Technical Guidance Note

(Monitoring)

M1

Sampling requirements for stack-emission monitoring

Environment Agency
October 2005
Version 3

Record of amendments
Version
number
2
3
3

Date

Amendment

July 2002
Oct 2005
Oct 2005

3
3
3

Oct 2005
Oct 2005
Oct 2005

Oct 2005

Added details on the use of temporary platforms to section 4.1.1.

Added details on conducting a swirl test (Section 2.1.3).
Added details on criteria for selection of gas sampling location (Section
2.2.3).
All references to BS 3405:1983 and BS 6069:Section 4.3:1992 deleted.
Amendments to positional requirements for CEMs (Section 3.2).
Section 4.1.1 amended to reflect requirements of HSE Work at Height
Regulations 2005.
Revised access port requirements (Figure 4.5).

Foreword
Purpose
This Technical Guidance Note (TGN) on sampling requirements for monitoring stackemissions to air from industrial applications is one of a series providing guidance to
Environment Agency staff, monitoring contractors, industry and other parties
interested in stack-emission monitoring. It is also a key reference document in
support of our Monitoring Certification Scheme (MCERTS) and Operator Monitoring
Assessment (OMA).
It provides guidance on:

the selection of the sampling position, sampling plane and sampling points;
access, facilities and services required;
safety considerations.

Throughout the document, references to sampling from stacks should be interpreted as

also meaning sampling from vents, ducts and flues, unless otherwise stated.
The section on safety draws attention to the importance of health and safety in stackemission monitoring, provides guidance on the hazards and risks associated with
stack-emission monitoring and highlights the relevant health and safety legislation.
How to use M1
The TGN is divided into four chapters. Chapter 1 describes the main monitoring
approaches, the fundamental principles of sampling and the importance of a
monitoring strategy. Chapters 2 and 3 give guidance on the sampling requirements
for periodic stack-emissions measurements and continuous emissions monitoring
systems, respectively. Chapter 4 provides guidance on sampling facilities and safety
requirements for stack-emission monitoring.

Contents
1. Monitoring approaches
1.1
1.2
1.3

1.4

Monitoring strategy
The different approaches to monitoring stack-emissions
The importance of representative sampling
1.3.1
Fundamental principle
1.3.2
Representative sampling of particulates and gases
Choice of sampling method, sampling technique and equipment

2. Sampling requirements: periodic measurements

2.1

2.2

Periodic sampling of particulates

2.1.1
Special characteristics
2.1.2
Positional requirements for sampling particulates
2.1.3
Criteria for locating the sampling plane
2.1.4
Surveying the proposed sampling plane
2.1.5
Preliminary velocity survey
2.1.6
Unacceptable characteristics
2.1.7
Number of sampling points
2.1.8
Position of sampling points along the sample lines
2.1.8.1 Position of sampling points for circular stacks
2.1.8.2 Position of sampling points for rectangular and square stacks
Periodic sampling of gases
2.2.1
Special characteristics of sampling gases
2.2.2
Location requirements for monitoring gases
2.2.3
Criteria for locating a sampling plane for gas concentrations
2.2.3.1 Procedure for conducting a stratification test
2.2.4
Number and position of sampling points for sampling gases

3. Sampling requirements: continuous emissions monitoring systems

3.1
3.2

CEMs measuring concentrations

CEMs measuring gas volumetric flow rates

4. Sampling facilities and safety requirements for stack-emission monitoring

4.1

4.2
4.3
4.4

Access and facilities for periodic monitoring

4.1.1
Safe working platform and means of access to the sampling position
4.1.2
Space for the equipment and personnel
4.1.3
Means of entry into the stack (sampling access ports)
4.1.4
Essential services
4.1.5
Other issues
Access and facilities for periodic monitoring of gases
Access and facilities for CEMs
The risk-management approach to site work
4.4.1
Safety has the highest priority
4.4.2
Prominent hazards encountered when stack-emission monitoring
4.4.3
The risk assessment process
4.4.4
Responsibilities
4.4.5
When to carry out a risk assessment
4.4.6
Getting the operator involved
4.4.7
Reporting of accidents and near misses
4.4.8
Further guidance

1
2
2
2
2
4
4
4
5
6
8
8
8
9
9
10
11
13
13
14
15
16
16
17
17
18
18
19
19
26
26
26
26
27
27
28
28
29
30
30
31
31

5. Status of this document

32

Annex 1 Sample point requirements as specified in ISO 9096:2003

33

References

34

Sampling requirements for stack-emission monitoring

1. Monitoring guidance
The Environment Agency publishes Technical Guidance Notes (TGNs) to provide
advice and support in relevant subject areas. This TGN covers sampling and safety
requirements for stack-emission monitoring. It is one of a series of TGNs in the M
series providing guidance on regulatory monitoring.
The Environment Agencys MCERTS scheme (see Box 1.1) includes both continuous
emission monitoring systems (CEMs) and manual stack-emission monitoring.
MCERTS for manual stack-emission monitoring provides for the assessment of the
management, quality and reliability of organisations, and the competence of
individuals carrying out stack-emission monitoring against published performance
standards for organisations1 and personnel2. This TGN is a key reference document
that underpins MCERTS for manual stack-emission monitoring and provides
guidance on suitable sampling locations for CEMs.

Box 1.1 MCERTS

MCERTS is the Environment Agency'
s Monitoring Certification Scheme for instruments,
monitoring and analytical services. The scheme is built on proven international standards
and provides industry with a framework for choosing monitoring systems and services that
meet the Environment Agencys performance specifications. MCERTS reflects the
growing requirements for regulatory monitoring to meet European and international
standards. It brings together relevant standards into a scheme that can be easily accessed by
manufacturers, operators, regulators and laboratories. Further information on MCERTS is
available at www.mcerts.net

1.1 Monitoring strategy

Emission monitoring carried out at a stack must be fit-for-purpose. To meet this
criterion, considerable thought must go into developing an appropriate monitoring
strategy for the particular application.
This monitoring strategy should be documented in a site-specific protocol (SSP),
which is generated by the monitoring organisation before monitoring work begins.
The SSP describes how the method will be employed in a given, specific situation and
should be undertaken for each site where monitoring is to take place. The MCERTS
performance standard for organisations1 sets out the requirements for the SSP.

Page 1 of 34

1.2 The different approaches to monitoring stack-emissions

Stack-emissions monitoring can be classified into two types:
a)

Periodic measurements a measurement campaign is carried out at

periodic intervals, for example, once every three months. The sample
is usually, but not always, withdrawn from the stack (extractive
sampling). An instrumental or automated technique may be used,
where the sampling and analysis of the substance is fed to an on-line
analyser. Alternatively, a technique may be used where a sample is
extracted on site and analysed later in a laboratory. Samples may be
obtained over several hours, or may be so-called spot or grab
samples collected over a period of seconds to several minutes.

b)

Continuous emissions monitoring systems (CEMs) automated

measurements carried out continuously, with few if any gaps in the
data produced. Measurement may be carried out in situ in the stack (for
example, cross-duct monitoring), or extractive sampling may be used
with an instrument permanently located at or near the stack. CEMs are
also referred to as Automated Monitoring Systems (AMS), particularly
in a European context.

This TGN uses the convenient division into periodic measurements and CEMs when
describing the requirements for sampling positions, access, facilities and services in
subsequent chapters.

