This part of EN 16407 covers the tangential inspection technique for detection and through-wall sizing of wall loss, including: a with the source on the pipe centre line, and b with the
Protection against ionising radiation
Exposure to X-rays or gamma-rays poses significant health risks It is essential to adhere to legal requirements when using X-ray equipment or radioactive sources Strict compliance with local, national, and international safety precautions is necessary when handling ionizing radiation.
Personnel qualification
Testing must be conducted by skilled and qualified personnel, supervised by competent individuals designated by the employer or the inspection company It is advisable for personnel to be certified under EN ISO 9712 or a similar recognized system to demonstrate their qualifications The employer should issue operating authorization for qualified individuals following a documented procedure Non-destructive testing (NDT) operations require authorization from a qualified NDT supervisor, unless otherwise specified.
3 or equivalent) approved by the employer
The personnel shall prove additional training and qualification in digital industrial radiology if digital detectors are being used.
Identification of radiographs
Each section of the object being radiographed must have symbols attached, ensuring that these symbols are visible in the radiograph outside the region of interest whenever possible This practice guarantees clear and unambiguous identification of each section.
Marking
Permanent markings should be made on the object to be examined in order to accurately locate the position of each radiograph
Where the nature of the material and/or its service conditions do not permit permanent marking, the location may be recorded by means of accurate sketches.
Overlap of films or digital images
When conducting radiography with multiple films or detectors, it is essential that they overlap adequately to cover the entire area of interest This overlap must be confirmed by a high-density marker placed on the object's surface, which should be visible on every film or detector In cases where radiographs are taken sequentially, the high-density marker must also be clearly visible on each radiograph.
Types and positions of image quality indicators (IQI)
Single wire or step hole IQIs
For tangential radiography, single wire or step hole IQIs are not applicable.
Duplex wire IQI (digital radiographs)
According to EN ISO 19232-5, IQIs should be utilized to measure the basic spatial resolution of the CR/DDA system in a reference radiograph The duplex wire IQI must be positioned next to the imaging plate or detector array, tilted at an angle of 2° to 5° relative to the digital rows or columns of the digital image.
6 Recommended techniques for making radiographs
Test arrangements
General
Normally radiographic techniques in accordance with 6.1.2 and 6.1.3 shall be used For both techniques, the film or digital detector shall be placed as close to the pipe as possible.
Radiation source located on the pipe centre line
In this setup, the source is positioned in front of the pipe, while the film or detector is located on the opposite side, as illustrated in Figure 1 The pipe can either be non-insulated (Figure 1 a) or insulated (Figure 1 b).
Figure 1 — Test arrangement and distances for tangential radiography with the source on the pipe centre line
Note that the wall loss can be located on either the inner diameter, outer diameter or both surfaces of the pipe.
Radiation source located offset from the pipe centre line
In this setup, the radiation source is positioned in front of the pipe, while the film or detector is placed on the opposite side, as illustrated in Figure 2 a) for the non-insulated pipe and Figure 2 b) for the insulated pipe.
Figure 2 — Test arrangement and distances for tangential radiography with the source offset from the pipe centre line
In this test setup, the source is positioned off the pipe's center line and aligned with the center of the pipe wall, as illustrated in Figure 2 It is important to note that wall loss may occur on the inner diameter, outer diameter, or both surfaces of the pipe.
Alignment of beam and film/detector
The beam of radiation shall be directed at the centre of the area being examined
The film or detector should be aligned to be orthogonal to the centre of the radiation beam
Modifications to these alignments and the test arrangements given in 6.1.2 and 6.1.3 may be needed in special cases, due for example to the presence of obstructions
Other ways of radiographing may be agreed between contracting parties.
Choice of radiation source
In tangential radiography, the selection of the radiation source is based on the maximum penetrated thickness of the pipe, denoted as \$w_{max}\$, which is achieved along the path that forms a tangent to the inner diameter of the pipe, as illustrated in Figure 3.
Figure 3 — Maximum penetrated thickness, w max, for the tangential technique
The maximum penetrated thickness, wmax, is given by: max =2 ( e− ) w t D t (1)
Table 1 gives recommended limits on the maximum penetrated thickness for different radiation sources
Some forms of insulation (e.g highly absorbing) may lead to reduction in the limits on maximum penetrated thickness, wmax, given in Table 1
The contracting parties may adjust these values, as long as the inner diameter edge's position can be accurately measured on the resulting radiograph or digital image, following the methods outlined in sections 7.6 or 7.7.
Table 1 — Maximum penetrated thickness range for different radiation sources for steel
Radiation source Limits on maximum penetrated thickness w max mm Basic
(for generalized wall loss) Improved
For digital radiographs, somewhat higher values for the limits on maximum penetrated thickness than those given in Table 1 may be used
To identify the suitable source(s) for a specific pipe, it is essential to calculate the maximum penetrated thickness, denoted as \$w_{max}\$, using Formula (1) and then compare the results with the values listed in Table 1 For a visual representation of this process, refer to Annex B.
