XXX Radiographs

HSE
Health & Safety

Executive
Recommended practice for the rapid inspection

of small bore connectors using radiography
Prepared by AEA Technology plc for the

Health and Safety Executive 2005
RESEARCH REPORT 294
HSE
Health & Safety

Executive
Recommended practice for the rapid inspection

of small bore connectors using radiography
S F Burch (BA, Ph D) and N J Collett

AEA Technology plc
551.11 Becquerel Avenue
Harwell International
Business centre
Harwell
Oxfordshire
OX11 0QJ
This document covers recommended practice for the radiographic inspection of corrosion in small bore
piping and in the small bore connections to larger diameter main pipes. Inspection of this type of
component is not covered by any internationally recognised standards or procedures, and there is
therefore a need for guidance on recommended practice.
Recommendations are given concerning techniques for detection of corrosion within different regions
of small bore connectors and their junction with main pipes. A distinction is drawn between corrosion in
the form of relatively uniform loss of wall and isolated corrosion pits. Recommendations are given
regarding the key technical aspects of in-situ radiographic inspection techniques including radiation
sources, film types, computed radiography systems, positioning of the source and film (or filmless
plate) and methods for reduction in exposure time.
This report and the work it describes were funded in part by the Health and Safety Executive (HSE). Its
contents, including any opinions and/or conclusions expressed, are those of the authors alone and do
not necessarily reflect HSE policy.
HSE BOOKS
Crown copyright 2005

First published 2005
ISBN 0 7176 2939 2
All rights reserved. No part of this publication may be
reproduced, stored in a retrieval system, or transmitted in
any form or by any means (electronic, mechanical,
photocopying, recording or otherwise) without the prior
written permission of the copyright owner.
Applications for reproduction should be made in writing to:
Licensing Division, Her Majesty's Stationery Office,
St Clements House, 2-16 Colegate, Norwich NR3 1BQ
or by e-mail to hmsolicensing@cabinet-office.x.gsi.gov.uk
ii
CONTENTS
Page
1
2
Introduction ........................................................................................................................... 1
Radiographic Methods For Corrosion Detection In Pipes .................................................... 2
2.1 Tangential method............................................................................................................. 2
2.2 Straight-through method ................................................................................................... 3
2.3 Both methods combined.................................................................................................... 5
2.4 Inspection of junction with main pipe............................................................................... 5
3
Radiography Equipment........................................................................................................ 9
3.1 Radiation Sources.............................................................................................................. 9
3.1.1 Type of source........................................................................................................... 9
3.1.2 Size and strength of sources .................................................................................... 10
3.1.3 Source containers and collimation .......................................................................... 10
3.2 In-situ inspection of plant................................................................................................ 11
3.3 Film and screens.............................................................................................................. 11
3.3.1 Conventional film.................................................................................................... 11
3.3.2 Fluorometallic screens............................................................................................. 12
3.4 Computed (filmless) radiography.................................................................................... 12
3.4.1 Image Processing on filmless systems .................................................................... 13
3.5 Dimensional Indicators ................................................................................................... 14
4
Equipment setup.................................................................................................................. 15
4.1 Inspection Techniques..................................................................................................... 15
4.1.1 Technique 1: Flat film, beam axis perpendicular to main pipe ............................... 15
4.1.2 Technique 2: Film flat against connector, beam axis perpendicular to main pipe .. 15
4.1.3 Technique 3: Film flat against connector, beam axis parallel to main pipe ............ 16
4.1.4 Technique 4: Angled Film flat, beam axis perpendicular to main pipe .................. 17
4.2 Source to film distance.................................................................................................... 17
4.3 Exposure times and film density ..................................................................................... 21
4.3.1 Fine grain film with lead screens ............................................................................ 21
4.3.2 Other film/detectors for reduced exposure times .................................................... 22
4.4 Image Quality Indicators................................................................................................. 23
5
Overall performance............................................................................................................ 24
5.1 Image quality................................................................................................................... 24
5.2 Examples ......................................................................................................................... 25
5.2.1 Comparison between Ir 192 and Se 75 using fine grain film with lead screens...... 25
5.2.2 Example of application to larger diameter (10 inch) main pipe specimens ............ 27
5.2.3 Example of use of fine grain film with fluorometallic screens ............................... 29
5.2.4 Examples obtained with computed radiography systems........................................ 30
5.3 Inspection of approximately uniform corrosion/erosion ................................................. 32
5.3.1 Wall loss external to the main pipe ......................................................................... 32
5.3.2 Wall loss within the main pipe wall ........................................................................ 32
5.4 Detectability of isolated corrosion pits............................................................................ 33
6
Summary of relevant safety standards and legislation ........................................................ 35
7
Main recommendations ....................................................................................................... 35
8
Recommendations for further work .................................................................................... 37
9
Acknowledgements ............................................................................................................. 37
10 References ........................................................................................................................... 37
iii
iv
EXECUTIVE SUMMARY
This document covers recommended practice for the radiographic inspection of corrosion in
small bore piping and in the small bore connections to larger diameter main pipes. Inspection of
this type of component is not covered by any internationally recognised standards or
procedures, and there is therefore a need for guidance on recommended practice.
The information obtained from the HOIS2000 radiographic evaluation trials is used as the basis
for this recommended practice document. Appropriate standards are also referenced, including
ISO 2919: 1999 and ISO 3999-1:2000, which cover radiation protection aspects of gamma ray
sources and containers. BS EN 1435 which is a standard for radiographic examination of welds
is also referenced.
Recommendations are given concerning techniques for detection of corrosion within different
regions of small bore connectors and their junction with main pipes. A distinction is drawn
between corrosion in the form of relatively uniform loss of wall and isolated corrosion pits.
Recommendations are given regarding the key technical aspects of in-situ radiographic
inspection techniques including radiation sources, film types, computed radiography systems,
positioning of the source and film (or filmless plate) and methods for reduction in exposure
time.
vi
1 INTRODUCTION
The inspection of complex geometries was identified as a key area for work within the
HOIS2000 core programme, at the start of the new joint industry programme (HOIS2000) in
April 1999. Standard NDT methods are routinely used to inspect vessels, tanks and medium and
large bore pipework. However, the standard methods have difficulties in application to more
complex components and fittings such as connectors, especially those of small bore, tight bends,
T-pieces, injectors, etc. Speed of inspection was also identified as a key parameter.
In the year 2001/02, an evaluation of radiographic techniques for the inspection of corrosion in
small bore connectors and weldolets was carried out (Burch & Collett, 2002). For these trials,
specimens containing simulated corrosion in the small bore connectors attached to larger
diameter main pipes were first designed and then manufactured. These specimens contained
controlled and accurately known levels of corrosion. The trials on these small connector
specimens involved evaluation of current radiographic techniques followed by an investigation
of more novel radiographic techniques which would be capable of faster application to plant.
These included the new computed (filmless) radiographic systems which are beginning to find
application to industrial NDT.
Inspection of this type of component is not covered by any internationally recognised standards
or procedures, and there is therefore a need for guidance on recommended practice.
The information obtained during the radiographic trials is used as the basis for this
recommended practice document, covering the inspection of small bore connectors for corrosion
using radiographic methods. Appropriate standards are referenced, including ISO 2919 and ISO
3999-1, which cover radiation protection aspects of gamma ray sources and containers, as well
as BS EN 1435 which is a standard for radiographic examination of welds.
Recommendations are given concerning techniques for detection of corrosion within different
regions of small bore connectors and their junction with main pipes. A distinction is drawn
between corrosion in the form of relatively uniform loss of wall and isolated corrosion pits.
Recommendations are given regarding the key technical aspects of in-situ radiographic
inspection techniques including radiation sources, film types, computed radiography systems,
positioning of the source and film (or filmless plate) and methods for reduction in exposure
time.
It is important to note that the trials were completed under laboratory conditions inside an
exposure compound or similar controlled and uncluttered environment. This made it
straightforward to set-up the radiography equipment in such a way as to achieve safe radiation
dose rates in a small controlled area, whilst using highly penetrating radioisotopes and the
source to film distances (SFD) recommended in this document. The practical application of
these recommendations under typical site conditions may need modification for particular
circumstances. In-situ, local access to small bore connectors is often restricted, limiting the
choices for the source position and shielding. However, the appropriate legislation requires the
radiography contractor to conduct a prior risk assessment of the work and it is at this stage that a
practical assessment should highlight potential difficulties so that solutions can be planned in
advance.
In addition it should be noted that the laboratory samples on which the trials were carried out
did not contain liquid or scale. However, due to the relatively small diameter of the small bore
piping and weldolets (the maximum internal diameter was 26 mm), it is not anticipated that the
presence of a low density liquid such as oil or water, or relatively low density scale would
significantly affect the performance of the radiographic techniques evaluated. Nevertheless

further trials could be carried out to check this, if considered necessary.
Note that specific recommendations are highlighted in the document using bold text.
2 RADIOGRAPHIC METHODS FOR CORROSION DETECTION