1.3 The importance of representative sampling

1.3.1 Fundamental principle
The fundamental principle behind any sampling activity is that a small amount of
collected material should be representative of all the material being monitored. The
number and location of samples that are needed to make up a representative sample
depends on how homogeneous the material is. If it is very homogeneous, only a few
samples may be required. If the material is heterogeneous, many more samples will
be required.
1.3.2 Representative sampling of particulates and gases
This fundamental principle applies as much to sampling stack gases as it does to any
other type of sampling. Although the gas in a stack might be thought of as being
more uniform than, for example, a stockpile of coal, gases in stacks can become nonhomogeneous. This may be due to differences in chemical composition, or
differences in temperature and velocity, which may lead to stratification and swirling.
Where the gas is also carrying particulates along the duct, there is likely to be even
less homogeneity. Here, special measures must be taken to ensure samples are
representative.

Page 2 of 34

For gases carrying particulates, the sampling approach has to address two effects.
Firstly, inertial effects introduced by gravity and the duct geometry lead to the
particles being unevenly distributed in the duct. Samples must be obtained from
multiple sample points (see to Box 1.2 for definitions of terms) across the sampling
plane to give an overall average of the particulate emission. Rules have been
developed specifying exactly where these sampling points should be located (these
are described in Section 2.1.8). In the case of a cross-duct CEM monitoring
particulates, the average particulate concentration is obtained as an integrated
measurement across the duct. Secondly, for extractive methods, the sample must be
collected isokinetically (see Box 1.3).
Box 1.2 Definition of important terms in stack emission monitoring
Isokinetic sampling is achieved when the gas enters the sampling nozzle at exactly the
same velocity and direction as the gas travelling in the stack..
Sampling location a suitable position on stack, duct or flue where representative
samples can be obtained.
Sampling plane or sampling section the plane normal to the centreline of the stack, duct
or flue at the sampling position.
Sampling point the specific position on the sample plane from where the sample is
extracted.
Sampling ports or access ports points in the wall of the stack, duct or flue through
which access to the emission gas can be gained.
Stack, duct, flue, vent a structure through which pollutants are released to atmosphere.
Typically stacks are intended to be of sufficient height to adequately disperse emissions
in the atmosphere. Ducts and flues are usually a lower height and often located before
entering a stack. Vents are not usually free-standing structures.
Sampling lines imaginary lines in the sampling plane along which sampling points are
located, bounded by the inner wall of the stack, duct or flue.

Box 1.3 The importance of isokinetic sampling for particulates

Due to the wide range of particle sizes normally present in process emission streams, it is
necessary to sample isokinetically to ensure that a representative sample of the particulate
emission is obtained.
If the sampling velocity is less than the isokinetic rate, at first sight it would appear, that
the emission will be underestimated. However, because the sampling rate is too low,
there is a divergence in flow around the sampling inlet. Small particles are able to follow
the flow and a percentage of them will not be sampled. Larger particles, on the other
hand, are not able to follow the flow because of their greater inertia and more of these
particles will enter the sampler. Thus a sub-isokinetic sampling rate will lead to a bias in
the sampled particle size distribution towards the larger particles. This could lead to an
overestimate of the particulate concentration depending on the original size distribution.
Sampling at a rate in excess of the isokinetic rate will lead to a bias in the sampled
particle size distribution towards the smaller particles. This could lead to an
underestimate of the emission concentration depending on the original size distribution.

Page 3 of 34

Where the measurement concerned is of concentrations of gaseous species alone, the

approach is to choose a sampling location where the gases are well mixed. Sampling
can often be from a single sampling point in the sampling plane. However, if the mass
emission rate is to be calculated, the gas volumetric flowrate will need to be
measured; this will require temperature and velocity measurements to be made at
several points across the sampling plane.
Some pollutants; for example, metals and dioxins, are present in both particulate and
vapour phases. Other pollutants for example, hydrogen chloride, may be present in an
aerosol phase and vapour phase. Aerosols are normally treated as particulates. In all
such cases isokinetic multi-point sampling is required, as described in Section 2.1.

1.4 Choice of sampling method, sampling technique and equipment

There is a wide choice of monitoring approaches, analytical techniques, published
methods and equipment that can be used to carry out stack-emissions measurements.
It is important that each of these is chosen to be suitable for the application in
question. TGN M2 Monitoring of Stack Emissions to Air3 gives detailed guidance on
deciding which approach, technique and method should be chosen.
The sampling approach, technique, method and equipment that are chosen can have
different effects on the requirements for access, facilities and services. Though the
precise requirements can vary, the following will always be required:

a safe means of access to the sampling position;

a means of entry for sampling equipment into the stack;
adequate space for the equipment and personnel;
provision of essential services, such as electricity.

Further details on such requirements are given in Chapter 4.

2. Sampling requirements: periodic measurements

2.1 Periodic sampling of particulates
2.1.1 Special characteristics
Periodic sampling of particulates requires extractive isokinetic sampling methods.
The principle of isokinetic sampling is that a sharp-edged nozzle is positioned in the
stack facing into the moving gas stream and a sample of the gas is extracted through
it, at the same velocity as the gas in the stack, for a measured period of time. To
allow for non-uniformity of particulate distribution, samples are taken at a preselected number of stated points across the sample plane. The particulates collected
in the sampler are later weighed, and the concentration of particulate matter in the
stack is calculated using the volume of gas sampled. The mass flow rate of particulate
matter in the stack can be calculated from the concentration and the velocity of gas in
the stack.

Page 4 of 34

Some of the older measurement standards are prescriptive (either explicitly or

implicitly) about the equipment that can be used for sampling particulates. The trend
now, however, is towards standards that lay down minimum performance and design
criteria for the equipment. Any piece of sampling equipment can be used if it meets
these criteria and is fit for purpose. This last caveat is important as each system may
have particular advantages and drawbacks. These need to be considered when
choosing the most suitable sampling equipment for the measurement in question.
The main differences between particulate sampling equipment, in so far as they affect
the access, facilities and services required, is the configuration of the sample probe
and particle-separator device. Sampled particulate is collected on a filter held in a
filter holder; the latter may be arranged on the probe so that it is in the stack (in-stack
filtration) or out of the stack (out-of-stack filtration).
2.1.2 Positional requirements for sampling particulates
There are a number of standard methods for monitoring particulates. The Comit
Europen de Normalisation (CEN) standard BS EN 13284-1:20024 details the
positional requirements for particulates and is referred to on positional requirements
by a number of other CEN standards. It should be noted that BS EN 13284-1:20024
along with another particulate standard BS ISO 90965 covers the measurement of
particulates in the range from 0 to 1000 mg m-3.
This chapter provides a single set of requirements for sampling positions for
isokinetic sampling of particulates and multi-phase pollutants. These requirements are
taken from BS EN 13284-1:2002.

Page 5 of 34

2.1.3 Criteria for locating the sampling plane

The general approach for periodic sampling of particulates and aerosols can be
summarised as:
Identify a potentially suitable sampling
location (covered in this section)

Access the stack and check that the gas-flow

criteria are met by carrying out an exploratory
survey (see Sections 2.1.4 and 2.1.5)

Decide the number and position of the

sampling points (see Sections 2.1.7 and 2.1.8),
and install ports and access to the sampling
location (see Section 4.1).

Carry out preliminary velocity traverse

(Section 2.1.5) to confirm satisfactory and
proceed with sampling.