In cases where radiographs are produced using gamma rays, the total travel-time to position and rewind the source shall not exceed 10 % of the total exposure time.
Film systems and metal screens
For radiographic examination, film system classes shall be used in accordance with EN ISO 11699-1
The radiographic film system class and metal screens to use with films for different radiation sources are given in Tables 2 and 3 See also EN ISO 17636-1:2013, Tables 2 and 3
When using metal screens, good contact between films and screens is required This may be achieved either by using vacuum-packed films or by applying pressure
Table 2 — Film system classes and metal screens for tangential radiography of steel, copper and nickel based alloy pipes
Radiation source Film system class a Type and thickness of metal screens
X-ray potentials ≤ 250 kV C 5 C 4 0,02 mm to 0,15 mm front and back screens of lead
X-ray potentials > 250 kV to 500 kV C 5 C 4
Front screens of lead range from 0.1 mm to 0.2 mm, while back screens of lead vary from 0.02 mm to 0.2 mm X-ray potentials exceed 500 kV and can reach up to 1,000 kV Additionally, front and back screens made of steel or copper measure between 0.25 mm and 0.7 mm.
Ir 192 C 6 C 5 0,02 mm to 0,2 mm front and back screens of lead b
Co 60 C 6 C 5 0,25 mm to 0,7 mm front and back screens of steel or copper c
X-ray equipment with energy from
1 MeV to 4 MeV C 6 C 5 0,25 mm to 0,7 mm front and back screens of steel or copper c
X-ray equipment with energy above
Up to 1 mm front screen of copper, steel or tantalum d Back screen of copper or steel up to
For optimal film systems, tantalum films can be utilized up to 0.5 mm, while ready-packed films with a front screen thickness of 0.03 mm are applicable if an additional lead screen of 0.1 mm is positioned between the object and the film In class TA, lead screens ranging from 0.5 mm to 2.0 mm are permissible, and screens between 0.5 mm and 1 mm may be used by mutual agreement between the contracting parties.
Table 3 — Film system classes and metal screens for tangential radiography of aluminium and titanium pipes
Radiation source Film system class a Type and thickness of intensifying screens
None or up to 0,03 mm front and up to 0,15 mm back screens of lead
0,02 mm to 0,2 mm front and back screens of lead b
For optimal performance, a film system can utilize front and back screens ranging from 0.02 mm to 0.2 mm in thickness Alternatively, instead of a single 0.2 mm lead screen, two 0.1 mm lead screens can be employed.
Different film systems may be used by agreement of the contracting parties, provided the required optical densities defined in 7.2 are achieved.
Screens and shielding for imaging plates (computed radiography only)
For optimal performance with metal front screens, it is essential to ensure proper contact between the sensitive detector layer and the screens This can be accomplished through vacuum-packed imaging plates (IPs) or by applying pressure Insufficient contact between lead screens and IPs can lead to image unsharpness Additionally, the intensification achieved with lead screens in contact with imaging plates is considerably less effective compared to traditional film radiography.
Low energy backscatter and X-ray fluorescence from lead can significantly affect image quality by causing edge unsharpness and reducing signal-to-noise ratio (SNR) in imaging plates (IPs) To minimize these effects, it is advisable to use steel or copper shielding directly behind the IPs Additionally, placing steel or copper shielding between a backscatter lead plate and the IP can further enhance image quality Modern cassette and detector designs may address these issues, potentially eliminating the need for extra steel or copper shielding outside the cassette.
The protective layer between the lead and the sensitive layer of an IP significantly diminishes the intensification effect of electrons, which becomes noticeable only at higher energies The degree of intensification varies based on the radiation energy and the design of the protective layer, ranging from 20% to 100% compared to the absence of a screen.
The minor intensification effect caused by a lead screen in contact with an imaging plate (IP) can be balanced by extending the exposure time or milliampere-minutes when lead screens are not utilized However, since lead screens can potentially scratch IPs if not properly separated during scanning, they should be employed for intermediate filtering of scattered radiation outside of cassettes.
Tables 4 and 5 outline the suggested screen materials and thicknesses for various radiation sources Alternative screen thicknesses can be negotiated between the contracting parties It is advisable to use metal screens in front of ionization chambers (IPs), as they can also mitigate the effects of scattered radiation when paired with digital dosimetry systems (DDAs).
Table 4 —Metal front screens for CR for tangential radiography of steels, copper and nickel based alloys Radiation source Type and thickness of metal front screens mm
X-ray potentials b ≤ 250 kV 0 to 0,1 (lead)
Ir 192, Se 75 b Class TA: 0 to 0,3 (lead)
Class TB: 0,3 to 0,8 (steel or copper)
Co 60 a 0,3 to 0,8 (steel or copper) + 0,6 to 2,0 (lead)
X-ray potentials exceeding 1 MV require specific shielding configurations, with steel or copper screens ranging from 0.3 to 0.8 mm and lead screens from 0.6 to 2.0 mm In setups using multiple screens, the steel screen should be positioned between the image receptor (IP) and the lead screen Alternatively, copper, tantalum, or tungsten screens can be utilized if they demonstrate adequate image quality Additionally, lead screens can be fully or partially substituted with iron or copper screens, with the equivalent thickness for iron or copper being three times that of lead.