IN PIPES
Using radiography, corrosion can be detected using two different exposure set-ups or modes,
both of which can often be used on the same radiograph of a small diameter component such as
a small bore connector.
2.1
TANGENTIAL METHOD
Firstly, when used in tangential mode, a radiograph shows a direct image of the pipe wall.
Corrosion producing loss of wall can then be detected and measured by the reduction in the
imaged wall thickness, as illustrated in Figure 2.1.
X or gamma-ray source
Pipe
Extended area of corrosion
Film
Image of pipe wall on film
Figure 2.1 Principle of tangential radiography

The extent of the loss of wall due to extended areas of either internal or external corrosion can
be measured directly from the radiographic images, provided appropriate calibration techniques
are used to allow for the enlargement of the radiographic image caused when the distance
between the object and film is non zero - see Section 3.4 for further details.
The tangential method inspects only a small extent of the circumference of the pipe for a single
source/detector position. Thus an isolated pit is likely to fall outside the inspection region,
unless a several radiographs are taken around the circumference of the pipe (by moving the
source and detector together circumferentially around the pipe).
For small bore connectors attached to larger diameter pipes, geometric constraints generally
preclude use of many different circumferential angles for the radiation beam (see Section 4.2 for
further details).
Also, even if an isolated pit is located within the inspection region, its through wall extent will
be underestimated unless it is positioned very close to the exact position at which the radiation
pipe forms a tangent to the pipe wall, as illustrated in Figure 2.2.
The tangential radiography method is recommended for the detection and through-wall
sizing of extended areas of corrosion, and should not generally be used for isolated
corrosion pits, unless their circumferential location has already been established using for
example the density difference or straight-through method (see below).
Figure 2.2 Illustration of coverage region for tangential radiography and the location of an
isolated pit for which the through wall extent would be underestimated
2.2
STRAIGHT-THROUGH METHOD
Radiography can also be used in a straight-through mode, whereby corrosion can be detected
by the increased film density produced by the loss of material (see Figure 2.3).
With this method, the size of the corrosion pits in the circumferential and axial directions can be
measured from the radiograph, provided methods are used for calibration of distances - see
Section 3.4. However, the through wall extent of the corrosion cannot be measured directly,
although some information on this can be gained from the magnitude of the change in film
density.
This method is recommended for the detection of isolated corrosion pits, and can also be
used for detection of extended areas of corrosion.
Pipe
Corrosion
Image of corrosion pit on

film
Film
Figure 2.3 Detection of corrosion by increased film density in straight-through mode
2.3
BOTH METHODS COMBINED
For relatively thin-walled, small bore pipes and weldolets, a single radiograph has sufficient
dynamic range and size to show the presence of corrosion by both of the tangential and density
difference methods. Note that the source position should then be central to the small bore pipe,
as shown in Figure 2.4.
Small bore pipe
Film
Figure 2.4 Radiography of a small bore pipe, combining both the tangential and straightthrough modes in a single radiograph
For this combined type of radiography, as shown in Figure 2.4, the tangential method can be
applied to both sides of the pipe on the same radiograph, and the region in between will show
any loss of wall by increased film density.
2.4
INSPECTION OF JUNCTION WITH MAIN PIPE
For small bore connectors, the junction with the main pipe leads to there being a distinction
between the sections of the connector which are external to the larger main pipe wall, and the
region within the main pipe wall, as illustrated in Figure 2.5.
For detection of corrosion within the wall of the main pipe, substantially larger metal paths need
to be penetrated, compared with the sections external to the main pipe wall.
Corrosion within
wall of main pipe
Corrosion
external to
main pipe wall
Source
Tangential
path
Film/detector
Figure 2.5 Small bore connector inspection showing different locations of potential corrosion
sites, and maximum path through wall of main pipe, T
For a pipe with wall thickness WT and outside diameter OD, the maximum metal path, T,
through the pipe wall occurs for a line forming a tangent with the inner diameter (as shown on
Figure 2.5). This maximum path is given by
T = 2 WT(OD - WT )
(2.1)
Note that this applies to any line drawn through the pipe, forming a tangent to the inner surface
of the pipe. Thus T is independent of the source position.
Values for the maximum path, T, through schedule 40, 80 and 160 pipes of various diameters
are given in Table 2.1, for ease of reference. Note that these paths are generally much larger
than twice the wall thickness of the main pipe. For example, T is 100 mm or larger for 8 inch
schedule 80 pipes and above.
Table 2.1 Maximum paths through schedule 40 & 80 pipes of various diameters
Nominal Bore
Outside
diameter, OD
(inches)
(mm)
60.3
88.9
114.3
141.3
168.3
219.1
10
273.0
12
323.8
Schedule
40
80
160
40
80
160
40
80
160
40
80
160
40
80
160
40
80
160
40
80
160
40
80
160
Wall
thickness, WT
(mm)
Max
Tangential
path
(mm)
3.9
5.5
8.7
5.5
7.6
11.1
6.0
8.6
13.5
6.6
9.5
15.9
7.1
11.0
18.3
8.2
12.7
23.0
9.3
15.1
28.6
10.3
17.5
33.3
29.7
34.7
42.4
42.8
49.7
58.8
51.0
60.3
73.8
59.6
70.8
89.3
67.7
83.2
104.8
83.2
102.4
134.3
99.0
124.8
167.2
113.6
146.4
196.7
The same information is presented graphically in Figure 2.6, which shows the maximum path
lengths as a function of pipe diameter.
Tangential path lengths