Sampling must be carried out at a suitable location on the stack. Bends, branches,
obstructions, fans and leaks can all cause undesirable variations in the temperature
and velocity profiles, which may make the location unsuitable for sampling.
For new installations, the sampling plane should be located according to the
requirements contained in Table 2.1. By adhering to these requirements, there is a
very good chance that the flow-stability criteria defined in Table 2.2 will be met. The
flow-stability criteria of a gas encompass velocity and direction. Flow criteria is used
in this TGN as a term to cover all these parameters. It should be emphasised that it is
the flow criteria that are important, not the adherence to the positional requirements
per se. Therefore on plants where the sampling plane location does not meet the
recommendations in Table 2.1, it is possible to locate the sampling plane at another
location as long as the flow criteria are met.

Page 6 of 34

Table 2.1 Recommended location for meeting sample location requirements for
periodic sampling of particulates
General location of
sampling plane

Location of sampling
plane in straight
section

The sampling plane must be situated in a length of straight duct (preferably

vertical) with constant shape and constant cross-sectional area. Where possible,
the sampling plane should be as far downstream and upstream from any
disturbance, which could produce a change in direction of flow (e.g. bend, fan or
a partially closed damper).
The sample plane criteria are usually met in sections of duct with five hydraulic
diameters (see Box 2.1 for definition) of straight duct upstream of the sampling
plane and two hydraulic diameters downstream. If the sampling plane is to be
located near the top of the stack outlet then the distance from the top should be
five hydraulic diameters (making a straight length of 10 hydraulic diameters).

Box 2.1 Definition of hydraulic diameter

It is not possible simply to use the term diameter in positional criteria, because not all stacks are of
circular cross-section. To allow all shape ducts to be included, the term hydraulic diameter is used,
defined as:
(4 x area of sampling plane) / length of sample plane perimeter

Table 2.2 Flow-stability criteria for periodic sampling of particulates

Angle of gas flow

15 from stack longitudinal axis

Flow direction

No local negative flow

Minimum velocity

Dependant on the method used (for Pitot tubes a differential

pressure larger than 5 Pa)
Ratio of highest to lowest local gas velocities less than 3:1

Gas velocities variations

Note: The reader is advised to consult the normative requirements contained within
Annex B of BS EN 13284-1 :20024 for guidance on how to carry out a swirl test with
either an L Type or and S Type Pitot tube.
Compliance with the flow criteria should be checked by a preliminary velocity survey
(Section 2.1.5). If the flow criteria in Table 2.2 are not met, the sampling location is
not in compliance and a more suitable sampling location should be selected.
However, problems with meeting the criteria sometimes occur with older installations
where it is not always practicable to move the sampling plane to a better location
because of safety considerations, cost, or the design of the stack. When this occurs it
may be possible to improve representative sampling by increasing the number of
sampling points above those specified in Tables 2.3 and 2.4.

Page 7 of 34

2.1.4 Surveying the proposed sampling plane

An exploratory survey of the proposed sample plane should be undertaken. This
should include a stack-gas velocity survey, as described in Section 2.1.5. For metal or
other thin-walled ducts or chimneys, 13mm (half inch) diameter pilot holes, drilled
through the flue wall on the centres of the proposed sample access holes, may enable
a small bore Pitot-static tube and thermocouple probe of suitable length to be used for
this purpose. The use of such temporary access holes distinguishes the exploratory
velocity traverse from the preliminary velocity survey. The latter must be performed
through permanent access ports meeting the requirements in Section 2.1.5.
It is important that this approach is followed, so that the sample plane location is
shown to be suitable before installing comprehensive and expensive sampling access
and facilities on a stack.
2.1.5 Preliminary velocity survey
A preliminary velocity traverse should be carried out before sampling is carried out at
a location for the first time. To establish that a sampling location is suitable,
measurements of gas velocity should be carried out at a minimum of ten equally
spaced points along each proposed sampling line. Sampling points should be located
either more than 3% of the sampling line length or more than 5cm whichever is the
greater value from the duct wall. Measurements at these sampling points must
demonstrate that the gas stream at the sampling plane meets the requirements
contained in Table 2.2. If the results do not conform to the flow-stability criteria,
another sample location should be found.
2.1.6 Unacceptable characteristics
Some older installations may not meet the flow-stability criteria, but sampling may
nevertheless still be required. Deviations from the standard can have significant
effects on measurement uncertainty.
Information on assessing the effects of deviations on accuracy is given in the Source
Testing Association (STA) Quality Guidance Note QGN001 r16

Page 8 of 34

2.1.7 Number of sampling points

The number of sampling points which is to be used is determined by the size of the
stack. The minimum number of sampling points and sample lines are shown in Tables
2.3 and 2.4, for circular and rectangular stacks respectively. The sample point
requirements contained in ISO 9096:20035 are detailed in Annex 1.
Table 2.3 Minimum number of sampling points for circular stacks
Range of sampling
plane areas
m2

Range of duct
diameters
m

Minimum number of
sampling lines
(diameters)

Minimum number of
sampling points per
plane

<0.1

<0.35

1a

0.1 to 1.0

0.35 to 1.1

1.1 to 2.0

1.1 to 1.6

>2.0

>1.6

At least 12 and 4 per m2 b

using only one sample point may give rise to errors that give an uncertainty above 10%.

for large ducts, 20 sampling points is generally sufficient

Table 2.4 Minimum number of sampling points for rectangular stacks

Range of sampling plane areas
m2
<0.1

Minimum number of side

divisions a
-

Minimum number of sampling

points per plane
1b

0.1 to 1.0

1.1 to 2.0

>2.0

At least 12 and 4 per m2 c

Note a: Other side divisions may be necessary, for example if the longest stack side length is more than twice the
length of the shortest side (see Figure 2.4). Sample lines for square and rectangular stacks are explained in Section
2.1.8.2.
Note b: Using only one sampling point may give rise to errors that give an uncertainty above 10%
Note c: For large ducts, 20 sampling points is generally sufficient.

2.1.8 Position of sampling points along the sample lines

The sampling plane is divided into equal areas and sampling is carried out from points
in the centres of these areas. This is shown in Figure 2.1 for circular stacks and in
Figures 2.2, 2.3 and 2.4 for square and rectangular ducts.
Sampling points should be located either more than 3% of the sampling line length or
more than 5cm whichever is the greater value from the duct wall.

Page 9 of 34

2.1.8.1 Position of sampling points for circular stacks

The sampling plane is divided into equal areas, the central one being circular (see
Figure 2.1). The sampling points are located along two or more sampling lines
(diameters) at the centres of these equal areas.
Figure 2.1 Sampling in circular stacks (positions shown are for stacks >2 m in
diameter)

1
2

Xi

3
4

The precise positions of the sampling points will depend on:

the number of sampling lines chosen;

the number of sampling points that need to be used (see Table 2.3).

A commonly encountered situation is a circular stack with two sample lines. In this
case, a simple expression can be used to calculate the distance of each sampling point
from the duct wall.
xi = Kid
where: xi is the distance (m) from the stack wall to an individual sampling point;
Ki is a calibration factor (a percentage of the diameter);
d is the duct diameter (m) at the sampling plane;
i is the sample point identification position on the sample line.
Values for the factor K are given in Table 2.5 for different numbers of sampling
points (nd) required along each of two sampling lines.

