, fewer diameters of clean upstream pipe may be necessary.

10.5 Maintenance

In all cases of flow measurement by pressure differential meters, a cleaning routine for the pipe, the primary
device and the pressure tappings should be established to suit the particular conditions. Where reasonable
accuracy is required in the measurement of the flow of dirty fluids, installations should be designed for easy
cleaning of the upstream pipe to an extent shown in Table 10.

30
 PD ISO/TR 12767:2007

Table 10 — Recommendations regarding the extent of cleaning

Upstream β Type of Approximate change Amount of cleaning (in multiples of D) to

internal roughness in discharge obtain roughness errors not exceeding:
pipe coefficient without
diameter D cleaning the pipe
mm ±3% ±2% ±1% ± 0,5 % Nil
76 0,5 to 0,59 7,0 mm spheres 9 % to 15 % 3 to 4 4 to 5 5 to 15 15 to 20 > 20
0,71 7,0 mm spheres 40 % 4 to 10 10 to 20 20 to 25 25 to 30 > 30
0,71 Sand 7% 3 to 5 5 to 25 25 to 30 > 30
152 0,5 to 0,59 7,0 mm spheres 4 % to 8 % 3 to 5 5 to 12 12 to 20 > 20
0,71 7,0 mm spheres 17 % 2,5 to 4 4 to 15 15 to 25 25 to 30 > 30
0,71 Sand 4% 1 to 3 3 to 4 4 to 20 > 20
305 0,71 7,0 mm spheres 8% 2,5 to 4 4 to 6 6 to 15 > 15
0,71 Sand 2% 1 to 3 3 to 5 >5

31
PD ISO/TR 12767:2007

Bibliography

[1] HOBBS, J.M. and HUMPHREYS, J.S. The effect of orifice plate geometry upon discharge coefficient. Flow
Meas. Instrum., 1990, 1(3), pp. 133-140

[2] HUSAIN, Z.D. and TEYSSANDIER, R.G. The effects of plate thickness and bevel angle in a 150 mm line
size orifice meter. In: Kinghorn, F.C., Gibson, E.E., editors. Flow measurement in the mid 80's:
International conference: Papers, paper 5.1, National Engineering Laboratory, Glasgow, 1986

[3] MARTIN, C.N.B. Effects of upstream bends and valves on orifice plate pressure distributions and
discharge coefficients. National Engineering Laboratory, Glasgow, 1986. 50 p. (NEL Report 702)

[4] STUDZINSKI, W., KARNIK, U., LANASA, P., MORROW , T., GOODSON, D., HUSAIN, Z. and GALLAGHER, J.
White paper on “orifice meter installation configurations with and without flow conditioners”. American
Petroleum Institute, Washington, DC, 1997. 252 p. (Available as GRI Report 99/0262 from Gas
Technology Institute, Des Plaines, IL, USA)

[5] STUDZINSKI, W., W EISS, M., ATTIA, J. and GEERLIGS, J. Effect of reducers, expanders, a gate valve, and
two elbows in perpendicular planes on orifice meter performance. In: Flow Measurement 2001:
Creating efficiency across industry sectors, international conference, Peebles, UK, May 2001,
paper 3.1. National Engineering Laboratory, Glasgow, 2001

[6] W EISS, M., STUDZINSKI, W. and ATTIA, J. Performance evaluation of orifice meter standards for selected
T-junction and elbow installations. In: Proceedings of the 5th International Symposium on Fluid Flow
Measurement, paper 5.1. Washington, DC, April 2002

[7] JEPSON, P. and CHIPCHASE, R. The effect of plate buckling on orifice meter accuracy, J. Mech. Eng. Sci.
1975, 17(6)

[8] NORMAN, R., RAWAT, M.S. and JEPSON, P. Buckling and eccentricity effects on orifice metering
accuracy, In: Proceedings of the 1983 International Gas Research Conference, London, UK,
13-6 June 1983, A22-83, 1983

[9] NORMAN, R., RAWAT, M.S. and JEPSON, P. An experimental investigation into the effects of plate
eccentricity and elastic deformation on orifice meter accuracy. In: Spencer, E.A., editor. Proceedings
of the International Conference on the Metering of Natural Gas and Liquefied Hydrocarbon Gases,
London, UK, 1-2 February 1984, paper 3.3. Oyez, London, 1984

[10] ISO/TR 9464:1998, Guidelines for the use of ISO 5167-1:1991

[11] ISO/TS 15510, Stainless steels — Chemical composition

[12] PRITCHARD, M., NIAZI, A. and MARSHALL, D. Assessment of the effect of contamination on orifice plates.
In: Proceedings of the 11th Flomeko Conference on Flow Measurement of Gas and Liquid, Groningen,
12-4 May 2003, paper 6.2. Gasunie Research, Groningen, 2003 (on CD-ROM)

