Skip to main content

Poster abstracts

Characterisation of CT Noise in Projection and Image Space with Applications to 3D Printing
S. Ip
Warwick Statistics, University of Warwick, Coventry, CV4 7AL, UK

X-ray CT can be used in defection detection for 3D printing. The object is scanned at multiple angles to reconstruct the object and any defects in 3D space. The process can be time consuming. To speed up the process it may be possible to conduct defect detection in projection space from a single scan. X-ray photons behave randomly, they arrive at a Possion rate with random energy. Hence to do defect detection pixel by pixel, uncertainty must be taken into account. One way is to model the greyvalue of each pixel as a compound Possion random variance to capture the behaviour of x-ray photons. This results in a linear relationship between the mean and variance of the greyvalue, which can be used for variance prediction. Methods for multiple hypothesis testing, when testing each pixel for defects, must consider dependencies between pixels. One way to test this is to estimate the spatial autocorrelation and it was investigated whenever shading correction would also correct correlation between pixels.

Return to top

Revealing the Complex Metal Carbide Morphologies of Cast Ni Superalloy by Synchrotron X-ray Microtomography
B. Koe1, 2, J.C. Khong1, 3, N. Green4, A.J. Bodey2, Z. Zhang5, J. Gebelin5, C. Rau2, J.Mi1
1 School of Engineering & Computer Science, University of Hull, Hull HU6 7RX, UK
2 Diamond Light Source Ltd, Harwell Science and Innovation Campus, Didcot OX11 0DE, UK
3 Department of Medical Physics & Biomedical Engineering, University College London, London WC1E 6BT, UK
4 High Temperature Research Centre, University of Birmingham, Birmingham B15 2TT, UK
5 Doncasters Group Technical Centre, George Baylis Rd, Droitwich WR9 9RB, UK

Nickel-based superalloys are widely used in the aerospace industry, particularly in high temperature applications, due to its’ exceptional high-temperature strength, toughness, creep properties, and resistance to degradation in corrosive and oxidising environments. Metal carbides are essential structural features for improving the high-temperature creep properties of these superalloys. Although the structures and morphologies of these metal carbides were widely studied in the past by using electron microscopy techniques, the true 3D morphology of these metal carbides have not been reported. In this study, we used synchrotron X-ray microtomography, at the Diamond-Manchester Imaging Beamline (I13-2), Diamond Light Source, UK, to reveal the influence of solidification time on the true 3D morphology and the complex network of metal carbides formed during the casting of a widely used IN713LC Ni superalloy. In order to maintain the high fidelity tomography datasets, the visualisation nodes of the University of Hull’s newly established high performance computer (Viper) were utilised to perform the graphics-intensive visualisation.

Return to top

The effects of continuous scan motion on surface measurement by X-ray computed tomography
L. Körner1, A. Thompson1, N. Senin1, 2, S. Lawes1 and R. Leach1

1Manufacturing Metrology Team, University of Nottingham, UK
2 Department of Engineering, University of Perugia, 06125, Italy

X-ray computed tomography (XCT) has recently gained attention as a tool for surface metrology. Particularly, XCT’s ability to measure internal features allows for non-destructive surface metrology of additively manufactured (AM) parts. While initial investigations have proven the feasibility of XCT for surface texture measurement, the various factors that influence XCT measurements have not yet been explored in depth. The parameter space of XCT measurements is large: the current guideline document VDI/VDE 2630-1.2 lists over sixty variables. Amongst such variables, the number of projections has been of frequent interest to the XCT community. For practicality and speed, continuous scanning, i.e. measurement with continuous rotation of the sample, is commonly used. With continuous scanning, projection data is averaged over a span of angular positions. The work presented in this paper explores the effects of changing the number of projections while operating in continuous scanning mode. A cubic sample of (10 × 10 × 10) mm, manufactured on a Realizer SLM 50, using a sieved Ti6Al4 powder to limit the grain size to below 32 µm, is scanned on a 225 kV source cone beam XCT system, with the longest dimension, the diagonal, in line with the axis of rotation. The number of projections and therefore the angular averaging of the projection data is varied. For each condition surface data are extracted using VGStudio MAX 3.0 and exported as triangulated models. These are raster-scanned into digital elevation models for calculation of areal surface texture parameters in MountainsMap. Surface topography results corresponding to each XCT measurement configuration are assessed through summary indicators such as texture parameters (e.g. those defined in ISO 25178-2). Texture parameter results are compared to the topography measured with a focus-variation microscope.

