Computed tomography (CT) on the way to measurement uncertainty estimation based on simulation

With the growing importance of industrial X-ray computed tomography (CT) in coordinate metrology, the need for precise and reliable CT measurements is increasing as well. Despite many enhancements achieved in the past 10 years regarding software and hardware as well as normative developments, dimensional CT has not yet reached the same level of and precision and reliability as tactile and optical coordinate measurement technologies. This is mainly because the traceability of CT measurements to the metre could not be yet stablished, especially because of the complex interplay of a large number of factors influencing the measurement results, which hinders (to a large extend) the task of measurement uncertainty determination.

State-of-the-art methods of measurement uncertainty estimation used in coordinate metrology are often very complex and require either extensive research work or large measurement effort by performing repeated measurements (at least 20 repetitions) of a calibrated workpiece.

The simulation of the CT measurement process is a promising and efficient way to estimate in future the measurement uncertainty numerically, saving time, costs and resources. However, there is currently no standardised procedure to verify and quantify the reliability of simulation softwares. Therefore, the main goal of the WIPANO project CTSimU was to create an objective and standardised procedure to test and evaluate CT simulation softwares. This standardised procedure tests the basic capabilities and functionalities as well as sufficiently correct basic physical effects of CT simulation softwares with respect to dimensional measurements of industrial workpieces. Besides that, the basic qualification provides simulation software users a certain level of standard or guideline safety through the use of standardised and verified procedures and it will facilitate, in a long term, the estimation of the measurement uncertainty using simulation.

Because of the complexity of the measurement process, CT simulation softwares currently available were basic qualified. The basic qualification checks whether the softwares are correct from metrological and physical perspectives. For that, a requirement analysis was carried out and a requirements catalogue was created as well. After that, a test framework was created, featuring representative measurement tasks to be carried out on a set of designed reference standards (see Figure 1) and their respective measurement procedures. The test framework serves to characterise the simulation softwares and can along with the requirements catalogue and the reference data be used to qualify further simulation softwares.

The main outcome of the project is a draft of the German guideline for VDI/VDE 2630 Part 2.2 “Basic qualification of software systems for the simulation of dimensional measurements with X-ray computed tomography”. This national guideline draft also represents a pre-draft for an international (ISO) standard. PTB's work in the CTSimU project focused in particular on the creation of the 2D and 3D test scenarios and their metrological simulation and evaluation as well as the writing process of the guideline draft.

The project CTSimU was financed by the federal ministry for economic affairs and energy (BMWi) in the funding guidelines WIPANO with the funding identification 03TNH026D.

Figure 1: Reference standard designs in use for the basic qualification of CT simulation softwares:
(a) Hole-sheet standard with conical holes; (b) spherical step-wedge; (c) Tungsten edge; (d) and (e) Front and back view of the multi-geometry cuboid.