Filtern
Erscheinungsjahr
Dokumenttyp
Referierte Publikation
- nein (24) (entfernen)
Schlagworte
- Ductile cast iron (3)
- Dynamic crack resistance curve (2)
- Aircraft crash (1)
- Brüche (1)
- Constitutive modeling (1)
- Constraint (1)
- Crack propagation (1)
- Crack resistance curve (1)
- Damage and fracture behaviour (1)
- Dampferzeuger (1)
Eingeladener Vortrag
- nein (8)
The finite element method was used to simulate the thermo-mechanical behaviour of a flip chip under thermocyclic loading between -55°C (-67°F) and 125°C (257°F). In order to simulate realistically the inelastic material response of the solder bump, Chaboche's viscoplastic material model instead of a classical ORNL-theory was implemented into the finite element code ABAQUS and applied to a flip chip configuration. This so-called unified constitutive model is able to describe the Bauschinger effect as well as the creep/plasticity interaction. Herewith the influence of the substrate material on the inelastic deformation behaviour of a tin-lead solder bump was analysed. A comparison is given with respect to the influence of the two substrate materials namely Al<sub>2</sub>O<sub>3 </sub> and our self-developed Low Temperature Co-fired Ceramics (LTCC) on the inelastic response of the Sn60Pb40 eutectic solder material. It may be shown that the thermoelastic behaviour of the LTCC substrate significantly reduces the cyclic stresses in the bump compared to Al<sub>2</sub>O<sub>3</sub> as well as the cumulative inelastic strain which is responsible for the development of damage and hence flip chip failure
Dynamic fracture mechanics characteristics are required for design and safety proof of advanced
ductile cast iron (DCI) components. In this paper, the analytical compliance ratio
(CR) and numerical finite element (FE) key curve methods were investigated for dynamic Rcurve
testing. Two DCI materials, a purely ferritic and a ferritic one with 18 % of pearlite,
were tested at room temperature (RT) and -40 °C. The special focus is on the impact of the
materials damage and fracture behaviour on performance and limitations of the investigated
methods.
Systematic metallographic and fractographic microstructural analyses were performed accompanying
the test program. Summarizing systematics of the specific damage behaviour
and fracture mechanisms were developed taking microstructure, loading rate, specimen geometry
and test temperature into account.
There has been a change from ductile to brittle fracture mechanism observed at -40 °C with
ferritic DCI and even at RT with ferritic-pearlitic DCI. The ferritic-pearlitic DCI did not show Rcurve
behaviour at -40 °C and only single Juc values could be determined. Furthermore, with
ferritic-pearlitic DCI, random pearlite shares in the ligament caused the relation between
loading rate, force-deflection record and crack extension not to remain uniquely defined.
Therefore, the FE key curve method proofed not to be applicable since the FE model was
based on homogeneous isotropic material and ductile failure mode only. The analytical CR
key curve method worked well only with selected combinations of microstructure, specimen
geometry and temperature.
In summary it has to be concluded that a robust, accurate and sufficiently simple dynamic
key curve method that was suited for application in quality control of industrial relevant DCI
qualities cannot be provided currently.
The finite element method was used to simulate the thermo-mechanical behaviour of a solder bump configuration on Al2O3 substrate under thermo-cyclic loading between -55 °C and 125 °C. Chaboche's viscoplastic material model was used to describe the Bauschinger effect and the creep/plasticity interaction for the tin-lead solder material. The influence of two different viscoelastic underfill materials (Epoxy+50 % SiO2, Epoxy+70 % SiO2) on the inelastic deformation of the tin-lead solder was analyzed and compared. It may be shown that the thermally matched 'hard' underfiller (Epoxy+70 % SiO2) significantly reduces the accumulated inelastic strain in the bump compared to 'soft' underfiller or no underfill material. On the other hand, the maximum lateral normal stress will increase slightly.