The T-stress as a measure for the multiaxiality or the degree of constraint is well-known in linear-elastic fracture mechanics. It is known to affect the size and shape of the plastic zone in front of the crack tip and is therefore expected to directly influence plasticity-induced crack closure (PICC). At the same time, individual overload cycles also influence the size of this plastic zone not only during the current overload cycle but also during subsequent normal loading cycles.The focus of the present study is to experimentally study the influence of multiaxial far field loading on the crack opening displacement (COD) and hence crack closure and opening in two materials commonly used in gas turbine engines. For this, room temperature crack propagation tests are performed on the titanium alloy Ti6246 and the tempered steel 26NiCrMoV14-5. The experiments are conducted using cruciform specimens in a biaxial test-rig. This experimental set-up allows for a systematic variation and comparison of different T values as well as combined overload and normal load cycles, enabling targeted investigation of their effects on the plastic zone in the crack tip area. The tests are monitored and analyzed using a newly developed Digital Image Correlation (DIC) setup. This enables to evaluate crack growth and crack flank opening as well as the measurement of crack closure forces at different crack flank positions. Ultimately, this approach allows for a systematic investigation of PICC.In conclusion, a workflow is developed to automatically extract COD, crack closure forces, and the compliance of both open and closed crack states, as well as the resulting compliance offset, from the DIC data. The results show that while the influence of isolated overload cycles is not pronounced, the T-stress has a clear and measurable effect on crack closure forces in biaxially loaded cruciform specimens.

We present a magnetometric imaging device that harnesses the sensitivity of free spin precession (FSP) optically pumped magnetometers alongside a high-speed camera. This approach enables the rapid imaging of magnetic field distributions at a frame rate of 120 Hz, achieving an intrinsic sensitivity of 111 pTHz and a spatial resolution of 280 μm, using a microfabricated rubidium (Rb87) vapor cell filled with nitrogen buffer gas to mitigate diffusion. Our device employs the parallel readout capability of the camera to effectively remove common-mode noise. The imaging capabilities were demonstrated by capturing the magnetic field distribution created by a defined current running through a pair of wires. Recording FSP signals using a commercial camera simplifies the experimental setup and allows for a larger number of pixels, paving the way for practical applications in industrial environments. Although FSP magnetometers exhibit lower sensitivity compared to their spin exchange relaxation free counterparts, they offer a broader tolerance to magnetic offset fields, making them suitable for deployment in less magnetically clean environments.

Non-destructive testing (NDT) techniques are important for the evaluation of properties in materials, such as crack initiation or heat stresses induced by weld seams, without causing further damage to the component itself. Optically pumped magnetometers (OPM) offer a high sensitivity in the range of 15 fT/√Hz to detect resulting localized changes of the magnetization in ferromagnetic materials. Here, the sensitivity of OPMs allows the detection of small variations in magnetic stray fields on the surface of the material originating from such stresses in a volume only 0.1 mm3. However, to measure stresses in larger devices for NDT applications, the volume in which the magnetization is measured must be controlled by special devices like flux guides. They are fabricated from soft magnetic ferrites to adapt its magnetic characteristics. This paper presents first results from a novel OPM flux guide design to effectively pick up local magnetic stray fields on the surface of specimen with an increased spatial resolution and directing them to the sensing cell of a commercially available OPM. The scanning system is shielded against environmental magnetic perturbations to exploit the OPM sensitivity even in industrial environments. It is demonstrated how the two components, OPM and flux guide, can be combined to build a system for high resolution non-destructive testing of ferromagnetic steel samples.

Microfabricated optically pumped magnetometers (OPM) are a novel class of commercially available quantum magnetometers combining high sensitivity with high dynamic range.In non-destructive testing (NDT) this should enable new applications like measuring stray fields arising from small stress concentrations affecting local magnetization.In ferromagnetic steel the sensitivity of OPM is sufficient to measure magnetization from critical stresses for crack initiation in submillimeter volumes.However, OPM, as they are built today, require additional components like magnetic flux guides to control the measurement volume in NDT systems.The improvement of spatial resolution with such a flux guide is demonstrated using neighbored weld seams as an example for stress concentrations.

Precise determination of the remaining service life of technical components requires sufficient knowledge of fatigue crack growth behaviour and the growth rate of defects. Cracks in real components often experience multiaxial far field stresses due to their complex geometry and composite loadings acting on it. Digital image correlation (DIC) is well established for crack length and displacement measurements, but it usually requires sample preparation with speckle paint and interferes with mechanical extensometers. To overcome these limitations, we use a novel 2D DIC system combining a graphics processing unit (GPU) with a CoaXPress 2.0 camera, acquiring up to 3 GB/s of image data. It enables real-time evaluation of both integral strain like an extensometer and full-field DIC on images selected automatically in real-time. This combination enables the use of one single sensor for strain-controlled testing and fatigue crack growth characterisation. The full-field displacement is compared to a finite-element model (FEM) simulating the actual crack contour measured by the DIC system. The results show that high-performance DIC has the potential to simultaneously simplify crack-growth experiments and provide comprehensive fracture mechanical information.