Page 10 of 34

Table 2.5 Position of sampling points as a percentage distance across a circular

stack
Values of K (as a % of diameter) for different numbers of points (nd)
Sample point
position (i) on the (central point included)
sampling line
nd = 5
nd = 7
nd = 9
nd = 3
1
2
3
4
5
6
7
8
9

11.3
50.0
88.7
-

5.9
21.1
50.0
78.9
94.1
-

4.0
13.3
26.0
50.0
74.0
86.7
96.0
-

3.0
9.8
17.8
29.0
50.0
71.0
82.2
90.2
97.0

Table 2.5 does not apply to circular stacks where more than nine sampling points are
needed or more than two sample lines are used. In such cases, the following general
formula should be used to calculate the distance of each sampling point in from the
duct wall.
xi = 0.5d [1-{((2nr-2.i+1)nd+1)/(2nr.ndia+1)}]
where: i is sample point identification number* along the sampling line;
d is the inside diameter of the stack, i.e. the length of the sampling line (m);
nr is the number of sampling points on a sampling radius (i.e. 0.5d); and
ndia is the number of sampling diameters (sampling lines).
There is also an alternative method, known as the tangential method, for determining
the position of sampling points in circular stacks. This rule allows the approach
common in the United States (that is, no sampling point at the centre of the stack).
Details of the tangential method can be found in BS EN 13284-1:2002.
2.1.8.2 Position of sampling points for rectangular and square stacks
The sampling plane should be divided up into the required number of equal areas by
lines parallel to the stack walls (see Table 2.4). The distance between each of these
parallel lines is known as a side division. Sampling points are located at the centres
of the equal areas, also known as partial areas. Figure 2.2 shows the relationship
between the sample points in the centres of equal areas, the side divisions and the
sample lines. Refer to Section 2.1 for details on where to install access ports to reach
these sampling points.

The first point in from the access port is identified as position 1.

Page 11 of 34

Figure 2.2 Equal areas, side divisions, sampling lines for square/rectangular
stacks
3 side divisions along
shortest side

3 side
divisions
along
longest
side

9 equal areas

9 sampling
points

Cross-section
of stack

3 sampling
lines

The most common case, shown in Figure 2.3, is for square and near-square stacks
(where the ratio of the longer sides l1 to the shorter sides l2 of the stack is no more
than 2:1). The longer and shorter sides of the stack are both divided into an equal
number of parts to meet the requirements of Table 2.4 for minimum numbers of
sample lines and sampling points. The number of partial areas, each having a
sampling point at its centre, is thus the square of the number of side divisions. These
areas have the same shape as the stack.
Figure 2.3 Sampling positions in square/near-square stacks where sides l1 to l2 is
2:1
l2

l1
2n1

l1
n1

l1
Cross-section
of stack

l2
2n2

l2
n2

Page 12 of 34

For stacks with a more extreme rectangular aspect ratio (ratio of longer sides l1 to the
shorter sides l2 is more than 2:1) a different approach is used, to avoid having long,
thin partial areas. The procedure is to give stack side l1 more side divisions than l2, so
that the partial areas have a length to breadth ratio of no more than 2:1. This is
illustrated in Figure 2.4.
Figure 2.4 Sampling positions in rectangular stack where sides l1 to l2 is >2:1
l2

l1
2n1

l1
n1

Cross-section
of stack

l1

l2
2n2

l2
n2

The requirements of Table 2.4 for minimum numbers of sample lines and sampling
points must be met. The number of sampling points is calculated as follows: if the
stack sampling plane sides l1 and l2 are divided into n1 and n2 parts respectively, then a
total of n1.n2 sampling points is obtained.
The distance between the stack wall and the first sample point will be l1/2n1 if
sampling was carried out from ports on side l2, or l2/2n2 if sampling was carried out
from ports on side l1.

2.2 Periodic sampling of gases

2.2.1 Special characteristics of sampling gases
For monitoring gases, the range of sampling equipment and apparatus is very wide.
However, they can be grouped conveniently into automated techniques and manual
techniques (see Box 2.2). The main steps in measuring gaseous pollutants are as
follows:

Page 13 of 34

Stage 1

Representative sample
of source gas extracted
through probe, filtered
and conditioned

Automated/ instrumental techniques

Stage 3

The sampled substance

is analysed by an
appropriate analytical
technique

Stage 2

Gaseous substances
adsorbed on, or
absorbed in, an
appropriate collection
medium
Manual techniques

When gases are measured using automated/instrumental techniques, such as gas

analysers, Stage 2 is omitted, and the sample goes directly to the analysis stage (Stage
3). In contrast, when gases are measured using manual techniques, Stage 3 is usually
carried out away from the site at an analytical laboratory.

Box 2.2 Automated and manual techniques

For automated techniques, the sampling and quantification stages are conventionally considered
to take place almost simultaneously, within an analyser. With manual techniques, the sample is
taken, then the quantification and analysis takes place as a discrete, later stage.

The particular sampling-train configuration will vary according to the substance and
the process conditions, for example whether the gas stream is hot or cool, wet or dry,
and whether sampling is to be performed in an intrinsically safe area. For instance, a
cool gas stream will require Stage 1 of the sample train to have a heated sample line;
very wet gas streams may require drop-out pots. The sample train configuration may
affect the requirements for access and facilities. For example, Stages 1 and 2 may
take place immediately after the sample probe, with the apparatus located on the stack
access platform. Alternatively, an inert (and usually heated) sample line may be used
to bring the sample gas from the probe to Stages 2 and 3 at the base of the stack or in
a mobile laboratory. It is also useful to consider the phase of the target species. For
example, dioxins may adsorb onto the surface of particulate species, if present.
Alternatively, if water droplets are present which may absorb hydrogen chloride,
isokinetic sampling may be required.
2.2.2 Location requirements for monitoring gases
In principle, the location requirements for measuring gas concentrations are less
exacting than for particulates. The criteria for measuring gas concentrations alone are
described in Section 2.2.3.

Page 14 of 34

However, in regulatory compliance monitoring it is common for mass emissions rates

(for example, g s-1) to be reported. Calculation of mass emissions rate requires the
measurement of gas volumetric flow-rate through the duct. This requires temperature
and velocity measurements to be taken at different points across the sampling plane.
Measurements to determine stack-gas velocity and volumetric flowrate should be
made in accordance with ISO 10780:19947. A suitable sampling location should
therefore conform to the flow-stability requirements described in Table 2.2.
Even in cases where no estimate of mass emissions rate is required, the process
operator should think carefully about upgrading the sampling plane in case mass
emissions are required in the future. For this reason, it is recommended that wherever
possible the sampling position is chosen to comply with the same criteria as when
sampling for particulates.
2.2.3 Criteria for locating a sampling plane for gas concentrations
Sampling should be carried out from a sampling position where the gases are
homogeneous and there is positive gas flow. This can be checked by a survey
following the basic principles in Section 2.1.4, but using a direct-reading instrument
to traverse the proposed sampling lines. The stack gas at the proposed sampling
position is classed as homogeneous if differences in concentration across the duct do
not exceed 10% as specified in ISO/DIS 103968 (see below). The proximity to bends,
branches, obstructions and fans is less important when measuring gas concentrations
only, as variations in the temperature and velocity profiles tend not to influence the
result. However, sampling in the vicinity of air in-leakage must be avoided. Single
point sampling for gaseous species should only be undertaken after an acceptable
degree of homogeneity has been demonstrated. Special care should be exercised if
gaseous sampling is being undertaken using access ports located in the vicinity of
those selected for particulate measurement. This because the in-duct conditions which
are desirable when sampling particulates, may also contribute to non-uniform
distribution of gaseous species.
Note: ISO 10396:19937 Stationary Source Emissions Sampling for the Automated
Determination of Gas Concentrations is currently being revised. The draft revised
standard ISO/DIS 10396 contains more-detailed requirements for sampling plane
selection for gas concentrations than the 1993 version. The account given below is
drawn largely from the revised, draft standard, ISO/DIS 10396.
Before any sampling is undertaken, it is necessary to determine any spatial or
temporal fluctuations in the gas concentrations, and to carry out a preliminary survey
of the gas concentration, temperature and velocity. Measure the concentration,
temperature and velocity at the sampling points several times to obtain their spatial
and temporal profiles. Conduct this survey when the plant is operating under
conditions that will be adhered to during the test, in order to determine whether the
selected sampling position is suitable and whether the conditions in the duct are
satisfactory. This survey could be omitted if the spatial or temporal fluctuations in the
duct can be determined from the owners investigation, a previous investigation or the
process characteristics in advance of the survey. In this case the information relating
to previous procedures for the determination of the sampling point and the adoption of
single point sampling shall be described in the report. It is necessary to ensure that the
Page 15 of 34