[13] PRITCHARD, M., MARSHALL, D. and W ILSON, J. An assessment of the impact of contamination on orifice
plate metering accuracy. In: Proceedings of the 22nd North Sea Flow Measurement Workshop,
St Andrews, UK, 26-9 October 2004, paper 2.2. National Engineering Laboratory, Glasgow, 2004 (on
CD-ROM)

[14] HUTTON, S.P. The prediction of Venturi meter coefficients and their variation with roughness and age.
Inst. Civil Eng. Proc. 1954, 3, pp. 216-241; 922-927

[15] SCHLICHTING, H. Boundary layer theory, 4th edition. McGraw-Hill, New York, NY, 1960, 647 p.

32
 PD ISO/TR 12767:2007

[16] READER-HARRIS, M.J. Pipe roughness and Reynolds number limits for the orifice plate discharge
coefficient equation. In: Proceedings of the 2nd International Symposium on Fluid Flow Measurement,
Calgary, AB, Canada, 6-9 June 1990, pp. 29-43. American Gas Association, Arlington, VA, 1990

[17] BEAN, H.S. and MURDOCK, J.W. Effects of pipe roughness on orifice meter accuracy, Report of
Supervising Committee on two-inch tests. American Gas Association, New York, NY, 1959 (American
Gas Association Research Project NW-20)

[18] BRENNAN, J.A., MCFADDIN, S.E., SINDT, C.F. and W ILSON, R.R. Effect of pipe roughness on orifice flow
measurement. National Institute of Standards and Technology, Boulder, CO, 1989 (NIST Technical
Note 1329)

[19] STUDZINSKI, W., BERG, D., BELL, D. and KARWACKI, L. Effect of meter run roughness on orifice meter
accuracy. In: Proceedings of the 2nd International Symposium on Fluid Flow Measurement, Calgary,
AB, Canada, 6-9 June 1990, pp. 1-15. American Gas Association, Arlington, VA, 1990

[20] CLARK, W.J. and STEPHENS, R.C. Flow measurement by square edged orifice plates: pipe roughness
effects. Proc. Inst. Mech. Eng. 1957, 171(33), pp. 895-904

[21] HERNING, F. and LUGT, H. Neue Versuche mit Segmentblenden und Normblenden [New research with
segmental diaphragms and standard orifices]. Brennst. — Wärme — Kraft, 1958, 10(5), pp. 219-223

[22] SPENCER, E.A., CALAME, H. and SINGER, J. Edge sharpness and pipe roughness effects on orifice plate
discharge coefficients. National Engineering Laboratory, Glasgow, 1969 (NEL Report No 427)

[23] THIBESSARD, G. Le coefficient de débit des diaphragmes, la rugosité et le nombre de Reynolds

[Diaphragm discharge coefficient, roughness and Reynolds number]. Chaleur Indust., 415, pp. 33-50,
1960

[24] W ITTE, R. Neue Beiträge zur internationalen Normung auf dem Gebiete der Durchflußmessung [New
contributions to international standardization in the flow measurement field]. Brennst. — Wärme —
Kraft, 1953, 5(6), pp. 185-190

[25] AKASHI, K., W ATANABE, H. and KOGA, K. Development of a new rectifier for shortening upstream pipe
length of flow meter. In: Yamasaki, H., editor. Proceedings of the Imeko Symposium on Flow
Measurement and Control in Industry, pp. 279-280, Tokyo, Japan, 13-6 November 1979. Society of
Instrument and Control Engineers, Tokyo, 1979

[26] BEAN, H.S. Indications of an orifice meter. Am. Gas Assoc. Month., Jul–Aug 1947, pp. 337-341, 349

[27] BEITLER, S.R. The flow of fluids through orifices in 6 inch pipelines. Trans. Am. Soc. Mech. Eng., 1929,
52, p. 751

[28] BLAKE, K.A. The design of piezometer rings. J. Fluid Mech., 1979, 78, pp. 415-428

[29] BRAIN, T.J.S. and REID, J. Measurement of orifice plate edge sharpness. Meas. Control, 1973, 6,
pp. 377-384

[30] CLARK, W.J. Flow measurement by square-edged orifice plate using corner tappings, Pergamon,
Oxford, 1965. 226 p.