The extreme sensitivity of quantum magnetometers enables new applications in material testing such as the identification of single defect events in the bulk of small volume specimen (0.1 mm³). Exposing ferromagnetic materials to strain alters their magnetic response. Due to uncompensated spins, defects arising from the fatigue process interact with magnetic domain walls. Optically pumped zero-field magnetometers (OPM) provide the sensitivity required to measure small variations in the magnetic response and potentially to quantify damage in the material. We provide first results of a novel micro fatigue setup with an integrated OPM to correlate variations of the magnetic response in a multimodal approach. The position of the Villari reversals within the magneto-mechanic hysteresis and the amplitude of magnetic field are potential candidates to estimate fatigue damage within the specimen.

Uniaxial fatigue testing of micro-mechanical metallic specimens can provide valuable insight into damage formation. Magnetic and piezomagnetic testing are commonly used for qualitative characterization of damage in ferromagnetic specimens. Sensitive and accurate measurements with magnetic sensors is a key part of such a characterization. This work presents an experimental setup to induce structural defects in a micro-mechanical fatigue test. Simultaneously, the resulting piezomagnetic signals are measured during the complete lifetime of the tested specimen. The key component is a highly sensitive optically pumped magnetometer (OPM) used to measure the piezomagnetic hysteresis of a small specimen whose structural defects can be analyzed on a small scale by other metallographic characterization methods as well. This setup aims to quantify the magnetic signatures of damage during the fatigue process, which could enable non-destructive mechanical testing of materials. This paper reports the initial results obtained from this novel micro-magneto-mechanical test setup for a ferritic steel specimen.

By means of fracture mechanics evaluation concepts, the most accurate possible estimation of the crack growth behaviour of complex and highly loaded structures is to be aimed for a more efficient use of components, a more precise planning of inspection intervals and for the realization of advanced design approaches. Standard fracture mechanics specimen used in materials testing are globally subjected to purely uniaxial loading. In contrast, the technical components usually experience multi-axial stresses due to the external loads and to their complex geometry. The influence of multi-axial far-field loads on the crack propagation rate and the crack closure behaviour is unclear and controversially discussed in literature. In this context, the crack propagation behaviour of uniaxially loaded corner-crack specimens and multiaxially loaded cruciform specimens is systematically studied in this investigation. Here, the multiaxiality condition is characterized by the so-called T-stress and systematically varied within the tests. The results so far show no significant influence of the far-field multiaxiality or the T-stress on the crack propagation behaviour. However, the evaluation of the crack closure behaviour indicates that an engineering approach for thin-walled structures leads to a non-conservative estimation.

Experiments to describe the crack growth behavior are expensive and complex, while the evaluation depends mainly on the determined crack lengths over the testing period. Digital image correlation (DIC) is well established for crack-length and displacement measurements, but it normally requires sample preparation with speckle paint and interferes with mechanical extensometers. Novel GPU-based DIC system capable of measuring total strain at rates up to 850 Hz and strain-fields overcome these limitations. We present first results of fatigue crack growth experiments with uniaxially und biaxially loaded specimens. The force controlled tests under room-temperature conditions were accompanied by conventional measuring systems comprising a side extensometer, an ACPD-measurement system and the GPU-based DIC system. The results indicate that DIC-measured crack depth correlates well with ACPD crack depth values. Furthermore, the crack-flank displacements derived from DIC-evaluation are in good agreement with 2D-FEM simulations. The GPU-based DIC-System appears as a promising measurement technique for crack growth investigations.

Digital image correlation (DIC) is a highly accurate image-based deformation measurement method achieving a repeatability in the range of σ= 10−5 relative to the field-of-view. The method is well accepted in material testing for non-contact strain measurement. However, the correlation makes it computationally slow on conventional, CPU-based computers. Recently, there have been DIC implementations based on graphics processing units (GPU) for strain-field evaluations with numerous templates per image at rather low image rates, but there are no real-time implementations for fast strain measurements with sampling rates above 1 kHz. In this article, a GPU-based 2D-DIC system is described achieving a strain sampling rate of 1.2 kHz with a latency of less than 2 milliseconds. In addition, the system uses the incidental, characteristic microstructure of the specimen surface for marker-free correlation, without need for any surface preparation—even on polished hourglass specimen. The system generates an elongation signal for standard PID-controllers of testing machines so that it directly replaces mechanical extensometers. Strain-controlled LCF measurements of steel, aluminum, and nickel-based superalloys at temperatures of up to 1000 °C are reported and the performance is compared to other path-dependent and path-independent DIC systems. According to our knowledge, this is one of the first GPU-based image processing systems for real-time closed-loop applications.