Table 5 — Metal front screens for CR for the digital tangential radiography of aluminium and titanium
Radiation source Type and thickness of metal front screens mm X-ray potentials < 500 kV ≤ 0,2 (lead) a , b
For radiation protection, if the lead thickness is 0.2 mm, a 0.1 mm screen with an additional 0.1 mm filter can be utilized outside the cassette Additionally, lead screens can be fully or partially substituted with iron (Fe) or copper (Cu) screens, with the equivalent thickness for Fe or Cu being three times that of the lead thickness.
Reduction of scattered radiation
Filters and collimators
In order to reduce the effect of back scattered radiation, direct radiation shall be collimated as much as possible to the section under examination
Alternative materials like tin, copper, tungsten, or steel can serve as effective filters For lead filters, it is advisable to include an additional steel or copper filter with a thickness ranging from 0.3 mm to 1.0 mm, positioned as close as possible to the sensitive plate and the detector.
Interception of back scattered radiation
Each new test arrangement must be verified for back scattered radiation using a lead letter "B," which should have a minimum height of 10 mm and a minimum thickness of 1.5 mm, positioned directly behind each film.
When using a CR cassette or DDA, a radiograph that displays a lighter image of the symbol (negative presentation) should be rejected Conversely, if the symbol appears darker or is not visible, the radiograph is deemed acceptable, indicating effective protection against scattered radiation.
In digital radiography, it is essential to shield the detector from back scattered radiation using a minimum of 1 mm of lead or 1.5 mm of tin placed behind the detector Depending on the configuration, up to 6 mm of lead may be required Additionally, a layer of steel or copper, approximately 0.5 mm thick, should be installed between the lead shield and the detector to mitigate the effects of lead X-ray fluorescence radiation It is important to note that lead screens should not be in direct contact with the back side of the detector for energies above 80 keV.
Source-to-detector distance
For tangential radiography, the source-to-detector distance, SDD, and the pipe centre to detector distance, PDD, are shown in Figures 1 and 2
The minimum source-to-detector distance, SDD, depends on the source size, d, the pipe outer diameter, De and on the pipe centre to detector distance, PDD
For tangential radiography with the source on the pipe centre line (as shown in Figure 1a) and Figure 1b)), the distances SDD shall be, where practicable, in accordance with the following
For the basic technique, TA, SDD should be the larger of the following two values (in mm):
For the improved technique, TB, SDD should be the larger of the value given by Formula (2) and:
In tangential radiography where the source is offset from the pipe's center line, as illustrated in Figures 2a) and 2b), Formula (2) is not applicable for calculating the minimum Source-to-Detector Distance (SDD) Consequently, Formula (3) provides the SDD for the basic technique TA, while Formula (4) specifies the SDD for the enhanced technique TB.
The geometric unsharpness values calculated from Formulae (3) and (4) are 0.6 mm and 0.3 mm, respectively, projected onto the plane at the pipe center, where measurements are taken using the tangential technique However, the unsharpness values observed at the detector are greater than these due to the influence of projective magnification.
Axial coverage and overlap
The maximum axial coverage of a pipe in a single image or film is determined by a 20% increase in the penetrated thickness at the edge of the inspection area, as shown in Figure 4.
Figure 4 — Axial cross section showing the maximum permissible axial length of the evaluated area for a single source position, on the detector, L d, and along the pipe, L p, at the tangent position
The total axial extent of the evaluated area on the detector, Ld, should be no greater than
The total axial extent of the evaluated area on the pipe, Lp, should be no greater than
The Lp formula is essential for calculating the exposure intervals along a pipe When the collimator angles of gamma sources or the window collimation of X-ray sources are less than ± 35°, it is necessary to adjust Lp and Ld to match the maximum opening angle of the radiation cone beam Additionally, separate films or digital images must overlap adequately to guarantee that every part of the component is examined, with a minimum axial overlap of 25 mm on each side unless stated otherwise.
Dimensional comparators
To accurately measure the remaining wall thickness, it is essential to dimensionally calibrate film radiographs or digital images This calibration corrects for the geometric magnification, or "blow-up," that occurs due to the arrangement of the source, pipe, and detector.
A common technique for dimensional calibration involves using a ball bearing or similar dimensional comparator This spherical object, which is effectively radiation opaque and has a known diameter, is positioned near the pipe and aligned with the tangent point on the pipe wall, as shown in Figure 5.