200
Co 60
180
Schedule 40
Schedule 80
Schedule 160
Limit for Se 75
Tangential path (mm)
160
140
120
100
Ir 192
80
60
40
Se 75
20
0
0
10
Pipe nominal bore (inch)
Figure 2.6 Maximum (tangential) path lengths through the walls of pipe of different diameter.
The maximum penetrated thicknesses for different isotope sources, as recommended in BS EN
1435 are also shown.
Also shown on Figure 2.6 are the maximum penetrated thickness for Se 75, Ir 192 and Co 60, as
recommended by BS EN 1435 for class A radiography. It can be seen that even the smaller
diameter main pipes have maximum path lengths which exceed the recommended limit of
40 mm for Se 75. For Ir 192, the limit of 100 mm is reached for a 6 inch 160 schedule pipe, a
8 inch schedule 80 pipe and a 10 inch schedule 40 pipe. Note that in some cases it is possible to
exceed the thickness limits shown in Figure 2.6, using for example computed radiography.
For larger path lengths, Co 60 has more penetrating power but is much less widely used in the
oil and gas industry due to greater radiation safety hazards.
For further information on wall thickness and source selection, see Section 3.
12
3 RADIOGRAPHY EQUIPMENT
3.1
RADIATION SOURCES
3.1.1
Type of source
The majority of in-situ on site radiography is carried out using gamma-ray emitting isotope
sources, although portable, light-weight X-ray sources can also be used in cases where the need
for electrical power and high voltages do not cause significant safety issues.
For the inspection of the sections of the small bore connectors external to the main pipe,
the two recommended isotope sources are Iridium 192 and Selenium 75.
Of these, Iridium 192 is a conventional isotope source used for inspection of medium steel
thicknesses (c. 10-100 mm, Halmshaw, 1995). The energy of the majority of the gamma ray
photons emitted is around 300 keV and the half-life is 74 days.
The isotope source Selenium 75 has been developed for industrial radiography more recently
than Ir 192. It has a lower energy gamma ray spectrum than Ir192 with main peaks at 137keV
and 265keV and a longer half life (120 days). Source strengths are available between about 2
and 80 Curies with physical sizes ranging from 1x1 mm to 3x3 mm.
Sources and their containers should comply with the requirements of ISO 2919: 1999 and ISO
3999-1:2000, which include integrity testing to 800 C.
Due to the lower gamma ray energies emitted by Selenium 75 compared with Iridium 192,
Se 75 gives radiographs with higher contrast on components with moderate steel thicknesses,
including the majority of small bore piping.
Source selection should be made according to the following criteria:
(i) Corrosion in small bore piping external to main pipe wall
For detection of small isolated corrosion pits and sizing in the circumferential and axial
directions, by the density difference (or straight-through) method described in Section 2,
use of Se 75 would be preferred to that of Ir 192 due to the higher contrast obtained with the
former source.
For detection and through wall sizing of generalised loss of wall by the tangential method
described in Section 2, both Ir 192 and Se75 have similar fit-for-purpose capabilities.
(ii) Corrosion in main pipe wall at junction with small bore connectors
For main pipes which have maximum penetrated paths less than 50 mm (see Table 2.1 and
Figure 2.6), both Ir 192 and Se75 have similar fit-for-purpose capabilities.
Ir 192 is recommended for main pipes for which the main path T exceeds c. 50 mm (see
Table 2.1 and Figure 2.6), but is less than c. 100 mm.
For maximum path lengths of order 100mm and larger even Ir 192 will not provide
adequate penetration and use of a gamma ray source generating a higher photon energy (e.g.
the 1 MeV photons generated by Cobalt 60) should be considered, although it is recognised
that there are additional safety issues involved in the use of this type of source.
Se 75 with computed radiography (see Section 3.4) may be applicable to paths greater than
50 mm (up to 70 mm), but this was could not be verified for small bore piping inspection
during the HOIS2000 radiographic evaluation trials due to the limited range of specimens
available.
3.1.2
Size and strength of sources
In the UK offshore environment, the limit for Ir 192 is normally 25 Ci, but depending on any
other sources are already in store, this may limit the maximum size of a single source. Thus
sometimes single sources can be limited to c. 10 Ci, but higher strength sources in the range 15
to 40 Ci have been used with special risk assessments.
For Se 75, higher activity sources can be used, due to the lower penetrating power of the
radiation produced. Source strengths of up to 30 curies can be used but there is no generally
accepted upper limit offshore.
Note that radiography contractors should state the maximum strength isotopes within their Local
Rules as required by IRR 1999.
The source size for inspection of small bore piping should be chosen taking account of the
limits on geometric unsharpness given in Section 4.2. Typically source dimensions for this
application will be about 2mm.
On occasion, use of smaller sized sources (e.g. 1 x 1 mm) may be considered preferable, as
these will lead to smaller geometric unsharpnesses for the same source to film distances. For
Ir 192, newly manufactured high activity sources of size 1.2 x 1.2 mm can have a strength of c.
8 Ci, and a 2 x 2 mm source can have an activity of c. 44 Ci.
The effective source size for geometric sharpness calculations should be used to calculate the
required source to film distance, as given in Section 4.1. Larger physical sized sources will
require greater source to film distances than smaller sized sources to obtain the same geometric
unsharpness.
3.1.3
Source containers and collimation
The source containers should conform with the requirements for source containers given by
ISO3999-1:2000 or BS5650:1978 ISO 3999-1977 and any applicable national standards.
Conventional projection equipment can be used for the radiography of small bore piping,
provided the requirements of the current radiation safety regulations are complied with. For
these systems, a large radiation controlled area (radius 100 m) is normally needed, which often
requires out of hours working, or even shutdown of plant.
Systems which keep the source within a single container or single container/collimator assembly
are now available (e.g. SCAR, SafeRad) which allow a much smaller size of controlled area (of
order 1m - 5 m). These systems reduce radiation doses to operators, and the small size of the
controlled area generally means that plant operation does not need to be interrupted when site
radiography is underway.
10
The selection of an appropriate source container and deployment system depends on the balance
of the above factors for each individual site and inspection application, together with economic
considerations.
Careful collimation of the sources is recommended to minimise unwanted radiation, and
reduce the effects of scatter on the radiograph.
3.2
IN-SITU INSPECTION OF PLANT
Use of gamma ray radiography equipment for in-situ inspection of plant involves significant
safety issues associated with the use of ionising radiation. The appropriate mandatory safety
regulations appropriate to the plant must be adhered to (IRR 1999 in the UK). These include the
construction and maintenance of radiation controlled areas, by means of appropriate barriers.
Pre-planning of the inspection work to be carried out on a plant is required, to include both a
risk assessment and a practical assessment of how the source container and shielding will be
placed (IRR 1999).
3.3
FILM AND SCREENS
3.3.1
Conventional film
Radiographs with the highest quality in terms of spatial resolution, contrast and granularity will
be achieved with a fine grain high contrast film.
All other detection techniques described in this document give some reduction in image quality,
but with the main benefit of reduced exposure time.
Use of different types of film for weld radiography is covered by BS EN 1435. In this standard,
two classes of radiography are defined, class A (basic) and class B (improved). The class B is
generally used for more demanding applications such as detection of fine cracks.
The detection of corrosion, which is a form of volumetric defect, is considered less demanding
for radiography than fine crack detection, and hence class A radiography, as defined by BS EN
1435 should generally be sufficient.
For Class A radiography and Se 75/Ir 192 sources, BS EN 1435 recommends a C 5 film class
(as defined is BS EN 584-1), common examples of which are Agfa D7, Fuji 100, Kodak AX
and DuPont 70. Use of front and back lead screens is also required; a thickness of 0.1mm for
both screens is consistent with the requirements of BS EN 1435.
Thus the recommended standard film class is C 5 (Agfa D7, Fuji 100, Kodak AX and DuPont
70), although it is recognised that for particular applications such as detection of very slight
corrosion in very small wall thickness pipes, a finer grain film may be required.
The advantages of fine-grained film used with lead screens are:
Provides the highest spatial resolution and overall image quality of all currently available
detection systems.
Low-cost, unless used in very large quantities.
Well established, well understood characteristics and universally accepted.
Forms a direct hardcopy which can be readily archived
Disadvantages of film include:
11
Relatively long exposure times

Need for film development which involves use of a dark room, and safe use and disposal of
the necessary chemicals.
Accurate dimensional measurements not straightforward
3.3.2
Fluorometallic screens
Fluorometallic screens provide an alternative to lead intensifying screens and, when used with
fine grain film (Agfa D7), give significant reductions in exposure times, albeit at the expense of
some loss of image sharpness.
The results of the HOIS2000 evaluation trials (Burch and Collett, 2002) showed that when used
with Ir 192, NDT 1200 fluorometallic screens with fine grain film gave approximately a factor
five reduction in exposure time compared with fine grain film used with lead screens (see
Section 4.3).
Use of fluorometallic screens with medium grain film (Agfa F8) was also investigated in the
HOIS2000 evaluation trials. A further reduction in exposure time was obtained, but with a
significant reduction in image quality. This combination is not therefore generally
recommended for applications requiring high sensitivity. However if the tolerable defect
sizes are large and hence the required sensitivity is low, then this technique could provide a fit
for purpose method for reducing exposure time.
3.4
COMPUTED (FILMLESS) RADIOGRAPHY
Technical developments have led to systems that do not rely on film as the detection medium.
These fall into two categories, those which use a storage phosphor technology and those
employing an array of semiconductor detectors. Only the phosphor technology has been tested
within this programme of work. Originally developed for medical applications the storage
phosphor technology is now finding use in industrial radiography.
Storage phosphor systems use a flexible plate coated with a type of phosphor that stores the
incident radiation as a latent image within the phosphor (composition used depends on the
manufacturer). The latent image can be read off the plate into a computer as a digital picture
using a special scanner. This system is sometimes referred to as computed radiography (CR) or
digital radiography.
Imaging plates have good linearity and high dynamic range, exceeding conventional film in
both cases. They also have a good sensitivity to gamma and X-rays. The same plate can be used
many times. The limiting factor is mechanical wear and damage to the plate rather than
degradation in the luminescent properties.
In practical implementations the phosphor is coated onto a flexible substrate and the plate can
be used like a conventional film. Readout is achieved by scanning over the plate with a fine spot
of laser light at a suitable wavelength to stimulate the photoluminescence of the phosphor
material. Typically the scan lines are 100 microns apart (10 lines per mm ) although some
scanners can achieve 50 microns ( 20 lines per mm ). After readout some residual image may
remain on the plate so to erase it fully a very bright light is used. Ambient light does not cause
an image on the plate unlike conventional film, although very bright light can cause erasure of a
stored image as noted. However overall the plates require less precautions against ambient light
exposure than conventional film.
12
The advantages of the computed radiography (filmless) systems include:
Significantly reduced exposure times compared with fine grain film used with lead screens.
Greater exposure latitude than film (i.e. acceptable images can be obtained over a broader
range of exposures than with film).
Greater exposure latitude also permits reasonable interpretation of a large range of material
thicknesses within the same image reducing the need for multiple exposures required by
film.
No chemical processing needed (no disposal of chemicals, no need for dark room).
The filmless plates can be re-used many times.
Images directly read into a computer for subsequent enhancement and straightforward onscreen measurements of wall-thickness.
Multiple film-like hard copies of images can be produced which are interpretable on
standard film viewers.
The disadvantages of this technology include:
Relatively new recording medium, characteristics not fully understood or documented.