This article reports a novel GPU-based 2D digital image correlation system (2D-DIC) overcoming two major limitations of this technique: It measures marker-free, i.e. without sample preparation, and the sampling rate meets the recommendations of ASTM E606. The GPU implementation enables zero-normalized cross correlation (ZNCC) calculation rates of up to 25 kHz for 256 × 256 pixel ROIs. This high-speed image processing system is combined with a high-resolution telecentric lens observing a 10 mm field-of-view, coaxial LED illumination, and a camera acquiring 2040 × 256 pixel images with 1.2 kHz. The optics resolve the microstructure of the surface even of polished cylindrical steel specimen. The displacement uncertainty is below 0.5 μm and the reproducibility in zero-strain tests approximately 10-5 (1 σ) of the field-of-view. For strain-controlled testing, a minimum of two displacement subsets per image are evaluated for average strain with a sampling rate of 1.2 kHz. Similar to mechanical extensometers, an analogue 0-10V displacement signal serves as a feedback for standard PID controllers. The average latency is below 2 ms allowing for cycle frequencies up to 10 Hz. For strain-field measurement, the number of ROIs limits the frame rate, e.g., the correlation rate of 25 kHz is sufficient to evaluate 10 images per second with 2500 ROIs each. This frame rate is still sufficient to compare the maximum and minimum strain fields within a cycle in real-time, e.g. for crack detection. The result is a marker-free and non-contact DIC sensor suitable for both strain-controlled fatigue testing and real-time full-field strain evaluation.

In macro welding-processes, monitoring and closed-loop control by using the so-called “full penetration hole” is well known. This paper reports first results for the transfer of this technology to laser micro-welding processes for lap joints as they are used for housings or fuel cells. The laser source was a 400 W cw fiber laser with a Gaussian beam profile and a spot size of 28 μm. A cellular neural network (CNN) camera was used to measure the image feature of the full penetration hole within the thermal image of the welding process. For full penetration weldings on stainless steel samples, the laser power was controlled by the rate of full penetration hole detection. The effect of the feedback system is that the laser power is automatically adapted to changes in sheet thickness or feeding rate. The sheet thickness was varied between 220 and 350 μm and the feeding rate between 10 and 40 m/min without significant change in the weld seam quality. The closed-loop system increases the robustness of the process against perturbations and the process is always guided at the minimum laser power necessary for full penetration thus reducing spatter and smoke residues.

In emitter wrap through (EWT) solar cells, laser drilling is used to increase the light sensitive area by removing emitter contacts from the front side of the cell. For a cell area of 156 x 156 mm(2), about 24000 via-holes with a diameter of 60 mu m have to be drilled into silicon wafers with a thickness of 200 mu m. The processing time of 10 to 20 s is determined by the number of laser pulses required for safely opening every hole on the bottom side. Therefore, the largest wafer thickness occurring in a production line defines the processing time. However, wafer thickness varies by roughly +/- 20 %. To reduce the processing time, a coaxial camera control system was integrated into the laser scanner. It observes the bottom breakthrough from the front side of the wafer by measuring the process emissions of every single laser pulse. To achieve the frame rates and latency times required by the repetition rate of the laser (10 kHz), a camera based on cellular neural networks (CNN) was used where the images are processed directly on the camera chip by 176 x 144 sensor-processor elements. One image per laser pulse is processed within 36 mu s corresponding to a maximum pulse rate of 25 kHz. The laser is stopped when all of the holes are open on the bottom side. The result is a quality control system in which the processing time of a production line is defined by average instead of maximum wafer thickness.

A continuous increase in production speed and manufacturing precision raises a demand for the automated detection of small image features on rapidly moving surfaces. An example are wire drawing processes where kilometers of cylindrical metal surfaces moving with 10 m/s have to be inspected for defects such as scratches, dents, grooves, or chatter marks with a lateral size of 100 μm in real time. Up to now, complex eddy current systems are used for quality control instead of line cameras, because the ratio between lateral feature size and surface speed is limited by the data transport between camera and computer. This bottleneck is avoided by “cellular neural network” (CNN) cameras which enable image processing directly on the camera chip. This article reports results achieved with a demonstrator based on this novel analogue camera – computer system. The results show that computational speed and accuracy of the analogue computer system are sufficient to detect and discriminate the different types of defects. Area images with 176 x 144 pixels are acquired and evaluated in real time with frame rates of 4 to 10 kHz – depending on the number of defects to be detected. These frame rates correspond to equivalent line rates on line cameras between 360 and 880 kHz, a number far beyond the available features. Using the relation between lateral feature size and surface speed as a figure of merit, the CNN based system outperforms conventional image processing systems by an order of magnitude.

