Contents of the publication
NDT Global reviews ultrasonic measurement of pipeline geometry, validation and application to dents and other shape changes. The authors single out the characteristics of detection and dimensionation, and discuss field verification. For engineering analysis, it is important that the quality of the profile is linked to the subsequent assessment of deformation, not just to the recording of the minimum diameter.
We will then consider the general requirements for geometric data. This is the self-explanatory explanation applicable to comparing different measurement principles. The published characteristics relate to the form of test facilities, the measurement grid and the conditions for verification. These parameters provide an insight into what details of the profile are accurately presented and where further assessment is required.
From distance to section shape
The sensor shall measure the distance or time of signal transmission relative to its position. To obtain a pipe shape, it is necessary to know the inter-channel position and the position of the measuring node. If the device is removed from the center, the same circular pipe creates different distances along the circle.
The reconstruction algorithm shall separate the movement of the instrument itself from the change of the wall. The model is geometric and over-observation. But if part of the circle is not measured, the division can become mixed. The result must therefore be accompanied by information on the channels available and the quality of recovery, especially in the vicinity of sharp distortions where the assumption of almost round sections works worse.
Minimum diameter does not describe dent
Two dents of the same depth may differ in length, width and local curvature. Accessable clearance is important for passing equipment; for assessing the deformed state — The shape of the surface and its change in space. So one depth indicator doesn't answer all engineering questions.
It is useful to keep the reference grid and surface construction rules. A highly smooth profile is suitable for viewing, but it can reduce local maximum curvature. The noise-intensive grid, on the other hand, creates artificial peaks when calculating derivatives. The choice of treatment should be consistent with the subsequent calculation and its influence should be checked on samples with known geometry, not only visually.
Sample step and accuracy — different properties
The axial step determines the distance between successive observations, the circle — Detailing the perimeter. Size error describes the difference between estimates and real values. A small grid helps to imagine a narrow feature, but does not correct the sensor position errors, sound speed or reconstruction model.
When comparing systems, you have to ask what geometry was used to confirm the characteristics. The result of smooth ovality is not necessarily carried over to a narrow sharp dent. The probability of detection and the size interval should also be distinguished. The more clearly described the area of confirmed objects, the less risk is to apply a beautiful specification table outside its meaning.
Joint geometry and wall thickness
Ultrasound can provide interconnected data on surface and thickness. This is useful in distinguishing between the shape and the metal. However, joint receipt does not eliminate the need to check each channel separately: loss of one reflection may affect thickness differently from geometry.
The data should retain the reliability and justification of the choice of reflection. Automatically setting the previous value can hide a local pass. If the geometric chart is continuous, the user must understand which points are measured and which ones are restored. For the analysis of the interacting corrosion and deformation, it is particularly important that both sets be tied to one real surface.
Form Field Check
During an external survey, the loading conditions may differ from that of the internal tube recording. The release of the tube from the support leg, pressure or temperature changes can change shape. Then the difference between profiles includes physical changes to an object, not just errors of two measurement methods.
It is useful to capture the modes and state of support before comparison, to combine the coordinates and to agree on the reference surface. Then they compare the same parameters, keeping the standard error. For a thorough analysis, it is desirable to see not only maximum depth but also the distribution of differences across the zone. This helps to distinguish the displacement of coordinates from systematic distortion of shape or insufficient resolution.
Implications for caliper tools
For OKB, it is of interest to arrange a complete path from sensor to engineering surface. Even if it were not physical, the same questions were useful: where the point of contact was located, how the movement of the instrument was taken into account, what details of the profile were confirmed and what happened when the channel was lost.
The drafting conclusion is to design the data format together with the caliper tool mechanics. The ability to check the source profile, repeat the treatment and evaluate the filtering effect is important for serious application as much as an attractive three-dimensional image. For the calculation model, it is particularly useful to return from smooth surface to reference points and see which local features are persistent when the processing parameters are changed reasonably.
Engineering output
A reliable profile requires knowledge of sensor position, processing and verification conditions. The minimum diameter and the beautiful image do not replace the tested spatial geometry.
Sources and further reading
- PPSA Technical Report 2018; open full textHenrik Witte, Roberto Yanez, Timo Schulten, Thomas Hennig, NDT Global — Ultrasonic Geometry: Proven accuracy for reliable assessment ↗
The article was originally written in Russian. It summarises the cited sources. For the original text and illustrations, consult the publications listed below.