Note that other dimensional calibration methods (see 7.5) do not require the use of these additional comparators
Figure 5 — Tangential radiography showing use of comparators for dimensional calibration (second comparator is optional)
Comparator(s) shall be placed in the tangent position, as close to the pipe wall as possible, without overlapping it
Measurements of the imaged size of the comparator then allow the pipe wall thickness measurement to be calibrated (see 7.5)
If the comparator cannot be positioned next to the pipe tangent due to external insulation, it is advisable to offset the source from the pipe centerline to align it with the pipe wall, as illustrated in Figure 6.
Tangential radiography, as illustrated in Figure 6, demonstrates the application of an offset source position alongside a comparator for dimensional calibration of insulated pipes It is essential to position the comparator as close to the outer edge of the insulation as feasible to ensure accurate measurements.
Dimensional comparators should generally not be wrapped in lead or similar materials However, when dealing with insulated pipes and saturated free beams, it is necessary to wrap the comparator in lead to prevent inaccurate calibration caused by the burn-off of its edges.
Image saturation and use of lead strips to avoid burn-off
In digital radiography (CR or DDA), it is essential to avoid using lead strips near the edge of the pipe to prevent image burn-off at the pipe's outer diameter To minimize burn-off, the exposure time should be carefully adjusted in accordance with section 7.1.1.
To achieve optimal imaging, it is essential to minimize the intensity difference between the free beam and the radiograph of the pipe This can be accomplished by utilizing prefilters near the source or intermediate filters positioned between the pipe and the imaging plate.
For insulated pipes, it is acceptable for the exposed beam beyond the insulation to be saturated, as long as the grey levels near the pipe wall do not exceed 90% of the digital system's dynamic range If dimensional comparators are utilized in this scenario, they must be wrapped in lead or a similar material to prevent image saturation around the comparator.
Selection of digital radiographic equipment
General
The spatial resolution of the detector, when divided by magnification (M = SDD/SPD), must not exceed 200 µm for class TA and 130 µm for class TB, and should remain within 5% of the nominal wall thickness (t) Alternative values may be negotiated between the contracting parties.
CR systems
The pixel size of the dimensionally calibrated CR scanner image must not exceed 100 µm While increasing the scanner gain or sensitivity can enhance grey levels for a specific radiographic exposure, it has little impact on image quality as indicated by the normalized signal to noise ratio (SNR N) To effectively improve SNR N, it is essential to increase the exposure rather than the scanner gain.
Using low gain or sensitivity in scanners with linear responses between radiation dose and grey level minimizes the risk of image saturation Conversely, higher scanner gains can lead to saturation, particularly in free beam areas during short exposures, which may not provide adequate image signal-to-noise ratio (SNR) values to satisfy the image quality standards outlined in section 7.1.
DDA systems
The pixel size of the dimensionally calibrated image (see 6.8) shall not exceed 200 àm for class TA and 130 àm for class TB Different values can be agreed by contracting parties
7 Radiograph/digital image sensitivity, quality and evaluation
Evaluation of image quality
General
In tangential radiography, conventional wire or step/hole Image Quality Indicators (IQIs) are unsuitable due to their inability to be placed close to the tangential pipe position Additionally, the rapid variations in penetrated thickness in this area of a radiographic image hinder the effective assessment of IQI visibility.
Evaluation of image quality is however required using the following methods to ensure reproducible results.
Maximum grey level in free beam (digital radiographs)
To ensure optimal imaging quality, it is essential to adjust the exposure time and system sensitivity so that the unimpeded radiation beam outside the pipe wall remains below 90% of the imaging system's saturation level.
Minimum normalized signal to noise ratio (digital radiographs)
Digital radiographic images can exhibit increased noise when captured under less-than-ideal conditions, such as during short exposure times This excessive noise can hinder the accuracy of measurements, posing a significant challenge in obtaining reliable results.
To maintain acceptable noise levels in digital images from CR and DDA systems, it is essential to measure the normalized signal-to-noise ratio (SNR N) using suitable software and methods.
EN 14784-1 using an area of at least 55 (vertically) x 20 (horizontally) pixels SNR N values shall be measured at a minimum of four separate positions, and the average value taken
To calculate the normalized SNR N value, the basic spatial resolution (SR b) of the imaging system must be assessed using the duplex wire IQI method outlined in EN ISO 19232-5 or a comparable technique This involves positioning the duplex wire next to the imaging plate or detector array for accurate measurement of the detector's basic spatial resolution.
In an unsaturated free beam area of the image outside the pipe, the average SNR N values must reach a minimum of 70 for the basic class and at least 100 for the improved class.
When measuring the pipe center line using combined tangential/double wall double image radiography, minimum SNR N values of 50 for standard quality and 80 for higher quality can serve as alternatives to the free beam SNR N values.
When measuring the Signal-to-Noise Ratio (SNR), it is crucial to ensure that the image's grey levels are directly proportional to the radiation intensity; otherwise, the resulting values will be inaccurate.