gas concentrations measured are representative of the average conditions inside the
duct or stack. The requirements for the extractive sampling of gas may not be as
stringent as those for particulate material. It is important that the sampling point is not
located near any obstructions that could seriously disturb the gas flow in the duct or
stack. The pollutant may have cross-sectional variation. The concentration at various
points of the cross-section shall first be checked, in order to determine if the presence
of gas stratification or air infiltration indicates that the gas to be measured is not
homogenous. If an alternative, acceptable, location is not available, multipoint
sampling is then required.
2.2.3.1 Procedure for conducting a stratification test:
With the process operating under steady-state conditions and at normal load, use a
traversing gas sampling probe to measure the pollutant and diluent (CO2 or O2)
concentrations at a minimum of twelve points located at sampling locations as
specified in ISO 9096. Use automated analytical methods for the measurement of the
gas concentrations. Measure for a minimum of two minutes at each traverse point.
While traversing, measure the pollutant and diluent gas concentrations from the centre
of the stack to determine if temporal, rather than spatial variations in the flue gas
concentrations are occurring. Calculate the average pollutant and diluent
concentration at each of the individual traverse points. Then, calculate the arithmetic
average concentrations for the gas from all of the traverse points. The pollutant or
diluent gas is said to be non-stratified (ie homogenous) if the concentration at each
individual traverse point differs by no more than 10% from the arithmetic average
concentration for all of the traverse points. Usually, the cross-sectional concentration
of gaseous pollutants is uniform, due to the effects diffusion and turbulent mixing. In
this case, it is only necessary to sample at one point within the stack or duct to
determine the average concentration. A gas sample should be extracted near the centre
of the sampling site, positioned one-third to halfway in the stack or duct. When using
non-extractive systems, a representative location should be similarly selected.
2.2.4 Number and position of sampling points for sampling gases
Providing the gases are homogeneous, sampling of the gaseous pollutant can be from
a single sampling point on the sampling line. For sampling pollutants in the gaseous
phase, ISO 10396:1993 recommends positioning the tip of the sampling probe
between one-third and half-way into the stack to give a representative sample,
provided:

an initial concentration traverse shows no significant differences in concentration

(as defined in Section 2.2.3); and
for cases where mass emissions are to be calculated, the preliminary temperature
and velocity traverse results meet the flow criteria in Table 2.2.

Some pollutants, such as heavy metals, dioxins and PAHs, are present in both the
particulate phase and the vapour phase, while hydrogen chloride can exist in both as
an aerosol and vapour phase. In most cases, the authorisation or permit will require
the total emission to be reported, in which case both phases should be sampled
simultaneously and isokinetically at the sampling points described in Section 2.1.7.

Page 16 of 34

3 Sampling requirements: continuous emissions monitoring systems

3.1 CEMs measuring concentration
For CEMs measuring particulates, the location of the CEM is determined by the
requirements of BS EN13284-1:20024 (see Section 2.1.3, Table 2.2).
For CEMs that monitor gases, the sample plane location must comply with the
requirements given in Section 2.2.3.
In many applications the calibration of the CEM is established by comparison with
simultaneous measurements obtained using a standard periodic manual sampling
method or an instrumental monitoring method. Wherever possible, such monitoring
should be undertaken in close proximity to the CEM and must comply with the
sampling requirements in Section 2.1 or 2.2 as appropriate. Further information is
contained in BS EN 14181:20049.

3.2 CEMs measuring gas volumetric flow rates

The positional requirements for CEMs measuring gas volumetric flow rate are
specified in BS ISO 1416410, and these are summarised in Table 3.1. The flowstability criteria are given in Table 3.2.
Table 3.1 Positional requirements for continuous gas-flow monitoring
General location of sampling plane
Length of straight section
Location of sampling plane in straight
section and in relation to bends fans
and the like

Representative of total volume-flow in the duct

At least ten hydraulic diameters long.
Note: this is more stringent than BS EN 13284-1:2002.
At least five hydraulic diameters upstream and at least five
hydraulic diameters downstream from any flow disturbance.

Table 3.2 Flow-stability criteria for continuous gas-flow monitoring

Angle of gas flow

10 from duct axis

Flow direction

No local negative gas flow

Minimum velocity

As specified by CEMs manufacturer

Gas velocity variations

Must be 10% at any sample point

Page 17 of 34

4. Sampling facilities and safety requirements for stack-emission

monitoring
4.1. Access and facilities for periodic monitoring
4.1.1 Safe working platform and means of access to the sampling position
Industrial processes are regulated by the Agency through authorisations or permits
issued under IPC or Pollution and Prevention Control (PPC) Regulations. These place
requirements on an operator to provide a safe and normally permanent means of
access to enable monitoring to be carried out at specified release points. However, in
exceptional circumstances, for example, at an old installation where limited access
prevents the installation of a permanent platform, temporary structures (such as,
scaffolding) may be used, provided it is fully justified by the operator.
All platforms, whether permanent or temporary, should meet the minimum weight
criteria required for sampling. This is defined as 400kg point load in BS EN 132841:2002 and for permanent platforms is achievable. However, temporary, scaffold
platforms cannot be constructed to this specification and must, instead, be constructed
to a specific minimum scafftag category of heavy duty or meet the requirements
stated in the monitoring standard, whichever is the greater. Temporary platforms must
also be tied to, or supported by, a permanent structure. Sampling from mobile access
platforms, cherry pickers or ladders is unacceptable. Sampling from roofs or
the tops of arrestment equipment, vessels and ducts is unacceptable unless they have
been assessed as being suitable by meeting the requirements for platforms described
in this Note and the Work at Height Regulations 200511.
The platform and access must meet all current legislative requirements regarding
dimensions and construction, be maintained to a safe standard and undergo regular
inspection by a competent person. For scaffolding, a properly completed and dated
Scafftag is one means of demonstrating and recording this inspection. For
permanently installed platforms, the structural integrity of the platform and any
supports and attachments shall be assessed periodically by the process operator. The
inspection shall also include the effects of weathering, corrosion and damage. The
frequency and comprehensiveness of inspection shall be commensurate with the risk
of failure and serious injury. For example, a steel platform at great height and in a
corrosive atmosphere may require more frequent and thorough inspections than a low
platform in a normal atmosphere. Records of inspections should be available for the
monitoring team to check before they ascend to the work area.
The Work at Height Regulations place specific stipulations that require the employer
to carry out inspections of work equipment and pre-use checks of places of work.

A permanent platform should be used where possible. If a temporary platform is required then the
process operator and sampling team must liase to ensure that the scaffold erected is of a sufficient
standard to support the minimum weight requirement. The sampling team leader should confirm this
before undertaking the work. Special or masonry independent tied scaffold would normally be
sufficient.