[31] DALL, H.E. The effect of roughness of the orifice plate on the discharge coefficient. Instrum. Eng.,
Apr 1958, 2(5), pp. 91-92

[32] GALLAGHER, J.E. and LANASA, P.J. Field performance of the Gallagher flow conditioner. In:
Proceedings of the 3rd International Conference on Fluid Flow Measurement, San Antonio, TX, 1995,
section: Flow Conditioning II; paper: 2

33
PD ISO/TR 12767:2007

[33] GALLAGHER, J.E., LANASA, P.J. and BEATY, R.E. The Gallagher flow conditioner. In: North Sea Flow
Measurement Workshop: Papers and programme, Peebles, UK, October 1994, paper 2.4. National
Engineering Laboratory, Glasgow

[34] HERNING, F. and W OLOWSKI, E. Die Kantenunschärfe von Normblenden und Segmentblenden und das
Ähnlichkeitsgesetz [The edge sharpness of standard and segment orifices and the laws of similarity].
Brennst. — Wärme — Kraft, 1963, 15(1), pp. 26-30

[35] IRVING, S.J. Effect of system layout of the discharge coefficients of orifice plates. Part II [British
Hydromechanics Research Association (BHRA) Report RR 1424, 1977]; Part III (BHRA Report
RR 1462, 1978)

[36] JENNER, S.R. An investigation of the influence of upstream fittings on the accuracy of flow
measurement using orifice plates. Hatfield Polytechnic, B.Sc. (Eng.) Project Report, 1977

[37] JEPSON, P. and CHAMBERLAIN, D. Operating high pressure orifice metering installations. Flow-Con 77.
Proceedings of a Symposium on the application of flow measuring techniques, Brighton, UK, April
1977. Institute of Measurement and Control, Gatton and Kent Sections

[38] KRETZSCHMER, F. and W ALZHOLZ, G. [Experiments on installation faults of standard orifice plates.]
Forschung, 1934, 5(1), pp. 25-35

[39] LAKE, W.T. and REID, J. Optimal flow conditioner. In: Proceedings of the 10th North Sea Flow
Measurement Workshop, Peebles, UK, 1992, paper 1.3. National Engineering Laboratory, Glasgow

[40] LAWS, E.M. Flow conditioning — A new development. Flow Meas. Instrum., 1990, 1, pp. 167-170

[41] LAWS, E.M. and OUAZZANE, A.K. Flow conditioning for orifice plate flow meters. In: Proceedings of the
3rd International Conference on Fluid Flow Measurement, San Antonio, TX, 1995, section: Flow
Conditioning I; paper: 3

[42] MCVEIGH, J.C. Further investigations into the effect of roughness of the orifice plate on the discharge
coefficient. Instrum. Eng., 1962, 3(5), pp. 112-113

[43] MASON, D. and W ILSON, M.P.JR. Measurement error due to the bending of orifice plates, ASME paper
No. 75-WA/FM-6, 1975

[44] MILLER, R.W. and KNEISEL, O. Experimental study of the effects of orifice plate eccentricity on flow
coefficients. Trans Am. Soc. Mech Eng Ser. D: J. Basic Eng., 1969, 91(1), pp. 121-131

[45] NAGASHIO, K. and KOMIYA, K. Effect of upstream straight length on orifice flowmeters. Report of the
National Research Laboratory of Metrology, 21-1, Japan, 1972

[46] NAGEL, P. and JAUMOTTE, A. Étude expérimentale de l’influence de singularités sur le coefficient de
débit d’un diaphragme normalisé [Influence of disturbances on the coefficients of a standardized orifice
plate]. Promoclim A: Appl. Therm. Aéraul., 1976, 7(1), pp. 57-84

[47] ORSI, E. Influence of special parts on the operation of standardized diaphragms. L'Energica Elettrica
NI, Italy, 1978

[48] READER-HARRIS, M.J. The effect of pipe roughness on orifice plate discharge coefficients, National
Engineering Laboratory, Glasgow, 1990 (Progress Report No. 9.). Also available as Report EUR
13763, Commission of the European Communities, Brussels, 1991 (negative microfiche)

[49] READER-HARRIS, M.J. Computation of flow through orifice plates. In: Taylor, C., Gresho, P., Sani, R.l.,
Haüser, J., editors. Numerical methods in laminar and turbulent flow, Volume 6, Proceedings of the.
6th International Conference on Numerical Methods in Laminar and Turbulent Flow, Swansea, Part 2,
pp. 1907-1917, Pineridge, Swansea, 1989

34
 PD ISO/TR 12767:2007

[50] READER-HARRIS, M.J., SATTARY, J.A. and SPEARMAN, E.P. The orifice plate discharge coefficient
equation, National Engineering Laboratory Executive Agency, Glasgow, 1992 (Progress Report
No. 14)

[51] READER-HARRIS, M.J. SATTARY, J.A. and W OODHEAD, E. The use of flow conditioners to improve flow
measurement accuracy downstream of headers. In: Proceedings of the 3rd International Conference
on Fluid Flow Measurement, San Antonio, TX, 1995, section: Flow Conditioning II; paper: 3

[52] ROARK, R.J. and YOUNG, W.C. Formulas for stress and strain, 5th edition, McGraw Hill, New York, NY,
1975. 624 p.