Resulting radiographs have lower spatial resolution than fine grain film used with lead
screens (comparable with fluorometallic screens used with fine grain film).
High capital cost of system.
Film-like hardcopying facility is non-standard and additional cost.
3.4.1
Image Processing on filmless systems
For computed (filmless) radiography, the raw digital images are generally stored on disc in a
proprietary 12-bit/pixel format.
On some systems, various image processing and enhancement techniques can then be applied to
obtain the image displayed on the PC monitor. Various enhancement options can be available,
including:
Contrast amplification
Edge contrast
Latitude reduction
Noise reduction
Usually the amount of enhancement within each of the above categories can be controlled by an
integer parameter.
This processing can have a significant effect on the on the perceived image characteristics. To
the eye, the processed image can be made to appear substantially sharper than the one without
processing, but the apparent noise level (granularity) on the processed image is then higher than
that without processing. By varying the processing parameters it is possible to obtain many
different effects, with different trade-offs between noisy/granularity level and apparent
image/edge sharpness.
It is clear that the values selected as optimum for the various image processing parameters are
somewhat subjective, and different operators may well prefer to use different combinations of
values. In addition, the values can be varied depending on the requirements of the inspection
application.
13
Nevertheless, this type of image processing can be an effective additional option, when using
computed (filmless) radiography. It should be noted that this processing is not necessarily
available on all computed radiography systems.
The above spatial image processing operations can be followed by interactive adjustment of
contrast and brightness (sometimes referred to as level control) to obtain the optimum display
of the processed images on the PC monitor. Images in standard file formats (e.g. jpeg) of the
current enhanced image can be created and stored on disc. Note that once in this format, only
very limited further processing or enhancement is possible.
3.5
DIMENSIONAL INDICATORS
For applications requiring accurate dimensional measurements, such as wall thickness

measurement by the tangential radiography technique, it is necessary to calibrate the dimensions
measured from a radiograph. For a film radiograph, this is because a non zero component to
film distance causes the image of the component to be magnified by the ratio (source to film
distance)/(source to component distance). For a digital radiograph, the pixel size needs to be
calibrated in terms of real dimensions (e.g. mm on the component) in order to make accurate onscreen dimensional measurements.
Techniques for calibrating dimensions include use of known pipe or small bore ODs or IDs if
available, but inaccuracies may arise due to possible presence of corrosion. An alternative is the
inclusion of an appropriate calibration object(s) in the radiograph of precisely known
dimension, such as a ball bearing of known diameter, at the same distance from the film/detector
as the object of interest, as illustrated in Figure 3.1.
Figure 3.1 Comparators for calibration of dimensional measurements taken from radiographs
Radiography contractors should produce a written procedure detailing how wall thickness
measurements will be made in a consistent and accurate manner, particularly if this
technique is used for condition monitoring of specific inspection points
14
4 EQUIPMENT SETUP
This section gives recommendations concerning the setup of the radiographic equipment
detailed in Section 3, for the inspection of small bore pipes, and their connection with the main
pipes.
4.1
INSPECTION TECHNIQUES
The junction between small bore piping and a larger diameter pipe can cause physical
restrictions on the placement of the source and film/detector plate. In particular, the presence of
the larger diameter pipe can make it difficult to place the film/detector plate in close proximity
to the small bore connector. Various different techniques can therefore be used to inspect these
components, as follows.
4.1.1
Technique 1: Flat film, beam axis perpendicular to main pipe
In technique 1, the radiation beam axis is arranged perpendicular to the axis of the main pipe,
and the film/detector is kept flat and perpendicular to the radiation beam, as shown in
Figure 4.1.
The simplest and most convenient location for the film/detector is then tangential to the OD of
the main pipe, which gives an appreciable stand-off between the film and the small bore
connector under inspection. This stand-off clearly increases with the diameter of the main pipe
and requires a corresponding increase in the source to film distance to maintain a small
geometric unsharpness as recommended in Section 4.2.
4.1.2
Technique 2: Film flat against connector, beam axis perpendicular to main pipe
To reduce the large source to film distances needed for the larger diameter main pipes,
technique 2 was investigated in the HOIS2000 evaluation trials (Burch & Collett, 2002). This
technique involves the film placed in direct (flat) contact with the small bore connector and
curved around the circumference of the adjacent section of the main pipe, as shown in
Figure 4.1. This allows smaller source to film distances to be used for larger diameter pipes,
while maintaining the recommended maximum geometric unsharpness of 0.3 mm.
Technique 2 is applicable to the detection of corrosion in the section of the small bore connector
external to the main pipe. A source to film distance of c. 300 mm is recommended.
Information can also be obtained on any corrosion within the wall of the main pipe, but as
shown in Figure 4.1, the change of direction of the film as it curves around the circumference of
the main pipe will lead to some distortion of the image.
In the HOIS2000 evaluation trials this technique was applied successfully to specimens
containing 10 inch schedule 40 pipes with a 300 mm SFD. Corrosion in the section of small
bore connector external to the main pipe wall was clearly seen, with significantly reduced
exposure times compared with Technique 1. Some information on the ID of the hole in the main
pipe was also obtained, but the large metal path (c. 100 mm) made this feature difficult to
image, even with increased exposure, in this case. With smaller diameter pipes (e.g. 6 inch or
8 inch) this would be a less significant problem.
However, by its nature Technique 2 requires the detector to be flexible, so that if film is used, it
cannot be housed in a rigid cassette. A possible further problem with the use of curved films, is
pressure marking of the films, particularly on tight bends. Faster films (e.g. Agfa F8) are
understood to be more prone to this than the fine grain films.
15
With computed radiography systems it can be difficult to use this technique as the filmless
plates, although in themselves are flexible, may be deployed within a rigid cassette. Some
computed radiography systems do not however have this restriction.
Technique 1
Technique 2
Film
Source
Technique 4
Technique 3
Figure 4.1 Different techniques for the radiographic inspection of the connections between
small bore piping and larger diameter pipes. Technique 1 - beam axis perpendicular to pipe, flat
film offset from small bore pipe; Technique 2 beam axis perpendicular to pipe, film flat against
the small bore connector and curved around the main pipe; Technique 3 beam axis parallel to
pipe, film flat against the small bore connector and main pipe, and bent at 90 at the joint.
Technique 4 - beam axis perpendicular to pipe, flat film angled to reduce gap behind the
connector
4.1.3
Technique 3: Film flat against connector, beam axis parallel to main pipe
Technique 3 is used to obtain radiographs with the beam axis parallel with the main pipe axis, as
illustrated in Figure 4.1. In this case, inspection of only the section of the small bore connector
external to the main pipe is possible, but small SFDs (e.g. 300 mm) can be used.
The comments about the use of curved films given in Section 4.2.2 are equally applicable to this
technique.
16
4.1.4
Technique 4: Angled Film flat, beam axis perpendicular to main pipe
Technique 4 is a variant of Technique 1. Here the film/imaging plate was kept flat but is angled
as shown to reduce the gap between the film and the connector. The reduction in the film to
object distance (OFD) with this technique depends on the angle used for the film, relative to the
beam direction. As shown in Figure 4.1, an angle of 45 would give an approximate factor 2
reduction in OFD, which would lead to a factor 2 reduction in the required SFD, resulting in an
reduction in exposure time by a factor c. 4.
The image distortion caused by the angled direction of the film was not a significant problem in
the result obtained using this technique in the HOIS2000 evaluation trials. However the use of
dimensional indicators (see Section 3.5) in this technique may lead to inaccuracies because of
the varying object to film distance.
Thus this technique is not recommended if the application requires accurate dimensional
measurements to be taken from the radiographs.
4.2
SOURCE TO FILM DISTANCE
In general, a large source to film distance will minimise the unsharpness in the radiograph
caused by the size of radiation source (known as geometric unsharpness). However, large source
to film distances can lead to very long exposure times, and increased shielding difficulties. Thus
trade-offs must be made, whilst ensuring acceptable image quality.
However, there is no universally accepted method for the choice of the source to film distance
which will provide a satisfactory radiographic technique (Halmshaw, 1995, p125) and existing
standards from various countries differ widely.
For example, BS EN-1435 uses unexplained formulae to determine the minimum source to film
distance, as follows:
For class A ("basic" techniques):
f
b
7.5
d
mm
2/3
For class B ("improved" techniques):
f
b
15
d
mm
2/3
where
f
d
b
is the source to object distance

is the source size
is the object to film/detector distance (which must be expressed in mm).
The basis for the above formulae is not clear, and appears to be empirical at best since the 2/3
power law produces a dimensionally inconsistent equation, and b must be expressed in mm.
17
An alternative approach for determining the minimum source to film distance is to ensure that
the source to film distance is large enough to keep the geometric unsharpness of the radiographs
below a certain value, hence giving radiographs with similar overall spatial resolution.
For inspection of a small bore connector, the distances which should be used in calculating the
geometric unsharpness due to the effective source size, d are shown in Figure 4.2. The
terminology follows that used in BS EN 1435.
The source to film distance (SFD) is the distance from the source position to the film or detector
plane. The distance between the source and the object, f, should be measured from the source to
the point on the connector nearest to the source, to give the most conservative value for the
geometric unsharpness.
For film/detectors which are not aligned normal to radiation beam, the distance to the film
should be taken as the maximum distance from the source side of the component to the position
on the film furthest from the component.
SFD
b
Source, size
d
Film/detector
Figure 4.2 Distances used in calculation of geometric unsharpness in small bore connector
inspection
The corresponding distance from the object to the film b is then simply given by:
b
= SFD f
The geometric unsharpness on the film/detector is then given by

