A Robot-Assisted Microscopy System for Digital Image Correlation in Fatigue Crack Growth Testing

Abstract

Background Digital image correlation (DIC) with microscopes has become an important experimental tool in fracture mechanics to study local effects such as the plastic zone, crack closure, crack deflection or crack branching. High-resolution light microscopes provide 2D images but the field of view is limited to a small area and very sensitive to its alignment. A flexible positioning system is therefore needed to collect such DIC data during the entire fatigue crack growth process. Objective We present in our paper a new experimental setup for local high-resolution 2D DIC measurements at any location and at any time during fatigue crack growth experiments with a non-fixed DIC microscopy system. Methods We use a robot to move the 2D DIC microscope to any location on the surface of the specimen. Optical and tactile methods automatically adjust the system and ensure highest image quality as well as accurate alignment. In addition, an advanced repositioning method reduces out-of-plane motion effects. Results The robot is able to achieve a repositioning accuracy of less than 0.06 mm in vector space, resulting in very low Von Mises strain scattering of 0.07 to 0.09% in the DIC evaluation. The system minimizes systematic errors caused by translation and rotational deviations. Effects such as crack deflection, crack branching or the plastic zone of a fatigue crack can be investigated with a field of view of 10.2 x 6.4 mm2. Conclusions The robot supported DIC system generates up to 8000 high-quality DIC images in an experiment that enables the application of digital evaluation algorithms. Redundant information create confidence in the results as all revealed effects are comprehensible. This increases the information content of a single fatigue crack growth test and accelerates knowledge generation.

Next generation fatigue crack growth experiments of aerospace materials

“…Figure 4 a presents the same data plotted as d a /d N vs. Δ K . For this experiment, we integrated our robot-based infrastructure into a servo-hydraulic uniaxial test rig 45 . The whole system is shown in Fig.…”

“…This second system can therefore be used to perform HRDIC by moving the microscope to a region of interest using the crack tip information or by scanning the entire specimen’s surface in a checker board pattern (if the specimen is small). To ensure that the region of interest appears sharply in the focus of the microscope, the robot’s position can be fine-adjusted fully automated according to the implementation of Paysan et al 45 . The hardware is fully automated for uniaxial test rigs (Fig.…”

“…The hardware is fully automated for uniaxial test rigs (Fig. 1 a) 45 and was used in this setup to obtain the data discussed in this study. Moreover, the whole system has recently been adapted for a large biaxial test rig (Fig.…”

“…Nevertheless, automated DIC remains very challenging, particularly for cases in which variable regions of interest much smaller than the specimen are studied, www.nature.com/scientificreports/ as it is the case of high resolution DIC (HRDIC). To address this challenge, Paysan et al recently developed algorithms for a robot-assisted test rig that can be used to automatically scan a specimen surface during fatigue crack growth experiments using an optical microscope for HRDIC analysis 45 .…”