Fraunhofer Institute for Mechanics of Materials IWMDetection of fatigue damage before crack initiation
With our multiaxial resonance fatigue apparatus, we detect early signs of fatigue such as pore, slip band, extrusion and crack initiation in individual grains as well as the growth of microcracks and short cracks via a characteristic drop in the resonance frequency of microsamples.
Two piezo actuators indirectly excite the microsample to vibrate at its natural frequency (100-2000 Hz, depending on the sample geometry), while the vibration amplitude is measured and controlled via a laser system. Thus, a symmetrical load (R=-1) is set in this fatigue apparatus. Fatigue phenomena such as extrusion formation and crack nucleation in individual grains or the growth of microcracks can be detected very early by a characteristic drop in the resonance frequency. In addition to grain boundaries and precipitations, extrusions are failure-relevant precursors of cracks in many materials. Imaging techniques can be used to record image series during the course of fatigue, from which information on the kinetics of damage accumulation can be obtained.
As an example, the cyclic experiment is explained below using the change in the relative resonant frequency ∆f/finitial as a result of a bending resonance load in the HCF (High Cycle Fatigue) section of a nickel sample, see figure on the left. During the first ten thousand cycles, the resonance frequency changes only very slightly by approx. ∆f/finitial =-1 x 10-4. In this regime, extrusions form in many grains, which occasionally lead to the initiation of the first microcracks. At load cycle numbers above ten thousand, the resonance frequency already changes by ∆f/finitial =-1 x 10-3. This frequency change indicates that grain boundaries are exceeded by microcracks. A relative frequency change of ∆f/finitial =-5 x 10-3 is reached when the first long cracks run through the entire sample. From this point onwards, the resonance frequency decreases more and more until the sample finally breaks violently.
The high sensitivity achieved can be attributed to the optimized sample geometry with a small, highly loaded volume and the control strategy.
While the resonance frequency provides integral information on the fatigue state of the specimen, damage can also be localized.
By capturing a series of images over the course of fatigue, damage locations can be assigned to individual microstructural units such as grain boundaries, grains and grain clusters, allowing conclusions to be drawn about microstructure-property relationships and the underlying fatigue mechanisms. Furthermore, information on the kinetics of damage accumulation can be obtained. Since the sample typically deforms in a bending mode at frequencies up to 2 kHz, imaging the surface is not trivial and requires the use of a precise stroboscope system. This emits light pulses of just a few µs in duration at the zero crossing of the symmetrical bending.
The simultaneous recording of global and local damage indicators during the measurement can be supplemented by additional external analytics. For example, multimodal image registration algorithms can be used to superimpose surface-sensitive scanning electron microscopy (SEM) images or microstructure data on the image series. This sensor fusion makes it possible to combine high temporal resolutions in damage measurement with high local resolutions and information about the underlying microstructure. The information regarding the microstructure of the microsample is obtained from electron backscatter diffraction (EBSD) or light optical microscopy of etched surfaces. This versatile data set can be used as a basis for data- or knowledge-driven lifetime models and for the validation of such models. Within numerical modelling, such data sets can be interesting as a reference and for optimization, especially for meso- and micromechanical models. These include, for example, crystal plasticity or discrete dislocation dynamics. The defined boundary conditions achieved, together with the small sample volumes, make it possible to map microstructures and loads in computationally complex simulations. The interaction of, for example, crystal plasticity models and experimental micromechanics can thus provide insights into failure mechanisms.
High-resolution damage analysis and extrusion formation kinetics with ROCS- microscopy
As part of a master's thesis in 2019, ROCS microscopy (Rotating Coherent Scattering Microscopy, Jünger et al. 2016) was combined with the fatigue of microsamples on the multiaxial resonance fatigue apparatus. The aim is to observe early damage mechanisms with high optical resolution. The recorded image series will be used to draw conclusions about the kinetics of the earliest damage mechanisms. ROCS microscopy, originally developed for the observation of biological cells, enables high-resolution imaging in terms of time and space for mospheric conditions, which complements in situ experiments in the scanning electron microscope under vacuum.
- Gublay, O.; Ackermann, M.; Gramlich, A.; Durmaz, A.R.; Krupp, U., Influence of transformation temperature on the high-cycle fatigue performance of carbide-bearing and carbide-free bainite, Steel Research International, Online first (2023) Art. 2300238, 14 Seiten Link
- Durmaz, A.R.; Hadzic, N.; Straub, T.; Eberl, C.; Gumbsch, P., Efficient experimental and data-centered workflow for microstructure-based fatigue data: Towards a data basis for predictive AI models, Experimental Mechanics 61/9 (2021) 1489-1502 Link
- Natkowski, E.; Durmaz, A. R.; Sonnweber-Ribic, P.; Münstermann, S., Fatigue lifetime prediction with a validated micromechanical short crack model for the ferritic steel EN 1.4003, International Journal of Fatigue 152 (2021) Art. 106418; 15 Seiten Link
- Thomas, A.; Durmaz, A. R.; Straub, T.; Eberl, C., Automated quantitative analyses of fatigue induced surface damage by deep learning, Materials 13/15 (2020) Art. 3298, 24 Seiten Link
- Buck, M.; Straub, T.; Eberl, C., Experimental investigation of damage detection and crack initiation up to the very high cycle fatigue regime; Fatigue of Materials at Very High Numbers of Loading Cycles : Experimental Techniques, Mechanisms, Modeling and Fatigue Life Assessment, Christ, H.J. (Hrsg.) Springer Fachmedien, Wiesbaden (2018) 365-393 Link
- Straub, T.; Berwind, M.F.; Kennerknecht, T.; Lapusta, Y.; Eberl, C., Small-scale multiaxial setup for damage detection into the very high cycle fatigue regime, Experimental Mechanics 55/7 (2015) 1285-1299 Link