Integrated Quality Control of Precision Assemblies using Computed Tomography

Computed Tomography (CT) is bringing about a profound change in the way that tolerance verification is performed in industry. CT allows the inner and the outer geometry of a workpiece to be measured without the need for external access or destructive testing. These are significant advantages over coor-dinate measuring machines (CMMs) when working with complex and fragile parts.

This Ph.D. project at DTU Mechanical Engineering concerns the applicability of CT for quality control of precision assem-blies. Investigations to quantify the accuracy of CT measurements, reference artefacts to correct sys-tematic errors in CT, and an international comparison on CT of Assemblies have been carried out during the project.

A series of investigations regarding the influence of the CT post-processing factors on the accuracy of CT measurements was carried out. Post-processing factors such as surface determination, data filtering and feature fitting were considered within the present investigations. The investigations were conducted on two CT systems, showing different metrological performances, and involved a variety of multi-ma-terial assemblies, having different shapes and materials.

The investigation results have showed that CT measurements on assemblies can be successfully conducted and that the surface determination method appears to be able to segment multi-material workpieces without any loss of accuracy. A novel type of artefact for calibration of the scale error in CT has been developed within the project.

One kind of artefact comprises a carbon fibre tubular structure on which a number of reference ruby spheres are glued. Another kind comprises an invar disc on which several reference ruby spheres are positioned at different heights using carbon fibre rods. The artefact is positioned and scanned together with the workpiece inside the CT scanner producing a 3D reference system for the measurement.

The two artefacts were calibrated on a tactile CMM and their applicability demonstrated using different cal-ibrated workpieces. The use of the developed artefacts ensures a considerable reduction of time by com-pressing the full process of calibration, scanning, measurement, and re-calibration, into a single imaging.

The use of the artefacts also allows a considerable reduction of the amount of data generated from CT scanning. Interlaboratory comparison on Computed Tomography of assemblies was carried out as a part of the Ph.D. study. The comparison involved 22 research laboratories from 7 countries, and was based on the circulation of two assemblies.

With respect to previous comparisons that only focused on physical items, the present comparison introduced a voxel item, which was an assembly scanned by the coordinator and distributed electronically to participants. The comparison results demonstrated that (i) length measure-ments are made without problems by most of the participants who corrected systematic errors effec-tively; (ii) CT post-processing has a sizable impact as the measurand complexity increases; (iii) the majority of the participants stated measurement uncertainties although many of them provided poor statements.