[53] W EST, R.G. Developments in flow metering by means of orifice plates. In: Flow measurement in closed
conduits, Glasgow, UK, 27-30 September 1960, Paper B-3, HMSO, Edinburgh, 1962

[54] W ILCOX, P.L., W EBERG, T. and ERDAL, A. Short gas metering systems using K-Lab flow conditioners.
In: Proceedings of the 8th North Sea Flow Measurement Workshop, 1990

[55] READER-HARRIS, M.J. and SATTARY, J.A. The orifice plate discharge coefficient equation — The
equation for ISO 5167-1. In: Proceedings of the 14th North Sea Flow Measurement Workshop,
Peebles, UK, October 1996, paper 24. National Engineering Laboratory, Glasgow, 1998

[56] HERNING, F. Untersuchungen zum Problem der Kantenunschärfe bei Normblenden und bei
Segmentblenden [Experiments on the problem of the edge sharpness of standard and segmental
orifice plates]. Brennst. — Wärme — Kraft, 1962, 14(3), pp. 119-126

[57] CROCKET, K.A. and UPP, E.L. The measurement and effects of edge sharpness on the flow coefficients
of standard orifices. Trans. ASME J. Fluids Eng., Paper No 72-WA/FM-4, 1972 (June 1973,
pp. 271-275)

[58] BENEDICT, R.P., W YLER, J.S. and BRANDT, G.B. The effect of edge sharpness on the discharge
coefficient of an orifice. Trans. ASME J. Eng. Power, Paper No 74-WA/FM-4, 1974

35
PD ISO/TR
12767:2007
BSI — British Standards Institution
BSI is the independent national body responsible for preparing
British Standards. It presents the UK view on standards in Europe and at the
international level. It is incorporated by Royal Charter.

Revisions
British Standards are updated by amendment or revision. Users of
British Standards should make sure that they possess the latest amendments or
editions.
It is the constant aim of BSI to improve the quality of our products and services.
We would be grateful if anyone finding an inaccuracy or ambiguity while using
this British Standard would inform the Secretary of the technical committee
responsible, the identity of which can be found on the inside front cover.
Tel: +44 (0)20 8996 9000. Fax: +44 (0)20 8996 7400.
BSI offers members an individual updating service called PLUS which ensures
that subscribers automatically receive the latest editions of standards.

Buying standards
Orders for all BSI, international and foreign standards publications should be
addressed to Customer Services. Tel: +44 (0)20 8996 9001.
Fax: +44 (0)20 8996 7001. Email: [email protected]. Standards are also
available from the BSI website at http://www.bsi-global.com.
In response to orders for international standards, it is BSI policy to supply the
BSI implementation of those that have been published as British Standards,
unless otherwise requested.

Information on standards
BSI provides a wide range of information on national, European and
international standards through its Library and its Technical Help to Exporters
Service. Various BSI electronic information services are also available which give
details on all its products and services. Contact the Information Centre.
Tel: +44 (0)20 8996 7111. Fax: +44 (0)20 8996 7048. Email: [email protected].
Subscribing members of BSI are kept up to date with standards developments
and receive substantial discounts on the purchase price of standards. For details
of these and other benefits contact Membership Administration.
Tel: +44 (0)20 8996 7002. Fax: +44 (0)20 8996 7001.
Email: [email protected].
Information regarding online access to British Standards via British Standards
Online can be found at http://www.bsi-global.com/bsonline.
Further information about BSI is available on the BSI website at
http://www.bsi-global.com.

Copyright
Copyright subsists in all BSI publications. BSI also holds the copyright, in the
UK, of the publications of the international standardization bodies. Except as
permitted under the Copyright, Designs and Patents Act 1988 no extract may be
reproduced, stored in a retrieval system or transmitted in any form or by any
means – electronic, photocopying, recording or otherwise – without prior written
permission from BSI.
This does not preclude the free use, in the course of implementing the standard,
of necessary details such as symbols, and size, type or grade designations. If these
details are to be used for any other purpose than implementation then the prior
BSI written permission of BSI must be obtained.
389 Chiswick High Road Details and advice can be obtained from the Copyright & Licensing Manager.
London Tel: +44 (0)20 8996 7070. Fax: +44 (0)20 8996 7553.
Email: [email protected].
W4 4AL