ContourMap is a unified system to carry out the data processing steps required in the contour method for determining residual stresses. These are:
- Carry out initial (pre alignment and averaging) data cleaning.
- For laser scanned data, determine the perimeters of the measured surfaces – for CMM data, these are usually available as part of the measurement.
- Generate triangular meshed surfaces from the raw measured data.
- Reflect the data for one of the surfaces, and align the data for the two surfaces using the perimeter data.
- Average the data from the two surfaces.
- Carry out post alignment and averaging data cleaning.
- Carry out data smoothing, either by simpler established methods (e.g. splines) or using a local polynomial smoothing algorithm.
- Impose the smoothed displacements on the appropriate surface of a finite element model, and run this to obtain the stresses.
Strain Scanning Simulation Software, (SScanSS)
Advanced experimental methodology for measuring strain by neutron diffraction developed by The Open University in collaboration with staffs at the ISIS neutron source and now used worldwide.
The SScanSS method was specifically designed to meet the needs of exacting, safety critical industries by offering:
- Maximal output of accurate strain measurements, using:
b) Optimised and automated instrument control
- A high level of traceability and QA through provision of:
b) Automatic recording and archiving of instrument configuration and instrument movements throughout experiment, enabling comprehensive record keeping and verification of measurement points etc.
Background and introduction
The ENGIN-X neutron diffraction instrument at the ISIS neutron source in Oxfordshire represents the state of the art in the non-destructive measurement of residual stresses, . Not only does it utilise one of the world's most powerful pulsed neutron sources, but more specifically it incorporates a powerful combination of hardware and software that enables it to provide services of particular appropriateness to the engineering community.
The technique of Neutron strain scanning (NSS), for determining the stress field deep inside engineering components or test samples, has evolved rapidly since its inception in the early 1980s. NSS is now an established tool for both academia and industry that commands substantial world-wide investment.
The ENGIN-X instrument at ISIS, was one of the first instruments designed with this ethos in mind, however it was recognized from the start that hardware improvements alone would not be sufficient to fully realize the potential of these instruments. In particular routine problems of experimental method needed to be overcome such as the difficulty of sample positioning and alignment and of estimating the time needed for an experiment. In addition the avoidance of collisions between the sample and elements of the instrument hardware was of prime importance. The possibility of using modern software techniques in the solution of these problems was realized in the development of the SScanSS software suite, [2, 3]. SScanSS utilizes virtual reality methods to provide comprehensive planning and execution tools for strain scanning experiments. The software provides comprehensive facilities for positioning measurement points with the sample and automating instrument movements to realize these measurements. In addition the neutron path length through the sample may be calculated and combined with the instrument gauge volume and material attenuation coefficient to provide estimates of the likely measurement time. In this way the temporal and spatial viability of the experiment can be verified in advance and maximum use made of the beam-time through planning and automation.
Figure 1: The ENGIN-X virtual laboratory: Three-dimensional models of the laboratory and sample are used for planning of experiments. (a) Simulated measurement of the axial direction of the strain tensor in a large pipe sample. (b) Once the position of the pipe has been measured the positioner movements for the actual measurement are determined automatically. (c) Measurement positions are placed within the three-dimensional model of the sample which is produced either from simple primitive objects or from detailed descriptions of the actual surface of the specimen obtained using the ENGIN-X coordinate-measuring machine and laser scanner. (d) Count time estimates are produced by combining knowledge of the material attenuation properties with calculated path lengths, (calculated intensity (symbols), measured intensity (continuous).
Use of robotics methods for modelling and controlling Neutron and Synchrotron diffraction instrumentation
The SScanSS software tools for planning and executing neutron strain scanning experiments were initially written specifically for the ENGIN-X engineering diffractometer at ISIS in the UK. However, recognition that a majority of the specimen positioning systems in use at strain scanning facilities are effectively serial robot manipulators, suggested that the methods of serial robot kinematic modelling might provide a means to generalize these tools for other facilities.
