Speaker
Description
Early detection of fatigue damage in metallic structures and complex components is limited by slow and often destructive procedures based on S–N curves, crack monitoring, and post-failure crack analysis. Infrared thermography (IRT) has recently been employed to identify ``invisible'' fatigue damage in thermosetting epoxy resins through a thermodynamic indicator based on entropy generation. This approach enabled non-destructive, full-field quantification of fatigue damage long before the onset of visible cracking. In that framework, equivalence is imposed between mechanically generated entropy, derived from cyclic stress--strain hysteresis, and thermal entropy, obtained from thermophysical properties measured by lock-in thermography. The present joint research investigates whether the same entropic damage indicator can be extended to structural metallic materials.
Fatigue tests are conducted on metallic specimens with notched geometries under controlled cyclic loading. In order to probe different fatigue regimes and damage mechanisms, specimens are intentionally loaded both below and above the fatigue limit. A key motivation comes from thermographic fatigue studies on steels showing that dissipative mechanisms change across the fatigue limit: anelastic dissipation is dominant below the endurance limit, whereas microplasticity and dislocation activity increasingly govern dissipation above it.
Based on previous studies, mechanical entropy generation was assumed to have the same value as thermal entropy generation.
Mechanical entropy was evaluated from the stress–strain response and thermal entropy was independently measured by means of lock-in thermography in vacuum conditions to minimise convective losses and improve repeatability. Measurements are acquired on specimens in the as-received state and after prescribed fatigue exposures. From amplitude and phase response, spatially resolved thermophysical properties (effective thermal diffusivity, thermal conductivity and volumetric heat capacity) are extracted, allowing computation of local thermal entropy generation. The temperature dependence of the specific heat capacity is described by a dedicated relationship formulated on the basis of the Debye model and the density is determined according to the Archimedes principle. The resulting entropy fields are then compared with conventional fatigue indicator in order to validate the applicability and sensitivity of the entropic metric to metallic materials.
To complement the entropic indicator, also a modified Two-Curve Method (TCM) is applied to the thermographic self-heating data to identify the transition between below-limit anelastic dissipation and above-limit microplasticity-dominated dissipation. Different thermal–stress relationships have been proposed in the literature to capture these two regimes and their transition. Below-limit conditions target run-out/high-cycle regimes where damage is subtle and distributed; above-limit conditions accelerate damage accumulation and promote early crack initiation and crack-tip dissipation.
The endurance limit is then obtained as the intersection of the two fitted regimes.
The overall aim is to assess whether the methodology can be generalised and thermography can be regarded as a suitable non-destructive technique for early detection and quantitative assessment of fatigue damage in metallic structures relevant to mechanical and aerospace engineering.