Ug
(d . b)/ f
However, the magnification, M, of the image is given by:
18
= SFD / f
Calculating the geometric unsharpness in the plane of the object then gives:
Ug'
= Ug / M
= (d . b) / SFD
Note that this measure of unsharpness refers to the value appropriate to the object under
inspection, and differs from the value Ug which refers to the value on the film/detector. For
large object magnifications, it is clear that Ug' is the appropriate measure to use for assessing
image quality in relation to object detail, and by extension, the same is true for the smaller
magnifications normally used in site radiography of small bore piping (and other components
for that matter).
For the HOIS2000 evaluation trials on the small bore connectors (Burch and Collett, 2002), the
geometric unsharpness in the plane of the object was c. 0.3 mm or better for most radiographs.
This is considered to be a reasonable target value for the detection of a volumetric defect such
as corrosion.
For a given object to film distance, b, and source size, d, the source to film distance, SFD,
should therefore be set so that Ug' is not more than 0.3 mm, i.e.
SFDmin
= (d . b)/0.3
It should be noted that fine grain film has a lower unsharpness than 0.3mm. For Ir 192
Halmshaw (1995, p 90), quotes a value of 0.17 mm for the unsharpness of fine grain film. The
film unsharpness for Se 75 is less than this due to the lower photon energies involved. Thus
larger SFD values than those given by the equation above would give some slight improvement
in detail resolution.
Table 4.1 shows calculated source to film distances needed to achieve the recommended
geometric unsharpness of 0.3 mm, assuming an effective source size of 2.3 mm. The same
information is shown graphically in Figure 4.3.
Table 4.1 Source to film distances (SFD) recommended for Technique 1, assuming an effective
source size of 2.3mm and a geometric unsharpness of 0.3 mm
Nominal Bore
Outside
diameter, OD
Small bore
piping OD
SFD required for

Technique 1
(mm)
(mm)
(mm)
60.3
88.9
17
17
296
405
(inches)
2
3
19
4
5
6
8
10
12
114.3
141.3
168.3
219.1
273.0
323.8
17
21
21
27
27
27
503
622
725
943
1150
1345
Note that for the 2 inch 3 inch OD main pipes, the SFD values are relatively small (300
400 mm), but for the 10 inch OD pipes and larger, the required SFDs exceed 1000 mm,
requiring substantial exposure times (see Section 4.3) when used with fine grain film.
Source to film distances recommended for Technique 1

1400
Source to film distance (SFD) (mm)
1200
1000
800
600
400
200
0
0
10
Main pipe nominal bore (inch)
Figure 4.3 Recommended source to film distances for application of Technique 1 to the small
bore connections, for an effective source size of 2.3 mm. The pipe nominal bore refers to that of
the main pipe to which the small bore pipe is connected. Smaller sized sources would require
proportionally smaller source to film distance to maintain a geometric unsharpness of 0.3 mm.
For Techniques 2 and 3, a constant SFD of c. 300 mm is recommended to ensure the
geometric unsharpness is within the limit set by the equation above.
For Technique 4, a larger SFD is required, intermediate between the minimum value of
300 mm, and the values shown in Figure 4.3. The exact value should be calculated
according the equation above and will depend on the angle of the film to the small bore
pipe, and hence the maximum distance between the small bore pipe and the film/detector.
Note that for a source size of smaller than 2.3 mm, the above recommended source to film
distances can be reduced in direct proportion to the source size relative to 2.3 mm. Thus a
1.2 mm source would allow the above source to film distances to be reduced by a factor of two.
20
12
4.3
4.3.1
EXPOSURE TIMES AND FILM DENSITY

Fine grain film with lead screens
For conventional radiography using fine grain film, exposures times should be chosen to
obtain a density of c. 3 in the centre of the small bore connection, so that the density of the
tangential wall thickness area is not too low.
Table 4.2 gives a guide to the exposure times required for the inspection of small bore
connectors using Technique 1 (i.e. flat film placed tangentially to the main pipe OD, radiation
beam perpendicular to the main pipe). The times given are for an Ir 192 source with a strength
of 10 Ci, and are based on exposures used for the HOIS2000 evaluation trials. For Se 75
sources, very similar exposure times were needed.
Exposure times for different strength sources can be readily calculated, as exposure time is
inversely proportional to source strength.
For weldolets with larger (10 mm) wall thickness, increased exposure times will be needed by
factors of c. 2 (for Ir 192) and c. 7 (for Se 75 larger factor due to greater attenuation with this
lower photon energy source).
For Techniques 2 & 3 (see Section 4.2), the SFD is maintained at c. 300 mm irrespective of the
nominal bore of the main pipe, so c. 3 min exposure time is required with a 10 Ci Ir 192 source.
Table 4.2 Approximate exposure times for fine grain film for small bore connector inspection
using Technique 1 with a 10Ci Ir192 source
Nominal Bore
SFD required for

Technique 1
Approximate exposure time for a

sch. 40-80 small bore connector
(10 Ci Ir 192 source)
(inches)
2
3
4
5
6
8
10
12
(mm)
296
405
503
622
725
943
1150
1345
(min)
3
6
8
13
18
30
45
60
The approximate exposure times are also shown in Figure 4.4 in graphical form.
21
60
Exposure time (min)
50
40
30
20
10
0
0
10
Main pipe nominal bore (inch)
Figure 4.4 Approximate exposure times for application of Technique 1 to schedule 40/80 small
bore connections, assuming use of fine grain film with lead screens and a 10 Ci Ir 192 source of
size c. 2 x 2 mm.
4.3.2
Other film/detectors for reduced exposure times
The exposure times given in the section above can be reduced using various alternatives to fine
grain film with lead screens, albeit with some reduction in image quality.
Table 4.3 gives a guide to the exposure times needed with different film/screen combinations
and also the newer computed (filmless) radiography systems. The exposure times are given for a
10 Ci source of either Ir 192 or Se 75 and for a source to film distance of 300mm.
Table 4.3 Approximate exposure times for different detection media and source combinations
using Technique 1 with 300 mm SFD (10Ci source in all cases).
Technique
Fine grain film with lead screens & Ir 192

Fine grain film with lead screens & Se 75
Fine grain film with fluorometallic screens &
Ir 192
Computed (filmless) radiography systems &
Se 75
Computed (filmless) radiography systems &
Ir 192
22
Exposure time for

10 Ci source, SFD
= 300
(min)
3
3
0.6
Exposure reduction
factor relative to fine
grain film
0.2 0.8
4 - 15
0.06
50
1
1
5
12
4.4
IMAGE QUALITY INDICATORS
In weld radiography, image quality indicators (IQIs) are used to verify that the sensitivity of the
radiographs obtained during an inspection is within acceptable limits. For small bore connector
radiography, this is most relevant to the detection of corrosion in the straight-through/density
difference method (Section 2.2), but is less critical for the tangential technique which images the
loss of wall directly.
Various different IQI types are in use world-wide, including single wire IQIs (EN 462 Part 1),
step hole IQIs (EN 462 Part 2) and duplex wire IQIs (EN 462 Part 5).
The IQI should generally be placed on the source side of the small bore pipe. For wire IQIs, the
wires should run across the pipe. The IQI needs to be positioned in a section of the pipe unlikely
to be affected by corrosion, otherwise it is possible that the IQI will mask the corrosion.
Table 4.4 is adapted from that shown in EN 1435 for class A radiography, and shows IQI
sensitivities which should be achieved as a function of the wall thickness of the small bore pipe.
Note that the minimum penetrated thickness is twice the values shown in Table 4.4, as the X and
gamma rays penetrate both walls of the piping.
For Ir 192, EN1435 allows lower sensitivities to be accepted for some thickness ranges, as
shown in Table 4.4.
Table 4.4 IQI sensitivities for radiography of small bore piping, adapted from EN 1435
Small bore pipe

wall thickness
(mm)
>2.5 to 3.5
>3.5 to 5.0
>5.0 to 7.5
>7.5 to 12.5
>12.5 to 16
For all radiation sources other than Ir