Page 18 of 34

Working platforms and associated guard rails, barriers, toe-boards, ladders etc, must
be inspected at the following frequencies:

after assembly/installation but before use;

at suitable intervals where there has been exposure to conditions causing
deterioration; and
at each time an exceptional circumstance has occurred which is liable to
jeopardise its safety and within seven days prior to use if the platform is one from
which a person could fall more than 2m.

The platform shall be provided with handrails and kick-boards that meet the
requirements of the Workplace (Health, Safety and Welfare) Regulations 1992,
(Regulation 13) for permanent platforms and the Health and Safety in Construction
Regulations (HS(G)150) for temporary platforms.
Where the selected sample plane is located in a horizontal section of a large
rectangular duct, and where some of the sample points are positioned above a
convenient and safe working height (nominally 1.75m maximum for sample probe
handling), it will be necessary to provide a dual-level sampling platform of adequate
design so that sampling staff can carry out the full range of sampling requirements in
a safe and satisfactory manner.
Note: performing monitoring in horizontal sections of very large circular ducts
presents severe practical difficulties in obtaining access to all sample points. Such
situations should be avoided.
Removable chains or self-closing gates shall be used at the platform to prevent
workers falling through access hatches or ladders.
The platform shall have suitable weather protection for personnel and equipment. The
platform shall not accumulate free-standing water and, if necessary, drainage is to be
provided.
BS EN 13284-1:20024 contains requirements related to the working platform. There
are also a number of CEN standards covering platforms and access12, 13, 14, 15.
4.1.2

Space for the equipment and personnel

The space/size requirements for platforms are shown in Figures 4.1 to 4.4. The
platform surface area should not be less than 5m2. The minimum width at any point
shall be 2m. The minimum length in front of the access port shall be 2m or the length
of the probe (including nozzles, suction/support tubes and associated filter holders)
plus 1m (whichever is the greater). The platform should be wide enough to prevent
sampling equipment extending beyond the platform.
4.1.3

Means of entry into the stack (sampling access ports)

The access ports shall be big enough for the insertion and removal of the equipment
used and to allow the sampling points to be reached. BS EN 13284-1:20024
recommends that access ports have a minimum diameter of 125mm (illustrated in
Page 19 of 34

Figure 4.5) or a surface area of 100mm x 250mm, except on stacks smaller than 0.7m
diameter. For small stacks (less than 0.7m diameter) a smaller socket (for example
50mm (2-inch) nominal BSP) may be necessary, but there must be close liaison with
the sampling team to ensure the required probes and Pitots can be accommodated.
The port socket shall not project into the gas stream (see Figure 4.5). Typical
arrangements of access fitments are shown in Figures 4.6 and 4.7. The operator must
maintain the ports in good condition and free them up prior to work being undertaken.

The UK standard type of sampling port is a 4-inch nominal British Standard pipe (BSP) socket.
The aim is to simplify the insertion of an in-line 47mm filter holder, which is approximately 150mm
long.

Page 20 of 34

Figure 4.1 Typical area required behind probe for handling

Sampling line
Minimum of 2m or probe length + 1m
Duct

Platform area

Sampling line

Plan view

Figure 4.2 Plan view of platform working area and orientation recommendations
for small vertical circular stacks (<3.6m diameter)

Stack
Sampling platform
subject to minimum of 5m2

Sockets

Minimum of 2m or probe length + 1m

Page 21 of 34

Figure 4.3 Plan and side view of platform working area and orientation
recommendations for large vertical circular stacks (
3.6m diameter)
Sampling should be carried out from
four sample holes on these large stacks

Platform
Minimum width of platform at
any point is 2m

Stack

'L'
L (Length) = Probe length +1m
Probe length = stack diameter / 2

MONORAIL SUPPORTS

PROVISION FOR
LIFTING EQUIPMENT

Page 22 of 34

Figure 4.4 Plan view of recommended platform working area for square or
rectangular ducts (up to l1 = 3.6 m)

Minimum platform area (a x

b) shall be 5m2
Duct

Sampling
platform

l1

Depth of platform (a) = Probe length +1m

Duct side length, l1 (m)

Duct side width, w (m)
Width of platform, b (m)

0.6
0.6
2.0

0.8
8.0
2.0

1.0
1.0
2.0

1.2
1.2
2.0

1.4
1.4
2.0

1.6
1.6
2.0

1.8
1.8
2.1

2.0
2.0
2.2

Duct side length, l1 (m)

Duct side width, w (m)
Width of platform, b (m)

2.2
2.2
2.3

2.4
2.4
2.4

2.6
2.6
2.5

2.8
2.8
2.6

3.0
3.0
2.7

3.2
3.2
2.8

3.4
3.4
2.9

3.6
3.6
3.0

Figure 4.5 Standard 125mm access port

2
STACK WALL
1

129.30

1.

Flange BS10 125mm (5)

2.

Pipe stub 125mm schedule 40

3.

Pipe stub length should be a minimum of 75mm from the stack wall (recommended is 100mm)

4.

Recommendation that 1000mm of Unistrut P1000 is fitted vertically on the centre line of the
sample port for positioning of the monorail

Page 23 of 34

Figure 4.6 Typical arrangement of access fittings on large circular stacks (

3.6m
diameter)

Parallel-sided socket

Centre line
90
Plan view
Parallel-sided socket

Side elevation

Parallel-sided socket

Note - for small circular stacks, two ports at 90 will usually suffice.

Page 24 of 34

Figure 4.7 Typical arrangement of access fittings on large square/rectangular

stacks

Sampling centre line

Plan view

Sampling centre line

Parallel-sided
socket

Sampling centre line

Side elevation

Note - for small stacks, ports on one side will usually suffice.

Page 25 of 34

4.1.4 Essential services

The sampling position should have artificial lighting or have facilities for temporary
lighting. The sampling position should be well ventilated. Single phase 110 V
electrical power of a suitable current shall be provided by a suitable number of
outdoor waterproof sockets at the platform. Water, drainage and compressed air shall
be supplied at the request of the sampling team. When undertaking sampling of gases
using instrumental techniques, many monitoring organisations use mobile emissions
monitoring laboratories, which may require a 32amp or even a 63amp power supply.
When this type of sampling is undertaken operators need to provide parking facilities
for these vehicles within close proximity to the sampling location.
Lifting equipment is required for the raising and lowering of apparatus where access
to the sampling platform is by vertical, or steeply inclined, ladders or stairs. In all
such cases, the lifting equipment (for example, hoists) and attachments (for example,
eyes) must be installed, inspected and maintained by the site operator.
A sampling monorail should be attached above the sampling ports to enable certain
designs of sampling train to be suspended.
4.1.5 Other issues
The platform shall be, as far as, possible free from obstructions that would hamper
sampling.
Protection from adverse weather will usually be required for an outdoor sampling
position.
An additional access port to vent sample air back into the duct may be necessary if the
flue gas presents an exposure hazard.

4.2 Access and facilities for periodic monitoring of gases

Access and facilities requirements for gases alone are usually less demanding than for
particulates. In practice, meeting the requirements for particulates in Section 4.1 will
also satisfy the requirements for gases. When gas concentrations only are to be
monitored, smaller access ports than those specified for particulates may be
permissible following consultation and agreement with the sampling team.

4.3 Access and facilities for CEMs

The requirement for access and facilities will be dependent on:

whether the CEM requires calibration by periodic monitoring using a standard

method;
the need for routine maintenance and operational function checks (such as span
and zero checks).