Density of film radiographs
In tangential film radiography, it is essential to ensure that the exposure conditions are optimized to achieve the minimum optical densities specified for the film system classes outlined in Table 2.
— optical density on the pipe centre line ≥ 1,5;
— optical density in the un-impeded beam (outside the pipe): 3,5 to 4 (max);
— optical density in a tangent position of the inner pipe wall ≥ 0,5
For optical densities greater than 4 in the un-impeded beam, one emulsion side must be removed at the outer tangential position to measure the outer wall position Conversely, if the optical density is 4 or less in the remaining emulsion, the radiograph is acceptable for wall thickness measurement.
A measuring uncertainty of ± 0,1 is permitted
To prevent excessive fog densities due to film aging, development, or temperature variations, it is essential to periodically check the fog density on a non-exposed sample of the films in use This sample should be handled and processed under the same conditions as the actual radiograph The fog density must not exceed 0.3, which is defined as the total density (emulsion and base) of a processed, unexposed film.
When employing a multi-film technique alongside the interpretation of individual films, it is essential that the optical density of each film adheres to the specified standards In cases where double film viewing is required, the optical density of at least one of the single films must not fall below 1.3.
Film processing
Films must be processed according to the guidelines set by manufacturers of both the film and chemicals to achieve the desired film system class Key factors such as temperature, developing time, and washing time require careful monitoring Regular control of the film processing is essential to ensure optimal results.
EN ISO 11699-2 The radiographs should be free from defects due to processing or other causes which would interfere with interpretation.
Film viewing conditions
Dimensional comparator
If the pipe outside diameter is not known accurately or reliably, then the alternative dimensional calibrator method, described in 6.8, shall be used
This method utilizes the measured dimension of the comparator, denoted as \$c'\$, to calibrate distances, allowing for the determination of the actual remaining wall thickness, \$t_{act}\$ This value is calculated from the measured remaining wall thickness, \$t_{meas}\$, by applying the ratio of the defined comparator dimension, \$c\$, to the measured dimension, \$c'\$: \$$t_{act} = t_{meas} \cdot \frac{c}{c'}\$$
Wall thickness measurements for film radiographs
Dimensional measurements from film radiographs can be accurately obtained using callipers to assess both the pipe wall thickness and a reference object with a known dimension, such as a ball-bearing comparator or the known outside diameter of a pipe.
Wall thickness measurements for digital radiographs
Interactive on-screen measurements
CR/DDA systems feature software that enables users to perform interactive dimensional measurements directly on digital images using a cursor, independent of the underlying grey level values Users visually assess the positions of the inner and outer edges of the pipe wall within the image.
The method may incur significant errors due to the dependence of apparent wall thickness on contrast and brightness settings in image display These errors are particularly pronounced for pipes with maximum penetrated thickness values, \( w_{max} \), nearing the recommended limits for the radiation source (refer to Table 1) To mitigate these errors, two techniques can be employed: a) High-frequency spatial filtering enhances the visibility of pipe wall edges while minimizing reliance on contrast and brightness settings; b) Utilizing a logarithmic relationship between radiation intensity and grey level can lower overall image contrast and better define the inner diameter position.
CR scanners produce logarithmic images directly For scanners and DDA systems that generate non-logarithmic response images, a suitable look-up table (LUT) can be utilized to convert the digital image into a logarithmic response format.
To ensure accurate measurements using the interactive on-screen method, it is essential to first verify its precision by applying it to a section of the pipe with a known wall thickness, such as an area confirmed to be free from corrosion or erosion, while using the current contrast and brightness settings of the displayed image.
Grey-level profile analysis methods
Many CR/DDA systems enable wall thickness measurement through the analysis of a grey level profile perpendicular to the pipe wall axis, in addition to on-screen measurements The software generates a grey-level profile along this line, typically displayed on-screen over the image, as shown in Figure 8 Automated routines facilitate this process.
Automated analysis routines enhance the reliability of wall thickness measurements, particularly when the maximum tangential penetrated thickness, \$w_{max}\$, is not close to the maximum limit of the radiation source used However, the accuracy of these automated processes can be compromised by factors such as external scale, corrosion products, or irregularities in internal and external corrosion.
Automated routines may encounter uncertainties, necessitating that the operator verify the consistency of the derived values with both the density profile and the digital radiographic image Additionally, interactive methods can be employed to enhance accuracy and reliability.
Figure 8 illustrates the interactive measurement of wall thickness by utilizing cursors on a grey level profile across the pipe wall This process follows the application of a logarithmic look-up table to the CR image and high-pass filtering to enhance details The outer diameter is indicated by a distinct peak in the profile, while the inner diameter is identified by a significant change in the gradient of the profile.
This method, combined with a visual assessment of the image, can in some circumstances give higher measurement accuracy than the automated routines described above
Figure 8 — Example of interactive wall thickness measurement using cursors superimposed on a grey level profile taken across the pipe wall
As the tangential penetrated thickness, \$w_{max}\$, nears the maximum recommended value for the isotope, the accuracy of all measurement methods declines This is primarily due to the challenges in reliably determining the location of the inner wall, which is exacerbated by reduced contrast and heightened noise levels.