192
Wire IQI value
Mean %
sensitivity
W14
W13
W12
W11
W10
2.7
2.4
2.0
1.6
1.4
For Ir 192
Wire IQI value
Mean %
sensitivity
W14
W13
W10
W9
W9
2.7
2.3
3.2
2.5
1.8
In the HOIS2000 evaluation trials(Burch and Collett, 2002), the sensitivities of the radiographs
were not assessed using IQIs, as the main aim of the trials was to establish detectability of
different levels of corrosion.
Further work is recommended to quantify the sensitivity of radiographs of small bore
connectors using IQIs, and to compare the values obtained with those given in Table 4.4.
For the computed (filmless) systems, use of Duplex IQIs which measure more directly the
spatial resolution of the image should be investigated.
23
5 OVERALL PERFORMANCE
5.1
IMAGE QUALITY
The different sources and detection media described in this document give differing image
qualities. Table 5.1 gives a qualitative comparison between the results obtained, in terms of
resolution/sharpness, contrast and noise/granularity level. These parameters were assessed using
a simple scoring system, involving a number of stars between 1 and 3, with the best being 3
stars. It should be noted that the range from 1 to 3 stars was qualitative, and chosen to span the
differences between the available results. The number of stars assigned to a particular technique
is not therefore intended to provide a quantitative or linear measure of the parameter concerned
(e.g. image resolution).
Table 5.1 Qualitative comparison between results obtained with different radiographic sources
and detection media on sections of small bore connectors external to the main pipe wall.
Technique
Fine grain film with lead
screens & Ir 192
Fine grain film with lead
screens & Se 75
Fine grain film with
fluorometallic screens & Ir 192
Computed (filmless)
radiography (Se 75) but without
spatial image processing
Computed (filmless)
radiography (Se 75) with
spatial image processing2
Computed (filmless)
radiography (Ir 192) with
spatial image processing2
Resolution/
sharpness +
***
Contrast +
**
Noise/
granularity +
***
***
***
***
**
**
**
**
***
**(*)
**
* to ***
depending on
processing
parameters
* to **
depending on
processing
parameters
Notes:
+
1
2
Qualitative measure only more stars the better (high contrast, high
resolution/sharpness, low noise)
Intrinsic contrast is difficult to quantify for computed radiography systems, due to use
of on-screen contrast/brightness control.
Spatial image processing can improve the apparent sharpness of the image, but often
at the expense of a significant increase in the noise/granularity level.
Note that the highest quality images were the fine grain film radiographs obtained with Se 75.
This is because, for the small bore connectors and weldolets (WT no more than 10 mm), which
have relatively low maximum metal paths, the Se 75 source gave noticeably higher contrast than
Ir 192 due to the lower mean gamma ray energies from Se75.
When fine grain film is used with fluorometallic screens, a significant reduction in exposure
time is obtained (see Section 4.3.2), but the resulting radiographs have appreciably lower spatial
resolution and somewhat reduced contrast.
24
The raw images obtained with two different computed radiography (filmless) systems were
similar in quality, and had lower resolution than fine grain film (but shorter exposure times, as
noted in Section 4.3.2). The spatial resolution of the raw images from the computed radiography
systems was comparable with that obtained with fine grain film and fluorometallic screens.
The spatial image processing package available with one of the computed radiography systems
allowed different trade-offs between noise level and apparent image resolution/edge sharpness
to be obtained, by varying the values of four available processing parameters. Thus, it was
possible to improve the apparent image sharpness but only at the expense of an increase in noise
level.
5.2
EXAMPLES
This Section gives examples of radiography of small bore connector specimens obtained during
the HOIS2000 evaluation trials (Burch & Collett, 2002). These specimens contained simulated
but realistic corrosion defects introduced into one end of the small bore pipes, prior to their
welding onto the main pipes. In each case, a length of about 10-15 mm of simulated corrosion
was introduced using a variety of ad-hoc techniques, which produced an irregular loss of wall
approximately evenly distributed around the circumference of each small bore pipe.
Superimposed on these relatively smooth variations (length scale several mm), were a number
of small isolated pits. At around 10 mm from the end of each pipe, the loss of wall tapered
down to zero.
5.2.1
Comparison between Ir 192 and Se 75 using fine grain film with lead screens
Figure 5.1 shows radiographs obtained using technique 1 (flat film, placed tangentially to main
pipe OD) on 3 inch main pipe specimens, with the small bore connectors containing different
levels of corrosion. The source used was Ir192, and the detection medium was fine grain film
with lead screens. For further details, see the figure caption. Note that inevitably the techniques
used to digitise and display the film radiographs in this report have caused a loss of image
quality, compared with viewing the original film radiographs on a light box.
Note that the wall loss on the moderate and severe cases of corrosion can be detected by the
tangential method. Also, the hole through the main pipe wall can also be clearly seen, and
corrosion here would have been detected, if present.
On the original film radiographs the increased density in the centre of the connector image
caused by wall loss in the straight-through radiography mode can also just be discerned for
both the moderate and severe cases of corrosion, although the presence of the screw thread
complicates matters.
25
(b)
(a)
(c)
Figure 5.1 Film radiographs obtained using Technique 1 with Ir 192 and fine grain film with
lead screens on threaded small bore connectors (OD 17 mm, WT 3.1 mm before introduction of
screw thread) connected to 3 inch pipe (WT 7.6 mm). (a) no corrosion, (b) moderate corrosion
(mean wall loss 24%), (c) severe corrosion (mean wall loss 43%). Results courtesy of
Oceaneering OIS plc.
Figure 5.2 shows results on a similar, but unthreaded set of small bore connectors using
technique 1 applied with a Se 75 source, again with fine grain film and lead screens. The use of
the lower photon energy (softer) Se 75 source has resulted in radiographs with significantly
improved contrast, compared with the corresponding Ir 192 results given in Figure 5.1.
Thus with Se 75, the corrosion can be readily detected by both the density difference and
tangential methods, although only the tangential method allows the wall loss to be measured
quantitatively.
26
(b)
(a)
(c)
Figure 5.2 Film radiographs obtained using Technique 1 with Se 75 and fine grain film with
lead screens on small bore connectors (OD 17 mm, WT 3.1 mm) connected to 3 inch pipe (WT
7.6 mm). (a) no corrosion, (b) moderate corrosion (mean wall loss 28%), (c) severe corrosion
(mean wall loss 52%). Results courtesy of SafeRad Ltd
5.2.2
Example of application to larger diameter (10 inch) main pipe specimens
For larger diameter main pipes, use of technique 1 (flat film, placed tangentially to main pipe
OD see Figure 4.2) can lead to long exposure times (e.g. c. 45 min for 10 inch pipes with a
10 Ci source), when using fine grain film and lead screens (Table 4.2). To reduce exposure
times, technique 2 can be used. This needs a flexible detection medium placed in direct contact
with the small bore connector, and curved around the main pipe surface (see Figure 4.2).
Corresponding exposure times are then only c. 3 min for fine grain film.
27
Results with Ir 192 for technique 2 applied to larger diameter 10 pipe specimens are illustrated
in Figure 5.3. In this case, the small bore connectors are set-through, and the majority of the
corrosion is within the wall of the main pipe. Nevertheless, on Figure 5.3, evidence of wall loss
in the tangential mode can be seen in the form of a general increase in the internal diameter of
the connector, close to the main pipe wall, for both moderate and severe levels of corrosion.
(b)
(a)
(c)
lead screens on plain small bore connectors (OD 17 mm, WT 3.9 mm) connected to
10 pipe (WT 9.3 mm). (a) no corrosion, (b) moderate corrosion (mean wall loss
23%), (c) severe corrosion (mean wall loss 53%). Note most of corrosion is
within wall of main pipe. Results courtesy of Oceaneering OIS plc.
For this larger diameter main pipe, the path length through the main wall of the pipe is large
(maximum 100 mm see Table 2.1) which leads to under-exposure on the radiograph in this
28
region of the image, and there is consequently little information concerning the
presence/absence of corrosion in this region. However, by increasing the exposure time by a
factor two or more, the hole in the wall of the main pipe can be imaged, allowing some
detection of corrosion in this region of the component, albeit at reduced sensitivity compared
with the sections of the connectors external to the main pipe wall.
5.2.3
Example of use of fine grain film with fluorometallic screens
Figure 5.4 shows a radiograph obtained with the faster detection medium of fluorometallic
screens used with the fine grain film, compared with the results obtained with conventional lead
screens. The specimen was that with severe simulated corrosion already shown in Figure
5.1(c).
It can be seen that the use of the fluorometallic screens has resulted in some reduction in image
resolution, but, in this case, this has not affected the detectability of the wall loss using the
tangential method. The exposure time was reduced by a factor 5 using the fluorometallic
screens, as shown in Table 4.3.
(b)
(a)
lead and fluorometallic screens on threaded small bore connectors (OD 17 mm, WT 3.1 mm
before introduction of screw thread) connected to 3 pipe (WT 7.6 mm) with severe corrosion
(mean wall loss 43%). (a) Fine grain film and lead screens, (b) fine grain film and fluorometallic
screens. Results courtesy of Oceaneering OIS plc
29
5.2.4
Examples obtained with computed radiography systems
(a) Computed radiography System 1