Page 26 of 34

If the CEM requires calibration by periodic monitoring, then the access and facilities
should as a minimum comply with the requirements given in Section 4.1 and 4.2, for
particulates and gases respectively. Routine maintenance and calibration may
necessitate additional access and facilities.
If the CEM is calibrated by a means other than periodic monitoring, then the access
and facilities required will be solely those needed for the maintenance and calibration
activities. Such activities will vary from one CEM to another. In practice, however,
it is unlikely that the standard of access and facilities will be less than that specified in
Section 4.1. The latter should be the default requirement, with any relaxation needing
careful assessment and justification.
The CEN standard which covers quality assurance of CEMs, BS EN 14181:20049,
requires good access to the CEM to allow frequent inspection and to minimise the
time needed for calibration tests. The standard reiterates many of the specific
requirements given in Section 4.1.

4.4 The risk-management approach to site work

4.4.1 Safety has the highest priority
Safety must be considered at the earliest point in the monitoring strategy, and must be
revisited whenever a decision is made on sampling positions, access and
requirements. Any change to the sampling positions, access and requirements is
subject to the overriding requirement for the work to be carried out safely for
example, a modification to the sampling position that would theoretically provide a
technical improvement must not proceed if it introduces an unacceptable safety risk.
There are many hazards associated with stack monitoring. The basic principles of
health and safety must be applied. In particular, a risk assessment must be carried out
before starting work - this is a legal requirement. The risk assessment should be
documented. Risk Assessments should also be performed during the site review
process. Further guidance is available in the MCERTS Performance Standard for
Organisations1.
The risk assessment will assess the level of risk from each of the various hazards
present. If the risk is unacceptable, control measures must be used to reduce the risk
to an acceptable level before starting work. It is recommended that, as a minimum,
the routine hazards described in Section 4.4.6 of this TGN should be included in the
risk assessment. However, there may be other hazards - every site is different. The
reader is advised to consult the Source Testing Associations (STA) booklet Risk
Assessment Guide: Industrial-emission Monitoring16 which is updated frequently.
The focus of this section is on periodic monitoring, as this approach requires a
considerable amount of manual work on-site. However, the principles apply equally
when using CEMs and should be followed whenever manual work on CEMs is
carried out, for example, during installation, maintenance and calibration.

Page 27 of 34

4.4.2 Prominent hazards encountered when stack-emission monitoring

The most prominent hazards are shown in Figure 4.8. It must be reiterated that this is
not an exhaustive list and there may be other hazards. Every site is different.

Figure 4.8 Most prominent hazards associated with stack monitoring

Temperature
extremes
Wind, rain,
lightning, snow
and ice
Sunburn

Fire and
emergency

Site traffic

Mechanical
operations

Weather,
environment
and welfare

General site
hazards

Lone working

Chemical
operations

RISK TO
HEALTH &
SAFETY

Tiredness

Chemical
hazards in lab

Lifting

Physical
hazards at stack

Burns

Chemical
hazards at stack

Exposure to
substances used in
analysis and
cleaning

Falling

Electricity
Compressed gases

Exposure to
substances used in
monitoring tests

Exposure to
substances from
flue gases

4.4.3 The risk-assessment process

The fundamental stages of the risk-assessment process can be summarised as:
Step 1

Step 2

Step 3

Identify the
HAZARDS

Assess the RISKS

Apply CONTROL
MEASURES

Page 28 of 34

Because the risk assessment forms the foundation of safe working on stacks, it is very
important to understand the distinction between hazard and risk. These terms are
defined in Box 4.1.
Box 4.1 The distinction between hazard and risk

Hazard - a substance or physical situation with inherent potential to cause harm

Risk - an estimation of the likelihood of that potential being realised, within a
specified period or in specified circumstances, and the consequence

Using an extreme example to illustrate this difference between hazard and risk,
working on stacks is associated with the hazard of slips on icy platforms. The risk of
slipping is high on a cold winters day, but is very low in summer.
The first step of the risk assessment is therefore to identify the hazards that will be
faced. Then, a judgement must be made on what the risk will be (that is, the
likelihood of an accident) in the light of all the relevant factors. If the risk is not
acceptable*, then control measures must be put in place to reduce the risk. Only then
should work commence. Control measures can be:

engineering - for example, a self-closing gate to reduce the risk of falls from the
platform;

procedural - for example, permit-to-work systems;

personal protective equipment (PPE) - for example, safety goggles to reduce the
risk of eye injury when opening access ports.

For many hazards, there will be a choice of control measures. PPE measures should
be used to reduce the risk further only when other measures fail to reduce the risk to
as low as reasonably practicable. PPE shall not be the control measure of first choice.
4.4.4

Responsibilities

The Management of Health and Safety at Work Regulations 1999 has many
requirements. Specifically, it places a duty of care on employers to:

ensure the competency of employees through training and experience to allow

them to work in a safe manner;
adequately assess the health and safety risks of employees and any other persons
not in his employment who may be affected by the works;
identify and implement appropriate risk-control measures to ensure the safety of
employees and other persons;
review risks and control measures regularly to ensure continued validity.

The criterion for the risk being acceptable is that the risk shall be as low as is reasonably
practicable, commonly known by the abbreviation ALARP.

Page 29 of 34

For operators carrying out stack-emissions monitoring using their own staff, the
responsibility for the risk assessment and safe systems of work is straightforward.
Where operators contract-out the monitoring, both the operator and monitoring
contractor have the responsibility as site occupier and employer to follow safe
systems of work.
The operator also has a responsibility to employ competent monitoring organisations
and the monitoring organisation has a responsibility to ensure employees are
competent to carry out the work safely.
Where work is being carried out by a monitoring organisation directly for the
Environment Agency, the monitoring organisation, the operator and the regulator all
have responsibilities in this regard. In practice, the monitoring team is usually best
placed to carry out a comprehensive risk assessment of the work they carry out on
site. Since they are also usually subject to the greatest risk, the site team should
always carry out their own risk assessment, regardless of whether the operator has
done so.
4.4.5 When to carry out a risk assessment
Where a monitoring organisation has not previously carried out sampling from a
particular stack, a site review will be necessary to:

carry out a risk assessment;

assess the suitability of the sampling position;
deal with practical issues to ensure subsequent monitoring proceeds smoothly (for
example, suitability of equipment, length of Pitots and probes required).

It is very important to understand that all safety improvements and control measures
must be in place before the monitoring team starts work. For this reason, a risk
assessment carried out as part of a site review before the monitoring visit is
preferable, as this gives the site operator time to implement any control measures
found to be necessary.
Where the monitoring organisation has carried out sampling at the site on previous
occasions, a full site review may not be necessary. However, the monitoring team
must still carry out a risk assessment before they start sampling work. It is important
that this risk assessment is carried out at the start of every monitoring campaign at a
site, even if the team has been before. This is a good discipline because it focuses the
teams attention on safety as the first issue to address on-site. It also reduces the
possibility of the staff becoming complacent after several visits because they feel they
know all the issues.
4.4.6 Getting the operator involved
When the monitoring team arrives on-site, its first task must be to conduct the riskassessment. If the team has not worked on the site before, the operators
representative should accompany them as they make the assessment for reasons of
safety and to provide knowledge of specialised site-specific issues (for example,