Figure 2: The robotics formulation enables the accurate modelling of arbitrary positioning systems:
A pipe sample with neutron access hole mounted for measurement on (a) the ENGIN-X (x, y, z, Ω ) table, (b) the KOWARI instrument at ANSTO (Australian Nuclear Science and Technology Organisation ), (c) the NRSF2 instrument at ORNL (Oak Ridge National Laboratory USA).
The numerical solution of the inverse kinematic problem allows specimens to be automatically positioned and orientated so that pre-determined strain components are measured. Using this approach the measurement positions and required strain components are established, prior to an experiment, on a virtual reality model of the sample to be measured. The software will then calculate the positioner movements that are required to execute these measurements, either in simulation mode, for planning purposes, or for automated instrument control at the time of the experiment.
A positioning system with sufficient degrees of freedom, such as the ENGIN-X (x, y, z, Ω ) table with the addition of 3-axis goniometer provides considerable flexibility and the option to i) measure the three orthogonal strain components typically required for stress determination to be measured consecutively at each measurement point or, ii) optimize a secondary characteristic of the measurement position such as the measurement count time.
Figure 3: Alignment of a complex sample on the ENGIN-X goniometer: (a) Measurement points and strain components are defined on the virtual sample model, (b) in the alignment step measurement vectors are aligned coincidentally with the instrument Q-vectors so that the required components are measured at each measurement point, (c) The ENGIN-X (x, y, z, Ω), table with the addition of a triple axis goniometer provides a very flexible positioning system enabling the three orthogonal strain components typically required for stress determination to be measured consecutively at each measurement point.
Measuring Stress on hidden features
The geometry of the sample being measured is usually captured by laser scanning. In some circumstances however it may be required that account is taken of the internal geometry of the sample, either in positioning measurement points or in optimising the beam path through the object during the measurements. If this facility is required and suitable CAD models are not available tomographic data may be utilised using neutron or synchrotron imaging. A new (£10M) instrument, ‘IMAT’ at ISIS is expected to be available from 2015 which will enable this technique to be conveniently applied, [4, 5].
Figure 4: Using tomography to position stress measurements on internal features: (a) The IMAT experimental hutch. IMAT will be the world’s first instrument capable of combined neutron imaging and diffraction, (b) Neutron tomography of turbine blade, . Neutron tomography provides the complete external and internal geometry of complex components, (c) Sections through the tomography derived model enable measurement points to be positioned in relation to hidden internal features.
1. J.R. Santisteban, M.R. Daymond, J.A. James and L. Edwards, ENGIN-X: A Third Generation Strain Scanner: J. of Appl. Crys, (2006), 39, 812-825.
2. J.A. James, J.R. Santisteban, L. Edwards and M.R. Daymond, A Virtual Laboratory for Neutron and Synchrotron Strain Scanning. Physica B, Condensed Matter (2004) 350 pp.743-746.
3. J.A. James and L. Edwards, Application of robot kinematics methods to the simulation and control of neutron beam line positioning systems. Nuclear Instruments and Methods in Physics Research A. (2007) 571, 709-718.
4. G Burca, J James, W Kockelmann, M Fitzpatrick, S Zhang, J Hovind, R.van Langh, A new bridge technique for neutron tomography and diffraction measurements. Nucl Instrum Meth A, DOI:10.1016/j.nima.2011.01.096
5. Burca, W. Kockelman, J.James, M.Fitzpatrick, Modelling of an imaging beam-line at the ISIS pulsed neutron source, Journal of Instrumentation. (2013) 8, 10001-DOI: 10.1088/1748-0221/8/10/P10001
6. S. Pierret, A. Evans, A Paradowska, A. Kaestner, J. James, T. Etter and H. Van Swygenhoven Combining neutron diffraction and imaging for residual strain measurements in a single crystal turbine blade. Journal of Non-Destructive Techniques & Evaluation" (NDT&E) International 45 (2012) 39–45