Figure 5.5 shows an example of results obtained using Se 75 with a computed (filmless)
radiography system on a 3 inch main pipe specimen with severe corrosion in the attached small
bore connector (for comparison with the corresponding radiographs obtained with Ir 192 and
film see Figure 5.4). This computed radiography system has additional facilities for spatial
image processing (see Section 3.3.1). To show the effects of this processing, Figure 5.5 shows
two images of the same component with and without spatial image processing.
The unprocessed image shows an appreciable loss of spatial resolution compared with the result
obtained with fine grained film used with lead screens (see Figure 5.1(c)) but the image
processing has a substantial effect on the perceived image characteristics. To the eye, the image
processed image appears substantially sharper than the one without processing, but the noise
level (granularity) on the processed image is higher than that without processing.
Thus, the effect of the image processing has been to improve the clarity of image detail (edges
etc), at the expense of an increase in noise level. However, by varying the image processing
parameters it is possible to obtain many different effects, with different trade-offs between
noisy/granularity level and apparent image/edge sharpness.
(b)
(a)
Figure 5.5
Digital images obtained using computed radiography system 1 (technique 1
with Se 75) on a threaded small bore connector (OD 17 mm, WT 3.1 mm before introduction of
screw thread) connected to 3 pipe (WT 7.6 mm) with severe corrosion (mean wall loss 43%).
(a) Computed radiography image without spatial image processing, (b) Computed radiography
image after spatial image processing using one combination of processing parameters selected
to highlight edge detail. Results courtesy of Oceaneering OIS plc.
(b) Computed radiography System 2
30
Figure 5.6 shows examples of results obtained with a second computed radiography system,
using Se 75 and technique 1 on a 3 inch main pipe specimen. A direct comparison can be made
with Figure 5.2 which shows the corresponding results obtained with fine grain film with lead
screen.
(b)
(a)
(c)
Figure 5.6 Digital images obtained using computed radiography system 2 (technique 1 with Se
75) on small bore connectors (OD 17 mm, WT 3.1 mm) connected to 3 pipe (WT 7.6 mm). (a)
no corrosion, (b) moderate corrosion (mean wall loss 28%), (c) severe corrosion (mean wall
loss 52%). For a comparison with the corresponding fine grain results, see Figure 5.2. Results
courtesy of SafeRad Ltd
The results from the computed radiography system clearly show the moderate and severe levels
of corrosion by both the reduced wall thickness in tangential mode and the increased image
density in straight-through radiography mode. However, as with computed radiography system
31
1, some loss of spatial resolution compared with fine grained film used with lead screens is
apparent.
5.3
INSPECTION OF APPROXIMATELY UNIFORM CORROSION/EROSION
Detectability and sizing of approximately uniform corrosion/erosion is best achieved using the
tangential radiography method by direct imaging of the resulting wall loss. Detectability is
affected by the location of the wall loss, as follows.
5.3.1
Wall loss external to the main pipe
If the wall loss is in the section of the small bore connectors external to the main pipe wall, then
the techniques 1-4 described in Section 4.2 are applicable, and detectability will not be affected
by the diameter and wall thickness of the main pipe to which the small bore piping is attached.
The HOIS2000 trials showed that all the source/detector media combinations described in this
document allowed both the moderate (20-30% loss of wall) & severe (40-60%) levels of
corrosion to be readily detected.
The minimum detectable wall loss is therefore appreciably less than 20%, especially for the
highest quality technique such as fine grain film used with lead screens.
5.3.2
Wall loss within the main pipe wall
Inspection of corrosion within the wall of the main pipe is affected by the significantly
increased metal paths that need to be penetrated by the gamma-ray radiation. For metal paths of
around 50 mm, both Ir 192 and Se 75 gave adequate penetration and hence detectability of wall
loss would be similar to the values for external corrosion/erosion given above.
Table 2.1 gives the maximum metal paths involved in the inspection of schedule 40 and 80
pipes of differing diameters. From this, it can be seen that metal paths of 50 mm or less are
obtained for pipes of diameter 3 inch or less.
For maximum metal paths of 100 mm, only marginal detection of the presence of
corrosion/erosion within the wall of the main pipe was obtained with Ir 192 used with fine grain
film and lead screens. No detection was demonstrated at such high metal paths using Se 75 and
either fine grained film with lead screens or with the computed radiography systems.
Thus, for maximum metal paths between 50 and 100 mm, detectability of corrosion/erosion is
predicted to decrease relatively gradually with increasing metal path with Ir 192, but more
rapidly for the less penetrating Se 75. Main pipes with maximum metal paths in this range are
typically between 3 inch and 10 inch in diameter (see Table 2.1).
For larger diameter pipes (>10 inch OD), neither of these isotopes would be capable of adequate
penetration of the main wall. More penetrating isotopes sources such as Co 60 could then be
used, but there are additional safety issues when using such sources.
32
5.4
DETECTABILITY OF ISOLATED CORROSION PITS
Detection, sizing and positioning of isolated corrosion pits as opposed to generalised wall loss
raises additional issues for radiographic inspection.
Isolated corrosion pits around the circumference of the connector are readily detectable with
radiography but cannot be correctly positioned and sized in the through-wall direction unless a
large number of radiographs are taken at different circumferential angles using the tangential
method.
However, the geometry of the small bore connections makes it difficult with flexible film, and
impossible with rigid cassettes, to position the detection medium in such a way as to make the
tangential method possible at all positions around the circumference, apart from the detector
positions described in this document, i.e. techniques 1-4 described in Section 4.2.
Thus, detection of an isolated corrosion pit at an arbitrary circumferential location would
generally be best achieved by the density difference produced on the radiograph in straightthough mode (see Figure 3.2).
To find the shallowest pits, use of Se75 instead of Ir 192, would then be preferred to obtain
higher contrast radiographs, provided the pitting is in sections of the small bore connector
external to the main pipe. To measure the through-wall depth of a corrosion pit, a tangential
radiographic image of the pit would still be needed.
Detectability of small corrosion pits using radiography was not assessed in the HOIS2000 small
bore connector trials, nor is there any published probability of detection (POD) information
relevant to small bore connector inspection.
However, some information on the smallest size of detectable corrosion pit can be estimated
from published IQI information. The step/hole type of IQI indicator consists of a cylindrical
hole having a height equal to its diameter, which is a reasonable approximation for the shape of
a small corrosion pit. However, the shape of the pit will affect its detectability. Very narrow
"worm-hole" type pits will be more difficult to detect than wider pits having the same depth.
For Ir 192 and the lower energy Yb-169, Halmshaw (1995, p 185) gives values for attainable
IQI sensitivity values in flat plate steel using fine grain film, which are shown here in Table 5.2.
Table 5.2 Attainable IQI sensitivity values in flat plate steel for Ir 192 and Yb 169 gamma ray
sources and fine grain film with thin lead screens, taken from Halmshaw (1995).
Steel thickness
6
6
12
12
25
40
50
Source
(%)
5.0
6.6
3.3
4.1
2.6
2.0
2.0
Yb169
Ir 192
Yb169
Ir 192
Ir 192
Ir 192
Ir 192
Step/Hole IQI
Hole diameter (mm)
0.32
0.4
0.4
0.5
0.63
0.8
1.0
For detection of corrosion pits in piping, the minimum steel thickness in the straight-through
technique is twice the wall thickness, so to convert the figures for percentage sensitivity into an
33
equivalent loss of (single) wall due to a corrosion pit would require multiplication by a factor of
at least two.
Thus, for example for a small bore pipe with a 3 mm wall thickness using Ir 192, the Step/Hole
IQI sensitivity is 6.6%, from which it can be inferred that a corrosion pit producing a 13.2% loss
of wall may just be detectable using fine grain film with thin lead screens. This assumes that the
corrosion pit is ideally placed in the centre of the pipe image, on the radiation beam axis. For
corrosions pits off-centre the metal path will be higher, and sensitivity will be reduced.
The sensitivity achievable with Se 75 should be somewhat higher than that for Ir 192, as Se 75
generates radiation with a mean gamma-ray energy approximately mid-way between Ir 192 and
Yb 169. Thus the corresponding IQI sensitivity for 6 mm of steel is likely to be c. 5.5%, which
converts to a 11% minimum detectable loss of wall for an idealised corrosion pit in the centre of
the pipe image.
For pipes with larger wall thicknesses, note that the sensitivity when expressed in percentage
terms improves. Thus for a pipe with 6 mm wall thickness, the corresponding minimum
detectable wall loss for an isolated corrosion pit would be 8.8%, with the same assumptions as
detailed above.
In practice, however, detection of corrosion pits in pipes will be more difficult than detection of
a step/hole IQI in a known location within a uniform plate, for the following main reasons:
(a) With an IQI device, the inspector knows exactly where to look for the object of interest.
Isolated corrosion pits on the other hand could occur in any location within a large region of
pipe material.
(b) With an IQI device on a flat plate, the density of the radiograph is almost constant over the
region of interest. With pipe inspection, the density of the radiograph changes appreciably
with distance from the centre of the image of the pipe.
(c) Corrosion pits can be irregular and less well defined than a machined cylindrical hole
In view of the above factors, more realistic conservative estimates for the minimum detectable
depth of a corrosion pit would be 15 20% loss of wall for pipes with wall thicknesses in the
range 3 - 6 mm, provided the pit is approximately centrally placed in the pipe image (not too
close to the tangential view).
Further investigation of this issue is recommended.
34
6 SUMMARY OF RELEVANT SAFETY STANDARDS AND