Page 30 of 34

process conditions, escape routes). It is not always necessary for the team to be
accompanied during the risk assessment when they return for repeat visits.
However, whether it is the teams first visit to the site or a repeat visit, the operators
representative should be briefed on the findings of the completed risk assessment.
This will give the operator the opportunity to raise any additional issues of concern,
and comment on the findings of the assessment and the control measures the
monitoring team has decided will be necessary. If the operator has any relevant
comments, these should be considered. They can be incorporated into the assessment
and its findings if they reduce the risk further, but not if they hinder the reduction of
risk.
Some of the control measures required may need to be put in place by the operator.
The operators confirmation that these have been completed should be obtained
before monitoring work is started.
When the site work is finished, any relevant comments should be noted on the risk
assessment based on any lessons learnt. This will be useful for the next visit.
4.4.7 Reporting accidents and near misses
There is a statutory requirement to report certain accidents under the Reporting of
Injuries Diseases and Dangerous Occurrences Regulations (RIDDOR) 1995. Any
incidents coming under these criteria must be reported to the Health and Safety
Executive. However, there are additional benefits to be gained from informing
relevant organisations such as the STA and the Environment Agency about such
events as well as other, non-reportable, accidents or near misses. Though some of the
occurrences may be minor, knowledge about them aids the development of
procedures and guidance to avoid future accidents.
The STA produces guidance for its members on safe working procedures for stack
monitoring. This is regularly revised and updated, taking into account legislative
changes and lessons learned from incidents and near misses. The STA also has a fasttrack system for disseminating urgent safety information to members. To enable the
STA to give maximum value to its members, it asks monitoring teams and operators
to submit details of incidents and near-misses. These submissions are treated in the
strictest confidence and can, if desired, be made anonymously.
The Environment Agency has its own reporting procedure for accidents, damage,
environmental impact and near misses involving its own staff, and/or monitoring
teams working under direct contract to the Environment Agency. The Environment
Agency passes details of all generic safety incidents and resultant lessons learned to
the STA to allow the source testing industry to benefit from this information. The
information passed to the STA does not contain details on the sites or the monitoring
teams involved.
4.4.8

Further guidance

Box 4.2 shows some of the major items of legislation and guidance relevant to the
hazards depicted in Figure 4.8. The STA has produced a health and safety booklet15,
which describes each of the most prominent hazards and the factors that affect the risk
Page 31 of 34

of an accident from them. This guidance, together with other relevant guidance,
should be consulted as part of the risk assessment process. However, safety
management is a fast-evolving subject and new guidance, and new and revised
legislation appears frequently. All parties involved in stack-emission monitoring
should ensure they take account of the most recent developments.

Box 4.2 Legislation and guidance relevant to stack-emission monitoring

health and safety
Legislation

Electricity at Work Regulations 1989

Personal Protective Equipment at Work Regulations 1992
Workplace (Health, Safety and Welfare) Regulations 1992
The Carriage of Dangerous Goods (Classification, Packaging and Labelling) and Use of
Transportable Pressure Receptacles Regulations, 2004
The Lifting Operations and Lifting Equipment Regulations 1998 (LOLER)
Control of Substances Hazardous to Health Regulations 1999 (COSHH)
Management of Health and Safety at Work Regulations 1999
Noise at Work Regulations 1989
Construction Health and Welfare Regulations 1996
Manual Handling Operations Regulations 1998
Confined Spaces Regulations 1999
Work at Height Regulations 2005

Guidance

HS(G)150, Health and Safety in Construction: guidance on small lifting equipment

HS(G)107, Maintaining portable and transportable electrical equipment
HSE guidance leaflet INDG308 5/00 C1200, The Safe Use of Gas Cylinders
HSE information leaflet IND(G)147(L), Keep Your Top On Health Risks from Working in
the Sun
HSE Guidance, 5 Steps to Risk Assessment
HSE Guidance, A Guide to Risk Assessment Requirements
HSC Safe work in confined spaces
BS4211 Specification for ladders for access to chimneys, other high structures, silos and bins
STA Health Safety Guidance Note HSGN0008, Hazards, Risks and Risk Control in Stack
Testing Operations

5.

Status of this document

5.1

Version 3 of TGN M1 replaces version 2, which is withdrawn and should be

discarded.

5.2

Version 3 may be subject to review and amendment following publication. For

the latest version go to the Environment Agencys website at :
www.environment-agency.gov.uk and put M1 into the Search facility.

Page 32 of 34

Annex 1 Sample point requirements as specified in ISO 9096:2003

Table 1 Minimum number of sampling points for circular stacks
Range of duct
diameters
(m)

Minimum
number of
sampling lines
(diameters)

Minimum number of
sampling points per
line,
Incl.
Excl.
centre
centre
point
point

Minimum number of
sampling points per
plane,
Incl.
Excl.
centre
centre
point
point

<0.35

1a

1a

0.35 to 0.70

0.70 to 1.00

1.00 to 2.00

13

12

>2.00

17

16

a. Using only one sampling point can give rise to errors greater than those specified in ISO 9096:2003

Table 2 Minimum number of sampling points for rectangular stacks

Range of
sampling plane
areas (m2)
<0.09

Minimum number of side

divisions a

Minimum number of sampling points

per plane

1b

0.09 to 0.38

0.38 to 1.50

>1.50

16

Note a: Other side divisions may be necessary, for example, if the longest stack side length is more than twice
the length of the shortest side (see Figure 2.4). Sample lines for square and rectangular stacks are explained in
Section 2.1.8.2
Note b: Using only one sampling point can give rise to errors greater than those specified in ISO 9096:2003

Page 33 of 34

References
1.

MCERTS, Manual stack-emission monitoring: performance standard for

organisations, Version 4, September 2003. Environment Agency.

2.

MCERTS, Manual-stack emission monitoring: personnel competency standard,

Version 3, July 2005, Environment Agency.

3.

Technical Guidance Note M2 Monitoring of Stack Emissions to Air, Version 3.1, June
2005 Environment Agency.

4.

BS EN 13824:2002 Stationary source emissions Determination of low range mass

concentration of dust Part 1 Manual gravimetric method..

5.

BS ISO 9096:2003 Stationary source emissions Manual determination of mass

concentration of particulate matter.

6.

Quality Guidance Note QGN001 r1, Guidance on Assessing Measurement

Uncertainty in Stack Monitoring, Source Testing Association, September 2004.

7.

ISO 10780:1994, Stationary source emissions Measurement of velocity and volume

flowrate of gas streams in ducts.

8.

ISO 10396: 1993, Stationary source emissions Sampling for the automated
determination of gas concentrations.

9.

BS EN 14181:2004. Stationary source emissions Quality assurance of automated

measuring systems.

10.

BS ISO 14164: 1999, Stationary source emissions Determination of the volume

flowrate of gas streams in ducts Automated method.

11.

Work at Height Regulations 2005. Statutary Instrument No. 735. ISBN 0 11 0725638.
HMSO.

12.

EN ISO 14122-1: 2001, Safety of machinery Permanent means of access to

machines and industrial plants Part 1: Choice of a fixed means of access between
two levels.

13.

EN ISO 14122-2: 2001, Safety of machinery Permanent means of access to

machines and industrial plants Part 2: Working platforms and gangways.

14.

EN ISO 14122-3: 2001, Safety of machinery Permanent means of access to

machines and industrial plants Part 3: Stairways, stepladders and guard-rails.

15.

EN ISO 14122-4: 1996, Safety of machinery Permanent means of access to

machines and industrial plants Part 4: Fixed ladders.

16.

Risk Assessment Guide: Industrial-emission Monitoring. Source Testing Association.

www.S-T-A.org (updated frequently).

Page 34 of 34