8 Digital image recording, storage, processing and viewing
Scan and read out of image
Detectors and scanners must be utilized according to the manufacturer's guidelines to achieve optimal image quality It is essential that digital radiographs are devoid of artifacts caused by processing, handling, or other factors that could hinder accurate interpretation The analysis can be based on various exposures, similar to multi-film viewing Additionally, software tools can measure wall thickness from individual or overlaid images The effectiveness of this technique and any associated software should be validated using a reference pipe sample with known dimensions.
When using DDAs, it is essential to follow the manufacturer's recommended detector calibration procedure This involves calibrating the detector with a background image (without radiation) and at least one gain image (with radiation and uniform exposure) While multi-gain calibration enhances the achievable signal-to-noise ratio (SNR) and linearity, it requires more time To minimize noise during calibration, all images should be captured with at least double the exposure dose (mA ⋅ min or GBq ⋅ min) compared to what will be used for production radiographs Calibrated images should be treated as unprocessed raw images for quality assurance, provided the procedure is documented Regular calibration and bad pixel interpolation are necessary, especially when exposure conditions change significantly.
Bad pixels are suboptimal detector elements in digital detector arrays (DDAs), as outlined in ASTM E2597 It is crucial to map the detector to identify the bad pixel map following the manufacturer's guidelines, and this map must be documented Interpolating bad pixels is a necessary procedure in radiography with DDAs It is advisable to use detectors that do not contain cluster kernel pixels (CKP) within the region of interest (ROI).
The evaluation of digital data from the radiographic detector involves a linearized grey value representation that correlates directly with the radiation dose, essential for determining SNR, SR b, and SNR N To achieve optimal image display, it is crucial that contrast and brightness are interactively adjustable The software should also incorporate optional filter functions, profile plots, and tools for SNR and SNR N to enhance image evaluation For detailed image analysis, operators must be able to zoom in with a factor ranging from 1:1, where one pixel of the digital radiograph corresponds to one monitor pixel, to 1:2, where one pixel of the digital radiograph is represented by four monitor pixels.
Further means of image processing applied on the stored raw data (e.g high pass filtering for image display) shall be documented, be repeatable and be agreed between the contracting parties
8.6 Digital image recording and storage
CR/DDA images should be stored in a file format which supports a minimum of 12-bits/pixel
Original images must be preserved in their full resolution as provided by the detector system Prior to storing the raw data, only image processing related to detector calibration—such as offset correction, gain calibration for equalization, and bad pixel correction (refer to ASTM E2597)—should be applied to ensure that the images are free of artifacts.
The data storage shall be redundant and be supported by suitable back-up strategies to ensure “loss-less” data storage
Any data compression techniques used in the storage of these files shall be “loss-less”, i.e it shall be possible to reconstruct the exact original data from the compressed data
The digital radiographs shall be examined in a dimmed room The monitor setup shall be verified with a suitable test image
The display used for image evaluation must meet specific minimum standards, including a brightness of at least 250 cd/m², the ability to show a minimum of 256 shades of grey, a light intensity ratio of at least 1:250, and a resolution of no less than 1 megapixel with a pixel pitch of less than 0.3 mm.
A test report must be generated for each exposure or set of exposures, detailing the radiographic technique employed and any special circumstances that could enhance the understanding of the results.
The test report must include essential information such as a reference to the applicable standard, the name of the examination body, and details about the object and pipe isometric along with pipe content It should specify the material type, outer diameter (De), and nominal wall thickness (t) of the pipe, as well as the material, thickness, and condition of insulation Additionally, the report must outline the material, dimensions, and position of the comparator (C), along with the examination specifications and acceptance requirements It should detail the radiographic technique and class, the test arrangement as per section 6.1, and the marking system used Furthermore, the report needs to include a detector position plan, information about the radiation source, type and size of the focal spot, and identification of the equipment utilized It should also cover the detector, screens, filters, tube voltage, current or source activity, CR system, IP type, scanner model, and scanner parameters such as scan speed, gain, laser intensity, laser spot size, and pixel size Lastly, it must specify the DDA type, operating parameters, pixel size, basic spatial resolution of digital detectors, and measured image parameters.
1) film densities measured at pipe centre (if applicable), in the pipe wall and outside pipe;
The article outlines key parameters for pipe inspection, including the Signal-to-Noise Ratio (SNR) measured at the pipe center and in the free beam, as well as the measured wall thicknesses, highlighting the minimum thickness and its location It addresses material loss, whether inside, outside, pitting, or generalized, and includes additional observations Any deviations from the standard must be documented through special agreements Finally, the report should include the operator's name, certification, signature, and the date(s) of exposure and testing.