Return to top

The life and death of pixels
J. Brettschneider
Warwick Statistics, University of Warwick, Coventry, CV4 7AL

While there are obvious global measures to quantify the amount of damage in an X-ray detector, natural questions also arise around the spatial distribution of the dysfunctional pixels and how observed patterns of dysfunctional pixels may be interpreted.After modelling occurrences of dysfunctional pixels as a planar point process we develop a higher level approach for analysing their spatial distributions. Key idea is to move from the notion of a dysfunctional pixel to the concept of a damage event defined by configurations of dysfunctional pixels using a typology based on local grid geometry. High density regions can be detected using density estimation of the damage event process, so remaining areas becomes suitable candidates for complete spatial randomness. This approach decouples observed damage from the detector resolution prescribed by I and from the exact shape of dysfunctional pixel configurations. We proposes a detector quality toolkit that allows users to monitor their technology following these principles. The methods allow users of detector based imaging technologies to detect, distinguish and monitor different types of quality damage and to identify the ones linked to specific causes. We apply our methods to a collection of bad pixel maps obtained as part of regular monitoring routines of a detector used in X-ray computed tomography.

Return to top

A Novel Surface Characterisation Strategy for the X-ray Computed Tomography Measurement of Complex Additively Manufactured Parts
S. Lou, W. Zeng, H. Abdul-Rahman, X. Jiang, P. J. Scott
EPSRC Future Metrology Hub, University of Huddersfield, Queensgate, Huddersfield, HD1 3DH

X-ray Computed Tomography (XCT) has the advantage over the traditional tactile and optical measurement systems in that it is the only valid non-destructive method to measure both external and internal geometries of complex functional parts, e.g. additively manufactured (AM) parts, whose intricate shape does not allow line-of-sight. However the XCT measurement posts many challenges to surface texture assessment of AM parts. One of the big issues is that XCT generated measurement data structures, i.e. point cloud and triangular mesh, are not straightforward compatible with the standard surface texture characterisation, which requires uniform sampled grid structure and also requires measured surface to be basically planar. This work proposes a novel strategy that the surface filtration and roughness parameterisation techniques can deal with triangular mesh. Based on the link between the Gaussian cutoff wavelength and the diffusion time, the proposed linear diffusion equation can achieve a Gaussian filtering effect on complex surfaces. Also the areal surface texture height parameters are extended to triangular mesh. With these two enhancements, it contributes to the solution to using XCT for a holistic and reliable measurement of surface texture of complex AM products.

Return to top

XCT for forensic engineering and failure analysis
N. Kourra, J. M. Warnett, A. Attridge, M. A. Williams
WMG, University of Warwick, Coventry, CV4 7AL

Forensic engineering and failure analysis is a necessary process for development that demonstrates the required changes to evolve designs, material selection, manufacturing processes, operation guides and maintenance procedures. Interpretation of the facts and reasons for failures is achieved by examining the physical evidence, verifiable facts and utilising scientific principles, knowledge, skills and methodologies. Analysis by non-destructive and destructive testing provides conclusions explaining the failures and evidences routes to mitigation in the future through design alterations. X-ray Computed Tomography (XCT) is applicable in non-destructive evaluation and metrology, using a series of radiographs to generate 3D models of the examined objects. The greatest advantage of this technology is the representation of outer and inner structures of the examined objects that assist in the identification of abnormalities, defects that can lead to failures and hidden failures. This technology was used to investigate failures non-destructively and avoid the collection of reduced data by destroying part of the problematic areas. This study examines a series of different failures analysis and utilises volumetric analysis to examine any abnormalities and defects. The results demonstrate the application of CT in forensic engineering and failure analysis as a non-destructive test while they would be unobtainable with any other non-destructive method.

Return to top