VERIFICATION OF X‐RAY COMPUTED TOMOGRAPHY FOR ADDITIVE MANUFACTURING DIMENSIONAL & SURFACE TEXTURE METROLOGY AND DEPLOYMENT TO INTERNAL GEOMETRY MEASUREMENT

Additive manufacturing (AM) possesses many advantages over conventional subtractive manufacturing techniques, including construction of highly complex geometries, shortened delivery cycle and reduced material waste. AM has found many applications in key industry sectors, particularly in aerospace and healthcare. However, many technical barriers hinder the uptake of AM. Two major challenges are poor surface roughness and a lack of dimensional accuracy. X‐ray computed tomography (XCT) is currently the only valid inspection tool that can non‐destructively measure AM internal dimension and surface texture. This thesis aims to verify XCT’s capability in dimensional and surface texture measurement by comparing it to the well-established metrology techniques. Furthermore, XCT will be employed to measure dimensional deviation and surface texture of internal geometries produced by selective laser melting (SLM).

The results show that XCT beam hardening corrections aid in obtaining accurate inner dimensions, but may also reduce the accuracy of external dimensions. it is necessary to ensure that these methods are based on an accurate reconstruction volume.

it is found that the XCT result is usually more blurred than the FVM result due to its limited resolution. The rougher the surface, the closer the XCT result to the FVM. XCT can capture deep valleys and re‐entrant features without the line‐of‐sight constraint. The local iterative surface determination is superior to the ISO 50 method.

The diameter and volume of the internal channels are all smaller than the nominals at every inclination angle due to the dross formation, the staircase effect and the laser spot size. The diameter and volume of 45° and 54° are the smallest mainly due to the staircase effect, whereas the shape deformations are most severe at 0° and 51°. At a small inclination angle, the effect of dross formation outweighs the staircase effect.

The cube channels are used to evaluate surface texture variation regarding the inclination angle. The main factor affecting surface roughness is the staircase effect, rather than the particles. XCT results are closer to FVM results on the waviness surface; on the particle surface, XCT captures only part of particle features – mainly large‐size particles – depending on the specific resolution. The inner surface and the outer surface share similar change trends in response to the inclination angle.