Low cycle lifetime assessment of Al2024 alloys
Loading...
Date
2011-12-19
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Summary and Conclusions
Many research groups around the globe take a direct route from failure undermonotonic loading to high cycle fatigue. Surprisingly, low cycle fatigue being
a border phenomenon between the aforementioned failure phenomena, is often
not considered. In the present thesis, the gap between failure under monotonic
loading and HCF was closed by thorough LCF analyses. As a practically
relevant material, the high-strength aluminium alloy Al2024 has been chosen.
The SRµCT investigations have made it evident that micro-mechanics based
porous plasticity damage models are not suitable for this particular alloy due
to a high number of dense particles. The particles are arranged in continuous
layers throughout the micro-structure. Studying the microstructure and
fracture mechanisms, it can be safely stated that bigger inclusions in Al2024
sheets and plates are responsible for damage initiation in the LCF regime.
Al2024 thick plate, due to its underlying manufacturing process develops precipitates
in the middle of the plate, creating layers within. A transition of
fracture mode, from surface to internal fracture, was observed with increasing
plastic range in cyclic experiments. In an ordinary low cycle regime (40-200
cycles), a fatal crack is generated by the propagation and frequent coalescence
of small surface cracks. Eventually, a mesocrack initiates from the surface of
the specimen due to the presence of bigger intermetallics.
For a 100 mm thick plate the S-direction has been found to be very brittle when
compared to L and T-directions (rolling directions) respectively. Although
there is a ductility observed, the resulting failure occurs without macroscopic
softening. An identical response was also observed in 4 mm thin sheets. For
allowing symmetric strain amplitudes (R = -1) in such sheets, a
oating antibuckling
guide has been proposed which successfully prevents buckling at high
compressive loading without affecting adversely the mechanical behaviour of
the specimen. In summary, the material behaviour in S-direction can be characterised
as neither completely brittle nor completely ductile. More precisely,
plastic strain accumulation as damage driving process as well as small fatigue
crack growth as brittle damage mechanism can be seen for LCF. This led to a
conclusion that for modelling low cycle fatigue for these materials, a material
model which incorporates both these effects (coupled ductile and brittle) was
required.
For that purpose, a phenomenological CDM approach was considered in the
present thesis. In our earlier analyses in the past, only ductile damage was
taken into account. However, since the microstructural arrangement of Al2024,
as observed in our experiments favour also brittle failure modes, a fully coupled
ductile-brittle model has been proposed. While ductile damage has been
modelled in a similar fashion as advocated formerly by other authors (see, e.g.
Lemaitre & Desmorat (2005)), a novel approach was elaborated for brittle
damage. In sharp contrast to the ductile damage model, material degradation
can already evolve below the yield limit. Since early work on a large number
of smooth and notched specimen demonstrated that wide variations in commercial
aluminium alloys caused little or no detectable differences in fatigue
strength, the novel coupled damage model is also applicable to a wider range
of aluminium alloys. The material parameters necessary to use the proposed
model can be easily obtained from a hysteresis loop test in order to determine
the cyclic stress-strain curve. The predicted LCF lifetimes for A12024
alloy are in good agreement the respect experiments verifying the underlying
assumptions.
Description
Table of contents
Keywords
Aluminium alloys, Continuum damage mechanics (CDM), Low cycle fatigue, Quasi-brittle damage