Determination of basic spatial resolution
Linearized grey levels are essential for accurately measuring basic spatial resolution values, as they ensure that grey values correspond proportionally to radiation exposure at specific image locations This process is usually facilitated by the software provided by the manufacturer.
The duplex wire IQI must be placed directly on the surface of the detector or cassette, and its reading should follow the guidelines set by EN ISO 19232-5 to accurately determine the basic spatial resolution (SR b) of the detector.
When the duplex wire IQI is placed on a test object rather than directly on the detector, the measurement obtained reflects the image basic spatial resolution, denoted as SR b image, instead of the detector basic spatial resolution, referred to as SR b detector.
If the first unsharp wire pair cannot be recognized clearly (see EN ISO 19232-5), the 20 % dip method shall be applied as follows:
In the digital radiograph, the initial wire pair that exhibits a modulation dip of less than 20% compared to the double peak size must be recorded as the outcome of the IQI test, such as D8 illustrated in Figure A.1a.
The image processing software will utilize a profile function to identify the initial wire pair, ensuring a dip of less than 20% when averaged across both minima (refer to Figure A.1(d)) Additionally, to enhance the signal-to-noise ratio (SNR) in the profile plot, the profile will be averaged over a minimum of 21 individual line profiles (see Figure A.1 b-c).
Using the duplex wire IQI in accordance with EN ISO 19232-5, the inherent image unsharpness (\$u_i\$) will be assessed, and the basic spatial resolution (\$SR_b\$) of the detector can be calculated using the formula: \$SR_b = b \cdot u_i\$.
Calibration of DDAs
When using DDAs, it is essential to follow the manufacturer's recommended detector calibration procedure This involves calibrating the detector with a background image (without radiation) and at least one gain image (with radiation and uniform exposure) Although multi-gain calibration enhances the achievable signal-to-noise ratio (SNR) and linearity, it requires more time To minimize noise during calibration, all calibration images should be captured with at least double the exposure dose (mA ⋅ min or GBq ⋅ min) compared to what will be used for production radiographs Calibrated images should be treated as unprocessed raw images for quality assurance, provided the procedure is documented Regular calibration and bad pixel interpolation are necessary, especially when exposure conditions change significantly.
Bad pixel interpolation
Bad pixels are suboptimal detector elements in digital detector arrays (DDAs), as defined by ASTM E2597 It is crucial to map the detector to identify the bad pixel map following the manufacturer's guidelines, and this map must be documented Interpolating bad pixels is a necessary procedure in radiography with DDAs It is advisable to use detectors that do not contain cluster kernel pixels (CKP) within the region of interest (ROI).
Image processing
The evaluation of digital data from the radiographic detector involves a linearized grey value representation that correlates directly with the radiation dose, essential for determining SNR, SR b, and SNR N To achieve optimal image display, it is crucial that contrast and brightness are interactively adjustable The software should also include optional filter functions, profile plots, and tools for SNR and SNR N to enhance image evaluation For detailed image analysis, operators must be able to zoom in with a factor ranging from 1:1, where one pixel of the digital radiograph corresponds to one monitor pixel, to 1:2, where one pixel of the digital radiograph is represented by four monitor pixels.
Further means of image processing applied on the stored raw data (e.g high pass filtering for image display) shall be documented, be repeatable and be agreed between the contracting parties.
Digital image recording and storage
CR/DDA images should be stored in a file format which supports a minimum of 12-bits/pixel
Original images must be preserved in their full resolution as provided by the detector system Only image processing related to detector calibration, such as offset correction, gain calibration for equalization, and bad pixel correction (refer to ASTM E2597), should be performed prior to storing the raw data to ensure artifact-free images.
The data storage shall be redundant and be supported by suitable back-up strategies to ensure “loss-less” data storage
Any data compression techniques used in the storage of these files shall be “loss-less”, i.e it shall be possible to reconstruct the exact original data from the compressed data.
Monitor viewing conditions
The digital radiographs shall be examined in a dimmed room The monitor setup shall be verified with a suitable test image
The display used for image evaluation must meet specific minimum standards, including a brightness of at least 250 cd/m², the ability to show a minimum of 256 shades of grey, a light intensity ratio of at least 1:250, and a resolution of no less than 1 megapixel with a pixel pitch of less than 0.3 mm.
A test report must be generated for each exposure or set of exposures, detailing the radiographic technique employed and any special circumstances that could enhance the understanding of the results.
The test report must include essential information such as a reference to the applicable standard, the name of the examination body, and details about the object and pipe isometric along with pipe content It should specify the material type, outer diameter (De), and nominal wall thickness (t) of the pipe, as well as the material, thickness, and condition of insulation Additionally, the report must outline the material, dimensions, and position of the comparator (C), along with the examination specifications and acceptance requirements It should detail the radiographic technique and class, the test arrangement as per section 6.1, and the marking system used Furthermore, the report needs to include a detector position plan, information about the radiation source, type and size of the focal spot, and identification of the equipment utilized It should also cover the detector, screens, filters, tube voltage, current or source activity, CR system, IP type, scanner model, and scanner parameters such as scan speed, gain, laser intensity, laser spot size, and pixel size Lastly, it must specify the DDA type, operating parameters, pixel size, basic spatial resolution of digital detectors, and measured image parameters.