LEGISLATION
The requirements of all relevant national and international safety standards relating to the
use of ionising radiation must be complied with. In the UK, the Ionising Radiations
Regulations 1999 (IRR 1999) are enforced by the Health & Safety Executive (HSE). These
are legal requirements and include the construction and maintenance of radiation controlled
areas, by means of appropriate barriers. Pre-planning of the inspection work to be carried
out on a plant, to include both a risk assessment and a practical assessment of how the
source container and shielding will be placed is also covered by this legislation.
The safe transport of radioactive material is covered by IAEA regulations, the latest version
of which is TS-R-1 (ST-1, Revised). Individual countries have their own legislation
covering this area.
Gamma ray sources and their containers are covered by the standards ISO 2919: 1999 and
ISO 3999-1:2000, respectively.
The standard BS 5650:1978, ISO 3999-1977 concerning specification for apparatus for
gamma radiography is also still current.
7 MAIN RECOMMENDATIONS
The Sections presents a summary of the main recommendations contained in this document.
Radiography methods
The tangential radiography method is recommended for the detection and through-wall
sizing of extended areas of corrosion, and should not generally be used for isolated
corrosion pits, unless their circumferential location has already been established using for
example the density difference method.
The straight-through or density difference radiographic method is recommended for the

detection (but not through-wall sizing) of isolated corrosion pits, and can also be used for
detection of extended areas of corrosion.
Both the tangential and straight-through methods can be combined on a single radiograph of
a small bore pipe.
Isotope sources
For the inspection of the sections of the small bore connectors external to the main pipe, the
two recommended isotope sources are Iridium 192 and Selenium 75.
Source selection should be made according to the criteria given in Section 3.1.1.
The source size for inspection of small bore piping should be chosen taking account of the
limits on geometric unsharpness given in Section 4.2. Typically source sizes for this
application will be about 2 x 2mm (effective size for calculation of geometric unsharpness
2.3 mm).
35
Careful collimation of the sources is recommended to minimise unwanted radiation, and

reduce the effects of scatter on the radiographs.
Wall thickness measurements
Radiography contractors should produce a written procedure detailing how wall thickness
measurements will be made in a consistent and accurate manner, particularly if this
technique is used for condition monitoring of specific inspection points
Source to film distances
The source to film distance, SFD, should be set so that the geometric unsharpness in the
plane of the object is not more than 0.3 mm. A minimum SFD of 300 mm should be used.
Exposure time reduction
For conventional radiography using fine grain film, exposures times should be chosen to
obtain a density of c. 3 in the centre of the small bore connection, so that the density of the
tangential wall thickness area is not too low.
Recommended measures to reduce exposure time include reduction of the object to film
distance, and hence SFD, by placing the detection medium in direct contact with the small
bore connector (a flexible detection medium is needed if the junction with main pipe is to be
inspected).
The highest quality radiographic images are obtained with fine grain film used with thin
lead screens. Alternative detection media can be used to reduce exposure time, including (a)
fine grain film with fluorometallic screens and (b) the reusable phosphor plates in computed
(filmless) radiography systems, provided the reduction in image quality is acceptable for the
application.
It is important to note that the above recommendations are based on trials which were completed
under laboratory conditions inside an exposure compound or similar controlled and uncluttered
environment. This made it straightforward to set-up the radiography equipment in such a way
as to achieve safe radiation dose rates in a small controlled area, whilst using highly penetrating
radioisotopes and the source to film distances (SFD) recommended in this document. The
practical application of these recommendations under typical site conditions may need
modification for particular circumstances. In-situ, local access to small bore connectors is often
restricted, limiting the choices for the source position and shielding. The appropriate legislation
requires the radiography contractor to conduct a prior risk assessment of the work and it is at
this stage that a practical assessment should highlight potential difficulties so that solutions can
be planned in advance.
36
8 RECOMMENDATIONS FOR FURTHER WORK

Further experimental work is recommended to provide further information which could be
incorporated into later versions of this recommended practice:
Quantify the sensitivity of radiographs of small bore piping using IQIs to compare the
values obtained with those given in EN 1435 for weld radiography. For the computed
(filmless) systems, use of Duplex Wire IQIs which measure more directly the spatial
resolution of the image should be investigated.
Assess the detectability of isolated corrosion pit type defects in small bore piping
specimens, as a function of defect size/depth and circumferential location in the pipe in
relation to the radiation beam axis direction.
9 ACKNOWLEDGEMENTS
This document has benefited from comments made on earlier drafts by Jim McNab of
Oceaneering OIS plc, Malcolm Wass of SafeRad Ltd and those HOIS2000 members who
participated in the Complex Geometry Working Group meeting in September 2002.
Jim McNab and Malcolm Wass are also thanked for carrying out the trials which resulted in the
example images. The assistance of BP Coryton with specimen manufacture is also gratefully
acknowledged. Arild Dalheim, of Norsk Hydro, is thanked for supplying a number of weldolets.
For the duration of this work, Raman Patel (HSE) acted as Project Champion and provided
valuable advice and support throughout, as well as commenting on an earlier draft of this report.
10 REFERENCES
Burch, S F and Collett, N J (2002) Evaluation trials of radiographic techniques for the rapid
inspection of small bore connector components, a report produced for HOIS2000,
HOIS2000(02)R1.
BS 5650:1978, ISO 3999-1977 Specification for apparatus for gamma radiography
BS EN 1435 (1997) Non-destructive examination of welds - Radiographic examination of
welded joints, European and British Standard.
Halmshaw, R (1995) Industrial Radiology Theory and Practice Second Edition, Chapman &
Hall
International Atomic Energy Agency. Regulations for the Safe Transport of Radioactive
Material, 1996 Edition (Revised), Safety Standards Series No. TS-R-1 (ST-1, Rev), IAEA,
Vienna (2000).
ISO2919 : 1999 Radiation protection -- Sealed radioactive sources -- General requirements and
classification
37
ISO3999-1: 2000 Radiation protection - apparatus for industrial gamma radiography - Part 1:
Specifications for performance, design and tests.
The Ionising Radiation Regulations 1999, Statutory Instruments 1999 No 3232, ISBN 0-11085614-7, The Stationery Office, P.O. Box 276, London SW8 5DT (referred to in this
document as IRR 1999)
38
Printed and published by the Health and Safety Executive

C30 1/98
Printed and published by the Health and Safety Executive
C0.06
01/05
ISBN 0-7176-2939-2
RR 294
10.00
9 78071 7 629398
Recommended practice for the rapid inspection of small bore connectors using radiography
HSE BOOKS

XXX Radiographs

Uploaded by

Copyright:

Available Formats

XXX Radiographs

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

XXX Radiographs

Uploaded by

Copyright:

Available Formats

HSE

Health & Safety

Recommended practice for the rapid inspection

Prepared by AEA Technology plc for the

RESEARCH REPORT 294

Health & Safety

Recommended practice for the rapid inspection

S F Burch (BA, Ph D) and N J Collett

Crown copyright 2005

significantly affect the performance of the radiographic techniques evaluated. Nevertheless

2 RADIOGRAPHIC METHODS FOR CORROSION DETECTION

Extended area of corrosion

Figure 2.1 Principle of tangential radiography

Image of corrosion pit on

Figure 2.3 Detection of corrosion by increased film density in straight-through mode

BOTH METHODS COMBINED

Small bore pipe

INSPECTION OF JUNCTION WITH MAIN PIPE

Tangential path lengths

Tangential path (mm)

Pipe nominal bore (inch)

Size and strength of sources

Source containers and collimation

IN-SITU INSPECTION OF PLANT

FILM AND SCREENS

Disadvantages of film include:

Relatively long exposure times

COMPUTED (FILMLESS) RADIOGRAPHY

The advantages of the computed radiography (filmless) systems include:

The disadvantages of this technology include:

Relatively new recording medium, characteristics not fully understood or documented.

Image Processing on filmless systems

For applications requiring accurate dimensional measurements, such as wall thickness

Technique 1: Flat film, beam axis perpendicular to main pipe

Technique 4: Angled Film flat, beam axis perpendicular to main pipe

SOURCE TO FILM DISTANCE

For class B ("improved" techniques):

is the source to object distance

The geometric unsharpness on the film/detector is then given by

However, the magnification, M, of the image is given by:

SFD required for

Source to film distances recommended for Technique 1

Source to film distance (SFD) (mm)

Main pipe nominal bore (inch)

EXPOSURE TIMES AND FILM DENSITY

SFD required for

Approximate exposure time for a

Exposure time (min)

Main pipe nominal bore (inch)

Other film/detectors for reduced exposure times

Fine grain film with lead screens & Ir 192

Exposure time for

IMAGE QUALITY INDICATORS

Small bore pipe

For all radiation sources other than Ir

Example of application to larger diameter (10 inch) main pipe specimens

Example of use of fine grain film with fluorometallic screens

Examples obtained with computed radiography systems

(a) Computed radiography System 1