A low-cost, high-throughput X-ray scanner for quantitative 3D imaging using biaxial computed laminography

X-ray radiography is widely used in non-destructive testing (NDT) since the penetrative nature of X-rays enables defect identification in a wide range of industries, from composites (carbon-fibre and glass-fibre) to metal manufacturing and electronics (printed circuit boards and component assembly). However, the nature of radiography also superposes all the volumetric details into a single 2D image, which can hide location and size information about defects. While 3D imaging like Computed Tomography (CT) can be used to address this gap, CT is associated with added cost, time, higher radiation dose and cannot be performed in-situ. This paper introduces a portable, lower-cost and high-throughput scanner which can compute quantitative 3D images using the well-known technique of computed laminography (also known as digital laminography or digital tomosynthesis). In this technique the relative position of the X-ray source and object is translated linearly - our innovation is to improve depth resolution of the images, while also reducing image artefacts by moving the X-ray source in 2 axes. We provide evidence of improved image quality, and demonstrate wide-ranging industrial applicability, evidenced by images of damage-assessment in composite (CRP and GRP), metal parts and electronics. In this work, 3D images have been obtained within 2-10 minutes, and we discuss a pathway to reduce scan times to less than 1 minute. In addition, the lower radiation power required in our method means that these scanners can be deployed in a minimally disruptive manner in-field and in- situ as a part of the assembly-line manufacturing process.