1) film densities measured at pipe centre (if applicable), in the pipe wall and outside pipe;
The article outlines key parameters for assessing pipe integrity, including the Signal-to-Noise Ratio (SNR) at the pipe center and in the free beam, measured wall thicknesses with a focus on minimum thickness and its location, and material loss types such as inside, outside, pitting, or generalized It also emphasizes the importance of additional observations, any deviations from established standards by special agreement, and requires the operator's name, certification, and signature, along with the date(s) of exposure and the test report.
Determination of basic spatial resolution
Linearized grey levels are essential for accurately measuring basic spatial resolution values, as they ensure that grey values correspond proportionally to radiation exposure at specific image locations This process is usually facilitated by the software provided by the manufacturer.
The duplex wire IQI must be placed directly on the detector or cassette surface, and its reading should follow the guidelines of EN ISO 19232-5 to determine the basic spatial resolution (SR b) of the detector.
When the duplex wire IQI is placed on a test object rather than directly on the detector, the measurement obtained reflects the image basic spatial resolution, denoted as SR b image, instead of the detector basic spatial resolution, referred to as SR b detector.
If the first unsharp wire pair cannot be recognized clearly (see EN ISO 19232-5), the 20 % dip method shall be applied as follows:
In the digital radiograph, the initial wire pair that exhibits a modulation dip of less than 20% compared to the double peak size must be recorded as the outcome of the IQI test, such as D8 illustrated in Figure A.1a.
The image processing software will utilize a profile function to identify the initial wire pair, ensuring a dip of less than 20% when averaged across both minima (refer to Figure A.1(d)) Additionally, to enhance the signal-to-noise ratio (SNR) in the profile plot, the profile will be averaged over a minimum of 21 individual line profiles (see Figures A.1 b-c).
Using the duplex wire IQI in accordance with EN ISO 19232-5, the inherent image unsharpness (\$u_i\$) will be assessed, and the basic spatial resolution (\$SR_b\$) of the detector can be calculated using the formula: \$SR_b = b \cdot 1 / u_i\$.
To prevent aliasing effects, the duplex wire IQI should be set at an angle of about 2° to 5° relative to the orientation of the pixel line or column, as illustrated in Figure A.1.
The determination of the basic spatial resolution for a digital detector system (SR b ) shall be performed under one of the following exposure conditions without object: a) Inspection of light alloys:
2) prefilter 1 mm Al b) Inspection of steel and copper alloys ≤ 20 mm penetrated thickness:
2) prefilter 1 mm Cu d) Gamma radiography or high energy radiography:
1) Use the gamma source as specified or X-ray source > 1 MV;
2) prefilter 2 mm Cu or 4 mm steel for Se 75, Ir 192, and 4 mm Cu or 8mm steel for Co 60 or X-ray voltage > 1 MV
The duplex wire must be placed directly on the detector or cassette surface, with a source-to-detector distance of (1,000 ± 50) mm In the digital image, the mean grey value should exceed 50% of the maximum grey value, or the signal-to-noise ratio (SNR) must be greater than 100 for standard systems with a pixel size of 80 µm or larger, and greater than 70 for high-resolution systems with a pixel size smaller than 80 µm in the reference radiograph Additionally, the basic spatial resolution, as defined in Formula A.1, should be documented in the examination report for the digital system and its settings used.
The basic spatial resolution of CR systems is determined by measuring it both perpendicular and parallel to the laser scanning direction The higher of the two spatial resolution values, referred to as SR b or SR b detector, is used as the resulting detector basic spatial resolution This process includes analyzing an image of the duplex wire IQI in a radiograph, averaging the profile from at least 21 lines, examining a zoomed profile of wire pairs D7 and D8, and utilizing a scheme for calculating the dip value.
Figure A.1 — Example for duplex wire IQI evaluation with resulting IQI value D8, being the first one with a dip < 20 %
To enhance the accuracy of measuring the SR b or SR b image value, it is essential to interpolate the 20% dip value from the modulation depth of adjacent duplex wire modulations Figure A.2 illustrates the procedure for a high-resolution CR system, showcasing a profile plot of the measured profile along with the determined modulation depths.
(dips) b) Interpolation of modulation depth vs duplex wire diameter
NOTE The 20 % value is determined from the intersection with the 20 % line resulting in iSRb = 66 àm
Figure A.2 — Example for determination of the interpolated basic spatial resolution (iSR b ) by interpolation from the measured modulation (dip) of the neighbour duplex wire elements
The relationship between modulation (dip) and wire diameter should be modeled using a second-order polynomial to determine the intersection with the 20% line, as shown in Figure A.2 Only modulation values greater than zero should be utilized for interpolation purposes.