Zusammenfassung Kameras, die auf Zellularen Neuronalen Netzwerken basieren, können neue Anwendungsgebiete für die industrielle Bildverarbeitung erschließen, denn Rechenelemente lassen sich direkt in die elektronische Beschaltung von CMOS-Kamerapixeln integrieren. Insbesondere Regel- und Steuerungssysteme profitieren von der hohen Rechenleistung und den kurzen Latenzzeiten dieser neuartigen Kameratechnologie. Was damit jenseits konventioneller Bildverarbeitungssysteme möglich ist, wird anhand von drei konkreten Anwendungen gezeigt: beim Laserschweißen, bei der Laserablation und beim Drahtziehen.

Although laser-welding processes are frequently used in industrial production the quality control of these processes is not satisfactory yet. Until recently, the "full penetration hole" was presumed as an image feature which appears when the keyhole opens at the bottom of the work piece. Therefore it was used as an indicator for full penetration only. We used a novel camera based on "cellular neural networks" which enables measurements at frame rates up to 14 kHz. The results show that the occurrence of the full penetration hole can be described as a stochastic process. The probability to observe it increases near the full penetration state. In overlap joints, a very similar image feature appears when the penetration depth reaches the gap between the sheets. This stochastic process is exploited by a closed-loop system which controls penetration depth near the bottom of the work piece ("full penetration") or near the gap in overlap joints ("partial penetration"). It guides the welding process at the minimum laser power necessary for the required penetration depth. As a result, defects like spatters are reduced considerably and the penetration depth becomes independent of process drifts such as feeding rate or pollution on protection glasses.

Laser beam welding (LBW) has been largely used in manufacturing processes ranging from automobile production to precision mechanics. The complexity of LBW requires the development of strategies for the real-time control of the process. Most of the available feedback systems lack of temporal and/or spatial resolution and, therefore, they hardly allow observing more than one characteristic of the process. In the last years, we proposed some high-speed visual algorithms for image feature extraction from process images. The detection of the full penetration hole (FPH) allowed controlling the laser power at rates of up to 14 kHz. Another strategy enables observing the occurrence of spatters at monitoring rates of 15 kHz. The achievement of these results was made possible by the adoption of a visual system including a focal plane processor programmable by typical Cellular Neural Network (CNN) operations. This paper is focused on a new visual algorithm for the simultaneous detection of FPH and spatters, which led to real-time control rates of about 8 kHz. Besides the algorithm description, some interesting experimental results will be presented.

Cameras based on Cellular Neural Networks can discover new applications for industrial image processing. Computing elements can be integrated directly into the electronic circuits of CMOS camera pixels. In particular, control systems benefit from the increased computing power and the latency of this new camera technology. Three specific applications show what's possible beyond conventional imaging systems: the cases of laser welding, laser ablation and wire drawing.

Zusammenfassung Kameras, welche auf Zellularen Neuronalen Netzwerken basieren, können neue Anwendungsgebiete für die industrielle Bildverarbeitung erschließen, weil diese Technologie die Integration von Rechenelementen in die elektronische Beschaltung der Pixel von CMOS-Kameras ermöglicht. Insbesondere Regel- und Steuerungssysteme profitieren von der hohen Rechenleistung und den kurzen Latenzzeiten dieser neuartigen Kameratechnologie. Dieser Artikel verdeutlicht das anhand dreier Anwendungen aus den Bereichen Laserschweißen, Laserablation und Drahtziehen, welche mit konventionellen Bildverarbeitungssystemen nicht möglich sind.

An important issue in industrial quality control is the inspection of rapidly moving surfaces for small defects such as scratches, dents, grooves, or chatter marks. This paper investigates the potential of the EyeRIS 1.3 camera as a state-of-the-art camera based on “cellular neural networks” (CNN) for this application in comparison to conventional image processing systems. Based on experimental data from an aluminum wire drawing process where defects with a lateral size of 100 μm have to be detected at feeding rates of 10 m/s, the potential specifications for other surface inspection applications are estimated. Using the relation between the lateral defect size and the feeding rate as a figure of merit, the CNN based system outperforms conventional image processing systems by an order or magnitude in this particular application. In general, the lighting system limits the performance at lower defect sizes and the computational power at larger defect sizes and fields of view.